EVOLVING PERSPECTIVES IN THE DEVELOPMENT OF INDIAN INFRASTRUCTURE

EVOLVING PERSPECTIVES IN THE DEVELOPMENT OF INDIAN INFRASTRUCTURE

Volume 1

Infrastructure Development Finance Company Limited ORIENT BLACKSWAN PRIVATE LIMITED

Registered office 3-6-752 Himayatnagar, Hyderabad 500 029 (A.P.), India Email: [email protected]

Other offices Bangalore, Bhopal, Bhubaneshwar, Chennai Ernakulam, Guwahati, Hyderabad, Jaipur, Kolkata, Lucknow, Mumbai, New Delhi, Noida, Patna

© Infrastructure Development Finance Company Limited 2012 First published 2012

All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Orient Blackswan Private Limited.

ISBN 978 81 250 4666 0

Typeset in Minion 11/14 by Trendz Phototypesetters, Mumbai 400 001

Printed in India at Aegean Offset Printers, Greater Noida

Published by Orient Blackswan Private Limited 1/24 Asaf Ali Road New Delhi 110 002 E-mail: [email protected]

The external boundary and coastline of India as depicted in the maps in this book are neither correct nor authentic. CONTENTS

List of Tables, Figures and Boxes vii Foreword by Rajiv B. Lall xv Acknowledgements xvii Introduction xix List of Abbreviations xxiii

ENERGY 1 Power Sector Reform: Policy Decisions in Distribution 3 2 Power Sector in India: A Summary Description 13 3 Power Sector Reform in Argentina: A Summary Description 22 4 Orissa Power Sector Reform: A Brief Overview of the Process 31 5 Power Sector Financing: A Note on Conditionalities 40 6 Six Steps to Accelerated Privatisation of Electricity Distribution 52 7 Regulation of Petroleum Product Pipelines 70 8 Incentives, Ownership and Performance in Power Sector: The Case of UP 108 9 Discussion Paper on Developing Power Markets 124 10 Captive Coal Mining by Private Power Developers: Issues and the Road Ahead 143 11 Power Distribution Reforms in Andhra Pradesh 163 12 Power Distribution Reforms in Maharashtra 181 13 Power Distribution Reforms in Gujarat 202 14 Barriers to Development of Renewable Energy in India and Proposed Recommendations: A Discussion Paper 222 15 Power Distribution Reforms in Delhi 273 16 Power Distribution: Being Driven to Insolvency by a Governance Crisis 290 17 India Solar Policy: Elements Casting Shadow on Harnessing the Potential 334

TELECOMMUNICATIONS 18 Telecom Sector Reform: Restructuring Telecommunications as if the Future Mattered 363 19 Transitioning from Administrative Allocation of Spectrum to a Market-based Approach 372

LIST OF TABLES, FIGURES AND BOXES

Tables Table 3.1 Total production of electricity and the share of different types of generation 22 Table 6.1 Timeline for accelerated privatisation of distribution 68 Table 7.1 Supply/demand—petroleum products (in mmt) 74 Table 7.2 Supply/demand—natural gas (in mmscmd) 76 Table 7.3 Gas demand estimates of different government agencies 78 Table 7.4 Total gas supply potential—Tenth Plan (in mmscmd) 79 Table 7.5 Total gas supply potential post K–G Basin find (in mmscmd) 80 Table 7.6 Projections for POL consumption 2001–2008 (in ‘000 metric tonnes) 97 Table 7.7 List of LNG terminals 97 Table 7.8 List of crude pipelines 98 Table 7.9 List of gas pipelines 98 Table 7.10 List of refineries 99 Table 7.11 List of oil product pipelines 99 Table 7.12 List of ports handling oil/petroleum products 100 Table 8.1 Cumulative commercial losses of consolidated UPPCL 111 Table 8.2 Performance parameters 112 Table 8.3 Performance of UPRVUNL generating stations 114 Table 8.4 Reduction of AT&C loss in North Delhi Power Ltd 117 Table 8.5 Collection efficiency (%)—governmental and non-governmental categories 119 Table 9.1 Recommended loan conditions 132 Table 9.2 Trading margins 136 Table 10.1 Coal blocks identified for the power sector 144 Table 10.2 Criteria for allocation of coal blocks 146 Table 10.3 Pre-production approvals for allottees of coal blocks 147 Table 10.4 Comparison of mining leases in Australia, Canada and India 151 Table 10.5 Normative time limit ceilings as provided in guidelines for allocation of captive blocks and conditions of allotment through the screening committee, Ministry of Coal 156 viii | Indian Infrastructure: Evolving Perspectives

Table 10.6 Minimum time frame of process 159 Table 11.1 Steps taken for power sector reforms in Andhra Pradesh 168 Table 11.2 Investment in infrastructure 172 Table 11.3 AT&C losses (%) of distribution companies 174 Table 11.4 Collection efficiency (%) of distribution companies 175 Table 11.5 Subsidy received 175 Table 11.6 Profit with and without subsidy 176 Table 11.7 Peak deficit and energy deficit in AP 177 Table 11.8 Subsidy received by distribution companies 178 Table 11.9 Sales mix (%) of distribution companies 179 Table 11.10 Revenue mix (%) of distribution companies 179 Table 11.11 Expenses as (%) of total cost for distribution companies 179 Table 12.1 Progress of Single Phasing Scheme 188 Table 12.2 Power scenario in Bhiwandi—before and after franchising 192 Table 12.3 Profits of MSEDCL 196 Table 12.4 MSEDCL’s arrears 196 Table 12.5 Sales mix (as percentage of total units sold) 200 Table 12.6 Revenue mix (as percentage of total revenue) 200 Table 12.7 Expenses as percentage of total expense 201 Table 13.1 AT&C losses for distribution companies 213 Table 13.2 Distribution losses of discoms 213 Table 13.3 Collection efficiency of distribution companies 214 Table 13.4 Subsidy received by distribution companies 215 Table 13.5 Gap between ARR and ACS for distribution companies without subsidy 216 Table 13.6 Profits of distribution companies without subsidy 216 Table 13.7 Profits of distribution companies with subsidy 216 Table 13.8 Gujarat state peak deficit and energy deficit 217 Table 13.9 Quality of Service parameters for discoms 218 Table 13.10 Sales mix of distribution companies 219 Table 13.11 Revenue mix of distribution companies 219 List of Tables, Figures and Boxes | ix

Table 13.12 Expenses as percentage of total cost for distribution companies 219 Table 14.1 RE potential and target cumulative capacity addition (in MWeq) 225 Table 14.2 Mismatch between RE capacity envisaged under policy and capacity addition targeted 229 Table 14.3 Policy instruments for promotion of RE 232 Table 14.4 Regulatory framework for promotion of RE 234 Table 14.5 Penalties for non-achievement of RPO 235 Table 14.6 Status of RPO across Maharashtra 235 Table 14.7 RET capacity added across states with tariff orders/FiTs 236 Table 14.8 Year-wise wind power capacity addition in Andhra Pradesh (in MW) 236 Table 14.9 Salient features of the schemes proposed under the solar power purchase policy of JNNSM 248 Table 14.10 Summary of RPOs at state level for select states 264 Table 14.11 FiTs for wind energy and assumptions for FiTs across states 266 Table 14.12 FiITs for solar power across states 267 Table 14.13 FiTs for SHP and assumptions for FiTs across states 268 Table 14.14 FiTs for biomass and bagasse and assumptions for FiTs across states 269 Table 15.1 Accepted bid loss rejection trajectory; minimum bid loss rejection trajectory 276 Table 15.2 Profits 280 Table 15.3 QoS parameters 281 Table 15.4 Peak and energy deficit 282 Table 15.5 Loan to Transco 282 Table 15.6 Expenses break-up of discoms 284 Table 15.7 Capital expenditure by discoms 287 Table 15.8 AT&C loss reduction by discoms 287 Table 15.9 Sales and revenue mix 288 Table 16.1 States exhibiting increases in losses from distribution business 292 Table 16.2 States exhibiting profits or decrease in losses from distribution losses 293 x | Indian Infrastructure: Evolving Perspectives

Table 16.3 Most utilities have shown considerable reduction in AT&C losses between 2005–06 and 2008–09 294 Table 16.4 Agricultural consumption continues to remain unmetered—status in select states 296 Table 16.5 Level of agricultural consumption in select states in 2008–09 297 Table 16.6 Status of implementation of select distribution reform initiatives as of April 2010 298 Table 16.7 Cost recovery in 2008–09 301 Table 16.8 Status of tariff revision in states/union territories at the end of 2009 307 Table 16.9 Increase in revenue gap without subsidy for utilities between 2005–06 and 2008–09 309 Table 16.10 Consumer tariffs as percentage of Average Cost of Supply approved by SERCs in FY 2008–09 311 Table 16.11 Funding of revenue gap of utilities in Uttar Pradesh 315 Table 16.12 Means of financing the revenue deficits of utilities (indicative) 318 Table 16.13 Outstanding bank loans and government guarantees in select states (Rs crore) 320 Table 16.14 Interest expenses disallowed by ERCs primarily on account of short-term loans taken by distribution utilities 323 Table 16.15 Estimated financial losses of utilities in 2012–13 325 Table 16.16 Tariff trends in UMPP bids 327 Table 17.1 Projected deployment of funds in SPSA 340 Table 17.2 List of projects selected under migration scheme of JNNSM 353 Table 17.3 List of projects selected under JNNSM for bundling scheme 354 Table 17.4 Allotment of solar capacities in Gujarat 356 Table 19.1 International experience in spectrum trading 379 Table 19.2 Example of spectrum trading with revenue share payments 384 Table 19.3 Example of revenue neutral differential pricing between SSUs 386 Table 19.4 Allocation of spectrum in other countries 388 Table 19.5 Allocation of spectrum in Indian telecom circles 389 Table 19.6 Public spectrum registry—example of contents 390 List of Tables, Figures and Boxes | xi

Figures Figure 2.1 Institutional structure of the Indian power sector 14 Figure 3.1 Restructuring of the Argentine electricity industry (federal assets) 24 Figure 3.2 Estimate medium monomial contract price in the market 26 Figure 3.3 Argentina’s transmission system 27 Figure 3.4 Evolution of capacity and energy prices 28 Figure 3.5 Outages as a per cent of energy demand 29 Figure 7.1 Growth of GDP versus POL consumption 75 Figure 7.2 Aggregate supply/demand POL 2001–02 76 Figure 7.3 Main consumers of natural gas 77 Figure 7.4 LNG prices, US gas prices and crude oil prices 81 Figure 7.5 Comparative transportation costs for road, rail and pipeline modes 84 Figure 7.6 Influence of amortization for a 12” diameter pipeline 85 Figure 7.7 OPEC supply/price dynamics 101 Figure 10.1 PERT chart for coal mine development 149 Figure 10.2 Flowchart of mining proposal approval process 158 Figure 11.1 Cost of power supply, average tariff and gap 164 Figure 11.2 Power sector: Structure pre- and post-reforms 169 Figure 11.3 AT&C losses 173 Figure 11.4 Collection efficiency 174 Figure 11.5 Subsidy received 175 Figure 11.6 ACS, ARR (without subsidy) and gap over the years 177 Figure 12.1 Average cost and realization of power in 2000–01 182 Figure 12.2 Restructuring of MSEB 185 Figure 12.3 AT&C losses 194 Figure 12.4 Collection efficiency 195 Figure 12.5 Subsidy from state 195 xii | Indian Infrastructure: Evolving Perspectives

Figure 12.6 Average revenue realised (ARR), average cost of supply (ACS) and the gap between them over the years for Maharashtra 196 Figure 12.7 Average revenue realised (ARR) without subsidy, average cost of supply (ACS) and the gap between them over the years for Maharashtra post-reforms 197 Figure 12.8 SAIFI 197 Figure 12.9 SAIDI 198 Figure 12.10 CAIDI 198 Figure 13.1 Average revenue realisation in Rs/kWh for various consumer categories 203 Figure 13.2 Restructuring of GEB 205 Figure 13.3 Improvement in cash collections over the years 210 Figure 13.4 AT&C losses 213 Figure 13.5 Subsidy received 214 Figure 13.6 ARR, ACS and gap between them over the years 215 Figure 14.1 Role of RE in India’s power generation capacity as on 31 March 2009 223 Figure 14.2 Technology-wise grid-interactive RE capacity in India as on 31 October 2009 224 Figure 14.3 Events influencing RE development and RE capacity addition 228 Figure 14.4 Bundling mechanism for sale of solar power under JNNSM 247 Figure 14.5 Time frame for completion of migration scheme under solar power purchase policy of JNNSM 247 Figure 14.6 Time frame for completion of scheme for new projects under solar power purchase policy of JNNSM 247 Figure 15.1 T&D losses and commercial losses pre-reforms 274 Figure 15.2 AT&C losses 280 Figure 15.3 ARR & ACS 281 Figure 16.1 Losses without subsidy for distribution utilities have risen sharply in 2008–09 291 List of Tables, Figures and Boxes | xiii

Figure 16.2 Cash losses before subsidy received for distribution utilities have trebled between 2005–06 and 2008–09 291

Figure 16.3 All-India AT&C losses are below 30% 295

Figure 16.4 Gap between ARR (without subsidy) and ACS at the all-India level has increased 300

Figure 16.5 In recent years, tariff increase has not kept pace with increasing ACS 302

Figure 16.6 Power purchase costs have increased 303

Figure 16.7 Procurement of short term power is increasing 303

Figure 16.8 Short-term power prices have shot up 304

Figure 16.9 Peak and energy deficit in India 304

Figure 16.10 Purchase of short-term power in select states (MU) 305

Figure 16.11 Coal imports for power plants have doubled between 2005–06 and 2008–09 306

Figure 16.12 Trends in price of imported coal 306

Figure 16.13 Increase in employee costs for distribution utilities in India 307

Figure 16.14 Subsidies booked by distribution utilities are rising but payment by state governments is inadequate 312

Figure 16.15 Top 10 states exhibiting the maximum increase in subsidy booked 312

Figure 16.16 States not paying full amount of subsidy to utilities 313

Figure 16.17 Distribution losses in Punjab 316

Figure 16.18 Distribution losses in Madhya Pradesh 316

Figure 16.19 Employee costs of PSEB 317

Figure 16.20 Debtors for sale/transmission of power for state-owned generation, trading and transmission companies 323

Figure 16.21 Private sector will lead future capacity addition in India 326

Figure 19.1 Variations in use of spectrum across circles and operators 378

Figure 19.2 Example of standard spectrum unit 381 xiv | Indian Infrastructure: Evolving Perspectives

Boxes Box 1.1 Take-or-pay in Indonesia 4 Box 1.2 Power sector reform in Argentina 9 Box 3.1 Objectives of CAMMESA 25 Box 3.2 Objectives of ENRE 25 Box 4.1 Timeline of key events in power sector reforms in Orissa 38 Box 6.1 Privatisation of generation assets of SEBs 56 Box 6.2 Limitations of the ‘mixed-zone’ structure 57 Box 6.3 Disadvantages of the single-buyer model 62 Box 7.1 New Exploration Licensing Policy 71 Box 7.2 Definition of petroleum products 73 Box 7.3 Impact of OCC on oil product prices 88 Box 7.4 Contract carrier versus common carriage carrier 90 Box 7.5 Main conversions used in the petroleum industry 102 Box 8.1 Memorandum of Understanding with GOI 113 Box 8.2 Noida Power Company (NPCL)— a successful distribution company 120 Box 8.3 KESCO privatization 121 Box 11.1 APSEB’s performance review 164 Box 12.1 MSEB’s performance review 183 Box 12.2 White paper on Maharashtra power sector reforms 184 Box 12.3 Load management 187 Box 14.1 Salient features of JNNSM 246 Box 14.2 Urjankur Nidhi Fund in Maharashtra 262 Box 14.3 Detailed provisions of National Solar Mission 271 Box 15.1 Computation and treatment of over/under achievement of target AT&C loss levels 275 Box 17.1 Solar Power: International Experience 348 Box 19.1 The spectrum is finally attached to the licence 374 Box 19.2 Defining spectrum trading units in Australia 381 Box 19.3 Ofcom’s proposed process for transacting a spectrum trade 382 FOREWORD Infrastructure Development Finance Company Limited (IDFC) was set up at the initiative of the Government of India with the mandate to ‘lead’ private capital to commercially viable infrastructure projects in India. Over the fifteen years of our existence, we have witnessed the share of private capital grow from an insignificant proportion to the current levels of almost 40 per cent of the total infrastructure investment. In the Twelfth Plan, the share of the private sector is expected to grow even further to 50 per cent of the investment. All through this period, IDFC has been actively engaged with governments, independent regulators, private developers, banks, financial investors and other stakeholders in the process of advocating and developing appropriate policy, and legal and regulatory frameworks for this purpose. We would, therefore, celebrate our fifteenth anniversary, later this year, with a great deal of satisfaction, now that private investment has become an important mode of investment across various infrastructure sectors. From the beginning, the role envisaged for IDFC was not limited merely to funding, but to lead private investment to this sector. In any case, there were very few private sector projects to finance in 1997. Leading capital to projects required intense preparatory work and active engagement with the government in policy formulation, in the preparation of legal and regulatory frameworks, and in the development of transparent procurement processes, objective evaluation criteria and equitable concession documents. IDFC, both through its policy advocacy and advisory services teams, provided inputs through a series of interventions. Some of these inputs were through specific advisory services transactions, for instance, in major ports and national highways. In power and telecom, inputs were often provided in response to the consultative process initiated by the regulator or government but were also through notes and opinions conveyed on specific issues. We believe that IDFC’s role was very useful and often pivotal in shaping opinion and in leading private investment to infrastructure. In fulfilment of its mandate to develop infrastructure, a large body of written material has been prepared over the years. These include sector studies, policy recommendations for resolving some of the difficult issues, overviews of emerging trends, outlines of good practices and issues in financing. Some of these find a place in the thematic India Infrastructure Report that is published every year. A few others have been published elsewhere, covered in IDFC’s quarterly policy reviews, or released as occasional papers. Many have remained in private domain. This publication is an anthology of some of the papers prepared by IDFC over the last fifteen years, and reflects its emerging views in the quest of nation building. It is a representative collection and not an exhaustive one. Last year we set up IDFC Foundation under Section 25 of the Companies Act 1956, as a wholly owned subsidiary of IDFC. IDFC’s development agenda is now being carried out through the IDFC Foundation. It is fitting that on the first anniversary of the Foundation, we are able to release this publication as a testament of IDFC’s contribution to infrastructure development in India.

RAJIV B. LALL March 2012 Managing Director and CEO

ACKNOWLEDGEMENTS It is always difficult to acknowledge all those who have contributed to a publication of this nature, which includes papers and notes written over a fifteen-year period. All the papers have received inputs of different kinds from various sources. The inputs include views, opinions, critical feedback, discussions, encouragement and sometimes just the opportunity to share ideas. In that sense, it would be impossible to acknowledge all the contributors. At the outset, it would only be fitting to acknowledge the Government of India and the various state governments that have, over the years, used IDFC as a sounding board for policy and regulatory advice. Within the organisation, at a strategic level, we would first of all acknowledge the encouragement and support of Deepak Parekh, Chairman, IDFC, Rakesh Mohan, who briefly served as Vice-Chairman, and its first two Managing Directors, D. J. Balaji Rao and Nasser Munjee. This support has been strongly continued by Rajiv B. Lall, the present Managing Director and CEO of IDFC, under whose stewardship the IDFC Foundation was set up. Acknowledgements are also due to Anil Baijal, Chairman, IDFC Foundation; Urjit R. Patel, who headed the policy group at IDFC for the first ten years; and Ritu Anand, who oversees it at present. The inputs received from the various members of IDFC’s Policy Advisory Boards from time to time, have certainly helped to clarify perspectives and shape policy opinions. For this, we thank each and every one of them. Where the papers already include the names of the authors, credits have been given at the end of the respective papers. Nevertheless, the following writers, too, need to be acknowledged: Rajiv B. Lall, Urjit R. Patel, Ritu Anand, Cherian Thomas, Partha Mukhopadhyay, Srikumar Tadimala, Anupam Rastogi, Nirmal Mohanty, P. V. Ravi, Saugata Bhattacharya, Piyush Tiwari, Sambit Basu, Shishir Mathur, Manisha Gulati, Kaushik Deb, Sunaina Kilachand, Aditi Jagtiani, Kunal Katara, Neeraj Sansanwal, Bhagyathej Reddy, Ashish Agarwal, Pritika Hingorani and Ranesh Nair. Their valuable contributions have gone into the compilation of these volumes, which provide both historical and current perspectives and could be useful points of reference in all our future efforts at nation building. Our thanks are due also to Orient BlackSwan, the publishers of this volume, who have been consistent in maintaining quality while accommodating the sometimes exacting demands made on them. Finally, we would like to thank all our colleagues at IDFC, past and present, whose consistent feedback and varying perspectives on the various themes have been of immense value. While we have taken care to include everyone who helped us in compiling this report, any omissions are unintentional and we hope that they would be construed as such.

INTRODUCTION Evolving Perspectives in the Development of Indian Infrastructure (in two volumes) is a compilation of several papers written by IDFC over the last fifteen years on the themes of infrastructure development and financing. It is not an exhaustive summary of all of IDFC’s work in infrastructure policy and regulation during this period. Much of IDFC’s policy contribution has been through inputs provided to the many task forces, working groups and high-level committees constituted by central and state governments from time to time, in which IDFC’s officials have been (or are being) represented. In many other instances, recommendations have been provided to governments and regulators as part of the consultative process followed while seeking views of stakeholders in these sectors. Some of IDFC’s development work has also been through specific outputs in the form of reports and documents prepared as part of advisory service transactions for governments. Another visible output that is published year after year is the India Infrastructure Report—a thematic report that discusses issues of contemporary concern across the infrastructure space. In some ways, this anthology captures some of IDFC’s emerging views over this period, which may be reflected in its recommendations and inputs in the activities listed above. This publication consists of two volumes. The first volume has two sections—energy and telecommunications. The second volume deals with transport, urban and other infrastructure, and infrastructure development and financing. It comprises 45 papers on a range of topics on various aspects of policy and financing. These cover the period from May 1998 to January 2012. The papers are of varying lengths, depending on the specific context in which they were written and the need sought to be addressed. Some of the issues debated in the earlier papers and the recommendations may still be relevant in the current context, and these volumes would have more than a historical significance. The power sector, not surprisingly, has the most number of papers—since it is the biggest of the infrastructure sectors—and these papers are reflective of the various emerging issues in the sector. A section- wise summary of the two volumes is set out below. The initial papers in the power sector (written 1998–2001) set out the issues and challenges for private investment in the sector. Drawing from international experiences in Argentina and Indonesia, right from the outset, IDFC highlighted the need to focus on reforms and on restructuring the sector with a view to increase efficiency and reduce losses. The suggestions for creating a more competitive market structure conducive to private sector participation— unbundling the monolithic electricity boards and creating independent regulatory authorities, separating network business from supply, and establishing power trading as a separate licensed activity—were forerunners to many subsequent changes, such as the introduction of open access. Equally noteworthy are the suggestions for private sector investment in power distribution, initially through franchising dense urban areas. The relative success of the franchisee model (in Bhiwandi and other places) in improving distribution efficiencies and collections reflect the merits of this suggestion. The separation of urban and non-urban areas and the development of a transparent mechanism for payment of subsidies will continue to be issues that engage us as the reform process in the sector moves ahead. xx | Indian Infrastructure: Evolving Perspectives

Some of the papers review the reforms and progress of the sector in the states of Orissa (February 2000), Uttar Pradesh (February 2005), Andhra Pradesh, Delhi, Maharashtra and Gujarat. The review includes an assessment of the gains made so far, as well as the shortcomings and the challenges ahead for each of these states. The most recent paper on the power distribution sector clearly highlights the severity of the losses in power distribution, the looming risk of insolvency of these utilities, and the urgent need for good governance to enable future capacity addition and to restore the overall health of the financial system. The wide gamut of subjects covered include the regulation of petroleum product pipelines, the conditionalities in power sector lending, and the challenges in implementing the captive coal mining policy. The section also includes a discussion paper on renewable energy and a specific paper on solar energy. The telecom section is rather small since much of the policy input was provided by IDFC as part of the consultative process in this sector with the government and the regulator. The initial paper on the sector (December 1998) set out some of the sectoral challenges, keeping in mind the new telecom policy that was to be announced. The second paper covered the principles for allocation of spectrum and also the issue of spectrum trading. The wisdom of using the market-based approach has been clearly vindicated by the subsequent events that have overtaken the sector. The second volume begins with a section on issues related to the transport sector. Much of the work done by IDFC in the transport sector was as Secretariat to the Task Force on Infrastructure and through specific advisory assignments in major ports and national highways sectors, in the initial years. The paper on ports sector reforms (December 1999) argued that there is little need for immediate additional public investment in the sector, but suggested a focus on efficiency in port services by private providers subject to competitive pressures. It also recommended a push to reform port labour, the need for strengthening hinterland connectivity, and that of putting in place a pro-competitive and stable regulatory system. All these represent a continuing agenda in the sector. A subsequent paper (November 2000) highlighted the need to integrate coastal shipping with the surface transport network. This is a challenge that has largely been ignored for several decades and could assume increasing importance given the enormous environmental benefits and lower transport costs that could accrue from increased coastal movement of goods. A paper on railways sets out the opportunities for public–private partnerships in the sector, and includes the experience of the Hassan–Mangalore railway project in Karnataka. Papers on the roads sector include a status note on the National Highway Development Programme (NHDP), a review of the challenges of financing highways, and, the most recent one, on the challenges for NHAI in financing the NHDP. The first paper in the urban sector covers the water sector. The paper (March 2001) highlighted the need for independent regulation, rational tariff setting and the appropriate role of the private sector. This is followed by a paper on special economic zones (SEZs) which argued that SEZs are more in the nature of ‘band aid’ fixes and that the focus of the policy should move from augmenting infrastructure facilities for export production to an overall focus on higher-quality infrastructure, growth and employment. A selection of papers Introduction | xxi

from IDFC policy quarterly research notes covers the issues of land pooling, bus rapid transit systems, green buildings, sewage water recycling for industrial use, and municipal borrowings using pooled finance mechanisms. The base note prepared on urban financing (for the High Powered Expert Committee) provides a comprehensive review of the financing for urban infrastructure. A paper on private healthcare in India (December 2002) broadly reviewed the key challenges of the sector at that point and identified a few business models for healthcare investment in future. The last section of the second volume includes papers on various cross-cutting issues in infrastructure development and financing. The earliest paper (May 1999) briefly reviewed the challenges of reforming the debt market in India, some of which still remain. A detailed paper on competitive bidding (August 2000) argued that competitive bidding provides the most efficient and cost effective method for procurement of infrastructure services—a lesson that has become systematised practice over the last decade across sectors. Two papers deal with the issue of regulation. The January 2005 paper reviews regulation across sectors and argues that regulation is one piece of the infrastructure puzzle and has to be complemented with an industry structure that aligns operators’ incentives towards pursuit of value for money. The more recent paper on regulation looks further at the future of regulation in India. Papers on financing comprise two independently written reviews: domestic financing and the role of IDFC, and infrastructure financing and the role of non- banking finance companies. Other papers include a short note on infrastructure development in India (prepared for the World Economic Forum), an analysis of the political economy of infrastructure development (which concludes that political logic would drive decisionmakers to deliver improved infrastructure with a greater reliance on private provision of these services), and a detailed comparison of infrastructure creation in China and in India. The last paper identifies some of the lessons that can be learnt from the Chinese experience, such as using improved technology, pursuing financially sustainable solutions, and improving efficiency and accountability in the government while pursuing solutions that are more implementable in our context. All in all, it is hoped that this publication would give the readers an insight into some of the evolving policy perspectives and recommendations that have emerged in the last fifteen years—a period when infrastructure development has received more focussed attention than at any time since Independence.

LIST OF ABBREVIATIONS

A$ Australian Dollar ACA Australian Communications Authority ACCC Australian Competition Consumer Council ACS Average Cost of Supply ADB Asian Development Bank ADF Airport Development Fees ADRD Alberta Department of Resource Development AEC Ahmedabad Electricity Company AEE Autorita per l'Energia Elettrica e il Gas AERA Aviation Economic Regulatory Authority AGCOM The Communications Regulatory Authority AGR Adjusted Gross Revenue AIM Alternative Investment Market AIP Administrative Incentive Pricing AMC Ahmedabad Municipal Corporation AP Andhra Pradesh APDRP Accelerated Power Development and Reform Programme APERC Andhra Pradesh Electricity Regulatory Commission APL Adaptable Programme Loan/Lending APM Administrative Price Mechanism APPSRP Andhra Pradesh Power Sector Restructuring Programme APSEB Andhra Pradesh State Electricity Board ARR Annual Revenue Requirement ARR Average Revenue Realised ASEAN Association of Southeast Asian Nations ASP Activated Sludge Process AT&C Aggregate Technical and Commercial ATE Appellate Tribunal for Electricity ATF Aviation Turbine Fuel AUDA Ahmedabad Urban Development Authority BBCD Bare-Boat-Charter-cum-Demise xxiv | Indian Infrastructure: Evolving Perspectives

bbl Barrels BCC Beneficiary Capital Contribution BCC Base Construction Cost bcm Billion Cubic Metres BCM Book Consolidation Module BEE Bureau of Energy Efficiency BEST Brihan Mumbai (Bombay) Electric Supply and Transport Undertaking BG Broad Gauge BIAL Bengaluru International Airport Limited BKCC B. K. Chaturvedi Committee BLD Billion Litres per Day BLT Build, Lease, Transfer BOD Biological Oxygen Demand BOLT Build, Operate, Lease, Transfer BOO Build, Own, Operate BOOM Build, Own, Operate, Maintain BOOST Build, Own, Operate, Share, Transfer BOOT Build, Own, Operate, Transfer BOQ Bill of Quantities BOT Build Operate Transfer BP British Petroleum BPCL Bharat Petroleum Corporation Limited BRPL Bongaigaon Refinery and Petrochemicals Limited BRTS Bus Rapid Transit System BS basic service BSES Bombay Suburban Electric Supply BSF Bond Service Fund BSNL Bharat Sanchar Nigam Limited BTS Bangkok Mass Transit System BUA Built-up Area BWSSB Bangalore Water Supply and Sanitation Board BYPL BSES Yamuna Power Limited List of Abbreviations | xxv

CA Constitutional Amendment CAA Constitutional Amendment Act CAA Civil Aviation Authority CAA Cost under the Annuity Approach CAGR Compound Annual Growth Rate CAIDI Consumer Average Interruption Duration Index CAM Common Area Maintenance CAMMESA Compañía Administradora del Mercado Mayorista Eléctrico CAT Consumer Analysis Tool CBDT Central Board of Direct Taxes CCA Cost under the Conventional Approach CCI Competition Commission of India CDB China Development Bank CDMA Code Division Multiple Access CDs Certificates of Deposits CEA Central Electricity Authority CEPZ Cochin Export Processing Zone CERC Central Electricity Regulatory Commission CESC Calcutta Electric Supply Company CESCO Central Electricity Supply Company CFC Consumer Facilitation Centres CFS Centre for Sight CGD City Gas Distribution CGHS Central Government Health Scheme CGWB Central Ground Water Board CIL Coal India Limited CLF Credit Local de France CMIE Centre for Monitoring Indian Economy CMS Cellular Mobile Service CMT Comisión del Mercado de las Telecomunicaciones CMTS Cellular Mobile Telephony Service CMW Chennai Metro Water xxvi | Indian Infrastructure: Evolving Perspectives

CNE Comisión Nacional de Energía COAI Cellular Operators Association of India CONCOR Container Corporation of India CONEA Coalition of North East Association CP Commercial Paper CPCB Central Pollution Control Board CPPs Captive Power Plants CPSU Central Public Sector Unit CPT Chennai Port CPUC California Public Utilities Commission CREF Credit Rating Enhancement Fund CRG Crisis Resolution Group CSE Centre for Science and Environment CST Concentrated Solar Thermal CSUs Central Sector Undertakings CTC Competitive Transition Charge DALY Disability Adjusted Life Years DBFO Design Build Finance Operate DELs Direct Exchange Lines DEPB Duty Entitlement Pass Book DERC Delhi Electricity Regulatory Commission DESU Delhi Electric Supply Undertaking DF Distribution Franchisee DFID Department for International Development DFIs Development Finance Institutions DFRC Duty Free Replenishment Certificate DGH Directorate General of Hydrocarbons DIAL Delhi International Airport Limited DIMTS Delhi Integrated Multimodal Transit System discom/distco Distribution Company DJB Delhi Jal Board DM De-mineralisation List of Abbreviations | xxvii

DMRC Delhi Metro Rail Corporation DoT Department of Telecommunications DSCR Debt-Service Coverage Ratio DSM Demand-Side Management DSR Debt Service Requirement DT/DTR Distribution Transformer DTA Domestic Tariff Area DVA Distribution Value Added DVB Delhi Vidyut Board DVP Delivery versus Payment DWT Decentralised Wastewater Treatment EA 03 Electricity Act 2003 EA Energy Audit EC European Commission ECB External Commercial Borrowings ECBC Energy Conservation Building Code EDENOR Empresa Distribuidora y Comercializadora Norte S.A. EDZ Economic Development Zone EIRP Equivalent Isotropically Radiated Power ENARGAS Ente Nacional Regulador del Gas ENRE Ente Nacional Regulador de la Electricidad EoD Event of Default EOU Export Oriented Unit EPC Engineering, Procurement and Construction EPZ: Export Promotion Zone ERC Electricity Regulatory Commission ESC Essential Services Commission ESIS Employees State Insurance Scheme ETDMA Extended Time Division Multiple Access ETOSS Ente Tripartito de Obras y Servicios Sanitarios EU European Union EUA Electricity Utilities Act xxviii | Indian Infrastructure: Evolving Perspectives

EUB Energy and Utilities Board EWS Economically Weaker Section FAA Federal Aviation Administration FAR Floor Area Ratio FCA Fuel Cost Adjustment FCC Federal Communications Commission FDI Foreign Direct Investment FERC Federal Energy Regulatory Commission FIE Foreign Invested Enterprises FIPB Foreign Investment Promotion Board FIs Financial Institutions FiT Feed-in Tariff FM Frequency Modulation FO Furnace Oil FOB Free-on-Board FOR Forum of Regulators FP Future Plot FPPPA Fuel and Power Purchase Price Adjustment FRP Financial Restructuring Plan FSA Fuel Supply Agreement FSA Financial Services Authority FSI Floor Space Index FTCs Foreign Trade Companies FY Fiscal Year GACL Gujarat Ambuja Cements Limited GAIL Gas Authority of India Limited GARR Guaranteed Average Revenue Realisation GBI Generation-based Incentives GBWASP Greater Bangalore Water Supply and Sanitation Project GDP Gross domestic product GIC General Insurance Corporation of India GNCL Gujarat NRE Coke Limited List of Abbreviations | xxix

GNCTD Government of National Capital Territory of Delhi GNIDA Greater Noida Industrial Development Authority GOG Government of Gujarat GOI Government of India GOK Government of Karnataka GOM Government of Maharashtra GQ Golden Quadrilateral GRIDCO Grid Corporation of Orissa GRIHA Green Rating for Integrated Habitat Assessment GSDP Gross State Domestic Product G-Secs Government of India Securities GSM Global System for Mobile Communications (formerly, Groupe Spécial Mobile) GSPC/GSPCL Gujarat State Petroleum Corporation (Limited) GSPL Gujarat State Petronet Limited GTPUDA Gujarat Town Planning and Urban Development Act GU Geographic Unit GUVNL Gujarat Urja Vikas Nigam Limited GWCL Ghana Water Company Limited ha Hectare HBEPL Hanzer Biotech Energies Private Limited HBJ Pipeline Hazira-Bijaipur-Jagdishpur Pipeline HCBS High Capacity Bus System HCCL Hindustan Construction Company Limited HERC Haryana Electricity Regulatory Commission HFCL Himachal Futuristic Communications Limited HIDRONOR Hidroeléctrica Norpatagónica Sociedad Anónima HMRDCL Hassan Mangalore Rail Development Company Limited HNIs High Net-Worth Investors HP Himachal Pradesh HPCL Hindustan Petroleum Corporation Limited HR Human Resources xxx | Indian Infrastructure: Evolving Perspectives

HSD High Speed Diesel HUDCO Housing and Urban Development Corporation HVDS High Voltage Distribution System IAAI International Airports Authority of India IARR Implied Average Revenue Realisation IAT Independent Assessment Team ICICI Industrial Credit and Investment Corporation of India ICRA (formerly) Investment Information and Credit Rating Agency of India Limited ICTSL Indore City Transport Services Limited IDBI Industrial Development Bank of India iDeCK Infrastructure Development Corporation (Karnataka) IDFC Infrastructure Development Finance Company Limited IFCI Industrial Finance Corporation of India IGBC Indian Green Building Council IGIA Indira Gandhi International Airport IGL Indraprastha Gas Limited IIBI Industrial Investment Bank of India IIFCL India Infrastructure Finance Company Limited IL&FS Infrastructure Leasing and Financial Services IOC IndianOil Corporation IPGCL Indraprastha Power Generation Company Limited IPPs Independent Power Producers/Projects IR/IRC Indian Railways (Corporation) IRA Independent Regulatory Agency IRBI Industrial Reconstruction Bank of India IRDA Insurance Regulatory and Development Authority IRR Internal Rate of Return IRRA Indian Rail Regulatory Authority IT Information technology ITU International Telecommunications Union IUP Intended Use Plans List of Abbreviations | xxxi

I-WIN ICICI-West Bengal Infrastructure Development Corporation Limited JCC Japanese Cocktail Crude JICA Japan International Cooperation Agency JNNSM Jawaharlal Nehru National Solar Mission JNNURM Jawaharlal Nehru National Urban Renewal Mission JNPT Jawaharlal Nehru Port Trust JV Joint Venture KERC Karnataka Electricity Regulatory Commission KESCO Kanpur Electricity Supply Company K-G Basin Krishna-Godavari Basin KINFRA Kerala Industrial Infrastructure Development Corporation K-RIDE Karnataka Rail Infrastructure Development Corporation KUIDFC Karnataka Urban Infrastructure Development and Finance Corporation KUWASIP Karnataka Urban Water Supply Improvement Project KUWSBD Karnataka Urban Water Supply and Drainage Board KWSPF Karnataka Water and Sanitation Pooled Fund LEED Leadership in Energy and Environmental Design LIC Life Insurance Corporation of India LNG Liquefied Natural Gas LoI Letter of Intent LPCD Litres per Capita per Day LPR Land Pooling and Readjustment/Reconstitution LPVR Least Present Value of Revenues LSHS Low Sulphur Heavy Stock M&A Mergers and Acquisitions MADP Maximum Alternative Distribution Payment MAGP Maximum Alternative Generation Payment MASTS Mobile Assignment Technical System MATS Monitoring and Tracking System MBR Membrane Bio Reactor mBtu/mmBtu Million British thermal units xxxii | Indian Infrastructure: Evolving Perspectives

MCA Model Concession Agreement MCB Miniature Circuit Breaker mcm Million Cubic Metres MCs Municipal Corporations MDF Municipal Development Fund MEPZ Madras Export Processing Zone MERC Maharashtra Electricity Regulatory Commission MFL Madras Fertilizers Limited MG Metre Gauge MMC Madurai Municipal Corporation MMDR Act Mines and Minerals (Development and Regulation) Act MMRDA Mumbai Metropolitan Regional Development Authority mmscmd Metric Million Standard Cubic Meters per Day Mmt Million Metric Tonnes Mmtpa Million Metric Tonnes per Annum MNRE Ministry of New and Renewable Energy MoCA Ministry of Civil Aviation MoD Ministry of Defence MoP Ministry of Power MoPNG Ministry of Petroleum and Natural Gas MoR Ministry of Railways MoRTH Ministry of Road Transport and Highways MoST Ministry of Surface Transport MoU Memorandum of Understanding MoUD Ministry of Urban Development MP Madhya Pradesh MPE Mumbai-Pune Expressway MPSC Model Production Sharing Contract MRTS Mass Rapid Transit System MS Motor Spirit MSB Minimum Subsidy Bidding MSEB Maharashtra State Electricity Board List of Abbreviations | xxxiii

MSEDCL Maharashtra State Electricity Distribution Company Limited MSRDC Maharashtra State Road Development Corporation MT Million Tons mt Metric Tonnes MTNL Mahanagar Telephone Nigam Limited MYT Multi-year Tariff NABARD National Bank for Agriculture and Rural Development NBFCs Non-Bank Financial Companies NBFI Non-Bank Financial Institution NCR National Capital Region NDPL North Delhi Power Limited NDRC National Development and Reform Commission NELP New Exploration Licensing Policy NEN National Expressway Network NEPZ Noida Export Processing Zone NESCO Northern Electricity Supply Company NFAP National Frequency Allocation Plan NHAI National Highways Authority of India NHDP National Highway Development Project NHPC National Hydroelectric Power Corporation NLD National Long Distance NMMC Navi Mumbai Municipal Corporation NMP National Mineral Policy NMPT New Mangalore Port Trust NPAs Non-performing Assets NPCL Noida Power Company Limited NPV Net Present Value NSDL National Securities Depository Limited NSE National Stock Exchange NS-EW North South-East West NTHS National Trunk Highway System NTPC National Thermal Power Corporation xxxiv | Indian Infrastructure: Evolving Perspectives

NUIF National Urban Infrastructure Fund NUTP National Urban Transport Policy NWP National Water Policy NWRC National Water Resources Council NYSPSC New York State Public Service Commission NZ$ New Zealand dollar O&M Operation and Maintenance OA Open Access OCC Oil Coordination Committee OECD Organisation for Economic Co-operation and Development OECF Overseas Economic Cooperation Fund OERC Orissa Electricity Regulatory Commission OFAPs Operational and Financial Action Plans OFCOM Office of Communications OFGEM Office of Gas Electricity Markets OFWAT Office of Water Services OHPC Orissa Hydro Power Corporation OIL Oil India Limited OMT Operate, Maintain, Transfer ONGC Oil and Natural Gas Corporation of India OP Original Plot OPEC Organization of Petroleum Exporting Countries OPGC Orissa Power Generation Corporation ORR Office of the Rail Regulator OSEB Orissa State Electricity Board OSN Obras Sanitarias de la Nación OUR Office of Utilities Regulation P&O Peninsular and Oriental Steam Navigation Company PBOC People’s Bank of China PBR Private Business Radio PCS Personal Communications Services PDCOR Project Development Company of Rajasthan List of Abbreviations | xxxv

PE Private Equity PFC Power Finance Corporation PFDF Pooled Finance Development Fund PFDS Pooled Finance Development Fund Scheme PFI Private Finance Initiative PfP Payment for Performance PFs Provident Funds PGCIL Power Grid Corporation of India Limited PIDB Punjab Infrastructure Development Board PIL Petronet India Limited PIL Public Interest Litigation PLF Plant Load Factor PMCs Project Management Consultants PMT Panna Mukta Tapti PNGRB Petroleum and Natural Gas Regulatory Board POL Petroleum, Oil and Lubricants PPA Power Purchase Agreement PPCL Pragati Power Corporation Limited PPFCA Power Purchase Fuel Cost Adjustment PPIAF Public-Private Infrastructure Advisory Facility PPP Public-Private Partnership PSA Port of Singapore Authority PSA Power Sale Agreement PSC Production Sharing Contract PSC Public Sector Comparator PSEB Punjab State Electricity Board PSP Private Sector Participation PSU Public Sector Undertaking PTC Power Trading Company PTIM Pre-tax Investment Multiple PwC PricewaterhouseCoopers PWLB Public Works Loan Board xxxvi | Indian Infrastructure: Evolving Perspectives

QoS Quality of Service Parameters QoSS Quality of Supply and Service R&D Research and Development RAPDRP Restructured-Accelerated Power Development and Reforms Programme RCF Rashtriya Chemicals and Fertilizers Limited RE Renewable Energy REBs Regional Electricity Boards REC Renewable Energy Certificate RERC Rajasthan Electricity Regulatory Commission RET Renewable Energy Technology RFP Request for Proposal RFQ Request for Quotation RFQ Request for Qualification RLDCs Regional Load Dispatch Centres RMC Rajkot Municipal Corporation RoC Regulation by Contract ROE Return on Equity ROR Rate of Return ROT Rehabilitate, Operate, Transfer ROW Right of Way RPOs Renewable Purchase Obligations SAA Simultaneous Ascending Auction SAIDI System Average Interruption Duration Index SAIFI System Average Interruption Frequency Index SBR Sequential Batch Reactor SCADA Supervisory Control and Data Acquisition System SCI Shipping Corporation of India SCM Subsidies and Countervailing Measures SEBI Securities and Exchange Board of India SEBs State Electricity Boards SEDs State Electricity Departments List of Abbreviations | xxxvii

SEEG Société d'Exploitation des Eaux de Guinée SEEPZ Santa Cruz Electronic Export Processing Zone SEGBA Servicios Eléctricos del Gran Buenos Aires SERC State Electricity Regulatory Commission SEZ Special Economic Zone SFC State Finance Commission SFCD State Finance Commission Devolution SGAs Specialised Government Agencies SGI Solicitor General of India SHP Small Hydro Power SKO Superior Kerosene Oil SLAUs Special Land Acquisition Units SOE State-owned Enterprise SONEG Société Nationalé des Eaux de Guinée SOUTHCO Southern Electricity Supply Company SPD Solar Power Developer SPFE State Pooled Finance Entity SPV Special Purpose Vehicle/Company SPV Solar Photovoltaic SSI Small Scale Industry SSUs Standard Spectrum Units STD Subscriber Trunk Dialing STP Sewage Treatment Plant STU Standard Trading Unit STW Sewage Treated Water SWM Solid Waste Management T&D Transmission and Distribution TA Technical Assistance TACID Tamil Nadu Corporation for Industrial Infrastructure Development TAMP Tariff Authority for Major Ports TAT Tourism Authority of Thailand TBA To Be Announced xxxviii | Indian Infrastructure: Evolving Perspectives

tcf Trillion Cubic Feet tcm Thousand Cubic Metres TDRs Transfer of Development Rights TDSAT Telecom Disputes Settlement and Appellate Tribunal TD-SCDMA Time Division Synchronous Code Division Multiple Access TEA Tirupur Exporters Association TERI The Energy and Resources Institute TEU Twenty-foot Equivalent Unit TFC Thirteenth Finance Commission TIMS Transformer Information Management System TN Tamil Nadu TNEB Tamil Nadu Electricity Board TNERC Tamil Nadu Electricity Regulatory Commission TNUDF Tamil Nadu Urban Development Fund TNUIFSL Tamil Nadu Urban Infrastructure Financial Services Ltd TNWSPF Tamil Nadu Water and Sanitation Pooled Fund TOU Time of Use TPAs Third Party Administrators TPC Total Project Cost TPO Town Planning Officer TRAI Telecom Regulatory Authority of India Transco Transmission Company TSS Total Suspended Solids TTRO Tertiary Treatment and Reverse Osmosis Plant TVEs Township and Village Enterprises UASL Unified Access Services License UDF User Development Fee UI Unscheduled Interchange ULBs Urban Local Bodies UMPPs Ultra Mega Power Projects UMTS Universal Mobile Telecom Service UPERC Uttar Pradesh Electricity Regulatory Commission List of Abbreviations | xxxix

UPPCL Uttar Pradesh Power Corporation Ltd UPRVUNL Uttar Pradesh Rajya Vidyut Utpadan Nigam Limited UPSEB Uttar Pradesh State Electricity Board US$ United States Dollar USAID United States Agency for International Development USF Universal Service Fund USFA Universal Service Fund Administrator VAT Value Added Tax VfM Value for Money VGF Viability Gap Funding VPT Village Public Telephone VSNL Videsh Sanchaar Nigam Limited WB West Bengal WBSEDCL West Bengal State Electricity Distribution Company Limited WESCO Western Electricity Supply Company WLL Wireless Local Loop WPC Wireless Planning and Coordination Wing WPI Wholesale Price Index WS&S Water Storage and Supply WSA Water Service Agency WSP Waste Stabilisation Pond WSPF Water and Sanitation Pooled Fund WSS Water Supply and Sewerage WTO World Trade Organization WUA Water Users Association WWD Water Works Department

POWER SECTOR REFORM: Policy Decisions in Distribution 1 May 1998

1. INTRODUCTION The investment needed in the power sector in the next five years is estimated to be in excess of Rs 200,000 crore (US$50 billion), largely in new generating plants. Over a longer term, 75,000 MW of new generating capacity is being contemplated in the form of mega power plants. Since each MW of generating capacity installed by an independent power producer requires an estimated commitment of about Rs 1.5 crore per annum (6000 MWhrs, at Rs 2.5 per kWh), the payment liability for this generating capacity would amount to Rs 112,500 crore (US$28 billion) per year. Considering that the total revenue of all electricity boards in the country is only around Rs 40,000 crore per annum, the additional liabilities are unmanageable under the current regime of revenue mobilisation. The government has provided many incentives to independent power producers (IPPs). These include a guaranteed rate of return of 16 per cent, counter guarantees from the central and state governments, escrow accounts for assured payments, etc. As a result, over 250 MoUs have been signed. However, the success rate from MoU to financial closure has been low. Only a few projects have been implemented in the past seven years. The primary hurdle in implementation is now recognised as the poor financial health of the electricity boards, who are currently the sole purchasers of electricity from the IPPs. In turn, this is attributable to the poor state of distribution, characterised by low quality of service, rampant theft, high subsidies and poor revenue recovery. Therefore, in order to make these planned MWs a reality, a substantially increased revenue generation from the distribution system assumes the highest priority. Additional megawatts can only be generated if more megawatt hours are delivered 4 | Indian Infrastructure: Evolving Perspectives and paid for, that is, either the number of paying consumers has to increase or the consumption of the currently paying consumers must rise. An efficient revenue generating system would also obviate the need for sovereign guarantees, escrow accounts and take-or-pay contracts, and permit the development of the sector on commercial lines. The consequences of ignoring additional revenue generation can lead eventually to major national problems, similar to that being currently experienced in Indonesia (see Box 1.1).

Box 1.1: Take-or-pay in Indonesia

In a situation very similar to that of India, the paucity of investible funds led Indonesia to invite the private sector to construct power plants. With optimistic growth forecasts, the National Power Corporation, PLN, set itself ambitious targets and signed 26 “take- or-pay” power purchase agreements with private IPPs. In most contracts, the price, which was linked to the US dollar, ranged from 5.7 to 8.5 US cents per kWh. The recent Southeast Asian currency crisis saw the Indonesian rupiah depreciate by over 70 per cent and the GDP growth rate turn negative. The growth in electricity demand slumped to zero and the dollar equivalent of the tariff fell sharply. At the prevailing exchange rate, PLN’s tariff rates are less than 2.0 US cents per kWh. Its liability for the next 15 years is estimated at US$43 billion, which it has no means to honour. PLN’s take-or-pay contracts are now a major national problem.

2. CHARACTERISTICS OF SEBs State Electricity Boards (SEBs) in India are beset with numerous problems. They are characterised by huge financial losses, high transmission and distribution losses comprising technical and non-technical components, unsatisfactory quality and low reliability of supply, undelivered and misdelivered bills, poor collection and unresponsiveness to consumers’ requirements. Financial losses: The annual commercial losses of the electricity boards in the country have increased from Rs 1565 crore in 1985–86 to nearly Rs 10,000 crore in 1996–97 (Planning Commission 1997). The average revenue collection per unit sold is 158 paise, against the average cost of 208 paise. A desirable scenario where the electricity boards can undertake a sustainable expansion programme would require surplus generation of about Rs 10,000 crore,1 which in effect means that the current shortfall in revenue is close to Rs 20,000 crore per year. Transmission and distribution losses: The transmission and distribution losses for the country were reported to be 21.2 per cent in 1994–95. However, it is widely believed, and substantiated by spot surveys, that the level of losses is considerably higher—in the range of 40 to 50 per cent—in many electricity boards, with a large Power Sector Reform | 5 degree of regional variations. It is alleged that lower losses are shown by falsely attributing higher energy consumption to unmetered consumers. In fact, the so- called losses are more likely to be due to uncalibrated and unsatisfactory meters, theft of energy by unauthorised connections, tampering of meters, and collusion between consumers and board staff to reduce the billing. Poor quality of service: The quality of service in electricity supply can be measured by the frequency, voltage and continuity of supply. On all these scores, the Indian distribution system rates extremely low. Frequency variations of ±2 per cent are quite common on Indian grids, whereas internationally even a 0.2 per cent variation is considered unacceptable. Voltage variations here regularly exceed ±10 per cent, whereas international norms prescribe ±5 per cent as the maximum permissible range. Consequently, agricultural and industrial electrical equipment and domestic electrical appliances suffer from frequent breakdowns. Continuity of service, that is, receiving uninterrupted power supply, is quite rare in India. Restoration of service after an unplanned or forced interruption is generally greatly delayed. The consumer response to the situation is the proliferation of “gensets”, involving significant investment in a relatively inefficient mode of power generation using scarce liquid fuels.

3. REASONS FOR THE PRESENT STATE OF SEBs The factors that have contributed to the present state of electricity boards can be broadly classified under two heads—management failure and inappropriate policies on the part of state governments. The potential for corruption in this highly capital- intensive sector has only added to its woe. Lack of management attention: The management of distribution facilities are not up to desired standards in terms of investment, manpower allocation and technological innovation. The share of transmission and distribution in the plan outlays has been only 26 to 28 per cent since the second five-year plan, against a desirable level of about 40 to 50 per cent. Generation projects requiring large investments, higher technological content, well-defined objectives and greater potential for job satisfaction and rewards have traditionally attracted more resources. Inappropriate emphasis: In distribution, the emphasis has been on extensive development rather than intensive development. Since targets focus on quantity rather than quality, the tendency has been to spread a low-cost and low-standard distribution network to as many households, villages and agricultural pump sets as possible, within the constraints of the budget. After construction, relatively little attention is devoted to preventive maintenance, as most distribution staff are engaged in new constructions. 6 | Indian Infrastructure: Evolving Perspectives

Metering, billing and collection practices: The present collection system causes large revenue losses. It is characterised by lack of proper customer information and reliance on antiquated billing and accounting systems that do not provide timely information. Existing labour policies provide no incentive for staff to produce results. Added to this is the persistent inability to take disciplinary action against defaulting consumers, many of whom are state and local government organisations or influential industrial and commercial consumers. This has led to estimated revenue arrears, receivable by electricity boards, of over Rs 13,000 crore in 1995–96 amounting to 37.5 per cent of the total annual revenues (Planning Commission 1997). Government subsidy policies: The government’s policy of supplying electricity to favoured consumers at lower costs is estimated to have resulted in a revenue loss of Rs 19,228 crore in 1996–97 (Planning Commission 1997). State governments are required to compensate the electricity boards for such losses. To the extent that this is done, it has a debilitating effect on state finances. Often though, the state government does not make the necessary budgetary transfers, resulting in paucity of funds for operation, maintenance and capital work. Apart from the revenue loss, a worse long-term consequence of this policy is the loss of incentive for electricity boards to meter such consumers, affecting the very basis of the revenue generation infrastructure. Cross-subsidisation: Electricity boards try to recover a part of the above revenue loss by charging higher tariffs to other sectors, such as industrial and commercial consumers. However, over-reliance on this mechanism results in unduly high tariffs, which affects industrial competitiveness and also drives industries to set up their own generation, through captive power plants.

4. SOLUTIONS To reiterate, in order to add additional megawatts to the sector, it is imperative that additional revenue be generated from the distribution system. Currently, the SEBs, who run the system, are characterised by massive financial losses, large transmission and distribution losses and poor quality of supply. They are not in a position to increase the resources mobilised from electricity consumers. In order to accomplish this task, it is thus necessary to remedy the fundamental causes of SEB failure, that is, ineffective management and inappropriate policies, as described above. The recommendations that follow seek to do just that. Separate the distribution system: The distribution system must be separated from the generation and transmission systems, and formed into commercial companies. This separation is essential to insulate the revenue-generating portion of the power Power Sector Reform | 7 sector from external pressure. It would also lead to more attention being paid to distribution issues, through a focused management.2 Privatise the management: The only way to incentivise the separated distribution system to mobilise additional resources is to ensure that it bears responsibility for its losses. The most credible manner of enforcing such a hard budget constraint is to privatise the management. This can be done in a variety of ways, such as management contracts, leases, joint ventures and outright disinvestment. The method is not as important as the principles on which such management control is transferred. The two crucial aspects are: • Transfer, whether by management contract or lease or any other method, must be for a long period (for example, 30 years), enough to make new investment remunerative. • The private management must have complete autonomy in all commercial decisions, subject to oversight due to the continuing monopoly status. Create an independent regulatory authority: Once management is transferred to private hands, there is a strong need for independent state-level regulatory commissions to monitor the private managements’ decisions in matters of tariff, investments, etc. They should be empowered to approve tariffs that will permit the licensees to earn a reasonable return on their investment, if they are operating efficiently. In addition, the commissions will provide a useful and necessary forum to resolve disputes between the government and the new private managers. Strategies to attract private investment: In the current environment, it may not be easy to induce private firms to take over the management of distribution companies, due to several perceived hurdles. These include the presence of mandated customers in distribution zones; a relatively low rate of return; absence of long-term incentives; unreliable power supply from the electricity boards; difficulty in carrying out asset valuation; and continued interference from the government and electricity board bureaucracies, leading to difficulty in dealing with the existing staff. In order to accelerate the process of improvement in distribution management and revenue mobilisation through private sector participation, these problems need to be addressed quickly. One way of doing so is outlined below. Offer urban franchise areas: The object of privatising the management of distribution is to generate efficiency gains leading to additional resource mobilisation. The franchise areas should be chosen to maximise these efficiency gains. For this, the area must have two properties. It must have a customer base of sufficient size, with the capability to pay, and it must have physical infrastructure in reasonable condition. It is, therefore, recommended that private investors should initially be offered dense urban areas, and other areas that satisfy the above properties, like industrial estates, for distribution.3 8 | Indian Infrastructure: Evolving Perspectives

Since urban demand is over 50 per cent of the total demand in the country, the ultimate scope of the proposed strategy will be large. In these areas, licensees would not be burdened with social obligations, nor would there be subsidisation of any consumers in the licence area. In view of the highly successful operation of licensees in urban centres, like Ahmedabad, Surat, Mumbai and Kolkata, it would be much easier to attract private investment in such centres. In order to provide sufficient incentive for licensees to meet inevitable urban expansion, a process must be simultaneously put in place for transferring contiguous urban agglomerations and other new concentrations of consumers to private management. Give full management autonomy: Since the thrust of this strategy is to increase resource mobilisation by reaping efficiency gains, it is essential that managements have complete autonomy over commercial decisions, including those relating to employment. Without such autonomy, managements cannot be held fully responsible, and this would dilute the incentives for efficient operation. Since the urban areas employ only a fraction of the total labour force of the SEBs, the problems of staff transfer need not form a bottleneck, and if required, the electricity board should retain the staff in the interim. Provide a reasonable rate of return: The permissible rate of return to private investors in distribution as per the Indian Electricity Act is set at RBI Bank Rate plus 5 per cent, which currently works out to 14 per cent. This is lower than the 16 per cent permitted to IPPs at 68.5 per cent Plant Load Factor (PLF). With an incentive bonus of 0.7 per cent for each per cent increase in PLF, IPPs can expect to earn returns of 22 to 23 per cent. Operation and maintenance of distribution facilities in India involve greater effort and greater degree of risk than what is involved in setting up generating plants. This skews the investment incentives towards generation. There is, therefore, a need to free the private investors in distribution from the provisions of the Electricity Act and permit the state regulatory commissions to determine tariffs to recover higher rates of return when warranted. Eventually, this could lead to a system where prices are regulated, instead of rates of return, and bulk consumers can strike customised deals with distributors and generators. Enable wheeling: In order for the distribution licensees to supply quality power, they need to be free from the vagaries of supply from unreliable generators. The simplest way would be to allow the distribution licensees access to efficient and reliable generators through wheeling, that is, by permitting the transport of power from a generating station in one area to a consuming centre in another area using the transmission network of the grid operator (SEB/Transmission Company/Power Grid Corporation). Permitting wheeling will also enable cost-effective captive generation plants to sell their surplus power to other consumers and expand their capacities to serve consumers in contiguous areas. This will add experienced Power Sector Reform | 9 producers to the pool of power generators.4 In the short term, however, lack of transmission capacity may prevent this from happening. This should not be a reason to hold back on the urban distribution strategy. Power trading corporation: A power trading corporation like the MEM in Argentina (see box below), which is a virtual spot market for power, could facilitate the process of wheeling considerably. By enabling efficient generators to supply electricity to creditworthy distributors at competitive rates, by enforcing commercial discipline on the distribution licensees and by excluding them from the market for non- payment, it creates a powerful force to generate additional revenues, thus bringing in extra megawatts. It is understood that there is a proposal to create Power Trading Companies (PTCs)5 for handling power wheeling. Furthermore, these are to act as long-term buyers of power and to sell it to existing HT consumers. However, unlike the example above, this is similar to escrowing the HT consumers for the IPPs that sell power to the PTC. This approach does not increase the total resource mobilisation from the distribution system, since the identified HT consumers are already paying customers. As such, it does not generate additional revenue from the distribution system, which is imperative in order to add additional megawatts.

Box 1.2: Power sector reform in Argentina

In the late 1980s, the Government of Argentina (GOA) reformed its power sector to achieve efficient pricing and sufficient investment levels. The reform process comprised unbundling, privatisation and regulation. The reform strategy simulated competition in natural monopoly segments, such as distribution and transmission, through regulation, award of concessions and the creation of a wholesale market for electricity. In distribution, the federal government broke up the Buenos Aires distribution area, which accounts for almost 60 per cent of Argentina’s electricity consumption, and awarded three exclusive concessions. To facilitate competition, a spot wholesale market for electricity, called the MEM, was created. Large consumers as well as distribution entities are free to negotiate power contracts directly with generators or fulfil their needs through the MEM. A National Regulatory Entity for Electricity (ENRE) was set up to ensure fair access to transmission and distribution networks and oversee all facets of the sector, including service quality. ENRE also sets maximum tariffs for transmission and distribution services under a price cap system (RPI-X), with the cap reset every five to eight years. Since the advent of the reform process in 1992, EDESUR, one of the major suppliers to the Buenos Aires area, has decreased its energy losses from 21 per cent to 12 per cent and reduced its outages from 39 to 6 hours per year. During the same period, the spot 10 | Indian Infrastructure: Evolving Perspectives

price of electricity on the MEM has decreased from 4.2 to 2.2 US cents per kWh, and thermal availability has increased from 48 per cent to 70 per cent. In transmission, forced outages have declined from 1000 to 300 hours. The privatised distribution companies have also made substantial investments in the sector. For example, the consortium owning EDENOR invested US$380 million till 1995, and is expected to invest an additional US$500 million by 2000. Private sector investment in the entire electricity sector is projected to reach US$7 billion by the year 2001.

Revenue implications for remaining areas: The strategy being proposed can be accused of “cherry picking” the lucrative areas for private management. The question naturally arises as to how the remaining distribution will be handled by the SEBs, in the absence of the “profitable” urban areas. At the outset, it is necessary to realise that many urban areas—for example, Lucknow and Bhubaneshwar— are not presently profitable. In fact, the suggested strategy focuses on “dense urban areas” because they have significant potential for efficiency gains. Their transfer will reduce losses to the boards and augment their revenues. In addition, bulk supply to urban licensees, wheeling charges and additional revenues in terms of taxes and excise duty on the sale of electricity, as being currently practised in cities like Ahmedabad and Mumbai, will also augment the SEB/government resources. These additional resources arising out of the expected efficiency gains will result in more revenue, not less, which can be used to support distribution in the remaining areas. Decentralised distribution and generation: Over time, other approaches will have to be explored to solve the problem of developing extensive distribution networks in far-flung areas and consumers with limited capability to pay. A model worth consideration is the one being thought about by West Bengal, that has created the Rural Energy Development Corporation, that will be responsible for the distribution of power under 11 kV to all rural areas (Gupta 1998). Decentralised distribution coupled with decentralised generation may be an answer to this problem. This approach could also use non-conventional energy sources, which are environmentally beneficial. Even if the generation cost is higher, this may be compensated by direct cost-savings on building long-distance transmission linkages and lower transmission losses.

5. NEXT STEPS AND CONCLUSION This section lays out the next steps to be taken in order to implement the identified strategy for improving resource mobilisation in the distribution systems of the Indian power sector. The first two are absolutely necessary, while the third helps to substantially increase efficiency. Power Sector Reform | 11

Selection of centres: The first priority is to select twenty or so centres to start the process. To begin with, areas contiguous with urban zones currently under private license, like Ahmedabad and Kolkata, can be offered. Urban centres in states that have completed their unbundling exercise are the next natural target. Their size would vary from region to region, but indicative examples are Pune and Nagpur in Maharashtra, Vadodara and Rajkot in Gujarat, and Hyderabad and Visakhapatnam in Andhra Pradesh. Industrial estates with sufficient power consumption can also be included. Mode of transfer to private management: The actual mode can be determined on a case-to-case basis (as noted on page 7: ‘Privatise the management’) using a transparent process like international competitive bidding. Transfer of assets, when necessary, should take place on the basis of the revenue stream it can be expected to generate. In all cases, management autonomy must be guaranteed. Facilitating wheeling: The Indian Electricity Act will need to be amended to permit wheeling transactions, and electricity boards and the Power Grid Corporation will have to be obligated to permit the use of their network for a charge to be determined on a case-by-case basis by the central or state regulatory commissions. This, however, is not a precondition for successful transfer to private managements and can be done in tandem with the above processes. Conclusion: In the above paragraphs, we attempted a quick review of the characteristics of the distribution system as it exists today, and the causes that have led to this situation. More to the point, we have outlined a strategy that is based upon increasing the size of the financial cake produced by the distribution system, rather than quarrelling over who gets the larger slice. The strategy relies upon transferring urban areas and other such dense concentrations of consumers to private management, and implementing appropriate arrangements for the remaining distribution areas. We feel this can be quickly implemented, and that this can revitalise the Indian power sector by enhancing distribution efficiency, thereby generating additional revenues to bring in extra megawatts, without sovereign guarantees and take-or-pay contracts. 12 | Indian Infrastructure: Evolving Perspectives

REFERENCES 1. “Annual Report on the Working of State Electricity Boards and Electricity Department” – by Planning Commission (November 1997). 2. “A Concept Note on Power Trading Corporation” – by T. L. Sankar (undated). 3. “Energising Power Sector: Light at the End of the Tunnel” – by R.K. Pachauri, Times of India (1 May 1998). 4. “PFC’s Initiative on Power Sector Reform” – by Power Finance Corporation (undated). 5. “Privatisation of Distribution in India: Issues, Options and Lessons from Other Countries – by International Resource Group Ltd (8 December 1997). 6. “Report of the Committee on Private Sector Participation in Power Distribution” – by S.J. Coelho (March 1998). 7. “Separate Power Corporation for Rural Sector in Bengal” – by Gautam Gupta, Economic Times (29 April 1998).

NOTES 1. This is based on a rough calculation. An annual increase of 8000 MW in capacity, at Rs 4 crore a megawatt, requires Rs 32,000 crore of investment. A 70:30 debt equity ratio would imply that SEBs should be able to commit around Rs 10,000 crore. 2. Studies conducted for various SEBs over the past few years have led to nearly similar reform structures. They all recommend the separation of the distribution system from the generation and transmission systems, and the establishment of commissions to set tariffs, license activities and perform other regulatory functions (Power Finance Corporation, “PFC’s Initiative on Power Sector Reform”; Coelho 1998). 3. The current approach is to create zonal distribution companies. The franchise areas include a mix of paying consumers and mandated customers, who are supplied electricity at subsidised or free rates. Each zone has a mix of urban and rural areas, with far-flung distribution networks, high cost of supply, and agricultural and irrigation loads from which recovery has traditionally been weak. The currently defined franchise areas thus add additional costs and social responsibilities that private investors are ill equipped to shoulder. Such privatisation needs transparent and guaranteed mechanisms for flow of subsidy from the state government to the licensee, which complicates the process and detracts from commercial orientation. 4. Some estimates indicate that there is about 15,000 MW of captive generating capacity which will be “decaptivated” by the wheeling provision (Sankar, “A Concept Note on Power Trading Corporation”). 5. Based on the “Annual Report on the Working of State Electricity Boards and Electricity Department” by Planning Commission (1997) and newspaper reports. Power Sector in India | 13

POWER SECTOR IN INDIA: A Summary Description 2 November 1998

1. INTRODUCTION Since Independence, in 1947, the Indian power sector has been mainly dominated by the public sector. As of 31 March 1997, about 95 per cent of the total installed capacity of 85,919 MW was owned by the central (32 per cent) and state (63 per cent) sector undertakings. In terms of fuel-type mix, thermal, hydro and nuclear capacities account for 72 per cent, 25 per cent and 3 per cent respectively. During the year 1996–97, the total generation in the country was about 394 billion units. Legislation: Under the Indian constitution, electricity is on the concurrent list— meaning that it comes under the purview of both the central (federal) and state (provincial) governments. The Indian Electricity (IE) Act 1910 and the Electricity Supply (ES) Act 1948 are the key Central Acts, under the broad ambit of which individual states have enacted their own laws. The IE Act 1910, which was enacted while India was still under the British rule, defines the obligations, powers and responsibilities of licensees in the electricity industry. Few such licensees exist today. The ES Act of 1948, passed soon after India achieved independence, set out the framework for establishing the Central Electricity Authority (CEA), the State Electricity Boards (SEBs) and generating companies. The first SEBs were established in the early 1950s, soon after the legislation had been enacted. The industrial policy resolution adopted in 1956 provided that generation and transmission of electricity would be reserved exclusively for the public sector. Accordingly, the law stated that no person or firm shall engage in the business of supplying energy to the public except with the previous sanction of the state government. Although the central government is apparently competent to give permission to generating companies 14 | Indian Infrastructure: Evolving Perspectives to sell energy, it can only do so for the sales aimed at SEBs or State Electricity Departments (SEDs).

2. INSTITUTIONAL STRUCTURE Although policy, generation and transmission activities are undertaken by both the central and state governments, distribution and supply of electricity to the final consumer and the associated tariff setting are exclusively under the purview of the state governments. A pictorial representation of the institutional structure of the Indian power sector is provided in Figure 2.1.

Government of India

National Ministry Department of State Development of Power Atomic Energy Governments Council

Planning Nuclear Department Commission Power Corporation of Energy

Central Electricity Authority (CEA)

Generating utilities Transmission utility Regional National Thermal Power Grid Electricity Boards State Electricity Private sector Power Corporation Corporation of (REBs) Boards (SEBs)/ licensees (NTPC) and others India Ltd (PGCIL) Departments (SEDs)

Figure 2.1: Institutional structure of the Indian power sector

Central level: At the federal level, the Ministry of Power and Non-Conventional Energy Sources is responsible for power policy as well as the generation and transmission capacity in the central sector undertakings (CSUs). Policy: In policy setting, the ministry is assisted by the Central Electricity Authority (CEA), which was formed in 1948 as the industry’s regulatory body.1 In addition to contributing to national power policy, CEA also undertakes other coordinating activities such as conducting appraisals of projects, granting clearances and issuing guidelines for setting tariffs. Generation: The generating capacity in the central sector accounts for about 30 per cent of the country’s total generating capacity. This capacity comes mainly from the National Thermal Power Corporation (NTPC) and the National Hydroelectric Power Corporation (NHPC). SEBs draw their share of power from these CSUs, and each CSU is entitled to a tariff that is an aggregate of three components—fixed cost Power Sector in India | 15

(capacity charge), variable cost (energy charge) and incentive. With a capacity of 16,795 MW, NTPC is the largest power generating utility in the country, contributing about 20 per cent of the total installed capacity in the country. In terms of Plant Load Factor (PLF), NTPC achieved a PLF of 77 per cent in 1996–97 as against the national average of 64 per cent. Transmission: In recognition of the need for greater integration of the transmission system, five regional electricity boards (REBs) were established in 1964, covering the northern, western, southern, eastern and north-eastern regions. These REBs co-ordinate system operations within each regional grid through regional load dispatch centres (RLDCs). In 1989, the Power Grid Corporation of India Limited (PGCIL) was established for managing the national transmission and dispatching system. As of March 1997, PGCIL was operating about 27,853 circuit kilometres (CKMs) of transmission lines, distributed over 54 substations with 23,331 MVA of transformation capacity. PGCIL’s performance in terms of overall average availability of its transmission lines—at above 98 per cent—is comparable to international standards. About 30 per cent of the country’s total installed generating capacity (81,500 MW) is connected to its network. Currently, PGCIL is in the process of linking regional grids, through a series of high voltage direct current (HVDC) interconnections. On the successful completion of this process, PGCIL’s network would have the capacity to transfer 1000 to 1500 MW of power from any one region to another. State level: State governments are exclusively responsible for the supply of power to final consumers located in the state, along with associated tariffs. In the states, the policy for the power sector is looked after by the departments of energy of the respective state governments and the generation, transmission and distribution facilities are owned and managed through the respective SEBs/SEDs. Although the legislation provides for financial autonomy to the SEBs, most of them work under the administrative control of the respective state governments and state ministries and are under the technical control of CEA, REB and RLDC. Much of the expansion of the electricity industry in India since the early 1950s has been carried out by the SEBs, which account for 63 per cent of the country’s generating capacity. Two-thirds of this capacity comes mainly from eight states— Andhra Pradesh, Gujarat, Karnataka, Madhya Pradesh, Maharashtra, Punjab and Tamil Nadu. Following their formation in 1948, SEBs established independent transmission networks to supply power within each state. However, most of the states have a very low HT:LT ratio and, hence, suffer high technical losses and non- technical losses (theft of electricity). Most of the states charge agricultural and domestic consumers tariffs that are far below the cost of generating electricity and seek to make up for this by charging higher tariffs to industrial consumers. 16 | Indian Infrastructure: Evolving Perspectives

Private sector: Although private sector utilities and local authorities provided around 80 per cent of the public supply in India prior to Independence, the state subsequently took over most licensees when their licences expired, and after 1956, no new licences were granted. Five private utilities remain in existence: Bombay Suburban Electric Supply Company, Tata Electric Companies, Ahmedabad Electricity Company, Surat Electric Company and CESC Ltd (formerly Calcutta Electric Supply Corporation). These five have a generating capacity of around 2800 MW. In addition, there exists about 12,000 MW of captive power capacity, which largely comprise coal-based plants (about 6000 MW) and diesel generating sets (about 4000 PW). Coal-based plants are owned mainly by power intensive industries like aluminium, steel, fertilisers, cement and petrochemicals. Diesel-based plants are used as a standby source of power in the event of non-availability of grid power.

3. PERFORMANCE Over the past five years, the physical performance of both the central and state sector utilities has either improved or held steady at the previous levels. On the other hand, the financial performance of the sector has deteriorated considerably, and the sector has failed to achieve the targeted additions to its capacity. It is important to note that the following paragraphs on performances are based on aggregates at the national level and there could be significant performance variations across different states. Physical performance: During 1992–97, the overall deficit in electricity supply (energy) increased from 8.3 per cent to 11.5 per cent, whereas peak-shortage has shown a marginal improvement from 20.5 per cent to 18 per cent. The sale of electricity increased from 213 billion kWh in 1992–93 to 275 billion kWh in 1996– 97, representing an annual growth rate of 6.6 per cent. Inadequate transmission infrastructure is regarded as one of the main reasons for the shortage of peaking power. In the absence of adequate transmission capacity to transfer power across the regions, there is limited scope for exploiting the differences in the time of occurrence of peak demand in various regions. Plant Load Factor (PLF) and plant availability: The PLF of thermal power plants increased from 45 per cent in 1980–81 to 64.4 per cent in 1996–97. However, it is significantly lower than the international standard of about 70–75 per cent. The lower PLF is mainly attributed to the lower availability of coal-based power plants. The average availability of Indian power plants is about 75–80 per cent. The PLF of power plants owned by the centre, the SEBs and private companies show significant variations. The central and private sector plants have consistently shown a higher PLF as compared to the plants owned and managed by the SEBs. Also, on an average, private and central sector plants have higher plant availability as compared to the Power Sector in India | 17

SEB-owned plants. The central sector plants supply power to more than one state in a region, which protects them from the reduction in demand from one of the states and enables them to maintain a high output level. Transmission and distribution losses: Of the electricity generated, about 21–22 per cent is lost in the process of transmission and distribution. A significant portion of these losses is attributable to inadequate metering and theft of electricity. The difference in the amount of electricity supplied and the amount actually metered is usually reported as T&D losses. In most of the states, the electricity supplied to agriculture is unmetered and is charged on the basis of the connected load. As the amount of electricity consumed by this sector is usually overestimated, actual T&D losses may be significantly higher than those reported. In addition, there is large scale theft of electricity through unauthorised connections. The main reason for high T&D losses in the Indian power system is the transmission and distribution of a large amount of power at low voltages—less than 11 kV or Low Tension (LT). Capacity addition: During the five-year plan period that ended in March 1997, only 16,742 MW of capacity could be added as against the target of 30,858 MW. Financial performance: The commercial losses of SEBs, without subsidy, have increased from Rs 45.6 billion (US$1.1 billion) in 1992–93 to Rs 98.0 billion (US$2.3 billion) in 1996–97, and these are projected to increase to Rs 101.65 billion (US$2.4 billion) in 1997–98. The Rate of Return on Capital, after taking into account the subsidy from the government, has deteriorated from (-)7.6 per cent in 1992–93 to (-)11.6 per cent in 1997–98. Without subsidy, the situation would have been far worse at (-)16.4 per cent in the year 1997–98. Subsidies and shortfall in tariffs: During 1991–97, the average unit cost increased from Rs 1.17 to Rs 2.08, whereas the tariff per unit increased from Rs 0.89 to Rs 1.58. Thus, the direct shortfall in revenue remained at about 24 per cent, which indicates that only three-fourths of the costs are being recovered through tariffs. Such shortfall is much more significant in the case of agricultural and domestic consumer categories. Due to political reasons, the tariffs for agricultural and domestic consumers have been very low in most of the states, at an average rate of Rs 0.22 per kWh and Rs 0.89 per kWh respectively, in 1995–96. In order to subsidise agricultural and domestic consumers, commercial and industrial consumers were charged much higher tariffs, at an average rate of Rs 2.24 and Rs 2.35 per kWh respectively, in 1995–96. According to an estimate, the effective subsidy for these sectors is expected to increase from Rs 159.5 billion (US$3.8 billion) in 1995–96 to Rs 192.3 billion (US$4.6 billion) in 1996–97 and Rs 217.9 billion (US$5.2 billion) in the following year. Cross-subsidisation of agricultural and domestic sectors is becoming an increasingly difficult proposition because the share of agriculture in the total 18 | Indian Infrastructure: Evolving Perspectives consumption increased from 17.6 per cent in 1980–81 to about 30 per cent in 1993–94, even as that of industry declined from 58.4 per cent to about 39.6 per cent in the same period. Revenue arrears and outstanding dues: The outstanding dues to be paid by SEBs to their central sector suppliers of electricity, equipment, finance and energy inputs are reported to be over Rs 122.5 billion (US$2.92 billion) as on 31 March 1997. On the other hand, as on March 1996, SEBs were to receive from their customers an amount of Rs 116.8 billion (US$2.78 billion), which accounts for 30–33 per cent of their annual sales turnover.

4. REFORM EFFORTS AND PRIVATISATION Since the economic liberalisation in 1992, both the central and state governments have sought to increase private sector participation in the provision of infrastructure services, including power. Initially, the focus of the reform was on encouraging private participation in generation. In response, nearly 127 expressions have been registered, aggregating Rs 2500 billion (US$59.5 billion) investment for setting up 69,000 MW. However, how many of these expressions would turn into reality remains a moot question, considering that only a handful of private projects have come into being over the past six years and most of the other projects are struggling to achieve financial closure. Private investors and lenders are wary of supporting power projects that have to rely exclusively on financially weak SEBs for evacuating their power. Hence, the state governments are gradually shifting the focus of their efforts to areas such as SEB-reform and regulation, whereas the central government has been focusing on the areas of regulation, transmission-privatisation, setting up of multi-state mega projects and power trading companies. Salient points of the reform and privatisation efforts undertaken by the central and state governments are summarised below. Private participation in generation: The reform policy introduced in 1991 allowed the private sector to set up companies to act as licensees, generating and distributing power, or simply as generators. Up to 100 per cent foreign equity participation was permitted, with a maximum debt to equity ratio of 4:1. The return on foreign equity was protected in foreign currency, and a number of tax concessions were also made. In order to determine the tariff for the purchase of power, a notification, which laid down the guidelines for a two-part tariff, was issued. The main features of the notification were: • The tariff would constitute two parts—a fixed part comprising return on equity (RoE), interest on loan capital, depreciation, operations and maintenance costs (O&M); and a variable part comprising fuel costs. Power Sector in India | 19

• A maximum of 16 per cent RoE was allowed to be included in the tariff (this was protected against fluctuations in the exchange rate). • Fixed cost could be recovered at a PLF of 68.5 per cent (equivalent to 6000 hours of operations) in the case of thermal plants, and at an availability factor of 90 per cent in the case of hydroelectric plants. The normative PLF for thermal plants was revised to 75 per cent in February 1997. • As an incentive, a maximum of 0.7 per cent additional RoE could be given for every 1 per cent increase in PLF or availability. Clearances and approvals: In order to clear a thermal project, the CEA requires an approval from the state government and the electricity board concerned, clearance of water availability and fuel linkage approval from the petroleum and natural gas ministry or from the coal ministry. Projects also require environmental clearance and chimney height clearance from the National Airports Authority of India. Fuel supply and transportation: Fuel supply has become a contentious issue as the supply of coal, naphtha and natural gas is controlled by public sector units (PSUs). The PSUs are not willing to enter into agreements for assured supply, mainly due to lack of experience with such contracts, particularly with regard to the evaluation and quantification of the associated risk and premium. Further, the Indian Railways, the principal carrier of fuel, is unwilling to assure uninterrupted supply. The IPPs insist that the penalty for supply interruptions should cover the loss of revenue (fixed cost component of tariff) attributable to the default in fuel supply; some of them suggest that the penalty could be on the basis of additional cost incurred in procuring fuel from alternative sources. PSUs and the Railways, however, contend that it should be related to the value of the fuel not supplied. In the absence of any satisfactory resolution to this problem, financial closure is getting delayed because the lending institutions, understandably, are reluctant to bank on a risky fuel supply agreement (FSA). SEB restructuring and reform: Some of the state governments are taking steps to introduce comprehensive reform legislation that could, eventually, lead to (a) unbundling of generation, transmission and distribution segments; (b) setting up of independent regulatory institutions; and, in a few cases, (c) privatisation. For instance, in the state of Orissa, the government has decided to limit its own role to policy making and regulation. Under a scheme endorsed by a council of ministers in 1994, the Orissa SEB is to be unbundled into new generation, transmission and distribution companies. Hydropower will remain in state ownership, as will transmission, but thermal generation has already been partially privatised and distribution is expected to pass into private hands by the middle of 1999. Similarly, in the case of Andhra Pradesh, the reforms envisaged include restructuring of the Andhra Pradesh SEB into a series of subsidiaries including the 20 | Indian Infrastructure: Evolving Perspectives

AP Power Corporation, the AP Transmission Corporation and a number of regional distribution companies. The transmission company would remain in public ownership while the distribution companies would be privatised gradually; full privatisation of the distribution companies was expected by 2000–01. Initially, the generating company would also remain in the public sector but operate as a commercial organisation. Regulatory commissions: The Indian Parliament recently enacted the Electricity Regulatory Commissions Act 1998, which provides for the setting up of independent regulatory commissions at the central and state levels. The independent regulatory commissions are expected to ensure that the industry develops on a sustainable basis. It is expected that the regulatory commissions, being independent of the government, would be able to effect tariff rationalisation, which is normally a politically sensitive and unpalatable decision for the governments. While the central and a few state governments such as Orissa and Haryana have already set up regulatory commissions, a few other states are in the process of setting up the same. The Ministry of Power (MoP) is assessing the need to redefine the role of the CEA as its role overlaps with that envisaged for the Central Electricity Regulatory Commission. Further, since power is a concurrent subject (under the jurisdiction of both the states and the centre), the respective roles of the central and state electricity regulatory commissions in the development of the power sector, especially with regard to the approval of large projects, need to be clarified. Transmission bill: The Electricity Laws (Amendment) Bill 1998 provides for the setting up of central and state transmission utilities for undertaking inter-state and intra-state transmission respectively. Under license from the respective electricity regulatory commissions/governments, transmission licensees can construct, maintain and operate transmission (inter-state or intra-state) systems under the direction, control and supervision of respective transmission utilities. Multi-state mega projects and power trading company: The Ministry of Power, Government of India, has been actively pursuing the concept of setting up mega power projects in the private sector, to (a) harness the economies of scale and locations associated with large pit-head plants; (b) take advantage of the low prices prevailing in the international markets for power equipment; and (c) quickly make up for the shortfall in capacity addition in the private sector at the state level. Considering that these projects would be of about 1500 to 2000 MW capacity, the output is expected to be shared by more than one state; so these projects are also referred to as multi-state mega projects. Private investors, however, have been reluctant to enter into multiple Power Purchase Agreements (PPAs) with several states and have indicated that they would prefer to deal with a single agency. In response, the Ministry of Power is considering setting up an intermediary—the Power Sector in India | 21

Power Trading Company (PTC)—that will undertake to buy all the power from the mega projects. PTC, in turn, will sell that power to State Electricity Boards, using the transmission network of PGCIL and arrange to collect the payments from them. To enhance PTC’s operational flexibility in evacuating power across various states and to shore up its stature, the shares of PTC are sought to be held by PGCIL (the national transmission utility), NTPC, SEBs and major Indian financial institutions.

5. CONCLUSION Clearly, the Indian power sector is undergoing a crucial phase of transition. Reform efforts to increase generating capacity have exposed the inadequacies in the remaining segments of the sector. Both the central and state governments are actively engaged in finding viable solutions to achieve sustainable development in the sector. As of now, regulation, rapid capacity addition and SEB-reform, with a specific focus on improving revenues from the distribution segment, are emerging as important areas of concern.

NOTE 1. India has recently set up the Central Electricity Regulatory Commission, as another regulatory authority, which is discussed later. 22 | Indian Infrastructure: Evolving Perspectives

POWER SECTOR REFORM IN ARGENTINA: A Summary 3 Description November 1998

1. INTRODUCTION The debt crisis of the 1980s severely impacted the Argentine economy, leaving it with a legacy of high inflation, low growth, a deteriorating infrastructure and limited resources available with the government. Argentina’s electricity industry, like many others, was completely state-owned. Its generation capacity was balanced between thermal and hydroelectric resources, some of which were shared with neighbouring countries like Uruguay and Paraguay. The nuclear programme, started in 1950, provided about a tenth of the energy needs (see Table 3.1). Immediate attention needed to be given to the efficiency and resource problems of the sector, if it was to develop further.

Table 3.1: Total production of electricity and the share of different types of generation

Year Thermal Hydro Nuclear Total (mn kWh) 1985 41% 47% 12% 43,587 1990 49% 37% 14% 48,945 1995 44% 46% 10% 65,720

Status prior to privatisation Argentina set up its first fully state-owned utility, AYEE, in 1947 in an attempt to provide more capital to the sector. At the beginning of the 1990s, the industry was completely state-owned. In 1991, immediately prior to privatisation, Argentina had four federal (equivalent to the central sector) utilities, 19 provincial (equivalent to the state sector) utilities and two binational bodies controlling large Power Sector Reform in Argentina | 23 hydroelectric projects jointly owned by Argentina and Uruguay, and Argentina and Paraguay respectively. The federal utilities owned a large proportion of the generation, including all the nuclear generation plants and transmission assets nationwide, and the distribution assets for the capital, Buenos Aires, and its surrounding region. The provincial utilities owned nearly all distribution assets outside the region around the capital. By the time of reform, the system had deteriorated badly and was characterised by considerable operational and financial problems. The cost of electricity was high (around US$60 per MWh), there were large commercial losses due to theft and non-payment and periodic threats of blackouts, aggravated in times of low rainfall by a large dependence on power from hydroelectric stations.

2. NATURE OF ELECTRICITY REFORM IN ARGENTINA Objective The goal of the reform process was to have an electricity industry that was capable of ensuring the economy sufficient energy at the best price that reflected the economic costs of maintaining and expanding the activity. It was also driven by the increasing inability of the government to service the public debt and the need to attract private investment to the sector. In January 1992, the electricity privatisation law was passed. The reform was based on principles of open access to the wholesale capacity, energy pool for generating facilities and least cost centralised dispatch. A national regulatory body, Ente Nacional Regulador de la Electricidad (ENRE), and a national wholesale market for electricity, CAMMESA, were established. Transmission and dispatch were mandatorily separated from generation and distribution, and no generator was allowed to control more than 10 per cent of the system’s capacity.

Restructuring Prior to privatisation, three federal companies,1 SEGBA, AYEE and HIDRONOR, were restructured by separating their generation, transmission and distribution activities (see Figure 3.1). Companies to be privatised were sold through auction, using a two-envelope process. This included a qualifying technical offer and a competitive financial offer. Usually, at least a bare majority (51 per cent) was offered for sale, and sometimes much more. For example, 98 per cent of a 448 MW hydroelectric plant was sold to a domestic aluminum company, retaining 2 per cent for the employees.2 In transmission and distribution, long-term concessions were awarded. The first of SEGBA’s generators was sold in April 1992, followed by two more in May and August. Two distribution companies were sold in September, followed shortly by a high-voltage transmission 24 | Indian Infrastructure: Evolving Perspectives company. The success of the privatisation depended to a great extent on the regulatory and commercial environment as determined by ENRE and CAMMESA.

1 6 National high Regional lower voltage (500 kV) voltage (220 kV) transmission transmission company companies

SEGBA HIDRONOR AYEE (Servicios Electricos del Gran (Hidroelectrica (Agua y Energia Electrica) Buenos Aires) Norpatagonica)

6 3 22 5 Thermal Distribution Thermal generation companies Hydroelectric generation companies 4 generation companies Hydroelectric generation companies companies

Figure 3.1: Restructuring of the Argentine electricity industry (federal assets)

ENRE The national regulator is responsible for all stages of the industry, with particular emphasis on transmission and distribution. It mediates disputes between electricity companies and enforces federal laws, regulations and concession terms. It also oversees the wholesale market, establishes service standards that must be met and sets the maximum price (price-cap) for transmission and distribution services. Generation is not subject to price regulation, as it is deemed to be a competitive activity with free entry.

CAMMESA An independent nonprofit operating agency, CAMMESA is directed by a board composed of two representatives of the federal government, power generators, transmission companies, distribution companies and large users. It has three primary tasks: dispatching power, determining fixed fees to be added to energy prices to cover capacity charges and the cost of transmission, and ensuring adequate reserve capacity in the system. CAMMESA dispatches power to the national grid by sending the cheapest power first, until current demand is satisfied. The payment to each generator is based on the highest cost producer whose power is dispatched. Power Sector Reform in Argentina | 25

CAMMESA’s budget is limited to 0.65 per cent of the total transactions in the wholesale market. Salaries, for around 150 staff members, represent almost three-quarters of its expenditure.

Box 3.1: Objectives of CAMMESA

• Dispatch energy optimally by minimising the total operation cost. • Maximise the security of the system and the quality of the electricity supplied. • Plan energy needs and forecast market prices. • Calculate economic transactions between market agents. • Bill, collect and make payments in the Wholesale Electricity Market. • Supervise option market operation and carry out technical dispatch of contracts. • Guarantee transparency and equity of Wholesale Electricity Market decisions.

Box 3.2: Objectives of ENRE

• Determine basis and criteria for assigning concessions. • Publicise and enforce regulatory structure, contracts and public service obligations. • Issue regulations on safety and technical procedures and monitor compliance. • Monitor billing, control, meter use and service quality. • Define basis for calculating tariff and ensure compliance. • Regulate sanction proceedings, impose penalties and take relevant issues to court. • Issue an annual report and recommend policy actions to the executive as needed.

Prices Demand and supply determine energy prices. The supply side of the wholesale market is composed of independent power producers, privatised generators, publicly owned generators and imported electricity. The demand side of the market is made up of private and public distribution companies, large users (currently more than 100 kWh annually) and foreign consumers. There are three main types of prices: contractual, seasonal and spot. Transmission and distribution prices (for supplies through distribution companies) are regulated. Contractual prices differ from contract to contract. They are negotiated between generators, distribution companies and large users for a minimum period of one year. Seasonal prices are determined by CAMMESA twice a year, based on parameters like forecasted demand, power availability, fuel prices, reservoir capacity, etc. Distribution companies purchasing power in excess of contracted amounts from the market pay the seasonal price. Spot prices vary hourly and are 26 | Indian Infrastructure: Evolving Perspectives paid by generators who cannot supply power they have contracted to sell, and large users who contracted for less power than they need. Generators who have power in excess of their contractual obligations, distribution companies and large users who have contracted to buy more than they currently require can sell power in the spot market. Obviously, price trends in the spot market affect the price paid for contracted power. 40

38

36

34

32 US$/MWh 30

28

26 My–Jl/94 My–Jl/95 Ag–Ot/94 Ag–Ot/95 Fb–Ap/94 Fb–Ap/95 Ap/92–Jn/94 Nv/94–Jn/95 Nv/95–Jn/96 Figure 3.2: Estimate medium monomial contract price in the market

Generation Argentina allows free entry into the generating market. Current and prospective generators make their own judgements, take their own risks on demand growth, investment levels, fuel market trends, etc. like producers of any other commodity. As noted above, there is no surety of dispatch and the price for energy dispatched depends on the highest cost producer whose power is dispatched. Full cost recovery is thus not guaranteed. Generators receive two types of payments from the market—one for the dispatched electricity and the other for the capacity offered to the grid. The capacity payment is sufficiently low so that generators have a dual incentive to reduce costs—first, to have their electricity dispatched and second, to increase their margins from electricity sales. Generating companies are prohibited from controlling more than 10 per cent of the system’s capacity, and cannot own majority shares in transmission facilities. However, they are assured open and equal access to the national grid. There are currently about 40 companies in Argentina, of which about 10 are still owned by federal and provincial authorities. These effectively act as independent power producers, selling their power through the wholesale market. Power Sector Reform in Argentina | 27

Transmission In contrast to generation, transmission is closely regulated. Firms enter after successfully bidding for a fixed-duration concession for a particular area and can charge no more than a regulated maximum price, providing an incentive to reduce costs. Currently, the concessions have been awarded for 95 years but the controlling shareholders’ package is rebid every 10 years. The charges have two components— one based on availability and the other on use. Concessionaires are required to allow third parties open access to their transmission network. They are not allowed to buy or sell electricity. The revenues of the transmission company are collected by the wholesale market through a surcharge on energy prices. The high-voltage (500 kV) transmission network was built by combining the transmission assets of the federal utilities into a single entity called TRANSENER. TRANSENER also operates one of the lower voltage regional systems. It serves 14 of Argentina’s 24 provinces, carrying most of its power. In addition, there are five other lower voltage (220 kV) regional systems (see Figure 3.3), of which three have been partially privatised.

Figure 3.3: Argentina’s transmission system Generators are required to provide new transmission facilities. A new transmission facility is being commissioned by a consortium of eight privatised electricity 28 | Indian Infrastructure: Evolving Perspectives companies, each of whom is legally prohibited from holding a majority share. They will fund the construction and award the concession. The project is expected to cost US$250 million. In addition, the same consortium is also planning a capacitor project. Distribution Like transmission, firms may enter distribution only by bidding for a concession. They also have regulated prices and a commitment to allow open access to third parties. The price caps are reset every five years; in the interim, the regulated company can avail of the benefits of cost reduction. Large users can, however, choose to be supplied directly by the generators or buy directly in the spot market, instead of through the distribution company. The number of large users in the wholesale market has risen rapidly. As distribution companies must supply to large users at the same rate they charge other customers, this helps to keep prices under control. While all federal distribution assets have been privatised, many provincial distribution companies remain in the public sector.

3. EFFECT OF REFORMS Lower prices and higher reliability Following the reform process, electricity prices fell sharply. After a period of turbulence, they stabilised at around half the pre-privatisation levels (see Figure 3.4). The extent of outages also reduced considerably (see Figure 3.5). In EDENOR’s distribution areas, outages fell from over 20 hours to around 5 hours a year. Generators were fitted with power system stabilisers, which permitted minimal disconnection of generating capacity while addressing transmission faults.

US$/MWh 50 45 Capacity 40 Energy 35 30 25 20 15 10 5 0 1992 1993 1994 1995 Figure 3.4: Evolution of capacity and energy prices Power Sector Reform in Argentina | 29

GWh 700 16% of demand 600

500

400

300

200

100

0 Jul-88 Jul-89 Jul-90 Jul-91 Jul-92 Jul-93 Jul-94 Jul-95 Jan-88 Jan-89 Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Figure 3.5: Outages as a per cent of energy demand

Increased efficiency Electricity loss for non-technical reasons, such as faulty billing and theft, came down and generation availability increased considerably. For example, the La Plata distribution company increased its customer base by over 20 per cent. Similarly, the availability of Costanera, a generator near Buenos Aires went up from 30 per cent to 75 per cent.

Increased investment Substantial investments were also made in upgrading the assets bought from the government. EDENOR, for instance, made capital investments of US$380 million over 1992 to 1995, and plans an additional US$500 million until the year 2000. It is anticipated that the total investment in the electricity sector by 2000 will be around US$7 billion.

4. CONCLUSION The Argentine experience demonstrates that it is possible to effect measurable changes in a state-owned electricity sector suffering from lack of funds and inefficient management within a reasonably short time frame. A number of the problems that plagued Argentina, such as high distribution losses and low generator availability, are similar to what India faces today. The Argentine power system is much smaller and has a better hydrothermal mix when compared to the Indian power system as a whole,3 but it may prove profitable to examine the 30 | Indian Infrastructure: Evolving Perspectives

Argentine experience to see whether it would help with the design of reform in India.

NOTES 1. The fourth, CONEA, was the nuclear energy utility, which was not privatised. 2. Prices could vary substantially. For example, a 20 MW oil and gas-fired generating plant was bought by a paper manufacturer for US$8.5 million. 3. The Argentine system is about 16,000 MW, of which nearly half is hydroelectric capacity. This compares favourably with some of the regional grids in India, though it is only a fifth the size of the national system. Orissa Power Sector Reform | 31

ORISSA POWER SECTOR REFORM: A Brief 4 Overview of the Process February 2000

In India, Orissa is the first state to have undertaken comprehensive reform of its power sector. Initiated in 1996, the programme is still under implementation.1 In fact, the crucial step of privatising distribution occurred only in 1999. Any attempt to evaluate the programme to determine its success or otherwise would therefore be premature at this juncture. Nevertheless, the reform experience to date has underscored several important lessons, which could be of value for other states that are seeking to reform their own power sectors.

1. SALIENT FEATURES OF THE ‘ORISSA MODEL’ The salient features of the reform programme included the unbundling of transmission, distribution, thermal power generation and hydel power generation into separate businesses; the establishment of an independent regulatory commission; and the subsequent partial divestment of equity in the thermal power generation and distribution business units to the private sector. As part of the first transfer scheme, effective from 1 April 1996, the State Electricity Board was split into three entities, viz., the Orissa Power Generation Corporation (OPGC for thermal power, the Orissa Hydro Power Corporation (OHPC), for hydel power and the Grid Corporation of Orissa (GRIDCO) for transmission and distribution. GRIDCO was to function as the single buyer of power within the state, for onward supply to the distribution companies. Under this transfer scheme, the state government also revalued the transmission and distribution (T&D) assets.2 According to GRIDCO, this was done in order to: (a) create a positive capital base for the new companies expected to raise substantial debt for making T&D investments; 32 | Indian Infrastructure: Evolving Perspectives

(b) set off some of the dues from the state government to the Orissa State Electricity Board (OSEB); and (c) raise some reasonable revenues from the sale of assets. A 49 per cent stake in OPGC was subsequently divested through an international competitive bidding process. Messrs AES Corporation bought the stake for Rs 603 crore. As part of the Second Transfer Scheme, effective from November 1998, the distribution-related assets, liabilities, proceedings and personnel of GRIDCO were transferred to four wholly owned subsidiary companies. These companies were subsequently privatised through the sale of 51 per cent GRIDCO’s equity to the private sector. Of these four distribution companies (distcos), three were bought by Bombay Suburban Electric Supply (BSES) in April 1999, whereas a joint venture between Jyoti Structures and M/s AES Corporation bought the fourth one in September 1999. The procurement as well as sale of power by various business units is regulated by the Orissa Electricity Regulatory Commission (OERC). The commission, while regulating retail tariffs, takes into account various parameters, including the cost of power procurement, capital base, reasonable return, acceptable level of losses, employee cost, interest and depreciation.

2. EXPERIENCE TO DATE The reform experience in Orissa to date has been rather painful. GRIDCO’s operating balance is turning out to be inadequate to service even present working expenses, leave alone servicing the past dues/liabilities which were loaded onto it at the time of disinvestment and restructuring. The cash deficit of GRIDCO is likely to increase from Rs 360.67 crore in 1998–99 to around Rs 500 crore in 1999–2000. This, in part, is due to GRIDCO’s arrangements with the distribution companies for deferred payments—which were accepted by GRIDCO in order to conclude their sale. On the distribution front, BSES, which bought three distribution companies, expects the combined loss figure in these companies to be about Rs 174 crore at the end of their first year of operation. The financial distress is mainly on account of discrepancies between the information provided in the Information Memorandum at the time of privatisation, the basis adopted for the regulatory decisions, and the reality. For instance, BSES finds that its subsidiaries’ operations are not viable at the tariffs fixed by OERC, which were based on a T&D loss level of 35 per cent, because the actual losses are higher at 45 per cent to 47 per cent. While the actual losses are Orissa Power Sector Reform | 33 about 5 to 6 per cent higher than the distribution losses indicated in the Information Memorandum, OERC has fixed tariffs based on a lower T&D loss level. According to OERC, this was done with a view to provide strong incentives to reduce T&D losses. Similarly, it is reported that GRIDCO, in its enthusiasm to show lower distribution losses, estimated higher billings that are not based on meter readings. These became part of ‘current assets’ transferred to the distcos at the time of handover. According to BSES, its companies would not be able to realise these doubtful receivables. Even if the regulatory commission is willing to accommodate the demand of the distcos to base tariffs on a more realistic estimation of losses and receivables, the scope for a drastic upward revision of tariffs is limited. Unless some of these issues are resolved, the distcos are bound to start with an inevitable sickness.

3. IMPORTANT LESSONS FROM THE ORISSA EXPERIENCE On the face of it, the reform programme in Orissa appears to have had all the ingredients, such as unbundling and privatisation, which were present in the successful reform efforts in other countries.3 A closer examination, however, reveals certain limitations, which offer important lessons from the Orissa experience.

3.1 Distribution should be privatised early on in the process Until recently, for all practical purposes, the Orissa model was almost a continuation of the previous SEB regime. The GRIDCO performed both transmission and distribution activities from April 1996 to November 1998, that too under state-ownership, with the attendant lack of incentives for improving efficiency. Moreover, allowing GRIDCO to perform both transmission and distribution complicated the process of separation further down the line, as one had to revisit the issue of apportioning the losses arising during this period of combined operation to the successor entities. A complete separation of transmission and distribution functions at the time of corporatisation would have helped the new entities to start with a clean slate. An analysis of the power sector reveals that losses have been heaviest in the distribution segment, the cash-generating end of the business, underscoring the significance of privatising distribution as soon as possible. Any interim arrangement such as creating a combined transmission and distribution company would not help stem the rot.

3.2 Pitfalls to avoid in the process of privatising distribution The process of distribution privatisation itself was fraught with several pitfalls such as regulatory uncertainty, information asymmetry and prior escrow commitments. 34 | Indian Infrastructure: Evolving Perspectives

At the time of privatising distribution zones, the private operators bid in an environment where they were uncertain about the regulator’s view regarding the valuation of the existing asset base, the likely profile of prices or performance levels. Some operators made their bids under the presumption that the regulator would use the sixth schedule of the Electricity (Supply) Act 1948 and accept the government’s overvaluation, whereas others did not put in their bids as they found the regulatory risk to be unacceptable. Understandably, those who bid did not see any apparent reason to contest the overvaluation of assets, and assumed that it would be reflected in the subsequent tariff. Similarly, the data provided in the Information Memorandum at the time of privatising distribution zones does not appear to conform with reality. The following observations of BSES indicate the extent of information asymmetry between the Memorandum and reality: Since authentic data on assets, receivable, etc., was not available, a full- fledged MIS was developed by consultancy agencies. Now, it is quite clear that the world of MIS and reality are significantly different. An unrealistically better picture of the distribution companies was presented through these information memorandum and documents. The future projections made on the basis of MIS are not only unrealistic but also too difficult to achieve within the time period set out. Perhaps if the consultants had done a better job, the exercise would have turned out to be more realistic and reliable, and the interest of privatisation process would have been better served. —BSES Limited 4 However, bidders took into account the assumption contained in the Information Memorandum since the government had set an implicit reserve price equal to the par value of the share, which was arrived at based on those assumptions. The combination of regulatory uncertainty, over-valuation and information asymmetry limited the number of bidders for the distribution companies. In addition to these problems, the cash flows of the Central Zone in Orissa were escrowed to meet power purchases from OPGC, where AES was a 49 per cent joint venture partner. Added to the other problems of the Central Zone, this pre-emption of cash flows affected investor sentiment so adversely that no responsive bid was received for the Central Zone.5 Finally, after a re-bid, the zone was sold to another AES joint venture in September 1999. In terms of the original reasons for revaluation, the government has benefited by being able to set off its dues to the erstwhile OSEB. It also managed to sell these assets for the stated book value. The revaluation has not made the distribution companies any more able to raise debt (if anything, it has made them less creditworthy). Orissa Power Sector Reform | 35

Post-privatisation, the regulator has been left to tackle the consequences of overvaluation of assets and the poor quality of data contained in the Information Memorandum. Not surprisingly, BSES, the successful bidder which took over three privatised distribution companies, and AES have already started raising these issues for consideration by the regulator. In this context, the government should appreciate that by opting to arbitrarily overvalue assets, it will be harming the process of privatisation.6 The primary aim of privatisation should be to improve the performance of the sector so as to provide quality power to consumers at economic prices, and not to increase revenues for the government. Accordingly, the government should resist any temptation to seek rents from the process of privatisation. One way of doing this is to encourage bids based on the potential of the assets being privatised for generating future cash-flows. In view of the above, privatisation of distribution should be preceded by a clear statement from the regulatory commission, detailing its regulatory approach and including its views regarding what constitutes an appropriate valuation of rate base, likely profile of prices and expected performance levels. The government’s reserve price, if any, should be arrived at based on the same transparent regulatory framework; this information should be available to all potential bidders. Such clarity would encourage serious private operators to participate, conduct due diligence of the existing assets, and submit a realistic bid. Otherwise, bids based on imperfect information and assumptions, though successful, are bound to create regulatory tussles and demands for re-negotiation and additional support. In addition, a clear advance indication of regulatory intent would help in preventing overzealous governments from excessively overvaluing their assets just prior to privatisation, with the hope to prevail upon the regulator to ‘grandfather’ such decisions. One might argue that it is the sole responsibility of the bidders to conduct their own due diligence rather than relying on the Information Memorandum. Nevertheless, in the interest of achieving successful and smooth privatisation, it would perhaps be more pragmatic to create appropriate conditions for encouraging more rational bidding in the first place.

3.3 Need for continuing financial support from the state government As any expectation for dramatic improvement in performance immediately after privatisation is unrealistic, one would have expected the government to continue with its support to the sector, which it had been providing in the form of mandatory subsidies to GRIDCO’s predecessor, OSEB. Instead, the government promptly withdrew its support, even though the hopes of GRIDCO to raise a structural 36 | Indian Infrastructure: Evolving Perspectives adjustment loan from the World Bank on soft terms, to tide over the problem of anticipated cash shortfall, did not materialise. GRIDCO’s problems were compounded by the fact that even at the time of privatising the distcos, it was asked to retain huge liabilities on its books in order to make the distcos financially attractive. An informal understanding of GRIDCO with the World Bank and the Department for International Development (DFID), to ensure its viability by a separate scheme, is yet to materialise. The above experience underscores the importance of a structured time-bound financial support mechanism, which should taper off over time, based on targeted improvements in revenue collection. It is important to provide such support to distribution companies to ensure specific targeting and to enable benchmarking performance across different distcos. Otherwise, providing financial support through a combined transmission and distribution company would weaken the incentives to improve performance at the distribution level.

3.4 Limitations of the single-buyer model The privatised distribution companies in Orissa are not free to source power from generating companies of their choice. GRIDCO, the state-owned transmission company, acts as the sole procurement agency on behalf of the distribution companies. This option is bereft of competitive pressures and leaves little scope for achieving procurement efficiency. The experience of power sector restructuring in countries such as Argentina, England and Wales squarely underlines the importance of competition. It is competition in power generation, coupled with the choice given to major consumers to source from the supplier of their choice, which has driven the wholesale prices of electricity in these countries.7 In order to facilitate competition, the unbundling of the natural monopoly segments (that is, transmission and distribution) from the segments that are amenable for competition (that is, generation) is necessary; but it is not enough. Hence, the distribution companies should be allowed to procure power from the generators of their choice or set up their own generating capacity, and the role of GRIDCO should be limited to providing transmission-related services only. Further, until the distcos have monopoly franchise for supply to consumers in their geographical areas, the regulator could oversee their procurement to ensure efficiency.

4. CONCLUSION While it might be a little premature to judge power sector reform in Orissa, the experience to date, however, clearly underscores a set of do’s and don’ts for other states which are engaged in the reform process. In summary, these are: Orissa Power Sector Reform | 37

Do’s: • Separate distribution from transmission and generation and privatise it early on in the process of reform. • Provide regulatory certainty prior to privatisation in order to encourage serious bidders and to avoid regulatory problems later. In particular, the regulatory commission should spell out its regulatory approach, including its views regarding the appropriate valuation of asset base, likely profiles of prices and expected performance levels. • The reserve price, if any, should be based on a clearly spelt out support that is likely from the government and a transparent regulatory framework; this information should be available to all potential bidders. • The government should continue its financial support to the sector over a limited time period. It is important to provide this support directly to distribution companies. • Plan for a quick transition away from a single-buyer model, which prevents distribution companies from procuring power from the generators of their choice, to an open-access model, where distribution companies and large consumers can source their own supply.

Don’ts: • Do not revalue assets prior to privatisation. This is, however, intimately tied to the regulator’s approach on tariff setting and the nature of relationship between asset-value and expected tariffs, i.e. whether it will use Schedule VI, which requires an asset base. • Do not escrow revenues from the distribution zones prior to privatisation.

NOTES 1. A timeline of key events in the restructuring is provided in the Annexure. 2. The state government took over the transmission and distribution assets of the Orissa State Electricity Board (book value plus capitalised expenses and interest at Rs 1200 crore) and revested them with GRIDCO after upvaluing by an additional Rs 1194 crore (additional 134 per cent). 3. For example, Argentina, England and Wales. 4. Vide a letter dated 27 October 1999, addressed to Special Secretary (Power), Government of India. 5. Tata Electric Company, which was selected as the partner for CESCO, did not pay the equity to GRIDCO, despite numerous extensions to the deadlines. 38 | Indian Infrastructure: Evolving Perspectives

6. Normally, the various valuation options available to the government include book (historic) value, replacement value, ‘re-valued’ value, or business value based on the potential for generating future cash-flows. 7. Following privatisation, Argentine wholesale electricity prices fell about 60 per cent from the pre-privatisation level of US$60 per megawatt hour in August 1992.

ANNEXURE Box 4.1: Timeline of key events in power sector reforms in Orissa Month and year Event November 1993 Power reforms programme announced by Chief Minister. April 1994 Reforms formally approved by the Council of Ministers. April 1995 Government issued a formal statement of its power policy. November 1995 State assembly approved the Orissa Electricity Reforms Act 1995. April 1996 The Orissa Electricity Reforms Act 1995 and the restructuring of the industry became effective. April 1996 The first Transfer Scheme. Assets, liabilities, proceedings and personnel of the Orissa State Electricity Board were transferred to Orissa Hydro Power Corporation (OHPC, for hydel generation) and the Grid Corporation of Orissa (GRIDCO, for transmission and distribution). August 1996 The Orissa Electricity Regulatory Commission (OERC) became functional. All three members took oath of office. March 1997 The Orissa Distribution and Retail Supply License and the Orissa Transmission and Bulk Supply License were issued to GRIDCO. These licenses became effective on 1 April 1997. March 1997 OERC issued an order on retail electricity tariffs for all types of consumers, effective from 1 April 1997. Orissa Power Sector Reform | 39

Box 4.1: Timeline of key events in power sector reforms in Orissa (contd...) Month and year Event November 1998 The second Transfer Scheme. Distribution related assets, liabilities, proceedings and personnel of GRIDCO were transferred to four wholly owned companies of GRIDCO—CESCO, WESCO, SOUTHCO and NESCO (Central, Western, Southern and Northern Electricity Supply Companies, respectively). November 1998 OERC issued its Orders on Retail Electricity Tariffs and Bulk Electricity Tariffs, effective from 1 December 1998. January 1999 BSES emerged as the top bidder for WESCO, SOUTHCO and NESCO. April 1999 BSES took over management of WESCO, SOUTHCO and NESCO, through acquisition of a 51 per cent stake. July 1999 Permission was granted to AES Transpower, for acquiring a 51 per cent stake in CESCO. September 1999 AES Transpower took over control of CESCO. December 1999 OERC issued Tariff Orders for retail supply for CESCO, WESCO, NESCO and SOUTHCO respectively and a Tariff Order for bulk supply and transmission for GRIDCO for 1999–2000. 40 | Indian Infrastructure: Evolving Perspectives

POWER SECTOR FINANCING: A Note on 5 Conditionalities June 2000

1. BACKGROUND Since 1991, when the power sector was opened to private investment, financial institutions (FIs) and banks have sanctioned assistance of Rs 39,853 crore to 57 independent power projects (IPPs) with an aggregate capacity of 21,393 MW in 15 states. Typically, these projects are financed on the strength of their power purchase agreements (PPAs) with the State Electricity Boards (SEBs) and are supported by a 3-tier security mechanism consisting of a letter of credit, escrow cover on SEB receivables, and state government guarantees. However, the progress of implementation of these projects has been rather slow. As many as 37 projects with an aggregate capacity of 14,618 MW are yet to achieve financial closure.1 One of the main reasons for this slow progress, as identified in previous meetings of the Crisis Resolution Group (CRG), is the non-availability of escrow cover. Most of the State Electricity Boards either have no capacity to offer escrow support to IPPs or have exhausted their escrowable capacity. At the meeting of the CRG on May 16, 2000, the Union Minister for Power suggested that the only alternative left is to link the security mechanism with reform, by setting up credible milestones. Strict adherence to reform milestones by state governments will, in due course, result in improved revenue realisation, which would improve security for lenders. It was therefore decided to derive a set of conditionalities including, but not limited to, reform milestones, which could be used by FIs and banks as a basis for investment in the Indian power sector. This note analyses the factors affecting the cashflow and viability of investments in the sector, with a view to establishing a set of Power Sector Financing | 41 conditionalities. It also draws upon prior experience with conditional lending, in order to identify possible pitfalls in the process.

2. FACTORS AFFECTING THE VIABILITY OF THE INDIAN POWER SECTOR This section draws upon the factors affecting the viability of the sector to delineate two types of conditionalities. The first is a set of prior actions (or pre-conditions) that are critical to establishing the government’s commitment to reform, while the second set of arrangements is required to attract private investment and participation in the sector.

2.1 Conditions precedent 2 A recent World Bank Policy Research Report concludes that conditionality is unlikely to bring about lasting reform if there is no strong domestic movement for change. Therefore, it is necessary to ensure that the state is committed to reform before embarking on a conditionality-based borrowing programme. This commitment of the state can be measured by its willingness to meet certain preconditions before disbursement.

2.1.1 Independent regulation The monopoly nature of transmission and distribution makes it possible for the business to exploit monopoly power, and also makes it vulnerable to interference by the state in view of the essential nature of the service. An independent regulator could address both these issues. The establishment of a credible regulatory institution is a critical requirement for sector reforms.

2.1.2 Unbundling The existing vertical integration extends the natural monopoly characteristic even to those segments that are amenable for competition, such as generation. Unbundling of generation, transmission and distribution is a pre-requisite for achieving competition-induced efficiencies.

2.1.3 Funding of existing liabilities The SEBs have a number of existing liabilities, many of which are unfunded, such as pensions, provident fund etc. They also have dues to central sector generating units, Coal India Limited, Railways, and other such organisations, along with receivables of doubtful quality, including payments from various state government bodies. It is necessary that these liabilities and assets be clearly provided for by the government, who is the owner of the SEB. One way of achieving this could be to 3 segregate the proceeds from privatisation in a separate fund to meet these liabilities before it is used for general budgetary support. 42 | Indian Infrastructure: Evolving Perspectives

2.1.4 Market structure A competitive market structure in both generation and supply is essential in order to confer benefits to the consumer. Avoidance of this issue merely increases uncertainty with respect to the timing of introduction of competition, which affects the bidding process for both generation and distribution assets. It is necessary for the government to spell out a time frame for the introduction and the extent of competition through the creation of a bulk market for power and the institution of open access arrangements for the grid.

2.1.5 Distribution privatisation Theft of power is currently the single most important cause of financial bankruptcy in the sector. Privatisation of distribution is considered to be the only credible alternative that can provide sufficient incentive to contain it. Accordingly, distribution privatisation is essential for improving cash flows in the sector.

2.1.6 Employee-related issues Significant over-manning in the sector appears to be one of the principal sources of inefficiency in the sector. Currently, resistance from the employees appears to be related to the lack of reassurance regarding payment of pensions, provident fund and other liabilities, which need to be provided for as mentioned above. In order to establish a viable power sector it will be necessary to allow flexibility in rationalising manpower. In addition, divestment proceeds can be used to make provisions for severance payments.

2.2 Conditionality for disbursement Apart from the conditions precedent detailed above, there is a need for an additional set of conditionalities in order to sustain the viability of the sector through appropriate allocation of risk among various stakeholders. The remaining part of this section looks at the broad risks in the sector, proposes an allocation of those risks and then suggests conditionalities aimed at achieving that allocation and, if possible, mitigating those risks. It focuses on distribution 4 and generation investments.

2.2.1 Distribution investments Distribution companies need to be reassured that if they manage an efficient distribution business, i.e., reduce distribution losses and avoidable outages, they will earn a reasonable return on investment. Viability of the distribution business is critical to ensure the growth of the sector. In broad terms, this will be affected by the following: (a) valuation of liabilities and assets handed over at the time of privatisation (b) level of distribution losses—actual and allowable Power Sector Financing | 43

(c) tariff level (d) extent of subsidy support from the government (e) timing of introducing choice to different classes of consumers Of these, to the extent that the liabilities and assets being handed over at the time of privatisation are clearly defined, the distribution company can be deemed to have control over their valuation. After all, it is free to conduct due diligence and incorporate its assessment of the impact of these assets and liabilities into its bid. Further, the distribution company can be expected to have considerable control over the actual level of distribution losses. On the other hand, the distribution company is unlikely to have control over the other items, viz., tariffs and allowable loss levels, which depend on the regulator, and subsidy support and timing of introducing choice, which depend on the government. A degree of certainty on these matters can be achieved through a combination of conditionalities and guarantees.

2.2.2 Regulatory certainty The level of tariffs and the allowable level of distribution losses are prone to uncertainty since they fall under the purview of the regulatory body, which is still 5 an evolving institution. Such uncertainty could be reduced if the regulator spells out its regulatory approach ex-ante, i.e., at the time of privatisation—with a medium-term five-year tariff profile, conditional on levels of performance such as distribution losses. It is possible to build in annual regulatory reviews in such a process in order to ensure that significant deviations from the assumptions underlying the tariff and performance profiles are accounted for. In case the government fails to convince the regulator to offer such certainty, an alternative route could be for it to offer a tailored guarantee to protect the distribution company against the risk of inadequate tariff being allowed by the regulator. A suggested structure is given in Annexure 1.

2.2.3 Subsidy commitments The responsibility of fulfilling subsidy commitments within a predetermined time frame should rest squarely with the government. In view of the possibility of non- fulfilment of such commitments, given the fiscal situation of most states, the commitment to make subsidy payments will need to be guaranteed by a credit- worthy body, such as the Government of India or the World Bank.

2.2.4 Timing of introducing retail choice The need to announce this at the time of privatisation has already been mentioned as a condition precedent, for the reasons mentioned therein. 44 | Indian Infrastructure: Evolving Perspectives

2.3 Generation investments The generating plant has three primary concerns—whether it will be dispatched, 6 whether it will be paid and how much it will be paid. The first relates to the overall demand and relative cost of the plant (assuming a merit order dispatch scheme), while the last two are intimately related to the viability of the distribution business and the structure of the power market, such as PPAs, bulk markets, etc. This section focuses on the first concern, as the remaining concerns are expected to be suitably alleviated with improvements in the viability of the distribution business and the freedom to enter into commercial contractual arrangements with each other.

2.3.1 Dispatch risk The dispatch risk can be further separated into two parts, namely, risk of overall demand realisation and risk of not being in the merit order. In the former, the risk is that the plant, which is built in anticipation of demand growth, may not be dispatched because growth in demand is lower than demand forecasts. In the latter, even if the growth in demand is as per the forecast, there may be less expensive 7 sources of supply that would be able to meet the demand. Ideally, both these risks should remain with the private sector, if they are allowed to make decisions on capacity and technology, etc. However, there may be a case for sharing the risks of non-realisation of demand, if the buyer (in the current scenario, the SEB and its owner, the government) insists on adding capacity that is higher than what the investor considers prudent. This risk may be mitigated to the extent that the power from the plant could be sold to buyers in other regions. The residual risk, if any, could be mitigated through a guarantee from the buyer (preferably, the state government), which would become effective only when the total allowed power purchase (by all distribution companies) by the regulator is less than the demand forecasted by the government. A suggested structure for this guarantee is attached in Annexure 2.

3. EXPERIENCE WITH CONDITIONAL LENDING This section examines the experience with employing the conditional lending approach, especially from the perspective of financial institutions and banks.

3.1 Multilateral conditional lending The concept of conditional lending is often employed by aid agencies, such as the World Bank, to foster reform, by making financing conditional on the adoption of certain policies. According to the World Bank Policy Research Report mentioned earlier, about a third of the World Bank’s adjustment loans fail to meet their reform objectives. In many cases loans were disbursed even though policy measures were Power Sector Financing | 45 not carried out. It also quotes a recent study that indicates that the number of conditions or resources devoted to preparation and supervision had no significant effect on the probability of success or failure of reform. In fact, it appears that the approach of the donor agencies in the past to ‘buy reform’ by offering assistance to government, that were not otherwise inclined to reform, has failed. According to the report, the primary benefit of conditionality-linked loans is that they provide a means by which reform-minded governments can publicly commit to policy measures and send a signal to the private sector that the reform programme is credible.

3.2 Conditional lending in India In the Indian power sector, the World Bank and the Power Finance Corporation (PFC) have tried out the conditional lending approach. The efficacy of their 8 approach can be gauged by the following assessment of the World Bank: … Operational and Financial Action Plans (OFAPs) from 1989 through 1997 (with Bank support under Ln.3436-IN from 1992) attempted revitalization of APSEB but could not stop its rapid decline into insolvency. It is perhaps too early to draw any meaningful conclusions on the effectiveness of the World Bank’s Adaptable Programme Lending (APL) approach in the states of Haryana and Andhra Pradesh. One of the salient features of the APL approach is that it provides a ‘stop-loss opportunity’ for the Bank at different stages of the project. In other words, the Bank can stop further disbursements if the government fails to achieve/fulfil the reform-linked milestones/conditionalities. This has happened in Haryana.

3.2.1 Possibility of replicating the conditional-lending approach of multilateral agencies Although tempting, it may be counterproductive for FIs/banks to replicate the conditionalities imposed by agencies such as the World Bank. These institutions, after all, differ significantly from the World Bank in more ways than one. First, it is important to note that the responsibility for repayment of World Bank loans is with the primary borrower, i.e., the Government of India (GOI) and not the state government. By contrast, the FIs/banks typically provide finance directly to public or privately owned utilities, whose creditworthiness is not backed by the GOI. The limited scope for sound state government guarantees is also fast disappearing. Second, the risk-absorption capability of the FIs/banks is more limited than that of agencies such as the World Bank, which operate on a much larger, more diversified and secure portfolio. 46 | Indian Infrastructure: Evolving Perspectives

Third, the success of the lending by FIs and banks is determined by the commercial performance of the loan, while the success of financing in the case of developmental projects supported by the World Bank encompasses a much broader definition of economy-wide impact, including social benefits. Fourth, they have to be satisfied with a lesser degree of leverage in order to effect their conditionalities as compared to aid agencies, who provide access to grant finance and long term low interest developmental loans, through windows such as the International Development Association. 3.2.2 Need for more caution It is clear, based on the experience and possibility of replication, that FIs/banks have to be more cautious while opting for conditionality-linked loans, as compared to multilateral agencies. This is especially so given the time-inconsistent nature of conditionalities, whereby the government can cease adhering to them, after having committed to them. The FIs would then be faced with accepting losses on disbursements already made to a project, or renegotiating with the state government from a weak position. Viewed from this perspective, the FIs/banks should insist on irreversible reform steps before committing their investments. Hence, a careful assessment of the states based on their commitment and capability to undertake reform should precede any attempt to structure a conditional lending programme. This is the basis for requiring prior action, based on the conditions precedent mentioned above. Concomitantly, the conditional lending programme should have the approval and commitment of the state legislature.

4. SUGGESTED CONDITIONALITY FOR POWER SECTOR FINANCING Conditional lending appears to be providing a window of opportunity for FIs/banks to facilitate private sector participation in the Indian power sector. However, in view of the low degree of leverage available to them and their low tolerance for risk of failure, they should be extremely cautious in choosing the states to which they would like to offer the conditional lending option. A suggested structure for conditional lending is given below. Conditions for lending to distribution companies Prior action 1. Formation of SERC 2. Unbundling of SEB 3. 100 per cent metering at 11 kV 4. Assumption of existing liabilities of the SEB by the state 5. Clearance or assumption (by the state government) of overdues of the SEB Power Sector Financing | 47

6. Creation of a fund to segregate divestment proceeds 7. Announcement of the time frame for introducing choice to consumers 8. Announcement of a medium-term tariff and loss profile by the regulator or execution of an average revenue realisation guarantee by the state, backed by the GOI or the World Bank 9. Privatisation of distribution

Conditionality 1. Creation of a fund to meet payment of subsidies or execution of Guarantee for payment of subsidies, backed by the GOI or the World Bank 2. Privatisation of generation

Additional conditions for lending to IPPs 1. Allowing IPPs to determine their plant capacity or execution of overall demand realisation guarantee by the state,9 backed by the GOI or the World Bank 2. Allowing use of transmission lines for wheeling of surplus power 3. Announcement of the time frame for establishing a bulk market

Further conditions for lending to state-owned generating plants 1. Improvement in plant availability

ANNEXURE 1 Average revenue realisation guarantee (limited to debt) Objective : To mitigate the risk of inadequate tariff being allowed by the regulator. Trigger : In the event that the average tariff allowed by the regulator is lower than a pre-specified level. Condition : The distribution company should meet pre-specified distribution loss levels. Explanation : This guarantee is triggered only if the cash flows of the distribution company are not sufficient to make debt service payments to lenders. In such an event, the following calculations and payments would be made: Method : a) Pre-specify a certain consumer ratio among various tariff categories and an overall level of Guaranteed Average Revenue Realisation (GARR), in rupees per unit. 48 | Indian Infrastructure: Evolving Perspectives

b) Multiply the tariff allowed by the regulator with the pre- specified consumer ratio and calculate the Implied Average Revenue Realisation (IARR), based on the power demand approved by the regulator. c) Deduct the IARR from the GARR to arrive at the ARR shortfall. d) If the ARR shortfall is positive, multiply the power demand approved by the regulator with the ARR shortfall and call this the ‘Maximum Alternative Distribution Payment’ (MADP). e) Make payments to each distribution company based on the amount needed to cover the minimum debt service or MADP, whichever is lower.

Example:

Years 1 2 3 4 5 1. Approved demand (MU) 32400 34800 37800 41400 45000 2. GARR (Rs per unit) 4.20 4.41 4.63 4.86 5.10 3. IARR (Rs per unit) 4.30 4.30 4.60 4.80 5.00 4. Target distribution loss (%) 35 30 30 25 20 5. Actual distribution loss (%) 36 32 28 24 20 6. ARR shortfall (Rs per unit) — 0.11 0.03 0.06 0.10 7. MADP (Rs crore) — — 113.4 248.4 450 8. Debt service (Rs crore) — 50 — 100 500 9. Guarantee payment (Rs crore) — — — 100 450 Note: In year 2, the distribution loss target is not achieved and hence there is no guarantee payment. In year 3, there is no debt service and hence there is no guarantee payment. In year 4, payment is made to the extent of debt service. In year 5, the debt service is 10 higher than the MADP, but payment is limited to the extent of MADP.

ANNEXURE 2 Overall demand realisation guarantee (limited to debt) Objective : To mitigate the risk of non-realisation of overall demand projections. Trigger : In the event that the overall power procurement allowed by the regulator is lower than the forecasted power demand. Power Sector Financing | 49

Explanation : This guarantee is triggered only if the cash flows of the IPP are not sufficient to make debt service payments to lenders. In such an event, the following calculations and payments would be made: Method : a) Compare the pre-announced forecast of power demand with the actual power procurement allowed by the regulator. Determine the extent of shortfall. b) Allocate the shortfall on a merit order basis to the various plants that are available for dispatch and with which guarantees have been signed. c) Multiply the allocated shortfall to each plant by the tariff of the highest cost plant in the list of procurements approved by the regulatory commission. Call this amount the ‘Maximum Alternative Generation Payment’ (MAGP) d) Make payments to each plant based on the minimum amount needed to cover debt service or the ‘Maximum Alternative Generation Payment’, whichever is lower. Example : There is an existing capacity of 5000 MW that supplies power at Rs 1.50 per unit. Additional capacity is expected to be added through Plant A (1000 MW starting from 2nd year supplying at Rs 2.50 per unit), Plant B (1500 MW starting from 4th year at Rs 2.00 per unit) and Plant C (1500 MW starting from 5th year at Rs 2.20 per unit). While Plants A and B have PPAs backed by overall demand realisation guarantees with the state government, Plant C is a merchant plant without similar guarantee cover.

Years 1 2 3 4 5 1. Total installed capacity 5000 6000 6000 7500 9000 2. Demand projected by government (MW) 5500 6050 6655 7320 8000 3. Demand allowed by the regulator (MW) 5400 5800 6300 6900 7500 4. Gap between projected and allowable demand (MW) 100 250 355 420 500 Merit order dispatch 5. Existing (5000 MW @ Rs 1.50 per unit) 5000 5000 5000 5000 5000 50 | Indian Infrastructure: Evolving Perspectives

Years 1 2 3 4 5 6. Plant B (1500 MW @ Rs 2.00 per unit) — — — 1500 1500 7. Plant C (1500 MW @ Rs 2.20 per unit) — — — — 1000 8. Plant A (1000 MW @ Rs 2.50 per unit) — 800 1000 400 0 9. Undispatched capacity of Plant A (MW) — 200 — 600 1000 10. Allocation of shortfall to Plant A (MW) (lower of 4 and 9, subject to merit order) — 200 — 420 500 11. Allocation in million units (@ 6 MUs/MW) — 1200 — 2520 3000 12. Tariff of the costliest plant dispatched (Rs/unit) 1.50 2.50 2.50 2.50 2.20 13. MAGP (Rs crore) — 300 — 630 660 14. Debt service (Rs crore) — — — 300 750 Guarantee payment (Rs crore) (lower of 13 and 14) ———300 660 Note: In year 2, there is no debt service and hence no guarantee payment. In year 3, Plant A is fully dispatched. In years 4 and 5, payments are limited to the gap between projected and allowable demand and also to the burden of debt servicing. In year 5, although 500 MW of Plant C also remains undispatched, it will not get any compensation, as it does not have guarantee agreements like Plants A and B.

NOTES 1. To date, 10 projects (aggregate capacity of 3,359 MW) have been fully commissioned, 1 project (420 MW) has been partly commissioned, and 9 projects (3,402 MW) are under various stages of implementation. 2. ‘Assessing Aid: What Works, What Doesn’t and Why’ published by the Oxford University Press (1998). A fuller discussion of the distinction between conditional multilateral lending and lending by FIs is given in Section 3. 3. Going by the dire financial condition of the sector, it would be a matter of surprise if the government could raise positive revenues from the process of privatisation of distribution. If it indeed does, it could be only on account of either expectation of a Power Sector Financing | 51

genuine turnaround, or sheer bravado on the part of the bidders or in anticipation of favourable re-negotiation with the government/regulator. The bulk of proceeds from privatisation should therefore be expected to come from privatisation of the generation segment. 4. The need for additional investment in transmission cannot be denied. However, in the initial phase of reform it is expected to remain with the state and the investment requirements would become clearer as the distribution and generation businesses evolve. 5. Similarly, the regulator may impact the valuation of liabilities and assets handed over at the time of privatisation depending upon the extent to which they are accommodated in tariff setting, either as pass-throughs or in the ratebase. 6. The equity investor is assumed to bear construction and technology risks, as is the current practice. Fuel supply risk is being ignored at this point, since the focus is on the state government, and because it is primarily a commercial risk to be borne by the equity investor, in the case of fuels such as imported coal. 7. This may come from new plants, renovated older plants or supply from other regions made possible by a more integrated grid. 8. The World Bank’s Project Appraisal Document on a proposed loan to India for the Andhra Pradesh Power Sector Restructuring Project, 25 January 1999. 9. This requires the state government to determine the ‘supportable capacity’ in order to structure the guarantee (see Annexure 2). 10. There may be a case here to capture the MADP due in earlier years into a notional reserve account in order to meet such contingencies. This would be a stronger guarantee than the one proposed in the note. 52 | Indian Infrastructure: Evolving Perspectives

SIX STEPS TO ACCELERATED PRIVATISATION OF 6 ELECTRICITY DISTRIBUTION January 2001

1. INTRODUCTION The power sector in India is facing a breakdown. Revenue arrears of State Electricity Boards (SEBs) grow by the hour, as consequently do their dues to central sector units like National Thermal Power Corporation Limited (NTPC) and Coal India Limited. Even in states like Karnataka, which are relatively well managed fiscally, the burden of power sector subsidies has grown to over a third of their fiscal deficit. Faced with this situation of a bankrupt SEB and a fiscally parlous state government, private independent power producers (IPPs) dare not invest and even if central sector units like NTPC make investments they are equally unlikely to be paid for the power they produce. Soon, power will become a binding constraint on our ability to maintain high growth rates. The need for reform in the power sector is therefore unarguable, especially in the distribution segment where the losses have been the heaviest. The estimated T&D losses, as is being demonstrated by the orders of State Electricity Regulatory Commissions (SERCs), in various states is closer to 40 per cent! This is true across the country, in states as varied as Andhra Pradesh, Gujarat, Haryana, Maharashtra, Orissa and Uttar Pradesh. Remedying this situation calls for the provision of adequate incentives and the transfer of risk along with reward that is, today, unfortunately not possible under public ownership in India. In order to leverage the accountability that is now possible as a result of the greater regulatory transparency, it is necessary to privatise the power sector, especially the customer end, i.e., distribution. This process of privatisation provides an opportunity to allocate the various risks in the power sector optimally, to the entity that can best control them. Accelerated Privatisation of Electricity Distribution | 53

Privatised distribution is not an alien concept in India. Today, private distribution licensees continue to operate in the cities of Mumbai, Calcutta, Ahmedabad and Surat. Some of these licensees also own generating assets. For instance, Bombay Suburban Electric Supply (BSES) sources about 50 per cent of its electricity requirements from its own plants. Similarly, the Ahmedabad Electricity Company (AEC) and the Calcutta Electric Supply Company (CESC) own generating capacity of 550 MW and 945 MW respectively. The fact that the tariffs of BSES, which caters to 19.8 lakh consumers in an area of 384 sq km, are competitive, and that its distribution losses are now only about 10 per cent, aptly underscores the superior performance of distribution under private ownership. Their successful track record highlights the possibility that privatisation of bundled distribution-cum- generation companies can be a viable and relatively quick option for privatising the power sector in India.

1.1 Need to accelerate the pace of privatisation Unfortunately, despite the need for privatisation of distribution in the reforming states and the impressive track record of private licensees, it is not taking place at the desired pace, for various reasons. On the one hand, reforming states are expending enormous time and effort in configuring distribution zones for privatisation, in the name of asset-valuation and in achieving the elusive ‘right’ balance between the subsidised and subsidising categories of consumers. For instance, in Orissa, the first state to have undertaken comprehensive restructuring of its power sector, although reforms were initiated in 1996 (in fact discussions started as early as 1992!), the crucial step of privatising distribution occurred only in 1999. Even in other reforming states such as Haryana, Andhra Pradesh and Uttar Pradesh, the privatisation of distribution is slated to take place only three to four years after the initiation of the restructuring process. On the other hand, as evidenced in Orissa and in the privatisation of Kanpur city in Uttar Pradesh, private investors are wary of participating in distribution privatisation, as they perceive several risks ranging from regulatory uncertainty to unreliable commitments on the part of the state governments.

1.2 A one-year timeline The adverse consequences of the delay in distribution privatisation are already visible in the continuing accumulation of losses in the SEBs and in the growing reluctance of investors to set up new generation capacity. Given the alarming pace of decay in the distribution systems and their management, quick privatisation of distribution is a sine qua non for the success of the reform process. Against this backdrop, this paper delineates a set of six inter-related steps for the successful and quick privatisation of distribution. It is our contention that it is possible to 54 | Indian Infrastructure: Evolving Perspectives privatise distribution within a year, through a diligent execution of the steps outlined below. An indicative timeline for this accelerated process is provided in Table 6.1.

2. SIX STEPS TO PRIVATISATION OF DISTRIBUTION Distribution privatisation is a difficult task in the best of times. It will be even more difficult in the current situation, given the low average tariffs and the decrepit distribution network. The private sector, when it enters the distribution business, is also required to make large investments. It makes these investments in the expectation of earning a fair return on these investments over the future. The private sector will not bring in investment unless there is a secure foundation for these expectations. This does not imply that the private sector needs to be protected from all risks. It does however imply that they should be provided with an environment that allows them to evaluate the various risks and price them accordingly. Like in any other financial transaction, in distribution privatisation too, success depends on the efficient allocation of risks. In the context of privatisation, the distribution business is fraught with several risks arising out of the following factors: • Existing commitments by SEBs, such as escrows and long-term power purchase agreements (PPAs) with IPPs • Unfunded liabilities from the past such as dues to suppliers of power and fuel and employee pension commitments • Regulation of tariffs, expected performance levels and allowable returns • Government support to meet subsidy commitments and • Market structure (input prices, competition and choice to consumers) In order to be effective, a distribution privatisation strategy should satisfactorily address these concerns. This is the purpose of this paper.

2.1 Step 1: Notify a policy eschewing escrows and long-term PPAs 2.1.1 Effect of escrows and long-term PPAs An escrow facility involves dedicating a stream of revenue from specified customers or regions into an escrow account maintained by an agent bank in order to meet the power purchase payment obligations of an IPP. This security mechanism was resorted to because the SEBs were perceived as poor credit risks. The primary claim on the revenue stream was therefore transferred from the distribution system to the IPP. In a situation where most of the privatised distribution regions would in any case be expected to have cash losses in the initial years, an additional pre-emption of cash flows would make it very difficult to run the distribution business. This negative effect of an escrow on the already low cash flow stream that would be received by Accelerated Privatisation of Electricity Distribution | 55 the prospective buyer makes it difficult to privatise a region that has been escrowed, as was seen recently with the Central Zone in Orissa. The escrow facility is also associated with very long-term power purchase agreements, extending up to thirty years, which were needed since the SEB was the sole buyer of power. In the context of transferring such contracts to a privatised distribution system, one solution that has been advanced is to retain a single bulk buyer, such as the transmission company. Here the experience of GRIDCO in Orissa indicates that such a measure may fail because the GRIDCO is unlikely to be a good credit risk, and the government would have to again step in to support the contract.1 Clearly, it is inappropriate to burden prospective private owners of distribution zones with fresh contracts, especially where they are not a party to the decision. More importantly, under the emerging market structures, such long-term contracts have a deleterious effect on competition as they prevent distribution companies from accessing the most competitive supplier of energy.

2.1.2 Breaking the vicious circle As long as the distribution business does not generate sufficient revenue and there remains a single buyer of power, there will continue to be a demand for escrows and long-term PPAs. This will continue until distribution is privatised and IPPs are allowed to enter into direct contracts with distribution companies and large consumers. At the same time, escrows and long-term PPAs adversely affect the attractiveness of the distribution zones and thereby obstruct the process of privatisation. The government should break this vicious circle and clearly opt for improving security through privatisation of the distribution business and eschew escrows and long-term PPAs by making a public notification of such policy. Newly privatised distribution companies should not be burdened with the erstwhile escrow arrangements and PPAs. Instead, they should have the freedom to enter into their own voluntary contracts, and develop other routes to reduce energy purchase costs.

2.2 Step 2: Financial restructuring 2.2.1 Retain unfunded liabilities to realise better value The SEBs have a number of existing liabilities, many of which are unfunded, such as pensions. In addition, they have dues to central sector generating units, Coal India Limited, Railways and other such organisations, along with receivables of doubtful quality, including payments from various state government bodies. In order to realise the best value for the business, these liabilities need to be retained by the current owner, i.e., the state government, under whose ownership they have been accumulated. Apart from the moral obligation not to transfer one’s sins onto others, 56 | Indian Infrastructure: Evolving Perspectives it also makes financial sense. In cases where the amount is undisputed, any transfer would result in an equivalent direct deduction from the bid amount. In cases where the liability is uncertain, such as future contractual commitments and pensions, the private sector would have a higher degree of risk aversion, given that they cannot control these risks. Transferring these liabilities to them will only result in a lower bid, which would be reduced by an amount greater than what the government believes to be the extent of the liability. Hence, as part of the Financial Restructuring Plan (FRP) prior to privatisation, these liabilities and assets should be clearly vested with the government, who is the owner of the SEB, and should not be passed on to the newly formed private companies. By retaining them, the government would improve the bid valuation, and can use a portion of the privatisation proceeds to meet the liabilities as described below.

Box 6.1: Privatisation of generation assets of SEBs

Privatisation of generating stations is necessary to ensure that the generation capacity of the SEBs, which is around three fourths of the installed capacity in the country, is utilised to the maximum possible extent. The current lack of appropriate incentives for good management in the public sector affects their investment and operating efficiency. As of date, the PLF (Plant Load Factor) for SEB plants is well below what is possible, even after accounting for the age of these plants. Improving their utilisation will enable an increase in the power supply that would mitigate the effects of the slower addition to new capacity. Generating stations often do not receive regular payments for the energy they supply. Their continued government ownership weakens the commercial environment in the power sector. There is no reason to continue to divert state resources to a commercially viable activity. As the distribution sector is privatised and begins to generate enough resources to permit financially sound arrangements for power purchase, these plants will gain in value. The value of these generating stations could be realised by selling the existing plants and the resources used to reduce the subsidy burden and finance the unfunded liabilities in the sector. Without their sale, it is unlikely that it will be possible to fully finance these liabilities.

2.2.2 Plough-back proceeds of privatisation In order to provide a secure financial source to meet the deferred payments to workers and creditors mentioned above, government should plough back all the proceeds realised by privatising the power sector, including those related to the sale of generation assets (see Box 6.1 above). In this manner, the government will be able to defray part of the liabilities assumed by it, on account of past dues, contractual obligations, pensions, etc. To meet the shortfall, if any, the government should use the low-interest long-term support from multilateral institutions. Indeed the Accelerated Privatisation of Electricity Distribution | 57 appropriate use of such multilateral funds should be for such purposes rather than to finance investment in physical assets, which can be undertaken by the private buyer through the local financial markets.

2.3 Step 3: Separate urban and rural zones for effective privatisation As of now, almost all the reforming states are seeking to privatise distribution zones as a mix of subsidising and subsidised categories of consumers. This leads to long delays as enormous time and effort is spent in configuring distribution zones for privatisation, in achieving the elusive balance between the subsidised and subsidising categories of consumers. It also increases uncertainty, given the lack of information about the distribution system in the non-urban areas. A private entrepreneur looking to buy the zones is concerned that his paying customers will turn to captive plants, driven away by high tariffs needed to raise revenue to meet the ‘social obligation’ of supplying non-paying consumers. In addition, this ‘mixed-zone’ structure is fraught with several limitations (see Box 6.2). By contrast, separation of urban and industrial zones from rural zones can not only result in a faster pace of privatisation; it also addresses the main reason for privatisation of distribution, which is to reduce the extent of theft and inefficiency, as explained below.

Box 6.2: Limitations of the ‘mixed-zone’ structure

A distribution company with ‘mixed’ zones will have two revenue streams, one from consumers, who pay their own user charges and the other from the government in the form of subsidy support, which would be determined by the regulator. The presence of a subsidy-stream provides an avenue for the company to camouflage theft and inefficiency, by over-reporting consumption under subsidised categories, which are often not metered in full and sometimes not at all, as seems to be the current practice in SEBs. Faced with the arduous task of improving the efficiency of collection from a large number of consumers, even the private distribution company may find it a more attractive proposition to try and convince the regulator that larger subsidy flows are called for, instead of expending effort to eliminate theft. Thus, the management of a ‘mixed’ zone company, whether public or private, would have little or no incentive to meter agricultural consumption, since it would take away the ability to camouflage the theft of power. Even if the regulator is aware of this problem, it can only address it through a burdensome exercise of verifying actual consumption through a programme of compulsory metering, which will take a substantial amount of time. Further, if the ‘mixed’ zone distribution company succeeds in its strategy of over-reporting subsidised consumption, the subsidy burden on the government will increase to that extent. A strategy of privatisation through segregated zones will curb this problem. 58 | Indian Infrastructure: Evolving Perspectives

Accelerated loss reduction: Apart from Delhi, even in other places, where there has been a certain degree of substation metering, such as Karnataka and Andhra Pradesh, it is evident that the extent of commercial losses in the urban and industrial areas is very high. Consequently, it can be inferred that pilferage and theft is concentrated in these areas. The privatisation of distribution in these areas along with regulatory stability2 through a medium-term tariff regime, as discussed in the next section, will provide strong incentives to distribution companies to find thieves and make them pay, since the reduction in theft, with stable tariffs, directly enhances profits. Faster pace of privatisation: If urban and industrial areas are offered for sale, the process of privatisation can proceed faster, as the need to discover the elusive balance between the subsidised and subsidising categories of consumers will no longer exist. This must however be done in conjunction with a stable regulatory regime and with realistic expectations, as outlined in our previous analysis of the delay in the privatisation of the Kanpur Electricity Supply Company (KESCO).3 The Request for Qualification (RFQ) can be issued as soon as the zoning is completed, as shown in Table 6.1. Improved revenue realisation and better bidders: The third strong reason to separate urban from rural zones is to improve the sale value for the government. As long as one of the revenue streams is from the government, a private bidder will expect that these payments to be delayed and will accordingly discount their availability, and reduce the value of its bid. By removing the uncertainty associated with subsidy support, the government can realise better value for urban and industrial zones at the time of privatisation. It will also attract companies whose expertise is distributing electricity and not those whose primary competence may be in persuading the government and the regulators to release subsidies.

2.3.1 Effect of segregation on the rural zones and subsidy bill It is now widely recognised that urban theft is often misreported as agricultural consumption, which inflates the amount claimed under agricultural subsidy. Separation of urban zones from the rural areas removes the incentives to camouflage theft and inefficiency under subsidised consumption since the distribution company serving the urban and industrial zones would not have any access to subsidy flows. It allows identification of the actual amount of subsidised consumption and permits a more accurate estimation of theft. The increased detection of theft, as a result of separating urban and industrial zones, would reduce the extent of such fraudulent subsidy claims. Combined with the better estimate of true non-urban consumption, this is likely to reduce the subsidy bill, and thus make it more likely that the state would be able to finance the subsidy. Accelerated Privatisation of Electricity Distribution | 59

The privatisation of urban and industrial areas does not at all imply that rural zones would be worse-off. Stand-alone urban zones will separate those now considered unable to pay, such as the rural poor, from those considered able to pay. In case the government wants to continue subsidies, it can do so through direct budgetary allocations. If the budget still comes up short, government can impose an across the board ‘surcharge’ on electricity consumption of the urban and industrial zones and use the resultant revenues for subsidy payments. The transparency of such a charge will soon generate the countervailing political pressure necessary to ensure that those judged unable to pay are truly incapable of paying.

2.4 Step 4: Ensure stable regulation for risk mitigation Traditionally, around the world, regulation in the electricity sector meant an intrusive cost-based regulation of vertically integrated utilities. In India, this is reflected in Schedule VI, which tries to assure a specific rate of return to the utility, which also involves a detailed and frequent examination of its costs. This regulatory regime is undergoing transformation, as it is now possible to subject more of the sector to competition. In a number of countries today, generation tariffs are completely deregulated, as also the tariffs for large consumers. Tariff regulation remains in place for the use of transmission and distribution wires, as well as for tariffs charged to smaller consumers. In India too, most of the central elements that affect the viability of the distribution business fall under the purview of regulatory commissions. These regulated elements include input prices, tariffs, allowable returns, and expected level of performance in terms of technical and commercial losses. Thus, under the current arrangements, allocation of risk to distribution companies is determined, to a large extent, by the regulatory philosophy and tariff orders of the SERCs. The release of a regulatory philosophy document, being only a statement of intent, only partially meets the objectives of providing information about risk allocation. The realisation of the stated intent is found in the tariff order. For this reason, prospective bidders for distribution zones often request that the process be deferred until the issue of a tariff order by the SERC. However, the Request for Proposals (RFP) can be issued and the process of due diligence can begin as soon as the tariff submissions are made for each separate zones. 2.4.1 Effect of recent tariff orders and risk allocation by the SERCs Almost all SERCs have realised that tariff increases alone are not the answer to solving the problem of SEB viability. Indeed, most SERCs have determined that it would be unfair to shift the burden caused by high T&D losses and low collection to billing ratios, i.e., a lack of efficiency on the part of the utility, completely onto the consumer by raising tariffs. SERCs have also realised that it is counterproductive to 60 | Indian Infrastructure: Evolving Perspectives charge higher tariffs that take the price of grid power well above the cost of generating captive power. Consequently, they have limited the extent of cross-subsidy from industrial and commercial consumers to other consumers, who are charged below the cost of supply. The owners of the SEB, i.e., the state governments, are now required, albeit partially, to come up with the requisite funds from the state budget to compensate the SEB, if they wish to charge tariffs below cost of supply. This direct impact on the budget, in contrast to earlier practice, has made many governments realise the financial unsustainability of insisting on tariffs that recover only a small fraction of the supply cost. This approach has however resulted in a difficult position for prospective private owners, who are now expected to bear the financial burden caused by the high T&D losses, low collection to billing ratios and limited extent of cross-subsidy, until such time as they can bring them down. While this admittedly provides a very strong incentive to the private company to reduce losses, coupled with the regulatory uncertainty caused by annual tariff orders, it also results in a degree of uncertainty that may be severe enough to drive investors away from the sector. It is thus necessary to devise a regulatory regime to balance the need to provide incentives to improve efficiency with the need to reduce the discouraging effect of regulatory uncertainty. This assumes additional importance in the current situation, when the process of transition from public to private ownership is about to take place and where sizable investments are necessary for systemic improvements. In effect, the regulator has twin roles, viz., an oversight role as well as an investment promotion role. It should therefore ensure a stable and sensible regime that will provide an environment conducive to such investments, without compromising the interests of consumers. 2.4.2 The need for medium-term tariff orders This balance is possible by instilling confidence in prospective investors, by giving them a degree of stability with respect to risk allocation. A good way to achieve this is to announce a medium-term, for example, five-year, tariff profile based on an associated loss profile, which provides tariff stability in the years immediately following privatisation, along with incentives to improve efficiency. For this, the government should ensure that a submission for medium-term tariff determination is made to the SERC, separately for each zone. The basis of determining this tariff profile would be the SERC’s judgement about the ‘base case’ for improvement. In arriving at this judgement, regulatory determinations on three variables are critical, viz., 1. The distribution losses that will be incurred. 2. The demand growth that will take place. 3. The investment that will be required. Accelerated Privatisation of Electricity Distribution | 61

While the first two of the above parameters would determine the amount of power purchase costs that the utility will be allowed to recover through the tariff, the third parameter would provide the utility with information about the regulator’s view as to the extent of investment that will be needed. In order to develop a credible ‘base case’, the regulator should ensure that these parameters are pegged at realistic levels. As regards the quantum of distribution losses, the regulator should rely mainly on the implied T&D loss, which is the difference between the amount of energy input into the distribution zone and the amount of revenue collected (deposited in the bank) from the distribution zone. In the existing system, these are the only hard, credible data points, as these are easily verifiable. The regulator and government should avoid spending time and effort in arriving at detailed but unverifiable estimates of a variety of losses, which, in any event, are not likely to be believed by potential bidders. Similarly, considering the decrepit distribution system and the other difficulties emanating from the change of ownership, loss reduction targets should be fixed at reasonable levels—say, none for the first year, followed by increased reductions from the second and third years. Finally, improving performance needs massive investments for system upgradation. These should be estimated and included in permissible investment. The power purchase and investment costs together constitute a large proportion of the controllable costs of the distribution company. Hence, an order specifying these costs provides the investor with a clear signal as to the expectations of the regulator regarding the kind of performance that is expected in order to earn a reasonable return. Based on such a profile over the next five years, a bidder for the distribution business, after due diligence, can arrive at his own judgement with regard to the reasonableness of the tariff profile. If it feels that it can do better than expected by the regulator, it will bid aggressively, in expectation of higher profits. On the other hand, if the regulator’s perspective is perceived as being too stringent, it will bid lower, since he would not expect to make large returns. Here, it is important to note that in case the initial regulatory order is too favourable to the private bidder, a portion of this will be recaptured in the higher price paid by the winning bidder, and thereby prevent excessive profits. Second, and more important, the returns to the investor are conditional on good performance. The investor makes more money only if he does better than expected, in terms of reducing losses, undertaking investments efficiently and increasing demand, e.g., by attracting high-tension users back to the grid. It is critical to acknowledge and reward such performance by assuring the investor that he would be able to retain such performance-linked driven profits. These returns are the very basis of the accountability that privatisation of distribution seeks to induct. 62 | Indian Infrastructure: Evolving Perspectives

Tariffs based on distribution value-added (DVA) charge: In case the regulator is reluctant to provide a medium-term profile of tariffs for want of adequate information, it could separate the costs associated with distribution services—that is, distribution value-added (DVA)—which is essentially a charge for using the distribution wires, and announce a medium-term profile only for that component. Under this arrangement, the remaining costs related to power purchase and transmission could be allowed as a pass-through, after appropriate regulatory scrutiny. Such a system would be compatible with an eventual structure of the sector, where a competitive power market determines power purchase costs. In the interim, this system would insulate private distribution companies from the vagaries of the bulk power market.4

2.5 Step 5: Market structure: avoid the single-buyer model The value of the various businesses in the power sector depends on the future cash flows that they generate. The extent and sustainability of these cash flows would depend critically on the market structure that is adopted. Given the international trends and the pointers in the proposed Electricity Bill 2000, India too will eventually witness the emergence of bulk power markets, merit-order dispatch, choice to consumers and open access to the transmission and distribution wires. Each of these is essential for the success of privatisation, as they form the very basis for competition. The single-buyer model is often sold as a transitional arrangement, ‘justified’ in order to maintain uniform prices across the state and form one agency responsible for honouring the existing PPAs entered into by the state. However, apart from its many disadvantages (see Box 6.3 below), its adoption instead increases the likelihood that, under pressure from vested interests, the next step toward liberalised electricity markets will be indefinitely delayed.5 This approach restricts distribution companies and large consumers from buying from the generators of their choice. Such restrictions reduce the scope for efficiency gains in the short term and vitiates the process of developing a healthy power sector in the long run.

Box 6.3: Disadvantages of the single-buyer model

The single-buyer model has major disadvantages, particularly in situations with high degree of corruption and low payment discipline. First, and most importantly, the single-buyer model weakens the incentives for distributors to collect payments from customers. The state-owned single buyer is often reluctant to take politically unpopular action against a delinquent distributor, and its aggregation of cash proceeds from distributors allows it to spread the shortfall caused by a poorly performing distributor among all generators. When distributors see that paying and non-paying distributors are treated alike, their motivation for cutting off non-paying customers weakens across Accelerated Privatisation of Electricity Distribution | 63

the board, vitiating the very reason for privatisation. Second, decisions about adding generation capacity are not made by distribution companies, who would have to bear the financial consequences of their actions. Instead government officials and ‘experts’, who do not have a direct commercial stake, have to make these decisions, and often find it difficult to resist powerful interest groups pushing for state- guaranteed capacity expansion. Third, PPAs create a contingent liability for the government, which is expected to step in if the state-owned transmission company is unable to honour its obligation to the generator. Loading the state-owned transmission company with all the agreements when it is hardly equipped to take on the risk arising from failure of the PPAs often leads to its financial ruin, as in the case of GRIDCO in Orissa. To the extent that these PPAs are a net cost, it is better that they be addressed directly rather than be hidden under the wraps of the state-owned transmission company. Fourth, the single-buyer model responds poorly when electricity demand falls short of projections (such as the recent experience with Dabhol). Ideally, electricity prices should fall, stimulating demand.6 Under the single-buyer model, however, wholesale electricity prices rise because take-or-pay quotas (or fixed capacity charges) must be spread over a shrinking volume of electricity purchases. When these high prices cannot be passed on to final consumers, taxpayers must bear the losses. Fifth, the single-buyer model hampers the development of electricity trading across state borders, by leaving it to the single buyer, a state-owned company without a strong profit motive, to enter into such arrangements. Based on The Single-Buyer Model: A Dangerous Path Toward Competitive Electricity Markets by Laszlo Lovei, Public Policy for the Private Sector Note No. 225, the World Bank, December 2000

2.5.1 Effect on distribution privatisation The choice of market structure has an impact on the cashflows of the distribution business. For example, a distribution business that enjoys a local monopoly, i.e., whose consumers cannot buy their power from other suppliers, would have a different valuation as compared to a business whose consumers have the option of purchasing from alternative sources. Similarly, a distribution company that can source its power requirement from a competitive bulk power market would have a different valuation as compared to one that has to buy from a single high cost supplier. Hence, the government should spell out the timing of introduction of the bulk power market as well as the choice to consumers at the time of privatisation. This will enable investors to factor in risks arising out of these developments into their bids. An illustrative schedule for introducing choice to consumers could be as shown below. Since permitting such transactions would require appropriate improvements in the areas such as billing software, interconnection and metering, the responsibility for such expenditure needs to be clearly and appropriately allocated between the distribution company and the transmission company. 64 | Indian Infrastructure: Evolving Perspectives

Illustrative time frame for introducing choice to consumers 1 year 3 years 4 years 5 years 7 years Above 10 MW Above 5 MW Above 3 MW Above 1 MW All

2.5.2 Own-generation and vesting contracts As noted before, the proposed Electricity Bill 2000 envisages open access to the network and the development of a bulk power market in the medium term. In a power market, there is usually a single market price for power, which may vary over the day—from hour to hour or at even finer intervals. In such a situation, the transmission company is responsible only for the transmission of power and not for bulk purchase of power for onward supply to distribution companies. Different distribution companies thus face the same price for power purchased from the market.7 However, in an environment of power shortage, either through the lack of sufficient capacity or due to transmission constraints, distribution companies are apprehensive about being held to ransom by generation companies and need assurance of power supply. The bargaining power of the generator is higher in the short term when the distribution company needs energy to meet its obligations and has little recourse to alternatives. In order to mitigate this uncertainty, distribution companies should be allowed to set up their own generation capacity or be sold bundled together with an existing generation facility. Alternatively, distribution companies may, at the time of privatisation, be ‘vested’ with a contract that allows them to source power from a designated generation facility, at a pre-specified price, and for a relatively short period such as five years. Such a vesting contract enables the distribution company to have contractual access to the energy produced by the designated generator in the immediate aftermath of privatisation. It also enables easier sale of the generating station by assuring that it has a ready off-take in the initial years.

2.5.3 Caveat on market structure choices While making market structure choices, care needs to be taken to ensure that the timing of these choices is realistic. Excessively zealous targets would only harm the process of privatisation. Understandably, private investors would be extremely reluctant to make the massive investments required in system improvement today, if their consumer base dents too soon and if they are not sure about the terms of open access. Similarly, one should avoid a mindless pursuit of the ‘unbundling’ doctrine, especially with regard to the existing, well-functioning private licensees, as we might end up wrongly ‘fixing’ what is not broken in the first place. In order to prevent such errors, the proposed Electricity Bill 2000 should permit the existing private licensees to maintain their status quo. Accelerated Privatisation of Electricity Distribution | 65

2.6 Step 6: Actual sale of the distribution business At the conclusion of Steps 1 to 5, the following should be in place: • The SEB has been unbundled into different generation units, a transmission company and different distribution zones, where urban and industrial zones have been separated from non-urban zones. • All commitments arising out of the previous PPAs, including escrow arrangements, are vested with the government and are isolated from the process of distribution privatisation. • All unfunded liabilities such as dues to power and fuel suppliers and employee pension commitments are retained with the government and arrangements are put in place to plough-back the proceeds from privatisation to meet these commitments. • A medium-term tariff order from the regulator, specifying a tariff schedule for the next five years, based, inter alia, on a benchmark profile of reduction in T&D losses. • The government, in consultation with the regulator, has instituted a mechanism to meet subsidy commitments through explicit budgetary allocations and, if necessary, through imposition of a transparent ‘surcharge’. • A time frame for open access to the transmission and distribution company wires has been announced.

2.6.1 Multiple-round ascending auction for urban and industrial zones At this stage the zones are ready for bidding. As noted above, the process of sale and due diligence can and should begin earlier. It is however important that the bidding for the zones be conducted appropriately. While it is not the intent of this paper to go into details, it should be noted that a single-round sealed bid auction is not necessarily the best option for bidding, since it does not generate sufficient information. An alternative method is the multi-round ascending auction. In this system, the bidders submit a sealed bid, which is then opened for all bidders. All bidders then have the option of revising their bid upward in the second round and resubmitting their bid. This method has recently been proposed by the Telecom Regulatory Authority of India (TRAI) for auctions to allocate the fourth cellular license. This method makes sense, especially in the current situation with paucity of information, because with each round of bidding new information will be revealed. While each bidder has its own estimates about the kind of investment that will be required and loss reduction that would be possible, all bidders are also trying to estimate the growth in the market size. The bids of the other bidders give each 66 | Indian Infrastructure: Evolving Perspectives bidder some information about their estimates of the growth in the market size, which in turn affects its own estimates in this regard. However, it is important to note that this may not be true in all situations.8

2.6.2 Minimum subsidy bidding for non-urban zones As regards non-urban zones, private sector efficiency in delivery can be exploited by bidding out subsidised zones for supply by private companies and cooperatives, with stringent but realistic improvements in performance parameters as regards quality of supply, as long as they demand lesser subsidy than what it currently costs the SEB. This will lead to a further reduction in the subsidy bill. More importantly, isolating these zones would help government in adopting a radically different approach to non-urban supply, which has its own special needs.9 Such interventions, for example, could range from allowing distribution through alternative ownership structures such as co-operatives to promoting appropriate, decentralised generation options such as photovoltaics and micro-turbines.

3. CONCLUSION In order to be successful, as well as politically feasible, it is essential that consumers benefit from the process of power sector reform. Concomitantly, the private sector must be able to earn adequate returns, commensurate with the risks of the distribution business. Most importantly, this process must take place quickly. This paper seeks to provide a set of six specific steps for accelerating this process, while balancing these two objectives. The first step is to notify a policy eschewing new escrows and long-term PPAs. They only add to the troubles of the SEB and make it even more difficult to undertake distribution privatisation, which is a fundamental reform. The second is to retain the existing liabilities with the current owner, i.e., the state government. These include existing financial and contractual commitments, e.g., past dues to the suppliers of power and fuel and existing long-term PPAs and, most importantly, deferred payments to workers, such as pensions. The proceeds realised by privatising the power sector, including the sale of generation assets, should be ploughed-back to meet these retained liabilities. The third step is to separate the distribution segment of the SEB into distinct urban and non-urban distribution businesses to obtain full benefit of private participation and quickly isolate and curb the rampant theft and pilferage in urban areas that is currently camouflaged as subsidised consumption. As opposed to the current practice of hidden cross-subsidy between consumers, government should make explicit budgetary provisions to meet subsidy commitments and, in case of shortfall, it may impose a transparent ‘surcharge’ on urban and industrial zones. The fourth step is to ensure that the regulatory regime Accelerated Privatisation of Electricity Distribution | 67 provides sufficient incentive to private entrepreneurs by allowing it to retain profits earned as a result of improvements in efficiency and bearing of risk. This is best done through a medium-term, rather than an annual, tariff determination process. The fifth step is to avoid the single-buyer model and spell out the market structure in terms of the timing of introducing markets for bulk power and choice to consumers. In order to contain the possible market power of generation companies owing to supply shortages in the short run, private distribution companies should be allowed to set up their own generation capacity or sold bundled together with an existing generation facility. Alternatively, they may be ‘vested’ with supply from a designated generation facility, at a specified price for a short period of five years. Finally, the urban and industrial distribution zones should be sold through a process of multiple-round ascending bid auctions. The non-urban or subsidised zones should also be privatised on the basis of a minimum subsidy bid, through a process designed to attract co-operatives, in addition to large and small private entrepreneurs. This entire process can be accomplished within a year, as shown in Table 6.1. As these steps are intrinsically inter-related, employing them selectively will not yield the intended results. However, taken together, they have the potential to transform our ailing power sector in a short span of two years. Further delay will only prolong the agony and lead to a slowdown in our overall economic growth. Either we act now or face more blackouts. Frankly, the choice couldn’t be starker! 68 | Indian Infrastructure: Evolving Perspectives nal bid privatise is ◆ is State Electricity Regulatory

◆ ◆ SERC ◆ ◆

◆ ◆ is Request for Proposals; RFP

◆ *

*

*

*

is Request for Qualification; *

RFQ *

*

Table 6.1: Timeline for accelerated privatisation of distribution *

is Expression of Interest; EOI

6h. Auctions (1 month) 4a. Preparing the tariff submission (4 months) 4b. Medium term tariff submission (7th month) 4c. Tariff examination by SERC (3 months) 4d. Medium term tariff order (10th month) 6a. EOI (1st month) 6b. Issue of RFQ (3rd month) 6c. Submission of RFQ (5th month) 6d. Evaluation of RFQ (2 months) 6e. Issue of RFP (7th month) 6f. Due diligence by bidders (4 months) 6g. Final bid preparation (1 month) announced. The RFQ can be issued when the zoning is decided and RFP tariff application submitted. fi preparation can begin as soon the medium-term tariff order is issued. Steps and sub-activities1. Policy preparation (1 month) 2. Financial restructuring (6 months) 3. Distribution zoning (3rd month) 4. Stable regulatory regime 5. Market structure: decision on open access (6th month) 6. Sale of distribution business 1 2 3 4 5 6Note: 7 8 9 10 11 12 Commission. * The sale process proceeds in lock step with the previous steps sequence. EOI can be issued as soon policy to Accelerated Privatisation of Electricity Distribution | 69

NOTES 1. The disadvantages of the single-buyer model are further elaborated upon later in the paper, in Section 2.5. 2. Under a medium-term tariff order, part of this recovery of stolen power by the private distribution companies and the associated increase in revenue would already have been accounted for by the SERC while deciding consumer tariffs and the benchmark T&D loss reduction profile. Additional reduction in T&D losses, over and above this benchmark, would thus result in higher profits for the distribution company, which is precisely the incentive for the distribution companies to control theft. 3. See Privatisation of Kanpur Electricity Supply Company (KESCO)—A Case Study, prepared by Infrastructure Development Finance Company Ltd, December 2000. 4. Given the perception of supply shortages, one can reasonably assume that a bulk power market would witness several price spikes in the initial phase of its operation. 5. Prima facie, while it is possible to de-license generation and institute a merit order dispatch regime almost immediately, it may not be possible to implement open access provisions as quickly, as it requires a reasonably reliable transmission and distribution network and a transparent energy accounting system for its implementation. 6. Revenue losses would be borne by private financiers, who are best equipped to manage market risks. 7. This is not strictly true, since there may be other products that are traded, such as reserve power. In addition, it may differ at different points of the grid, in case there is a nodal system for transmission pricing, which attaches different prices for transmission services when off-take is from different points of the grid. 8. For example, this method does not make sense for a situation where the bid does not depend on others’ information, e.g., when a construction contract is awarded. In this case, each bidder knows the design and physical inputs required, and forms its own estimate about the cost. Knowing the bids of other bidders provides no new information that can lead to a revision of the bid. 9. Even in the mecca of capitalism, the USA, literally hundreds of cooperatives and small companies are responsible for most of the rural electric supply. 70 | Indian Infrastructure: Evolving Perspectives

REGULATION OF PETROLEUM PRODUCT 7 PIPELINES February 2003

1. INTRODUCTION The New Exploration Licensing Policy (NELP) introduced in 1997 has given rise to an increase in proven reserves of domestic gas. The exploration of crude oil too may turn out to be fruitful. The oil and gas blocks auctioned under NELP I and NELP II have struck gas in significant quantity in the Krishna– Godaveri Basin (K–G Basin) and some private companies have found gas onshore in Rajasthan and Gujarat. Further auctioning of exploration blocks is on the anvil (Box 7.1: New Exploration Licensing Policy). The Directorate General of Hydrocarbons—the regulator of the upstream hydrocarbon sector—has stated on record that there is ample geological evidence that India’s hydrocarbon reserves can increase by two to three times.1 A whole host of technical innovations—including 3D seismic imaging, horizontal drilling, deepwater platforms, multi-phase pumps, and tertiary recovery techniques—have enabled oil companies to increase exploration success, reduce development costs, boost recoveries, cut downtime, and improve access to new exploration areas. A pipeline is a safe, convenient, reliable and environment-friendly mode of transport for bulk liquids. Several developments in the last fifty years, such as the use of ‘pigs’ to clean the interior of pipelines, the use of ‘batching’ to transport different petroleum products through the same pipeline, the use of cathodic protection to reduce corrosion of pipelines, and the use of computers and communication technologies to monitor and control pipeline operations, have seen pipelines emerge as the preferred mode of transport for large volumes of petroleum products. It is generally argued that in the relevant economic markets, in which product pipelines seek to meet transportation demand, there is no Regulation of Petroleum Product Pipelines | 71 effective competition; hence, regulatory restrictions upon the structure and conduct of firms are required to reap benefits for society. New refineries built after the deregulation of the sector are currently able to meet the oil product demand of the country. The basis of industry structure of crude/oil product pipelines has been primarily to decongest surface transport and to complement rail transport. The access control and tariff applicable to pipeline transportation of oil products is a result of centrally planned development of refinery capacity, marketing and transportation of oil and oil products. Box 7.1: New Exploration Licensing Policy

The widening gap between domestic production and consumption of hydrocarbon products has put a serious strain on the Indian economy. Keeping in view the rising import bill of Petroleum Oil and Lubricants (POL), GOI has taken several initiatives to increase the pace of exploration and secure increasing participation of the private sector

in exploration and production of oil and gas. India has around 3.14 million sq km of sedimentary area in 26 basins. Only one-third of this area has been explored so far. The ratio of reserves to production of oil is 17.8 while that of natural gas is 24.5 [BP (2002)]. Efforts have been made by the national oil companies and by the Directorate General of Hydrocarbons (DGH) to open up unexplored areas by providing detailed seismic data. The government liberalised the terms of offer for exploration blocks by unveiling the NELP in 1997. Apart from better terms offered to bidders, an outstanding feature of the new policy is that it offers a level playing field to national oil companies and private sector participants. The new terms offered by the government include the following: • The possibility of seismic option in the first phase of the exploration period • No minimum expenditure commitment during the exploration period • No signature, production, or discovery bonus • No mandatory state participation • No carried interest by national oil companies • Income tax holiday for seven years from start of commercial production • No customs duty on imports required for petroleum operations • Biddable cost recovery limit up to 100 per cent • Option to amortize exploration and drilling expenditures over a period of 10 years from first commercial production • Sharing of profit petroleum, based on pre-tax investment multiple achieved by the contractor and biddable • Freedom to contractor to market oil and gas in domestic market Since 1997, three rounds of NELP have been made, and 22, 23 and 23 blocks have been allocated under NELP I, II and III respectively. In March–April 2003, another 23 blocks 72 | Indian Infrastructure: Evolving Perspectives

are likely to be offered under NELP IV. Investments of around US$340 million (Rs 16.25 billion) were made during the first phase of exploration in NELP I blocks till September 2002. This exceeded the envisaged expenditure of US$250 million (Rs 12 billion). In the second and third phase, the expenditure estimated is about 2 US$913 million (Rs 43.8 billion). Allocation of exploration blocks under NELP I, II and III Company/consortium NELP I Reliance Industries & Niko Resources 12 ONGC 5 IOC & ONGC 2 ONGC & GAIL 1 OIL 1 Cairn Energy 1 Total 22 Company/consortium NELP II ONGC 6 ONGC & consortium 10 Reliance Industries & Hardy Oil 4 Niko Resources 1 OIL 1 GSPC–GAIL–Joshi Tech. 1 Total 23 Company/consortium NELP III ONGC & consortium 13 Reliance Industries & Hardy Oil 9 GSPC consortium 1 Total 23 Sources: CMIE and media reports Unlike oil products, pipeline is the only mode of transportation for large quantities of natural gas. The Hazira–Bijaipur–Jagdishpur (HBJ) pipeline—the only natural gas transmission pipeline in the country—is owned and operated by Gas Authority of India Limited (GAIL). The gap between the supply and demand of gas, estimated to be 55 metric million standard cubic metres per day (mmscmd) in ‘Hydrocarbon Vision 2025’, and the exhaustion of existing gas resources, situated in the western oil Regulation of Petroleum Product Pipelines | 73 fields, led the government to provide fiscal incentives to import liquefied natural gas (LNG). This in turn guided the development of LNG terminals in the country to supply regassified LNG through the HBJ pipeline. Natural gas from Bangladesh gas fields, using a trans-national pipeline, will become a reality soon.3 With substantial gas findings in the K–G Basin and a number of LNG facilities being established in the country, many issues related to gas pipelines are coming to the forefront. Unlike industry nomenclature to use natural gas distinctly and have separate regulation for it, GOI has chosen to define petroleum products to include natural gas and all products derived from it (Box 7.2: Definition of petroleum products).

Box 7.2: Definition of petroleum products The definition of petroleum includes natural gas and refinery gas in the Petroleum Act, 1934 (35 of 1934). The same definition is adopted in the Petroleum and Minerals Pipelines (Acquisition of Right of User Inland) Act, 1962. The Petroleum Regulatory Board Bill 2002 also includes natural gas in the definition of petroleum and the expression ‘petroleum product’ means any product manufactured from petroleum. Generally, natural gas is recovered from gas lakes along with other hydrocarbons, water and sand. Natural gas is extracted from this and refined to remove impurities like water, other gases and sand. In its pure state, natural gas is a complex hydrocarbon vapour. This vapour can be separated or ‘fractionated’ into several components that may be utilised as fuels or as raw materials for other products. In India too, C1, C2, C3, C4 and C5 fractions are known by their common names—methane, ethane, propane, butane and other fractions respectively—and used as fuel, feedstock for urea plants and fuel for power plants, production of petrochemicals, LPG and industrial fuels and solvent respectively. To all intents and purposes, legal definition of petroleum products pipeline is inclusive of natural gas pipeline.

The purpose of this paper is to critically evaluate the demand and supply scenario of oil products and gas emerging in the wake of the deregulation of the sector and to suggest appropriate regulation of the pipeline industry for the benefit of consumers. Our methodology is to focus on the economic characteristics of the pipeline industry and their implications for the design of efficient regulatory policy. Demand for pipeline capacity being a derived demand from the demand and supply of oil products and gas, regulatory framework should be convergent with the regulatory framework of oil and gas.

2. DEMAND AND SUPPLY OF PETROLEUM PRODUCTS ‘Hydrocarbon Vision 2025’ [GOI (2000)] has played a decisive role in the development of the hydrocarbon sector in the last few years. Its aim was to have a free market and promote healthy competition among players and to improve customer services. 74 | Indian Infrastructure: Evolving Perspectives

2.1 Demand for oil products Almost all analysis starts with the demand estimates (up to 2025) of the Hydrocarbons Group. The report suggested a large gap in the demand and supply of oil products and assumed an income elasticity of one.4 Table 7.1 gives the supply/ demand estimates given in ‘Hydrocarbon Vision 2025’. Almost all official documents and expert analyses since 2000 refer to these demand estimates.

Table 7.1: Supply/demand—petroleum products (in mmt) Year Demand (without Demand (with Estimated Estimated meeting gas meeting gas refining crude deficit) deficit) capacity requirement 1999–2000 91 103 69 69* 2001–2002 111 138 129 122* 2006–2007 148 179** 167 173 2011–2012 195 195*** 184 190 2024–2025 368 368 358 364 * Estimates given in the report. Actual crude oil consumption was 86 and 107 mmt respectively. ** Assuming 15 mmtpa of LNG import by 2007 (15 mmtpa = 57 mmscd) *** Assuming adequate availability of gas through imports and domestic sources Source: Report of the Group on ‘India Hydrocarbon Vision 2025’ (2000) The report asserted that the gap will have to be met through imports and an increase in domestic production from the existing oil and gas fields. The report did not assume that the success of the NELP would increase supply of crude oil and gas in the country. India has installed refining capacity of 115 million metric tonnes per annum (mmtpa) and another 10–15 mmtpa is likely to come on stream by December 2004. Detailed estimates of oil products consumption suggest that India will be able to meet all its oil product consumption needs from domestic refineries (Table 7.6: Projections from POL consumption). In FY 2001–02, India consumed 100.4 million metric tonnes (mmt) of oil products and had a surplus of 3.8 mmt oil products which were exported. In FY 2003, domestic consumption of oil products is estimated to be 103 mmt with a surplus of 5 mmt. The refining capacity is expected to increase to 162 mmtpa by 2008. The demand, on the other hand, is estimated to grow to 128 mmtpa over this period. Refining margin in India is US$2–3/bbl compared to under a dollar margin elsewhere. Hence, refineries in India would prefer to satiate domestic demand before exporting5 [ICRA (2003)]. Historically there has been a strong correlation between Gross Domestic Product (GDP) growth rate and POL consumption due to the large share of the industrial and agricultural sectors. The slowdown in the industrial and agricultural sectors and the Regulation of Petroleum Product Pipelines | 75 phenomenal growth of the services sector, especially IT-related services, have changed the composition of the GDP pie. For the past two years, GDP has been growing at 5–6 per cent and POL consumption is growing at less than 2.6 per cent per annum. 12

10

8

6

4 % growth rate 2

0

1990–911991–921992–931993–941994–951995–96 1996–971997–981998–99 2000–01 1999–2000 2001–02*2002–03* POL growth (%) GDP growth (%) Figure 7.1: Growth of GDP versus POL consumption Sources: CMIE and MoPNG (various issues) *Quick estimates Figure 7.1 suggests that the demand elasticity of POL consumption has declined rapidly, and in the first half of the fiscal year 2002–03, the demand of POL consumption is likely to grow at less than 2.5 per cent compared to a GDP growth of 4.4 per cent. The demand estimates for oil products and for gas provided by ‘Hydrocarbons Vision’ are far more optimistic than what the economy can absorb. Many LNG suppliers, who rushed to construct LNG terminals and regassification facilities, have shelved their plans in the absence of demand from paying customers, such as independent power producers who abandoned their projects (see Table 7.7: List of LNG terminals).

2.2 Regional demand and supply of oil products The total demand of POL products is not an appropriate indicator of the pipeline capacity required for different products and of the competitive pressure faced by the pipeline industry. The demand for petroleum products is largest in the north and in the west due to the extent of industrialisation and the higher concentration of vehicles on the road. As against the above demand pattern, the maximum supply of products comes from refineries situated in the western and southern regions. 76 | Indian Infrastructure: Evolving Perspectives

50 45 40 35 30 25 20

% demand/supply 15 10 5 0 North East West South % demand % supply Figure 7.2: Aggregate supply/demand POL 2001–02 Source: India Infoline estimates Figure 7.2 shows that the northern region is in a state of deficit, the southern and eastern regions are barely in surplus, and the western region alone has substantial surpluses.6 The mismatch of regional demand and supply requires modes of transport to bridge the demand and supply gap. The regional disparity is likely to continue until refineries in the northern region are able to expand their capacity and a new refinery in the northern region gets constructed.7 The implication for oil product pipelines is that pipeline capacity may not be fully utilised at all times and that demand for oil product pipeline capacity may shrink even though demand for petroleum product increases in the region. 2.3 Demand for natural gas The natural gas market in India is supply constrained and the latent unmet demand for gas has attracted unprecedented attention from the LNG industry after it was provided fiscal incentive in Budget 2002. Natural gas in India is an intermediate industrial product, mainly used by power generating plants and fertilizer plants. Table 7.2: Supply/demand—natural gas (in mmscmd) Year Demand 1999–2000 110* 2001–2002 151* 2006–2007 231 2011–2012 313 2024–2025 391 *Estimates given in the report. Actual consumption was 73.6 and 76.8 mmscmd respectively. Source: ‘Report of the Group on India Hydrocarbon Vision 2025’ [GOI (2000)] Regulation of Petroleum Product Pipelines | 77

‘Hydrocarbon Vision 2025’ suggested a surge in the demand for natural gas on the assumption that the power sector would be a large consumer of gas and that the fertiliser industry would substitute natural gas as raw feedstock in place of naphtha after the deregulation of gas prices in January 2003 (Table 7.2). It asserted that the gap will have to be met from imports, an increase in domestic production, and by switching to liquid fuels. As against the estimated demand, the present domestic gas supply and consumption is 76.8 mmscmd. 12,000

10,000

8,000

6,000

4,000 million cubic metres

2,000

0 1990– 1991– 1992– 1993– 1994– 1995– 1996– 1997– 1998– 1999– 2000– 91 92 93 94 95 96 97 98 99 2000 01

Power Fertilisers Others Figure 7.3: Main consumers of natural gas Source: CMIE The two biggest consumers of natural gas are power plants and fertiliser plants. Approximately two-thirds of the natural gas produced in the country is consumed, in equal proportion, by the power sector and by the fertiliser sector, and the remaining one-third is used by domestic consumers, other industries, the transport sector and by the gas pipeline industry itself. The proportion of other users has been rising steadily through the nineties (Figure 7.3). It is expected that gas use would spread wider into the residential, commercial, small industrial, and transport sectors [GOI (2002a)]. The spatial analysis of gas-based fertilizers and power plants suggests that there is a shortage of natural gas in western India, while the demand for gas on the eastern side is minimal, as there they use coal of which there are large proven reserves in the region. The demand for gas may not pick up if investment in the power and fertiliser sectors does not materialise due to delay in sectoral reforms. In the ‘Report of the Sub-group on Natural Gas Availability’, the MoPNG has re-emphasised that the power sector and fertiliser demand would drive the gas demand in the country in the foreseeable future. Given such uncertainties of gas demand, 78 | Indian Infrastructure: Evolving Perspectives the official estimates of gas demand vary from 135 mmscmd to 231 mmscmd— a large variation for any investor to take the financial risk of investment in gas pipelines (Table 7.3).

Table 7.3: Gas demand estimates of different government agencies Terminal year of plan 2006–07 Gas linkage committee allocations + potential demand by existing market 180 mmscmd Hydrocarbon Vision 2025 231 mmscmd ADB’s Gas Master Plan 185 mmscmd Initial assessment by user ministries + other sectors 135 mmscmd Source: GOI (2002) Many prospective investors in the power and fertiliser sectors have postponed their investment in these sectors because major policy issues concerning their viability are yet to be sorted out.8 In addition to policy risk, gas demand is sensitive to the delivered price and the price regime for competing alternatives. In the case of fertilizers, there could be a ‘make’ versus ‘buy’ option available, and there are strong opinions for and against the ‘buy’ option. The overall economics of the domestic production of fertilizers, fertilizer pricing, and other such factors, would therefore need to be addressed in the context of the deregulation of oil and gas prices [GOI (2002)]. In ‘Hydrocarbon Vision 2025’, large gas requirements have been projected for the fertilizer sector to augment agricultural production to ensure food security. In ADB’s study on the ‘Gas Master Plan’, it was observed that it would be relatively cheaper to buy fertilizer and/or manufacture it abroad and import to India, rather than manufacture it domestically, based on the high prices of imported gas [GOI (2002a)]. The demand of gas from the fertilizer industry after the dismantling of subsidies on urea may get stunted if the import of urea is allowed freely. The demand from other industries depends on the long-term price of gas at the factory gate and not at the land-fall price. Many industries can substitute fuel oil and electricity with gas as heating media, only when it is economical.9 The aforesaid discussion shows that natural gas is mainly an industrial product and the future demand of natural gas depends crucially on gas price and on the removal of subsidies in the fertilizer and power sectors. Gas pipeline capacity, being a derived demand from the gas demand, need not show a secular increase in demand as suggested by various government agencies and face unquantifiable demand and market risk. Regulation of Petroleum Product Pipelines | 79

2.4 Supply of natural gas At present, Oil and Natural Gas Corporation of India (ONGC) and Oil India Limited (OIL) are the main producers of natural gas in India, accounting for almost 98 per cent of the gas produced in the country. GAIL is a dominant player in the transportation and distribution sector, accounting for over 95 per cent of the gas sold in the country. Gas production in the country stood at 27.9 billion cubic metres (bcm) in 2000–01. The three main producing basins in the country—the western offshore region, the Cambay basin in Gujarat, and the Upper Assam region—are in the mature phase of exploration. As per the projections of the Sub-group on Utilisation of Natural Gas constituted under Hydrocarbon Vision 2025, domestic gas availability is expected to decline to about 16 bcm by 2011–12. Reserves at ONGC’s Gandhar fields are also declining. ONGC has informed large consumers that it may not be able to meet their entire requirements. Concerned by the growing gap between demand and supply of gas, the government waived the countervailing duty on LNG in Budget 2002 in order to encourage the building of LNG terminal facilities. Several public sector units and private companies have proposed to construct LNG terminals. An estimated 37.5 million metric tonnes per annum (mmtpa) of LNG terminal capacity has been planned (Table 7.7: List of LNG terminals). The first LNG Terminal at Dahej, owned by Petronet LNG, a PSU under the administrative control of the MoPNG, is expected to become operational in the year 2004. A couple of more LNG terminals may become operational in the year 2005. In view of the proximity to enormous gas reserves in Iran, Qatar, Bangladesh and Burma, the option of pipeline gas is economically superior to that of LNG imports.10 However, transnational gas pipelines hinge critically on the geo-political situation and hence, the import of gas using pipelines is still a few years away. Keeping these constraints in mind, the MoPNG estimated the potential gas supply by 2006–07 to be approximately 140 mmscmd. Their estimates of gas supply potential between 2002–03 to 2006–07 is given in Table 7.4.

Table 7.4: Total gas supply potential—Tenth Plan (in mmscmd) Source 2002–03 2003–04 2004–05 2005–06 2006–07 Domestic (firm) 71.99 75.73 81.53 78.89 79.80 Domestic (firm + possible) 71.99 77.13 89.67 87.47 85.76 LNG imports – – 20.00 40.00 50.00 Gas imports – transnational gas pipeline – – – – 10.00 Total (domestic firm + imports) 71.99 75.73 101.53 118.89 139.80 Total (domestic firm + possible + imports) 71.99 77.13 109.67 127.47 145.76 Source: GOI (2002a) 80 | Indian Infrastructure: Evolving Perspectives

But, the possibility of gas supply from domestic sources has changed considerably after the gas findings in the Krishna–Godavari Basin. The in-place volume of natural gas in D-6 Block of the Krishna–Godavari Basin has been estimated to be in excess of 7 trillion cubic feet (tcf). The recoverable reserves have been estimated to be over 5 tcf. Once production begins in two years’ time, 40–50 mmscmd of natural gas will be added to the current domestic output; in 10 years, production is expected to go up to 100 mmscmd.11 An estimated production of 100 mmscmd in two years’ time and 150 mmscmd in ten years’ time can only be absorbed if thermal power plants substitute natural gas in place of coal as the primary fuel. On the basis of public announcements of the exploration companies and DGH’s confirmation, we estimate the potential gas supply by the year 2006–07 to be as high as 179 mmscmd (Table 7.5). These estimates have not taken into account the gas findings in the Bassein (Vasai) gas field, the K–G Basin, and in Rajasthan by ONGC and Cairn Energy because these gas reserves have not yet been certified by a competent authority.

Table 7.5: Total gas supply potential post K–G Basin find (in mmscmd) Source 2002–03 2003–04 2004–05 2005–06 2006–07 Domestic (firm) 72 75 81 78 79 Domestic (firm + possible) 72 77 129 129 129 LNG imports – – 20 20 50 Total (domestic firm + possible + imports) 72 77 149 149 179

Another development which is going to have a long-term impact on the supply of oil and gas is the development of proven reserves of Sakhalin oil and gas fields. These reserves were found in 1970 and are being developed now. Oil majors such as BP and Royal Dutch/Shell group are developing these oil and gas fields.12 Sakhalin oil and gas are going to be important sources of energy for Japan, South Korea and China. An LNG facility at Sakhalin, having a capacity of 9.6 mtpa, will be the nearest LNG plant to the Japanese, Korean and Chinese market. ONGC Videsh Limited holds a 20 per cent stake in the Sakhalin-1 project. Production of oil and gas from these fields is expected to start in 2005 and in 2008 respectively.

2.4.1Gas prices The business of gas has three important elements—production of gas, consumption of gas and pipelines connecting gas fields to consumers. These elements are bonded by ‘delivered’ price of gas to consumers. The price of gas paid by consumers, in turn, has three components—price of calorific value of gas, cost of carrying gas from gas fields to consumers including production cost, and statutory levies and Regulation of Petroleum Product Pipelines | 81 local taxes. It is because of the inflexibility of pipelines as a means of transportation that investment in pipelines is a sunk cost and pipelines are laid only when producer is assured of gas off-take over a long period of time. The necessity of upfront large amount of money sunk in gas pipelines to enable gas to reach to consumers has led to long-term contracts of 25–30 years’ duration between gas producers and consumers. On the other hand, presence of gas producers in the entire value chain, i.e. exploration and development of gas fields, LNG plants, LNG ships and LNG degassifcation facilities at consumers’s end has led to reduction in contract period and development of spot market in LNG. So far, industry practice has been to link LNG prices to crude oil prices. Japan being the largest consumer of LNG, reference price for LNG has been determined with respect to Japanese Cocktail Crude (JCC) index as Japan used regassified LNG to generate electric power. The dynamics of crude oil prices is based on revenue maximisation of OPEC countries and a veiled attempt to dampen development of alternative energy source such as natural gas (Appendix B). It should be noted that the LNG price in Japan has been substantially higher than the US gas prices because shipping and insurance cost from Middle East to Japan is included in this price. In the US—a significant producer and consumer of gas—the prices of natural gas have not mimicked the crude oil prices. The natural gas prices have been market- determined and are less volatile compared to crude oil prices (Figure 7.4). The cost structure of natural gas production and of its transportation being different from those of crude oil prices, many countries are choosing market-determined rates for natural gas.13 Hence, in India also, the linking of gas prices with crude oil prices or fuel oil prices is not in the interest of consumers. 5.5 30.00 5.0 4.5 25.00 4.0 3.5 20.00 3.0 US$/bbl US$/mBtu 2.5 2.0 15.00 1.5 1.0 10.00

1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Japan (LNG cif - US$/mBtu) USA Henery Hub (US$/mBtu) Dubai (US$/bbl) Figure 7.4: LNG prices, US gas prices and crude oil prices Source: BP (2002) 82 | Indian Infrastructure: Evolving Perspectives

2.4.1.1 Gas price scenario in India The price of natural gas produced by ONGC and OIL are fixed by the government. In 1997, the government linked the landfall price of gas to a basket of international fuel oil prices with the ceiling of Rs 2850/1000 m3. The Gas Pricing Committee had recommended that natural gas prices be increased gradually so that they are at par with the import prices of LNG by 2002. The MoPNG was to review recommendations of the Gas Pricing Committee in April 2000 but it postponed the decision on the revision of natural gas prices. The government was supposed to raise gas prices to 55 per cent, 65 per cent and 75 per cent of the basket price in FY98, FY99 and FY00 respectively but did not implement it. However, with the rise in global prices the landfall prices of gas hit the ceiling in October 1999.14 The consumer price along the HBJ pipeline was fixed at Rs 4000/1000 m3, which included a flat transportation charge of Rs 1150/1000 m3.15 Currently, the city gate price of gas is fixed and the factory gate price includes royalty, taxes and other statutory levies.

2.4.1.2 Gas price post K-G Basin gas discovery It has been reported that Reliance will price its K–G Basin gas at US$3/mBtu on a ‘delivered’ basis. The price of $3/mBtu offered by Reliance is substantially lower than the price of naphtha and fuel oil (a sizeable proportion of existing fertilisers and power capacity is based on these) at about US$7–7.5/mBtu and US$5.5– 6.5/mBtu respectively. This means that all users, irrespective of their location, will get gas at this price. The corresponding price currently charged by GAIL for supply of domestic gas from ONGC and other producers in the private/joint sectors is US$1.9/mBtu to plants at landfall point/receiving on-shore gas, and US$2.5/mBtu to plants along the HBJ pipeline. The local taxes are different in different states and quite substantial in some states. For example, while Madhya Pradesh and Rajasthan charge 4 per cent sales tax on natural gas or LNG, Maharashtra charges 15.2 per cent, Andhra Pradesh wants 16 per cent and Gujarat charges 20 per cent. For imported LNG, Petronet-LNG has indicated that its price (benchmarked to the Japanese Cocktail Crude Index) at the Dahej terminal will be US$4/mBtu.16 At the user point, the price will be higher. The industries located in the ‘hinterland’ may end up paying an amount in excess of US$5/mBtu. Apart from Qatar, Australia, Russia and other Middle East countries are ready to supply LNG to India at a competitive price. Piped gas from Bangladesh could, perhaps, match the RIL price. Iran and Burma are also competing to provide piped gas to India using transnational gas pipelines. The issue of gas pipelines has been intrinsically linked to ‘delivered’ gas prices. The cornerstone of the success of the NELP is that the companies investing in exploration and development of oil and gas fields are allowed to sell oil and gas at Regulation of Petroleum Product Pipelines | 83 market price. The regassified LNG also can be sold at market price but the price of the gas produced by ONGC and OIL is still controlled. The MoPNG is in favour of deregulating the gas prices but the main user industries—power and fertilizer— have raised serious concerns. It is reported that the MoPNG would have liked to deregulate gas prices and it has proposed to raise the ceiling on gas prices to 75 per cent of fuel oil prices by 1 January 2003 (Rs 4300 per mcm) and to 100 per cent by 1 April 2003 (Rs 5800 per mcm) before deregulating completely in October 2003. This was not acceptable to user industry ministries and the government constituted a Group of Ministers Committee to recommend gas price.17 The implication of all these developments is that gas prices linked to fuel oil prices are not in the long-term interest of Indian consumers. The government should abandon its existing approach of linking the price of gas to the imported price of fuel oil. The pricing of gas should stand on its own, especially when global trade in gas is poised to take a quantum jump. Gas prices locked at US$3/mBtu would lead to supply of electricity at a competitive rate of Rs 2.30 to Rs 2.70 per unit and would ensure a competitive position on the merit order of most State Electricity Boards.18 Even the fertiliser industry stands to gain substantially when gas prices are allowed to be market-determined [ICRA (2003)]. An important element in this development is that the producers of gas are ready to keep the city gate price of gas same all over the country. This is at variance from government thinking of doing away with the flat transportation charge for gas and introducing a new distance-based charge for gas supplied through long distance pipelines. From the above unfolding scenario, two issues are important for the gas pipeline industry. First, there will be a number of suppliers of gas in the country. Second, there will be a keen competition among suppliers to supply gas at competitive rates and they would like to lock in ‘delivered’ gas price to ensure that investment in the development of gas fields and the sunk cost in laying of pipelines yield an average rate of return over its life commensurate with the risks assessed at the time of investment.

3. ECONOMICS OF PIPELINES

3.1 Crude pipelines Crude oil is the most important input for any refinery. Crude pipelines are largely owned and operated by the refinery/oil field owner. The pipeline capacity is in line with the refinery capacity and, therefore, it is an integral part of the refinery. These pipelines are generally not regulated. 84 | Indian Infrastructure: Evolving Perspectives

3.2 Product pipelines The traditional methods for inland bulk transportation of petroleum products involve the use of road tankers, rail tank wagons and pipelines. For transporting relatively small volumes up to 300 km, road tankers are the most cost-effective method, even though their unit cost (ton/kilometre) is the highest among all the above-mentioned methods. Beyond 300 km, if volumes are good enough for full dedicated trains, rail tank wagons could be the best solution, provided there is an existing railway system and it is efficiently managed. When the railway is competently operated, the ton/kilometre cost of using a rail tank wagon is roughly one-half the ton/kilometre cost of using a road tanker. Transportation of oil products via pipelines require relatively large initial investments, and generally are attractive only for large volumes travelling long distances. However, once the initial investment costs are amortized, pipeline transportation offers the cheapest ton/kilometre costs in the long term because the basic operating cost for the pipeline, without amortization costs, is less than one-half the rail operating cost [World Bank (2001)]. Figure 7.5 compares typical petroleum transportation costs for various levels of throughput for road, rail, and pipeline modes. As the graph shows, the specific transportation cost via pipeline decreases with increasing transport volume, whereas product transport by road or railway is not dependent on the transport volume. Figure 7.5 also shows that the pipeline option becomes competitive once transport volumes rise above 700,000 tonnes per year and that with a transport volume of three million tons per annum the specific transportation cost of using a pipeline is about one-fourth the transport cost of using a road tanker. 14

12 Pipeline Road tanker 10

8 Railway - mixed 6

4 Railway - unit 2

0 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Specific transp. cost (US$/tonne/100 km) Transport volume (1000 tonnes/year) Figure 7.5: Comparative transportation costs for road, rail and pipeline modes Note: These are economic costs and no margin for profit has been added Source: World Bank (2001) Regulation of Petroleum Product Pipelines | 85

For the pipeline, the amortization of initial capital expenditure has a major influence in determining tariff. Once the debt repayment and capital cost depreciation is completed, transportation of oil products by pipeline can provide a competitive edge. Figure 7.6 shows the specific transportation cost (before and after amortization of capital cost) for a pipeline (12” diameter). The comparison demonstrates that even if the cost of preventive maintenance increases, the specific transportation cost decreases by about 50 per cent as compared to the cost during the amortization period. 9 8 specific 7 transp. cost (OPEX + 6 CAPEX) 5 4 3 specific transp. cost 2 (OPEX only) 1 Specific transp. cost in US$/100 km 0 012345678910 Transport volume (million tons/year) Figure 7.6: Influence of amortization for a 12” diameter pipeline Source: World Bank (2001) Clearly transportation of large volumes of oil products favours pipelines which are fixed point-to-point link; and over a short distance, road tankers, which are flexible link, are economical. In the transportation of products, road tankers are an integral part of the distribution network for any oil company to reach demand centres.

3.3 Gas pipelines Gas is usually transported using pipelines. For transportation of gas, transmission lines are required to transport pressurised gas over long distances and supply lines having low pressure are required to supply gas to end consumers.19 Investment is required in development of gas pipeline facilities, comprising of high pressure, medium pressure and low pressure system and associated gas compressors, telecommunication system, etc. Though the economic profile of a gas pipeline is similar to that of an oil pipeline, there are some major disparities between oil and gas pipelines which alter the 86 | Indian Infrastructure: Evolving Perspectives break-even point of gas pipelines. First, the unit investment in gas pipelines is three times that of oil pipelines (without terminal facilities). Second, the life span of gas pipelines is at least three times that of oil pipelines. Third, for oil product pipelines to work efficiently, a large sum of investments is required in terminal facilities including storage tanks whereas gas pipelines are designed to take care of short-term storage requirements. Fourth, gas pipelines consume gas to run gas pumps to push gas down the pipeline and this internal consumption is accounted as shrinkage. This internal consumption of gas is directly proportional to the distance over which gas is to be transported20 [Gardel (1981)]. The sunk cost and longevity of a gas pipeline suggest that the gas pipeline industry is one in which contestability analysis cannot be applied and these sunk costs generally deter the entry of a new pipeline. This implies that small gas finds cannot be commercially exploited and the life of gas pipelines may be longer than the life of gas lakes.

4. MAIN ISSUES FOR PETROLEUM PRODUCT PIPELINES IN INDIA

4.1 Oil product pipelines In India, prima facie pipelines would appear to have a fragmented network of various types, thereby suggesting little planning, if any. However, when one superimposes railways and road network on the pipeline network with the volume of crude/products to be pushed through the transport network, the picture becomes clearer and the product pipeline network emerges as an integral part of transport logistics and in congruence with the economics of product pipelines. The basis of the industry structure of crude/product pipelines has been primarily to decongest surface transport and to complement rail transport. The access control and tariff applicable to pipeline transportation of oil/products is a result of centrally planned development of refinery capacity, marketing and transportation of oil and oil products. The current utilisation of oil pipelines is between 50 per cent– 60 per cent of the available capacity.21 The main issues in product pipelines are ownership, access and transmission charges.

4.1.1 Ownership and access Under the Administrative Price Mechanism (APM), marketing of petroleum products was undertaken only by public sector oil companies and directed by the MoPNG. The existing pipelines owned by different marketing companies used to have a ‘hospitality arrangement’ to carry products of other refineries. That is, if Regulation of Petroleum Product Pipelines | 87 refineries had a marketing arrangement with the pipeline operating company, they had access to the pipeline. Post APM, product pipelines are being used strategically to deny access to other refiners by the operating company and hence the issue of access has become important.22 Some pipelines which were product pipelines have been converted into crude pipelines, thus leaving investments in pipelines connected to it, but owned by other operators, stranded.23 Product pipelines generally carry multiple products and, therefore, tap-off points need to have substantial investment in storage and quality control facilities.

4.1.2Transportation charges Less economical pipelines (owing to lower volume traffic or demand) were cross- subsidised by pipelines enjoying better economies. Once the asset was created, it was serviced by the oil pool account administered by the government. Under APM, pipelines complemented the railway network to reach oil products to main demand centres. Being captive users and the transportation charges being paid from the oil pool account, marketing companies were immune to transportation charges paid to railways. The POL used to attract the highest rail tariff rate under the APM.24 But sensing tough competition from the oil pipeline, the railways has already begun slashing freight tariff. In some sectors, the railways has reduced rates by 20 per cent or so.25 Post APM, the scenario has changed with transport charges being paid by the marketing companies26 (see Box 7.3).

The laying of product pipelines needs upfront investment; hence tariff charges should be in two parts—capacity charge and user charge. The capacity charge is to cover the fixed cost of laying the pipeline and the user charge is a variable charge in proportion to the volume of material carried through the pipeline. This second part of the tariff is to cover the operational cost of the pipeline and ancillary services. An oil product pipeline also has business risk. Capacity augmentation of an existing refinery in a particular region or the construction of a new refinery changes the demand and supply scenario of oil products in the region. Minor suppliers of oil products do not want to give a long-term take-or-pay commitment to the pipeline company because the contingent liability of such providers increases. A refinery having a dedicated pipeline to evacuate products, however, has the comfort of pre-empting product demand in the region as well as being able to meet the growing demand of oil products by augmenting the capacity of the pipeline. 88 | Indian Infrastructure: Evolving Perspectives

Box 7.3: Impact of OCC on oil product prices

The Oil Coordination Committee (OCC) was a de facto regulator for the oil industry. The OCC was established in 1975 to implement the elaborate oil pool account mechanism recommended by the Oil Prices Committee. As the nodal agency for the sector, key functions of the OCC included administration of pool accounts, coordination of the transportation of crude oil imports for PSU refineries, inland distribution of petroleum products, deciding on allocation of crude oil, product pattern of refineries based on imports, exports and national/regional demands, logistics of transportation, and other allied matters. Thus, the OCC under the administrative control of the MoPNG, performed the functions of a planner, coordinator, advisor, and regulator in the downstream sector.27 The OCC performed its assigned role using the Administrative Price Mechanism, which ensured * orderly growth of the oil industry; * continuous availability of petroleum products to consumers at fairly stable prices; * continuous availability of crude oil to refiners; and * achievement of socio-economic objectives of the government. At the heart of the APM was the oil pool account. This is the account, where the subsidies and contributions were netted off each other. Under the APM, refined products were purchased from refineries and transferred to marketing companies. The refineries were paid at international parity price of the products. The OCC decided the amount marketed by each marketing company and allowed 12 per cent post tax return on funds deployed by marketing companies. The oil pool deficit increased substantially year after year. The APM had many unintended consequences. Oil product pricing was divorced from underlying crude prices. The prices of politically-sensitive products did not reflect the economic costs of the products. Subsidies and cross-subsidies resulted in a wide distortion of consumer prices and led to wastage of energy. The APM provided little incentive for improving productivity or efficiency as returns were guaranteed on the capital employed. Competition was stifled with marketing companies acting as mere distribution companies. The industry is just trying to get out of this mindset. The administrative control of the ministry on utilisation/misutilisation of various assets can be seen from the following anecdote: The MoPNG would like to ship crude oil from Revva field offshore Andhra Pradesh to the Bongaigaon Refinery and Petrochemicals Ltd (using the Haldia–Barauni–Bongaigaon crude pipeline) and bring the refined products back to the northern region (using Bongaigaon–Barauni product pipeline). Such a move is being opposed by AP as the crude is used by HPCL’s Vizag refinery.28 Regulation of Petroleum Product Pipelines | 89

4.2 Natural gas pipelines In India, natural gas pipelines mainly serve industrial consumers. The transportation of natural gas through pipelines involves significant economies of scale and very little inter-modal competition. The nature of natural monopoly of gas transport calls for a set of regulations not required for oil products pipeline. However, the issues related to gas pipelines are the same as those related to oil product pipelines, namely, ownership, access and transmission charges. The issue of state–centre jurisdiction will get settled when the Supreme Court pronounces judgement on the rights of state to pass legislation related to gas.

4.2.1Ownership and access At present, the HBJ pipeline is the only transmission gas pipeline in the country and all the gas from ONGC’s western gas fields is marketed by GAIL. But, as a number of LNG terminals come up along the western coast, the issue of access to HBJ pipeline will become important. Unlike oil product pipelines the tap-off point in a gas pipeline can be easily provided as natural gas is a homogeneous product and priced on the basis of energy content.

4.2.2 Transportation charges Gas uses gas to transport gas and, hence, operational cost of a gas pipeline is a function of volume of gas and distance over which the gas is to be transported.29 But, in India, gas is sold to users as bundled product where gas is priced according to calorific content of gas and flat transportation charges. The MoPNG has fixed the transportation charges as Rs 1150 per thousand cubic metres (tcm) along the HBJ pipeline, with effect from 1 October 1997.30 While the transportation charge will remain static until the Group of Ministers decides on decontrolling gas prices in the country, GAIL will get compensation for cost escalation directly from the gas pool account.31 In a supply constrained scenario, transportation charges were fixed by the MoPNG to ensure that GAIL does not exploit its monopoly power and at the same time earns ‘adequate’ return on its investment. In a few years’ time, when there will be multiple suppliers of gas, a gas transport operator could be exposed to uncontrollable risk if producers and importers do not supply adequate quantity. On the other hand, if flat transport charges are unremunerative, it would not be keen to increase the pipeline capacity. Long term contracts between producer and consumers mitigate the supply risk but expose a pipeline company to credit risk.

4.2.3 State–centre jurisdiction Gujarat is a key state for the gas industry. Besides exploration and production from the western onshore and offshore gas fields, Gujarat is the gateway for gas imports 90 | Indian Infrastructure: Evolving Perspectives in the country. The state also houses a significant number of large downstream consumers. Consequently, the state government passed the Gujarat Gas Act to facilitate the development of transmission and distribution (T&D) infrastructure in the state. The Act has come in for criticism from the centre and GAIL, and it has been referred to the Supreme Court for its opinion. Maharashtra, Andhra Pradesh, Punjab, Tripura and Assam are also keen to expand gas distribution pipeline network in their states but they have not shown any inclination to establish a state level authority to regulate gas T&D infrastructure. Potential centre–state areas of conflict relate to access and transportation charges of inter-state transmission lines.32 National PSUs, such as GAIL, have argued that one company regulated by two regulatory authorities may give rise to operational difficulty if access code and transportation charges are divergent. For example, a gas pipeline built on contract carriage principle cannot function on open carrier principle (see Box 7.4: Contract carrier versus common carriage carrier).

Box 7.4: Contract carrier versus common carriage carrier

Contract carriers are only obligated to provide service for those who have contracted for their services. Under contract carriage, individual transmitters need to provide additional facilities only where users are willing to sign firm contracts for their use. In the case of existing facilities, transmitters are required to provide transmission services up to the extent of any spare capacity. This means that system development generally takes place in response to demand from users, and this in turn facilitates the financing of new developments. The users pay for capacity on pipelines and receive in return a defined set of rights enshrined in a contract. These rights set out in these contracts cannot, in general, be undermined by later developments outside the control of the contracting parties. The main advantage to construct a pipeline on contract carrier principle is that the pipeline company is paid for capacity and demand risk is shifted to users. Capacity is built in line with the committed demand and this eliminates financial risk attached to building overcapacity. Regulatory risk is also limited as contracts define transportion charges. The disadvantage under this system is that the producer has captive consumers, thus, reducing the choice which consumers may get in future. Common carriers are obliged to render service to all comers on a non-discriminatory basis and have the responsibility to reasonably anticipate future demand for their services. If the regulator feels that the transmitter is underestimating future needs, he may require that the transmitter put up pipelines with adequate capacity. The transmitter would, of course, have the right to recover all costs and earn a reasonable rate of return on all authorised assets. Mandatory open access means that the regulator should have the power to ensure that any spare capacity in a pipeline be offered on a Regulation of Petroleum Product Pipelines | 91

non-discretionary basis to anyone who wishes to make use of that capacity. The regulator should also have the power to provide interconnection between existing and other pipelines. The key features of common carriage are that while transmission is provided on an ‘as required’ basis, users are not committed to long-term use-of-system contracts. This means that the transmitter must construct additional capacity to cope with all anticipated demands for its services. The transmitter, therefore, has significant demand forecasting and investment obligations. The main advantage is to producers and consumers as they do not have to pay or commit for the pipeline capacity. A consumer has a choice to use an alternative supplier of the product. In the absence of committed consumers, producers are reluctant to sign take-or-pay contract which increases financing risk. Under the open carrier system, a regulator determines transportation tariff and he needs detailed capacity utilisation and financing cost to determine the tariff. Additionally, the regulator can force the pipeline owner to increase the pipeline capacity if he feels that the capacity is inadequate.

5. POLICY AND REGULATORY DEVELOPMENTS Instead of having separate regulatory authorities for natural gas and the downstream petroleum sector, the MoPNG, on recommendations from user ministries, has decided to have a single authority in the country for the natural gas and downstream petroleum sector.33 A bill to constitute a downstream petroleum regulatory board was introduced in the Parliament on 8 May 2002, and has been referred to the Parliamentary Standing Committee on Petroleum and Chemicals [GOI (2002b)]. The Bill would be tabled in Parliament as soon as the Standing Committee approves it. In the interim period, the union government will continue to act as the regulator.34

5.1 The Petroleum Regulatory Board Bill, 2002 The proposed bill is to provide for the establishment of a Petroleum Regulatory Board to regulate the refining, processing, storage, transportation, distribution, marketing and sale of petroleum and petroleum products excluding the production of crude oil and natural gas. The purpose of the bill is to promote competitive markets. The salient features of the bill related to petroleum product pipelines are as follows: • The bill will authorise the central government to constitute a single authority for the downstream petroleum sector for the entire country. • The regulator will have authority over new, as well as existing, pipelines. 92 | Indian Infrastructure: Evolving Perspectives

• The regulator would regulate gas and oil product pipelines which have been laid on common carrier principle. Captive pipelines and crude oil pipelines would not come under its purview. New pipeline owners/operators would have right of first use. • In practice, pipelines will be laid on ‘modified’ contract carriage principle and existing owners and operators of pipelines would keep the right of first use. Only that capacity which is not being used would come under common carrier principle. Even expansion of capacity will carry the right of first use (see Box 7.4: Contract carrier versus common carriage carrier). • The regulatory body would have powers to declare a pipeline as common carrier, and to authorise laying, building, operating or expanding a pipeline as common carrier, or for establishing a liquefied natural gas terminal, or for marketing notified petroleum and petroleum products. • Before declaring a pipeline common carrier, the owner would be given a proper hearing and fix the term and conditions subject to which the pipeline is to be declared as a common carrier. The entity laying, building, operating or expanding a pipeline shall have the right of first use. • The authority would permit pipeline-on-pipeline competition and invite open offers. • The regulator would have powers to regulate any distribution or marketing company.

5.2 Guidelines for laying petroleum product pipelines The Ministry of Petroleum and Natural Gas has notified a new policy for laying petroleum product pipelines in the country through the Gazette of India Extraordinary dated 20 November 2002 [GOI (2002c)]. The salient features of the Guidelines are as follows: • There would be three categories of petroleum product pipelines, namely, (i) pipelines originating from refineries upto a distance of around 300 km; (ii) captive pipelines, originating either from a refinery or from an oil company’s terminal, of any length; and (iii) pipelines exceeding 300 km in length and pipelines originating from ports. Category (i) and (ii) pipelines would be for the exclusive use of the proposer company and owned by the company. Any legal entity can propose and own a category (iii) pipeline. Three-fourths of the designed capacity of a category (iii) pipeline would be reserved for the owner and ‘take or pay’ contracts and only one-fourth of the designed capacity would be made available for use by anyone at a government-approved tariff. Regulation of Petroleum Product Pipelines | 93

• Through this notification, the government has taken away the monopoly of Petronet India Limited for laying product pipelines in the country. • ‘Contract carriage principle’ has effectively replaced the ‘Common carrier principle’ for pipelines.35 • The authority to grant right of use inland, i.e. to give license to lay a pipeline, will remain with the ministry. • The Guidelines have a sunset clause. The regulatory functions of product pipelines will be passed on to the regulatory board constituted under the Petroleum Regulatory Board Bill 2002. Through the bill and the guidelines, the government has tried to address the issues of investment efficiency and operating efficiency, and tried to make the pipelines bankable. The government has recognised that pipeline developments are long- term investments, reliant on market growth for their viability. As a consequence, returns are likely to be poor during the initial years of a project. For a project to proceed and to attract investment in the sector, the estimated average rate of return of the project over its life must be commensurate with the risks as assessed at the time of the investment. The government recognised the nature of these risks and their consequences and, thus, came out with the bill and further liberalised the sector through the guidelines. Among the classes of risk that face the developer of a pipeline project in India, credit risk, market risk and regulatory risk predominate. All these risks have a significant effect on the investor’s assessment of project viability. The current level of credit risk is high and it needs to be mitigated and the guidelines have given a free hand to developers to manage captive pipelines as well as pipelines upto 300 km in length. The investor expects to accept market risk along with construction, technological and general economic risks. Market risk is about whether sufficient load will materialise to make the project viable. For a greenfields project, there are two sources of load—first, penetration of the existing available market, i.e. the conversion of existing consumers to alternate fuels; and second, generation of new demand from new consuming businesses to the region. The bill and the guidelines have been able to take care of the market risk and credit risk; but the bill, which has introduced regulatory risk emanating from the regulator’s power to declare a pipeline as a common carrier, should be addressed by the government and the regulatory authority. 94 | Indian Infrastructure: Evolving Perspectives

6 . THE WAY FORWARD FOR PIPELINE REGULATION A regulatory environment must be created to encourage pipeline infrastructure development and make an allowance for efficient market growth. The Petroleum Regulatory Board Bill 2002 is clear that there will be a single regulatory authority to regulate downstream oil and natural gas sector. As and when the bill becomes an Act, the government should establish a regulatory authority for the downstream petroleum sector under the Act which has power to regulate petroleum product pipelines. The ‘Guidelines to Lay Petroleum Product Pipelines’ has further liberalised the pipeline sector and the sunset clause in the guidelines ensures that the authority to regulate pipelines is automatically transferred to the regulator established under the Act. The following major tasks in the pipeline industry should be addressed by the regulator: • Introduce distance-based transportation charges for natural gas. This would ensure that small onshore gas fields are viable and have access to the gas pipeline network. • The regulator must distinguish transmission gas pipelines from distribution and supply gas pipelines. Although, at present there is only one supplier of gas in metropolitan cities, the scenario could change rapidly if new suppliers of gas enter the distribution business. The distribution and supply pipelines should be licensed on open access principle right from the beginning to ensure that retail household consumers and the transportation sector is able to choose the supplier of gas at a later date. The gas supply companies must be asked to maintain separate accounts to show the cost of energy and cost of supply and distribution. • The regulator should encourage a secondary market in the product pipeline capacity as the market matures.36 In pipeline regulation, there are two issues which need to be addressed to mitigate regulatory risk. First, the possibility of mandated access adds to the regulatory risk for investors and the pipeline owner. If the terms and conditions of any mandated access are perceived as likely to be unduly favourable to the access seeker, the attractiveness of investment will be reduced. Regulatory risk can be reduced significantly if the regulatory rate of return were indicated for the life of the project at its inception on the basis of the risks as assessed at that time. Second, to give right of use still remains with the central government and from this authority the government derives the power to license a petroleum pipeline [GOI (1962)]. This power should be vested in the regulator so that the regulator has enough teeth to ensure that interconnection and the power to declare a pipeline a ‘common carrier’ can be enforced. Regulation of Petroleum Product Pipelines | 95

7. CONCLUSION The petroleum product market is enjoying the fruits of economic deregulation set in motion in 1991. Private sector investment in oil refineries has resulted in a substantial increase in refining capacity in the last two years and the country is running a surplus of refined oil products. Post the dismantling of APM, the transportation of oil products, which was overlooked earlier, has acquired importance. The failure of Petronet India Limited (PIL) to achieve financial closure has led the government to change its regulatory practice of petroleum product pipelines. The ‘Guidelines to Lay Petroleum Product Pipelines’ has taken away PIL’s monopoly over laying new product pipelines and new pipelines can be constructed by anyone using a ‘modified’ contract carriage principle rather than being compelled to adopt the common carrier principle. The modification in the contract carrier principle is that one-fourth of the designed pipeline capacity should be made available on open carrier principle. The natural gas market in India is in transition as it attempts to move from a fully centralised, government-controlled business to one that relies increasingly upon reduced government controls and a more market-responsive pricing climate to encourage foreign and private investment in upstream exploration and development of oil and gas. Natural gas pipeline projects are built on trust because nearly all of the cost is incurred at the beginning and the revenues come only over the next couple of decades, usually from long-term contracts signed with gas users. Pipeline construction and development of gas fields is undertaken only after these contracts are signed. The Petroleum Regulatory Board Bill 2002 and the ‘Guidelines to Lay Petroleum Product Pipelines’ have addressed some of these issues. The scenario which may become a reality in a couple of years’ time is the supply of gas from multiple sources using different processes having different cost structures. In such circumstances, gas pipelines have been allowed to use contract carriage principle as well as to lay captive pipelines. The driving force to use pipelines will be the sunk cost and long-term contracts with the users.37 Captive gas transmission pipelines would not be regulated. The distribution and supply gas pipelines are being laid by individual gas companies and should be licensed on open-carrier principle right from the beginning to allow choice at a later date to retail consumers and to the transport industry. Only a few metropolitan cities have started to have piped gas for domestic use and it faces competition from bottled LPG gas which is a ‘notified’ product and would continue to be subsidised for some more time. The Petroleum Regulatory Board Bill provides adequate power to regulate access and the price of gas, and to regulate distribution companies and ‘notified’ products. The gas distribution and supply pipeline regulation is going to remain in animated 96 | Indian Infrastructure: Evolving Perspectives suspension until the issue of centre–state jurisdiction over gas is decided by the Supreme Court. We have critically examined the Petroleum Regulatory Board Bill 2002 and the ‘Guidelines for Laying the Petroleum Product Pipelines’ which have addressed the issue of investment and operational efficiency and also made pipelines investment bankable. The government should get the bill through Parliament and constitute a petroleum regulatory authority. The new guidelines point towards a great deal of reliance on market forces to discover product prices and transportation tariffs. Oil product pipelines already face strong competitive pressures from other modes of transportation—such as trucking and railways. In order to create a regulatory environment which encourages pipeline infrastructure development, the authority should provide an indicative rate of return for the life of the project at the time of granting right of use to mitigate regulatory risk. The authority to acquire land under the Petroleum and Minerals Pipelines (Acquisition of Right of User Inland) Act, 1962, should also vest in the regulatory authority in order to unify all regulatory and licensing powers in one authority. Regulation of Petroleum Product Pipelines | 97

APPENDIX A Table 7.6: Projections for POL consumption 2001–2008 (in ‘000 metric tonnes) 2001 2002 2003 2004 2005 2006 2007 2008 LPG 7021 7583 8189 8844 9552 10316 11141 12033 MS 6625 7138 7692 8288 8930 9622 10368 11171 SKO 11295 11323 11606 11896 12194 12499 12811 13131 HSD 38203 37439 38188 39906 41901 43997 46196 48506 ATF 2234 2301 2370 2441 2514 2590 2668 2748 FO/LSHS 12645 12930 13220 13518 13822 14133 14451 14776 Lubes 1027 1053 1079 1106 1134 1162 1191 1221 Bitumen 2677 2784 2895 3011 3132 3257 3387 3523 Others 6266 6391 6519 6650 6783 6918 7057 7198 Total 102000 102956 105780 109689 113997 118536 123319 128363 Overall growth rate % 0.9 2.7 3.7 3.9 4.0 4.0 4.1 Source: India Infoline Estimates Table 7.7: List of LNG terminals Location Capacity (mmtpa) Promoter Dahej (Gujarat) 5.0 Petronet LNG (January 2004) Ennore (Tamil Nadu) – Petronet LNG Mangalore (Karnataka) – Petronet LNG Kochi (Kerala) 2.5 Petronet LNG Vizag (Andhra Pradesh) 2.0 Total/HPCL Trombay (Maharashtra) 2.5 Total/Tata Electric (shelved) Pipavav (Gujarat) 2.5 BG/Gujarat Pipavav Port (shelved) Dabhol (Maharashtra) 5.0 Enron (suspended) Ennore (Tamil Nadu) 2.5 TIDCO/Unocal (shelved) Kakinada (AP) 2.5 CMS Energy/Unocal/GVK Industries Kakinada (AP) 2.5 IOC/Petronas/BP Hazira (Gujarat) TBA Mobil/Gujarat Maritime Board Hazira (Gujarat) 5.0 Reliance/Elf Aquitaine Hazira (Gujarat) 2.5 Shell/Essar (proposed) Total 37.5 Source: Petroleum Economist (December 1997), media reports 98 | Indian Infrastructure: Evolving Perspectives

Table 7.8: List of crude pipelines Pipeline Length (km) Capacity (mmtpa) Owner Nahorkatiya–Barauni 1156 5.5 OIL Salaya–Mathura 1881 21.0 IOCL Ankleshwar–Koyali 95 2.0 ONGC Kalol–Navagam–Koyali 127 2.0 ONGC Bombay High–Uran (offshore) 203 15.0 ONGC Haldia–Barauni 506 4.2 IOCL Total 3968 49.7 Source: IOC Presentation, Seminar on Pipelines (2000) Table 7.9: List of gas pipelines Name Length (km) Owned by Trunk Lines Hazira–Bijaipur–Jagdishpur 2300 GAIL Goa–Sangli–Hyderabad– ~1300 Gas Transportation Vijayawada–Kakinada; Spurs to (proposed) and Infrastructure Co. Mumbai and Chiplun (Phase I) Ltd (Reliance) Vijayawada–Chennai–Bangalore – GTIC (Reliance) –Kayamkulam (Phase II) Mumbai–Delhi–Kolkata– ~6400 GAIL Chennai (proposed) Dahej–Hazira–Uran ~600 GAIL (proposed) Distribution lines Paguthan–Vadodra 68 Gujarat State Petronet Ltd Vadodra–Ahmedabad 85 Gujarat State Petronet Ltd (proposed) Regulation of Petroleum Product Pipelines | 99

Table 7.10: List of refineries Owner Capacity as of December 2001 (mmt) IOC, Guwahati 1.00 IOC, Barauni 4.20 IOC, Koyali 12.50 IOC, Haldia 3.75 IOC, Mathura 7.50 IOC, Digboi 0.65 CPCL, Manali 6.50 CPCL, Nasimanam 0.50 BRPL, Assam 2.35 IOC, Panipat 6.00 IOC, Gujarat 3.00 BPCL, Mumbai 6.90 KRL, Kochi 7.50 NRL, Assam 3.00 HPCL, Mumbai 5.50 HPCL, Vizag 7.50 MRPL, Bangalore 9.60 RPL, Jamnagar 27.00 Total 114.95

Table 7.11: List of oil product pipelines Pipeline Length (km) Capacity (mmtpa) Owner Guwahati–Siliguri 435 0.82 IOCL Koyali–Ahmedabad 116 1.10 IOCL Barauni–Kanpur 669 1.80 IOCL Haldia–Barauni 525 1.25 IOCL Haldia–Mourigram–Rajbandh 277 1.35 IOCL Mathura–Jalandhar 526 3.70 IOCL Kandla–Bhatinda* 1443 7.50 IOCL Digboi–Tinsukhia 75 0.73 IOCL Bombay–Pune 161 3.67 HPCL 100 | Indian Infrastructure: Evolving Perspectives

Table 7.11: List of oil product pipelines (contd...) Pipeline Length (km) Capacity (mmtpa) Owner Mumbai–Manmad 252 4.33 BPCL Vizag–Vijayawada 356 4.10 HPCL Vadinar–Kandla 113 11.50 PIL Jalandhar–Udhampur 233 – IOCL Koyali–Sidhpur – – IOCL Mangalore–Hassan–Bangalore 332 4.20 HPCL/PIL Chennai–Trichy–Madurai 505 1.40 IOCL/PIL CIPL (Jamnagar–Rajkot– 1760 – RPL/IOCL/ Koyali–Ratlam) PIL(shelved) Paradeep–Rourkela – 5.00 IOCL/PIL Jamnagar–Bhopal (Phase I) ~2500 – GTIC (Reliance) Bhopal–Raipur–Cuttack– (proposed) – Kolkata (Phase II) Chennai–Bangalore – – RIL (proposed) Total 4935 41.94 * Being converted into a crude pipeline Source: IOC Presentation, Seminar on Pipelines (2000), media reports

Table 7.12: List of ports handling oil/petroleum products Category Ports Crude oil Salaya (Gujarat), Jamnagar (Gujarat), Mumbai, Mangalore, Kochi, Chennai, Vizag, Haldia Petroleum products Kandla (Gujarat), Okha (Gujarat), Mumbai, Goa, Mangalore, Kochi, Tuticorin, Chennai, Vizag, Paradeep, Haldia, Port Blair Regulation of Petroleum Product Pipelines | 101

APPENDIX B Crude oil price web and natural gas prices Apart from price spikes during brief political and economic crises, oil prices, since the early 1970s, have tended to move closely in line with OPEC producer capacity utilisation. Significant price increases have generally occurred when OPEC production exceeds 90 per cent of capacity utilisation (roughly 28 to 30 million barrels per day). When utilisation is high, prices stay higher. When utilisation falls, producers have been unable to sustain high prices. OPEC is crumbling now and OPEC producers have realised that substantially higher prices are not in the producers’ long-term interests. Major Middle East producers have tried to follow a strategy of maintaining moderate prices and high market share that will maximize their returns over time. This strategy has allowed non-OPEC economies to grow and prosper, whereas excessive energy prices have led to reduced economic activity and lower oil consumption. It also discourages a number of activities that run counter to OPEC interests: oil conservation and the substitution of alternative energy sources; synthetic fuel research and LNG developments; tertiary recovery with tax and fiscal incentives; and exploitation of non-conventional hydrocarbons (heavy oil, oil shale, deepwater oil, and sources in Arctic or near Arctic regions). A price of US$25 per barrel (in 1994 prices) is considered to be optimal and US$20 per barrel discourages oil conservation and substitution of alternative energy resources. [Conn and White (1994)]. Equivalent gas prices are in the range of US$3–4 per mBtu.

60 81 80 50 82

40 83 79 84 75 74 85 76 30 78 77 90 96 2000 87 02 92 93 01 20 91 97 89 86 94, 95 99 88 10 73, 98 1970 71 72 Price (US$ per bbl in 1994 prices) 0 15 20 25 30 35 Crude oil production (Mbbl per day) Figure 7.7: OPEC supply/price dynamics Source: Oil Industry Outlook, BP Statistical Review 2002 102 | Indian Infrastructure: Evolving Perspectives

APPENDIX C Box 7.5: Main conversions used in the petroleum industry Crude oil 1 tonne = 7.33 barrels = 1.165 cubic metres (kilolitres) 1 barrel = 0.136 tonnes = 0.159 cubic metres (kilolitres) 1 cubic metre = 0.858 tonnes = 6.289 barrels 1 million tonne = 1.111 billion cubic metres natural gas = 39.2 billion cubic feet natural gas = 0.805 million tonnes LNG = 40.4 trillion British thermal units = 0.805 million tonnes LNG Natural gas 1 billion cubic metre = 35.3 billion cubic feet natural gas = 0.90 million tonnes crude oil = 0.73 million tonnes LNG = 36 trillion British thermal units = 6.29 million barrels of oil equivalent LNG 1 million tonne = 1.38 billion cubic metres natural gas = 48.7 billion cubic feet natural gas = 1.23 million tonnes crude oil = 52 trillion British thermal units = 8.68 million barrels of oil equivalent Source: BP Statistical Review of World Energy 2002

NOTES 1. India, which imports 70 per cent of its energy requirement, is projected to have the capability of producing up to 3 per cent of the world’s oil and gas output, with a sedimentary basin region of around 5.37 per cent, in the coming few years. In the year 2001, India produced 36.1 million tonnes of crude oil (1per cent of world output) and 26.4 billion cubic metres of natural gas (1.1 per cent of world output). Compared to this, India’s proven reserves were reported to be 0.62 billion tonnes oil and oil equivalent gas (O+OEG) [BP (2002)]. The Directorate General of Hydrocarbons (DGH) has informed the Petroleum Ministry that India’s prognostic hydrocarbon resources are about 28 billion tonnes of O+OEG. After the recent discoveries in the Krishna–Godavari deepwater area, the resources could go up by another 4 billion tonnes (Business Standard, 25 February 2003). Regulation of Petroleum Product Pipelines | 103

2. This information was given by the Union Minister for Petroleum and Natural Gas, Shri Ram Naik, to the Parliamentary Consultative Committee (Infraline, 24 February 2003). 3. US energy giant Unocal has offered to sell piped natural gas from Bangladesh to markets in northern India at US$3.5–4.5 per million British thermal units (mBtu). The landed cost of Bangladesh gas in Delhi would be between US$3.5 and US$4.5 per mBtu, compared to the subsidised current natural gas price of US$2.5–2.7 per mBtu (Business Standard, 10 December 2002). Bangladesh is looking for export of around 500 million cubic feet of gas per day. A proposal approved by Bangladeshi Prime Minister, Begum Khaleda Zia, in February 2003 is expected to be presented in the Parliament soon for ratification. The ruling four-party coalition government has a comfortable number of seats in the houses to get the proposal through (Infraline, 18 February 2003). 4. The report did refer to change in industry structure in India and suggested that the income elasticity may reduce to 0.7 by the year 2025 [GOI (2000)]. 5. ‘Industry Experts Look at Downstream Future in Hydrocarbon Processing’ (January 2003) published by Gulf Publishing Company. 6. This should not surprise anyone because as a crude oil importing country it is advantageous to have refineries in the coastal region. Incidentally, three out of four major metropolitan cities have major ports which are served by oil tankers. 7. The IOC refinery at Panipat is going to expand its capacity from 6 mtpa to 12 mtpa by 2004. HPCL’s Bhatinda refinery and BPCL’s Bina refinery will take 6–7 years to come on stream. Reliance is likely to expand capacity of its Jamnagar refinery from 27 mt to 37 mt in two years time. BPCL’s Mumbai refinery expansion from 8 mtpa to 12 mtpa is expected to be completed in FY05. 8. The new domestic gas finds or more such discoveries in the future would not in any significant way diminish the prospect of import of piped gas or LNG. Demand for gas will grow in the country from 110 million standard cubic metres per day (mmscmd) in 2001–02 to 145 mmscmd in 2006–07, 225 mmscmd in 2011–12 and to 325 mmscmd in 2019–20, according to GAIL estimates. These estimates are modest compared to the estimates given in Table 7.3. 9. India will continue to remain a net deficit country in natural gas production despite the discovery of world-class gas reserves off the east coast according to the Minister of State for Petroleum and Natural Gas, Santosh Kumar (written reply to the Lok Sabha, 12 December 2002). 10. Sale of condensate, a by-product of gas extraction, has reduced the breakeven point of Qatargas landed in Japan to US$13.6/bbl. After adding around US$3/bbl for regassification cost, the cost of gas could be approximately US$17/bbl or 104 | Indian Infrastructure: Evolving Perspectives

24. For petrol and HSD, ‘class’ for wagon load has been reduced from 300 to 250, and for crude and gas it has been reduced from 270 to 250 in the Railway Budget 2003–04. Reducing the class leads to lesser freight rates. Carrying petroleum products on the Railways will now be 10.7 per cent cheaper (Business Standard, 27 February 2003). 25. The Economic Times (23 October 2002). 26. The government would continue to provide freight subsidy in far-flung areas in the north and east regions, but the quantum of sales in these region is very less. Bongaigaon Refinery and Petrochemicals Ltd, having an installed capacity of 7 mtpa, is working below capacity as crude availability from Assam oil fields is pegged at 5 mtpa. 27. The Directorate General of Hydrocarbons (DGH) was established under the administrative control of the MoPNG in 1993 to manage Indian petroleum and gas resources. The DGH acts as a regulator for reservoir management of oil and gas fields; monitors public service centres on behalf of the Government of India; and acts as a project facilitator, assisting companies to get clearances/approvals from various ministries. One may simply call him an upstream hydrocarbon sector regulator. 28. The Business Standard (25 December 2002). 29. In 1999–2000, gas losses due to shrinkage were 7 per cent [GOI (2002a)]. 30. GAIL’s charges along non-HBJ pipelines are significantly lower (Rs 165/tcm). These pipelines, which were primarily transferred from ONGC in the fiscal year 1993, operate on cost plus margin basis. 31. Under the existing pricing mechanism, Rs 350 accrued to the gas pool for every ‘000 m3 of gas sold. This money is collected by GAIL and is shown as a liability on its balance sheet. At the end of the fiscal year 1997, the gas pool account was Rs 11.3 billion in surplus. The gas pool money is supposed to subsidise gas consumption in the states of the Northeast. However, low consumption in these areas results in relatively small outflow from the gas pool. 32. The Gas Authority of India (GAIL) has finally started work on the Rs 26 billion, 600 km Dahej–Vijaipur gas pipeline. It hopes to complete work on the pipeline by April 2004. For months, the Gujarat State Petroleum Corporation Limited (GSPCL) had tried to block GAIL’s plan to lay this 600 km pipeline from Dahej to Vijaipur to carry regassified LNG from the Dahej terminal to customers on the HBJ. GSPCL had argued that GAIL had no right to build pipelines in the state (The Economic Times, February 23, 2003). Even after the passing of almost two years, Gujarat State Petronet is still waiting for the bidders to finalise their bids to pick minority stake in a gas grid promoted by its subsidiary Gujarat State Petronet Limited (GSPL). Gujarat State Petroleum Corporation (GSPC) is offering 49% equity of GSPL for sale with a cap of 106 | Indian Infrastructure: Evolving Perspectives

11 per cent on each equity holder. GSPC wants to retain management control. The reports say that pending clarity on the constitutional validity of the Gujarat Gas Act, none of the interested parties–Shell, British Gas, GAIL, Indian Oil, Bharat Petroleum and KRIBHCO—are ready to commit anything. Also, the Supreme Court is yet to pronounce judgment on the rights of the state to pass legislation related to gas (Infraline, 18 February 2003).

33. The rationale for having the Gujarat Gas Act according to the state government has been that the item ‘gas and gas works’ figures in the State List of the Constitution. The item ‘gas and gas works’ pertains to synthetic and industrial gas and does not relate to natural gas. The central government is of the opinion that natural gas is part of mineral oils, which is under the Central List. The issue of centre–state jurisdiction would be settled after the Supreme Court has given its opinion on this issue to the government.

34. Business Line, 22 January 2003.

35. Gas Transportation and Infrastructure Co. Ltd—a subsidiary of Reliance India Ltd— has been given permission to acquire land under the Petroleum and Minerals Pipelines (Acquisition of Right of User Inland) Act, 1962, to lay a gas pipeline for Goa–Hyderabad– Kakinada and an oil product pipeline for Jamnagar–Bhopal (Government of India, Ministry of Petroleum and Natural Gas, Lok Sabha, starred question no. 2763, answered on 5.12.2002). Reliance intends to extend the Jamnagar–Bhopal oil product pipeline to Kolkata passing through Raipur and Cuttak. The pipelines are expected to be ready by the end of 2004 to supply gas from the K–G Basin to users in Andhra Pradesh, Maharashtra and Goa and to evacuate oil products from the Jamnagar refinery to users in Gujarat and Madhya Pradesh. The guidelines have enabled Reliance India Ltd to lay five pipelines for moving products from its 27 million tonne refinery to various parts of the country. These pipelines will connect Jamnagar to Delhi, Goa to Hyderabad through Sholapur, Delhi to Patiala, and one connecting Chennai to Bangalore (source: industry reports).

36. Oil product pipelines owned by PIL and other oil companies will benefit from the secondary market as it has happened in the US [IDFC (2002)]. The HBJ pipeline capacity is 65 mmscmd whereas it transports roughly 21 mmscmd. The pipeline can function on open carrier principle for quite some time. Sunk cost of this pipeline is so large that even at flat tariff rate, it will be beneficial for GAIL. The HBJ is a gift of a powerful politician to his constituency at the cost of the nation. Should consumers continue to suffer for profligation of politicians or move ahead is a question which the regulator has to tackle.

37. The HBJ pipeline capacity of 64 mmscmd compared to usage of ~21 mmscmd would restrain development of secondary market for sometime. Regulation of Petroleum Product Pipelines | 107

REFERENCES 1. BP (2002): ‘BP Statistical Review of World Energy’, British Petroleum, London. 2. Conn Charles and White David (1994): ‘The Revolution in Upstream Oil and Gas’, The McKinsey Quarterly Number 3. 3. ICRA (2003): ‘The Indian Oil and Gas Sector’, ICRA Limited, New Delhi. 4. IDFC (2002): ‘International Best Practices in Pipeline Regulation’ (Mimeo). 5. Gardel Andre (1981): Energy—Economy and Prospective, A Handbook for Engineers and Economists, Pergamon Press, Oxford. 6. GOI (1962): The Petroleum and Minerals Pipelines (Acquisition of Right of User Inland) Act, 1962, GOI, New Delhi. 7. GOI (2000): ‘Report of the Group on India Hydrocarbons Vision – 2025’, GOI, New Delhi. 8. GOI (2002a): ‘Report of the Sub-Group on Natural Gas Availability’, Tenth Five Year Plan 2002–02, Ministry of Petroleum and Natural Gas, Government of India, New Delhi. 9. GOI (2002b): The Petroleum Regulatory Board Bill 2002, Bill No. 38 of 2002, Government of India, New Delhi. 10. GOI (2002c): ‘Guidelines for Laying Petroleum Product Pipelines’, Ministry of Petroleum and Natural Gas, Government of India, New Delhi. 11. GOM (2001): ‘Report of the Energy Review Committee (Part 1)’, Government of Maharashtra, Mumbai. 12. World Bank (2001): First World Bank Workshop on the Petroleum Products Sector in Sub-Saharan Africa, World Bank, Washington. 108 | Indian Infrastructure: Evolving Perspectives

INCENTIVES, OWNERSHIP AND PERFORMANCE IN 8 POWER SECTOR: The Case of UP

February 2005

1. INTRODUCTION Like in the case of all other states in India, the power sector in Uttar Pradesh (UP) has been traditionally characterized by lack of competition, high transmission and distribution losses, irrational tariff structure and inadequate government support. These problems manifested themselves in huge cash losses for Uttar Pradesh State Electricity Board (UPSEB) year after year. By March 1999, the accumulated losses of UPSEB were Rs 10,300 crore or 6 per cent of SGDP and payables to power suppliers were about Rs 3400 crore (almost 20 months of power purchases). The poor financial performance of UPSEB, which performed as a single buyer, deterred private investment, while public investment was slow due to the resource crunch of the state. The inadequate and often distorted investment in turn led to poor performance. The sector was thus caught in a downward spiral of poor performance, low revenue and sluggish investment. It is against the background of bankruptcy of UPSEB, a near halt in investment and unsustainable fiscal pressures, that the power sector reforms were introduced, with the government issuing a power sector policy statement in January 1999. Realizing that UPSEB was operating as an extension of the state government and that the organizational, institutional, financial and ownership arrangements were not conducive to the realization of reform goals, the state government decided to distance the power industry from state administration and provide the power sector with the autonomy required to operate on commercial principles. The UP Electricity Reforms Act was notified in July 1999 to support the reform process. A number of significant initiatives entailing changes in organizational, institutional and financial structures have been taken in the last six years (see below). Power Sector: The Case of UP | 109

Two significant areas that have remained untouched, however, are the ownership arrangement and market structure. The power industry in UP is still predominantly government-owned in all its segments. So far, there has been no generation capacity (other than captive power) created by the private sector in the state. Except for a small part of the state, i.e. Greater Noida (connected load of about 35 MW) which is operated by a private distribution company, power distribution and transmission in the state are carried out by utilities owned by the state government.1 It may also be noted that the reform measures taken so far have virtually done nothing to change the market structure which could introduce competition into the sector. There has been no scope for competition for the distribution market as the recently formed distribution companies continue to be owned by the government. Further, the two state generating companies have signed long-term PPAs with UPPCL, whereby, generation tariff is determined on cost-plus basis and there is no scope for competition among generating companies. In the absence of competition, there has been a tendency on the part of the regulator and policy makers to resort to incentives to improve sector performance. The objective of this note is to assess how the power sector has responded to the various incentives in the last few years and examine whether government ownership has been the dominant factor influencing the nature and extent of this response.

2. FACILITATING MEASURES

2.1 Multi-year tariff (MYT) An independent regulatory commission, Uttar Pradesh Electricity Regulatory Commission (UPERC), was established in September 1999, with a mandate to adopt a tariff structure that would meet the objectives of efficiency and equity. The UPERC has issued five tariff orders so far in pursuit of its mandate. To enhance the predictability of the basis for tariff setting and ensure that consumers gain from reforms, the UPERC has adopted a multi-year tariff framework since 2002, which has been working as the prime incentive system for the utilities. Under the framework, annual performance targets for the utility have been fixed for five years in terms of T&D losses and collection efficiency, assuming 2000–01 as the base year.2 If the utility fails to achieve the targets and hence incurs a loss, the regulator would not treat the loss as a regulatory asset, implying that the consumers will not be required to bear the burden (in the next year or any time in future) resulting from the failure of the utility to achieve targets. On the other hand, if the utility exceeds the targets, it would retain the resultant profits. 110 | Indian Infrastructure: Evolving Perspectives

2.2 Initiatives to improve performance To facilitate improved performance of the sector, three important initiatives have been taken.

Unbundling With the ultimate aim of introducing competition in generation and distribution, the government unbundled UPSEB into three functionally separate, autonomous and separately accountable corporations: a thermal generation company (UPRVUNL), a hydro company (UPJVNL) and a company responsible for managing the transmission and distribution system (UPPCL). The assets, liabilities and staff of the UPSEB were transferred to these three corporations under a statutory transfer scheme. These companies continue to be state-owned. In a second round of unbundling, the state was divided into four geographically contiguous zones (barring Noida and Kanpur) and a separate distribution company was created in each. These four companies were carved out from UPPCL through the notification of a transfer scheme in August 2003.3 The discoms are managed by boards and have organic links with the UPPCL. To strengthen governance in the discoms, the selection of the MDs has been done through open advertisement.

Financial restructuring In January 2000, a clean up of the balance sheet of the UPSEB was carried out as a prelude to the transfer of business to successor utilities to enable the sector to inherit a relatively healthy opening balance sheet, which would facilitate a quick restoration of the sector’s creditworthiness (World Bank, 2000). The restructuring was done by write-off and provisioning of doubtful and obsolete assets, recognition of liabilities that were either understated or not reflected in the balance sheet, and settlement of cross dues between the government and UPSEB.

Tariff rationalization Tariff proposals are now subject to public scrutiny and the utilities have to defend their requests for tariff revision in open hearings. To reduce cross subsidy, the tariff increase for the subsidizing segments (industrial, commercial, railway traction, etc.) has been kept at lower levels than those for the subsidized sectors. Further, to the extent the tariffs suggested by the state government deviates from that fixed by the regulator, the state government has been required to fill the revenue gap through subsidies. Power Sector: The Case of UP | 111

3. IMPACT ON PERFORMANCE 3.1 Immediate results Let us first discuss the facilitating measures. Unbundling and corporatization has facilitated the emergence of a clearer picture and helped identify the sources of inefficiency, which would not have been possible under the vertically integrated entity. Thus, the true cost of generation and distribution (of different discoms, which have recently been created) is now revealed. It is now possible to find out, for example, the T&D loss at each discom. Financial restructuring has provided a pragmatic solution to dealing with the problem of past liabilities, which could have potentially become a hindrance to reforms. As a result of restructuring, the balance sheet size of the utility fell from Rs 33,800 crore to Rs 14,500 crore; the debt equity ratio fell from 23:1 to 3:1; net receivables for sale of power declined from 440 days to 61 days because of provision made for doubtful receivables; and payables on power came down from 615 days to 52 days (World Bank, 2000). The restructuring has clearly improved the financial viability of the sector. Tariff rationalization process is underway. Cross-subsidies are getting reduced. For example, the cross-subsidization by railway traction has fallen from 47 per cent in 2000–01 to 30 per cent in 2002–03; similarly, the cross-subsidy received by domestic consumers fell from 41 per cent to 25 per cent over the same period. Further, tariffs in successive years are reflecting increasing levels of efficiency on the part of the utility. Finally, the subsidy as determined by the UPERC is being paid by the government on a regular basis. Clearly, as a result of the on-going rationalization, tariffs are sending less distorted signals for production, maintenance and use.

3.2 Impact on performance While the immediate results of the facilitating measures have been in the right direction, they have not been translated into better performance—financial or technical—of the utilities. The UPPCL, like its predecessor (the UPSEB), continues to be in financial trouble. Total accumulated loss of consolidated UPPCL had risen to Rs 5072 crore in March 2003, up from Rs 3753 crore in March 2002 and further to RS 7400 crore (estimated) by January 2005 (see Table 8.1). Table 8.1: Cumulative commercial losses of consolidated UPPCL (Rs crore) Jan 2000 Mar 2000 Mar 2001 Mar 2002 Mar 2003 Mar 2004 Jan 2005 0 142 2353 3753 5072 6156 (estd.) 7400 (estd.) Source: PWC 112 | Indian Infrastructure: Evolving Perspectives

In the past few years, the UPPCL has not been collecting enough revenue to even pay for its power purchases. A large part of the commercial losses of the UPPCL can be attributed to the repeated failure of UPPCL to reach target levels of T&D losses and collection efficiency—which are the basis for tariff setting and which UPPCL has committed itself to. The T&D loss has fallen from 41.5 per cent in 1998–99 to 33 per cent in 2003–04 and collection efficiency has risen from 78 per cent in 2000–01 to 85 per cent in 2003–04 (Table 8.2). Even this slow and modest improvement claimed by UPPCL is suspect. The UPERC, for example, has raised doubts about the UPPCL’s claim relating to loss levels in 2003–04. The performance can be evaluated only when actual consumption data for the whole year is available to the Commission. Further, the sharp fall in revenues as compared to the approved levels in the tariff order … does raise serious doubts about the maintainability of the stand of the licensees that the loss position has considerably improved, as compared to the previous year. Till the time that there is credible estimation of unmetered consumption and the billing data on slab-mix can be relied upon, the stand of the licensees has little value. (UPERC Tariff Order 2004–05) In terms of physical performance parameters, while there has been moderate improvement in PLF and oil consumption since 2000–01, the increase in generation has been insignificant (Table 8.2). Table 8.2: Performance parameters 2000–01 2001–02 2002–03 2003–04 Actual Target Actual Target Actual Target Estd. T&D loss (%) 39.0 33.4 41.0 31.3 36.0 30.4 32.8 Collection efficiency (%) 78.3 85.0 81.0 88.0 79.0 91.0 84.0 AT&C loss (%) – 45.9 52.7 41.4 49.2 36.7 43.6 Generation (billion units) 19.6 – 20.5 – 20.9 – 20.7 PLF (%) 57.2 – 59.8 – 61.2 – 60.2 Oil consumption (KL/MU) 2.7 – 2.3 – 2.2 – 2.1 Note: AT&C: Aggregate Technical & Commercial Generation, PLF and oil consumption relate to thermal plants. Source: UPPCL, UPRVUNL and UPERC’s tariff orders (2001–02, 2002–03 and 2003–04) Further, UPPCL has not made satisfactory progress in most of the directions issued by the Commission, which ranged from introduction of MCBs to database Power Sector: The Case of UP | 113 management (UPPCL Tariff Order 2003–04). It has also failed to honour the commitments it made to the Government of India as per the MOU signed in February 2000 (See Box 8.1). For example, although the MOU required the UPPCL to introduce online billing in 20 selected towns by March 2001, only one locality of the city of Lucknow is reported to have made some progress by that date.

Box 8.1: Memorandum of Understanding with GOI

The Government of India has signed a Memorandum of Understanding with the UP government to facilitate further reforms in a time-bound manner. The memorandum signed on 24 February 2000, inter alia states: Energy audit will be undertaken at all levels in order to reduce system losses. This would be done in a time-bound manner with the following milestones: 1. Installation of metering at all 11 kV feeders by September 2000. 2. 100 per cent metering of all consumers by December 2000. 3. Online billing in 20 selected towns through computerization by 31 March 2001. The Government of India would provide financial assistance/loans to the tune of around Rs 7000 crore for renovation and modernization of thermal generation stations, repair and maintenance of hydro-electric stations, repairing critical transmission and sub-transmission lines, etc.

Source: http://powermin.nic.in/

4. OTHER INCENTIVES AND UTILITIES’ RESPONSE

4.1 Generation The Commission allows return on equity at the rate of 14 per cent for UPRVUNL plants if stations operate at higher than UPERC benchmark PLFs and 80 per cent, whichever is higher, and at the rate of 8 per cent when they are operating at or higher than UPERC benchmark PLF, but lower than 80 per cent. In the financial year 2003–04, four out of seven generating stations of UPRVUNL did not qualify for any return on equity (Table 8.3). Similarly, the Regulatory Commission has considered it appropriate to provide incentives (to be assessed on the basis of prescribed norms) to the generating stations for better performance. Incentives are determined on the basis of actual performance as compared to the benchmark PLFs. In 2003–04, only one generating station (Anpara B) qualified for incentives (Table 8.3). 114 | Indian Infrastructure: Evolving Perspectives

Table 8.3: Performance of UPRVUNL generating stations

Name of the Actual Projected Benchmark Return on Incentive station PLF 2003–04 PLF 2004–05 PLFs equity (%) (%) (%) (Rs crore) (Rs crore) Harduaganj 22.3 24.4 25.0 0.0 0.0 Panki 50.1 49.0 49.0 3.2 0.0 Paricha 33.9 53.0 53.0 4.2 0.0 Obra A 20.2 30.0 50.0 0.0 0.0 Obra B 62.2 57.7 65.0 0.0 0.0 Anpara A 78.9 77.7 80.0 0.0 0.0 Anpara B 86.7 80.8 80.0 177.6 4.1 Note: For estimating revenue requirement on account of return on equity and incentives for 2004–05, projected PLF for 2004–05 (based on actual PLF in 2003–04) was compared with benchmark PLF. Source: UPERC Tariff Order 2004–05

4.2 APDRP The GOI’s Accelerated Power Development and Reform Programme (APDRP) aims at using the fiscal leverage of the GOI to encourage reforms at the distribution level. Funding under the APDRP has two components: the incentive component and the investment component (for upgradation and modernization of sub-transmission and distribution networks). The incentive component rewards the utilities for actual cash loss reduction by way of grants (50 paisa for every 1 rupee reduction), while the investment component makes resources available for investment geared towards cash loss reduction. As part of the investment component, the GOI provides an assistance of 50 per cent of the project cost, of which 25 per cent is a grant and 25 per cent a loan.4 Under the incentive component, UP has not benefited in any single year because of its inability to reduce its cash losses. As regards investment component, in 2001–02, upgradation projects at a cost of Rs 124 crore were sanctioned with the target date of completion being March 2004. Of this, Rs 30 crore has been released by the GOI, but no expenditure appears to have been incurred so far. Similarly, in 2002–03, upgradation projects worth Rs 306 crore were sanctioned and the GOI has already released money against the above sanction. But the state-owned discoms have been unduly “tardy about utilizing the funds” (UPERC Tariff Order 2004–05). Power Sector: The Case of UP | 115

4.3 Captive power In UP, the industrial sector is one of the largest consumers of electrical energy. A number of industries are, however, relying on their own generation (captive and cogeneration) rather than on grid supply, primarily because of: • Non-availability of adequate grid supply • Poor quality and lack of reliability of grid supply • High tariff as a result of heavy cross-subsidization. Realizing their inability to meet the demands of the industry, state governments (including UP) have traditionally been taking policy initiatives to promote captive power, but not to the extent that it would paralyze their respective utility. UP, for example, has been following a transparent, but restrictive captive power policy. The threat of loss of revenue due to existing or new industries opting out of the grid for self-generation is generally expected to create competitive pressures, especially when the sector is corporatized. In that sense, the captive policy is supposed to create incentives for the utilities to improve their performance. But this has not been borne out by experience. UPPCL has made little attempt to improve availability, quality and reliability of grid supply to retain its (existing and potential) high-paying customers. Instead, it has responded on certain occasions by urging the regulatory authorities (unsuccessfully) not to permit the creation of captive capacity. As a result, captive capacity continued to grow from an estimated 1240 MW in 1998 (Captive Report 1998, Power Line Research) to 1907 MW in 2003 (Central Electricity Authority), while addition to capacity by state generating companies (and earlier by UPSEB) during the period was negligible.5 This has been the case despite the fact that the industrial tariff over the past few years has been relatively stagnant! 4.4 One-time settlement of UPSEB dues Following the formulation of a well-designed scheme by an expert group set up by the GOI, recommending a one-time settlement of outstanding dues (as on 1 October 2001), a tripartite agreement (between each state government, GOI and RBI) incorporating the scheme is in operation.6 The key feature of the scheme is that it brings into focus the payment of current dues in future by linking it to the settlement of outstanding dues through an incentive mechanism. If states adhere to some specified conditions, which include making timely payments of current dues in future and achieving certain performance milestones, 60 per cent of the surcharge currently outstanding will be waived and some cash incentives will also be given to them.7 If, however, they default, they would be penalized through graded reduction in the supply of power from central power stations and through suspension of APDRP grants.8 116 | Indian Infrastructure: Evolving Perspectives

UP is one of the many states that have signed this agreement. Consequently, power purchase payables of erstwhile UPSEB to central generating stations have been securitized. As the state government is servicing this liability, the burden on the sector has been considerably reduced. But, how has the UPPCL responded to the incentive system underlying the agreement? The nature of the response can be gauged from the UPPCL’s submission to the UPERC in 2003 that it purchased less power during 2002–03 than its own projection, so as to meet the payment conditions in the tripartite agreement. The UPPCL did this by cutting down power supply rather than executing measures to improve T&D losses and collection efficiency. Furthermore, while the UPPCL’s payables to central PSUs have remained under control, its payables to state generating stations have tended to go up.

5. ASSESSMENT Reform initiatives taken in recent years have been in the right direction. While financial restructuring and one-time settlement of UPSEB dues have substantially reduced the burden of past liabilities of the utilities, making the utilities more amenable to future reforms, initiatives such as unbundling and public scrutiny of tariff proposals have resulted in greater transparency. Tariff setting has been substantially insulated from political interference and some degree of tariff rationalization has been achieved. There is no doubt that these measures have together created a facilitating framework. A number of incentives are currently in operation to complement the framework. Some of them are designed to improve overall performance, while others are in specific areas. The incentives are largely well-designed and a strong and positive response by the utilities would have certainly helped increase sector efficiency. But the utilities’ response has been weak and often perverse. Even though a number of years have passed since these measures were taken, the sector efficiency has remained abysmally low as evidenced by grossly inadequate investment, high T&D losses, low collection efficiency and consumer dissatisfaction. Why has this been the case? 5.1 Contrast with Delhi The answer is illustrated most strikingly by contrasting UP’s response to the MYT approach to that of Delhi. It has been noted that although UP had adopted MYT in 2002, privatization of distribution is yet to occur. It was made clear to UPPCL from the beginning that underperformance (vis-à-vis targets) would lead to commercial loss in any given year (because tariff for a particular year is set by the regulator by taking into account pre-determined performance targets) and would also make the challenge for the next year even tougher. Yet, the UPPCL not only underperformed persistently in the face of progressively stiffer targets (see Table 8.1), but also failed Power Sector: The Case of UP | 117 to comply with even the routine directives given by the UPERC. The UPPCL’s explanation for its repeated poor performance has been “an attempt to blame extraneous factors for its … low level of efforts” (UPERC Tariff Order 2003–04). Delhi, on the other hand, decided to privatize its distribution zones at about the same time that it adopted the MYT approach. The experience of the past two years shows that the privatized distribution zones met the targets and in some cases, exceeded them (see Table 8.4). Table 8.4: Reduction of AT&C loss in North Delhi Power Ltd (% points) July 02–March 03 2003–04 2004–05 Committed reduction 0.5 2.3 4.5 Effective reduction 2.6 6.2 6.1* *Up to August 2004 Source: North Delhi Power Ltd The contrast weakens the argument made by some that it is too early for UP to expect any substantial improvement in operations, and that the investment in recent years in the primary and secondary systems (including metering of feeders, implementation of energy audits, etc.) would show results only in the coming years. More importantly, the contrast provides evidence that MYT system can hardly work as an incentive scheme in a setting such as UPPCL, which lacks commercial orientation.

5.2 Contrast with NTPC As regards efficiency in generation, a contrast of UPRVUNL stations with those of NTPC, which have similar kinds of incentive systems, can be illustrative. The PLF of NTPC power stations have historically been much higher than those of UP state generating stations. In 2003–04, for example, the overall PLF of NTPC was 84.4 per cent as compared to 60.2 per cent for UPRVUNL. One major factor explaining the difference in performance is resource (revenue and capital) availability. Resource abundance helped NTPC—whose revenues are now protected by the tripartite agreement—to respond positively to the tariff system, which linked profitability to physical performance. A large part of the rising profits were ploughed back into investment, encouraging lenders to lend more and at fine rates. In contrast, UPSEB and later UPRVUNL lacked resources, primarily due to absence of reforms at the distribution end. It is well known how this has led to a vicious circle of deteriorating performance. It may be pointed out that a frontier production function model study carried out by the UPERC shows that on an average, the power stations of UPRVUNL can 118 | Indian Infrastructure: Evolving Perspectives increase their existing output levels by 37 per cent without additional resources, simply by proper utilization of technology and adoption of best practices. But the study perhaps presupposes a well-trained and motivated work force and management and a corporate culture (such as that of NTPC), which UPRVUNL does not have. It is, of course, debatable whether it is possible to have good training, sound human resource management and a corporate culture in the face of acute and persistent shortage of resources.

5.3 Is open access the answer? Can the problem be taken care of by open access per se? It is expected that as open access is phased in as envisaged in the Electricity Act 2003, competitive pressures would be created. If, however, the generation and distribution assets remain predominantly with the government, the government would have a vested interest not only in delaying the phase-in of open access, but also in rendering it ineffective to the extent possible. Even if open access succeeds in facilitating the creation of private generating capacity, which in turn attracts away high paying consumers from government-owned distribution companies, it is doubtful that the latter would be motivated to improve performance to remain competitive, as evidenced from the utilities’ experience relating to captive power. In fact, the open access regime would entail lesser losses (and therefore weaker pressure) for the utilities than self- generation, as the former requires cross-subsidies to be paid by consumers to the affected distribution company, while the latter does not.

6. SHORTCOMINGS OF THE CURRENT DISPENSATION Shortcomings of the current dispensation that impede adequate response of utilities to reform stimuli can be mainly attributed to government ownership, as can be seen from the discussion given below.

6.1 Political patronage There are major shortcomings in the accountability framework in government- owned utilities, even though they are managed by a board. Corporatization does not help in altering the orientation of accountability from internal hierarchy to power consumers or regulators. Year after year the regulator has been reprimanding the utility for its poor performance, but to no avail. At the root of the problem is the political patronage of payment indiscipline, which has been possible mainly because of government ownership. Given the political patronage, there is hardly any incentive for the management to bring dishonest staff to book or to cut off connections to non-paying consumers. Besides, the maximum punishment given to government employees is usually not deterring enough. Under such circumstances, tariff rationalization can hardly serve any useful purpose: neither can it create incentives Power Sector: The Case of UP | 119 for consumers to use electricity more efficiently nor can it boost revenues for operators to expand access or improve services. 6.2 Government interference The government has continued to interfere in the day-to-day operations of the newly formed corporations, whose managements hold the same bureaucratic attitudes and promote the same organizational cultures as before. Their relationship vis-à-vis the state government has also remained unchanged. For example, the government of UP, in an effort to stall tariff increase, had given a direction to the UPPCL to file their tariff application to the SERC for 2000–01 with reduced T&D loss target, without giving any strategy for achieving the target. The utility, being a government- owned company, had to oblige.9 Although the immediate result was that the tariff hike was moderated, ultimately, the T&D losses remained at the previous year level and the UPPCL incurred large commercial losses. Not surprisingly, with government interference eroding the autonomy of the utility, it has been difficult to establish accountability for the utility’s performance. 6.3 Dual role for government The government’s role as a consumer of power compromises its position as an operator. The government is itself one of the biggest defaulters. The collection efficiency of the government category has been fluctuating over the years: for example, it fell from 87 per cent in 1999–2000 to 42 per cent in 2000–01. True, the government is a relatively small consumer of power accounting for 11–12 per cent of the total bill and therefore cannot possibly drag down the overall collection efficiency substantially. The important point is that the government, which is constrained by its fiscal situation, cannot provide moral leadership in payment discipline because of the poor example it sets for other consumers. Table 8.5: Collection efficiency (%)—governmental and non-governmental categories Billing 1997–98 1998–99 1999–2000 2000–01 2001–02 Government 74 52 87 42 52 Non-government 87 86 84 83 82 Overall collection efficiency 86 82 84 78 78 Source: UPPCL/PWC 6.4 Erosion of hard budget constraint The government ownership has led to an absence of hard budget constraints. It has been noted that UPPCL’s payables to the state government’s generating stations has 120 | Indian Infrastructure: Evolving Perspectives been rising (from 112 days’ power purchases in 2003–04 to 169 days’ by September 2004), while those to the central PSUs have remained under control (30 days). Similarly, in the repayment of loans, UPPCL has given the lowest priority to the government of UP amongst all its lenders. Yet, the government continues to be its dominant lender.

7. WAY FORWARD The government had formally recognized that privatization of the distribution business was critical to the viability of the sector in its Power Sector Reform Policy Statement in January 1999. In fact, the privatization of Greater Noida was done as early as 1992 and the results were encouraging (see Box 8.2). There was a subsequent attempt to privatize distribution in Kanpur. While the first attempt by the state government to privatize KESCO (Kanpur Electricity Supply Company Ltd) was unsuccessful, the subsequent decisions to invite private bids have been postponed several times (see Box 8.3). Recognizing that privatization is the answer is not enough; the task has to be implemented quickly. While privatization is getting delayed, commercial losses of the UPPCL have been mounting and the benefits of the balance sheet clean-up in 2000 are getting wiped out (Table 8.1). Since at the time of privatization, these losses would have to be dealt with, delays in privatization will increase the financial burden on the government. Box 8.2: Noida Power Company (NPCL)— a successful distribution company

Background NPCL is the first private distribution company in India, which took over a network from a state undertaking. It was jointly promoted by New Okhla Industrial Development Authority (Noida) and Greater Noida Industrial Development Authority (GNIDA) in 1992 to take over distribution of the new industrial township. Currently, NPCL has an equity base of Rs 9.2 crore, of which 73 per cent is held by the RPG group and the balance by GNIDA. Performance The company inherited a dilapidated distribution network, inadequate to meet the rising load growth. Through extensive operational revamping and high consumer focus, the company has been able to achieve a turnaround. Between 1994–95 and 2002–03, its asset base has grown from Rs 14 crore to Rs 60 crore and sales revenues from Rs 19 crore to Rs 70 crore. Its T&D loss level has been consistently about 8 per cent, one of the lowest in the country. NPCL also has one of the lowest distribution manpower cost (at Rs 0.05 per unit sold). In 2000–01, the company made a net profit of Rs 2 crore, up from Rs 0.5 crore in 1996–97. Power Sector: The Case of UP | 121

Minimizing revenue loss To minimize revenue loss, the company follows a thorough energy auditing process, which entails aggregation of the quantum of energy consumed in downstream distribution on a periodic basis for reconciliation with input energy. The 11 kV feeders are provided with electronic meters at substations, which enable accurate assessment of energy sent out to the system. To develop the rural distribution network, NPCL has developed the concept of “cluster supply” in villages, whereby multiple small-sized transformers are introduced for providing supply to localized groups of consumers. By extending the high tension network to almost the doorstep of consumers, NPCL has reduced energy pilferage opportunities. The company’s consumer focus is reflected in the fact that connections are activated within 6 days of application for domestic consumers and 15 days for industrial consumers. Source: NPCL Annual Report (various issues), UPERC Order, 2003–04, Prayas Occasional Report No.2 (2003)

Box 8.3: KESCO privatization

The government of UP indicated its intention to privatize power distribution in Kanpur city in the first quarter of 1999. Although more than five years have since passed, distribution in Kanpur is yet to be privatized. In April 1999, the government had pre-qualified four bidders for the privatization procedure—BSES Limited (BSES), Calcutta Electricity Supply Company Limited (CESC), Larsen & Toubro Limited and AES Combine (L&T–AES) and Tata Electric Companies (TEC). The bidders sought and obtained a postponement of the final date for submission of bids until after the issue of the first tariff order—which came in July 2000—since bidders (rightly) expected future viability of KESCO to be contingent on regulatory decisions on a number of issues such as the bulk tariff payable by KESCO to UPPCL, the consumer tariffs chargeable by KESCO and the allowable level of T&D losses. Bidding took place in July 2000. However, since only one company submitted its bid, the bid was not opened. Since then, although the bidding deadline has been postponed a number of times, bidders have not responded.

Source: Tadimalla, Sri Kumar, “Privatization of Kesco—A Case Study”, 2000 To extract best results from privatization, distribution zones need to be appropriately designed. Two methods are generally considered: mixed zones and concentrated zones. The latter is a superior method, for the following reason. The option of claiming subsidy encourages distribution companies in mixed zones—regardless of whether they are government-owned or owned by private players—to camouflage theft and inefficiency rather than to improve distribution efficiency, by over- reporting consumption of subsidized categories, and thereby raising the subsidy burden on the government.10 (Concocting false consumption data is particularly 122 | Indian Infrastructure: Evolving Perspectives easy in states that have a large number of agricultural consumers, such as UP.) Such options do not exist for concentrated zone distribution companies, who, by definition, would have no access to subsidy flows. In fact, these zones can be made to cross-subsidize rural zones through a transparent electricity surcharge. Mixed zone privatization would thus weaken the motivation for the distribution companies to respond strongly to the incentive systems. UP had started the right way by attempting to privatize KESCO. Potential private investors showed little interest because of the lopsided risk allocation that was attempted and the absence of regulatory certainty, and not due to the fact that KESCO was a relatively small, concentrated zone. It would not be appropriate to abandon this strategy in favor of mixed zone model for distribution, as UP appears to have been doing. Unless UP rectifies this flaw at this stage, it would lose substantial benefits of privatization.

8. CONCLUSION As stated earlier, there are two areas that have remained untouched by the recent reform process: ownership arrangement and market structure. While the Electricity Act, 2003 entails provisions to radically alter market structure, it allows a number of options for ownership and does not mandate any changes in the existing ownership structure. So, UP as well as a number of other major states, which are in the same stage of reform as UP (such as Rajasthan, Karnataka, Haryana and Andhra Pradesh) are well within their rights to continue with government ownership of distribution business. The case of UP, however, shows that it is futile to attempt to achieve higher productivity through multi-year tariff regime and other incentive schemes if the distribution business continues to be owned by the government. Furthermore, it would be naïve to believe that with the onset of open access, the sector efficiency will increase even if distribution is not privatized. The UP experience shows that continued government ownership would lead the newly formed utilities to deeper and deeper financial trouble, while the sector would continue to ail. There is even a danger that reforms may be discredited. To make utilities more responsive to incentives and to take advantage of the upcoming open access regime, states need to privatize distribution at the earliest and do it the right way.

NOTES 1. The right to distribute power in Greater Noida was sold in 1993 to Noida Power Company Ltd. 2 . The process of tariff setting on the basis of performance targets set each year by the regulator increases the uncertainty of investor/utility about their respective future revenue Power Sector: The Case of UP | 123

streams. This is borne out by the KESCO privatization exercise. The practice also increases the burden on the financial and human resources of the utilities. Finally, such a process fails to offer correct incentives for a long-term view of investment, maintenance and use. A multi-year incentive-based approach to regulation can rectify these shortcomings. 3. They are Varanasi, Agra, Lucknow and Meerut distribution companies. In addition to these four, UP has two more discoms operating at Kanpur and Noida respectively. 4. The balance 50 per cent is to be arranged by the utilities either through internal resource generation or as counterpart funding from financial institutions such as the Rural Electrification Corporation and Power Finance Corporation. 5. The UP Government Energy Policy 2003 states that the captive power capacity in UP is higher than the industrial load contracted with the grid. 6. For details, see Report of the Expert Group on Settlement of SEB Dues, March 2001. 7. The balance arrears would be securitized through tax-free bond issued by respective state governments. 8. If defaults exceed 90 days from the date of billing, the Ministry of Finance should recover these dues through adjustments against releases due to them from the centre. 9. The SERC, on its part, had even felt that the target spelt out in the tariff application was inadequate and called for even higher loss reduction target. 10. To scrutinize the validity of the claims for subsidy by distribution companies, the regulator will have to verify the actual consumption by subsidized categories, which is a very cumbersome exercise. 124 | Indian Infrastructure: Evolving Perspectives

DISCUSSION PAPER ON DEVELOPING POWER 9 MARKETS September 2008

1. INTRODUCTION In the electricity sector, the totally steel-jacketed arrangement of long-term contracts which defined the rules of the game for several years has not allowed a competitive market structure to develop. Even though power trading has been recognized as a distinct licensed activity under the Electricity Act, 2003, the volume of trade has been so insignificant and far from a critical mass that it has not created any visible impact. Though the National Electricity Policy, 2005, provides for 15 per cent of the total capacity to be developed as merchant capacity which could play a meaningful role in market development, even this alternative has not taken off in a credible way. Therefore, definite action needs to be taken to ensure that (a) merchant capacities develop at least to a level of about 15 per cent of the total capacity, (b) trading is encouraged to occupy a much larger space, thereby giving options to distribution licensees for competitive procurement, and (c) availability of open access, firstly to the transmission system and subsequently to the distribution network, does not emerge as a constraint on facilitating these processes.

2. MERCHANT POWER PLANTS The Electricity Act envisages the bulk of the development of power plants through long-term power purchase agreements (PPAs) with tariff determination as prescribed under Sections 62 and 63 of the Act. Consistent with the provisions of the Act, the Guidelines for Determination of Tariff by Bidding Process for Procurement of Power by Distribution Licensee stipulates PPAs for periods exceeding one year. Section 62 of the Act also permits sale or purchase of electricity between a generating company and a licensee or between two licensees for a period not exceeding one year. Section 66 of the Act explicitly provides for development of the power Developing Power Markets | 125 markets. With a view to developing the electricity market, it would be essential that while there are power projects which are developed on the basis of long-term PPAs, lasting over project life cycles under which capacities developed are fully tied up with the procuring agencies, there are also capacities developed to cater to needs which are short-term in nature. The competitive advantage of the electricity market would accrue to consumers only when a reasonable quantum of merchant power generation from a number of producers is also available for purchase on the basis of competitive tariff through trading or equivalent arrangements. Some of the key benefits of setting up merchant power plant capacities, inter alia, include: a. Merchant power plants provide virtual capacities for regions/areas in need of short-term power due to temporary demand–supply mismatches. b. The entire risk for offtake of power is carried by the merchant power plant, thereby obviating the need for procurers to enter into firm power purchase agreements for their requirements. Procurers could enter into firm PPAs for their base load requirements and meet the peaking, short-term requirements from merchant power plants. c. Power plants operating on a merchant basis, necessarily need to ensure that the power produced is cost-effective/reliable, failing which these capacities will not be dispatched in preference to other available power supplies. d. These power plants can also help meet peak load demand in the system. Development of merchant power capacities to the extent of about 15 per cent of the installed base would go a long way in terms of development of trading, introduction of competition, and development of the electricity markets. Project-specific merchant capacity should be determined on the basis of the location of the project, type of project fuel, etc. However, there are certain issues hampering the development of power projects in general, which are discussed below: a. In the recent past, a number of promoters have announced power generation capacities in the range of 1000 MW or above. The majority of these capacities are based on coal or hydel and are expected to be developed in the eastern part of the country for coal (Jharkhand, Chhattisgarh or Orissa) and in the states of Arunachal Pradesh, Himachal Pradesh, Uttaranchal or Sikkim for hydro power projects. Currently, many of the proposed generation capacities are in the development stage and as such are yet to tie up any offtake of power. Typically, most of these projects would tie up offtake for 60 per cent to 70 per cent of the capacity with the balance capacity to be committed on a merchant basis. The tie-up of 60 per cent to 70 per cent of the capacity on 126 | Indian Infrastructure: Evolving Perspectives

long-term PPA basis will be essential to service the debt for the project. In the case of hydro projects, government policy recently notified 40 per cent of the capacity on merchant basis. At the time when promoters approach financiers for tying up equity/debt funding, the generation capacities have not tied up long-term offtakers of power. The tie-up of capacity on long-term basis with the offtakers, in some cases, may take considerable time, which may delay the financing/equity tie-up. The Guidelines for Determination of Tariff by Bidding Process for Procurement of Power by Distribution Licensee makes it mandatory for the state discoms to competitively procure the power for their requirements more than one year. The procurement process of the discoms is carried out under Case I bids (where the location, technology or fuel is not specified by the procurer) or Case 2 bids (for hydro power projects, load centre projects or other location-specific projects with specific fuel allocation, such as captive mines available, which the procurer intends to set up under the tariff-based bidding process). The time taken for finalization of the bids is in line with those stipulated by the Ministry of Power (MoP)—(240 days for single-stage bid to 425 days for two- stage bid (RFQ and RFP)). Additionally, the bids called by various states do not occur within a specific time period and can occur any time during the year. These factors add to the uncertainties of the outcome associated with the bidding process, wherein the promoter may not be successful in some or all the bids, which themselves are a time-consuming process and therefore may extend the planned financial tie-up date. The financiers/investors would need to evaluate the credit risk associated with the offtakers, and hence it becomes imperative that the intended capacity tie-ups (60 per cent to 70 per cent) occur prior to financial closure. If it is possible for the power plant developer to indicate tied-up capacity and the states in the respective regions, then this would enable the CEA and the CTU to prepare a plan for additional transmission capacity needs of these developers. b. The delays in tying up of firm capacities may have implications in terms of evacuation arrangements for the project. The inter-regional transmission infrastructure may not be adequate to ensure evacuation of power primarily from the Eastern/North-Eastern to the North/West and Southern regions, assuming that a significant part of the capacities planned in these regions materialize. The existing transmission capacities are tied to generation capacities which evacuate power from specific power plants to different regions. The transmission capacities that are expected to be developed within the next few years (Eleventh Plan) will evacuate the power from identified projects (78,000 MW planned to be developed in the Twelfth Plan period) to specific Developing Power Markets | 127

regions with marginal redundant capacities. The process of planning and developing evacuation infrastructure, by the Central Transmission Utility (CTU), requires the identification of the project location and the offtaker. This process of planning and development of the transmission infrastructure may take about 30 to 36 months from the time the offtaker is identified for a specific project. In a situation wherein the developer is unable to tie up the offtaker of the capacity within a suitable timeframe, it will lead to delays in the development and commissioning of evacuation infrastructure. Such delays in establishing evacuation infrastructure will lead to part or the complete power generation capacity being commissioned prior to the transmission infrastructure being operational and create a stranded asset for a limited period of time. If developers, at the time of commencing the development activities, indicate tied-up capacity and the states in the respective regions, it would enable the CEA and the CTU to plan for additional transmission capacity needs of these developers. In the case of coal-based plants that are coming up in Orissa, Chhattisgarh and Jharkhand, which will become the hub for the new capacities of the IPP developers, a two-track policy approach may be required. First, there is need to have a dedicated line up to the identified pooling point, and its cost should be shared by the developers. Second, from the common pooling point a separate common corridor should be planned, and its cost should be shared by all the states in the region. This approach may be further discussed with the IPPs, CEA and the states concerned in the region. c. In the recent past, the Ministry of Coal, Government of India, has allotted coal mines to private developers of power projects. The developers are expected to use the coal from the mines to operate their power plants. The process of developing the mine ranges from conducting geological investigations, submission of mining plans, obtaining statutory approvals to land acquisition and obtaining mining lease. This process of development of the mines is time- consuming, and can take about four years before the extraction of coal from the mine can commence. The power project is, however, expected to commence generation prior to the extraction of the coal from the allotted mine considering the timelines involved in the development of the power project and the coal mine development. In such a situation, the developer is expected to have some alternative tie-up for fuel till such time as the captive mine is operational. d. Land acquisition is one of the most critical activities that can delay project development substantially and have major implications on the project costs. Currently, land for the private power projects is being acquired by the state governments, and the procedures identified under the Land Acquisition Act 1894 are followed. These procedures are enshrined in various sections, starting 128 | Indian Infrastructure: Evolving Perspectives

from Section 4(1) which relates to public notice by the state government informing the general public of the land identified for acquisition in the public interest, followed by Section 5(1) and Section 6(1) notifications which deal with identification of the land records, and fixing and approval of the compensation. There are further notifications which need to be made and finally culminate in obtaining a notification from the state government under Section 17(1), wherein the possession of the land is actually handed over to the developer. The entire process of land acquisition by the state government can take up to two years, assuming the processes are not marred by controversies/litigation as has been observed in some of the projects being implemented in Orissa. The delays in acquisition of land in turn impact on the receipt of various statutory clearances/licences (mining lease), etc. that further delay the projects.

Recommendations The importance of merchant power projects for the development of the power sector in India cannot be disputed. It should be ensured that capacity additions in merchant generation take place at a rapid pace. If the stakeholders ensure that visibility with reference to the variables under their control is enhanced, as have been identified herein below, thereby reducing the associated risks, the process of funding of the projects by both the equity investors and the funding institutions would be smoothened. A. Developers: a. Land acquisition: Acquisition of land is a time-consuming process. It is suggested that the developers acquire at least 40 per cent of the land with a firm schedule to acquire the balance land if the developer is acquiring the land on its own or deposit about 80 per cent of the cost of the land (with the state government) if the same is acquired by the local authority for the power project before approaching the funding agency (equity investor/lending institution). The above will provide visibility as regards the land acquisition timelines and eliminate the land-related variables in the risk perception of the funding agency. b. Environment: Prior to approaching funding agencies for financing (equity and debt), the developer should ensure that the following activities have been completed: i. Terms of reference of the environment study are approved; and ii. Public hearing process is either completed or is scheduled. The above will obviate to a certain extent the risks associated with the visibility on the key statutory clearances. Developing Power Markets | 129 c. Firming up power offtake: The funding agencies (equity and debt) place a great deal of importance on the credit risk associated with the offtaker of power. For the projects supplying power to financially stressed discoms, the equity investors are willing to participate in the project at relatively higher rate of returns and the financial institutions providing debt do so with additional covenants and at higher rates of interest. For mega power projects (greater than 1000 MW capacity), the developer should make firm arrangements through medium-long-term power purchase agreements for 60 per cent to 70 per cent of the capacity at the time of approaching the funding agencies for tying up equity and debt. The balance could be left open, and tied up progressively over the next few months with discoms which are not financially stressed. In the case of small to medium projects (less than 1000 MW), the percentage of power offtake to be tied up at the time of the developer approaching the funding agencies could be lower at 40 per cent to 50 per cent, with the balance being tied up later. The above provides visibility to the funding agencies as regards the credit risks associated with the project during its operations phase. For Case I bidding, however, no PPA could be even discussed unless the bidder quotes a price and wins the bid. The bidder cannot quote a price unless he has an idea of cost of funds. Therefore, the requirement of offtake should be made a pre-disbursement condition and not pre-commitment condition. This approach will facilitate capacities through Case I bidding. d. EPC/major equipment contract(s): The ability of the developer to ensure successful implementation of the project to a greater extent rests on the selection of the contractor. The developer of the merchant project, by selecting the key equipment supplier or the EPC contractor (or at least placing the letter of intent), will help in the assessment of the construction risks on the basis of the reputation of the contractor. Hence it is suggested that the developer should have signed the agreement with the proposed equipment supplier(s) or the EPC or package contractor prior to any loan disbursement. Issuance of the letter of award of the main plant should be a precondition to loan finalisation. e. Coal availability: The list of activities to be carried out post-allocation of the coal block ranges from conducting geological investigations, submission of mining plans and obtaining statutory approvals to land acquisition and obtaining mining lease. These processes can take up to four years to conclude prior to commencement of mining operations. In the event of part of the power plant capacity achieving commercial operations prior to commencement of mining activities, the fuel source needs to be firmed up. Developers should have a clear-cut plan on the timelines within which they will obtain all necessary approvals or have a back-up in terms of a coal linkage. In addition, the 130 | Indian Infrastructure: Evolving Perspectives

developers of merchant capacities should have provided a bank guarantee of the requisite amount to the Ministry of Coal prior to approaching funding agencies for tying up equity and debt.

B. Regulators and Central Government/ministries a. Coal linkages: In some of the coal blocks which have been allocated, the timelines involved in land acquisition and obtaining all clearances leading to mining are very long. This would lead to part or total capacity achieving commercial operation prior to the commencement of mining activities. It is recommended that in such cases the ministry of coal should provide tapering linkages to projects where there has been serious project development work and where delays in commencement of mining operations and extraction of coal from the allotted mine(s) are anticipated. b. Open access: As of now open access in transmission is available either for a period of up to 3 months or greater than 25 years. This poses a problem to developers as they need open access for short/medium terms and for varying capacities to sell power from their proposed project. It is recommended that open access be made available for varying terms: short term, up to 3 months; intermediate term, 3 months–5 years; medium term, 5 years–15 years; and long term, over 15 years, to merchant power plant developers to enable them to dispatch their power for varying terms and capacities to offtakers for the untied capacities. c. Exemption from cross-subsidy in the case of financial event of default: Currently, the power procurement for the state discoms is handled by the aggregator(s) at the state level who invite(s) bids for power on behalf of all or some of the discoms. The financial position of all the discoms, for whom the state aggregator may procure power may not be similar. This will result in some of the financially weak discoms incurring a financial default. Though the model PPAs stipulate third party sale, in reality, the sale of capacity defaulted on by the discom is generally made to another discom despite the existence of financially sound HT consumers in the distribution area of the defaulting discom. This is due to the prevalence of cross-subsidy surcharge. It is recommended that if a state-owned procurer defaults on the payment for power procured from any power plant, the power plant should be allowed to sell power to third parties within the state (like HT consumers), and the discom provides the wheeling services for the power of the project at wheeling charges as decided by the regulator. The levy of cross-subsidy surcharge in such cases should be waived and in other cases be decided in a rational manner, as the state procurer has already defaulted on obligations to the seller, and the surcharge should reflect the cost of efficient/ uninterrupted supply. It should, however, be noted that in order to encourage Developing Power Markets | 131

Open Access, surcharge should not be in excess of the surcharge as per the formula laid down in para 8.5.1 of the National Tariff Policy. d. Mega power status for tied capacities: The mega power policy of the MoP, Government of India, exempts the equipment for power projects above a capacity threshold (1000 MW for coal-based and 500 MW for hydel) from customs duties and excise duties, provided such projects supply power to more than one state and the beneficiary states have constituted their regulatory commissions with full powers to fix tariffs as envisaged in the Central Act. The procuring state would also have to privatize distribution in the cities having a population of more than one million. In the case of a project where at least 1000 MW of capacity is proposed to be sold through long-term power purchase agreements to utilities in more than one state, the benefits of the mega power project should be automatically accorded to such projects. This will have a clear implication for the project cost, since the taxes and duties constitute a substantial element of the cost, and consequently on the means of finance. For the merchant power projects, tying up funding on the basis of a mega power project’s benefits, in the absence of visibility on the applicability of the benefits to the project, runs the risk of an increase in the project cost if the mega power project’s benefits are not available.

C. Governments a. State governments should continue to assist the developer in land acquisition and allocation of water to the merchant power plants. b. The state governments over the past few years have had the benefit of experiencing the patterns of power demand, and supply and growth in the power demand. Various states together should be in a position to project their demand for the next few years and arrive at an assessment of supply. On the basis of this, the states can call for competitive bids for supply of power at the same time to mitigate the risks associated with visibility on the power capacity tie-ups by the developer (as highlighted above). c. MoP or the CEA could coordinate this, particularly with respect to the timing of the bids. d. An in-principle support for transmission of power based on requisition from the developer, CEA or PGCIL, could provide this comfort.

D. Financing agencies Non-finalization of loan agreements also emerges as a constraint in many cases. Financing agencies can encourage development of merchant plant/capacities for which it may be necessary to adopt the following approach for some of the critical issues: 132 | Indian Infrastructure: Evolving Perspectives

Table 9.1: Recommended loan conditions

Issues Conditions for final loan agreements Conditions for loan disbursement 1. Land Land should have been identified and The developer should Section 6(1) notification should have been have the possession of the issued for the entire land and the entire land for the project. developer should have deposited 80% of the cost of the land with the government authorities.

2. MoEF/ The terms of reference for the MoEF clearance and Forest environment studies should have been Forest clearance should clearance approved and the public hearing process have been obtained. should have been completed. In addition, the developer should have obtained consent from the state pollution control board.

3. Power Initiated a process of bidding for offtake Finalised off-take of power offtake of power through the competitive process for at least 60% to 70% of the capacity for major projects and 50% to 60% of the capacity for smaller projects (up to 500 MW).

4. EPC/Major The developer should have issued the The EPC or the major equipment letter of award for the entire scope of the equipment contracts, as contracts project (EPC basis) or major critical the case may be, should equipment for the power project. have been finalized (However, for commencing the discussions incorporating the inputs with the lenders, the promoter should have from the lenders' at least initiated the tendering process for engineer and all the major equipment/EPC contract and should conditions precedent for have achieved significant progress). effectiveness of the EPC/ major equipment contracts should have been fulfilled. Developing Power Markets | 133

Table 9.1: Recommended loan conditions (contd...)

Issues Conditions for final loan agreements Conditions for loan disbursement 5. Fuel tie-ups The developer should have obtained Letter Fuel supply agreement of Assurance from the Ministry of Coal should have been signed and deposited the bank guarantees with with the state mining the ministry towards linked supplies. agencies for projects For captive mines, in addition to obtaining where fuel linkages are a fuel linkage for the period between the available (and for projects time the plant achieves commercial with captive mines operation and the time the captive mine wherein fuel linkages shall commences operations, the developer be required for the period should have initiated the process of GR commencing from the preparation, mining plan preparation etc. project COD till the In addition, the heads of agreement should commencement of have been entered into with the fuel mining operations). supplier before the financing documents For captive mines, the GR, were finalized. mining plan should have been prepared and Currently, a single coal block is allotted to approved by the a number of allottees. It becomes essential relevant agency. that the company approaching the lenders for financing its power project has at least formed the company for the coal mining and should have entered into a share- holders agreement with the other allottees of the coal block.

6. Power The developer should have initiated The evacuation evacuation discussions with the CTU for evacuation arrangements/agreements of power from the project site to the for evacuation of power prospective states in line with 3 above. from the project to the respective procuring states should have been firmed up with the CTU.

The above will reduce to a large extent the risks associated with financing and development of merchant power plants and ensure that the capacities announced by the various developers materialize. 134 | Indian Infrastructure: Evolving Perspectives

3. POWER TRADING Prior to enactment of the Electricity Act 2003, there was no concept of trading of power. The power sector was a natural monopoly with state-owned entities, viz. State Electricity Boards (SEBs) which were vertically integrated (i.e. generation, transmission and distribution housed in one entity) and accorded the right to carry out their businesses in a geographically defined territory with pre-approved tariffs. Each SEB had an allocated share in a central/ jointly-owned power station. The entire power sector operated on a fixed return basis and interplay of market forces remained non-existent. Utilities would back down their generating stations in case of low demand and resort to load- shedding in case of excess demand. Thus the concept of power sector operations (generation, transmission and distribution) as a commercial activity which could utilize the surpluses of other regions and use the fixed assets for better returns did not exist. With a view to developing the electricity market, power trading was introduced as a concept in 2001 to tap the power from surplus regions of the country to the deficit regions. This resulted in better utilization of existing capacities by way of creating virtual capacities and also added to the cause of substantial improvement in plant load factors of the existing generation units. The initial growth of the short-term power trading market was appreciable. The share of short-term traded power increased to about 2.5 per cent, while the balance grid power continued under long-term power purchase agreements. The participants in the power sector for the first time started to look at power as a source of revenue. In the present situation, many states like Chhattisgarh, Jharkhand, Orissa, Himachal Pradesh, J&K and Uttaranchal have devised policies and planned large capacity additions to become power hubs. This will lead to rapid capacity addition, including merchant power plant capacities.

Characteristics/Advantages of power trading After due consideration over a long period of time, power trading is now recognized as a distinct activity, hence it may be worthwhile outlining the characteristics/merits of electricity as a trading activity: a. Electricity trading facilitates sale of power from a surplus region to a deficit region, mitigating shortage. In effect, electricity trading enables better utilization of existing generation assets, without requiring huge investments to be made in setting up generation assets. b. Electricity traders act like market makers. They provide a single-point specialized service and enable buyers and sellers to transact their business. In the absence of traders, both buyers and sellers shall have to incur significant Developing Power Markets | 135

expenditure for training manpower, locating sellers/buyers and establishing the trading infrastructure. c. Unlike an intermediary who does not acquire interest and incur liability, the electricity trader purchases electricity and assumes all the risks of the transaction, especially the payment risk by way of adequate payment security mechanism. An electricity trader acts as a specialist who accepts liability and responsibility for the transaction to be completed, unlike a person who gets a commission for bringing seller and buyer together. d. By assuming counter-party credit risk, electricity traders provide comfort to the seller of electricity by facilitating sale and purchase of power at a predetermined price, thereby insulating the purchaser and the seller from the financial risk of transacting through the Unscheduled Interchange (UI) market, where the frequency and rate realised are variable. Such predetermined sale of power also aids better grid discipline and leads to better grid parameters. e. With the development of the electricity markets, the number of active participants in the trading segment would increase. This would create, though in a limited way, an environment of competition in which the procurer distribution companies do have the option to choose from various trading agencies which, in turn, leads to reduced burden on ultimate consumers. Going ahead, there is a potential for credible trading agencies to act as catalysts/ facilitators of new generation capacity, wherein the trader could facilitate tie-up of the bulk of the capacity on a long-term basis with a utility and take a principal risk for the balance untied amount. f. Even the limited volume of trading activity has been able to establish the value/price of power and has brought in a commercial sense among, if not all, a large number, of state utilities.

Regulatory orders on power trading The increased demand for power as well as inadequate new generation capacity being set up has led to a rise, over the past two to three years, in the short-term price of the power being traded, causing concern to purchasing utilities as well as to the regulatory bodies. To address the situation, the CERC/Appellate Tribunal for Electricity (ATE) have passed three orders, and highlights of which are presented below: a. The trading margin, which can be charged by trading licensees for inter-state trading of power was capped at 4 paise/kWh on power traded on a short-term basis (not exceeding a year), including all charges, except charges for scheduled energy, open access and transmission losses. b. The price at which a generating company can trade power should not exceed the base price plus 4 per cent thereof. 136 | Indian Infrastructure: Evolving Perspectives c. Trader-to-trader transactions are not permitted. The above matters are sub judice, and final decisions on the same are still awaited.

Current trading scenario a. The capping of trading margin has not led to price stability. The transaction prices quoted below indicate that there has been no price stabilization post fixation of trading margin by CERC.

Table 9.2: Trading margins

Year 2005-06 2006-07 Weighted average purchase price 3.14 4.47 Weighted average sale price 3.23 4.51 Trading margin 0.09 0.04 b. The present short-term bilateral market prices have reached levels of Rs 7 to Rs 8/kWh, which is less than the present maximum Unscheduled Interchange (UI) price. The present maximum UI price is in the range of Rs10/kWh, with many of the state utilities continuing to overdraw from the grid, causing substantial variation in the grid frequency. The short-term prices (UI price/ bilateral contract prices) are sending out price signals of a massive deficit power situation in the country. c. Pursuant to the above trading restrictions, the growth rate in short-term trading volumes has declined, and the market appears to have stagnated. Few players in the trading industry have remained active, and many are reworking their business plans to ensure that viability/profitability levels are maintained. Market making/trading as an activity is not making an impact or contributing to the extent envisaged. d. The role of a trader, in accordance with the Act, is also to promote electricity markets and develop new and innovative products, which may inherently carry higher risks. Capping of trading margin will reduce the risktaking appetite of traders and discourage them from assuming a principal role in trade and push them to a broking role, wherein they would have back-to-back arrangements tied up. e. Traders bring a substantial level of comfort to both buyers and sellers of electricity by way of providing the counter-party credit risk in even long-term power purchase and sale agreements. The trader effectively assumes a principal position and takes high payment risks toward open access charges payable to Central Transmission Utility, full fixed charges during plant life (35/25/15 years) Developing Power Markets | 137

even in the event of failure of power off-take, and prompt payment to the seller even if it receives delayed or no payment from the buyers. Such comfort by the trader to the project developer helps finance new and upcoming projects. In absence of such a counter-party guarantee and such trading restrictions over a prolonged period, investment in new and upcoming projects may get adversely affected, which will become a barrier to the overall development of the electricity market in the country.

Recommendations a. Trader-to-trader transactions: Pursuant to the Electricity Act 2003, the power sector has been unbundled and reorganized. The monolithic SEBs in many states have been split up into distinct generation, transmission and distribution companies. In many of the states, a state-owned trading company has been incorporated which is involved in bulk purchase/sale of power. The state-owned trading entity procures power in bulk for onward sale to the distribution companies (discoms) in the states. In the present situation, wherein the financial position of the discoms is weak, it has become necessary to aggregate/facilitate power procurement at a nodal level, i.e. bulk purchaser/state-owned trading entity, at a common price from generators/inter-state traders and avoid duplication of expertise and effort. Also, in most of the states, financial health varies significantly across discoms, which results in higher power purchase cost for a financially weaker discom as compared to a discom having a better credit rating. Imposition of restrictions on trader-to-trader transactions may also affect the progress of reform of state utilities. Even though by virtue of an intervention by the Hon’ble Supreme Court, technically, restrictions on trader-to-trader transactions for purchase/sale of power do not exist, in view of an earlier order of the CERC, many of the states have been avoiding trader-to-trader transactions. The trader-to-trader transaction should be freely allowed, so long as it is unidirectional, and the discoms/state-owned bulk procurers of power should be permitted to buy power from either generators or traders; however, the purchase consideration should be governed solely by the cost competitiveness of the procured power. Restricting state-owned traders to procuring power only from generators will constrain the development of the electricity markets in the future and minimize options available to the purchaser of power. The Act does not disallow trader-to-trader transactions; rather, it is silent on this issue. Furthermore, if the trader is not broking a deal as an intermediary, but assuming the role of a principal, then he bears the risks associated with the 138 | Indian Infrastructure: Evolving Perspectives

trade and hence enjoys the returns thereof. Traders cannot force utilities to buy power at high rates as the economics of demand and supply dictate the sale and purchase of electricity. To prevent any price escalations arising due to the operation of a trading cartel, trader-to-trader transactions could be allowed, provided the trade conducted is unidirectional in nature. b. Generator-to-trader transactions: The restriction on the generator-to-trader transactions at a price not exceeding 4 per cent over the cost of generation may prove to be counterproductive in terms of developing the electricity market. In a situation of extreme power shortage, when part of the proposed/planned capacity is merchant in nature (i.e. where power sale is not tied through long-term PPAs), with associated risks to be borne by developers, any exercise to cap the price through regulatory intervention could only hamper the process of addition of such capacities coming into the grid. Since the cost of generation is not determined for competitively bid out projects and merchant power plants, the restriction of 4 per cent over base price on the sale of power by generators to traders should not be applied to such plants. With the introduction of the power exchange, the sale/purchase of power will be market-driven, and price restrictions on sale of power may be difficult to implement. It needs to be recognized that one of the reasons for trading volume not increasing substantially is insufficient power available outside long-term PPAs. Such power can be available only when the regulator facilitates merchant capacity and duly recognizes the risks which developers may have to take while developing such projects. Further, this may not send the right investment signals to the states which have already planned for power hubs or are willing to set up such hubs in the near future. Though technically this restriction does not exist currently by virtue of the intervention of the Hon’ble Supreme Court, in view of an earlier judgment of the Appellate Tribunal of Electricity, there has been uncertainty on the final outcome of the case. It may be reiterated that the objective of power trading outside the long-term power purchase agreements (PPAs) is to permit the development of electricity markets and enable competitive forces to determine price. The National Electricity Policy (NEP) recognises that, to promote market development, a part of new generating capacities, say 15 per cent, may be sold outside long- term PPAs, i.e. effectively on market-based commercial principles. In the coming years, a significant portion of the installed capacity of new generating stations could participate in competitive power markets. Regulators may need Developing Power Markets | 139

to be more proactive towards promotion of the power market and take additional steps to bring in captive, renewables, etc. into the fold. This will increase the depth of the power markets, provide alternatives for both generators and licensees/consumers and, in the long run, lead to reduction in tariff. c. Cap on trading margin: In the initial years, the trading companies commenced operations as intermediaries, without assuming any principal risk and as such the operating/financial risk for traders was significantly lower. Past experience indicates that trading licensees could be encouraged to undertake a certain amount of risk, and they will be prepared to do so. Of the over two dozen licensees (those authorized by CERC), there are some who have not only imbibed operational capabilities in trading but are also financially sound. These trading companies may not necessarily assume an intermediary position but may, for part of the trading volume, assume a principal position wherein they would not always have back-to-back arrangements with the distribution companies for any commitment which they make with the generators. In such a situation, the traders would bear significantly higher risks and as such would expect commensurate returns. Therefore, any restriction on the trading margin needs to be removed. Any cap on the margin admissible to trading licensees may only prove counter-productive with regard to the process of trading and its becoming an instrument/ catalyst for capacity development. In the most simplistic model of a trader only functioning as an intermediary, such a stipulation could perhaps be justified, but unless we think of different alternatives and structure appropriate risk–reward models which could permit various types of traders with their meaningful contribution, we may end up defeating the objective of trading being a distinct licensed activity and the resultant outcomes expected from such institutions. In short, the removal of the restriction on the trader-to-trader transaction, and the lifting of the cap on the trader’s margin will lead to the development of trading markets that will ensure market price–based competition amongst the traders. In addition, the traders will be encouraged to introduce new products in the electricity markets and differentiate their offerings to the end-consumer, resulting in benefits to the latter.

4. OPEN ACCESS IN TRANSMISSION In the Electricity Act, “open access” has been defined as “the non-discriminatory provision for the use of transmission lines or distribution system or associated 140 | Indian Infrastructure: Evolving Perspectives

facilities with such lines or system by any licensee or consumer or a person engaged in generation in accordance with the regulations specified by the Appropriate Commission”. It may be seen that the option has been provided to consumers, to intermediaries like trading licensees or distribution licensees or even to generating companies. Each one of them is entitled to the use of transmission systems or distribution infrastructure in a non- discriminatory manner.

Advantages of open access a Enables power to be sold from surplus regions to deficit regions, thereby enabling overall economic growth in both regions. b Encourages merchant power capacities to come up, and thereby leads to competition in the power markets and development of power markets. c Provides freedom to procurers to choose their suppliers, promoting competition in the sector and reduction in the cost of procurement. d Provides flexibility to generators to sell their power to procurers of their choice. The present dispensation on provision of open access on transmission is so structured that it is almost impossible to develop capacities to larger volumes outside long-term contracts. Open-access requests are entertained for periods (a) up to 3 months and (b) 25 years or more. No trader can be in a position to make any long-term arrangement beyond three months unless he takes the risk of beyond 25 years (a highly unrealistic proposition). Unless transactions of various types, short-term, medium-term, long-term, and for intervening durations are permitted, it will be difficult for any meaningful market development outside PPA to materialize. A number of project developers, keen to take investment risks, are facing difficulties at the time of achieving financial closure due to restrictive conditionalities in providing open access to the grid. Unlike other commodities, electricity does need to have a dedicated transmission system. Project developers can take risks in terms of likely consumers and market prices. However, to expect the developers to accept the risk of evacuation infrastructure/open-access availability is an unrealistic proposition in a situation of power shortages.

RECOMMENDATIONS 1. Open access period The present arrangement of short term up to 3 months and long term of 25 years needs to be modified if we have to take into account the practical Developing Power Markets | 141

requirements of industry and consumers. Open access should be available for the following periods: • Short term: up to 3 months • Intermediate term: 3 months to 5 years • Medium term: 5 years to 15 years • Long term: over 15 years The above will ensure contracts (trading and transmission) for intervening durations (3 months to less than 25 years) and increase competition amongst the various players in the trading/generating sphere to the benefit of the end user.

2. Transmission hubs A large number of merchant power plants are coming up in the states of Orissa, Chhattisgarh, Jharkhand, Sikkim and Arunachal Pradesh. The Central Electricity Authority (CEA) and the Power Grid Corporation of India Limited (PGCIL) should take up the lead in developing transmission hubs in these states as the development of large merchant capacities calls for large transmission capacity additions. Such transmission capacities should be made available to the merchant generation capacities on a non-discriminatory basis. Project developers could at most be expected to connect to the transmission hubs (maximum 100 to 150 km). The CEA and the PGCIL will be in the best position to determine the direction of flow of power from the transmission hubs to the load centres, and they shall keep creating the transmission network. The contention that aiming at eliminating congestion with large investments may not be optimal needs to be challenged. This approach may be applicable to a system which is stabilized and augmentation requirement is minimal. In the Indian context, when the growth of the power sector is expected to be 8 per cent to 10 per cent a year, the stakeholders involved need not be concerned about excess build-up on transmission. Any excess would get absorbed within the network in a couple of years, particularly when the demand for power is growing at 8 per cent to 10 per cent per year. In the unlikely event that the PGCIL, for some reason, is constrained to bring in requisite funds due to the absence of beneficiary state utilities, i.e. offtake not being tied up, then it is recommended that the funds be sought from the Plan Allocation on a need basis. 3. Open access charges Pancaking of transmission charges, i.e. adding up of transmission charges of intermediate regions in case of transfer of power from one region to another, 142 | Indian Infrastructure: Evolving Perspectives

leads to increase in cost of power. The same renders the generated power from efficient generators located in other regions/states expensive. It is recommended that a postage stamp pricing mechanism be adopted to both simplify the tariff mechanism and encourage open access.

4. Timely open access permission by state load dispatch centres (SLDCs) There are instances where it is seen that when captive power plants approach the SLDCs for open-access permission, the SLDCs take considerable time in arriving at a decision, leading to uncertainties for the developer. Many states are yet to fix the transmission charges and wheeling charges. This works as a disincentive to merchant power plants and restricts competition and development of the power market.

5. Cross-subsidy charges Cross-subsidy surcharge may need to be reworked in a manner that encourages and facilitates open access rather than constrains this initiative. Some state regulatory commissions have notified cross-subsidy charges which are so high that open access would practically not be possible. Captive Coal Mining by Private Developers | 143

CAPTIVE COAL MINING BY PRIVATE POWER DEVELOPERS: 10 Issues and the Road Ahead October 2009

The rapid growth in the Indian economy has led to a robust growth in the demand for power, which has been constantly outpacing supply. With a view to reducing the demand–supply gap, large additions to generation capacity have been planned. Notwithstanding efforts to have a diversified portfolio of fuel options in power generation, the share of coal-based generation has increased over the years, and with large capacity additions envisaged over the next decade, coal will continue to remain the primary source of fuel for power generation. Besides substantial coal-based capacity additions, the increase in demand for coal has been accentuated by improved utilization of plants, diminishing quality of coal and inadequate availability of gas. With rapid growth in demand for coal, coal supply is proving to be a major cause for concern. The Government of India (GOI) is targeting coal-based generation capacity addition of about 53,000 MW by 2012 (end of the Eleventh Plan period); coal-based installed capacity would then be more than 125 GW by 2012.1 In this scenario, there is an apprehension that coal companies may not be able to cater to the enhanced coal requirement due to resource and other constraints. To augment the coal supply, the Ministry of Coal (MoC), GOI, decided to allocate captive mines to bulk users of coal, in the public and private sectors, and consequently many captive coal blocks have been allocated to power developers.2 In May 2007, at the Parliamentary Consultative Committee meeting of the MoC,3 it was announced that, to meet the coal demand, about 81 coal blocks with geological reserves of about 20 billion tonnes had been identified for allocation to companies, both government and private, for permissible end uses. Of these, 41 coal blocks, with geological reserves of about 15.7 billion tonnes, were earmarked for the power sector. Currently, the allocation of coal mining blocks to companies, other than Coal India Ltd (CIL), is done either under the government company 144 | Indian Infrastructure: Evolving Perspectives dispensation route4 or through the captive dispensation route. These blocks for the power sector have been further categorized in three separate lists on the basis of method of allocation, namely government company dispensation route, screening committee5 route and tariff-based bidding as per the Ministry of Power (MoP) guidelines. The details of these blocks identified for the power sector are as follows:

Table 10.1: Coal blocks identified for the power sector

Method of allocation No. of blocks Total reserves (billion tonnes) Government dispensation route 10 6.1 Tariff-based bidding as per Ministry of Power guidelines 16 6.0 Screening committee route 15 3.6 Total 41 15.7

According to the MoC, till 31 December 2007, 170 captive coal blocks had been allocated, of which 15 blocks allotted to three PSUs and nine private companies had already started producing coal. Of the 170 captive coal blocks allotted (with reserves of 39.3 billion tonnes), 76 coal blocks with reserves of about 23.6 billion tonnes had been allotted to the power sector (24 coal blocks were allotted in 2007).6 It may be noted that production of coal through open cast mining may not need a lengthy lead time unlike in the case of underground mining, which involves a lengthy gestation period, particularly when the stripping ratio is high. Given the foregoing, one may be misled into believing that coal production can start within a short period from the allotted coal blocks. This is not the case, as the allocated mines could also involve underground mining. Discussions with private sector and public sector coal mine allottees have revealed that the lead time is at least four to six years on account of initial planning, conducting geological studies to authenticate quantum and structure of reserves, obtaining several statutory approvals from a multitude of authorities and agencies, formulating mining plans and getting approvals, land acquisition, relief and rehabilitation issues, and infrastructure development. This note attempts to examine the problems faced by allottees of the captive coal blocks and suggests recommendations which could shorten the actual lead time involved in commercial production of coal from these coal blocks. Captive Coal Mining by Private Developers | 145

GUIDELINES FOLLOWED FOR IDENTIFICATION OF COAL BLOCKS FOR CAPTIVE ALLOCATION The guidelines adopted for demarcating the blocks are such that the developers would face a number of problems in quickly bringing the allotted blocks to the production stage. The MoC relies on Coal India Ltd (CIL) and Singareni Collieries Company Ltd (SCCL), for identifying the captive coal blocks. The guidelines adopted by CIL and SCCL for identifying and allotting coal blocks for captive mining are as follows7 : • The blocks offered to the private sector should be at reasonable distance from existing mines and projects of CIL in order to avoid operational problems. • Preferably, blocks in greenfield areas with little or no development of basic infrastructure, like road or rail links, may be allotted to the public/private sector for captive mining. The areas where CIL has already invested in creating such infrastructure for opening new mines should not be handed over to the private sector, except on reimbursement of costs. • Blocks already identified for development by CIL, where adequate funding is on hand or in sight should not be offered to the private sector. • The public/private sector should be asked to bear the full cost of exploration in the blocks offered. • For identifying blocks, the requirement of coal for about 30 years would be considered. • Others, which include mine plan approval under the provisions of the Mines and Mineral (Development and Regulation) Act 1957, approval of the Directorate General of Mine Safety, and inspection by the Coal Controller for appropriate enforcement of conservation measures under the provisions of the Coal Mines (Conservation and Development) Act 1974. Given the fact that the blocks are identified at a distance from the existing infrastructure of the CIL and also that coal blocks are located in remote areas devoid of all basic infrastructure, like roads, rail links and electricity it is difficult for the coal block allottees to quickly bring the coal blocks to production stage. The development of infrastructure on a piecemeal basis, i.e. individually on a block-by-block basis, may cause a drain on the resources of the developer, would not bring in economies of scale, and could be onerous as well. Further, the blocks identified are only regionally explored with inadequate information, which adds to the risk and causes a delay in the development of the blocks. Thus, to facilitate speedy development of coal blocks, it is recommended that the following be considered in identifying and allocating the blocks: • If there are coal blocks in the vicinity of the CIL/SCCL—blocks which are not included in the expansion plans of CIL/SCCL—then these blocks should also 146 | Indian Infrastructure: Evolving Perspectives

be included in the list of captive blocks for allocation and not excluded simply because of their proximity to CIL/SCCL blocks. • In identifying captive coal blocks at a ‘reasonable’ distance from existing mines and projects of CIL/SCCL, it should be ensured that the distance should not put the captive coal block developer at a disadvantage in terms of available infrastructure and other facilities. • Without disturbing the present procedure of coal block allotment, the future allotment should be on the basis of better investigation, for which CMPDIL and other agencies should be mobilized. Further, in the context of inadequacies in infrastructure associated with the captive blocks identified for allocation, it is recommended that the central/state government agencies facilitate development and creation of infrastructure in the mining areas, particularly in providing right of way, railway clearances, water, electricity, etc. The Government could also consider pre-identification of non-coal bearing corridors to be used for rehabilitation colonies and townships. The captive block developers should coordinate with other coal block developers in proximity to jointly fund the development of infrastructure based on a master plan prepared by an independent agency.

ALLOCATION OF COAL BLOCKS Allocation of coal blocks should not only look into the promoter background and the end-use but should also give importance to the technical and financial capability of the applicants for timely development of the blocks. The process of allocation of the blocks could also take into consideration the extent of the preparedness of the developers and the projects. Further, the auctioning approach may be adopted for allocation of coal blocks. However, it may be noted that auctioning of the block to the highest bidder may not be economically viable, since it would get translated into a pass-through in the cost of the end-use product and thereby adversely affect the ultimate consumer. The following approaches, depending on the level of information available for the coal blocks, may be adopted in taking the auctioning route for allocation of blocks: Table 10.2: Criteria for allocation of coal blocks Status of block Possible criteria for allocation Fully explored Lowest cost of power generated Partly explored Maximum estimated production Totally unexplored Production-sharing formula Auctioning on the basis of the maximum proposed production or a production- sharing formula may be workable till sufficient data on the depth, seam thickness, Captive Coal Mining by Private Developers | 147 and quantity and quality of the coal is available for the blocks put up for auction. However, once sufficient data is available for the blocks put up for auction, the criteria for grant of block could be linked to the lowest cost of power generated from the captive coal block.

APPROVALS AND CLEARANCES In the present legislative and regulatory framework, the allottee of a captive coal block has to obtain a multitude of clearances and approvals as stipulated under the provisions of the Coal Mines (Nationalisation) Act, the Colliery Control Rule 2004, the Coal Mines (Conservation and Development) Act 1974 and rules thereunder, the Mines and Minerals (Development and Regulation) Act 1957 (MMDR), the Mineral Concession Rules 1960 (MCR), the Environment Protection Act with its rules and procedures, and the Forest Conservation Act with its rules and procedures. In the federal structure of India, the state government is the owner of the minerals located within the boundaries of the state. Thus, though the central government allocates the coal blocks for captive mining, the state government grants the reconnaissance permit (RP), prospecting licence (PL) and mining lease (ML) under the provisions of the MMDR Act and Mineral Concession Rules. However, the state government can grant the mining lease only with the prior approval of the central government as provided under Section 5(1) of the MMDR Act. The central government approves the application only after the coal block allottee obtains the mining plan approval and clearances from several authorities at the central, state and district levels. Although broadly three clearances are required—grant of RP or ML, environmental clearance and forest clearance, depending on whether the allotted block is explored or unexplored, in forest or non-forest areas, etc.—clearances may be required from multiple authorities. Table 10.3 below indicates the authorities and agencies at the central and state levels from whom the approvals have to be sought by coal block allottees before actual production can begin: Table 10.3: Pre-production approvals for allottees of coal blocks Approvals/Clearances Authority/Agency involved Mining Lease Approval or purchase of CMPDIL (purchase could also be from geological report SCCL, MECL) Directorate General of Civil Aviation and Ministry of Defence (for unexplored blocks if aerial reconnaissance is conceived) Mine plan CMPDIL Coal Controller 148 | Indian Infrastructure: Evolving Perspectives

Table 10.3: Pre-production approvals for allottees of coal blocks (contd...) Approvals/Clearances Authority/agency involved Mine safety Directorate General of Mine Safety Mining technology & Coal Controller (under the provisions of conservation measures; Colliery Control Rules and the Coal Mines coal categorisation (Conservation & Development) Act) Mining Lease State Government (Mining Department), Ministry of Coal (GOI) – Reviewed at various levels within the departments at the state & central government levels Environment EIA/EMP studies State Pollution Control Board; State Environmental Impact Assessment Authority; State Water Resource and Water Supply Department; District Administration (for various aspects of site clearance); Coal Controller; Department of Environment (MoEF). Forest Forest clearance & valuing Committee to advise GOI (MoEF); compensatory afforestation Office of Chief Conservation of Forests, (Regional Office of MoEF); State Forest Department & District Authority; Department of Environment & Forests (MoEF); State Revenue Department; Hon’ble Supreme Court Land Acquisition Ministry of Coal (under provisions of CBA); State Department of Revenue Infrastructure (electricity, Appropriate departments of the state water, railways, roads, etc.) government & ministries of central government It may be noted that of all the clearances, the MoEF clearance is the most time- consuming, since many departments and issues are involved in getting environmental clearances and also the vast majority of the coal blocks are situated on forest land. Even geological investigations (which require Captive Coal Mining by Private Developers | 149 drilling for exploration) in these areas require MoEF approval. This is a lengthy process. The Guidelines for Allocation of Captive Blocks & Conditions of Allotment through the Screening Committee, MoC, provide for the normative time limit ceilings (Appendix, Table 10.5) to ensure that coal production from the allocated captive blocks commences within 36 months (42 months in case the area is in forest land) of the date of issuance of letter of allocation in the case of open cast (OC) mines and within 48 months (54 months in case the area falls under forest land) from the date of the said letter in respect of underground (UG) mines. Figure 10.1 (drawing upon the ceiling time limit provided in the guidelines) is an indicative depiction of the schedule of commencement of mining operations from the time the coal block is allocated. 0 3 6 9 12 15 18 21 24 27 30 32 36 39 42 44 48

GR purchase

Submission & approval of mining plan EIA/EMP studies & approval

Forest clearance

Land acquisition mining lease grant Infrast Coal mining Coal p Start of oining op. Start of prod.

Letter of allotment coal block (zero date) Source: Infraline

Figure 10.1: PERT chart for coal mine development The MoC guidelines mentioned above are to ensure timely development and operation of the allocated captive blocks. In order that timelines are adhered to, the allottee has to provide a bank guarantee, and the encashment of the bank guarantee is dependent on achievement of the milestones consistent with the normative time limit ceilings. Time limits are specified in the Mineral Concession Rules and other legislation for maximum time permissible for grant of approval to an application. This is based on the time to be taken from the receipt of the completed application. Since basic information required for processing the application is available at the district level, particularly for forest clearance, unless and until all such information is available, the application is considered incomplete. The time taken at various levels of scrutiny delays the process. Thus, in view of the large number of approvals required, it is unlikely that the timeline specified by the MoC for development of the coal block 150 | Indian Infrastructure: Evolving Perspectives would be adhered to by the coal block allottees. The existing provisions of the acts do not provide for deemed approval status to applications if the timelines are not adhered to, and thus there is no urgency in disposing of the applications within the specified time. From the above, the following limitations are thus evident in the process of grant of clearances and approvals, which would cause significant delay in production from the allotted captive blocks: • Several parallel clearance and approval processes are to be pursued at the central and state government levels, and the allottee has to follow up the applications with various authorities. • Environment and Forest clearance is extremely time-consuming, and significant delays arise in the public consultation process, valuation of compensatory afforestation, identification of non-forest land for compensatory afforestation, enumeration of trees and completion of cost– benefit analysis by the forest departments. • The entire process of seeking approvals lacks clarity, and often delays in one process cause delays in the others. Besides, conditions of the mining lease are not standardized and could be significantly influenced by individual judgment of the granting authorities. A comparative study on grant of mining leases in Australia, Canada and India, included in the Report of the Expert Group, constituted by the Ministry of Steel for formulating Guidelines for Preferential Grant of Mining Leases (2005), suggests that the time taken for grant of mining lease in Australia is about one to two years (12 + months) and in Canada about two to three years (12 to 36 months) as compared to seven to eight years in India (though the Mineral Concession Rule8 provides that the state government shall dispose of the application for grant of mining lease within 12 months of the date of receipt of application). The observations of the study are listed in Table 10.4. Based on the above observations and discussions, the following suggestions may be considered to speed up the approval process and grant of mining lease: • A single-window approach through nodal agency set up at the state level under the department of mines with representation from all the departments concerned in various ministries. The nodal agency can predetermine the conditions for each category of land based on environmental sensitivity and the nature of the proposed activity (prospecting, mining, etc.). The nodal agency may complete the requirements of identification of land for compensatory afforestation, enumeration of trees, cost–benefit analysis, etc. before inviting application for ML. Captive Coal Mining by Private Developers | 151

Table 10.4: Comparison of mining leases in Australia, Canada and India

Australia Canada India State is fully empowered to State is fully empowered to State to grant mining lease grant mining lease grant mining permit (MP) with the approval of the Central Government

Single-window process, Single-window process, Approvals are required involving four agencies: involving three government from a multitude of • Department of Minerals agencies: authorities at the central, & Energy • Department of Natural state and district levels, with • Department of Resources the different authorities Environment • Department of having little or no • National Native Title Environment coordination amongst Tribunal • Department of Labour them. Broadly three • Land Acquisition different clearances Authority in Local requiring submission of Government separate applications: 1)Approval of grant of ML Minister of Minerals and 2) Forest clearance Energy is the final 3) Environmental approving authority, and clearance takes decisions in consultation with other Central Government grants Ministries. Close interaction forest and environmental between Department of clearances on the Minerals and Energy and recommendations of the agencies responsible for state government. protection of environment.

Mining Lease: Time taken Mining Lease: The time Mining Lease: Although for grant of mining lease is taken for grant of MP is two time frames for clearances one to two years. Time to three years. Fixed time and approvals are taken for activities involved frames are complied with. specified as less than a year are fixed and largely in the various rules, the adhered to, except in cases The project can be rejected actual time taken is usually where public consultation if there are strong chances seven to eight years for process and stakeholder of adverse socio-economic grant of ML. participation is involved. and environmental impacts. Identification of non-forest Applicants submit land for compensatory application for grant of afforestation is a lengthy 152 | Indian Infrastructure: Evolving Perspectives

Table 10.4: Comparison of mining leases in Australia, Canada and India (contd...)

Australia Canada India lease to the Mining process, and sometimes Registrar in the Department takes more than a decade. of Minerals & Energy. Requirement of enumeration of trees; cost- benefit analysis carried out by the State Forest Department is time- consuming (10 to 12 months)

Mining areas are The project is first assessed — categorized as per from an environment angle, environmental sensitivity. before processing for Conditions and procedures mining permit (MP). Once for each category are the project gets predetermined and environmental clearance, identified in each case. the proponent makes an Grant of mining leases in application for MP. very sensitive areas requires approval from both Houses of Parliament.

Applicant is required to Department of Natural Limited emphasis being submit a bond to take care Resources requires a bond given to the technical and of environmental and or security to ensure that financial strength of the rehabilitation reclamation work is applicant. considerations. carried out.

Financial and technical Information not available strengths of the applicant are taken into consideration before grant of mining lease.

Minimum term of mining The minimum term of Mining lease is granted for lease is granted for 21 years. mining permit is 20 years. 20 to 30 years. Captive Coal Mining by Private Developers | 153

Since a single-window approach may require change in legal procedure, alternatively a Public–Private Partnership model can be adopted in the form of a shell company formed for each of the captive blocks allocated, as has been done in the case of UMPPs in the power sector, wherein shell companies (formed by the public sector) are bid out to the private sector only after obtaining clearances and completion of the land acquisition process. • As an immediate remedial measure, MoEF should map and segregate the entire coal-bearing areas into ‘Go’ and ‘No-Go’ areas for each type of lease (reconnaisance, prospecting, and mining) on the basis of forest cover, and environmental and ecological sensitivity. The GOI/MoC, by not allotting the blocks under the ‘No-Go’ areas, would prevent wastage of resources and also speed up the approval process for grant of lease. Thus, for areas defined as ‘Go’, it may be prudent to allow minimal or no forest and environmental clearances for investigative or prospecting purposes, with the conditionality that cutting of trees without prior permission is not allowed. Thus, although the need for detailed environmental impact studies, assessment of compensatory afforestation and enumeration of trees and cost–benefit analysis for forest clearance is required for mining approval, the same level of detail may not be required for limited drilling involved in prospecting or investigative purposes. • Forest clearance is a contentious issue, and the problem has been further compounded by the development which requires, in each case of forest clearance, concurrence by the forest advisory committee or the empowered committee, which has to then send a report to the Hon’ble Supreme Court before sanction is accorded. This procedure needs to be reviewed, and the empowered advisory group or the empowered committee and the MoEF could be delegated authority to accord approval in certain defined categories of forest areas. • MoEF clearances for projects which have a greater probability of commencing operations before the Eleventh Plan should be given priority. • Approval for the prospecting licence by the state government should be issued as the foremost requirement within a minimum time. The developer could then develop the coal block in two or three phases, i.e. identify comparatively easier areas within these blocks (keeping in view government/private land, forest/non-forest land, etc.) and get on with the required investigations, mining plan, approach, etc. for the first phase so that production could start early, and also repeat the cycle of activities in the subsequent phases. This approach will require due consideration and consent by the Ministry of Coal as well as the MoEF. 154 | Indian Infrastructure: Evolving Perspectives

LAND ACQUISITION Unlike in the case of other industries, siting of a coal mine does not leave room for choice in the land to be acquired. Land has to be acquired where coal exists, irrespective of population base or existence/density of forest land. The present procedure regarding land acquisition is lengthy and is prone to litigation. In a great number of cases, a large part of the land belongs to the government, and in such cases the government should take measures to transfer the land to the developers in a shorter time frame.

Rehabilitation of project-affected people As regards rehabilitation of project-affected people (PAP), there are many prevalent resettlement and rehabilitation (R&R) policies. The central government has a National R&R Policy 2007, which provides for the state governments to have their own policies. The R&R policies for the benefit of the PAPs are more stringent on the project developers. Also, providing employment to all PAPs may not always be possible as new technology-driven mining methods are less labour intensive. It is recommended that the MoC coordinate with the state governments to align the state R&R policies with the national R&R policy. Further, the state governments should help developers negotiate the compensation package with the PAPs. For general development and improvement in the coalfield areas, the state government could consider creating an Area Development Fund by applying a levy on each tonne of coal produced. This fund could be utilized for social welfare of PAPs, including their health and education, improvement of infrastructure, roads, water supply, etc.

Geological investigations and mining plan—Alternative agencies In most coal blocks allocated to companies, detailed geological investigations have not been done. At present, exploration for coal in India is carried out by the Geological Survey of India (GSI), Mineral Exploration Corporation Ltd (MECL), Singareni Collieries Company Ltd (SCCL), and directorates of mines and geology of some states. These agencies have limited capacity for drilling of areas (for collating data) and, currently, are already fully stretched. Out of 22,400 km2 of the coal-bearing sedimentary formations identified by the GSI, only about 10,200 km2, or only 45 per cent of the total area, has been systematically explored through regional and promotional drilling. Apart from geological investigations, in the present legislative framework, the mining plan is approved by the MoC (with technical inputs from CMPDIL). This further delays the whole process. Captive Coal Mining by Private Developers | 155

Thus, new agencies with the competence to perform geological investigations need to be set up and accredited by the GOI. These agencies should be independent and unaffiliated to Coal India Limited or other public sector coal companies. These agencies or an independent expert group should also be empowered to review and approve the mining plan, which would be an input to the MoC.

Tapering coal linkage/marketing of surplus coal The time frame for development of the coal block is likely to be much longer than for a power plant in view of the various problems faced by the developers. Therefore, there is a risk that the power plant would be commissioned prior to the commercial operations date of the coal mine. In order to mitigate this risk of delay in the commercial operations of the coal mine, the MoC has decided to provide coal linkage on a tapering basis to the power producers who have been allotted coal blocks for captive use. The tapering linkage is being considered by the MoC to facilitate the working of end-use plants in case development of coal blocks allocated to such consumers does not synchronize with the operation of end-use plants. In this regard, the MoC, in December 2007, came out with a guideline relating to issuance of LoA/allocation of coal on a ‘tapering basis’ to various consumers. However, the application for such tapering linkage would only be considered if the applicant has an approved mining plan for the coal block allocated. This is tricky, and most of the captive block owners are unlikely to qualify for tapering linkage even if their end-use plant is in an advanced stage of development. The Ministry of Coal should consider reviewing this condition for grant of tapering linkage to facilitate rapid capacity addition in power (considering the huge deficit situation in power and sustenance of economic growth), and should consider granting tapering linkage based on the technical and financial capability of the developer and preparedness of the end-use project. Instead of keeping approved mining lease as the criterion for considering the application of tapering linkage, the MoC should provide for the condition that the mining lease be approved before the power plant commences operations.

Joint allotment of coal blocks In some cases, a coal block has been allotted to a number of companies and the allottees have been required to submit bank guarantees to safeguard commercial obligations and ensure timely development of the allotted coal blocks. If some partner companies of the proposed joint venture do not furnish the bank guarantee, then the development of the mine and its associated power plant is held up. 156 | Indian Infrastructure: Evolving Perspectives

It is recommended that if in the consortium of companies granted a coal block, one or some of the partners are not serious, but the remaining partners are serious, then the progress of the development of the block by the consortium should not be jeopardised. Thus, companies in the consortium who furnish the bank guarantee should be allowed to proceed with development of the coal block, and the partners who fail to provide the bank guarantee should be replaced with other companies from among the applicants whose applications are pending with the ministry. However, pending the replacement of partners, the development of the coal block should proceed as usual. In this regard, the company allocated the coal block, having the requisite financial strength, could also be given the freedom to choose partners from among the applicants whose applications are pending with the ministry.

Infrastructure status for coal industry The coal industry needs to be given infrastructure status so as to attract more players into this industry and to incentivise domestic production of mining equipment. The infrastructure status could be for both coal mining and coal washeries. Infrastructure status for the coal sector would bring it on a par with other sectors such as roads, railways and oil, and the laying of oil and gas pipelines. In the past, this proposal was rejected as the bulk of coal mining was under the public sector. Coal mining, however, is expected to undergo a major change in the coming years, with production from captive blocks mainly by private sector companies. The main beneficiaries would be power companies as concessions would make coal production economical and ultimately help in keeping power tariffs low. Coal companies would be also able to import capital equipment and spares at concessional rates.

APPENDICES APPENDIX 1 Table 10.5: Normative time limit ceilings as provided in guidelines for allocation of captive blocks and conditions of allotment through the screening committee, Ministry of Coal Sr. Event Time limit no. in months from ‘0’ date 1 Allocation 0 2 Purchase of GR 1.5 3 Bank guarantee 3 4 Mining lease application 3 5 Mining plan submission 6 Captive Coal Mining by Private Developers | 157

Table 10.5: Normative time limit ceilings (contd...) Sr. Event Time limit no. in months from ‘0’ date 6 Mining plan approval 8 7 Previous approval application 11 8 Previous approval 11 9 Forest clearance application 12 10 Forest clearance 18 11 Environment clearance application 12 12 Environment clearance 18 13 Mining lease grant 24 14 Land acquisition begins 9, 19 15 Land acquisition 30, 36 16 Opening permission application 34, 40 for OC 17 Opening permission grant 35, 41 for OC 18 Production 36, 42 for OC 48, 54 for UG 158 | Indian Infrastructure: Evolving Perspectives

APPENDIX 2 (A) Applicant Initial stakeholder consultation Responsibilities PIRSA Assess risks with key stakeholders (including PIRSA) Responsibilities Prepare draft proposal Other agency No Responsibilities Is proposal suitable for public circulation? Return to Yes applicant Formally lodge proposal with PIRSA

Proposal publicly advertised and circulated to other government agencies

Responses to consultation collated and referred to applicant for response

Applicant prepares response

Is response adequate? No Yes Applicant Yes revises Has proposal changed materially proposal No Assessment report drafted including lease conditions

Assessment report endorsed by PIRSA TRC? Yes No Is referral to DAC or Minister for River Lease grant intention endorsed by Director of Mines. Murray or Minister for Environment and Conservation required? No Yes Yes Yes Minister for Environment and EIC of DAC Minister for River Murray Conservation recommends recommends recommends grant? Yes grant? grant? No Yes No Yes Matter is No Is Native title referred to agreement required? Cabinet or Native title agreement negotiated Yes proposal No modified Native title Agreement registered with PIRSA

Lease offered to applicant No Applicant appeals to Are Lease conditions accepted by applicant? Minister Formally apply for Yes EPA Licence (if required) Minister or delegate grants Lease. MARP prepared Is consultation with other agencies required? Yes No EPA Licence MARP/SEB plan assessed and Comments provided by other agencies granted approved Waivers obtained (if required) MINING MAY COMMENCE Bond paid. Figure 10.2: Flowchart of mining proposal approval process Source: Mining Approvals in South Australia, Regulatory Guideline No.1 of the Division of Minerals and Energy Resources, Government of South Australia, June 2007 Note: PIRSA - Department of Primary Industries and Resources; TRC - Tenement Review Committee - PIRSA independent peer review committee for lease application assessments; DAC - Development Assessment Commission; EIC - Extractive Industries Committee - a subcommittee of DAC; EPA - Environment Protection Authority; MARP - Mining and Rehabilitation Program; SEB - Significant environmental benefit (offset for native vegetation clearance) Captive Coal Mining by Private Developers | 159

APPENDIX 2 (B) Table 10.6: Minimum time frame of process Step Activity Duration Comment 1 Applicant engages and Duration dependent on applicant’s undertakes consultation with resources, complexity of project, and relevant stakeholders and stakeholder sensitivity. acquires relevant studies. 2 Applicant organises risk Duration dependent on applicant’s workshop with key stakeholders, resources, complexity of project, and including PIRSA and other stakeholder sensitivity. relevant government agencies. 3 Applicant prepares proposal, Duration dependent on applicant’s taking into account these resources, complexity of project, and guidelines and identified stakeholder sensitivity. stakeholder concerns. 4 Applicant submits draft 1 day Nominal timeframe proposal 5 PIRSA reviews draft 2 weeks proposal 6 If acceptable - Step 7. Otherwise back to Steps 1 to 3 7 Applicant formally submits Duration dependent on applicant’s lease application and proposal resources and efforts so far 8 PIRSA commences statutory processing of mining lease application. This includes Statutory 14-day time frame to preparation of advertisement for 2 weeks notify landowner and council. relevant newspapers and formal Approximately 2 weeks to prepare notification to landowner and for formal consultation. council that an application has been received. 9 Formal public consultation 2 to 8 The statutory time frame is at least period. (Written submissions weeks 14 days. Dependent on complexity received from agencies and or more of project and PIRSA assessment of public). consultation undertaken by applicant, the time frame may be extended at request of individual stakeholders. 10 Applicant formally asked to Depends on extent of stakeholder respond to issues raised during concern. consultation. PIRSA may 160 | Indian Infrastructure: Evolving Perspectives

Table 10.6: Minimum time frame of process (contd...) Step Activity Duration Comment summarise issues raised in public 2 weeks consultation and recommend or more actions to be taken by applicant to address issues raised. 11 Applicant produces response Duration at applicant’s discretion. document. Document will be made public. 12 If acceptable, Step 13. Otherwise (major issues are raised, or major material changes to the project are proposed) back to Step 10, or Steps 1-3. 13 PIRSA assesses all documentation This is a major task for PIRSA. This (including proposal, time frame cannot be reduced submissions and response) 4 weeks without compromising quality of and produces formal Assessment assessment. Assessment report is Report, including detailed made public. lease conditions. 14 Consideration of Assessment PIRSA will manage as far as possible Report and Draft Lease 2 weeks to keep to a minimum. Conditions by Tenement Review Committee and Director of Mines. 15 Consideration of lease Not under PIRSA’s control. Subject conditions by Minister for to EIC/DAC meeting arrangements, Planning (Schedule 20 area of and Minister for Environment and Development Act); Minister for Conservation or River Murray Environment and Conservation time frames. If the recommendation (regional reserve or jointly of PIRSA is rejected by DAC or proclaimed park under National Minister for Environment and Parks and Wildlife Act); and/or Conservation or Minister for River Minister for the River Murray Murray, the matter may be resolved (River Murray Protected Area). in Cabinet. 16 Applicant negotiates native title access and/or aboriginal Duration at applicant’s discretion heritage access and registers with PIRSA (if subject to native title). 17 Formal offer of mining lease 1 week with detailed conditions. 18 Applicant accepts offer of 21 days is a statutory time frame to mining lease terms and 3 weeks which a response must be made by conditions. the applicant. The time frame may be Captive Coal Mining by Private Developers | 161

Table 10.6: Minimum time frame of process (contd...) Step Activity Duration Comment extended at the discretion of PIRSA. If applicant does not accept terms and conditions offered, the applicant may appeal directly to the Minister. 19 Formal grant of mining lease. 1 week 20 Company formally applies for Duration at applicant’s discretion EPA licence and works approval (if required). 21 Company finalises MARP Duration at applicant’s discretion and SEB plan. 22 Consultation on MARP with Not under PIRSA’s control. Subject to other government agencies, if other agency timeframes. required by lease conditions or for other reasons. 23 PIRSA assesses draft MARP 2 weeks and sets bond. 24 PIRSA formally approves MARP 1 week and SEB plan (if required). 25 EPA works approval/ Not under PIRSA control. licence issued. 26 Leaseholder pays bond to PIRSA Leaseholder obtains waivers and Duration at leaseholder’s discretion. provides copy to PIRSA (if applicable). 27 Leaseholder may commence mining TOTAL minimum timeframe approximately 6 months from formal submission of an acceptable mining lease proposal, dependent on complexity of project, quality of stakeholder engagement undertaken by applicant, PIRSA and other government agency workloads. If any part of this process covers the Christmas – New Year period, a further month can be added to the approximate time frame. Source: Mining Approvals in South Australia, Regulatory Guideline No.1 of the Division of Minerals and Energy Resources, Government of South Australia, June 2007 162 | Indian Infrastructure: Evolving Perspectives

NOTES 1. Power Scenario at a Glance for All India, CEA, GOI, July 2008; White Paper on Strategy for 11th Plan, CEA & CII, August 2007. 2. Under the provisions of the 1993 amendment to Coal Mines (Nationalisation) Act 1973. 3. 81 coal blocks identified for allocation under captive END-USE, Press Release, PIB, GOI, 18th May 2007. 4. Under the government dispensation route, the block is allocated to a government company and this company has the right to be used for a specific end-use such as power, steel and cement. 5. A screening committee has been set up in the MoC for screening the proposals received for captive mining of coal and lignite. The screening committee comprises members representing the Ministry of Coal, the Ministry of Power, the Ministry of Railways, the Ministry of Steel, state governments concerned, CIL, CMPDIL, and the Department of Industrial Policy & Promotion (Ministry of Industry). 6. Coal Directory of India 2006–07, Part-I: Coal Statistics, Ministry of Coal, GOI 7. Coal Directory of India 2006–07, Part-I: Coal Statistics, Ministry of Coal, GOI 8. Mineral Concession (Amendment) Rules 2002. 9. The mining proposal approval process and minimum timeframes of process as provided in the Regulatory Guideline No.1 of the Division of Minerals and Energy Resources, Government of South Australia, June 2007, are presented in Appendix 2. The regulations laid down by the Department of Industry & Resources, Government of Western Australia, are similar. Distribution Reforms in Andhra Pradesh | 163

POWER DISTRIBUTION REFORMS IN ANDHRA 11 PRADESH October 2009

INTRODUCTION The Andhra Pradesh State Electricity Board (APSEB) was formed on 1 April 1959. Until its unbundling in February 1999, APSEB was responsible for electricity generation, transmission, distribution and supply. It functioned under the overall guidance of the state government, interacting with the central power agencies for planning and coordination. The APSEB enjoyed a good reputation amongst the other utilities in India. The plant load factor (PLF) of APSEB-owned generation stations was 83.2 per cent in 2000, much higher than the national average of 67 per cent. The Vijayawada Thermal Power Station (VTPS) received the productivity award in 2000 (PLF of 86.9 per cent) and the Rayalaseema Thermal Power Plant (RTPP) won the incentive award. Other aspects of good performance include rapid erection of power stations, and low employee/consumer ratio.1 The APSEB was the third-largest SEB in terms of units of power sold, next only to Maharashtra and Gujarat.2

REFORMS IMPERATIVE IN THE STATE Though APSEB’s performance on the generation side was far better compared to other state electricity boards, its performance on distribution and financial aspects was poor.1 By the late nineties, the state was facing both energy and peak shortages and the quality of power supply had deteriorated; the power utility’s financial losses had grown to Rs 39 billion and new investments were not financeable. The power subsidies had increased to 1.6 per cent of the Gross State Domestic Product (GSDP), while on the other hand the combined public expenditure on health and education had declined from 4.7 per cent of GSDP in FY1987 to 164 | Indian Infrastructure: Evolving Perspectives

3.6 per cent of GSDP in FY1998.3 The gap between average cost of supply (ACS) and average revenue realized (ARR) grew from 4.2 paise/kWh in 1990–91 to 138.8 paise/kWh in 1999–2000.2 Power sector reforms became imminent as in their absence the need for subsidies from the government’s budget would have continued to grow and crowd out social sector investments.

350

300

250

200 138.8

150 122.3

Paise/kW h

100 73.1 59.0 57.8 36.0 50 4.2 5.9 5.8 10.4 0 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–971997–98 1998–99 1999–2000 Cost of power supply Average tariff Gap Figure 11.1: Cost of power supply, average tariff and gap Source: Annual Report (2001–02) on the Working of State Electricity Boards & Electricity Departments, Planning Commission (Power & Energy Division), Government of India, May 2002

Box 11.1: APSEB’s performance review

• Power deficit: The power shortage faced by the state kept on increasing despite significant growth witnessed in generation in Andhra Pradesh. The total deficit increased from 6.7 per cent in 1991–92 to 8.5 per cent in 2001–02. During the same period, peak power deficit saw an increase from 15.8 per cent to 19.9 per cent.2 • Transmission and distribution losses (T&D losses): Inadequate infrastructure, low investments in new infrastructure and improper O&M of the network led to an increase in T&D losses reported by APSEB. T&D losses as a percentage of availability increased from 19.2 per cent in 1992–93 to 35.2 per cent in 1999–2000.2 • Gap in cost and revenue realized: The gap between cost of power supply and the average tariff realized from the customers denotes the margin for a power distribution business. This gap grew from 4.2 paise/kWh in 1990–91 to 138.8 paise/kWh in 1999–2000.2 Distribution Reforms in Andhra Pradesh | 165

• Increasing losses and subsidy: Increasing T&D losses combined with the large gap in cost of supply and revenue realized led to deterioration in the financial health of APSEB. Commercial losses without subsidy increased from Rs 4 crore in 1992–93 to Rs 3117 crore in 1999–2000. Subsidy received from State Government increased during the same period from zero to Rs 3064 crore.2 T&D losses as % of availability 40

35

30

25

20

15

10

5

0 1992–93 1993–94 1994–95 1995–96 1996–97 1997–981998–99 1999–2000 Source: Annual Report (2001–02) on the Working of State Electricity Boards & Electricity Departments, Planning Commission (Power & Energy Division) Government of India May 2002 Commercial loss (without subsidy) 3500 3117 3000 2679 2500

2000 1376 1500 1255 981 939 1000

500 4 23 0 1992–93 1993–94 1994–95 1995–96 1996–97 1997–981998–99 1999–2000 Source: Annual Report (2001–02) on the Working of SEBs & Electricity Dept., Planning Commission (Power & Energy Division), Government of India May 2002 The deteriorating situation on the power front in Andhra Pradesh had a number of causes. Some of the main reasons are: • Change in hydro-thermal energy mix: In Andhra Pradesh, historically, installed capacity of hydel power used to be greater than that of thermal power. 166 | Indian Infrastructure: Evolving Perspectives –

In 1960–61, hydel power accounted for 58.2 per cent and thermal power for 41.8 per cent of the installed capacity. Over time, this mix has changed in favour of thermal power. In 1990–91 the proportion of installed capacity constituted 50.1 per cent hydel power and 48 per cent thermal power, which further changed to 36.5 per cent for hydel power and 42.9 per cent for thermal power by 1997–98. The remaining power capacity in 1997–98 was in gas projects.1 As the proportion of cheaper hydel power declined over time and the proportion of costly thermal power increased, so did the average unit cost of power. With the rising average cost of power supply, the gap in ACS and ARR widened, leading to deterioration in the financial health of APSEB. • Change in load mix: There was a huge disparity between agricultural and industrial tariffs over the years, putting pressure on the industrial sector and leading to stagnation in industrial consumption. Slowly, industry moved towards cheaper captive generation. Sale of power to industry declined from 35 per cent in 1993–94 to 24 per cent in 1999–2000, while sale to agriculture remained unchanged at around 40 per cent during the same period. The average tariff charged from industry and agriculture during 1999–2000 was 394.9 paise/kWh and 15.35 paise/kWh respectively2.

REFORMS UNDERTAKEN In the background of the deteriorating situation on the power front and the new initiatives by the Government of India to attract private investment, the state government contemplated restructuring the power sector. Reforms were brought about in multiple steps.

Hiten Bhaya Committee The Government of Andhra Pradesh constituted a high-level committee under the chairmanship of Hiten Bhaya, a former chairman of the Central Electricity Authority, to suggest reforms in the power sector. This committee was constituted in January 1995, and it submitted its report in June. The important proposals made by the Hiten Bhaya committee were: • A tariff structure which covers production costs • Restructuring APSEB on a functional basis to promote efficiency and functional specialization by unbundling APSEB. Constituting separate companies for each function (namely generation, transmission and distribution) and putting them in the hands of different companies • Keeping the companies thus formed as subsidiaries of APSEB Distribution Reforms in Andhra Pradesh | 167

• Running the companies on commercial lines • Retaining the board as a holding company in charge of long-term sector planning, supervision and co ordination of the subsidiaries • Government to retain role of monitoring reform implementation and advising on policy • Setting up a regulatory commission to fix tariff structure and keep licensing powers with the state government • The committee did not recommend outright privatization of public utilities and cautioned that substituting private monopoly for public monopoly would only make the situation worse. The committee felt that the privatization initiative should start initially with management contracts in the distribution business.

The World Bank and the AP power sector restructuring programme After Chandrababu Naidu became chief minister in September 1995, the Government of Andhra Pradesh (GoAP) approached the World Bank for a structural adjustment loan to tide over the fiscal crisis that it was facing. In response, the World Bank brought out a comprehensive report, AP—Agenda for Economic Reforms, in January 1997, outlining its approach to reforms, including the power sector. The bank suggested comprehensive reforms in the power sector going beyond the recommendations of the Hiten Bhaya Committee. Some important components of the reforms proposed by the World Bank were: • Defining a structure for the sector consistent with privatization of distribution and private sector development in generation • Corporatizing power utilities and ensuring that they operate without governmental interference • Creating an independent and transparent regulatory system for the sector with a broad range of responsibilities, including granting of licences and enforcing them • Enacting comprehensive reform legislation to establish the new regulatory framework and implement restructuring measures • Increasing the tariff rate for agriculture to at least 50 paise/kWh in the near term and continuing to adjust tariffs to cover costs and reduce cross-subsidies. 168 | Indian Infrastructure: Evolving Perspectives

Table 11.1: Steps taken for power sector reforms in Andhra Pradesh 1995 June Hiten Bhaya Committee Report 1996 September World Bank’s Agenda for Economic Reforms in Andhra Pradesh 1997 March AP State Government’s Policy Statement on Power Sector Reforms 1998 April Passing of AP Electricity Reforms Bill in the State Legislative Assembly 1999 January World Bank’s PAD on AP Power Sector Reforms Programme (APPSRP) 1999 February AP Electricity Reforms Act 1998 comes into force 1999 February APSEB unbundled into APGENCO and APTRANSCO 1999 April AP Electricity Regulatory Commission starts functioning 2000 March APTRANSCO further unbundled into APTRANSCO and four discoms 2002 April Financial autonomy to discoms 2002 August Employee division (option process) among APGENCO, APTRANSCO and discoms on permanent basis 2003 June Enactment of Electricity Act, 2003 2003 August Suspension of the World Bank loan after the first stage itself citing high interest rate and unacceptable conditions 2004 May Change in government and the announcement of free power to the agricultural sector. The bank's approach was driven by the idea of changing the ownership from public to private in a span of eight to ten years. The AP Power Sector Restructuring Programme (APPSRP) was to be implemented over a ten-year period, starting February 1999. The Adaptable Programme Loan (APL) scheme was planned in five stages, APL-1 to APL-5. The total loan amount was US$4460 million, with the World Bank contributing 22 per cent of the amount. The other international lending agencies included Department for International Development (DFID) and Overseas Economic Cooperation Fund (OECF). The Indian agencies included the GoAP, the Power Finance Corporation and the Rural Electrification Corporation. This loan had several preconditions which were to be satisfied so that the utility became eligible for the next stage of the loan. These conditions included privatization of distribution and generation, average annual tariff hikes, implementing cost-based tariff and reducing government subsidy to zero.

Reforms undertaken by the AP Government Within six months of the World Bank recommendations, on 14 June 1997, the GoAP released a power sector policy statement indicating proposed policy and Distribution Reforms in Andhra Pradesh | 169 structural changes in the power sector. In order to give concrete shape to this policy, the GoAP enacted the Electricity Reforms Act of 1998. The Reform Bill was introduced in the Legislative Assembly on 27 April 1998 and was passed on April 28. It was notified on 29 October 1998 and came into effect in February 1999.

APSEB (Andhra Pradesh State Electricity Board) Structure prior to reforms

Structure post-reforms

APGENCO Generation Company

APTRANSCO Transmission Company

APCPDCL APEPDCL APNPDCL APSPDCL (AP Central Power (AP Eastern Power (AP Northern Power (AP Southern Power Distribution Company Ltd) Distribution Company Ltd) Distribution Company Ltd) Distribution Company Ltd)

Figure 11.2: Power sector: Structure pre- and post-reforms The APSEB was unbundled into APGENCO and APTRANSCO in February 1999. The Electricity Reforms Act provided for the constitution of the Andhra Pradesh Electricity Regulatory Commission (APERC). In April 2000, APTRANSCO was further unbundled into a transmission company and four distribution companies (discoms) managing distribution in the four zones of the State—Central, Eastern, Northern and Southern. The state government signed an MOU with the Ministry of Power, Government of India, on reform and restructuring which had the road map for reform, plans for tariff rationalization, metering and maintaining grid discipline. As part of the distribution sector reforms, in April 2001 the four discoms were issued independent licences for distribution. Andhra Pradesh took to power sector reforms much earlier than most other states in the country. However, the pace of reforms slowed down by year 2004. Signs of the slowdown were visible in suspension of the World Bank loan after Stage I itself, and no attempt was made to privatize distribution. The reasons identified for this slowdown were opposition to the reform agenda, failure of the World Bank-led reform process in Orissa and the national-level rethinking on World Bank-led reforms. In May 2004, the Congress came to power in AP, replacing Chandrababu Naidu’s TDP. It announced free power to agriculture and promised to review the reforms, including power purchase agreements (PPAs) with private generators. 170 | Indian Infrastructure: Evolving Perspectives

Initiatives taken during the reform process Power distribution sector reforms in Andhra Pradesh were concentrated on reducing losses and improving commercial viability through improved infrastructure, better auditing, and use of information technology. The initiatives taken during the reform process were: • Theft control: The GoAP enacted an anti-theft legislation in July 2000, laying down stringent penalties for theft of electricity, including mandatory imprisonment of offenders. The legislation enabled constitution of special tribunals and courts for speedy trial and recognized collusion of the utility staff as a punishable offence. The enforcement efforts made by both government and distribution companies have shown positive results in controlling theft. According to a report by the World Bank, about 5 million out of 12 million metered services have been inspected. About 150,000 cases of theft were registered during FY2000–03 compared to 9200 cases during FY1998–2000. Also, 4100 consumers and about 50 employees were arrested.3 The success of the theft-control initiatives was built on proactive measures taken, including: • Communicating to the stakeholders the objective, intent and the enforcement of the new act • A regularization drive was launched to provide a one-time opportunity to the unauthorized consumers to register legally. About 2 million3 residential consumers were regularized. • Implementing institutional, management and administrative changes in the power distribution companies to ensure effective enforcement of the act • Legal support system geared up to provide implementation support • The vigilance department has been strengthened with appointment of the Inspector General of Police as the joint managing director in the company. • The organizational structure was modified to strengthen coordination between various departments like operations/technical department, commercial departments and vigilance department. • Special police stations were set up to deal with electricity theft cases. Initially, statewide inspections and revenue collection drives were launched targeting large industrial and commercial consumers, and were gradually extended to the rural areas. This was supported by a comprehensive programme of consumer metering and energy audit. Discoms have developed specialized IT-based tools for Distribution Reforms in Andhra Pradesh | 171 institutionalization of theft control and monitoring measures. These efforts at theft control led to improvements in billing and collection by the utilities. • Energy audit and metering: To measure power transacted at various levels and determine reliable extent of transmission and distribution losses in the system, a new metering drive was launched. High-quality meters were installed on the interface points between the power sector companies. The 11-kV feeders were metered and data loggers were installed to monitor the supply of electricity to agricultural consumers. For realistic estimation of agriculture consumption, about 30,000 distribution transformers (out of 190,000 DTRs) supplying power to predominantly agriculture consumers were metered3. A programme to improve the consumer metering system was also started by distribution companies. For high-value customers, the existing meters were replaced with high-accuracy electronic meters. • Consumer Analysis Tool (CAT): CAT is a customer database used to analyze customer information in order to identify trends in metering, billing, and collections. A key feature of CAT is to risk-profile customers to enable the management to design strategies for efficiency improvements targeted at specific customer groups. The customers are grouped into different categories based on their payment history—a matrix of proportion of bill paid and number of defaults in a twelve-month payment track record. CAT addresses the needs of a number of departments, including operations, vigilance, and regulatory affairs departments. It acts as an effective monitoring tool and helps exercise control. The results are reflected in improved billing and collection efficiency. • Monitoring and Tracking System (MATS): MATS has been designed as a tool to assist in monitoring and tracking various cases of irregularities, like theft, malpractice and back-billing. The objectives of the system are to: • Streamline the regularization system by reducing the effort and time required to inspect and follow up on irregularities • Enable process automation to reduce high documentation requirements and loss of records • Enable a performance-based monitoring system to track and take action on irregularities. MATS is based on a workflow process where each completed document is automatically sent to the appropriate reviewing officer based on the defined process. Training is given to all officers on the usage of the system to enable them to fulfil their role. 172 | Indian Infrastructure: Evolving Perspectives

• The system also enables performance review of the employees by tracking the period of time each case is kept pending at each level. • The system is highly role-based with each officer being able to track as well as take action only on cases that lie within his/her jurisdiction. • Investment in infrastructure: The distribution infrastructure was modernized to bring down technical losses and improve the performance of the system. Under the first power sector restructuring loan under the adaptable loan programme of the World Bank (APL1), funds were provided for financing high-priority investments in the T&D system. Till 2006, a total of Rs 6652 crore was spent on improving the distribution system.8 These funds were utilized for installing new meters, replacing or repairing old meters, installing new transformers and upgrading distribution lines. The yearly capital expenditure of various discoms is shown in Table 11.2: Table 11.2: Investment in infrastructure (Rs crore) 2001 2002 2003 2004 2005 2006 Total APCPDCL 404 397 445 426 456 483 2611 APEPDCL 186 199 151 198 222 167 1124 APNPDCL 327 207 195 194 287 329 1537 APSPDCL 261 171 195 194 283 277 1381 TOTAL 1178 974 985 1012 1249 1255 6652 Source: Power Sector Reforms in Andhra Pradesh: Their Impact and Policy Gaps: B. Saranga Pani, N. Sreekumar and M. Thimma Reddy • Transformer Information Management System (TIMS): TIMS enables effective tracking of the distribution transformers as well as the associated structures throughout their life cycle and analyzes the information to: • Improve asset tracking, utilization and maintenance • Improve customer service through deployment of transformers and exception handling (in case of failures) • Enable greater visibility leading to improved decision making. Discoms benefit financially by reducing inventory carrying costs. Operationally, they help reduce transformer downtime and evaluate vendor performance, and enable these through various reports which indicate data like: • Availability of stocks of DTRs in stores • Locations with maximum transformer failures Distribution Reforms in Andhra Pradesh | 173

• Identification of transformers and structures which show frequent failure • Time required to repair transformers at each repairing location • Book Consolidation Module (BCM): BCM is a tool aimed at reducing the time taken and the manual intervention involved in the consolidation of accounts from the circle level. The tool generates balance sheets and profit/ loss for the discoms after consolidation of all accounting units. In addition to consolidation, the tool also provides for various variance reports: • Budgeted versus actual comparisons: These reports compare monthly variance between budgeted and actual income or expenses from various account codes. • Comparative statements: These reports compare income/expense for current month with the previous month or the previous year to calculate appropriate variances.

Outcome of reforms and initiatives The reform process in Andhra Pradesh was started under the guidance of the World Bank. However, the loan was suspended after the first tranche, and the regime also changed in AP. The new regime changed the course of the reform process, and this was reflected in a number of parameters used to assess the distribution sector. Some of these are discussed below: • Aggregate technical and commercial losses (AT&C): AT&C losses in AP have decreased from 27 per cent in 2002–03 to 16 per cent in 2007–08. Such a decrease is attributable to the focus of reforms on reducing losses through better auditing and investment in infrastructure. However, AT&C losses have not declined much since 2005–06. 30 27.0

25 21.2 20 17.9 16.5 16.7 16.2 15

10

5

0 2002–032003–04 2004–05 2005–06 2006–07 2007–08 Figure 11.3: AT&C losses (%) Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited 174 | Indian Infrastructure: Evolving Perspectives

The performance of the discoms is shown in Table 11.3. While eastern and northern region discoms have reported low losses significantly, AT&C losses reported by central and southern region discoms are still high. Also, post- 2004–05, loss reduction has not been significant.

Table 11.3: AT&C losses (%) of distribution companies 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 APCPDCL 30.19 18.99 23.95 18.99 18.32 19.23 APEPDCL 17.61 16.57 14.27 12.19 12.09 7.46 APNPDCL 27.09 9.79 21.91 11.82 23.28 11.92 APSPDCL 27.45 17.06 20.55 19.23 17.47 20.02 Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited • Collection efficiency: The collection efficiency in Andhra Pradesh has been above 90 per cent post-reforms. High collection efficiency has been possible due to the energy auditing and metering drive and implementation of tools like MATS and CAT. Collection efficiency of individual discoms is shown in Table 11.4. While collection efficiency of all discoms has been high, there has been a lot of variation in collection efficiency reported by APNPDCL. 120 102.9 100.2 96.5 97.2 98.1 100 92.4

80

60

40

20

0 2002-032003-04 2004-05 2005-06 2006-07 2007-08

Figure 11.4: Collection efficiency (%) Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited Distribution Reforms in Andhra Pradesh | 175

Table 11.4: Collection efficiency (%) of distribution companies 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 APCPDCL 90.52 102.39 94.81 99.04 98.51 97.46 APEPDCL 96.95 96.63 99.54 99.18 99.22 99.88 APNPDCL 92.57 113.10 96.65 104.27 89.41 96.90 APSPDCL 92.11 102.83 97.03 100.58 98.53 99.23 Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited • Subsidy by state: Subsidy provided by the state government to the power distribution sector has come down significantly from the pre-reforms period. In 1999–2000, subsidy by the state was over Rs 3000 crore while the subsidy received in 2004–5 was Rs 1303 crore. However, post-2004–05, subsidy has started increasing. Subsidy reported for the year 2007–08 was Rs 2408 crore. 3000

2500 2408

2000 1842 1509 1515 1483 1500 1303

1000

500

0 2002–032003–04 2004–05 2005–06 2006–07 2007–08 Figure 11.5: Subsidy received (Rs crore) Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited Subsidy received by individual discoms is given below: Table 11.5: Subsidy received (Rs crore) 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 APCPDCL 506 579 464 261 499 1108 APEPDCL 212 227 194 31 8 0 APNPDCL 353 307 311 639 839 733 APSPDCL 438 402 334 552 496 567 Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited 176 | Indian Infrastructure: Evolving Perspectives

• Financial viability of AP discoms: 1. Profitability: Discoms in AP have been registering losses without subsidy. While these losses were around Rs 1200 crore in 2004–05, they started increasing in later years. Aggregate losses registered without subsidy for the year 2007–08 were Rs 2526 crore. With the help of subsidy, discoms have been able to cover the losses made. However, in 2007–08, aggregate losses with subsidy were Rs 118 crore.

Table 11.6: Profit with and without subsidy 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 With subsidy 16 -135 26 308 145 -118 Without subsidy -1232 -1579 -1194 -1241 -1697 -2526 Source: Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited 2. Gap in cost and revenue realized: Post-reforms, the gap in ACS and ARR has decreased significantly for AP post-reforms. The gap reported for the year 1999–2000 was 138 paise/kWh as compared to 48 paise/kWh in 2007–08. However, this gap has started increasing in the last few years.

CONCLUSION Andhra Pradesh is among the states that initiated reforms in the power sector. The first steps were taken way back in 1995 with the formation of the Hiten Bhaya Committee. Actual unbundling took place in the year 2000. It has been over eight years since the power sector has been unbundled in AP. Over this period, AP has seen a change of regime which has also brought a change in the way reforms have been pursued in the state. At the time of unbundling of the APSEB, the reforms were driven by World Bank guidelines, and minimization of cross-subsidization and privatization of discoms were considered eventual outcome of reforms. However, the World Bank's failure in Orissa led to AP withdrawing from the programme and to change in the course of reforms. All the discoms are still under government ownership. Also, the new government announced free power for agriculture. At the time of introduction of free power to agriculture, the four discoms together were receiving nearly Rs 400 crore1 as revenue from agricultural connections. It was claimed that the same amount would be saved by renegotiating the PPAs with the IPPs in the state. However, Distribution Reforms in Andhra Pradesh | 177 increasing subsidy and the gap between ACS and ARR after introduction of free power shows that it had a negative impact on the financial health of the sector. 3.5

3.0

2.5

2.0

1.5

1.0

0.37 0.38 0.35 0.48 0.5 0.29 0.30

0.0 2002–032003–04 2004–05 2005–06 2006–07 2007–08 ACS ARR Gap Figure 11.6: ACS, ARR (without subsidy) and gap over the years Source: Report on the performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited The impact of distribution reforms had been positive and can be seen in decreased losses, improved collection efficiencies, and smaller gap between ARR and ACS compared to what existed in the state prior to reforms. Also, the deficit position of the state has improved over the years. Peak deficit for the state declined from 19 per cent in 2002–03 to 7.6 per cent in 2008–09. Overall energy deficit declined considerably till 2003–04 but rose again and stood at 6.8 per cent in 2008–09.7 Table 11.7: Peak deficit (%) and energy deficit (%) in AP 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Peak deficit 19.2 10.5 2.3 5.1 15.4 8.8 7.6 Energy deficit 6.8 2.9 0.7 1.3 4.4 4.1 6.8 Source: Power Sector at a Glance, April 2009, Central Electricity Authority Performance, both in financial and operational terms, has varied across the discoms. While APEPDCL has been able to considerably reduce its AT&C losses and subsidy received, APCPDCL and APSPDCL have not been able to maintain a similar performance. 178 | Indian Infrastructure: Evolving Perspectives

Table 11.8: Subsidy received by distribution companies APCPDCL APEPDCL APNPDCL APSPDCL 2002–03 506 212 353 438 2003–04 579 227 327 402 2004–05 464 194 377 334 2005–06 261 31 639 552 2006–07 499 8 839 612 2007–08 1108 0 733 567 Source: Report on the performance of the State Power Utilities for the Years 2002–03 to 2004–05 and 2005–06 to 2007–08, Power Finance Corporation Limited

The good performance of APEPDCL can be attributed to the customer mix, given the high proportion of industrial mix it inherited from APSEB (see Annexure 1). However, APNPDCL has performed even better despite having a high proportion of agricultural customers.

In the long run, there is a question mark over the commercial viability of various discoms without subsidy, especially in view of the rise in subsidy requirements in the last three financial years of this study. Also Quality of Supply (QoS) targets need to be set and measured in order to usher in a power distribution sector which is both commercially viable and consumer friendly.

ANNEXURE 1 Sales mix, revenue mix and cost components • Sales mix: This was inherited by discoms from APSEB depending on their region. While APEPDCL inherited a favourable mix due to very low agricultural component, APNPDCL sells approximately 50 per cent of its power to the agricultural sector. Over the years, the sales mix for various discoms has remained the same. • Revenue mix: The revenue mix of various discoms given below shows that maximum contribution to revenue is made by the industrial sector, irrespective of the proportion of sales accounted for by them. Also, it can be seen that the agriculture sector contributes very little to the revenue of discoms. Contribution to revenue by the commercial sector is also higher than the proportion of sales accounted by them. This revenue mix clearly shows the cross-subsidization being done by the industrial and commercial sectors. Distribution Reforms in Andhra Pradesh | 179 Table 11.9: Sales mix (%) of distribution companies Table 11.10: Revenue mix (%) of distribution companies Table 11.11: Expenses as % of total cost for distribution companies 2005–062005–06 2006–072005–06 2006–07 2007–08 2006–07 2007–08 2007–08 4.6%1.2% 5.7%4.9% 0.6% 6.2% 4.1% 1.5% 5.2%0.8% 4.5% 1.2% 5.4% 0.7% 4.5% 1.4% 7.2% 0.7% 3.9% 0.5% 8.0% 4.6% 4.5% 1.3% -2.3% 5.4% 2.7% 1.0% -3.9% 5.3% 4.6% 1.4% 1.2% 7.8% 1.7% 0.5% 2.0% 6.2% 4.4% 1.3% 9.3% 7.1% 2.9% 1.2% -2.1% 4.7% 3.4% 1.9% 83.8% 81.7% 80.6% 77.5% 86.8% 84.1% 80.6% 79.8% 78.4% 83.7% 80.7% 79.3% APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL APCPDCL APEPDCL APNPDCL APSPDCL Report on the Performance of State Power Utilities for Years 2005–06 to 2007–08, Finance Corporation Limited Report on the Performance of State Power Utilities for Years 2005–06 to 2007–08, Finance Corporation Limited Report on the Performance of State Power Utilities for Years 2005–06 to 2007–08, Finance Corporation Limited DomesticCommercialAgriculturalIndustrial 18.8 9.2Others 36.1 26.0 32.7 5.0 19.0 19.4 3.3 41.7 3.3 48.7Domestic 27.0Commercial 16.0 8.5 34.8Agricultural 5.8Industrial 17.8 19.4 24.0 21.3 12.7Others 37.1 10.0 3.6 26.2 19.6 10.1 32.7 8.4 52.5 18.1 5.2 0.8 18.5 23.6 10.6 42.0 3.3 46.9 50.7 2.4Power purchase 3.2 25.7 24.4 2.0 12.8Employee cost 22.5 15.3 38.7 35.8O&M cost 8.5 18.7Interest cost 5.6 0.8 19.9 25.1 23.4 25.0 38.7Depreciation 30.0 12.3Admin & General 28.4 20.7 6.2 23.3 10.6 42.0expenses 1.4 52.7 16.1Other expenses 7.9 17.4 26.9 11.8 37.3 3.4% 5.4 49.1 2.7 0.8 49.2 3.1 27.5 24.1 5.8% 13.7 15.0 38.6 18.8 3.3 1.3% 30.1 1.8 4.9% 13.0 19.2 23.5 28.3 42.1 1.6% 20.9 6.1 0.5 5.3% 15.1 19.9 10.4 1.5% 51.3 19.6 3.5% 1.3 1.6% 26.7 11.2 8.0 46.0 6.0% 4.7 1.2% 0.7 26.8 15.1 4.8% 34.9 23.1 1.6% 5.7% 1.6 46.9 1.4% 25.6 3.2% 0.5 1.5% 10.7 4.2% 0.8% 4.5% 1.5% 4.7% 1.0% 1.1% Source: Source: Source: 180 | Indian Infrastructure: Evolving Perspectives

• Cost components: The cost components as percentage of cost have remained the same over the years. Interest cost for APCPDCL and APNPDCL has declined. APEPDCL has seen a continuous increase in employee costs.

REFERENCES 1. Saranga Pani, B., N. Sreekumar and M. Thimma Reddy. Power Sector Reforms in the State of Andhra Pradesh in India. 2. Government of India. 2002. Annual Report (2001–02) on the Working of State Electricity Boards & Electricity Departments. Planning Commission (Power & Energy Division). 3. Government of Andhra Pradesh. 2004. World Bank implementation completion report on a loan in the amount of US$210 million to the Government of India for Andhra Pradesh power sector restructuring project. 4. Power Finance Corporation Limited. Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05. 5. Power Finance Corporation Limited. Report on the Performance of the State Power Utilities for the Years 2005–06 to 2007–08. 6. Chatterjee, Rachel. 2003. Presentation. National Conference on Reforms in Infrastructure Sectors: Impact Assessment and Governance. 7. Central Electricity Authority. 2009. Power Sector at a Glance. 8. Saranga Pani, B., N. Sreekumar and M. Thimma Reddy. 2007. Power Sector Reforms in Andhra Pradesh: Their Impact and Policy Gaps. Distribution Reforms in Maharashtra | 181

POWER DISTRIBUTION REFORMS IN 12 MAHARASHTRA October 2009

INTRODUCTION Traditionally, the power sector in Maharashtra, excluding Mumbai, has been served by Maharashtra State Electricity Board (MSEB) which was set up in 1960 to generate, transmit and distribute power to all consumers in Maharashtra excluding Mumbai. Mumbai is served by three power utilities—Tata Power Company Ltd, Bombay Suburban Electric Supply (BSES) Ltd and Bombay Electric Supply & Transport Undertaking (BEST). MSEB was the largest SEB in the country in terms of units of power sold till 2005–06.5 Its generation capacity grew from 760 MW in 1960–61 to 9771 MW in 2001–02. MSEB's customer base of 107,833 in 1960–61 grew to 14,009,089 in 2001–02. MSEB's thermal power stations were also efficient as they achieved high power availability of 86 per cent and plant load factor of 74 per cent in 2001–02 (the average PLF [thermal] for various utilities was 69.9 per cent in 2001–02).2 By 2001–02 MSEB had a large Transmission & Distribution (T&D) network of 6.67 lakh ckt km.1

IMPERATIVE FOR REFORMS IN THE STATE Over time, the predominance of social objectives led to a lack of commercial orientation in the operations of MSEB. Further, tariffs for domestic, power looms and agricultural segments were lower than the average cost of supply of power, and were subsidized by industrial and commercial consumers. For the year 2000–01 average cost of supply for MSEB was Rs 3.65 kWh whereas average realization was 1 Rs 2.93/kWh. As shown in Figure 12.1 revenue realized from agricultural connections was far lower than the cost of supply. The distorted tariff structure led to more and more high-paying industrial consumers setting up their own captive generating stations. This led to decline in consumption 182 | Indian Infrastructure: Evolving Perspectives of power from the MSEB grid by high-paying industrial consumers, while consumption by subsidized consumer categories grew over the years. Share of electricity sold to agricultural customers grew from 25 per cent in 1993–94 to 34 per cent in 1998–99 in Maharashtra. During the same period, share of industry 2 fell from 35 per cent to 32 per cent.

5.0 4.62

4.5 4.20 4.0 Average cost of supply Rs 3.65/kWh 3.5

3.0 Rs 2.92/kWh Average realization 2.5 2.25 Rs/kWh 2.0

1.5

1.0 0.86 0.5

0.0 Commercial Industrial Residential Agriculture Figure 12.1: Average cost and realization of power in 2000–01

Further, the low tariff for subsidized consumers led not only to deterioration in financial performance, but also to wasteful consumption from these consumers. The impact of the lack of commercial focus was reflected in both the quality of supply and the performance of MSEB as shown in Box 12.1. The above factors contributed to MSEB's decline in financial health. MSEB made commercial profits without subsidy till 1994–95. Commercial profits (without subsidy) as reported in 1994–95 were Rs 276 crore. These profits declined over time and MSEB reported a commercial loss (without subsidy) of Rs 1479 crore in the 2 year 1999–2000. However, MSEB shows commercial profits during the year if the subsidy of Rs 2084 crore provided by the state government is taken into account. The commercial profits stood at Rs 605 crore. With deterioration in its financial health, MSEDCL found it difficult to invest in maintenance and upgradation of infrastructure. This led to further deterioration in the quality of supply and increase in technical losses. Caught in this downward spiral, MSEB was finding it hard to escape from declining performance. Due to financial deterioration and ever increasing need for subsidies, need for reforms became eminent. Distribution Reforms in Maharashtra | 183

Box 12.1: MSEB’s performance review

• Power deficit: The state faced a shortage in meeting overall as well as peak load requirements. Energy deficit grew from 4.5 per cent to 8.8 per cent between 1991–92 and 2001–02. The peak deficit during the same period grew from 8.7 per cent to 12.5 per cent.2 • Transmission and distribution losses (T&D losses): Ageing infrastructure, inadequate O&M of the network and low investments in new infrastructure and increased power theft led to an increase in T&D losses from 17.7 per cent in 1995–96 to 30.5 per cent in 1999–2000.2 • Gap in cost and revenue realized: The difference between the cost of power supply and the average tariff realized from the customers denotes the margin for a power 2 distribution business. This difference grew from 16.3 paise/kWh in 1995–96 to 48.8 paise/kWh in 1999–2000.2 • Increasing subsidy: The increasing difference between the cost of supply and revenue realization per unit led to increase in subsidy requirements. From 1994–95 to 1999–2000 the subsidy provided by the state government to MSEDCL increased from nil to Rs 2084 crore.2

REFORMS UNDERTAKEN Given the deteriorating financial health of MSEB and its impact on the state, the Government of Maharashtra (GOM) decided to review the power situation in the state and undertake reforms. The GOM constituted the State Electricity Restructuring Committee and the Energy Review Committee (ERC) to review the power situation in the state and suggest broad future course of reforms for the power sector in the state. The GOM came up with a white paper in August 2002, indicating reforms to be undertaken and the timelines for the same. The summary of various suggestions made in this white paper are discussed in Box 12.2. This white paper specifically mentioned that employees and unions of MSEB were opposed to unbundling and/ or privatization and stated that full operational autonomy must be given to MSEB and internal reforms should be carried out first. In less than a year of this initiative taken by the GOM, Electricity Act 2003 was passed. As per the Act, states were required to unbundle SEBs and, at the minimum, the transmission activity was to be separated from SEBs. Consequently, the GOM unbundled the MSEB in June 2005 into one holding and three subsidiary companies. The new entities formed were: • MSEB Holding Company • Maharashtra State Generation Company • Maharashtra State Transmission Company • Maharashtra State Electricity Distribution Company 184 | Indian Infrastructure: Evolving Perspectives

Box 12.2: White paper on Maharashtra power sector reforms1

Participants: Prior to the preparation, the Government of Maharashtra (GOM) invited suggestions from various stakeholders comprising industry, employees, consumers and the Maharashtra Electricity Regulatory Commission (MERC). Reform requirements identified: The reform process was expected to bring about changes under three broad categories: • Internal reforms: These reforms were expected to focus on developing human resources, implementing loss-reduction measures and anti-theft measures. To improve the quality of service, demand side management and consumer grievance redressal system were to be set up. • Independent regulatory mechanism: GOM did setup MERC under the provisions of the Electricity Regulatory Commissions Act, 1998. GOM made a commitment to ensure smooth and independent functioning of MERC. Tariff rationalization was also considered as an important measure to ensure the recovery of the cost of power supply. • Structural changes: It was identified that a vertically integrated MSEB catering to the diverse needs of a customer base has inherent limitations. GOM proposed that MSEB be restructured in order to promote and encourage efficiency, autonomy and accountability in decision making and functional specialization. Milestones: GOM identified the following milestones: • Legislative milestones: To make anti-theft legislation effective from October 2002 and pass the Maharashtra Electricity Reforms Bill in December 2002 • Efficiency improvement milestones: • To develop Consumer Charter of Rights in six months • To reduce technical losses by 1 per cent and commercial losses by 3 per cent per year in urban areas. In rural areas, technical losses to be reduced by 0.5 per cent and commercial losses by 2 per cent per year • To increase overall collection efficiency to 94 per cent in two years • To ensure metering of all agricultural consumers by December 2004

MSEB Holding Company was expected to function as a think tank and take necessary decisions relating to investment in the three companies. Maharashtra State Electricity Distribution Company Limited (MSEDCL) came into existence on 6 June 2005 as a result of this unbundling. MSEDCL is also known as Mahavitaran or Mahadiscom. MSEDCL inherited a number of problems from its predecessor, MSEB, which showed in its results for the year 2005–06: • Low collection efficiency: Collection efficiency reported by MSEB for the year 2005–06, when unbundling was done, was less than 90 per cent.5 Distribution Reforms in Maharashtra | 185

• Inadequate distribution infrastructure: MSEDCL also inherited an inadequate infrastructure which had a negative impact on technical losses as well as reliability of supply. The LT to HT ratio of distribution lines at that time was high at about 2:1 which led to high technical losses. • Aggregate Technical and Commercial Losses (AT&C): AT&C losses for the year 2005–06 stood at 36.74 per cent for MSEDCL. • Power deficit: By the year 2005–06 the peak deficit reached 23.1 per cent.3 • Gap in cost and revenue realized: For the year 2005–06 average cost of supply (ACS) and average revenue realized (ARR) stood at Rs 2.49/ kWh and Rs 2.43/ kWh respectively. • Consumer-related problems: When MSEDCL came into existence, a number of problems existed on the consumer front: • No separate consumer care centres • No call centre for complaints • No system to give feedback to consumers • Delay in supply restoration against complaints • No system for tracking status of consumer complaints Due to high level of consumer dissatisfaction caused by low quality of supply and high losses, MSEDCL decided to undertake a number of initiatives.

MSEB (Maharashtra State Electricity Board)

MSPGCL MSETCL MSEDCL (Maharashtra State (Maharashtra State (Maharashtra State Power Electricity Electricity MSEB Holding Co. Generation Co. Ltd) Transmission Co. Ltd) Distribution Co. Ltd)

Figure 12.2: Restructuring of MSEB 186 | Indian Infrastructure: Evolving Perspectives

INITIATIVES TAKEN POST-REFORMS MSEDCL decided to undertake a number of initiatives to bring about concerted changes in the distribution business and power scenario in the state. These initiatives were combined under what is called as a "ten-point programme". The programme is as follows: 1. Preventive maintenance 2. Distribution network planning 3. Consumer grievances redressal systems 4. Distribution system loss reduction 5. Improvement in collection efficiency 6. Circles to act as profit centres 7. Efficient use of technology 8. Improved services to ag. consumers 9. Improving working conditions of employees 10. Demand side management It can be seen from the ten-point programme that the initiatives taken by MSEDCL focused on three broad areas: • Initiatives to improve Quality of Supply (QoS) • Initiatives to minimize AT&C losses • Customer-centric initiatives

Initiatives to improve Quality of Supply (QoS) Quality of Supply (QoS) is determined by keeping in view a number of factors like reliability of supply, load shedding, etc. To improve QoS, MSEDCL decided to start initiatives that can ensure better management of existing infrastructure by optimizing allocation of infrastructure and better load management on demand side. Initiatives introduced to achieve better QoS by MSEDCL are: • Gaothan Feeder Separation Scheme: Under this scheme, MSEDCL is separating the rural feeders that service homes from those that feed agricultural pumpsets. Expected to be completed in two phases, this scheme sought to ensure better power supply to homes in rural areas. Guaranteed 8 hours of electricity to agricultural water pumps would help in shifting the agricultural load to non-peak hours, thereby enabling better load management on the part of MSEDCL. Gaothan Feeder Separation Scheme has been planned for more than 15000 villages. This scheme will be implemented Distribution Reforms in Maharashtra | 187

18 with an estimated total cost of Rs 2389 crore for both the phases. This scheme is expected to provide multiple benefits to MSEDCL and is also expected to bring relief to consumers by providing reliable power supply through load management. Some of the benefits this scheme is expected to provide are: 1. Uninterrupted power supply to homes in villages and suburban areas 2. As agricultural connections are low-tariff connections, separating them will lead to better power accounting. 3. Flattening of load curve: A typical load curve in MSEDCL shows peak demand for the day at around 22:00 in the evening. At that time of the day, demand-supply gap is the largest for MSEDCL. Schemes like Gaothan Feeder Scheme can help MSEDCL to achieve the flattening of this load curve in a judicial way, thereby reducing the cost on power purchase and reducing the penalty for unscheduled interchanges. Phase-I of Gaothan Feeder Separation Scheme has covered 5185 villages till April 2009. 1318 feeders have been commissioned and a load management of the order of 1592 MW has been achieved under this phase.18 Box 12.3: Load management

Load gap on a typical day in MSEDCL (6 July 2009) 3000 2573 2500 2055 2000 1500

MW 1000 500 0 -500 21–22 00–01 01–02 02–0303–04 04–05 05–06 06–07 07–08 08–09 09–10 10–11 11–12 12–13 13–14 14–15 15–16 16–1717–18 18–19 19–2020–21 22–2323–24 Time during the day

The graph above shows the gap between power supply and demand for MSEDCL on a typical day. Power requirements are high at certain times of the day, due to which peak daily deficit is high. As shown in the graph, the peak deficit is of the order of 2573 MW at 10:00 pm, whereas there is no deficit at 3:00 am.7 Schemes like Gaothan and Akshay Prakash Yojana allow MSEDCL to shift the agricultural load from high deficit time to non-deficit periods, thereby reducing daily peak load requirements. This leads to reduced load shedding.

• Akshay Prakash Yojana: Akshay Prakash Yojana (APY) is a demand side management measure, whereby MSEDCL has attempted to restrict rural feeder loads to 20 per cent of actual value by reducing pilferage, removing inefficient devices and using better load management. The programme rests on the collective responsibility of the inhabitants of the village and is carried out voluntarily for ensuring better quality of supply. 188 | Indian Infrastructure: Evolving Perspectives

This scheme was triggered by children demanding electricity for education from MSEDCL. Some MSEDCL officials made them understand that electricity is a scarce commodity and is being used through pilferage and inefficient devices in their village. The school students then approached the gram panchayat with the MSEDCL officials where it was communicated that if the villagers prudently managed their consumption, the entire village could have better quality of supply. Later, this initiative spread to other villages through the efforts made by MSEDCL. Under this scheme, villagers voluntarily restrict the use of any 3-phase load during 5pm–11pm on week days. Only lighting load is utilized. During 5pm–11pm, the load is restricted to 20 per cent of the full load. Load restrictions are supplemented by the removal of hooks and unauthorised heavy consumption devices like heaters and hotplates. Apart from this, the scheme envisages adoption of energy-saving lighting and pumps and use of capacitors. To supervise the usage restriction and reduction in unauthorised access, surveillance committees (Veej Dakshata Committee—VDC) have been formed by the villagers. This scheme was intended to benefit both the consumers as well as the utility by 1. Reducing transformer breakdowns through reduction in usage of high consumption devices and unauthorised connections 2. Reducing load shedding through better load management 3. Increasing supply of power and lesser load shedding improve conditions for rural people and help cottage industries. 4. Reducing commercial losses and maintenance costs • Single Phasing Project: The Single Phasing Scheme is also aimed at providing rural areas with uninterrupted power supply. It envisages supplying single phase rural lighting load through three single-phase transformers. Table 12.1: Progress of Single Phasing Scheme

Phase I Phase II Phase III* No. of substations 424 296 459 No. of feeders 1186 768 — No. of villages covered 8085 3877 1536 Expected load management 1153 MW 722 MW — Project cost 235 cr. 213 cr. 205 cr. * All figures for Phase III are expected numbers Source: MSEDCL Website Distribution Reforms in Maharashtra | 189

Single phasing of the selected rural mixed load feeders is carried out by using changeover switches at the sub-station. During the normal operation, the agricultural load continues to be supplied from the three-phase transformers. On operation of the changeover switch, there is no supply to the 3-phase load on the 11kV distribution network whereas single phase supply is available to the lighting and fan load. On revising changeover switch, normal 3-phase supply is restored. This scheme is being implemented in three phases out of which two have been completed already. Under these two phases, 11,962 villages were covered. The third phase is expected to cover 1,536 more villages. • Pune load shedding model: This distributed generation model was designed to achieve zero load shedding in the Pune urban circle. It was developed by MSEDCL in consultation with the Confederation of Indian Industry (CII). CII proposed to utilise surplus power available from Captive Power Producers (CPPs) during peak hours by implementing a workable alternative for harnessing distributed generation on a pilot basis. CII proposed that industries with captive or standby gensets that were drawing power from the MSEDCL grid on a 24-hour basis should reduce their off-take of power from the grid during certain specified peak periods and instead operate their own generators. The additional grid power made available through this strategy could then be diverted by MSEDCL to low voltage customers to mitigate load shedding. This would eliminate the need for load shedding in the Pune urban circle. The CPPs were reimbursed the incremental cost for electricity they generated on-site during the specified peak periods. Initiatives to minimize AT&C losses In order to minimize AT&C losses MSEDCL had to concentrate on its ageing infrastructure. High technical losses result from inadequate and sub-optimal infrastructure. The commercial losses are majorly caused by power theft. As the existing AT&C losses were very high, it was essential for MSEDCL to take a number of steps to contain and minimize them: • Investment in infrastructure: MSEDCL has started a three-year infrastructure upgradation plan, to be executed in two phases. The objective of this project is to improve existing infrastructure along with increasing the capacity of the current system in order to support expected future demand. Improved infrastructure is expected to reduce losses that arise because of outdated equipment and over-loading across the grid lines. Heavier loads result in frequent tripping of power along with transformers burning out. Hence improvement in infrastructure will lead to increased reliability of power along with lower distribution losses. 190 | Indian Infrastructure: Evolving Perspectives

MSEDCL has set the following milestones for this scheme: • Reduce loading of the distribution transformers to 80 per cent in the horizon year from the present level ranging from 100 per cent to 150 per cent. • Use of SCADA, call centres, consumer facility centres in all municipal corporation areas. • Bring down the DTC failure rate from 16.14 per cent in FY 2005–06 to 5 per cent for urban areas and 7 per cent for rural areas. • Bring down the Ag. pending connections to one month, in three years. • Meet the standard of performance given by MERC. • Power factor to be brought to 0.90, 0.95 and 0.99 in rural, urban and industrial areas respectively. As per the plan, the company is expected to set up 76,182 km of power distribution lines in order to improve the HT–LT ratio of the distribution network. MSEDCL is also expected to set up 565 new substations and augment the existing substations. The total project cost for infrastructure investment in 19 119 divisions is expected to be Rs 8918 crore. For speedy and qualitative development of the electricity distribution grid in Maharashtra and to ensure speedy implementation of its ambitious infrastructure development and upgradation plans, the MSEDCL has decided to engage professional services from project management consultants (PMCs). To ascertain cost control and quality management under this program, MSEDCL has taken the following steps: • Setting up of six Quality Control Labs with state-of-the-art testing equipments at Kolhapur, Pune, Bhandup, Nasik, Aurangabad and Nagpur • Formation of a quality control department to ensure purchase of the best quality material • Formation of a material specifications cell • Theft detection drive: MSEDCL launched a theft detection drive in order to improve its collection efficiency. Six dedicated police stations have been established in Maharashtra to handle power theft cases only. During FY 2007–08 about 90,000 cases of power thefts amounting to Rs 55.41 crore were detected. Speedy disposal of vigilance cases and strict action against defaulters were ensured. As a result, more than 9000 FIRs were registered against people accused of power theft. This drive was implemented consecutively for 15 days every alternate month. During April to September 2008, the drive resulted in 36383 cases and recovery of Rs 25 crore as penalties and FIRs against 3559 persons.4 Distribution Reforms in Maharashtra | 191

MSEDCL also took strict disciplinary action against delinquent employees. This is evident from the fact that, in the initial days, FIRs were filed against 22 employees. Also, disciplinary action against 389 employees was taken. • Metering and energy audit related initiatives: MSEDCL started an initiative for the metering of agricultural consumers and feeders. It also started carrying out feeder-wise energy audit (EA) to obtain feeder-wise distribution loss data. MSEDCL started to undertake Monthly Energy Accounting at division, feeder and DTC levels. By 2009, MSEDCL achieved the following milestones: • Over 5 million old consumer meters replaced in three years. This addressed concerns/complaints about meters not being read, under- or over- reporting, and manipulations. • Metering of 9339 feeders completed and carried out feeder-wise EA for all of them • Metering of 150,000 distribution transformers completed • Monthly Energy Accounting at division and DTC levels • Distribution franchisee (DF) arrangement: MSEDCL was the first distribution utility in the country to implement urban distribution franchising (DF) arrangement, wherein it franchised the circle of Bhiwandi to the private sector. Under the franchisee agreement, MSEDCL is to supply power at specified input points as per MERC regulations and directives (viz. load shedding schedule) and DF to pay the agreed input rate. DF was allowed to procure power and supply additional power over and above the supply received from MSEDCL; but no guidelines were given for such power procurement or for the recovery of related costs from consumers. DF was to pay to MSEDCL wheeling charges specified by MERC for distribution of such power. The DF is required to bring about the reduction of T&D losses to 10 per cent and increase collection efficiency to 98 per cent at the end of the franchise period. The DF arrangement at Bhiwandi has yielded successful results as shown in Table 12.2. Plans are afoot to give other circles (Nagpur, Aurangabad, Jalgaon, etc.) to private parties on similar terms. • Performance based incentives: MSEDCL introduced the concept of annual performance reports based on improvements in area-specific Aggregate Technical and Commercial (AT&C) losses and collection efficiency of its employees. Such initiatives have led to the involvement of employees in the reform process. The company also conducts management classes for its staff and sends them for training courses, besides sharing the best practices with the employees. 192 | Indian Infrastructure: Evolving Perspectives

Table 12.2: Power scenario in Bhiwandi—before and after franchising

Handover to At the end DF – Dec 2006 of 2008–09 Aggregate Technical & Commercial (AT&C) losses 58% 24% Transformer failure rate 40% 7.5% Status of consumer metering Poor with few 95% meters accurate meters accurate Indicators for Quality of Supply System Average Interruption Frequency Index (SAIFI) 47.63# 13.57* System Average Interruption Duration Index (SAIDI) 23.56# 3.55* Consumer Average Interruption Duration Index (CAIDI) 0.49# 0.26* # Feb 2007 * Jan 2009 Source: MSEDCL, TPL Customer-centric initiatives Along with concentrating on technical issues, MSEDCL introduced a number of initiatives to provide better customer services. With the usage of information technology tools, MSEDCL has been able to connect to its customers and handle their issues in a much better way. MSEDCL has established a customer database of 12 over 15 million customers. Initiatives taken by MSEDCL to serve its customers better are: • Photo metering: To address billing complaints, wrong meter readings and excessive consumption by consumers, MSEDCL took a first-of-its-kind initiative in the country. MSEDCL has started taking digital photographs of energy meters and displaying these images on energy bills. Billing is done as per the meter reading shown in the photograph. This has a number of inherent benefits. 1. It results in higher customer satisfaction as actual readings are printed on the bill. 2. It also ensures reduction in the chances of malpractices like over or under billing by MSEDCL's own employees. For this purpose, it has developed an indigenously devised software programme that captures the details which, besides showing the readings, also ensure that the meters are not tampered with or manipulated by magnetic devices. This scheme has reduced the number of disputes over billing, which ultimately leads Distribution Reforms in Maharashtra | 193

to better collections and reduced litigations. MSEDCL has covered more than 4 1 crore consumers under this scheme till September 2008. The new format also included past consumption patterns in the bill sent to consumers. This was helpful as earlier, consumers were not sure of past consumption and this had led to disputes. • Complaint handling initiatives: MSEDCL has initiated various activities for improving its response to consumers and improving complaint handling: 1. To take care of consumer complaints and give feedback to the consumers, 11 Consumer Grievance Redressal Forums have been established at various locations and internal grievance redressal units were established in all 4 O&M circles. Central Grievance Redressal System has been set up at the Head Office. 2. For handling supply-related consumer complaints 15 call centres have been 4 set up. 3. About 50 Consumer Facilitation Centres (CFCs) have been set up for 4 resolution of billing and related matters at sub-division level. 4. Grievance redressal meetings with industries and consumer associations are also organized. 5. Single coordinating agency set up to deal not only with customers but also to monitor the operational resolution of the complaints within MSEDCL. • Ease of billing: MSEDCL has started a number of initiatives to make it easy for consumers to access and pay their bills. All bills have been put on the internet to provide easy access to consumers. The payment gateways available to consumers have been increased by commissioning ATM cash collection centres, drop boxes and offering consumers the facility to make e-payments. MSEDCL started all these initiatives to ensure the economic viability of the business and to provide better services to its customers. The decision to unbundle MSEB had made employees apprehensive, as they saw unbundling as the first step towards the eventual privatization of the utility. Employees also feared mass layoffs from the utility. Therefore, one of the most important tasks before MSEDCL was to increase employee morale. To ensure the success of the initiatives, MSEDCL's decisions had to be backed by the dedication and drive of its employees. Hence, the first thing that MSEDCL did was to strengthen communication initiatives towards internal employees. To this end, it undertook 194 | Indian Infrastructure: Evolving Perspectives

• Workshops with field staff and with unions • Workshops conducted by unions for members • Staff meetings at division/circle level • In-house journal Maha Vitaran Veej Varta was used to convey the management's viewpoint to the employees, communicate new initiatives/plans, etc.

OUTCOME OF REFORMS AND INITIATIVES The overall impact of reforms and initiatives taken by MSEDCL has started producing favorable results. While the success of each initiative cannot be measured individually, a number of parameters indicating the overall health of power dstribution sector in the state are discussed below: 60.0 54.3 50.4 50.0 39.4 40.0

26.3 30.0 24.8

20.0

10.0

0.0 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 12.3: AT&C losses (%) 5 Source: PFC report, MSEDCL website • Aggregate technical and commercial losses (AT&C): AT&C losses 5 reported by MSEB in 2005–06 were very high at 50.4 per cent. These losses have been reduced to 24.8 per cent through various initiatives taken 8 by MSEDCL. • Collection efficiency: The collection efficiency has improved to the level of 8 4 96.57 per cent in FY 2008–09 from 82.96 per cent in 2005–06. • Subsidy by state: The subsidy provided by the state government to MSEDCL has been increasing over the years as shown in the graph below. Growth in subsidy despite the reforms is a cause of concern for the state. Distribution Reforms in Maharashtra | 195

120.0 96.6 93.8 97.4 100.0 83.0 80.9 80.0

60.0

40.0

20.0

0.0 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 12.4: Collection efficiency 5 Source: PFC report, MSEDCL website

2000 1,829.2 1800 1,684.0 1600 1,553.5 1,562.5 1400

1200 1,100.8 1000 800 600 400 200 0 2003–04 2004–05 2005–06 2006–07 2007–08 Figure 12.5: Subsidy from state (Rs crore) 5 Source: PFC report, MSEDCL website

• Financial viability of MSEDCL 1. Profitability: MSEDCL registered a profit of Rs 117 crore during the year 2007–08 as compared to losses of Rs 303 crore in 2005–06. However, without subsidy from the state, MSEDCL is still making losses. This loss has decreased since MSEDCL has come into existence but the change has not been significant. 2. Arrears: MSEDCL's arrears have increased from Rs 9288 crore in March 2007 to Rs 12547 crore by March 2009. Increase in arrears over the years reflects MSEDCL's inability to collect earlier dues. 196 | Indian Infrastructure: Evolving Perspectives

Table 12.3: Profits of MSEDCL (in Rs crore) 2005–06 2006–07 2007–08 Profits with subsidy (-) 303.4 (-) 133.9 117.2 Profits without subsidy (-) 1865.9 (-) 1817.9 (-) 1712.1

5 Source: PFC report, MSEDCL Annual Report 2007–08

Table 12.4: MSEDCL’s arrears (in Rs crore) March 07 March 08 March 09 Receivables 9288.5 10719.0 12547.4 Source: MSEDCL Annual Report 2007–08, MSEDCL website

3. Gap in cost and revenue realized: The gap in ACS and ARR has declined from 11 paise/kWh in 2004–05 to 3 paise/kWh in 2007–08 as shown in the graph below. 3.00 0.12

2.50 0.10

2.00 0.08

1.50 0.06

Rs/kWh

Rs/kWh 1.00 0.04

0.50 0.02

0.00 0.00 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08

ACS ARR Gap

Data prior to unbundling Data post unbundling

Figure 12.6: Average revenue realised (ARR), average cost of supply (ACS) and gap between them over the years for Maharashtra

5 Source: PFC report, MSEDCL Annual Report 2007–08

The gap between ARR without subsidy and ACS however is still high at 26 paise/kWh in 2007–08. Distribution Reforms in Maharashtra | 197

3.00 0.35

2.50 0.30

0.25 2.00 0.20 1.50 0.15

Rs/kWh Rs/kWh 1.00 0.10

0.50 0.05

0.00 0.00 2005–06 2006–07 2007–08 ACS ARR Gap Figure 12.7: Average revenue realised (ARR) without subsidy, average cost of supply (ACS) and gap between them over the years for Maharashtra post-reforms 5 Source: PFC report, MSEDCL Annual Report 2007–08

• Quality of Supply (QoS): Investment in infrastructure and other load management initiatives have led to better quality of supply. Reliability indicators like SAIFI, SAIDI and CAIDI have improved due to decreased load shedding and tripping. SAIFI, SAIDI and CAIDI improved from 10.30, 292.96 and 28.46 in April 2007 to 8.02, 159.78 and 19.92 respectively in March 2009. 12.0 10.3 10.0 9.0 8.0 8.0

6.0

4.0

2.0

0.0

April 2007 March 2008 March 2009 Figure 12.8: SAIFI 198 | Indian Infrastructure: Evolving Perspectives

350.0 293.0 300.0 238.50 250.0 200.0 159.8 150.0 100.0 50.0 0.0 April 2007 March 2008 March 2009 Figure 12.9: SAIDI 28.5 30.0 26.6 25.0 19.9 20.0 15.0 10.0 5.0 0.0 April 2007 March 2008 March 2009 Figure 12.10: CAIDI CONCLUSION MSEDCL has been in existence for a little over four years and it will be too soon to comment on the sustainability of the reforms undertaken. However, given the scale and the kind of problems it inherited, MSEDCL has done a commendable job in improving the situation of power distribution sector in Maharashtra. This improvement has been reflected by decrease in AT&C losses, improved collection efficiency and better QoS parameters. However rising arrears are a cause of concern for MSEDCL. Also, the subsidy from the state has risen over the years for MSEDCL. MSEDCL is still far from breaking even without subsidy and the gap in ARR (without subsidy) and ACS is quite high. MSEDCL's sales mix has not reflected any significant change over the years whereas revenue mix shows a declining contribution from agriculture and domestic sectors (refer to Annexure). This reflects that commercial and industrial sectors are still cross-subsidizing these sectors. Revenue contribution from agriculture and domestic sector has to increase to improve MSEDCL's financial viability. Distribution Reforms in Maharashtra | 199

Power deficit faced by Maharashtra from April 2008 to March 2009 was 15 21.4 per cent. Going forward, MSEDCL has to cater to remote and rural areas facing load shedding and bad quality of supply. With increasing arrears and huge gap in ARR and ACS, MSEDCL has a tough task ahead of itself to achieve the target of being financially viable and of providing quality customer service.

REFERENCES 1. Maharashtra Power Sector Reforms, White Paper: Industries, Energy and Labour Department, 28 August 2002 2. Annual Report (2001–02) on the Working of State Electricity Boards and Electricity Departments: Planning Commission (Power and Energy Division) Government of India, May 2002 3. All India Electricity Statistics General Review 2006, CEA 4. http://www.mahadiscom.in/aboutus/abt-us-01.shtm 5. Report on the Performance of the State Power Utilities for the Years 2004–05 to 2006–07, Power Finance Corporation Limited 6. Report on the Performance of the State Power Utilities for the Years 2002–03 to 2004–05, Power Finance Corporation Limited 7. http://www.mahadiscom.in/interpole_upload/Dailygap.pdf 8. http://mahadiscom.net/emp/Sale_Demand_%20Collection_Loss_Report/ STATE/COLLEFF.htm 9. http://www.indiaenvironmentportal.org.in/node/28548 10. Problems before Mahavitaran – Action Plan, Achievements and Future Plans towards Reforms 11. Demand Side Management to Support Electricity Grids, MSEDCL's Perspective, 26 March 2008 12. Challenges of Electricity Sector in a Developing Economy, Maharashtra Case Study, 23 April 2009 13. http://www.karmayog.org/library/libartdis.asp?r=152&libid=655 14. http://www.mahadiscom.in/AnnualPerformanceReview_13may.shtm 15. http://www.cea.nic.in/god/gmd/Monthly_Power_Supply_position/ Energy_2009_03.pdf 16. http://www.mahadiscom.in/soa/Final_statementofaccounts0607.pdf 17. http://www.mahadiscom.in/soa/final_statementofaccount0506.pdf 18. http://www.mahadiscom.in/Gaothan_Feeder_Separation_Scheme_ Project-01.shtm 19. http://www.mahadiscom.in/Infrastructure_Project-02.shtm 200 | Indian Infrastructure: Evolving Perspectives

ANNEXURE Sales mix, revenue mix and cost components of MSEDCL • Sales mix: Sales mix has not changed significantly over the years after the reforms. As shown in the table below, the share of agriculture has declined and that of industrial and commercial segment has grown over time. Table 12.5: Sales mix (as percentage of total units sold) 2004–05 2005–06 2006–07 2007–08 Domestic 16.7% 16.6% 17.1% 16.4% Commercial 4.3% 4.2% 4.5% 5.1% Agricultural 22.5% 21.8% 19.2% 22.1% Industrial 43.0% 45.6% 50.2% 47.8% Others 13.6% 11.9% 9.0% 8.6% 5 Source: PFC report, MSEDCL annual reports • Revenue mix: After the reforms, MSEDCL has witnessed a change in its revenue mix. Revenue contribution of industrial sector has gone up from 48 per cent in 2004–05 to 55 per cent in 2008–09. Consequently, despite insignificant change in sales mix, the revenue share of agriculture and domestic sector has decreased over the years. Table 12.6: Revenue mix (as percentage of total revenue) 2004–05 2005–06 2006–07 2007–08 Domestic 16.1% 15.7% 15.8% 15.0% Commercial 6.7% 6.3% 6.5% 6.8% Agricultural 12.9% 12.6% 11.0% 11.1% Industrial 48.4% 51.0% 56.3% 55.2% Others 15.9% 14.4% 10.4% 11.9% ,5 Source: PFC report MSEDCL annual reports • Cost components: The cost components of MSEDCL's total expenses are as shown in the table below. Over the years, power purchase cost has not changed much as a percentage of total cost. However, MSEDCL has seen a significant rise in its interest and financing costs. Rise in interest costs is due to capital expenditure being incurred by MSEDCL for upgrading its infrastructure. MSEDCL's administration expenses have also increased due to rise in vigilance activities. Distribution Reforms in Maharashtra | 201

Table 12.7: Expenses as percentage of total expense 2005–06 2006–07 2007–08 Purchase of power 83.01% 81.09% 81.76% Repairs and maintenance 1.49% 2.07% 2.53% Employee costs 9.34% 10.15% 8.63% Admin & general expenses 0.91% 1.03% 1.32% Depreciation 2.89% 2.50% 2.60% Interest and finance charges 2.36% 3.15% 3.17% 5 Source: PFC report, MSEDCL annual reports 202 | Indian Infrastructure: Evolving Perspectives

POWER DISTRIBUTION REFORMS IN GUJARAT 13 October 2009

INTRODUCTION The Gujarat Electricity Board (GEB) was established under Section 5 of the Electricity (Supply) Act 1948 along with the formation of Gujarat State in the year 1960. It commenced operations with a generation capacity of 315 MW and a consumer base of 1.40 million. During the 1970s and 80s, the major thrust was on the supply of electricity in the rural areas. It was largely due to GEB’s unwavering focus on rural electrification that Gujarat became the first state to achieve the landmark of “100 per cent electrification of villages”. As per the 1991 Census, 17,940 out of 18,028 villages were electrified—which was notified as close to 100 per cent.1

The impetus for reforms Over time, the emphasis of GEB on electrification, particularly in the rural areas, new connections and maintenance activities resulted in divergence from concentrating on profitability. Recovery of revenue was then considered as a secondary function. As a result, GEB faced minimum growth of revenue, rising arrears and heavy financial losses. It was also a drain on public resources due to the state’s policy of supplying electricity to agricultural consumers at extremely subsidized levels. The tariff for about 0.5 million agricultural consumers, prior to October 2000, was Rs 350 per horsepower of load connected per year, which led to a revenue realization of only Rs 0.15 per unit during 2000–01. Each incremental unit of agricultural consumption required a subsidy of at least Rs 3.00 per unit. The provision of heavily subsidized electricity to agriculture consumers boosted its share of consumption Distribution Reforms in Gujarat | 203 from 16.7 per cent of all electricity sold in the state in 1970–71 to 43 per cent in 1999–00. The loss incurred by GEB on this account was estimated at Rs 14 billion during 1999–00.2 Though GEB started certain initiatives in 2000, conditions did not improve much by the year 2004–05 (as shown in Figure 13.1).3 As a result, GEB faced recurring financial deficits and was unable to raise resources for investments. 5.00 4.65 4.50 4.30 4.00 3.50 2.96 3.03 3.00 2.77 ACS = Rs 2.49/kWh 2.50

Rs/kWh 2.00 ARR = 1.50 Rs 2.05/kWh 0.97 1.00 0.50 0.00 Domestic Commercial Agricultural Industrial Bulk Others Figure 13.1: Average revenue realization in Rs/kWh for various consumer categories The inefficiencies in the sector manifested themselves in the form of chronic shortages and unreliable service. During FY 1998–99, load shedding ranging between 50 MW and 1,450 MW was experienced on 362 days of the year.4 The Government of Gujarat (GoG) initiated an ambitious policy of inviting private sector participation (PSP) in the power sector. But the desired PSP did not materialize because the revenues generated by the sector were insufficient to service the large inflow of capital that was required.5 Due to the drain on its resources caused by supporting an inefficient power sector, the GoG was not able to increase spending on other important areas of infrastructure as well as for social services. In view of the above, GoG decided to reform the power sector in the state with the following objectives: 1. Addressing the concerns of the investors 2. Creating a business environment conducive to improving the sector’s operational efficiency, financial viability, and service to consumers GoG proposed to achieve its objectives through a number of reforms. Some of the important measures which GoG decided to take in order to achieve the targets were: 1. greater competition at all levels of the sector wherever practicable 204 | Indian Infrastructure: Evolving Perspectives

2. corporatization and commercialization of existing sector entities 3. private sector participation in the generation and distribution segments 4. tariffs enabling cost recovery as well as reasonable profits 5. an independent regulator 6. transparent, reasonable, direct, and quantified subsidies to vulnerable sections of consumers.

Implementation of reforms The promulgation of the Gujarat Electricity Industry (Reorganization and Regulation) Act in 2003 for reorganization of the electricity industry in Gujarat and for establishing an Electricity Regulatory Commission in the state paved the way for the organizational restructuring of GEB. The vertically-integrated GEB was unbundled into seven companies; one each for generation and transmission, four distribution companies (discoms) and a holding company known as Gujarat Urja Vikas Nigam Limited (GUVNL). The generation, transmission and distribution companies have been structured as subsidiaries of GUVNL. GUVNL acted as the planning and coordinating agency in the sector when reforms were undertaken. It is now the single bulk buyer in the state as well as the bulk supplier to distribution companies. It also carries out the function of power trading in the state. All companies became fully operational from April 2005 and began conducting their activities independently. Distribution in the cities of Ahmedabad and Surat has historically been with a private sector entity, viz. Torrent Power, through its fully-owned subsidiaries, Ahmedabad Electricity Company and Surat Electricity Company. A noteworthy feature of reforms in Gujarat was inclusion of representatives of the unions and associations of the staff in the restructuring process from the initial stage, i.e., from the time the decision was taken on reforming the sector. It convinced the staff that the GoG and GEB were not pursuing any hidden agenda. It thus cultivated a high level of trust and confidence amongst the staff about the aims and objectives of reforms and the process proposed to be followed to achieve them. This ensured full co-operation of the staff of GEB in the reform process. No cases of strikes or protests by employees of the erstwhile GEB were reported.

Transition support by the state government As is typically the case with structural reforms of power utilities, GoG prepared a Financial Restructuring Plan (FRP) to enable the newly formed distribution companies to start with a clean balance sheet. Under this FRP, the losses of the Distribution Reforms in Gujarat | 205 erstwhile GEB were inherited by GUVNL. GoG took over the debt payment liability of GEB. It settled outstanding dues of Rs 1627.71 crore payable to central public sector units (CPSUs) up to September 2001 and in lieu, issued bonds to these CPSUs. This payment to CPSUs since then has been regularly made through Letter of Credit without having any further problems resulting in zero outstanding dues payable to any of CPSUs. Further, GoG converted its loan to GEB, aggregating to Rs 623 crore, into equity shares in GUVNL. It also allowed a moratorium period of six years (from FY 2005–06 to FY 2010–11) on interest payment liabilities on the remaining outstanding loan of Rs 842 crore.1 The objective of this moratorium period was to enable early recovery of financial health of GUVNL. Besides this, GoG sanctioned a capital grant of Rs 250 crore per annum from FY 2005–06 to FY 2010–11 with the objective of strengthening the power sector. Such grants can be utilized for capital expenditure purposes, rural electrification projects, maintenance of quality human resources and expansion of generation capacity.1

GEB (Gujarat State Electricity Board) Structure prior to reforms

Structure post reforms

GUVNL (Gujarat Urja Vikas Nigam Limited) Holding company

GSECL GETCO DISTRIBUTION (Gujarat Energy Transmission Corp. Ltd) (Gujarat State Electricity Corp. Ltd) COMPANIES Generation company Transmission company

VGVCL DGVCL MGVCL PGVCL (Uttar Gujarat Vij (Dakshin Gujarat Vij (Madhya Gujarat Vij (Paschim Gujarat Vij Company Ltd) Company Ltd) Company Ltd) Company Ltd)

Figure 13.2: Restructuring of GEB

EARLY REFORM INITIATIVES IN GUJARAT Several states had undertaken the process of structural reforms to varying degrees before Gujarat embarked on this path. These states include Orissa, Haryana, Andhra Pradesh, Delhi, Karnataka and Uttar Pradesh. On the other hand, states like Madhya Pradesh and Maharashtra were undergoing this process around the same time as Gujarat. However, unlike other states that waited to complete structural reforms 206 | Indian Infrastructure: Evolving Perspectives before taking up comprehensive measures to address the problems facing them, Gujarat started the process of reforms in early 2000 during the GEB days. GEB undertook several initiatives to improve revenue as well as efficiency and control expenditure. A brief overview of the measures undertaken by GEB is provided below.

Revenue improvement measures GEB made significant efforts to improve its revenue through greater monitoring of the revenue situation and fixing of accountability for the same on its employees. It started monthly meetings at the zonal levels which were attended by the chief engineer and the members of the board. The objective of the meetings was to familiarize all officers concerned with the extent of the problem, fix performance parameters for the succeeding month and monitor past performance. GEB adopted the feeder manager approach to make field-level officers accountable and through monitoring of their performance, achieve results through reduction in transmission and distribution (T&D) losses. Similarly, it held deputy engineers and junior engineers responsible for sub-division-wise revenue performance parameters such as reduction in arrears.

Efficiency improvement One of the biggest achievements of GEB was its drive against power theft. GEB took stringent measures to curb theft of power and dealt sternly with cases of theft and non-payment of bills, whether by individuals or by companies. It appointed 500 retired army personnel to check power offenders and set up a vigilance department headed by an IPS officer in the rank of additional director general of police on deputation from GoG. Further, it introduced a cash reward scheme (based on the recovered amount due to submission of information) as an incentive to encourage people to come forward and submit information on theft. The informer was required to submit detailed information in a prescribed format. The name, address and amount paid to the informer were kept confidential. Besides this, GEB formed 74 inspection squads under this vigilance department. Eleven squads were dedicated to checking HT installations and the remaining were required to check LT industrial, commercial, and residential installations. These squads conducted raids during odd hours. Cases of theft led to disconnection immediately upon detection; reconnection happened only after arrears were paid by violators; many violators were convicted by the court. Managers were appointed by GEB to look into settlement of cases, and support was also to be had from GoG in the form of five dedicated police stations at Surat, Baroda, Sabarmati, Rajkot and Bhavnagar which were set up exclusively to Distribution Reforms in Gujarat | 207 deal with cases of power and power property theft. Officers of the rank of DSP, PI, PSI, and ASI from the state police department are working on deputation to these police stations. Some retired officers from the state police department are also posted here as officers on special duty. In a span of four years, almost all connections, both High Tension (HT) and Low Tension (LT), have been checked and verified. Consequently, sealing of connections was carried out and by 2005 GUVNL had sealed almost 13.89 lakh connections. In 2004–05, GUVNL recovered Rs 16 crore by settling 36,982 civil suits of power theft and malpractice.1

DISTRIBUTION REFORMS IN GUJARAT AND THEIR IMPACT The focus areas of distribution reforms in Gujarat have been as follows: • Reduction of distribution losses • Commercial loss reduction • Improvement in revenues • Improvement in customer services

Reduction of distribution losses The distribution companies in Gujarat have focused on reducing distribution losses by a combination of measures such as implementation of technology, strict measures to tackle theft, strengthening of the network, and changing processes and procedures.

Jyoti Gram Yojana Though the villages in the state were largely electrified as per prescribed parameters, there was a significant gap in the quality of power supplied to the villages. This was attributable to the use of power through illegal means resulting in frequent transformer failures, poor voltage stability and unreliability of supply. Further, there was a rapid increase in demand for power in the rural areas. Against this backdrop, the GoG launched the Jyoti Gram Yojana (JGY) as a pilot initiative in eight districts in September 2003 with the objective of supplying reliable and quality power. This scheme was part of the bigger objective of facilitating growth of the rural economy in the state. The pilot was successfully completed in October 2004 and in November 2004 the scheme was extended to the entire state. The JGY had the following characteristics: • Bifurcation of rural feeders into: • agricultural feeders catering solely to demand for agricultural purposes • rural feeders catering to load other than agriculture 208 | Indian Infrastructure: Evolving Perspectives

• erection of 11/22 kV HT lines in rural areas to separate the agriculture load from the village transformer centre • metering of transformers on JGY feeders • providing round-the-clock 3-phase power supply to consumers other than agricultural consumers while ensuring improved quality of a minimum of 8 hours’ continuous power supply at a pre-determined schedule to agriculture. At the end of FY 2007–08, 17,839 villages were covered under the JGY. It involved laying a parallel rural transmission network across the state involving the erection of 15,500 transformers and 75,000 km of lines at an investment of Rs 1,200 crore. The investment was almost entirely funded through grants from GoG (Rs 1017 crore) with the remaining funds being contributed by the discoms concerned, the Asian Development Bank and schemes such as the APDRP, the MLA Fund, etc.6 The Jyotigram Yojana has been successful in providing multiple benefits to both the residents of villages as well as the discoms. Some of the prominent effects seen as a result of implementation of this scheme are: • Improved standard of living: The Jyotigram Yojana has led to a substantial improvement in the standard of living of the people in the rural areas, as they are now able to access and use a wider variety of goods and instruments. • Development of small-scale industries in the rural sector has come about due to better and improved availability of power supply. • Local employment: The industrial and economic development in rural areas has led to greater employment opportunities in villages. • Reduced emigration from rural areas: Schemes like Jyotigram Yojana help check rural–urban migration as a result of the above-mentioned benefits. Non-farm activities, both trade and industry, have benefited significantly from the scheme. The rural population has been provided avenues to increase earning power and improve standards of living. Besides, there has been improvement in the availability of medical, water supply and sanitation services. This check on migration out of rural areas has also eased the pressure on urban infrastructure.

Initiatives for technical loss reduction In order to minimize distribution losses, various distribution companies in Gujarat started upgrading their infrastructure. A number of steps were taken to adjust existing infrastructure so as to optimize costs and minimize losses. Distribution Reforms in Gujarat | 209

• Feeder bifurcation: Feeder bifurcation done under JGY has had an impact on technical losses as well by reducing load on transformers and conductors. Also, reduced ampere loading of feeders has meant reduction in I2R losses. • Reduction of HT/LT ratio: Distribution companies in Gujarat took steps in order to improve their HT/LT ratio in order to reduce technical losses. Some of the steps taken were: • Use of HVDS: Discoms introduced use of high voltage distribution system (HVDS) in Gujarat. HVDS is a practice in which the HV line is extended up to the load. Required supply is then tapped off from 3-phase HV mains in proximity of a load point through a distribution transformer of lower capacity. This kind of arrangement reduces the length of LT line to just that of the service cable. Implementation of HVDS also leads to better reliability in the system as the unauthorized connections, if any, are now connected to HV line. The HV line has the capacity to handle the extra load during peak hours and hence tripping of electricity caused due to overloading is reduced. • Replacement of low capacity lines with HV lines and usage of conductors of adequate size. • Optimum loading of transformer: The loading and positioning of distribution transformers was done so as to reduce copper losses/iron losses. Usage of amorphous transformers was also introduced.

Commercial loss reduction The main reason for commercial losses suffered by distribution utilities is power theft. Gujarat was no exception; strict measures were taken to check theft. The initiatives started by GEB have been continued by GUVNL as well as the discoms. The Vigilance Department is now part of GUVNL and keeps a watch on pilferage of electricity in the state. It continues to have provisions for submission of information regarding power theft. Engineers from the discoms were deputed to GUVNL to coordinate centralized mass checking drives in so-called strong areas. Based on the consumption patterns of the feeders, theft-prone areas were identified and massive anti-theft drives organized with the help of police squads. The move was unpopular and there was stiff resistance from the people, so much so that in one instance an official was kidnapped by some locals. Other steps that were taken to avoid future occurrences of theft were: • Installation of new meters: Approximately 11.8 lakh1 metal meter boxes were installed for better energy audit and prevention of power theft. Meters were 210 | Indian Infrastructure: Evolving Perspectives

shifted outside the premises with separate services, particularly in towns. For heavy consumers or seasonal consumers, meter reading was done on a weekly basis. Metering of feeders and transformers was also undertaken. Energy audit to ensure zero theft was done regularly. Also, billing data was analyzed for irregularities. • Improved cash collection services: To boost cash collection, GUVNL set up almost 1,000 centres, outsourced to private agencies. To further improve cash collection, 9000 rural post offices were also used. The collective result of these efforts was increase in collection from Rs 10,204 crore in 2004–05 to Rs 14,767 crore in 2007–08.7 16000 1400 1,231

14000 1,092 1200

12000 959 1000 850 10000 765 800 8000 14767 13101 600 6000

Rs crore per year 11506 10204 Rs crore per month 9176 400 4000

2000 200

0 0 2003–04 2004–05 2005–06 2006–07 2007–08

Collections per year Collection per month Figure 13.3: Improvement in cash collections over the years Source: GUVNL Annual Report 2007–08 • Insulated/aerial bunch conductor: Insulated conductors were used to reduce the instances of power theft by unauthorized connections. In aerial bunched conductors, three conductors are twisted into a thicker insulated cable which makes tampering with the power line difficult.

Revenue improvement measures To improve its financial health, Gujarat discoms took a number of steps starting from reducing costs to increasing the number of connections. Some of the revenue Distribution Reforms in Gujarat | 211 improvement measures taken up along with reduction in distribution and commercial losses were: • Reduction in power purchase cost: In 2003–04, renegotiating of power purchase agreements (PPAs) began with the four independent power producers (IPPs) — Essar Power, Gujarat Paguthane, GIPCL and GSEG. This led to savings of Rs 4.95 crore in that financial year. Considering the fact that dues to the IPPs were at a staggering Rs 1,300 crore in 2003–04, IPPs were asked to work out a compromise. The alternative would have been yet another sick state corporation and suspension of operations. In another round of negotiations in 2005–06, they managed to get a further reduction of Rs 64 crore.1 • Centralized purchase cell: A centralized purchase cell was created to take the responsibility of timely and cost-effective procurements of materials and inventory planning. A development which came more in the form of a boon was the notification from the Ministry of Environment and Forests to use washed coal with ash content of less than 34 per cent, which is less polluting for power plants. This step led to savings of almost Rs 137.93 crore over the period 2002–06.1 • Releasing new connections: Camps were arranged in poor areas and slums for on-the-spot sanctioning of of connections. A number of schemes, like TASP (Tribal Area Sub-Plan), Kutir Jyoti and Zupadpatti, were started to provide connections to the poor. • Settlement of old dues: Voluntary disclosure schemes and one-time settlement schemes were also initiated to clear old dues. One-time settlement scheme was availed of by 28,793 consumers.6

Customer service improvement Providing better consumer services is one of the major challenges faced by distribution companies in the country. Consumers face a number of problems like delay in release of new connections, delay in redressal of complaints and replies to queries. To address these, discoms in Gujarat took the following measures: • Customer care centres: were set up at all sub-divisions, divisions and circle offices of distribution companies. These centres took care of all the customer queries related to new connection, billing, change of name procedures, technical parameters, etc. • Trouble call management centres: were set up to register and resolve power-supply – related complaints through telephone. Consumers can register 212 | Indian Infrastructure: Evolving Perspectives

their complaints which are dispatched to the sub-division concerned. Once the problem is resolved, the status is updated by site/sub-station. Consumers can check status of their complaints and other details through their customer number. • Bill collection arrangements: Distribution companies in Gujarat have taken a number of measures to facilitate bill payment by consumers. ‘All-time’ payment centres have been set up to allow payment 24 hours a day. Bill collection arrangements have been made with post offices, banks and other private agencies to increase the number of collection centres. Retired employees have been hired so as to increase the number of booths and shorten queues. • Introduction of Geographical Information System (GIS): GIS is a technology which integrates diverse information within a single system by putting maps and other kinds of spatial information in digital form, making connections between activities based on geographic proximity and helping decision making for system planning and maintenance. GIS is helpful in locating consumer complaints immediately as it can index consumers directly to poles based on the geographical and spatial data available. GIS also allows identification of voltage and regulation problems relating to HT & LT network.

OUTCOME OF REFORMS AND INITIATIVES Post reforms, Gujarat has turned out to be one of the few states in India which can boast power availability round-the-clock in most of its towns, cities and villages. Reforms in Gujarat led to the formation of four power distribution companies. Performance across these companies, however, varies according to the demography of the distribution areas. Various parameters indicating the performance of power distribution companies in the state are discussed below: • AT&C losses: The distribution companies in Gujarat inherited a distribution network with high AT&C losses. In the year 2004–05, GEB reported AT&C losses at 35.2 per cent.3 Post reforms, a number of initiatives were taken to reduce both technical and commercial losses. AT&C losses for the state were reported at 22.6 per cent in the year 2007–08 by GUVNL.7 AT&C losses reported by individual distribution companies have also declined over time, as shown in Table13.1. While other distribution companies have AT&C losses less than 20 per cent, PGVCL still has very high losses. PGVCL reported AT&C losses at 33 per cent for the year 2006–07.3 Distribution Reforms in Gujarat | 213

40.0 35.4 35.2 35.0

30.0 26.5 25.0 23.7 22.6

20.0

15.0

10.0

5.0

0.0 2003–04 2004–05 2005–06 2006–07 2007–08 Figure 13.4: AT&C losses (%) Source: GUVNL Annual Report 2007–08 Table 13.1: AT&C losses for distribution companies 2005–06 2006–07 2007–08 DGVCL 18.1% 16.5% 15.2% MGVCL 19.74% 15.2% 17.2% PGVCL 37.1% 35.8% 32.7% UGVCL 23.6% 15.9% 17.2% Source: PFC Report on Performance of the State Power Utilities for the Years 2005–06 to 2006–08 • Distribution losses: Gujarat Electricity Regulatory Commission (GERC) sets target distribution losses for each discom separately. While distribution losses for the discoms have decreased from what they had inherited, all discoms, except DGVCL, reported increase in distribution losses for the year 2007–08.9 Table 13.2: Distribution losses of discoms 2005–06 2006–07 2007–08 DGVCL 20.0% 16.5% 15.5% MGVCL 20.2% 15.1% 15.9% PGVCL 38.7% 32.5% 32.8% UGVCL 23.0% 15.8% 17.3% Source: GERC Website 214 | Indian Infrastructure: Evolving Perspectives

Table 13.3: Collection efficiency of distribution companies 2005–06 2006–07 2007–08 2008–09 DGVCL 102.5% 100.1% 99.8% NA MGVCL 99.8% 97.2% 100.3% 98.7% PGVCL 98.1% 97.0% NA 100.3% UGVCL 99.2% 99.8% 87.1% 100.0% Source: PFC Report on Performance of the State Power Utilities for the Years 2004–05 to 2006–07 GERC Website • Collection efficiency: The collection efficiency of GEB in Gujarat has been high at about 97 per cent in 2003–04 and 2004–05. Post-reform efficiency has improved and was reported around 98.6 Data for Gujarat has been calculated on the basis through weighed average using units sold per cent for the year 2006–07.3 Collection efficiency for different distribution companies is as shown in Table 13.3. • Subsidy: Subsidy received by GEB was Rs 1,527 crore in 2003–04. After unbundling this declined to Rs 1,178 crore in 2005–06.3 However, there has not been any significant reduction in subsidy received since then. The total subsidy received by four distribution companies for the year 2007–08 was Rs 1,182 crore8. Subsidy received by individual distribution companies has not declined significantly since reforms. Only MGVCL has seen a decrease—from Rs 101 crore in 2005–063 to Rs 58 crore in 2007–08.8 2500

2017 2026 2000

1527 1500 1178 1206 1182

1000

500

0 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Figure 13.5: Subsidy received (in Rs crore) Source: PFC Report on Performance of the State Power Utilities Distribution Reforms in Gujarat | 215

Table 13.4: Subsidy received by distribution companies 2005–06 2006–07 2007–08 DGVCL 74 79 81 MGVCL 101 58 58 PGVCL 427 474 466 UGVCL 576 595 578 Gujarat 1,178 1,206 1,182 Source: PFC Report on Performance of the State Power Utilities for the Years 2004–05 to 2006–08, Annual Report of Distribution Companies. • Financial viability of the distribution companies • Gap in cost and revenue realized: Gap in ACS and ARR has declined from Rs 0.51/kWh in 2002–03 to Rs 0.24/kWh in 2007–08* (as shown in Chart 5). This decrease is attributable to a number of initiatives taken to reduce commercial as well as technical losses. 3.5

3.0

2.5

2.0

Rs/kWh 1.5

1.0 0.70 0.51 0.44 0.5 0.26 0.26 0.24

0.0 2002–03 2003–04 2004–05 2005–06* 2006–07* 2007–08*

ACS ARR Gap Figure 13.6: Average revenue realised (ARR) without subsidy, average cost of supply (ACS) and gap between them over the years for Gujarat (in Rs/kWh) *Data for year 2005–06, 2006–07 and 2007–08 has been calculated through weighted average using units sold. Source: PFC Report on Performance of the State Power Utilities The gap in ARR and ACS for distribution companies varies by a great margin. While DGVCL had a gap of 5 paise/kWh in 2007–08, the gap for UGVCL was 44 paise/kWh. 216 | Indian Infrastructure: Evolving Perspectives

Table 13.5: Gap between ARR & ACS for distribution companies without subsidy (in Rs/kWh) 2005–06 2006–07 2007–08 DGVCL 0.07 0.06 0.05 MGVCL 0.16 0.10 0.09 PGVCL 0.25 0.27 0.26 UGVCL 0.47 0.48 0.44 Source: PFC Report on Performance of the State Power Utilities for the Years 2004–05 to 2006–07, Annual Report of Discoms and GUVNL • Profitability: Without subsidy, distribution companies in Gujarat have been making losses. These have increased in the case of DGVCL and PGVCL. MGVCL has seen a decrease in losses from Rs 84 crore in 2005–06 to Rs 55 crore in 2007–08. For UGVCL, losses have increased marginally from Rs 575 crore in 2005–06 to Rs 577 crore in 2007–08. Table 13.6: Profits of distribution companies without subsidy (in Rs crore) 2005–06 2006–07 2007–08 DGVCL (64) (63) (79) MGVCL (84) (69) (55) PGVCL (401) (457) (465) UGVCL (575) (578) (577) Source: PFC Report on Performance of the State Power Utilities for the Years 2004–05 to 2006–07, Annual Report of Distribution Companies However, when subsidy has been provided all the discoms have shown profits. Table 13.7: Profits of distribution companies with subsidy (in Rs crore) 2005–06 2006–07 2007–08 DGVCL 10 16 1.6 MGVCL 17 (11) 2.4 PGVCL 27 18 1.2 UGVCL 2 17 0.9 Source: PFC Report on Performance of the State Power Utilities for the Years 2004–05 to 2006–07, Annual Report of Distribution Companies Distribution Reforms in Gujarat | 217

CONCLUSION Gujarat is one of the few states where several reforms were initiated before the actual unbundling of the SEB. A number of new initiatives were taken post restructuring by the discoms to tackle the many problems facing the distribution system. Some of these have now been adopted by other states in one form or other. The impact of distribution reforms can be felt in decreased losses and improved collection efficiencies. Also, the deficit condition of the state has improved over the years. Peak deficit for the state declined from 30 per cent in 2006–07 to 24 per cent in 2008–09. Energy deficit for the year reduced from 13 per cent in 2006–07 to 10 per cent in 2008–09. Table 13.8: Gujarat state peak deficit and energy deficit 2006–07 2007–08 2008–09 Peak deficit 30.2% 26.7% 24.3% Energy deficit 13.4% 16.2% 9.8% Source: Central Electricity Authority (CEA) Individual performance of distribution companies in Gujarat has varied. Among the factors responsible is the different customer mix they inherited (refer to Annexure). The agricultural sector’s contribution to revenue has been less than their percentage share in the sales mix; the industrial sector contributed to more than half the revenues for all the companies. This reflects the cross-subsidization being done and hence an advantage for companies having higher percentage of involvement with the industrial sector. DGVCL and MGVCL require lower subsidy than PGVCL and UGVCL which can be attributed to lower percentage of agricultural sector in their sales mix. The performance of DGVCL and PGVCL has lagged behind that of MGVCL and UGVCL respectively in terms of reduction in AT&C losses and containment of subsidy requirement. Difference in performance is also apparent with regard to Quality of Service (QoS) parameters. In the absence of data on the QoS parameters prior to restructuring, and, currently, quality data on QoS parameters on an annual basis, it is difficult to comment on the extent of improvement. Finally, the profitability of discoms without accounting for subsidy remains in doubt. Lowered dependence on the state government as far as subsidies are concerned is a must in order to achieve the ultimate goal of a financially sustainable power distribution system. 218 | Indian Infrastructure: Evolving Perspectives

Table 13.9: Quality of service parameters* for discoms Mar–08 Mar–09 SAIFI SAIDI CAIDI SAIFI SAIDI CAIDI DGVCL 13.9 0.5 0.0 5.0 3.9 0.8 MGVCL 1.7 0.3 0.2 0.8 0.3 0.4 PGVCL 0.6 2.7 4.8 NA NA NA UGVCL 0.4 0.5 1.2 0.8 1.2 1.6 Source: http://www.gercin.org/sop1.php * SAIDI and CAIDI are in hours

ANNEXURE • Sales mix: The distribution sector in Gujarat was divided into four companies and all these companies inherited a different mix of consumers depending on the area served. While DGVCL and MGVCL got a mix with predominantly industrial consumers, PGVCL and UGVCL inherited a consumer mix with a high percentage of agricultural consumers. Over the years, all the companies have seen a marginal decline in sales to agricultural consumers as percentage of total sales. The share of industrial consumers has increased significantly for PGVCL—from 32.4 per cent in 2005–063 to 41.6 per cent in 2007–08.9 • Revenue mix: The revenue mix of companies varies according to their sales mix. However, approximately half of the contribution to revenue is made by the industrial sector for all the companies. Contribution by the agricultural sector is quite low despite its accounting for a large share of sales in PGVCL and UGVCL. • Cost components: The cost components as percentage of total costs have not varied much for the distribution companies. However, most of the companies have seen a rise in employee costs as percentage of total costs. Distribution Reforms in Gujarat | 219 Table 13.10: Sales mix of distribution companies Table 13.11: Revenue mix of distribution companies 2005–06 2006–07 2007–06 Table 13.12: Expenses as percentage of total cost for distribution companies 0.3% 0.9% 1.2% 0.8% 0.4% 1.2% 1.0% 0.7% 0.7% 1.5% 1.2% 0.9% DGVCL MGVCL PGVCL UGVCL DGVCL MGVCL PGVCL UGVCL DGVCL MGVCL PGVCL UGVCL GEB DGVCL MGVCL PGVCL UGVCL DGVCL MGVCL PGVCL UGVCL DGVCL MGVCL PGVCL UGVCL GEB DGVCL MGVCL PGVCL UGVCL DGVCL MGVCL PGVCL UGVCL DGVCL MGVCL PGVCL UGVCL 2004–05 2005–062004–05 2005–06 2006–07 2007–06 2006–07 2007–06 PFCR Report on Performance of the State Power Utilities for years 2004–05 to 2006–07, GERC website PFCR Report on Performance of the State Power Utilities for years 2004–05 to 2006–07 and 2005–06 2007–08, GERC website PFCR Report on Performance of the State Power Utilities for years 2004–05 to 2006–07 and 2005–06 2007–08, GERC website DomesticCommercialAgricultural 12.4% 4.0%Industrial 14.0% 32.1%Others 5.0% 7.4% 23.6% 34.9%Source: 7.5% 69.1% 18.0% 16.5% 16.6% 40.8% 5.4% 42.8% 8.4% 4.5%Domestic 14.8% 32.4% 61.5% 2.5% 10.0%Commercial 24.0% 23.2% 7.0% 5.2%Agricultural 12.9% 6.6% 15.3% 68.6% 2.9%Industrial 15.7% 11.0% 11.0% 7.8%Others 42.9% 6.0% 8.6% 37.5% 4.3% 1.9% 21.0% 52.3% 39.4% 5.2% 15.3% 59.4% 10.4% 4.3% 75.9% 25.0% 24.1% 18.1% 2.6% 4.9% 17.3% 6.6% 67.9% 15.6% 9.5% 9.7% 51.0% 11.0% 15.1% 5.6% 5.2% 15.9% 43.1% 8.9% 11.0% 2.6% 34.7% 53.2% 5.8% 8.2% 28.3% 41.6% 12.8% 56.9% 20.3% 4.3% 48.9% 27.3% 6.1% 5.5% 1.8% 15.3% 75.3% 3.1% 4.6% 10.4% 2.8% 10.7% 4.3% 53.6% 9.6% 6.1% 8.4% 11.9% 13.4% 60.5% 20.4% 2.5% 5.8% 5.6% 27.5% 51.0% 15.3% 4.1% 11.5% 74.5% 6.7% 1.9% 10.8% 53.0% 10.9% 2.4% 4.2% 61.0% 8.6% 12.8% 5.2% 52.0% 26.4% 5.9% 5.0% 11.5% 2.3% 4.9% Power purchaseEmployee costO&M costInterest cost 93.6%DepreciationAdmin & general expenses 85.3% 1.6% 82.7% 5.2% 0.3% 2.5% 86.7% 1.2% 5.9% 90.9% 2.2% 4.3% 2.2% 81.6% 3.7% 1.9% 5.4% 83.0% 3.5% 3.6% 84.7% 1.6% 4.5% 9.0% 90.8% 2.7% 0.7% 2.4% 82.6% 6.3% 1.6% 2.1% 84.8% 3.3% 6.9% 86.1% 2.6% 1.9% 3.6% 3.5% 3.3% 1.5% 3.6% 8.0% 2.6% 1.0% 2.1% 6.2% 1.8% 1.8% 3.1% 5.5% 2.8% 1.8% 3.0% 3.0% 2.2% 2.6% 2.6% Other expenses 0.5% 0.1% -0.8% 0.1% 1.0% 0.2% 0.9% 0.0% 0.0% 0.1% 0.0% 0.1% Source: Source: 220 | Indian Infrastructure: Evolving Perspectives

REFERENCES 1. Presentation by Madhya Gujarat Vij Company Ltd. Available in the Report on “Loss reduction strategies”, September 2008, Forum of Regulators. 2. “Co-management of Electricity and Groundwater: Gujarat’s Jyotirgram Yojana, Strategic Analyses of India’s NRLP”, Tushaar Shah, Regional Workshop at Hyderabad, August 2007. 3. Presentation by Gujarat Urja Vikas Nigam Limited on “Gujarat Power Sector Initiatives” at the Regional Conference on “Excellence In Public Service Delivery”, YASHADA, Pune, October 2007. 4. http://www.karmayog.org/electricitynews/electricitynews_11067.htm 5. “Study on Impact of Restructuring of SEBs” by the Indian Institute of Public Administration (IIPA), September 2006. 6. “Gujarat electricity board’s turnaround: Complete rural electrification in Gujarat”, London Business School, October 2008. 7. “Reforms and Loss Reduction Strategies—Gujarat Experience”, P. R. Chaudhary, Officer on Special Duty, Uttar Gujarat Vij Co. Ltd. 8. “Co-Management of Electricity and Groundwater: An Assessment of Gujarat’s Jyotirgram Scheme”, Economic & Political Weekly, February 2008. 9. Asian Development Bank: Reports available under Technical Assistance: 29694, Gujarat Power Restructuring 10. http://www.gercin.org/orders_tariff.php 11. http://www.gercin.org/sop1.php 12. http://www.gercin.org/rims1.php

NOTES 1. Report on Gujarat Electricity Board —A benchmark in the progress of SEB reforms, by the Indian Institute of Planning and Management (IIPM) Ahmedabad, 2006. 2. Asian Development Bank, RRP:IND 29694, Report and recommendation of the President to the Board of Directors on proposed loans and technical assistance grants to India for the Gujarat Power Sector Development Program, November 2000. 3. Report on the Performance of the State Power Utilities for the years 2004–05 to 2006–07, Power Finance Corporation Limited. 4. Asian Development Bank, RRP:I ND 29694, Report and recommendation of the President to the Board of Directors on proposed loans and technical assistance grants to India for the Gujarat Power Sector Development Program, November 2000. Distribution Reforms in Gujarat | 221

5. Asian Development Bank, RRP:IND 29694, Report and recommendation of the President to the Board of Directors on proposed loans and technical assistance grants to India for the Gujarat Power Sector Development Program, November 2000. 6. “Reforms and Loss Reduction Strategies—Gujarat Experience” by P.R. Chaudhary, Officer on Special Duty, Uttar Gujarat Vij Co. Ltd 7. GUVNL Annual Report 2007–08 8. Annual reports for the financial year 2007–08 of the UGVCL, DGVCL, PGVCL and DGVCL 9. http://www.gercin.org 222 | Indian Infrastructure: Evolving Perspectives

BARRIERS TO DEVELOPMENT OF RENEWABLE ENERGY IN 14 INDIA AND PROPOSED RECOMMENDATIONS: A Discussion Paper February 2010

INTRODUCTION India has set out on the path of harnessing renewable energy (RE) sources like never before. The reasons are many and not hard to find—chronic shortage of power, energy security and environmental concerns. The country’s energy strategy is moving strongly in favour of RE technologies. This strategy has made India a leader in a number of renewable energy technology (RET) applications such as grid-connected wind energy generation, decentralised solar PV for rural applications, decentralized distributed generation, etc. India has now set itself very aggressive targets for RE capacity addition. The Eleventh Five-Year Plan (FYP) (FY 2007–12) envisages the addition of 14,050 MW of additional capacity, which means adding, in five years, more capacity than what India has added since Independence. However, RE capacity addition and development of the sector suffers on account of a number of constraints, overlaps and gaps prevalent in the current policy and regulatory environment. It is becoming clear that the policy and regulatory framework introduced so far has been appropriate only for accelerating the early growth of the sector from a small base and helping mainstream RE. However, this policy and regulatory environment has now (with changing market conditions and imperatives) become outmoded for the sector. Though the Ministry of New and Renewable Energy (MNRE) has been taking proactive steps to improve this environment, its initiatives have been able to address specific problems and constraints but have not been successful in helping the RE sector as a whole in India to leapfrog ahead. Barriers to Development of Renewable Energy | 223

There is, therefore, a need to review the existing environment for development of RE and propose a new approach to the development of this sector. With this objective in view, this paper examines the current status of RE development in India and the existing environment for such development. It examines the barriers to further development as well as gaps constraining investments in this sector of renewable energy. It then makes recommendations towards removing such barriers and adopting new mechanisms for the promotion of RE. In sum, the paper identifies the issues that have to be addressed in order to achieve a widespread use of RE, so that determined and practical steps can be taken to increase their application substantially. RE technologies (RETs) in India can be divided into two categories: 1. near- commercial and commercial technologies such as wind, small hydro power (SHP), solar PV, biomass and co-generation (cogen) that have matured and are being deployed or are close to deployment, and 2. emerging technologies such as solar thermal and biofuels that will need time to mature. The latter will also have to undergo pilots before commercial deployment. This paper focuses on the RETs that fall in the first category. The paper also restricts itself to grid-connected RE.

STATUS OF RE DEVELOPMENT IN INDIA Today, the RE sector contributes a very small percentage of the total installed power capacity of the country (approximately 9 per cent at the end of FY 2008–09) (see Figure 14.1). The share of different technologies in the total RE capacity existing in the country is presented in Figure 14.2. It is clear that wind energy makes up the

36878

93725 13242

4120

Hydro RES Nuclear Thermal Figure 14.1: Role of RE in India’s power generation capacity as on 31 March 2009 (in MW)1 Source: MoP 224 | Indian Infrastructure: Evolving Perspectives

RET-wise installed capacity 0.4% 5.3% 0.04% RET Installed Capacity Small hydro power 2,519.88 16.2% Cogeneration- bagasse 1,241.00 8.0% Wind power 10,891.00 70.1% Bio power 816.50 Waste to energy 67.41

Solar power 6 Biomass (5.3%) Waste to Small hydro power (16.2%) energy (0.4%) Total 15,541.79 Solar power (0.04%) Cogeneration- Wind power (70.1%) bagasse (8.0%) Figure 14.2: Technology-wise grid-interactive RE capacity in India as on 31 October 2009 (in MW) Source: MNRE largest proportion of RE. It has also overtaken the installed nuclear power capacity by nearly a factor of two. On the other hand, solar power—whether PV or thermal— is yet to gather momentum. The growth in RE capacity addition picked up pace during the Tenth Plan. In this Plan, the sector was not only able to achieve its targets but also exceeded them by almost 120 per cent. A review of the physical achievements during this Plan indicates that RE capacity of 6795.44 MW was added as against a target of 3583.50 MW. Of this, 5426.4 MW came from wind power, 536.83 MW from small hydro, 785 MW from bio-energy and 46.58 MW from waste to energy.

RE potential in India The contribution of renewable energy to the power sector has increased and is expected to increase in the future. MNRE is targeting a huge capacity of renewable energy and aims to add almost four times the present capacity by 2017. Table 14.1 highlights the potential and target cumulative capacity addition for each of the RETs in India till FY 2016–17. It is evident that wind will continue to dominate the future capacity addition from RE and the country is expected to harness around 88 per cent of its available potential of wind by 2022. SHP is also expected to be harnessed up to 43 per cent of its potential. Further, the potential for each of the RETs is expected to increase in future with more resource assessments and technological advancements. Barriers to Development of Renewable Energy | 225

Table 14.1: RE potential and target cumulative capacity addition (in MWeq)

Type of RET Estimated Target Target Total potential as on addition addition capacity March 31, till 2011 till 2017 in 2017 2009 Wind 45,195 17,600 35,000 45,243 SHP 15,000 3,376 6,500 8,930 Biomass 16,881 1,025 1,500 2,203 Cogeneration- bagasse 5,000 2,016 3,400 4,449 Waste to energy 2,700 244 600 659 Solar 50,000 53 10,000 10,002 Total 1,34,776 24,314 57,000 71,485 Source: MNRE Table 14.1 makes it clear that there is a huge potential for RE, and only a small part of it has been tapped so far. Going by the past record, these anticipated capacity additions may not materialize in their entirety as the development of RE is critically dependent on a variety of factors (which will be touched upon later in this paper). To get realistic estimates about the capacity addition that would be possible, it may be useful to consider that only 15,000 MW2 of the planned incremental capacity would be added in the country by 2017.

Drivers of RE in India The main driver for RE at the global level, particularly in Europe and North America is the reduction of emissions. Increased levels of greenhouse gases have primarily been held responsible for global warming and, consequently, climate change. Europe and North America being the largest emitters of greenhouse gases in the world, the need to reduce emissions of these gases provides a very compelling reason for them to make use of alternative and cleaner sources of energy. While the need to protect and preserve the environment has come to the forefront in India, concerns over energy security and the stability of the energy supply continue to be the main drivers of RE in the country. The Expert Committee of GOI on Integrated Energy Policy (IEP) notes that to deliver a sustained growth of 8 per cent through 2031, India would need to augment its primary energy supply by three to four times and electricity supply by five to seven times the 2003–04 levels.3 The country currently imports about 72 per cent of its oil consumption and this is expected to reach 90 per cent by 226 | Indian Infrastructure: Evolving Perspectives

2031–32. The scenario for coal imports is not going to be very different. It is envisaged that India will import 50 to 60 million tons (MT) of coal every year by the end of the Eleventh Five-Year Plan. According to scenarios developed by the Expert Committee on IEP, imports could increase to as much as 45 per cent of the total coal requirement. Besides issues of energy security, such growing dependence on imports also raises concerns of price shocks and vulnerability to supplying countries. Long-term energy security is just one aspect. The country also needs to address the shortage of power that has engulfed it over the years and access to electricity. The peak power shortage in June 2009 was 14 per cent and has been upwards of 11 per cent every year since 1997–98.4 The scenario varies from state to state with some states facing a peak power shortage of 35 per cent in June 2009.5 While the level of village electrification for the country as a whole reached 83 per cent at the end of June 2009,6 the level of household electrification continues to remain poor. Last available estimates indicate that 90 per cent of urban households and only about 55 per cent of rural households are electrified.7 No doubt, efforts are being made to increase electrification. But given the shortage of power prevailing in the country, increased electrification would perhaps make no difference. Allied benefits of energy security are savings in foreign exchange on account of reduction in import of conventional fuels. Another offshoot of any scale-up in RE investment and development would be more investment in RE manufacturing. This, in turn, would lead to savings in foreign exchange (from import of RE equipment), spur development of equipment manufacturing and ancillary industries specific to renewable energy technologies, and generate employment. Promoting renewable energy resources also has a positive impact on the net creation of jobs. Rough estimates indicate that a 4000 MW ultra mega power project (thermal power) would create employment for approximately 300 people. One MW of RE necessitates the employment of a minimum of five people, which means that about 20,000 people would get employment through 4000 MW of RE. International experience also bears this out. For instance, RE jobs in Germany shot up from 160,500 in 2004 to 249,300 in 2007.8

Policy and regulatory framework for RE Overall environment for development of RE India is one of the few countries in the developing world which has pioneered the development of renewable energy. Following the first oil shock in the 1970s which brought to light concerns about energy access and energy security, India recognized the relevance of these natural sources of energy. Thereafter, the sector witnessed Barriers to Development of Renewable Energy | 227 slow but steady growth over the next three decades. The milestones in the RE sector in India can be summarized as follows: • Establishing the Commission for Additional Sources of Energy in 1981 for promoting research and development in renewable energy. • Establishing the Department of Non-conventional Energy Sources (DNES) in 1982 in the Ministry of Energy • Wind-resource assessment and publication of a databook in the early 1980s • Research and development, capacity building and demonstration programmes in the areas of biogas, cooking stoves and solar energy in the 1980s • Installing the first grid-connected wind turbine in 1985 and beginning of the demonstration programme by DNES in 1986 • Establishing the Indian Renewable Energy Development Agency (IREDA) in 1987 to finance renewable energy projects. • Upgrading DNES into a full-fledged Ministry of Non-conventional Energy Sources (MNES; now MNRE) in 1992. • Recognition of renewable technologies for power generation in 1992, by their inclusion in the Eighth Five-Year Plan (1992–97) • Policy to encourage private sector investment in renewable energy and guidelines for renewable energy tariffs by MNES in 1993 Until 1993–94, the primary approach for development of RE was through the provision of subsidies. After that, the approach has shifted to include the provision of fixed tariffs for purchase of power from RE. In 1993, the MNRE issued policy guidelines prescribing a price of Rs 2.25/kWh with a 5 per cent annual escalation (with 1993 as base year). It also allowed wheeling and banking of energy generated by RE sources to facilitate private investments in the sector. These guidelines were valid for a period of 10 years. The guidelines were adopted with variations by utilities in different states. Several states even brought out their policies for RE, based on the MNRE’s guidelines; some even offering additional incentives for RE investments. The enactment of the Electricity Act 2003 (EA 03) has radically changed the legal and regulatory framework for this sector by providing for policy formulation by the Government of India and making it mandatory for state electricity regulatory commissions (SERCs) to take steps to promote renewable and non-conventional sources of energy within their area of jurisdiction. Section 3 of EA 03 clearly mandates that the formulation of the National Electricity Policy (NEP), Tariff Policy and Plan thereof for development of power systems shall be based on optimal utilization of all resources, including renewable sources of energy. Further, EA 03 has specific provisions for determination of feed-in tariffs for renewable energy sources as well 228 | Indian Infrastructure: Evolving Perspectives as for creation of renewable portfolio standards for states. Annexure 1 describes the provisions of EA 03 and the policies formulated therein for RE. Figure 14.3 maps the capacity addition in RE with the main events that have driven this sector forward. MW 3500 3213 3250 Generation- based 3000 Policy framework Policy on incentives 2750 to push hydro power indigenisation 2500 development 2250 Guidelines 2138 Electricity 2011 National wind for tariff & 1899 2000 Act 2003 resource monitoring interconnection 1750 & demonstration for captive & 1500 programme third-party 1366 1250 MNRE policy sales & tariff 1000 849 guidelines 750 411 431 500 266 298 337 220 142 172 250 26 67 0 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 2000–01 2001–02 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 1999–2000

• Renewable Purchase Obligations • Feed-in tariffs Figure 14.3: Events influencing RE development and RE capacity addition (in MW) Source: MNRE, others Policies for promotion of RE Over the years, the GOI through the Ministry of New and Renewable Energy (MNRE; earlier known as the Ministry of Nonconventional Energy Sources), the Ministry of Finance, and the state governments have used a number of policy instruments towards promotion of RE. A summary of these initiatives is provided in Table 14.3. Annexure 2 discusses these initiatives in detail.

Regulatory framework for promotion of RE Regulatory measures have mainly taken two forms: Renewable Purchase Obligation (RPO) and Feed-in tariffs. A summary of these measures is provided in Table 14.4. Annexure 3 discusses the initiatives in detail.

Barriers to development of RE The barriers to development of RE in India, in general, are described below. Some of these may be specific to a technology, while some may be specific to a policy, site or a region. Barriers to Development of Renewable Energy | 229

Policy and regulatory barriers Policy framework for RE There is no single comprehensive policy statement for RE in the country. Policies have been issued as and when necessary to facilitate the growth of specific RETs. Further, the plans for development of RE do not match these policies. Table 14.2 indicates that the RE capacity addition targeted by MNRE and the capacity addition planned under the Jawaharlal Nehru National Solar Mission (JNNSM), also known as Solar India, is inadequate to meet the target for RE generation mandated under the National Action Plan on Climate Change (NAPCC) Table 14.2: Mismatch between RE capacity envisaged under policy and capacity addition targeted 2009–10 2010–11 2011–12 2016–17 Energy Requirement (in MU)a 820920 891203 968659 1392066 Share of RE as mandated under NAPCC (in %)b 5% 6% 7% 12% Quantum of RE required (in MU) 41046 53472 67806 167048 RE capacity addition targeted by MNRE (in MW) 15542c 20376 25211 57000 Solar capacity targeted under JNNSM (in MW) 1000 10000 Quantum of RE available (in MU)d 29952 39269 50514 129122 Additional RE required to meet RE share mandated under NAPCC (in MU) 11094 14203 17292 37926 a. As per 17th EPS b. 5% in 2009–10 & 1% increase each year c. As on 31.10.2009 d. Assuming a capacity utilization factor of 22% The policy framework at the state level is no better. In fact, in many states policies have only created uncertainty for investments in RE. For example, in Madhya Pradesh, the policy for promotion of non-conventional energy sources waives 230 | Indian Infrastructure: Evolving Perspectives wheeling charges as well as cross-subsidy surcharge for RE. The state government, to encourage RE, would provide subsidy to the distribution utilities towards wheeling charges at 4 per cent of the energy injected at the rate of prevailing energy charges for the user. The policy also exempts wind energy from the payment of electricity duty for a period of five years from CoD provided actual generation is at least 70 per cent of the energy generation declared in the DPR. However, the policy is applicable only for a period of five years. Therefore, there is a high degree of policy and regulatory uncertainty for investment in RE. In case of biomass projects, most states do not have defined policies with regard to the radius within which such plants can be established. It has been seen that biomass plants have come up in close proximity to each other, thereby affecting the availability of fuel to each other. As a result, these plants have been rendered unviable.

Provision of accelerated depreciation to wind developers Wind power growth has hinged on the 80 per cent accelerated tax depreciation that is provided by the GOI. In view of this, a bulk of wind power capacity has been set up on the balance sheets of existing companies which wanted to save income tax. Many of these projects are in fact located in low wind speed areas and have failed to deliver on the kind of energy production that was expected of them. Foreign investors who had no income tax to save did not find it lucrative to invest in wind energy assets. It has also been seen that buyers take decisions from investment in wind power projects at the last moment (just before September 30 and March 31 every year to avail themselves of the accelerated tax depreciation); the equipment suppliers in the country have evolved as developers themselves and typically undertake all development activities, including land acquisition, construction, PPA finalization and transmission tie-up. In fact, many equipment suppliers have bought vast areas of land in high wind potential sites and sold these as part of the deal to buyers. After the commissioning of the project, they even undertake operation and maintenance (O&M) for the buyers. Therefore, readymade projects are sold off-the-shelf by equipment suppliers. Since the equipment suppliers are undertaking the functions of developers as well, buyers are forced to pay a premium for the wind power projects. This has resulted in wind power projects being more expensive and even restricting competition for equipment supplies.

Regulatory framework for promotion of RE Definition of RPO A review of the RPO determined by different SERCs indicates that there are differences in the definition of the framework for RPO. There is little consensus Barriers to Development of Renewable Energy | 231 on whether a single RPO percentage should be specified for all RE sources, or RE source/RE technology-specific percentage needs to be specified. There are some issues which merit discussion here. In case technology-specific RPO is specified and there is limited availability of a particular RE source in a year, will the SERC allow such shortfall in RE procurement to be met through another type of RE source? If not, the discoms concerned may have little incentive to explore other RE sources; indirectly limiting investments in such other RE sources/RETs. Further, if the discom can meet its RPO through RETs/RE sources not specified by the SERC, it should not be liable to pay penalty for non-achievement. Another issue is that of the level of RPO. This has to be carefully determined by SERCs. While a high RPO target would incentivize discoms to purchase more RE power, thereby encouraging investments, such targets may be ambitious in the short term. On the other hand, a low target may put a restriction on the amount of energy purchased by a discom from RE sources. This was the case in Gujarat where the discoms reportedly stopped signing energy purchase agreements with wind developers as they had met their RPOs (2.28 per cent as against the mandated 2 per cent in FY 2007–08). Moreover, the discoms currently have little incentive to exceed their RPO. Finally, some states such as Maharashtra, Gujarat, Madhya Pradesh and Karnataka do not allow the procurement of RE power from outside the state. This is detrimental to the overall development of RE in the country.

Applicability of RPO Section 86(1)(e) of EA 2003 provides for specification of RPO on ‘consumption’ within the area of discoms. This implies that the RPO should be applied on entire consumption in the area of discoms and not to procurement of energy by the discoms alone. Currently, only Maharashtra, Rajasthan and Andhra Pradesh impose RPO on open access (OA) and captive consumers.

Enforcement of RPO Thus far, only a few states such as Rajasthan and Maharashtra have specified penalty mechanisms on distribution licensees in case the RPO is not met by them. Table 14.5 indicates the extent of penalty levied in these states. In the case of Rajasthan, the penalty is called an RE surcharge and is to be paid to the State Transmission Utility (STU). The surcharge so collected will be credited to a fund to be utilized for creation of transmission system infrastructure of RE plants. 232 | Indian Infrastructure: Evolving Perspectives (till 2006) currently and wind power – Solar power applications – Solar power applications Applicability intervention power generation reduce cost of capital and inturn life cycle cost of projects in small hydro, biomass and wind power production of power Objective of which incentivize RE-based down upfront investment costsbiomass hydro, small in State nodal agencies promote investments in RE Table 14.3: Policy instruments for promotion of RE cess, exemption from Ministry of Finance, Lower the gap between RE- –and wind for only AD — VAT/sales tax & electricity duty,exemption from import/excise duty, state governments based power and conventional solar technology; neutral powerother of case in Name of instrument Primary responsibility Accelerated Depreciation (AD) interventions d. Interest subsidies3 MNRE RE funds Provide a subsidy on interest to – Demonstration projects State governments and Provide low cost funds to Technology neutral 2 Production subsidies (GBI) MNREforincentive an Provide wind and Solar b. Indirect taxes c. Direct tax exemptions/tax holidays Ministry of Finance Provide direct tax exemptions Technology neutral 1a. Fiscal interventions Capital subsidy MNRE Provide a subsidy to bring – Demonstration projects Barriers to Development of Renewable Energy | 233 inviting investments Provide a financial incentive Technology neutral encouraging clean power generation development with the aim of Ministry of Environment and Forests for carbon mitigation, thereby Table 14.3: Policy instruments for promotion of RE (contd...) (compiled from various sources) development of transmission networks toconnect RE projects, and wheeling &banking, third party sale) encouraging RE investment in the state Name of Instrument Primary responsibility Objective of Intervention Applicability 5 Carbon trading 6 State RE policies (including issues such as State governments Provide a policy framework for Technology neutral 4 Demonstration projects and R&D grants MNRE Showcase technology Technology neutral Source: 234 | Indian Infrastructure: Evolving Perspectives Technology neutral Technology neutral for RE for RE projects feeding into the grid Provide an assured price RE generation purchase higher cost green power and provide incentives for RE generation distribution to encouragestate on depending Provide an assured price Table 14.4: Regulatory framework for promotion of RE determination CERC (compiled from various sources) such as open access, development oftransmission networks to connectRE projects, and wheeling & banking, SERCsthird party sale plants, and allow RE generators sale of power flexibility in generation and Type of regulationof tariff Primary responsibility Objective of the regulationRenewable portfolio standards Applicability andgeneration power technology specific projects feeding into the grid b. Terms and conditions for 3. Green power (voluntary purchase) SERCs4. Regulations addressing systemic issues State government/ Facilitate development of REtochoice consumers Allow Technology neutral Technology neutral 2. Renewable purchase obligations/ SERCs Provide a target of RE share in Technology neutral or 1. Tariff related a. Feed-in tariffs (FiT)/ Preferential SERCs Source: Barriers to Development of Renewable Energy | 235

Table 14.5: Penalties for non-achievement of RPO

Penalty clause Levied on Maharashtra Rs 5/unit for FY 2007–08; Distribution licensee, open access Rs 6/unit for FY 2008–09 consumer & captive power plant Rajasthan RE surcharge of Rs 3.59/ Distribution licensee, open access unit for FY 2007–08; consumer & captive power plant to continue until revised Source: SERC orders in states concerned Few instances of penalties on distribution licensees (discoms) for non-achievement of RPOs have been reported so far. Only the Maharashtra Electricity Regulatory Commission (MERC) has penalized the discoms for not meeting the mandated RPOs (refer Table 14.5). MERC has introduced an enforcement charge for shortfall in compliance with RPS obligations at the rate of Rs 5.00/kWh during FY 2007–08, at Rs 6.00/kWh for FY 2008–09, and at Rs 7.00/kWh for FY 2009–10. It was further clarified that this enforcement charge, if levied, shall not be allowed as ‘pass through’ expense while approving the annual revenue requirement (ARR) of the discom.

Table 14.6: Status of RPO across Maharashtra

Licensees RPO (4% RPO Shortfall Actual Penalty quantified (MU) (MU) percentage in MU) achieved (Rs crore)

MSEDCL 3058.07 2658.52 399.54 3.48 % 199.77 TATA Power 107.54 125.00 N/A 4.65 % REL 368.29 1.02 367.27 0.01 % 183.63 BEST 184.33 3.49 180.84 0.08 % 90.42 MPECS 24.04 0 24.04 0.00 % 12.02 Total 3742.27 2788.03 971.69 2.98% 485.84 Source: MERC Role of FiTs/tariff orders Table 14.7 provides an analysis of the impact of a sound regulatory framework on RET-wise capacity added at the state level. The table maps the RET-wise capacity addition with the tariff orders/FiTs issued.9 It is clear that almost all biomass and wind potential addition has been in states which have determined a FiT. In case of small hydro, nearly 83.4 per cent of the capacity has been added in states which have issued a tariff order. 236 | Indian Infrastructure: Evolving Perspectives

Table 14.7: RET capacity added across states with tariff orders/FiTs

RET Number of states Number of these Percentage of with potential states with capacity added of >100 MW tariff orders across states with tariff orders Small hydro 25 24 14 Wind 10 9 10 Biomass 16 15 13 Source: MNRE, SERCs Orissa makes for an interesting case as far as the impact of the regulatory framework on RE investments is concerned. The state has a wind potential of 255 MW. The information on state-wise cumulative wind generation available from MNRE indicates that Orissa does not (perhaps barely) contribute to wind generation. Likewise for SHP, only 44.3 MW is installed against a potential of 295 MW. It is important to note that the state does not have a proper regulatory framework for these RETs. While determination of FiT is one aspect of the regulatory framework, adequacy of this FiT is another important aspect to be addressed. In May 2006, TNERC determined the FiT for wind as Rs 2.9 a unit in 2006. This tariff was largely perceived as inadequate and is reflected in the pace of capacity addition of wind power in the state. While the state added 858 MW of wind capacity in FY 2005–06, capacity addition fell to 577 MW in FY 2006–07 and 381 MW in FY 2007–08.10 TNERC, in March 2009, revised the FiT to Rs 3.39/unit for windmills commissioned after 1 April 2009. Another case of inadequate FiT can be seen in AP. FiT determined by the Andhra Pradesh Electricity Regulatory Commission (APERC) in 2004 were ceiling tariffs. As a result, the discoms in AP were offering to procure power from RE projects at rates that were much lower than those prescribed by APERC. These tariffs, not being economically viable, several developers dropped their investment plans. Table 14.8 provides an overview of wind power capacity addition in AP to illustrate this point.

Table 14.8: Year-wise wind power capacity addition in Andhra Pradesh (in MW) 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Capacity addition 0.0 6.2 21.8 0.5 0.8 0.0 Barriers to Development of Renewable Energy | 237

It has been seen that many SERCs tend to use proxies instead of working out tariffs on a cost-plus basis. For example, among the nine SERCs that have issued tariff orders for solar PV-based power generation, only two have provided details of the parameters used for determining the tariff (Chhattisgarh and Uttar Pradesh). Some SERCs have taken the tariff of other RE technologies in the state as a reference point and added the level of subsidy (as declared by MNRE) to arrive at the tariff for solar PV-based IPPs (West Bengal). There are two issues here. Similarly, there are SERCs that have provided details of parameters used for determination of FiT of some other RETs. First, if the FiT does not reflect the underlying costs, no capacity addition will take place. Second, subsidies—where declared by MNRE—are payable for limited periods; beyond that the FiT will be inadequate. Finally, many SERCs have determined FiTs for limited periods. For instance, the West Bengal Electricity Regulatory Commission (WBERC) has prescribed FiT for five years from the date of regulations coming into force. This FiT is clearly inadequate given that the life of the power plants is 20 years, thereby giving rise to regulatory uncertainty and impacting on the bankability of the project. For any project developer and lender, it is important that the project revenue stream be known in advance, at least for the period of debt servicing (10 to 12 years). While CERC has issued regulations on the terms and conditions for determination of tariffs for different RETs, many states are yet to decide on adopting CERC tariffs. The states and developers have argued that the tariff norms adopted by CERC are inadequate. For example, in the case of solar projects, CERC does not distinguish between location/region or technology as far as PLF is concerned. In reality, the PLF would vary across locations/regions and technologies. This is also true of wind projects. Therefore, they have argued for a review of the CERC’s tariff framework. The impact of regulation and FiTs needs specific mention in case of SHPs. Almost all states determine cost plus tariffs for SHPs. As a result, to ensure high returns through high tariffs, developers have been pushing up capital cost or have not been allowing capital costs to decrease.

Third party sales through open access Section 39 of the Act directs the State Government to set up an STU which shall own the transmission network in the state and provide non-discriminatory Open Access (OA) to its network. Section 42 of the Act directs the discoms to provide non-discriminatory OA to its distribution network to eligible customers on payment of wheeling charges and other applicable surcharges. This has been imbibed in the regulatory framework in many states. However, state utilities are averse to allowing OA to RE sources. In fact, it has been seen in many cases that the distribution utilities are resenting export of RE power to 238 | Indian Infrastructure: Evolving Perspectives other states and are trying to block OA to RE-based generators with the intent of forcing them to sell this energy to the state utilities at dictated prices.11 In some states, OA for RE generators is restricted to consumer categories with lower tariffs. For example, in Gujarat, though the wind power policy of the state allows third- party sale (TPS) to any consumer in the state, in practice RE generators are allowed to sell power only to industrial consumers and not to commercial consumers. This is because commercial consumers have higher tariffs than industrial consumers and the state does not want to lose such high-paying consumers to OA. In a case such as this, developers/investors have no incentives by way of profit margins to supply to industrial consumers. Besides the above issue, states also restrict OA citing difficulty in scheduling due to the unpredictable nature of such generation. Further, even though RE generation can be predicted with 70 per cent probability, utilities insist on predictability of generation with 90 per cent probability. Another issue is the absence of firm policies or regulatory framework for wheeling charges for RE. The case of Madhya Pradesh as highlighted earlier, is an example of the lack of regulatory certainty on wheeling charges for OA in the case of RE. In the case of SHP, TPS becomes unviable after taking into account wheeling charges and free power to the state government. Institutional barriers Inter-institutional coordination Lack of coordination and cooperation within and between various ministries, agencies, institutes and other stakeholders delays and restricts the progress in RE development. A case in point is the implementation of the GBI announced for wind projects by the MNRE. IREDA started accepting applications from wind projects under the GBI scheme soon after the announcement of this scheme. However, the GOI has rejected applications that were made before the notification of the scheme through the gazette and is considering only applications made after such notification. While this justification may hold ground in principle, in practice IREDA should not have started accepting applications before the notification through gazette. Such gaps in implementation of policies on account of absence of inter-institutional coordination reduce the faith of investors in the investment climate for RE. Single-window clearance system Several states have adopted a single-window project approval and clearance system for RE. These include Punjab, Himachal Pradesh, Haryana, Rajasthan and Uttarakhand. However, the effectiveness of this system is questionable. The issue is sometimes complicated by the fact that delays in obtaining clearances for projects awarded through competitive bidding (such as SHP) result in the levy of a penalty on the developer. It is understood that Punjab is one of few states where this system Barriers to Development of Renewable Energy | 239 is robust as it sets a time-bound target for getting all approvals (a period of 60 days has been specified by Punjab) and RE developers do not have to follow up with different state government departments.

Pre-feasibility reports for hydro It is well known that in most states, the state utility or a government-owned entity entrusted with the responsibility of hydro power development is given the task of preparing the pre-feasibility reports (PFRs) for hydro projects. Several problems have been observed in this arrangement. First, the PFR is not a big priority for many state utilities, thereby leading to a delay in the preparation of the PFR and allocation of the project. Second, most of these entities follow very conventional norms which do not incorporate possible innovations from the perspective of cost reduction or capacity enhancement. As a result, developers find it difficult to rely on the state nodal agency’s PFR and have their own PFR developed. Development of PFRs may be a barrier to small and local developers seeking to implement such projects. Examples of this can be seen in Sikkim where Polyplex Industries was allocated three project sites of 40, 80 and 90 MW each. The developer was able to convert these into three projects of 100 MW each through better engineering.

Fiscal and financial barriers Budgetary constraints The GOI has announced BGI for wind, rooftop PV and for solar power plants that do not qualify under the JNNSM and sell to the state utilities. However, the extent of fund allocation towards payment of such GBI remains to be seen. The budget for FY 2010–11 is awaited in this regard and would be an indicator of GOI’s seriousness to this end. Financing of RE projects RE projects face several difficulties as far as financing is concerned. In general, the development of RE faces barriers in obtaining competitive forms of finance due to lack of familiarity with and awareness of technologies, high risk perception, and uncertainties regarding resource assessment. These have been elaborated below: • RE projects tend to have little or no fuel costs and low operation and maintenance (O&M) costs but their initial unit capital costs tend to be much higher than those of fossil generation systems. The higher ratios of capital cost to O&M cost are significant because they indicate that these projects carry a disproportionately heavy initial burden that must be financed over the life of the project. This makes exposure to risk a long-term challenge (which also has policy and regulatory-risk implications). 240 | Indian Infrastructure: Evolving Perspectives

• The risk of non-provision of subsidies on account of limited or non-availability of resources with the government is also significant since these subsidies may be the lifeline of the project. • The generally smaller nature of RE projects results in lower gross returns, even though the rate of return may be well within market standards of what is considered an attractive investment. • Developers of RE projects are often small, independent and newly established developers who lack the institutional track record and financial inputs necessary to secure non-recourse project financing. Lenders therefore perceive them as being high-risk and are reluctant to provide non-recourse project finance. They wish to see experienced construction contractors, suppliers with proven equipment, and experienced operators. • Some RETs are newly commercial and are, subsequently, not widely known among lenders (although this is changing rapidly). This results in inaccurate perceptions of risk with respect to these projects amongst lenders, thereby making financing difficult. • The intermittent generation characteristics of RETs may place them in an unfavourable position regarding structuring of contracts for power transmission as compared to conventional power projects. • For small and local developers seeking to implement RE projects, the lack of financial support for working capital requirements hinders O&M of the equipment. • The development and operation of small-scale RE projects involves the same business and financial risks as any enterprise. Variability in earnings and therefore in returns to the equity investors does not enthuse many local entrepreneurs to get involved with such projects. • The paperwork and soft costs associated with identifying and obtaining access to financing for small- and medium-scale RE projects is high relative to the financing needs. • Issues relating to underperformance or below par PLF and uncertainties inherent to such projects (like those on account of hydrology or wind pattern assumed at the time of financing) also pose a barrier to funding of projects. • Since any delay in payments by offtaking state utility would directly impact debt serviceability, lenders often seek credit enhancement mechanisms such as Debt Service Reserve Account (DSRA) or bank guarantee (BG) which may be beyond small and local developers’ reach. • Limited understanding/expertise on RE in the financial institutions also acts as a barrier to financing. Barriers to Development of Renewable Energy | 241

Market-related barriers Level playing field for RE The current structure of subsidies in the power sector is such that subsidies are effectively being provided to conventional fossil fuel resources. These give conventional fuels an unfair advantage over RE, thereby giving the impression that the difference between the price of conventional power and RE based power is too high.

Market for RE The market for RE projects/products in India can be classified into four segments: • Government market: where the government buys the output of the projects as a consumer, often providing budgetary support for it • Government-driven market: where the government pursues the use of RE in establishments outside its control for social reasons, often providing budgetary support or fiscal incentives for the same. For example, the government promotes the use of solar applications in schools, malls and hospitals. • Loan market: where people take loans to finance RE-based applications since self-financing is limited • Cash market: where high networth individuals (HNI) can buy RE-based applications to meet personal energy needs India is currently at an initial stage of the first two segments. The GOI is not focusing on promoting the third and fourth categories of RE, which may offer high potential for RE-based applications.

Fuel costs for biomass In the case of biomass-based projects, unreliable biomass supply, absence of an organized fuel market and frequent price fluctuations threaten project viability. It was seen that in early 2000, the cost of biomass was nominal (Rs 300 to Rs 400 per tonne) in most states. Over the years, the increase in demand of biomass for power generation as well as alternative uses has resulted in a demand–supply gap in this sector and resulted in spiralling biomass prices. The type of biomass available differs from state to state and so do the alternative uses for biomass. For instance, in Andhra Pradesh, rice husk is used by the fisheries industry for packaging for export purposes. But by and large biomass is alternatively used by the SME sector to replace coal for heating (operating boilers), cattle fodder and household usage in rural areas. The availability of biomass and coal in a state determines the change in price of biomass. For example, in Chhattisgarh where coal is available in abundance and at no transportation cost, 242 | Indian Infrastructure: Evolving Perspectives the degree of price elasticity for biomass is relatively very low in comparison to coal. Not surprisingly, the price of biomass is at around Rs 1615 per MT as compared to the coal price of Rs 2100 per MT (including cost of transportation).12 On the other hand, due to their geographical location, northern states such as Punjab and Haryana, have a critical constraint of coal supply to small scale industry and, therefore, the demand and price elasticity of biomass is very high. The biomass price has reached Rs 3500 per ton in these states. It may be argued that the potential for biomass-based power projects in both these states is significantly high. While that is true, the alternative usage of such biomass is also high. For instance, almost the entire quantity of wheat stalks collected is used as cattle fodder in both the states.

Inadequate market prices The price of RE power is determined on a cost-plus basis with the objective of ensuring cost recovery for RE projects. Pricing does not reflect environmental costs, thereby masking the striking environmental advantages of the new and cleaner energy options. As a corollary, it can be said that undertaking life cycle assessment costs of fossil fuels and RETs would serve to reduce the gap between the price of fossil fuel – based power and RE power. Transmission network Availability of evacuation infrastructure and grid integration are amongst the biggest problems affecting the development of RE projects, particularly SHP projects or wind projects that are located at remote locations with limited or no evacuation infrastructure. Though states are required to provide the infrastructure for evacuation of power from RE projects, in practice it is the RE developer who has to provide for such infrastructure. This has an impact on the cost of the project. Even where states provide evacuation infrastructure, such infrastructure is inadequate. In fact, instances of scaling down of RE projects due to inadequate evacuation infrastructure have come to light. For example, Sai Engineering’s 20 MW project in Toos, Kullu (HP), was scaled down to 10 MW due to the absence of adequate evacuation infrastructure. Similarly, in FY 2007–08 Tamil Nadu was unable to utilize all the power generated from wind due to lack of adequate evacuation capacity.13 It had to consequently buy more expensive power from other states to meet its needs. The small size of many RE projects and seasonality of generation add another dimension to the problem as the size of the project does not lend adequate economic viability for extending transmission lines for such projects. The issue needs to be addressed. The development of evacuation infrastructure and provision of measures for connectivity to the grid for RE sources is considered the responsibility of the Barriers to Development of Renewable Energy | 243 transmission utility. However, the distribution licensees also have a major role to play in evacuation of RE generation, as many RE sources are often connected at distribution voltages. The Forum of Regulators (FOR) has noted that except for a few utilities, such as Maharashtra State Electricity Transmission Company Ltd (MSETCL), Rajasthan Vidyut Prasaran Nigam (RVPN) and Himachal Pradesh State Electricity Board, others have not included evacuation infrastructure for RE as part of their overall transmission or distribution capex plans. Even in the case of these utilities, lack of funds is a major issue in being able to realize such plans. It is also understood that the utilities are not well aware of the transmission requirements for evacuation of RE-based power.

High equipment costs It is generally believed that volumes and advancement in technology would drive down capital costs. However, this is not always true. Several examples exist to this end—the automobile sector being, perhaps, the best example. Similarly, it has been observed that the capital cost of even the commercially deployed RETs has not declined over the years, despite increasing capacity. On the contrary, it has been observed that developers or equipment providers have been quoting increasing capital costs over the last few years. For example, a trend analysis in terms of movement of capital cost for wind projects funded by IREDA for the period from the FY 2004–05 to FY 2008–09 indicates that the average capital cost has gone up from Rs 4.79 cr/MW to Rs 5.76 cr/MW.14 Several reasons have been cited for this—the main ones being the huge demand–supply gap due to exports, inadequate built-up capacity in the Indian RE equipment industry and cartelization of equipment suppliers. As a result, the cost of power from these RETs remains high.

Inputs for RE plants Many RE projects suffer from problems similar to those faced by conventional power plants. Wind and solar thermal projects require vast areas of land. In addition to land, solar thermal projects also require huge quantities of water. The absence of water in several states having high solar power potential such as Rajasthan may complicate the task of capacity addition.

Absence of serious developers for SHP The SHP segment has seen several non-serious players who have primarily bid for projects or entered MoUs and got project allocations only to make short-term gains through the sale of projects—post-clearance—to the buyer who pays the largest premium to them. 244 | Indian Infrastructure: Evolving Perspectives

Technological barriers Technology risk In the case of many new RETs such as solar thermal, the risks related to technology are high. Since the technology is at a development stage, the risks are not clearly known. Further, even though the technology may have been deployed elsewhere in the world, its performance under Indian conditions remains to be seen. Moreover, the risk of technology obsolescence is high.

Absence of minimum standards Some RETs are characterized by the lack of minimum standards in terms of durability, reliability, performance, etc., thereby affecting their large-scale commercialization.

R&D and manufacturing capabilities One of the biggest problems confronting RETs such as solar power is the high upfront cost of establishing a solar plant. Investments in R&D with the objective of cost reduction and scaling up of operations to utilize economies of scale have long been advocated as solutions to these problems. Around the world, companies and government-backed research projects are engaged in advanced R&D and are continuously setting up bigger, more advanced manufacturing facilities. In India, however, manufacturing facilities are only focused on replicating existing technologies and are limited to small processing units. India’s manufacturing capacity is about 700 MW for PV modules as compared to facilities in countries like USA, China, Germany, Malaysia, etc., capable of multi-giga watt production. India is relying on international suppliers for equipment as well as technology. However, there is no indigenous capacity/capability for solar thermal power projects. Non-availability of local technology In many cases, the technology or equipment is imported. This implies that spare and replacement parts when required may not necessarily be readily available especially in more remote locations.

Information barriers Lack of skilled manpower Lack of trained personnel for training, demonstration, maintenance and operations, along with insufficient awareness and information programmes for technology dissemination, impedes renewable energy penetration. Experience indicates that subsequent to installation of RE projects/applications, no proper follow-up or assistance was available for their maintenance, thereby impacting their working. The impression that has been formed from such experiences is that RE installations do not work. Barriers to Development of Renewable Energy | 245

Lack of information and awareness General information and awareness in relation to new technologies and understanding the practical problems in implementing and maintaining RE projects is limited.

REVIEW OF SELECT PROGRAMMES/POLICIES OF GOI FOR PROMOTION OF RE Jawaharlal Nehru National Solar Mission As mentioned earlier, the JNNSM announced in November 2009 has brought about significant changes in the way solar power will be developed in the country. Key provisions of JNNSM are summarized in Box 14.1 (for more details, refer to Annexure 4). While the intent of the JNNSM is not under doubt, there are several issues that require greater clarity and further action. This section discusses the status of the key policy and regulatory actions being undertaken to implement the JNNSM and the concerns emerging therein.

Status of implementation of JNNSM Solar power purchase policy The GOI has appointed NVVN as the nodal agency for purchase and sale of grid- connected solar power under Phase I of the JNNSM. The solar plants participating under the scheme have to be connected to the grid at 33 kV and above. For each MW of solar power procured by NVVN under a PPA, NVVN will be allocated an equivalent amount of capacity from the unallocated power of NTPC coal-based stations. The tariff for the sale of this bundled power will be determined by CERC. In addition to this tariff, utilities will have to pay a facilitation charge to NVVN. MNRE and NVVN have estimated that the bundled power would be sold in the range of Rs 5 to Rs 5.50 per unit (see Figure 14.4). Since this price would be lower than the price of electricity purchased through the power market, discoms would be willing to buy this bundled power (see Table 14.9). Prima facie, this seems to be a good solution. By purchasing this bundled power, states/discoms would get thermal power to meet some amount of the power shortage faced by them. At the same time, they would be able to meet their RPO. MNRE and NVVN are in the process of devising guidelines for the implementation of this policy. Two distinct schemes have been devised: a Migration Scheme for existing projects, i.e., Solar Power Developers (SPDs) who have already initiated a definite process of setting up solar power plants and have made arrangements for sale of power to utilities and a scheme for new projects, i.e., SPDs approaching with new proposals for setting up solar power projects. The time frame for the signing of PPAs and power sale agreements (PSAs) under both schemes is indicated in Figures 14.5 and 14.6 and the 246 | Indian Infrastructure: Evolving Perspectives

Box 14.1: Salient features of JNNSM

• Achieving installed capacity of 20,000 MW in a phased manner by the end of the 13th Five-Year Plan in 2022

Phases Target for grid solar power including Target for off grid rooftop solar applications Phase I (2010–13) 1000–2000 MW 200 MW Phase II (2013–17) 4000 MW (10,000 MW based on 1000 MW enhanced international finance & technology transfer) Phase III (2017–22) 20,000 MW 2000 MW

• Demonstration projects focused on CSP in Phase I – 50–100 MW solar thermal plant with 4–6 hours’ storage (which can meet both morning and evening peak loads and double plant load factor up to 40%) – A 100–MW capacity parabolic trough technology based solar thermal plant – A 100–150 MW solar hybrid plant with coal, gas or bio-mass to address variability and space constraints – 20–50 MW solar plants with/without storage, based on central receiver technology with molten salt/steam as the working fluid and other emerging technologies • Shift away from GBI-based framework to one that relies on reducing the cost of delivered solar power for grid-connected solar projects – NTPC Vidyut Vyapar Nigam (NVVN) to purchase the 1,000 MW solar power (connected to 33 kV or more grid) planned in Phase I – GOI to allot another 1,000 MW capacity of thermal power from unallocated quota of NTPC stations, i.e. from power available under GOI’s discretion to allocate to states that are in shortage – NVVN to bundle this power and sell it at a rate determined by CERC • Solar-specific RPO to be fixed for states after modification of the National Tariff Policy 2006 – RPO may start with 0.25% in Phase I and increase to 3% by 2022 • Provision of a GBI to rooftop solar PV and other small solar power plants connected to LT/11 kV grid – GBI rate: tariff fixed by CERC minus notional tariff of Rs 5.5 per unit, with 3% annual escalation • Provisions for technology development, fiscal incentives, indigenization requirement and human resource development Barriers to Development of Renewable Energy | 247

Solar power developer NTPC unallocated power

PV: Rs 18.44/unit Rs 2/unit CSP: Rs 13.45/unit X kWh 3X kWh*

NVVN Price of bundled power or weighted price PV: (18.44X + 6X)/5X = Rs 6.1/unit CSP: (13.45X + 6X)/5X = Rs 4.9/unit

Sale price of power Utility If ratio of PV to CSP is 40:60 = Rs 5.36/unit If ratio of PV to CSP is 50:50 = Rs 5.49/unit Figure 14.4: Bundling mechanism for sale of solar power under JNNSM * Given the higher capacity factors of coal-based generation For illustration purposes only minimum requirements for SPDs to qualify under the schemes is listed in Table 14.9. It is important to note that out of the 1000 MW proposed to be developed under this route, 250 MW will be contributed by NTPC from its proposed solar thermal plants at Anta and Suratgarh.

MNRE Circulation requested of draft Meeting of MNRE, MoP, Signing of states for Feb 1st MOU Feb 24- Jan 11, information to be NVVN, SPDs Mar 31, PPAs & 2010 week 26, 2010 on SPDs 2010 signed 2010 & discoms PSAs interested in with SPDs for signing migrating & discoms MoUs

Figure 14.5: Time frame for completion of migration scheme under solar power purchase policy of JNNSM

Confirmation of SPDs to submit preparedness of SPDs Invitation documents & recommendations Signing of MoUs of EOI required for MOU by states with SPDs

Mar 10, Apr 30, Jun 30, Jul 15, Aug Sep Sep 30, Oct 31, 2010 2010 2010 2010 2010 2010 2010 2010

Last date for Forwarding of Selection of SPDs Signing applications & details to states by a Central of PPAs registrations for validation/ Empowered & PSAs recommendation Committee Figure 14.6: Time frame for completion of scheme for new projects under solar power purchase policy of JNNSM 248 | Indian Infrastructure: Evolving Perspectives

Table 14.9: Salient features of the schemes proposed under the solar power purchase policy of JNNSM

Migration scheme Scheme for new projects Minimum – – Net worth of the SPD for the past three qualification years (level to be determined) requirement – Turnover of SPD for the past three years for SPDs (level to be determined) – Technical requirement (to be determined) – Confirmation for plant CoD to be on or before March 31, 2013

Criteria for – Clear title and – Confirmation from STU for evacuating participation possession of land power to the grid at 33 kV and above under the (say, @ approx. two – Availability of statutory and other scheme hectares/MW) clearances as applicable – Approval from STU – Complete Detailed Project Report for evacuating power – Letter of comfort for equity/debt from to the grid at 33 kV promoter(s)/financial institution(s) and above – Letter of confirmation from the state – Confirmation from authority regarding identification/ states concerned/ notification/allotment of land for setting discoms for purchase up of the solar power plant of all the power from – Necessary water linkage from the the solar power plant state authorities concerned through NVVN (for CSP plants) – Necessary water – Submission of bank guarantee (to be linkage from the determined) state authorities – No change in equity holding permitted concerned from MOU signing till PPA execution (for CSP plants) – Letters of comfort for funding the project – Bank guarantee @ Rs 50 lakh per MW to NVVN, out of which Rs 25 lakh per MW to be given at the Barriers to Development of Renewable Energy | 249

Table 14.9: Salient features of the schemes proposed under the solar power purchase policy of JNNSM (contd...)

Migration scheme Scheme for new projects time of signing of MOU & balance Rs 25 lakh per MW to be given at the time of signing of PPA – No change in equity holding permitted from MOU signing till PPA execution

Features of – PPAs to be the same as – PPA for 25 years as per CERC regulations PPA with that signed with – Tariff to be determined by CERC SPDs discoms – SPD to deliver power at 33 kV or above substation of discom/STU – Discoms to bear transmission charges, losses, RLDC/SLDC charges, scheduling charges or any other charges for supply of solar power beyond delivery point – Billing & payment cycle as per energy accounts issued by RPC/SLDC – Scheduling of power as per Indian Electricity Grid Code – NVVN to establish irrevocable revolving letter of credit in favour of SPD prior to commencement of electricity supply from the plant – Payment Security Mechanism as per Tripartite Agreement and open irrevocable revolving letter of credit – Aggregate shareholding of promoter not to go below 51% for 3 years after COD. Any reduction thereafter to be approved by NVVN 250 | Indian Infrastructure: Evolving Perspectives

Generation-based incentive Under the JNNSM, GBI is available to solar PV installed on residential rooftops, and commercial, institutional, industrial and on other rooftops, as well as to tail- end grid-connected projects. The payment of GBI to these installations will be made by utilities on net metering basis and the GBI will be paid/reimbursed to utilities by the GOI through IREDA. The duration of the PPA to be signed between the establishments deploying rooftop PV and the utilities would be determined by the SERCs. The SERCs are also required to formulate guidelines/regulations for the metering and billing arrangements between the utility and the rooftop PV operator. It is learnt that the FOR is formulating standardized billing and payment guidelines in this regard. The FOR is also preparing a Model PPA which would be issued to the utilities by March 2010. The Central Electricity Authority (CEA) is in the process of issuing technical guidelines for the connection of such installations with the grid.

Technology to be promoted In the case of the scheme for new projects under the Solar Power Purchase Policy, the ratio of solar PV to concentrated solar power (CSP) or solar thermal is proposed at 40:60.

Demonstration projects The MNRE aims to set up these projects through the competitive bidding route. The Power Finance Corporation (PFC) is preparing bid documents for award of these projects and the bidding is expected to be initiated by the end of 2010.

Key concerns over JNNSM While the objectives of JNNSM are laudable, and so too the GOI’s intention and efforts towards its implementation, there are several concerns that need to be addressed to ensure its smooth implementation. These concerns are as follows: • The target of 20,000 MW under the JNNSM appears unrealistic given the domestic capability for solar projects, technology risks associated with solar energy and financial implications of the target. Further, the JNNSM does not provide concrete plans on the manner in which the target of 4000 MW would be scaled to 10,000 MW at the end of the Twelfth Plan. • What should be the appropriate technical and financial criteria for qualification of SPDs under the scheme for new projects in the Solar Power Purchase Policy? The GOI is keen to encourage only serious players to come forward under this scheme and is therefore contemplating rigid qualification criteria. This may, however, be detrimental to smaller players who may have genuine intent as well as the ability to set up projects. Barriers to Development of Renewable Energy | 251

• Some developers have proposed the setting up of projects over smaller areas of land than contemplated by GOI. For example, while GOI is considering land requirement of approximately two hectares/MW, some developers have come forward with proposals involving land requirement of 1.2 hectares/MW. Lower land requirement translates into lower time for land acquisition and, therefore, faster project implementation. The question, therefore, is how the GOI should choose developers who propose projects with lower land requirements. • Several developers have expressed concern over the GOI’s focus on solar thermal technology in the first phase of JNNSM. It has been argued that the GOI is ignoring global trends which indicate that solar thermal is a new and yet to be commercialized technology. However, some experts do not subscribe to this. They assert that solar thermal technology is in no way new or unproven. Another argument against the focus on solar thermal technology is the long time frame for commissioning of projects. This has an implication for commissioning solar thermal plants by 2013 to meet the targets of the first phase of JNNSM. Therefore, there is a need for GOI to back its focus on solar thermal techonology with adequate evidence. • The time frame proposed for the application and registration of new projects under the scheme for new projects in the solar power purchase policy is aggressive. This is also significant in light of the proposed conditions of the PPA which state that the aggregate shareholding of the promoter has to be more than 51 per cent for three years after COD. Given the little time given for making applications under the scheme, developers who are not able to meet the qualification requirements may tie up with anyone meeting the same to ensure qualification. However, this may become a problem in view of the PPA’s conditions. Moreover, bringing in a serious investor or partner may become difficult. Therefore, there is a need to examine the proposed time frame. • Though the mission lays a lot of thrust on R&D, the R&D strategy may actually take a long time to finalize. The GOI must fix a time frame for the setting up of the high-level research council described in the Mission, the development of the technology road map by them to achieve more rapid technological innovation and cost reduction, and the establishment of the National Centre of Excellence (NCE) to implement the technology road map. • Another aspect where the Mission is weak is the scaling up of R&D demonstration projects or pilots to the commercialization stage. The Mission talks about funding support from the NCE for performance-linked solar R&D programmes, including such demonstration projects. However, it remains silent on the support required to take projects or R&D to commercialization stages. 252 | Indian Infrastructure: Evolving Perspectives

Such support is important in keeping with the past experience of demonstration projects in RE in India where a number of pilot projects have been undertaken but have not been pursued further. • The implementation of the Mission may also face barriers due to the absence of the solar supply chain in India. There are no manufacturers for silicon crystals and wafers in the country and limited number of equipment suppliers for cells (this is true even globally). The quality of equipment is often uneven due to the absence of industry standards and the spares are expensive. In the case of rooftop PV, the lead time for balance of supply (BOS) materials such as inverters is very high. These problems are compounded by the absence of local service capability. While the Mission focuses on the creation of local capacity to build technically qualified manpower, interventions such as creation of a supply chain are weak. • The issues related to land acquisition for solar thermal projects would be the same as those for any conventional power project. Unless the issues surrounding land acquisition are addressed in the overall context of infrastructure projects, it is unlikely that the capacity addition targeted for the first phase would be met. • The Mission does not provide any incentive to states to enable speedy implementation. The only perceptible role for states is provision of land and water, and provision of infrastructure for evacuation of power from solar projects. No clarity has been provided and no mechanism spelt out on the manner in which power purchased under the solar power purchase policy by NVVN would be allocated to states. Certainty in this regard is crucial for states as they would be required to meet solar RPOs determined by SERCs going forward. A related issue here is the migration of existing projects to the solar power purchase policy. Developers are concerned that in the absence of clarity on allocation of power by NVVN, states may not allow them to migrate under this policy. It may be argued that in the event of shortage in procurement of solar power, states can fulfil their RPOs by purchase of solar RECs. But the uncertainty in the level of RPOs in future, quantum of RECs generated and price of RECs poses a barrier here. • The Mission only recommends the provision of fiscal incentives from the Ministry of Finance (MoF) in the form of customs duties and excise duties concessions/exemptions on specific capital equipment, critical materials, components and project imports. There is no commitment from the MoF to this end. This issue is critical for developers in light of the deadline for application and registration of new projects proposed by MNRE and NVVN. Barriers to Development of Renewable Energy | 253

• The Mission proposes the creation of a single-window clearance mechanism for solar power projects. However, states as well as the CERC have opined that provision of single window clearance is difficult. • Increased RPOs for solar energy would imply the need to increase consumer tariffs which may be difficult for most utilities. It may be argued that the share of solar energy in the total power purchased by a state utility would be relatively low and therefore should not have a significant impact on consumer tariff. However, in a scenario where state regulators and utilities increasingly find it difficult to make any tariff hikes, they would find it difficult to pass on even the smallest impact of increased solar RPOs to consumers. • The availability of transmission infrastructure to evacuate and transmit power that would be generated if the planned capacity addition materializes is in doubt. • The Mission provides for GBI for rooftop PV. The extent of GBI has been linked to the tariff to be determined by CERC. However, there is no mention of the time frame for which this GBI would be provided. Further, this GBI would be routed through the state utilities. The intention of utilities towards this end is doubtful. The absence of a regulatory framework for rooftop PV at the state level is also a cause for concern. Unless states are active and issue regulations enabling rooftop PV in line with the guidelines issued by FOR, the deployment of rooftop PV would not make much headway. Finally, an issue that is of concern is that projects that do not qualify under the solar power purchase policy will have to supply directly to state utilities. Even though the GOI has stated that GBI would be available to these projects as per the GBI scheme announced by GOI before the announcement of the JNNSM (see Annexure 2), developers are concerned about the poor payment security mechanism made available by state utilities as well as the financial viability of utilities.

GBI for wind As indicated in Annexure 2, MNRE is providing GBI to grid-connected wind power projects at Rs 0.50 a unit for a period not less than four years and a maximum period of 10 years in parallel with accelerated depreciation on a mutually exclusive manner, with a cap of Rs 62 lakh/MW. However, GBI will be provided only to wind projects selling power to state utilities as well as captive wind power projects. Projects undertaking TPS by way of merchant power or open access are excluded from the purview of the scheme. The GOI should reconsider the exclusion of projects undertaking TPS as this would enable the expansion of the wind power market. 254 | Indian Infrastructure: Evolving Perspectives

RECOMMENDATIONS Policy • GOI must formulate a comprehensive policy or action plan for all-round development of the sector, encompassing all the key aspects. The action plan should be prepared in consultation with the state governments. It is understood that the Energy Coordination Committee of GOI has approved the preparation of an umbrella RE law to provide a comprehensive legislative framework for all types of RETs, their usage and promotion. However, GOI has fixed no time frame for the formulation and enactment of such a law. The GOI must speed up this task and ensure that the desired law be enacted expeditiously. • ‘Must Run Status’ for RE—GOI should accord a ‘Must Run Status’ for RE- based power to ensure effective utilization of this power. The CERC—under the (Terms and Conditions for Tariff determination from Renewable Energy Sources) Regulations, 2009—has determined that all RE plants except for biomass power plants with installed capacity of 10 MW and above, and non- fossil fuel-based cogeneration plants shall be treated as ‘must run’ power plants and shall not be subjected to ‘merit order despatch’ principles. To ensure that states adopt this provision in their regulatory framework, a statement to this effect in a comprehensive policy for RE by GOI would be more effective. • States must be encouraged to remove policy and regulatory uncertainty surrounding RE. They must be encouraged to identify their thrust areas as far as RE development is concerned. Punjab is a good example here. The NRSE Policy of the state clearly specifies the objective, targets, thrust areas, and measures to achieve the targets. It also provides short-and long-term targets for the RE sector in the state. Gujarat is another example. The state government has identified wind and solar power as its thrust area. In the case of biomass, states must be encouraged to have clear policies on the radius for setting up biomass plants. Strict adherence to such a policy must be encouraged in order to ensure the viability of biomass projects. • Provision of GBI may be considered for SHPs as well. The tariff determined by CERC with normative capital cost may be adopted for this purpose. • In the case of solar thermal, a UMPP-like mechanism may be adopted for the award of projects. A special purpose vehicle (SPV) may be set up for a project wherein this SPV is responsible for all initial project activities, including land acquisition, obtaining of clearances, preparation of DPRs, tying up of basic facilities required for the implementation of the project, etc., before handing over the project to a selected developer. Barriers to Development of Renewable Energy | 255

• In the case of biomass projects, developers must be encouraged to involve the farming or fuel supply community by providing them with a share in the revenues earned from the project. • There is a need for stronger initiatives at local body levels for the promotion of RE. For example, local bodies must be discouraged from granting municipal approvals for commercial buildings in urban areas unless they house a solar application. Solar installations should be a precondition for a power connection from the utility. • The commercial success of RETs depends significantly on adoption and enforcement of appropriate standards and codes. GOI must prescribe minimum performance standards in respect of durability, reliability, and performance for different RETs to ensure greater market penetration.

Regulation • As discussed earlier, only sixteen states have notified RPO. States must be mandated to set RPO targets in a defined time frame, failing which the CERC may be given the task of determining the RPO for them. • There is an urgent need for clarity on the RPO framework. It may be better to specify the overall RPO percentage rather than technology-specific percentages. This in turn would encourage investments in RE on the basis of techno- economic analysis. Further, there should be no cap on RPO. • RPO must be levied on OA and captive consumers as well. • For RPO to be effective and their objectives to be met, it is imperative that an enforcement mechanism be introduced in all states. • SERCs must monitor the compliance of RE obligation through the ARR/Tariff approval process. Further, SERCs must consider monitoring compliance with RPO, subject to availability of energy from renewable sources (not restricted to the state), by invoking Sections 142 and 146 of EA 03 against the responsible officer of the utility. • Suitable incentives should be devised to encourage utilities to procure RE power over and above the RPO mandated by the SERC. • SERCs may amend the licence for power distribution which should be amended to include fulfilment of RPO. This would imply that non-fulfilment of RPO would be treated as violation of licence conditions and would attract suitable actions under EA 03. • A number of states (such as Maharashtra, Gujarat, Madhya Pradesh and Karnataka) do not allow the procurement of RE power from outside the state. This raises an artificial barrier in the way of RE power generation and investment 256 | Indian Infrastructure: Evolving Perspectives

across the country. Instead, regulators can identify ways and means of selling this power to neighbouring states short on RE resources or RPO at a mutually agreed upon rate. • All state governments/SERCs may consider concessional transmission on RE being sold within the state. • CERC has issued tariff guidelines covering critical aspects related to renewable energy sources from the long-term perspective of harnessing of available renewable energy potential. These guidelines provide clarity on each component of the FiT for different RETs as well as the useful life of different RETs. The control period for these guidelines is three years. States must align their FiTs to the provisions of these guidelines.

Transmission requirements • Grid connectivity to RE generation should be provided by STUs through their capex plans that are approved by the SERCs. Transmission system plans prepared by STUs should cover evacuation and transmission infrastructure requirements for RE sources. • There is a need to provide funds and capacity to STUs for this purpose. • STUs should also be made accountable and penalized if they fail to fulfil this responsibility. A possible penalty mechanism in this regard can be making the STU responsible for deemed generation if evacuation is not in place by the time of commissioning of the projects. This mechanism has been adopted in Himachal Pradesh. • There is a need to establish specific norms for grid connectivity for RE projects. SERCs can take this up under Section 86(1)(e) of EA 2003. However, since these aspects would need to be addressed as part of the larger issue of grid standards and standards for construction of transmission lines, the CEA may undertake this exercise under Sections 34 and 73(b) of EA 03. • There is a much stronger need for coordination and consultation between the STU and the nodal agency responsible for development of RE at the state level for the development of transmission infrastructure for RE projects that are in the process of being allotted or developed or are likely to be bid out in the near future.

Fiscal incentives GOI may consider fiscal incentives in the form of excise and customs duty reduction/ exemption for RE equipment. Barriers to Development of Renewable Energy | 257

Financing of RE • In order to increase the availability of funds for RE projects, GOI may consider mandating insurance companies and provident funds to invest 10 per cent of their portfolio into RE. Such investments, in fact, make business sense for the insurance companies. RE, given its benefits, will cause less damage to the environment and human health, thereby implying a lower risk of insurance payouts for these companies. • RE should be declared a priority sector. At present the priority sector broadly comprises agriculture, small-scale industries and other activities/ borrowers (such as small business, retail trade, small transport operators, professional and self-employed persons, housing, education loans, microcredit, etc.). The inclusion of RE in priority sectors will increase the availability of credit to this sector and lead to greater participation by commercial banks in this sector. • GOI should ask banks to allow an interest rebate on home loans if the owner of the house is installing an RE application such as solar water heater, solar lights or PV panel. This would incentivize people to integrate RE applications into their home, thereby encouraging the use of RE. The rebate could vary depending on the number of applications installed or the type of installations installed. • GOI may consider allowing a higher exemption on the rate of interest of home loans under income tax rebates for individuals who instal RE applications in their homes. Once again, the extent of rebate could vary depending on the number of applications installed or the type of installations installed.

Manufacturing • To achieve low-cost manufacturing and therefore lower capital costs, and to capitalize on its inherent advantages in the solar sector, India needs to consider revamping and upgrading its solar R&D and manufacturing capabilities. In this regard, GOI may consider promoting a core company to produce wafer and silicon. This will enable substantial reduction in the costs of solar technologies. • Given the continuing high capital costs of even the commercially deployed RETs despite increasing capacity, there is an urgent need to encourage price- reduced capital cost manufacturing through policy. 258 | Indian Infrastructure: Evolving Perspectives

Development of a fuel cost adjustment methodology for biomass projects In a scenario of fixed FiT, the volatility in biomass prices suppresses the PLF of power plants. Though many SERCs have revised the FiT and would do so in future, they may consider putting in place a fuel cost adjustment methodology to pass through any increases in fuel costs in tariff as has been done in the case of coal- based plants. The CERC—under the (Terms and Conditions for Tariff determination from Renewable Energy Sources) Regulations, 2009—has specified the price of biomass and bagasse for different states and determined a fuel price indexing mechanism. It has also provided the option of normative escalation of 5 per cent per annum for each subsequent year of the three-year control period. SERCs need to adopt this approach. Such fuel cost adjustment in tariff will, however, need a strong institutional set-up for monitoring the price of biomass as well as the costs of its collection, transport and storage.

Better location analysis for biomass projects In order to achieve continuous and reliable fuel supply for biomass plants, their location must be optimized. State nodal agencies must, therefore, develop a plan for development of biomass projects indicating the number and location of such plants by considering the total biomass potential available in each district, the density of such availability and potential collection centres.

Capacity building and information dissemination • There is an urgent need for technical assistance programs designed to increase the planning skills and understanding of RETs by utilities, regulators, local and municipal administrations, and other institutions involved. • Information specific to viable RETs needs to be made easily accessible both to increase general awareness and acceptability as well as to aid potential investors and sponsors of such projects. • Capacity-building initiatives should be undertaken to train people/workers to operate and maintain RE facilities • There is a need to improve the maintenance support mechanism for RE products/plants to redress the post-installation problem faced by the users. For RE plants, the after-sales service network can be strengthened by encouraging the setting up of service centres by the manufacturers which are involved in the supply of the systems. For RE applications, the same can be done through the Akshaya Urja shops. Barriers to Development of Renewable Energy | 259

ANNEXURE 1 Provisions for development of RE under Electricity Act 2003 and policies issued therein

Electricity Act 2003 The EA 03 has the following provisions for promotion and development of RE: • Section 3(1) requires GOI to prepare the National Electricity Policy and tariff policy, in consultation with the state governments and the authority for development of the power systems based on optimal utilization of resources such as coal, natural gas, nuclear substances or materials, hydro and RE. • Section 61(h) requires electricity regulatory commissions (ERCs) to consider the promotion of co-generation and generation of electricity from RE when determining the terms and conditions for the determination of tariff in their jurisdictions. • Section 86 promotes RE by ensuring grid connectivity and sale of renewable electricity. It mandates SERCs to promote cogeneration and generation of electricity from RE by providing suitable measures for connectivity with the grid and sale of electricity to any person, and also specify, for purchase of electricity from such sources, a percentage of the total consumption of electricity in the area of a distribution licensee (discoms). National Electricity Policy The NEP was notified by GOI in February 2005 as per provisions of Section 3 of EA 03. Clause 5.12 of NEP contains several conditions in respect of promotion of RE. The salient features of the said provisions of NEP are as follows: • Clause 5.12.1 targets the reduction in capital costs of RETs and identifies competition as one of the means for such reduction. It also specifies the need for adequate promotional measures for development of RETs and their sustained growth. • Clause 5.12.2 requires SERCs to determine tariffs for purchase of power from RE by discoms (until RE can compete with conventional sources in terms of cost), specifying percentages that progressively increase the share of electricity purchased by discoms from renewable sources. • Clause 5.12.3 highlights the benefits of cogeneration and promotes its use by suggesting that SERCs promote arrangements between a co-generator and a discom for purchase of surplus power from such plants. • Clause 5.2.20 states that efforts will be made to encourage private sector participation through suitable promotional measures to increase the overall share of non-conventional energy sources in the electricity mix. 260 | Indian Infrastructure: Evolving Perspectives

Tariff policy The National Tariff Policy (NTP) was notified by GOI in January 2006 as per provisions of Section 3 of EA 03. This policy has further elaborated the role of regulatory commissions, the mechanism for promoting harnessing of renewable energy and the time frame for implementation, etc. The salient features of NTP with regard to RE are as follows: • SERCs to specify minimum percentages for electricity to be purchased from RE sources by 1 April 2006. • Future procurement of RE by discoms to be done, as far as possible, through competitive bidding process (as specified under Section 63 of EA 03) by suppliers offering energy from same type of RE sources. • GOI to lay down guidelines within three months for pricing non-firm power, especially from RE sources, to be followed in cases where such procurement is not through competitive bidding.

ANNEXURE 2 Policy interventions for promotion of RE

Fiscal incentives • Capital subsidy: The MNRE has been running several capital subsidy programmes. These subsidies are provided on installation of the equipment; they are not linked to the use or performance of the equipment. • Interest subsidy: The GOI has been providing subsidies in the form of reduction in the interest rate for financing installation of equipment. Currently, interest subsidy is available to end-users of solar thermal programme, for both domestic and commercial applications. • Direct tax benefits: The GOI has offered a 10-year tax holiday under Section 80 IA of the Income Tax Act for all RE projects, including solar. It also has a scheme for accelerated depreciation under which tax savings can be claimed against investments in solar up to 80 per cent of the asset value, starting from Year I. • Indirect tax benefits: Indirect tax benefits such as reduction or exemption of electricity duty (ED), VAT, octroi or other local taxes have been used as an instrument by state governments for reducing the price the consumer pays for using RE-based power, including solar. States such as Madhya Pradesh and Punjab have exempted such projects from the payment of VAT and octroi or other local taxes. Others such as Gujarat and Madhya Pradesh (MP) have exempted consumption of electricity generated by solar power projects from payment of ED. In the case of MP, the exemption of electricity duty and cess is applicable for the first five years of the project. In Rajasthan, consumption of Barriers to Development of Renewable Energy | 261

electricity generated by solar power projects for captive use or for sale to a nominated third party attracts reduced ED (50 per cent of ED that would otherwise be applicable) for a period of seven years from the date of commissioning of the project.

Production subsidies The MNRE introduced generation-based incentives (GBI) to back up the feed-in tariffs for grid-connected solar and wind power in 2008. GBI is an attempt to change the nature of the RE industry in India, especially wind. Hitherto, wind investors primarily included Indian corporations or individuals who could offset their income tax liabilities by investing in wind or solar power through accelerated depreciation. However, few foreign firms or Independent Power Producers (IPPs) found this market attractive on account of limited or no income tax to offset. Brief descriptions of the GBI schemes are provided below.

GBI in solar projects Prior to the announcement of JNNSM, the GOI had announced the provision of GBI for grid interactive solar projects up to a maximum capacity up to 50 MW (including solar photovoltaic as well as solar thermal power generation) during the Eleventh Plan period subject to minimum installed capacity of one MW per plant. Under the scheme, a maximum cumulative capacity of 10 MW of solar PV power generation projects and 10 MW of solar thermal power generation projects could be set up in a state. The scheme envisaged provision of GBI of a maximum of Rs 12 a unit for solar PV and Rs 10 a unit for solar thermal after taking into account the per unit power purchase rate provided by the SERC or utility for that project. The GBI for a project would be determined after deducting the power purchase rate offered by the utility under the PPA from a notional amount of Rs 15 a unit for solar PV and Rs 13 a unit for solar thermal projects. The power generation plant is to be commissioned by 31 December 2009 after which the incentive will reduce by 5 per cent and the ceiling rate for the incentive would become Rs 11.40 a unit for solar PV and Rs 9.50 a unit for solar thermal projects.

GBI in wind Under this scheme, the MNRE is providing GBI to grid-connected wind power projects at Rs 0.50 a unit for a period not less than four years and a maximum period of ten years parallel with accelerated depreciation in a mutually exclusive manner, with a cap of Rs 62 lakh/MW. This implies that companies may avail themselves either of accelerated depreciation or GBI, but not both. Once a company has opted for one benefit, it cannot change the option later. The total disbursement in a year will not exceed one-fourth of the maximum limit of the incentive, i.e., 262 | Indian Infrastructure: Evolving Perspectives

Rs 15.50 lakh/MW during the first four years. The scheme will be applicable to a maximum capacity limited to 4000 MW during the remaining period of the 11 FYP. The provision of GBI will continue till the end of the Eleventh Plan. The GBI will cover wind projects selling power to state utilities as well as captive wind power projects. But projects undertaking third party sale by way of merchant power or open access are excluded from the purview of the scheme.

State-specific policies for promotion of RE A number of state governments (Karnataka, Punjab, Rajasthan, and Madhya Pradesh, to name a few) have introduced state-level policies for the promotion of RE. Some have issued policies specific to certain RETs. Gujarat and Maharashtra are cases in point, with Gujarat having issued policies specific to wind and solar energy, and Maharashtra policies for wind, waste-to-energy and cogen. The state policies encourage investments in RE through measures such as single-window clearance system, creation of green energy funds, streamlined procedures for allocation of RE projects and project sites, and other incentives such as relaxation in state taxes, etc.

RE funds In an effort to promote investment in RE, states like Maharashtra and Rajasthan have created Green Funds to provide soft loans for RE technologies. The Maharashtra Energy Development Agency (MEDA) has created a Clean Energy Fund by taxing conventional energy sources (see Box 14.2). In the case of Rajasthan, the Rajasthan Electricity Regulatory Commission has determined that any shortfall to meet the RE obligation by the distribution licensees, open access consumers and captive power users involves the payment of an RE surcharge to the State Transmission Utility (STU). The RE surcharge will be as notified by RERC from time to time. This surcharge collected by the STU is credited to a fund to be utilized for creation of a transmission system infrastructure of RE-based power plants. The state of Madhya Pradesh is also in the process of setting up a green energy fund. The fund would be financed through the cess collected from power consumers within the state.

Box 14.2: Urjankur Nidhi Fund in Maharashtra The Government of Maharashtra and the Infrastructure Leasing & Financial Services (IL & FS) have jointly promoted the Urjankur Nidhi Trust Fund to boost non- conventional energy projects in Maharashtra. This fund would develop and take up equity in RE projects. The fund has a corpus of Rs 418 crore of which Rs 218 crore would be contributed by the Government of Maharashtra. This fund would be replenished through the imposition of a green cess of of 4 paisa/unit on industrial and commercial power consumers in Maharashtra. The other Rs 200 crore would be contributed by private institutional investors. Barriers to Development of Renewable Energy | 263

The fund would initially promote bagasse based cogeneration power projects which have a significant potential in Maharashtra. These projects will be developed, implemented and operated through separate Special Purpose Vehicle (SPV) on BOOT basis, and the Urjankur Nidhi Fund along with financial institutions and private investors will take up equity in the SPV. The Trust has identified nearly 18 sugar factories and three of these sugar factories have already entered into project development agreement with the Trust. The trust would provide financial support in the form of equity with maximum support per project of up to 20 per cent of the project cost or 20 per cent of the corpus, whichever is lower. The fund will also provide crucial support functions during project development, project management and distribution of resulting power. Source: MEDA

Demonstration programmes • Tail-end Grid Connected Solar Power Generation: The tail-ends of the grid in rural areas experience voltage drops and power outages. A solar PV plant connected at the tail-end can provide power there and also improve the quality of power in the grid. In order to meet these objectives, the MNRE started a new demonstration programme, permitting utilities, generation companies and state nodal agencies to set up grid-connected solar PV plants of 25 kW to 1,000 kWp capacity. MNRE provides support of 50 per cent of the basic cost of the plant, subject to a maximum of Rs 10 crore per MWp. Assistance will be available to set up 4 MWp aggregate capacity projects in the country during the Eleventh Plan period. • Wind power: About 26 project sites have been developed in states with high potential for wind power under the Demonstration Programme to establish technological viability of wind farms, resulting in the establishment of around 57 MW of capacity. Others • Use of solar water heating systems in buildings: The GOI has been promoting solar water heating systems (SWHS). However, implementation of the scheme is tardy as several authorities are involved in implementation of any scheme involving SWHS. First, the states have to issue orders to their respective municipalities on making the SWHS compulsory. As of date, thirteen states and two union territories have issued orders making installation of the SWHS mandatory in certain categories of new buildings. The states are Andhra Pradesh, Chhattisgarh, Delhi, Haryana, Himachal Pradesh, Madhya Pradesh, Maharashtra, Nagaland, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh, and Uttarakhand, the Union Territories being Chandigarh and Dadra and Nagar Haveli. 264 | Indian Infrastructure: Evolving Perspectives

ANNEXURE 3 Regulatory framework for RE

Renewable purchase obligations Section 86(1)(e) of the Electricity Act 2003 (EA 03) empowers SERCs to specify the percentage of electricity to be procured by the obligated entities (distribution licensees, open access consumers and captive power users) from the RE sources. Accordingly, many SERCs have issued orders/regulations specifying such percentages. This percentage is referred to as Renewable Purchase Obligation (RPO). At present, 17 SERCs have notified RPO targets for their respective states. Most states have remained technology neutral while specifying these RPOs. But a few, such as Rajasthan, Madhya Pradesh, Karnataka and Chhattisgarh, have specified RPOs from individual RE sources. While most states have advocated an RPO between 1 per cent and 5 per cent, MP has advocated a 10 per cent RPO. Moreover, the states of Karnataka and Rajasthan have specified a maximum cap for RE-based procurement. Table 14.10 provides an overview of the RPOs in different states and their achievement. With the exception of Andhra Pradesh, Maharashtra and Rajasthan where RPO has been levied on discoms, open access consumers and captive power plants, it has been levied only on discoms in other states.

Table 14.10: Summary of RPOs at state level for select states

States RPO (in %) 2007-08: 4.88% 2008–09: 6.25% Rajasthan 2009–10: 7.45% 2010–11: 8.50% 2011–12: 9.50% 2007–08: 1% 2008–09: 1% Punjab 2009–10: 2% 2010–11: 3% 2011–12: 4% 2007–08: 3% Haryana 2008–09: 5% 2009–10: 10% Barriers to Development of Renewable Energy | 265

Table 14.10: Summary of RPOs at state level for select states (contd...)

States RPO (in %) Maharashtra* ‘Percentage RPO’ for each licensee shall be the same as the ‘Percentage RPO’ for the state as a whole. The ‘Percentage RPO’ for the State for a financial year shall be the ratio of ‘total RE generation’ in the state to the ‘sum of gross input energy units’ for all licensees for that financial year, excluding any inter-se sale/ consumption of electricity amongst the licensees Gujarat 2007–08: 1% 2008–09: 2% Chhattisgarh Biomass-based plants: 5% each year from 2008–09 to 2010–11 Small hydel plants: 3% each year from 2008–09 to 2010–11 Solar, wind, bagasse-based cogeneration & others: 2% each year from 2008–09 to 2010–11 Andhra Pradesh 5% each year from 2009–10 to 2013–14 Karnataka Minimum of 5% and a maximum of 10% Uttar Pradesh 7.5% 2008–09: 4.8% West Bengal** 2009–10: 6.8% 2010–11: 8.3% 2011–12: 10% Himachal Minimum 20% of total consumption during a year Pradesh

** For the purposes of determination of ‘Percentage RPO’, generation from all types of renewable energy sources as approved by MNRE is considered; Only ‘RE generation’ from grid-connected RE projects is considered; ‘RE generation’ excludes RE generation by developers meant for self- consumption and third-party sale purposes to a licensee’s consumers. * For WBSEDCL

Feed-in tariff (FiT) or preferential tariff The existing regulatory framework requires SERCs to determine FiTs for procurement of RE power by the distribution licensees under the RPO regime. It is envisaged that SERCs will determine tariff separately for each type of technology adopted for harnessing any of the RE sources. Accordingly, many SERCs have determined the FiTs for various RETs. These SERCs have generally followed a ‘cost-plus’ approach 266 | Indian Infrastructure: Evolving Perspectives Tariff plants 3.50–5.30 (Rs/kWh) per annum tariff for new up to 2011–12. projects: Tariff as Escalated at 1.5% 2006–07) with five Rs 4.08 for 2007–08, Rs 3.49 (with base year per PPAs entered earlier annual escalations @ 5% For new projects: Rs 3.37 Rs 2.50 for 2005–06 with an escalation of 4% per annum 5 Rs 3.37 Period for which tariff (from 13 6th year) 20 2.9 Escalation 3 yrs, 5% 1.25% 5.72% Project-specific capital expenses is specified 15% is allowed 1.1% of 85% of85% of 1.1% 5% Rs 3.24 if commissioned of the remaining consum- expenses O&M in 18th year 10th, 14th & (after 10 yrs) 2% in yr 4 4th year) CUF from 6th, (pre-tax) (annually) 0% (for 5 yrs) (post-tax) (annually) For old 17.63% pre tax after 31.3.2009 and civil works 0.22% 1.4.2009 19.85% pre-tax Maintenance of land if commissioned after up to 31.3.2009; capital investment; up to 31.3.2009; Rs 3.39 Table 14.11: FiTs for wind energy and assumptions across states 45 16% 16% 5% 0% 1% 1.5% for 1.10% 5% (from 4.25 16% 0% 0.50% 1.25% 5% (annual) 10 3.4 5.35 cost on equity 4.65 14% 0% 0% 1.5% 5% 20 4.30 16% 5.25 16% 1.25% of MW) cost) (in years) Capital Return Derating Auxiliary O&M (Rs crore/ ptionof (% Maharashtra Karnataka TN Andhra Pradesh Tamil Nadu Gujarat Haryana Uttar Pradesh Punjab Rajasthan Barriers to Development of Renewable Energy | 267 for determination of FiTs. States that are yet to adopt FiTs include Orissa, Bihar, Jammu and Kashmir, Jharkhand, and the Northeastern states. Tables 14.11 to 14.14 summarize the FiT determined by different SERCs for different RETs.

Table 14.12: FiTs for solar power across states

Preferential Solar PV Solar Thermal tariffs for solar energy CoD up to CoD after CoD up to CoD after Dec 2009 Dec 2009 Dec 2009 Dec 2009 Rajasthan Covered under Rs 15.78 /kWh Rs 15.18 /kWh Rs 13.78 /kWh Rs 13.18/kWh* GOI Policy Not covered Rs 15.60 /kWh Rs 15 /kWh Rs 13.60 /kWh Rs 13 /kWh under GOI Policy West Bengal Covered under Rs 4/ Kwh + GBI Rs 4/ kWh + GBI Not determined GOI Policy Not covered under Rs 11 / kWh Rs 10 / kWh Not determined GOI Policy Uttar Pradesh Rs 15/ kWh Rs 15 / kWh# Rs 13 / kWh Rs 13 / kWh# Gujarat Rs 13 / kWh Rs 12 / kWh Rs 10 / kWh Rs 9 / kWh (1–12 years) (1–12 years) (1–12 years) (1–12 years) Rs 3 / kWh Rs 3 / kWh Rs 3 / kWh Rs 3 / kWh (13–25 years) (13–25 years) (13–25 years) (13–25 years) Haryana* Rs 15.96 / kWh Rs 15.16 / kWh Andhra Pradesh *Rs 3.70/ kWh + WPI Not determined Maharashtra** Rs 3/ kWh + GBI Rs 3/ kWh + GBI Rs 3/ kWh + GBI Rs 3/ kWh + GBI Punjab Rs 7/kWh (with base year 2006–07) + annual escalation @ 5% up to 2011–12 Karnataka Rs 3.40/ Kwh + Rs 3.40/ kWh + Rs 3.40/ kWh + Rs 3.40/ kWh + GBI GBI GBI GBI Chhattisgarh*** Rs 15.84/kWh Rs 13.26/kWh Tamil Nadu Rs 3.15/kWh # commissioned before 31.12.2011 * 5 years ** 10 years; commissioned up to 31.3.2010 *** commissioned up to 31.12.2010 268 | Indian Infrastructure: Evolving Perspectives

Table 14.13: FiTs for SHP and assumptions for FiTs across states

Tariff (Rs/kWh) Capital Return Auxiliary O&M Escal- cost on consu- expenses ation in (Rs/MW) equity mption (% of O&M capital expenses cost) Punjab Rs 3.49 (with base year 2006–07) with five annual escalations @ 3% up to 2011–12. Haryana Rs 3.67 for 2007–08, Escalated at 1.5% per annum 10.25 16% 0.5% Maharashtra Tariff of Rs 2.84 in first year, which increases by Rs 0.03/unit every year till the debt repayment is over (10th year) Tariff of Rs 3.11 between years 10–15 after which it again increases annually at a constant rate of Rs 0.03/unit 4.4 16% 0.5% 3% 4% Andhra Tariff from yr 1–10: Pradesh 2.60, 2.52, 2.44, 2.36, 2.27, 2.19, 2.11, 2.03, 1.95, 1.88 3.625 15% 1% 1.5% 4% Karnataka Rs 2.80 without any escalation for the first 10-year period from the year of commercial operation of the plant 3.9 16% 0.5% 1.5% 5% Uttar Tariff determined for Pradesh each of the 20 years of the life of plant for plants commissioned between 2005–06 to 2009–10 Himachal Rs 2.87 for SHP Pradesh projects up to 5 MW; project specific rates for SHP with capacity more than 5MW and up to 25 MW 6.5 14% 0.5% 2% 4% Barriers to Development of Renewable Energy | 269

Table 14.14: FiTs for biomass and bagasse, and assumptions for FiTs across states Biomass tariff (Rs/kWh) Bagasse tariff (Rs/kWh) Rajasthan Project specific tariff for new plants Project-specific tariff for plants Punjab Rs 3.49 (with base year 2006–07) Rs 3.49 (with base year 2006–07) with five annual escalations @ 5% with five annual escalations @ 3% up to 2011–12 up to 2011–12. Haryana Rs 4 for 2007–08, escalated at Rs 3.74 for 2007–08, escalated 2% per annum at 2% per annum Fixed: (1–1.70; 2–1.67; 3–1.63; 4–1.59; 5–1.54; 6–1.49; 7–1.43; 8 –1.37; 9–1.32; 10–1.25; 11–1.18; 12–1.11; 13–1.02); Maharashtra Variable: 2005–06 1.34; 2006–07 1.41; 2007–08 1.48; 2008–09 1.55; Rs 3.05 for the first year of operation, 2009–10 1.63; 2010–11 1.71; escalation of 2% per annum 2011–12 1.80; 2012–13 1.89; 2013–14 1.98; 2014–15 2.08; 2015–16 2.18; 2016–17 2.29; 2017–18 2.41) Gujarat Rs 3.08 for entire project life Rs 3.00 for entire project of 20 years life of 20 years Chhattisgarh Fixed (1–1.78; 2–1.75; 3–1.73; 4–1.68;5–1.63; 6–1.58; 7–1.53; 8–1.48; 9–1.43; 10–1.38); 75:25–Variable: 2005–06 1.20; 2006–07 1.26; 2007–08 1.32; 2008–09 1.39; 2009–10 1.46; 2010–11 1.53; 2011–12 1.61; 2012–13 1.69; 2013–14 1.77; 2014–15 1.86); 75:15–Variable (2007–08 1.27; 2008–09 1.34; 2009–10 1.40; 2010–11 1.47; 2011–12 1.55; 2012–13 1.62; 2013–14 1.71; 2014–15 1.79) Karnataka Rs 2.85 per unit in the first year of Rs 2.80 per unit in the first year of commercial operation of the plant, commercial operation of the plant, annual escalation of 2% per annum annual escalation of 2% per annum for subsequent period of 9 years for subsequent period of 9 years Fuel price of Rs 1000/- per MT escalated at 5% Uttar Variable cost for 2005–06 Rs 1.2821, Variable cost for 2005–06 Rs 1.2821, Pradesh escalation of 6% per annum for each escalation of 6% per annum for each subsequent year. subsequent year. 270 | Indian Infrastructure: Evolving Perspectives

Green power Green power is a concept wherein the utility supplies consumers with RE-based power and charges the consumers the actual cost of this power, the supply of which is aimed at consumers who are environmentally conscious and is priced higher than normal retail tariffs. Within India, only the Andhra Pradesh Electricity Regulatory Commission (APERC) has introduced the Green Power under its FY 2008–09 retail tariff order. APERC has fixed the tentative Green Power tariff at Rs 6.70/kWh for FY 2008–09, and the difference between this tariff and the normal tariff would be used to create a ‘Green Power Fund’. It has further determined that consumers buying green power have the option of obtaining Clean Development Mechanism (CDM) benefits and Renewable Energy Certificates (RECs), whenever these are introduced. Barriers to Development of Renewable Energy | 271

ANNEXURE 4 Box 14.3: Detailed provisions of National Solar Mission Solar power purchase policy Solar RPS NVVN appointed the nodal agency for Renewable Purchase Obligation of utilities purchase & sale of grid-connected solar to be split into solar and non-solar power at 33 kV & above under Phase –I • RPO may start with 0.25% in Phase I and • For each MW of solar power, MOP to increase to 3% by 2022 allocate equivalent MW capacity from • RPO to be fixed after modification of the unallocated quota of NTPC stations National Tariff Policy 2006 • NVVN to bundle solar & thermal • RE Certificates to meet RPO power & sell it at regulated tariff plus facilitation charges

Generation-based incentive Demonstration projects Provision of GBI to 100 MW capacity solar Technology configurations not covered under project sconnected to LT/11 kV grid 1,000 MW capacity • Eligibility: own consumption as well as • Projects to be set up following competitive power fed into the grid. bidding to enable price discovery. • GBI rate: tariff fixed by CERC minus • Maximize indigenous content notional tariff of Rs 5.5 per unit, with 3% • Technology transfer annual escalation

Off-grid opportunities Fiscal/financial incentives Promote solar home lights & other power Increase competitiveness of solar projects & applications to cover 10,000 MW villages & provide enabling environment for solar hamlets. manufacturers • Refinance facility/soft loans up to 5% • Recommendation to MoF for customs and annual interest rate by IREDA. excise duty concessions/exemptions on • 30% subsidy for select applications. specific capital quipment, critical materials, • 90% subsidy for niche applications to components & project imports. special category areas • SEZ-like incentives to manufacturing parks

HRD R&D Build technically qualified manpower of Improve efficiency of existing/new materials international standards & applications & develop cost effective storage • Develop specialized courses at engineering technologies. colleges. • Development of National Centre of • Ministry of Labour to introduce training Excellence & Centres of Excellence to modules/course materials for technicians. undertake & fund R&D. • 100 fellowships a year to support students/ • High-level Research Council to guide groups. overall strategy. • National Centre for PV Research & • Support incubation & innovation through Education at IIT, Mumbai a venture fund 272 | Indian Infrastructure: Evolving Perspectives

NOTES 1. According to the MNRE, the total RE capacity in the country as on 31 March 2009 was 14,485 MW. The difference in magnitude of RE capacity addition as reported by different agencies in the country arises on account of discrepancies in reporting of data and the different time frames when such data is reported. The objective here is to give a fair idea of the extent of RE capacity existing in the country. 2. Current RE capacity is a little less than 15,000 MW and it may be assumed that only another 15,000 MW is doable till 2017. 3. Integrated Energy Policy, 2006. 4. Ministry of Power, Annual Report, 2008–09. 5. Central Electricity Authority, Power Supply at Glance, July 2009. 6. Ministry of Power, www.powermin.nic.in accessed at 2.55 p.m. on October 23, 2009. 7. Shonali Pachauri and Adrian Muller, ‘A Regional Decomposition of Domestic Electricity Consumption in India: 1980–2005’. Presented at the annual IAEE conference at Istanbul on 20 June 2008; available at http://www.iiasa.ac.at/Research/PCC/recent-events/ Pachauri&Mueller_Istanbul_June2008_Final.pdf 8. http://www.renewableenergyworld.com/rea//news/article/2008/04/renewable-energy- jobs-soar-in-germany-52089 9. While this may be sufficient to indicate the importance of the state-level regulatory framework on RE capacity addition, it is important to note that the MNRE-determined tariff which was valid till 2004 may have played a significant role in this capacity addition. 10. Most of the capacity additions in FY 2006–07 and FY 2007–08 would be those where installation started in earlier years. Capacity addition in future years would reflect the impact of TNERC’s order. However, given the time required to commission wind projects, this may broadly reflect the impact of TNERC’s order. 11. CERC’s discussion paper on promotion of co-generation and generation of electricity from renewable sources of energy (May 2008). 12. http://cserc.gov.in/pdf/25-2009-Interim.pdf 13. It is understood that TNEB is addressing this problem now by developing the requisite transmission network. 14. Explanatory Memorandum issued by CERC for Draft Terms and Conditions for Determination of Tariff for Renewable Energy Sources, May 2009. Distribution Reforms in Delhi | 273

POWER DISTRIBUTION 15 REFORMS IN DELHI April 2010

INTRODUCTION The electricity distribution business has been privatized in only two states in the country—Orissa and Delhi. The privatization in Delhi has been successful as far as the efficiency improvements are concerned. Irrespective of improved efficiency, divergent views exist on the success of privatization. Many believe it has not been effective in Delhi. Against this backdrop, this study attempts to assess and evaluate the privatization of the distribution business in Delhi.

IMPERATIVE FOR REFORMS IN THE STATE Traditionally, power supply in Delhi was the responsibility of the Delhi Electricity Supply Undertaking (DESU), which was an integrated utility, with generation, transmission and distribution functions, serving the entire State of Delhi except the New Delhi Municipal Corporation (NDMC) and Military Engineering Services (MES), or Cantonment areas (to which DESU supplied power in bulk). The poor performance of DESU led to its being succeeded by the Delhi Vidyut Board (DVB) in 1997, which was established as a State Electricity Board under the Electricity (Supply) Act 1948. The creation of DVB proved to be merely a change in the legal status of the organisation without any structural changes. The change did not affect the functioning and the work culture of the organization. Its performance continued to deteriorate. The power sector suffered from problems of high technical losses due to poor maintenance of existing infrastructure and very little augmentation, high commercial losses attributable to theft, and high receivables for DVB. During FY 1995–96 to FY 1999–2000, operating losses of DESU/DVB rose from Rs 578 crore to Rs 1100 crore.1 274 | Indian Infrastructure: Evolving Perspectives

Transmission and distribution (T&D) losses during the same period were around 50 per cent. This poor commercial performance of the DVB due to high T&D losses made it incapable of raising the resources required to improve its services and rendered it a drain on the public exchequer.

T&D losses as % of availability Commercial loss (without subsidy) in Rs cr. 60 1200

50 1000

40 800

30 600

20 400

10 200

0 0 1995–96 1996–97 1997–98 1998–99 1999–2000 1995–96 1996–97 1997–98 1998–99 1999–2000 Figure 15.1: T&D losses and commercial losses pre-reforms Source: Annual Report on the Working of State Electricity Boards & Electricity Departments: Planning Commission, Government of India, May 2002

The state also suffered from a severe demand–supply imbalance. Generating stations in Delhi had an installed capacity of 694 MW, but availability was lower at 300–350 MW. Capacity addition remained relatively stagnant, leading to overdependence on purchased power. Not surprisingly, public discontent continued to rise. All these factors required that the Government of the National Capital Territory of Delhi (GNCTD) formulate a strategy to bring about a structural change in the Delhi power sector. It was perceived that the present vertically integrated structure had failed to deliver the desired outcomes and the time was ripe to restructure the sector.

THE REFORM STRATEGY In view of the above considerations, the GNCTD brought out a Strategy Paper on Power Sector Reforms in February 1999 for reforming the power sector in the state. A fast-track reform process was followed that ultimately resulted in the unbundling of DVB into seven companies: one holding company called the Delhi Power Company Limited, two generation companies, viz. Indraprastha Power Generation Company Limited (IPGCL) and Pragati Power Company Limited (PPCL), one transmission company, viz. Delhi Transco Limited (TRANSCO), and three distribution companies (discoms). Subsequently, the GNCTD issued policy directions indicating its intent to disinvest majority shareholding in the discoms to private investors, with the balance 49 per cent remaining with the GNCTD. Distribution Reforms in Delhi | 275

The GNCTD believed that a long-term definitive loss reduction or efficiency gain programme needed to be settled at the very start of the privatization process to reassure private sector investors. It further believed that to attract them it would be appropriate that reduction in losses/efficiency gains be determined through market forces rather than being predetermined unilaterally in the bidding documents. Therefore, loss reduction or efficiency gain to be achieved by the distribution companies in the first five years, viz. 2002–03 to 2006–07, became the basis of privatization or the bidding criteria for the selection of the private entity. The GNCTD was of the view that losses of any kind—technical, non-technical or non-realisation of payments—ultimately amount to loss in revenues. Efficiency gains must embrace all these aspects. Therefore, losses should be measured as the difference between the units input and the units realised (units billed and collected), wherein the units realised will be equal to the product of units billed and the collection efficiency, where collection efficiency is defined as the ratio of actual amount collected to amount billed. The difference between the units input and the units realized are referred to as aggregate technical and commercial (AT&C) losses. The GNCTD, as a matter of policy, decided that the AT&C loss shall be the basis for determination of tariffs and also for computation of incentives for better performance. It asked the Delhi Electricity Regulatory Commission (DERC), vide Policy Directions, to determine the opening level of AT&C losses and the bulk supply tariff (BST) for each discom.

PRIVATIZATION OF DISCOMS GNCTD initiated the bidding process for selection of the private investors for each of the three discoms. Upon opening of bids, it emerged that the loss-level trajectory as submitted by the bidders was not in the range acceptable to the GNCTD. Consequently, the bidders were called for negotiations. After a number of discussions, the bidders and the GNCTD came to an agreement on the accepted year-wise AT&C loss reduction trajectory over the five-year period. These discussions also resulted in other changes to the terms and conditions of privatization. Prime amongst these was the method of computation and treatment of overachievement and underachievement for the years 2002–03 to 2006–07 (see Box 15.1). Box 15.1: Computation and treatment of over/underachievement of target AT&C loss levels i. In the event the actual AT&C loss of a distribution licensee in any year is better (lower) than the level based on the minimum AT&C loss reduction levels stipulated by the government for that year, the distribution licensee shall be 276 | Indian Infrastructure: Evolving Perspectives

allowed to retain 50% of the additional revenue resulting from such better performance. The balance 50% of additional revenue from such better performance shall be counted for the purpose of tariff fixation. ii. In the event the actual AT&C loss of a distribution licensee in any year is worse (higher) than the level based on the AT&C loss reduction levels indicated in the Accepted Bid for that year, the entire shortfall in revenue on account of the same shall be borne by the distribution licensee. iii. In the event the actual AT&C loss of a distribution licensee in any year is worse (higher) than the level based on the minimum AT&C loss reduction levels stipulated by the Government for that year, but better (lower) than the level based on AT&C loss reduction levels indicated in the Accepted Bid for that year, the entire additional revenue from such better performance shall be counted for the purpose of tariff fixation. Provided further that for paras (i), (ii) and (iii) above, for every year, while determining such additional revenue or shortfall in revenue, the cumulative net effect of revenue till the end of the relevant year shall be taken, in regard to overachievement/underachievement and appropriate adjustments shall be made for the net effect.

After the negotiations and the incorporation of the resulting changes, revised bids were submitted by the bidders. Consequent to this round of bidding, GNCTD appointed TATA Power as the distribution utility for the North & North-West circle and BSES for two circles—Central & East and South & West. The opening loss levels as well as the loss reduction trajectory that were accepted by the investors for each discom and the minimum AT&C loss reduction level as indicated in the accepted bids are as shown in Table 15.1. The three distribution companies were privatised with effect from 1 July 2002.

Table 15.1: Accepted bid loss reduction trajectory 2002–03 2003–04 2004–05 2005–06 2006–07 Total BSES Rajdhani 0.55 1.55 3.30 6.00 5.60 17.00 BSES Yamuna 0.75 1.75 4.00 5.65 5.10 17.25 NDPL 0.50 2.25 4.50 5.50 4.25 17.00 Minimum bid loss reduction trajectory BSES Rajdhani 1.50 5.00 5.00 5.00 4.25 20.75 BSES Yamuna 1.25 5.00 4.50 4.50 4.00 19.25 NDPL 1.50 5.00 4.50 4.25 4.00 19.25 Source: Jagdish Sagar: Power Sector Reforms in Delhi: The Experience So Far Distribution Reforms in Delhi | 277

INITIATIVES TAKEN POST-REFORMS The discoms undertook a number of initiatives to bring about changes in the distribution business and achieve the reduced loss-level targets set for privatisation.

Initiatives taken by NDPL

Automation initiatives and GIS The NDPL embarked on automating all its 66 kV and 33 kV grids, and in line with this has already automated 34 grids with a view to operating all equipment from the central command centre. This has expedited the resolution time for faults. The entire electrical network has been mapped through GIS to enable quicker fault location and speedy redressal; the outage management system is being upgraded for automation on a GIS platform.

Complaint management system The NDPL has a unique SMS-based fault management system using GSM which ensures that the ‘No supply’ complaints lodged by a consumer get addressed quickly and consumer feedback is also institutionalized as part of the process. In July 2002, the NDPL had very poor consumer care facilities. Now, each of the 12 districts has an online consumer care centre, each handled by customer care executives under the supervision of customer relation officers and customer service officers.

Online connection management by consumer The NDPL uploaded the billing details of all its consumers on its website, www.ndpl.com. Consumers can view their bill, know the consumption pattern and even print duplicate bills and make online bill payments.

Doorstep delivery of new connections To ensure hassle-free new connections, an NDPL representative visits the consumer’s premises and completes all formalities there itself.

Privileged consumer scheme As incentive for prompt payment, NDPL has institutionalized a privileged consumer scheme through which discounts are offered. The NDPL has also institutionalized a structured approach towards consumer relationship management as it organizes regular meetings with consumer representative groups, such as RWAs and IWAs, on the first Saturday of every month in each district. 278 | Indian Infrastructure: Evolving Perspectives

Automated bill payment kiosks for consumer convenience The NDPL has introduced Automated Bill Payment Kiosks, a first in Delhi and the National Capital Region (NCR). These unique ATM-like kiosks accept both cash and cheque payment towards electricity bills and even issue a receipt to the consumer. They are operational 365 days a year from 8 a.m. to 8 p.m.

Initiatives taken by BSES Consumer related • Setting up customer help desks (CHDs) in all the 33 divisions for online registration of consumer complaints, and commercial call centres to address complaints • In-house developed CRM software (CAS) which provides the facility of single-window resolution of complaints • Setting up of a 24-hour control room at the CM’s office to provide immediate information on faults/breakdowns • Bifurcation of the existing 21 districts into 33 divisions • Centralised helpline numbers have been created for all queries related to power supply/meter/billing/anti-corruption/vigilance/anti-power theft/enforcement • Starting of the ‘Bill Amendment Module (BAM)’, a billing software module through which customer complaints are addressed on the spot • Facility of raising one composite bill for bulk consumers like MCD and DJB for convenient payments • Meter reading is now being done by meter reading instruments (MRIs). • Location of a BSES office in a 3-km radius from each point. Consumers may go to any of the 143 BSES offices. • The following facilities for key consumers have been created: • Creation of a special cell for customers having a load of over 45 KW • Single-window ease and preparation of composite bill for bulk consumers like DJB and MCD for convenient payments • Bill dispatch by e-mails and special couriers Distribution Reforms in Delhi | 279

Supply and streetlighting related • Approximately hundred breakdown vans; cable restoration on a 24 by 7 basis • Online connectivity between technical call centre, system and circle control • 25 Genset vans and mobile transformers made available round the clock • A special drive, ‘Roshini’, which provided uninterrupted illumination during the entire festive season

Metering related • Mass meter replacement drive was undertaken for all large industrial power (LIP) consumers. Electronic meters have also been introduced for domestic consumers. • Faulty meters are being replaced on a priority basis with tamper-proof electronic meters. A Special Meter Testing Drive was undertaken. Only 0.01 per cent of electronic meters were found defective. • A month-long voluntary disclosure scheme (VDS) was undertaken.

Payment related • Point-of-sales (POS) machines installed at the cash collection counters with bar code scanners for speedy service; tie-ups with collection agencies, like Easy Bill and Skypak, for cheque payments and with bill desk and bill junction for online payments • 120 cash collection centres, 150 Skypak drop boxes and 1050 Easy Bill outlets across BSES; 150 drop boxes installed on RWA premises

OUTCOME OF PRIVATIZATION AND REFORMS The overall impact of reforms and initiatives taken has started producing favourable results. Various parameters indicating the overall health of the power distribution sector in the state are discussed below: • Aggregate technical and commercial losses AT&C loss reduction was the parameter used for privatisation of distribution companies in Delhi. Opening AT&C levels were given for the three distribution companies, and annual targets for reduction were set. Post-privatisation, a number of initiatives were taken by distribution companies to reduce AT&C losses. The AT&C losses for Delhi have declined from 56 per cent in 2002–03 to 38 per cent in 2007–08.2 The AT&C losses reported by various discoms are as shown in Figure 15.2. The NDPL has reported the lowest AT&C losses for the year 2008–09. 280 | Indian Infrastructure: Evolving Perspectives

70

60

50

40

30

20

10

0 2002–03* 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 BSES Rajdhani BSES Yamuna NDPL Figure 15.2: AT&C losses (%) Source: Forum of regulators (FOR)3 • Financial viability a. Profitability The financial position of discoms has improved post-privatisation. All the discoms started reporting profits from 2004–05. While NDPL has shown a consistent improvement in profits generated, BSES slipped into losses in 2007–08. Table 15.2: Profits (in Rs crore) 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 BSES Rajdhani -57 -32 60 89 27 -449 BSES Yamuna -101 -55 7 46 48 -55 NDPL 22 29 57 113 186 282 2 Source: PFC Report Both NDPL and BSES registered a significant increase in the power purchase cost in 2007–08 over 2006–07. The power purchase cost per unit of energy input was higher for BSES as compared to NDPL. BSES discoms also reported increase in their interest costs for FY 2007 and FY 2008.2

b. Average cost of supply (ACS) and average revenue realized (ARR) NDPL has seen an increasing difference between ARR and ACS from 10 paise/kWh in 2002–03 to 50 paise/kWh in 2007–08.2 Distribution Reforms in Delhi | 281

BSES discoms registered better revenue realized than cost of supply for FY 2005 and FY 2006. However, in FY 2007 and FY 2008 their cost of supply was higher than revenue realized. This deterioration was reflected in their financials as they slipped into losses for FY 2008. • Quality of Supply (QoS) Quality of supply in Delhi has improved post reforms. NDPL registered best figures for quality of supply parameters in FY 2007 (namely, outage duration, number of outages and average duration of outage). Data is available from FY 2005 to FY 2007, and BSES had much worse conditions than NDPL at the start of this period. 4.0 4.0 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 BSES Rajdhani ACS BSES Yamuna ACS NDPL ACS BSES Rajdhani ARR BSES Yamuna ARR NDPL ARR Figure 15.3: ARR & ACS (Rs per unit) 2 Source: PFC Report Deficit situation The deficit situation in Delhi has improved in terms of both peak and energy deficit. Peak deficit has reduced from 9.2 per cent in 2002–03 to 0.0 per cent in 2008–09. Energy deficit has reduced from 1.9 per cent in 2002–03 to 0.6 per cent in 2008–09. Table 15.3: QoS parameters Outage duration per feeder (hh:mm) 2004–05 2005–06 2006–07 NDPL 49:33 13:34 3:54 BYPL 105:14 77:40 37:05 BRPL 98:07 82:31 33:29 282 | Indian Infrastructure: Evolving Perspectives

Table 15.3: QoS parameters (contd...) No. of outages per feeder 2004–05 2005–06 2006–07 NDPL 26 10 4 BYPL 68 77 38 BRPL 60 76 34 Average duration of an outage (hh:mm) 2004–05 2005–06 2006–07 NDPL 1:55 1:22 0:54 BYPL 1:33 1:01 0:58 BRPL 1:38 1:05 0:59 Source: http://www.cea.nic.in Table 15.4: Peak and energy deficit (%) Peak Energy deficit deficit 2002–03 9.2 1.9 2003–04 3.0 1.4 2004–05 1.9 1.0 2005–06 3.3 1.5 2006–07 6.6 1.7 2007–08 1.1 0.6 2008–09 0.0 0.6 Source: http://www.cea.nic.in Table 15.5: Loan to TRANSCO (in Rs cr.) FY 2003 FY 2004 FY 2005 FY 2005 FY 2006 Total 1364 1260 690 138 0 3450

CONCLUSION Distribution reforms in Delhi and privatization of discoms have led to positive results. This is being reflected in reduced AT&C losses, low deficit situation, improving quality of supply parameters and no subsidy to discoms. The only subsidization of discoms post-privatisation was a loan of Rs 3450 crore from GNCTD to TRANSCO during 2003 and 2006 (Table 15.5). This loan was given to bridge the gap between Distribution Reforms in Delhi | 283 their revenue requirement and revenue received from discoms through bulk supply tariff. However, compared to annual losses of the order of Rs 1000 crore that existed before reforms, the financial burden has reduced significantly. NDPL has reported minimum losses and best figures for QoS parameters amongst the discoms in Delhi. It has also been the only discom to register profits for all the years from FY 2003 to FY 2008. All the discoms have made capital investments to achieve the target reduction in loss levels (Annexure 2). Capital expenditure per unit of energy input is higher for NDPL. Actual AT&C loss reduction by discoms along with government-specified minimum bid and accepted bid is shown in Annexure 3. For the period FY 2003 to FY 2007, total loss reduction by NDPL was more than BSES discoms. Also the total reduction was more than the government- specified minimum bid. While the distribution reforms have given positive results in Delhi, there are a few areas of concern. For the first five years (FY 2003 to FY 2007), loss level targets and bulk power purchase tariffs were specified. Going forward, discoms will be vulnerable to fluctuations in bulk purchase tariffs. Further, AT&C loss reduction will be harder to achieve as they have already reduced considerably, and collection efficiencies will not remain more than 100 per cent for a long period. In 2007–08, both the BSES discoms registered financial losses driven by high power purchase cost and interest cost (Annexure 1). Only NDPL registered profits after tax of Rs 282 crore, that too on the back of Rs 225 crore recoverable by truing up of earlier year revenues. This income resulted from NDPL winning an appeal to use 6.69 per cent depreciation for FY 2003, FY 2004 and FY 2005 as compared to the 3.75 per cent rate specified by DERC. Hence, the financial viability of discoms needs to be examined under high power purchase cost scenarios. 284 | Indian Infrastructure: Evolving Perspectives

ANNEXURE 1 Table 15.6: Expenses break-up of discoms (in Rs crore) BSES Rajdhani 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 846 1,244 1,654 1,877 2,103 2,899 Employees 97 119 113 122 147 164 O&M 76 80 92 69 88 70 Interest 1 2 6 32 153 205 Depreciation 82 115 125 116 139 155 Admin & general 11 32 41 60 63 67 Other 0 133 0 -23 77 92 Total 1,113 1,724 2,031 2,253 2,770 3,652 BSES Yamuna 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 479 638 798 922 993 1,973 Employees 80 100 92 94 111 129 O&M 43 47 65 52 47 49 Interest 2 8 13 29 76 132 Depreciation 20 32 42 49 57 72 Admin & general 7 21 27 35 39 45 Other 0 95 0 -31 63 77 Total 632 941 1,037 1,150 1,386 2,476 NDPL 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 599 868 1,105 1,204 1,309 1,882 Employees 82 103 131 146 155 152 O&M 20 91 59 53 51 57 Interest 0 4 20 23 57 75 Depreciation 66 87 113 110 129 155 Admin & general 13 19 24 29 30 33 Other 45 40 27 2 -3 -202 Total 824 1,213 1,478 1,566 1,727 2,152 Source: Report on the Performance of the State Power Utilities for the Years 2004–05 to 2007–08, Power Finance Corporation Limited Distribution Reforms in Delhi | 285

Expenses break-up (as % of total expenses) BSES Rajdhani 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 76.0 72.2 81.4 83.3 75.9 79.4 Employees 8.7 6.9 5.6 5.4 5.3 4.5 O&M 6.8 4.6 4.5 3.1 3.2 1.9 Interest 0.1 0.1 0.3 1.4 5.5 5.6 Depreciation 7.4 6.7 6.2 5.1 5.0 4.2 Admin & general 1.0 1.9 2.0 2.7 2.3 1.8 Other 0.0 7.7 0.0 -1.0 2.8 2.5 Total 100.0 100.0 100.0 100.0 100.0 100.0 BSES Yamuna 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 75.8 67.8 77.0 80.2 71.6 79.7 Employees 12.7 10.6 8.9 8.2 8.0 5.2 O&M 6.8 5.0 6.3 4.5 3.4 2.0 Interest 0.3 0.9 1.3 2.5 5.5 5.3 Depreciation 3.2 3.4 4.1 4.3 4.1 2.9 Admin & general 1.1 2.2 2.6 3.0 2.8 1.8 Other 0.0 10.1 0.0 -2.7 4.5 3.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 NDPL 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 72.7 71.6 74.8 76.9 75.8 87.5 Employees 10.0 8.5 8.9 9.3 9.0 7.1 O&M 2.4 7.5 4.0 3.4 3.0 2.6 Interest 0.0 0.3 1.4 1.5 3.3 3.5 Depreciation 8.0 7.2 7.6 7.0 7.5 7.2 Admin & general 1.6 1.6 1.6 1.9 1.7 1.5 Other 5.5 3.3 1.8 0.1 -0.2 -9.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 Source: Report on the Performance of the State Power Utilities for the Years 2004–05 to 2007–08, Power Finance Corporation Limited 286 | Indian Infrastructure: Evolving Perspectives

Expenses break-up (in Rs/unit of energy input) BSES Rajdhani 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 1.51 1.54 1.97 2.17 2.31 3.13 Employees 0.17 0.15 0.13 0.14 0.16 0.18 O&M 0.14 0.10 0.11 0.08 0.10 0.08 Interest 0.00 0.00 0.01 0.04 0.17 0.22 Depreciation 0.15 0.14 0.15 0.13 0.15 0.17 Admin & general 0.02 0.04 0.05 0.07 0.07 0.07 Other 0.00 0.16 0.00 -0.03 0.08 0.10 Total 1.99 2.13 2.42 2.61 3.04 3.94 BSES Yamuna 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 1.32 1.23 1.49 1.71 1.87 3.66 Employees 0.22 0.19 0.17 0.17 0.21 0.24 O&M 0.12 0.09 0.12 0.10 0.09 0.09 Interest 0.01 0.02 0.02 0.05 0.14 0.25 Depreciation 0.06 0.06 0.08 0.09 0.11 0.13 Admin & general 0.02 0.04 0.05 0.06 0.07 0.08 Other 0.00 0.18 0.00 -0.06 0.12 0.14 Total 1.74 1.82 1.94 2.13 2.62 4.60 NDPL 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Power purchased 1.52 1.56 2.00 2.11 2.19 3.00 Employees 0.21 0.19 0.24 0.26 0.26 0.24 O&M 0.05 0.16 0.11 0.09 0.09 0.09 Interest 0.00 0.01 0.04 0.04 0.10 0.12 Depreciation 0.17 0.16 0.20 0.19 0.22 0.25 Admin & general 0.03 0.03 0.04 0.05 0.05 0.05 Other 0.11 0.07 0.05 0.00 -0.01 -0.32 Total 2.10 2.18 2.67 2.75 2.89 3.43 Source: Report on the Performance of the State Power Utilities for the Years 2004–05 to 2007–08, Power Finance Corporation Limited Distribution Reforms in Delhi | 287

ANNEXURE 2 Table 15.7: Capital expenditure by discoms Capex (in Rs crore) 2003 2004 2005 2006 2007 2008* Total BSES Rajdhani 72.00 115.00 538.00 711.00 399.00 239.00 2464.00 BSES Yamuna 58.00 85.00 416.00 357.00 283.00 164.00 1663.00 NDPL 49.00 299.00 338.00 431.00 271.00 248.00 1899.00 *Provisional data Capex (in Rs/unit of energy input) 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 Total BSES Rajdhani 0.13 0.14 0.64 0.82 0.44 0.26 2.43 BSES Yamuna 0.16 0.16 0.78 0.66 0.53 0.30 2.60 NDPL 0.12 0.54 0.61 0.76 0.45 0.40 2.88 Source: http://www.derc.gov.in/ordersPetitions/orders/Misc/2009Order%20on% 20Physical%20Verification%20of%20Assets.pdf

ANNEXURE 3 Table 15.8: AT&C loss reduction by discoms BSES Rajdhani loss reduction trajectory 2002–03 2003–04 2004–05 2005–06 2006–07 Total Govt. specified minimum bid 1.50 5.00 5.00 5.00 4.25 20.75 Accepted bid 0.55 1.55 3.30 6.00 5.60 17.00 Actual 0.70 2.35 4.41 5.11 5.62 18.19 BSES Yamuna loss reduction trajectory 2002–03 2003–04 2004–05 2005–06 2006–07 Total Govt. specified minimum bid 1.25 5.00 4.50 4.50 4.00 19.25 Accepted bid 0.75 1.75 4.00 5.65 5.10 17.25 Actual -4.69 7.60 4.16 6.26 4.84 18.17 NDPL loss reduction trajectory 2002–03 2003–04 2004–05 2005–06 2006–07 Total Govt. specified minimum bid 1.50 5.00 4.50 4.25 4.00 19.25 Accepted bid 0.50 2.25 4.50 5.50 4.25 17.00 Actual -1.02 4.26 11.05 7.29 2.78 24.36 288 | Indian Infrastructure: Evolving Perspectives

ANNEXURE 4 Table 15.9: Sales & revenue mix NDPL Sales mix: Sale in MkWh to total sales (%) 2003–04 2004–05 2005–06 2006–07 2007–08 Domestic 47.1% 42.1% 41.5% 41.9% 37.5% Commercial 20.1% 18.7% 18.7% 19.2% 17.3% Agricultural 0.9% 0.6% 0.5% 0.2% 0.2% Industrial 26.8% 32.4% 33.9% 33.0% 32.2% Others 5.1% 6.1% 5.4% 5.7% 12.8% Revenue mix (%) 2003–04 2004–05 2005–06 2006–07 2007–08 Domestic 31.8% 28.8% 28.4% 27.5% 29.0% Commercial 29.3% 25.7% 25.6% 27.4% 24.9% Agricultural 0.3% 0.2% 0.2% 0.1% 0.1% Industrial 33.4% 40.0% 43.2% 40.3% 39.0% Others 5.1% 5.3% 2.7% 4.7% 7.1%

REFERENCES 1. Planning Commission (Power & Energy Division) Government of India. 2002. Annual Report (2001-02) on the Working of State Electricity Boards & Electricity Departments. 2. Power Finance Corporation Limited. Report on the Performance of the State Power Utilities for the Years 2004–05 to 2007–08. 3. http://www.forumofregulators.gov.in/Data/policy_Imp/AT% 20&% 20C% 20LOSS% 20DATA% 20-% 20STATE% 20&% 20UTILITES% 20WISE.pdf 4. Sagar, Jagdish. Power Sector Reforms in Delhi: The Experience So Far. 5. http://delhigovt.nic.in/power.asp 6. Central Electricity Authority, Planning Wing. 2009. Power scenario at a glance. 7. http://www.ndpl.com/Display Content.aspx? RefTypes=3 & RefIds = 149 & page = Pioneering Initiatives 8. http://www.bsesdelhi.com/Aboutus/in_undertaken.asp 9. http://www.bsesdelhi.com/Aboutus/bsesataglance.asp Distribution Reforms in Delhi | 289

10. http://www.cea.nic.in/god/dpd/RELIABILITY_INDICES_MONTHLY.pdf 11. http://www.cea.nic.in/power_sec_reports/executive_summary/2009_04/25-26.pdf 12. http://www.derc.gov.in/ordersPetitions/orders/Misc/2009/Order%20on% 20Physical% 20Verification%20of%20Assets.pdf 13. Tariff orders of discoms; from DERC website (http://www.derc.gov.in) 290 | Indian Infrastructure: Evolving Perspectives

POWER DISTRIBUTION: Being Driven to Insolvency 16 by a Governance Crisis April 2011

INTRODUCTION The financial health of distribution utilities in the country is deteriorating at an alarming level once again. A handful of states account for the lion’s share of increased financial losses. There is a tendency to attribute the rising financial losses to increasing Aggregate Technical and Commercial (AT&C) losses.1 But AT&C losses have reduced from 33 per cent in 2005–06 to 28 per cent in 2008–09. Clearly, the problem lies elsewhere. Analysis indicates that the sector is undergoing a severe governance crisis. Inefficiency of utilities is another, albeit lesser, problem. Losses of distribution utilities are estimated to range between Rs 93,000 and Rs 150,000 crore in 2012–13. Given that state governments are not in a position to support the sector in the long run, the impact of the rising losses on the financial system of the country is a matter of grave concern. The financial system would get affected not only because of its direct exposure to the distribution sector but also because of debt servicing by generation project developers. Therefore, there is an urgent need to restore good governance in the sector and prevent the insolvency of the distribution business from jeopardizing not only future capacity addition but also the health of the financial system.

2 Financial losses of distribution utilities are mounting once again The increase in losses of distribution utilities from Rs 3,000 crores to Rs 30,000 crores in the ten years between 1991–92 and 2001–02 had forced the attention of policy makers on reforming the distribution sector with the objective of reducing transmission and distribution (T&D) losses and improving the commercial viability of the distribution utilities (herein after referred to as utilities). The many reforms initiated in the early 2000s helped reduce and contain financial losses between 2002–03 and 2005–06. However, between 2005–06 and 2008–09, losses (without subsidy) Power Distribution: Being Driven to Insolvency | 291 have more than doubled (see Figure 16.1) and are expected to reach Rs 68,600 crore at the end of FY 2010–11.3 A bigger concern is the rising level of cash losses4 before subsidy received. These losses have trebled from Rs 14206 crore to Rs 44059 crore between 2005–06 and 2008–09 (see Figure 16.2). Further analysis indicates that Rajasthan, Tamil Nadu (TN) and Andhra Pradesh (AP) have shown the maximum increase in losses between 2005–06 and 2008–09. Table 16.1 provides a snapshot of states that have exhibited significant worsening of losses. On the other hand, Chhattisgarh, Gujarat, Himachal Pradesh (HP), Kerala, West Bengal (WB), and most of the north-eastern states have reduced their financial losses or are earning profits (see Table 16.2). 60000 Losses on subsidy booked basis 50000 Losses on subsidy received basis Losses without subsidy 40000

30000 50585

20000

34237

27893

23371

19579 10000 20914

17555

0 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.1: Losses without subsidy for distribution utilities have risen sharply in 2008–09 (Rs crore) Source: Power Finance Corporation Ltd 44059 45000 40000 35000 Cash losses - on subsidy received basis Cash losses - before subsidy received 30000 27341 25671 25000 21119 20000 14206 15000 12790 12022 10352 10869 10000 8283 5000 3268 1213 229 350 0 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.2: Cash losses before subsidy received for distribution utilities have trebled between 2005–06 and 2008–09 (Rs crore) Source: Power Finance Corporation Ltd, IDFC analysis 292 | Indian Infrastructure: Evolving Perspectives * For MSEDCL 2005–06 2008–09 Increase 2005–06 2008–09 Increase 2005–06 2008–09 Increase Losses on subsidy received basis Losses without subsidy Cash losses before subsidy received Table 16.1: States exhibiting increases in losses from distribution business (Rs crore) Power Finance Corporation Ltd, IDFC analysis

Sign indicates surplus;

UttarakhandMaharashtra*JharkhandBihar 215Uttar Pradesh 594Punjab 469 902 588Karnataka 3388 254Haryana 1240 308Madhya Pradesh 916 4239 215Rajasthan 652 594 1005 -13Tamil Nadu -91 851 469 593 951 902Andhra Pradesh 323 4302 640 1667 89 254 2784 1320 308 5821 1483 1273 1758 616 1329 653 2191 -41 369 121 1519 1136 1725 1160 1423 6604 7382 -76 953 3321 1612 3118 379 877 452 3242 5988 3769 6053 1653 3728 3362 4121 1982 1629 1819 1172 2508 5249 258 1156 2775 1442 1729 2509 7655 8964 840 1669 1480 7936 295 839 633 6026 1464 6456 2913 6494 2564 513 3270 3946 1321 1326 2073 1725 954 2637 7269 2482 8194 7157 5948 6868 6203 Sources: Note: Power Distribution: Being Driven to Insolvency | 293 -402 -774 -372 -402 -774 -372 -455 -1015 -560 2005–06 2008–09 Increase 2005–06 2008–09 Increase 2005–06 2008–09 Increase Losses on subsidy received basis Losses without subsidy Cash losses before subsidy received Table 16.2: States exhibiting profits or decrease in losses from distribution (Rs crore) Power Finance Corporation Ltd, IDFC analysis : Sign indicates surplus or positive movement. Chhattisgarh GujaratHimachal PradeshKeralaWest Bengal -20 -56 -32 -15 257 -12 43 -39 41 56 -217 1124 -296 -260 -32 1085 257 43 -88 -39 -39 -217 -296 3 -260 872 -129 -140 845 -349 -297 -132 -652 -27 -157 -303 Sources Note: 294 | Indian Infrastructure: Evolving Perspectives

AT&C losses have reduced significantly It is commonly believed that high levels of AT&C losses are the main reason for the high financial losses of utilities. There is no doubt that AT&C losses are high in absolute terms; with 23 out of 52 utilities recording AT&C losses over 30 per cent in 2008–09 (see Table 16.2). This is resulting in revenue losses to the utilities. But overall AT&C losses have reduced over the years and now stand at less than 30 per cent (see Figure 16.3). A closer look at AT&C losses since 2005–06 indicates that the majority of utilities have shown considerable reduction in AT&C losses, though the pace of reduction may still be slow (see Table 16.3) and need to be accelerated. Amongst the states that have exhibited high increases in financial losses, AP and TN have low AT&C loss levels. Utilities in Rajasthan and some utilities in Karnataka, Haryana and Madhya Pradesh (MP) exhibit high loss levels, but they have been reducing losses.

Table 16.3: Most utilities have shown considerable reduction in AT&C losses between 2005–06 & 2008–09 AT&C loss Increase in Percentage points reduction in AT&C losses levels in losses 2008–09 Up to 5% 5–10% 10–15% 15–20% > 20%

< 20% Reliance Punjab, NDPL, Uttar Bangalore Upper Mumbai, All AP utilities, Madhya Gujarat, Assam Torrent Tamil Nadu, Gujarat Lower Assam Ahmedabad, Dakshin Torrent Gujarat, Surat BEST Mumbai, Mangalore

20–25% Kerala Himachal BRPL BYPL Pradesh, MSEDCL

25–30% WBSEDCL, Ajmer, Jodhpur, Dakshin Haryana, Jaipur, Chamund- Paschim UP eshwari

30–40% Chhattisgarh Northern Orissa, Western Orissa, Meghalaya Dakshin UP Bihar Uttar Haryana, Hubli, Madhya UP, Paschim Gujarat, Uttarakhand, Madhya MP, Paschim MP Central Assam, Tripura

> 40% Southern Central Orissa, Arunachal Gulbarga Jharkhand Orissa, Purv MP, Pradesh Manipur, Poorv UP, Mizoram, Nagaland Sikkim

Sources: Forum of Regulators, Power Finance Corporation Ltd, IDFC analysis Power Distribution: Being Driven to Insolvency | 295

Note: * AT&C loss levels for some utilities were fairly low in 2005–06 because they had reduced their losses in earlier years. The extent of reduction indicated here may not reflect this. Further, for some utilities AT&C losses rose between 2005–06 and 2008–09. In some cases, this was not a result of poor performance but a result of better availability of data on the exact level of AT&C losses prevailing in their service area. Reliance Mumbai, Torrent Ahmedabad and Torrent Surat have traditionally had low levels of losses. Some fluctuation in losses is bound to occur and this fluctuation does not indicate poor performance in AT&C loss reduction. For detailed names of utilities, refer to Annexure.

40 38.77 37.75

36 34.33 33.02 % 32 30.62 29.58 28.44

28 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.3: All-India AT&C losses are below 30% Source: Power Finance Corporation Ltd

However, AT&C loss levels as reported here may not reflect the true situation5 since the information base of distribution utilities still remains poor and sales to the agriculture sector continues to remain unmetered (see Table 16.4) though agricultural sales form a sizeable portion of sales in many states (see Table 16.5). Agriculture accounts for 23–24 per cent of electricity consumption in the country.6 Utilities such as Punjab, Uttar Haryana, Uttar Gujarat and Paschim MP, where agriculture accounts for over 25 per cent of total electricity consumption, have poor levels of metering as far as agricultural consumers are concerned. Even in TN where agriculture accounts for 22 per cent of total electricity sales, metering is abysmally low. It is also interesting to note that utilities that have very low levels of agriculture consumption have higher levels of metering for agriculture consumers. Examples to this end can be seen in Uttarakhand and West Bengal, and in utilities in Orissa and Gujarat. 296 | Indian Infrastructure: Evolving Perspectives

Table 16.4: Agricultural consumption continues to remain unmetered— status in select states

State Distribution utility Percentage of agricultural consumers that are metered Haryana Dakshin Haryana 61% Uttar Haryana 32% Gujarat Uttar Gujarat 27% Paschim Gujarat 35% Madhya Gujarat 56% Dakshin Gujarat 43% Mysore 24% Madhya Pradesh Purv MP 30% Madhya MP 37% Paschim MP 2.3% Maharashtra MSEDCL 44% Orissa Central Orissa 9% Northern Orissa 21% Western Orissa 95% Southern Orissa 96% Punjab PSEB 9.7% Tamil Nadu TNEB 3%* Uttarakhand Uttarakhand 87% West Bengal WBSEDCL 61.5%

Source: Forum of Regulators Note: *As mentioned in the tariff order issued by the Tamil Nadu Electricity Regulatory Commission on July 31, 2010, meters are provided to 3% of the service connections in each distribution circle of TNEB and consumption is recorded on a sample basis. Power Distribution: Being Driven to Insolvency | 297

Table 16.5: Level of agricultural consumption in select states in 2008–09 Agricultural sales as Utilities/states a share of total sales < 10% WBSEDCL, Orissa utilities, Kerala, Dakshin Gujarat, Himachal Pradesh, Uttarakhand, Jharkhand 10–25% Bihar, UP utilities, Eastern AP, Tamil Nadu, Madhya Gujarat, Purv MP, Chhattisgarh 25–40% Dakshin Haryana, Punjab, Ajmer, Jaipur, Central & Southern AP, Bangalore, Mangalore, Paschim Gujarat, Madhya and Paschim MP Over 40% Uttar Haryana, Jodhpur, Northern AP, Gulbarga, Hubli, Chamundeshwari, Uttar Gujarat Sources: Power Finance Corporation Ltd, IDFC analysis Poor levels of metering of agricultural consumption may also impact the financial health of the utilities on account of under provision of subsidy by state governments. In the absence of metering of agricultural consumption, it is not possible to determine the accurate consumption in each service connection. Since the subsidy provided by state governments towards agricultural consumption is based on the estimation of this consumption by utilities, poor metering may lead to a situation where the subsidy is under estimated.7 Reduction in AT&C losses requires investments in augmentation and modernization of the distribution infrastructure for reducing technical losses, energy audits and improved governance for controlling electricity theft. The momentum of loss reduction made possible by these measures has slowed down. Several Electricity Regulatory Commissions (ERCs) have made observations to this end in the orders issued by them. For example, the ERC in Haryana has lamented the absence of remedial measures to address high feeder losses in some districts in the states. The ERC has also observed that theft cases have been rising and that the utilities do not fully implement the capital expenditure plan drafted by them. The ERC in MP has also pointed out the abnormally low progress in capital expenditure by the utilities. Actual investments vis-à-vis investment plan prepared by utilities in MP varied between 26 per cent to 54 per cent during 2007–08 and 2008–09.8 Progress on metering of distribution transformers (DTs) remains poor across states (see Table 16.6). Though energy audits have been initiated in many states, these audits can be carried out only in feeders where all DTs are metered. Thus, in the absence of 100 per cent DT metering, energy audit would have limited benefits. Moreover, in several states energy audit is being conducted only on a sample basis. 298 | Indian Infrastructure: Evolving Perspectives

These initiatives which had gathered impetus under the Accelerated Power Development and Reforms Program (APDRP) that aimed to improve the financial viability of utilities and reduce AT&C losses to around 10 per cent has also slowed down. Given the shortcomings of APDRP and the problems in its implementation of reasons,9 the GOI restructured the programme into the Restructured–Accelerated Power Development and Reforms Programme (RAPDRP) in December 2008. The objective of RAPDRP is to reduce AT&C losses to less than 15 per cent over five years, by automating and integrating various utility processes like asset management; maintenance management; metering, billing and collection; energy audit; and GIS-based consumer indexing. However, achieving the targets would also take time given that the establishment of reliable and automated systems for sustained collection of accurate base line data, and the adoption of information technology in the areas of energy accounting takes time. Further, the implementation of the programme has already seen several ups and downs.10 Unless the strategy for reduction of AT&C losses is redefined and initiatives stepped up, AT&C loss reduction would continue to be slow and the financial health of the distribution would take time to improve.

Table 16.6: Status of implementation of select distribution reform initiatives as of April 2010 Extent of metering Extent of metering Initiation of energy for 11 kV feeders for distribution audits for segregation transformers of technical & commercial losses Haryana Uttar Haryana: Uttar Haryana: Uttar Haryana: 99.74%, 2987 nos. Initiated in Dakshin Haryana: Dakshin Haryana: Gurgaon and 100% 20.88% Faridabad, Dakshin Haryana: for inter utility interface points of feeder Punjab 100% 5% Yes Rajasthan 91% NA No Uttar Pradesh 100% In part in urban areas Yes Madhya Pradesh 100% Most urban DTRs Yes, at division level are metered; for rural DTRs: work in progress Power Distribution: Being Driven to Insolvency | 299

Table 16.6: Status of implementation of select distribution reform initiatives as of April 2010 (contd...) Extent of metering Extent of metering Initiation of energy for 11 kV feeders for distribution audits for segregation transformers of technical & commercial losses Maharashtra* 99.93% NA Yes, division-wise Gujarat 100% Uttar Gujarat: Not by state owned 52.3%, utilities Paschim Gujarat: 17.31%, Madhya Gujarat: 54.54%, Dakshin Gujarat: 56.18% Andhra Pradesh Being done on All have not been Yes sample basis provided with meters Karnataka 100% for Bangalore Bangalore: Yes and Chamun- 39.5%, deshwari, NA for NA for other other utilities utilities Kerala 100% 31.92% In 46 out of 65 divisions Tamil Nadu 100% 49.68% No West Bengal# 92% 26% Yes Source: Central Electricity Authority Notes: * For MSEDCL; # For WBSEDCL; NA – Information not available So why are utilities incurring rising financial losses? AT&C losses, though important as an indicator of efficiency, are not a complete indicator of the financial health of distribution utilities. The financial health of utilities is also dependent on the extent to which they can recover their costs through the tariffs charged for electricity consumption. Since the establishment of the independent regulatory framework in the late 1990s and the enactment of the Electricity Act 2003 (EA 03), ERCs have been entrusted with the mandate of examining and approving the costs incurred by utilities on the basis of certain norms and electricity sales, and setting efficiency parameters such as target distribution loss levels. However, given that costs such as those for power purchase, which form 300 | Indian Infrastructure: Evolving Perspectives the majority of the costs, are often beyond the control of the utilities and sales cannot always be accurately forecast, costs may exceed the revenues of utilities, leaving a revenue gap. To allow the utilities to recover all genuine and uncontrollable increases in expenses, a process of truing up is followed wherein the variances in actual costs incurred by utilities or revenues earned vis-à-vis those projected by utilities and approved by ERCs are allowed to be recovered by utilities. This truing up takes place with a lag because the audited accounts of utilities that form the basis of truing up are finalized only after the end of a financial year. To put it simply, the annual accounts for a year, say FY 2006–07, would be available only during FY 2007–08. Therefore, variation in expenses incurred by a discom during FY 2006–07 would be recovered only in FY 2008–09. This indicates that variations in costs and revenues do get adjusted and the financial losses of utilities should not escalate to unmanageable levels. The question that then arises is why losses have escalated. The gap between tariffs (represented by average revenue realized or ARR) without subsidy and average cost of supply (ACS) has doubled between 2005–06 and 2008–09 (see Figure 16.4). Further, cost recovery through tariffs deteriorated from 85 per cent to 77 per cent in this period. The situation is not much different if subsidies are considered. ARR with subsidy continues to be inadequate to cover ACS and the revenue gap with subsidy has doubled from Rs 0.16/kWh to Rs 0.33/kWh in this period. 4 3.40 2.93 2.76 3 2.60 2.38 2.39 2.54 Gap 2.62 Average cost of supply 2 2.39 2.21 2.27 2.03 2.09 Average revenue realized 1.95 without subsidy 1 0.78 0.54 0.43 0.36 0.45 0.39 0.49

0 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.4: Gap between ARR (without subsidy) and ACS at the all-India level has increased (Rs/kWh) Sources: Power Finance Corporation Ltd, IDFC analysis Detailed analysis indicates that the majority of utilities recover less than 90 per cent of costs incurred (see Table 16.7). Amongst the states that have exhibited increasing Power Distribution: Being Driven to Insolvency | 301 financial losses, utilities in Rajasthan, TN and MP exhibit cost recovery of less than 70 per cent. The utilities in Haryana, AP and Karnataka have varying levels of cost recovery, but cost recovery remains below 80 per cent. A few points need to be noted here. First, some utilities belong to states that are resource-rich and therefore have surplus power (governed by the policy of the state government to take free power or first right of refusal for a certain quantum of power at cheap rates). These utilities end up selling power at high rates in the market, adding to their revenues. Second, there is a trend to maintain uniform tariffs within a state despite the existence of more than one distribution utility. Even though utilities have different sales mixes and cost structures, and consequently revenue mixes, tariffs are set using the weakest utility as the benchmark. As a result, the utilities with a favourable consumer mix show better cost recovery and are in a better financial position while utilities which do not have such a consumer mix continue to perform poorly. Exceptions to this trend of uniform retail tariffs include Karnataka.

Table 16.7: Cost recovery in 2008–09 Cost recovery through Utilities revenue realized (without subsidy) < 50% Northern AP, Manipur, J&K 50–70% Bihar, Jharkhand, Mizoram, Nagaland, Ajmer, Jodhpur, Jaipur, Dakshin UP, Madhya UP, Poorv UP, Hubli, Uttar Haryana, Central AP, Southern AP, Northern AP, Chamundeshwari, Tamil Nadu, Madhya MP, Paschim MP, Purv MP 70–80% Arunachal, Dakshin Haryana, Punjab, Paschim UP, Uttarakhand, Gulbarga, Mangalore 80–90% Kesco, Eastern AP, Bangalore, Puducherry, Uttar Gujarat 90–100% Central Orissa, Northern Orissa, Southern Orissa, Meghalaya, BRPL, Dakshin Gujarat, Madhya Gujarat, Paschim Gujarat, MSEDCL > 100% Western Orissa, Sikkim, WBSEDCL, Tripura, BYPL, NDPL, HP, Kerala, Chhattisgarh, Goa Sources: Power Finance Corporation Ltd, IDFC analysis Note: For detailed names of utilities, refer to Annexure 302 | Indian Infrastructure: Evolving Perspectives

Historically, one of the major problems facing the power sector is the inadequacy of tariffs to meet the cost of supply. In the initial years of regulatory reforms, tariffs were keeping pace with costs (see Figure 16.5). But, in recent years, tariff increase has once again not kept pace with the increasing ACS. While ACS increased by 31per cent between 2005–06 and 2008–09, ARR without subsidy increased by only 17 per cent. 18 16.0 16 Increase in average cost of supply 14 Increase in average revenue realized 12 without subsidy 9.6 10 % 8 6.3 5.7 6.2 6.2 6 5.3 4.1 4 3.0 2.4 2.7 2 0.4 0 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.5: In recent years, tariff increase has not kept pace with increasing ACS Source: Power Finance Corporation Ltd, IDFC analysis

Why have costs risen? Power purchase costs have increased sharply The cost of supply has been increasing mainly due to the increases in power purchase costs. A sample analysis of the power purchase costs of utilities in 12 major states in India (see figure 16.6) indicates that power purchase costs have increased by 47 per cent between 2005–06 and 2008–09. This increase was mainly on account of increase in short-term power prices, increase in fuel costs, and higher cost of power from new projects. Of course, increase in fuel costs would also lead to some increase in the price of short-term power as well as in new power being more expensive. Since power purchase costs account for the lion’s share of costs of utilities (89 per cent of the total expenses in 2008–09), increases in these costs are bound to lead to a huge increase in the overall ACS. Power Distribution: Being Driven to Insolvency | 303

4.50 4.00 2005–06 2008–09 3.50 3.00 2.50 2.00 1.50

Rs/Unit 1.00 0.50 0.00

AP AP

UP UP

BYPL

BRPL

Hubli

Jaipur

Ajmer

NDPL

Punjab

Gujarat

Jodhpur

Purv MP

Gulbarga

MSEDCL

Poorv UP

Bangalore

Mangalore

Eastern

Central AP

WBSEDCL

Madhya

Tamil Nadu

Dakshin

Paschim UP

Madhya MP

Paschim MP

Southern AP

Northern

Uttar Gujarat

Uttar Haryana

Madhya Gujarat

Paschim Gajarat

Dakshin

Dakshin Haryana

Chamundeshwari Figure 16.6: Power purchase costs have increased (Rs/kWh) Note: For SEBs, cost of post purchase includes the cost of own power generation Source: Power Finance Corporation Ltd, IDFC analysis

Short-term power procurement by utilities Power procured by utilities through short-term trades increased by nearly 75 per cent in volume terms between 2005–06 and 2008–09, though its share as a percentage of total electricity generation in the country remained low. There has been a significant jump in this procurement between 2008–09 and 2009–10; with short-term power accounting for 8.6 per cent of the total electricity generated in FY 2009–10 (see Figure 16.7).

70 8.63 10 Volume of short term 60 transactions 8 As % of electricity 6.76 50 generated 40 6 % 30 3.15 4 Billion units 2.45 20 2.41 2

10 20.96 14.19 15.02 65.90 0 46.70 0 2005–06 2006–07 2007–08 2008–09 2009–10 Figure 16.7: Procurement of short-term power is increasing* Source: Central Electricity Regulatory Commission, IDFC analysis Note: * Includes UI 304 | Indian Infrastructure: Evolving Perspectives

The price of short-term power has also exhibited a significant upward trend over the years, with FY 2008–09 registering a particularly sharp increase (see Figure 16.8). This may be attributed to the shortage of power (see Figure 16.9), increase in maximum Unscheduled Interchange (UI) 11 charges to Rs 10/kWh in January 2008 from Rs 7.45/kWh, and the socio-political compulsions on utilities to provide power to agriculture or compulsions to ensure uninterrupted power supply in the run up to general elections (held in April–May 2009) even at the cost of sourcing expensive short-term power.

8 7.11 7

6 5.09 4.52 5 4.51

4 3.23 3

2

1

0 2005–06 2006–07 2007–08 2008–09 2009–10 Figure 16.8: Short-term power prices have shot up (Rs/kWh)* Sources: Central Electricity Regulatory Commission, IDFC analysis Note: * Average price of electricity transacted through UI during the calendar years 2008 and 2009 have been used as proxy for FY 2008–09 and FY 2009–10, respectively for analysis

18 Peak deficit 16.6 Energy deficit 16

13.8 14 % 12.7 12.3 11.9 12

10 11.1 8.4 10.1 9.6 9.9 8 2005–06 2006–07 2007–08 2008–09 2009–10 Figure 16.9: Peak and energy deficit in India (%) Source: Central Electricity Authority Power Distribution: Being Driven to Insolvency | 305

Though short-term power accounts for less than 10 per cent of the power generated in the country, the high rates at which this power is procured have significantly affected the financial viability of many utilities. Amongst the states that have shown high increases in financial losses, AP, MP, Rajasthan, Karnataka, and TN have seen short-term power purchase volume double between FY 2007–08 and FY 2008–09 (see Figure 16.10). Rajasthan particularly stands out here. In FY 2008–09, between August to March, Rajasthan procured the highest quantum of short-term power amongst all states, accounting for 14 per cent of the total short-term power purchase in the country. The Government of Rajasthan, in a note dated December 2009,12 has identified this as one of the main problems affecting the financial viability of utilities in the state. It has noted that the utilities procured short-term power at an average rate of Rs 9/kWh during April to December 2009 and estimates that a subsidy of Rs 2254.25 crore would need to be provided to the utilities against this power purchase. 6000

5000 4905

4488

3891 4000 3731

3000 2776

2469

2502

2402

2365

2060

1907

1892

2000 1590

1484

1393

1362

1041

856 1000 718

508

434

396

246

41 0

Kerala

Punjab

Gujarat

Haryana

Rajasthan

Karnataka

Tamilnadu

West Bengal

Maharashtra

Uttarakhand

Uttar Pradesh

Andhra Pradesh 2007–08Madhya Pradesh 2008–09 Figure 16.10: Purchase of short-term power in select states (MU)* Sources: Central Electricity Regulatory Commission, IDFC analysis Note: * Includes UI, 2008–09 data is for the period August to December

Reliance on imported coal Over the years, domestic availability of coal has become inadequate for meeting the growing requirement for electricity generation. Therefore, power plants are increasingly resorting to coal imports. The Ministry of Power even assigns generating entity-wise targets for coal imports in consultation with the Central Electricity Authority. Though imported coal accounts for only 3.2 per cent of the total coal consumed by power plants in the country, the import of coal for power plants has more than doubled since 2005–06 (see Figure 16.11). The increase in imports has 306 | Indian Infrastructure: Evolving Perspectives been particularly significant in FY 2008–09 and FY 2009–10. The price difference between domestic and imported coal after adjusting for calorific value can be estimated at about 26 per cent.13 25 23.2

20 16.1 15

10.4 9.7 10.2 10

Million tonnes 5

0 2005–06 2006–07 2007–08 2008–09 2009–10 Figure 16.11: Coal imports for power plants have doubled between 2005–06 and 2008–09 Source: Ministry of Power

180 163 Coal, Australia 150

120 116 95 71 90 68 48 47 50 53 72 60

30

0

Jul- Jan- Jul- Jan- Jul- Jan- Jul- Jan- Jul- Jan- Sep Mar Sep Mar Sep Mar Sep Mar Sep Mar

2005–06 2006–07 2007–08 2008–09 2009–10 Figure 16.12: Trends in price of imported coal (in rupees) Source: Ministry of Finance Employee costs have increased At the all-India level, employee costs for distribution utilities have increased by 76 per cent between 2005–06 and 2008–09. The increase has been largest in 2008–09 Power Distribution: Being Driven to Insolvency | 307

(see Figure 16.13) and can be attributed to the revision in employee cost as per the recommendations of the Sixth Pay Commission. 40 34 35

30

25 20 20

15 9 10

5

0 2006–07 2007–08 2008–09 Figure 16.13: Increase in employee costs for distribution utilities in India Sources: Power Finance Corporation Ltd, IDFC analysis Why are tariffs not keeping pace with costs? Tariffs are inadequate to cover costs for two reasons. First, tariff increases have not been timely. Some states have not raised tariffs for the past five years (See Table 16.8). There are instances of utilities not filing tariff determination petitions before state electricity regulatory commissions (SERCs) or delaying such filings.

Table 16.8: Status of tariff revision in states/union territories at the end of 2009 Tariffs last revised No. of states/UTs States/UTs 1 year 13 Andhra Pradesh, Assam, Chhattisgarh, Gujarat, Himachal Pradesh, Karnataka, Madhya Pradesh, Orissa, Punjab, West Bengal, Arunachal Pradesh, Sikkim, Delhi 1–2 years 6 Bihar, J&K, Maharashtra,* Meghalaya, Uttar Pradesh, Uttarakhand 2–3 years 2 Kerala, Tripura 3–5 years 5 Rajasthan, Jharkhand, Mizoram, Nagaland, Chandigarh > 5 years 5 Haryana, Tamil Nadu, Goa, Manipur, Puducherry Sources: Economic Survey of India 2010–11, IDFC analysis Notes: *for MSEDCL; States such as Haryana and Tamil Nadu have undertaken tariff revisions during 2009–10 or 2010–11. However, given the huge revenue deficits to be 308 | Indian Infrastructure: Evolving Perspectives

covered through tariff hikes and the consequent tariff shock to be faced by consumers, some ERCs have had to resort to the creation of regulatory assets. Examples include Haryana and Tamil Nadu. Second, even when tariff revisions have taken place, the gap has not been reduced. This is on account of several reasons. • Lack of information – The absence of reliable and adequate data or studies on the part of the utilities implies that ERCs do not have a solid basis to determine and approve costs and efficiency-improvement targets. – The annual accounts of utilities are not finalized on time, implying the absence of authentic information on the costs and consumption base of utilities.

• Delays in tariff orders – The issue of tariff orders is often delayed due to utilities not following the process timelines for submission of related petitions and additional information or due to inadequate data submission by utilities. Tariff revisions are therefore ineffective.

• Limitations of truing up – ERCs may not follow the practice of truing up. For instance, in case of power purchase costs, only few states such as AP, Assam, Gujarat, Haryana, Kerala, Maharashtra and Punjab have laid down principles or methodology for automatic adjustment of fuel and power purchase cost in tariff. These principles are known as Fuel Cost Adjustment (FCA) formula or Fuel and Power Purchase Price Adjustment (FPPPA) formula or Fuel Surcharge formula. However, states have either recently started cost recovery through such a formula (such as Kerala in January 2010 and Assam in December 2010) or face delays in cost recovery on account of delays in filing of claims and obtaining necessary regulatory approvals. If there is no automatic adjustment of these costs, utilities have to wait for the annual truing up exercise to recover them in case such an exercise is conducted. – ERCs do not allow the inefficiency of utilities to be passed to consumers. ERCs set efficiency targets for utilities. Further, they allow only uncontrollable costs to be trued up. In the event that utilities are not able to achieve their performance targets or contain costs, and they are not able to justify such non-achievement, ERCs do not allow the under-performance to be passed on through higher tariffs. For instance, the level of distribution Power Distribution: Being Driven to Insolvency | 309

losses affects the power purchase quantum and costs of a utility. If a utility does not achieve the distribution loss reduction set by the ERC, the ERC disallows the power purchase expenses on account of high distribution losses. Another example of inefficiency of utilities is not following the merit order schedule for procurement of power i.e. cheapest sources of power be offtaken first by the utility. ERCs may not accept purchase of expensive power in such case. – ERCs may not recognize the true extent of even the genuine and uncontrollable costs. For example, in the case of states such as Assam and Maharashtra, the FPPPA is subject to a ceiling of the variable component of tariff. In Assam the adjustment is subject to a ceiling of 25 per cent of the variable component of tariff while in Maharashtra, the ceiling is 10 per cent. Similarly, ERCs may disallow unmetered sales by utilities and consequently allow lower power purchase requirement and costs. – Utilities do not approach ERCs for truing up of actual costs and revenues either due to non-finalization of annual accounts or to avoid further tariff hikes. Examples of states where utilities have not approached ERCs for truing up include UP, Haryana, and TN. – Utilities do not implement truing up as allowed by the ERCs even though this is in their own commercial interest. Examples here include UP. • Creation of regulatory assets – ERCs may resort to the creation of regulatory assets wherein the recovery of costs through tariff hikes is postponed to future years to avoid tariff shocks to consumers. Examples of states where SERCs have created regulatory assets are Bihar, Haryana, Tamil Nadu, West Bengal and Orissa. To sum up, the inadequacy of tariffs has led to an increase in revenue gap in case of many utilities (see Table 16.9).

Table 16.9: Increase in revenue gap without subsidy for utilities between 2005–06 and 2008–09*

Increase in revenue Utilities gap without subsidy Decrease in gap Southern Orissa, Western Orissa, Sikkim, WBSEDCL, Arunachal, Lower Assam, Upper Assam, Manipur, Meghalaya, Nagaland, Tripura, HP, J&K, Paschim UP, 310 | Indian Infrastructure: Evolving Perspectives

Table 16.9: Increase in revenue gap (contd...)

Increase in revenue Utilities gap without subsidy

Poorv UP, KESCO, Kerala, Chhattisgarh, Goa, All Gujarat utilities

< 10 paisa Central Orissa, Northern Orissa, BYPL, Gulbarga

10–20 paisa Jharkhand, Dakshin UP, Uttarakhand, MSEDCL

20–50 paisa Bihar, Central Assam, BRPL, Dakshin Haryana, Punjab, Bangalore, Hubli, Puducherry

50–100 paisa Mizoram, Madhya UP, Eastern AP, Southern AP, Mangalore, Chamundeshwari, Tamil Nadu, All MP utilities

> 1 Rupee Uttar Haryana, Ajmer, Jodhpur, Jaipur, Central AP, Northern AP

Sources: Power Finance Corporation Ltd, IDFC analysis Notes: *NDPL reported zero revenue gap in 2005–06 and 2008–09 and has therefore not been included in the above table. Decrease in revenue gap does not imply that these utilities are necessarily performing better than the others. It only indicates the trend. These utilities may continue to have substantial revenue gaps. Therefore, this table needs to be read along with Table 16.6. For detailed names of utilities, refer to Annexure.

Tariff rationalization has also been slow An important contributor towards the financial viability of the distribution business is tariff rationalization. The goal of tariff rationalization is cost-reflective tariffs and reduction in cross subsidies. The National Tariff Policy mandates SERCs to notify a road map for tariffs to be within ± 20 per cent of the average cost of supply by end of FY 2010–11. While SERCs have taken steps to rationalize tariffs, the progress is inadequate. In 2008–09, the cross subsidy levels were high as compared to the target laid down by the Policy (see Table 16.10). Progress in tariff rationalization would help mitigate the problem of rising financial losses, particularly when agricultural sales increase. More importantly, it would reduce the dependence of utilities on subsidies from state governments. Power Distribution: Being Driven to Insolvency | 311

Table 16.10: Consumer tariffs as percentage of Average Cost of Supply approved by SERCs in FY 2008–09 Domestic Agriculture Non-domestic/ HT industry commercial Andhra Pradesh 88% 4% 214% 140% Assam 80% 72% 130% 110% Bihar 52% 27% 116% 101% Chhattisgarh* 58% 54% 145% 115% Delhi# 76% 41% 145% 129% Gujarat 82% 27% 129% 144% Haryana 80% 6% 100% 100% Himachal Pradesh 50% 20% 154% 111% Jharkhand# 42% 48% 155% 124% J&K* 31% 46% 53% 60% Karnataka 100% 17% 162% 129% Kerala 59% 26% 150% 155% Madhya Pradesh 92% 72% 148% 128% Maharashtra 100% 40% 170% 120% Punjab 93% 73% 138% 126% Rajasthan 90% 41% 131% 99% Uttar Pradesh 71% 49% 96% 137% Uttarakhand 69% 24% 123% 116%

Sources: Forum of Regulators/CRISIL Risk and Infrastructure Solutions Ltd report on Study on Analysis of Tariff Orders & Other Orders of State Electricity Regulatory Commissions Notes: *FY 2007–08; # FY 2006–07

Subsidies required from state governments are rising Another area of concern is the sharp rise in subsidies from state governments to utilities. The subsidy booked by utilities to the state governments has shot up from Rs 12,000 crore in 2005–06 to almost Rs 30,000 crore in 2008–09 (see Figure 16.14). But the subsidy payouts by the state governments are much less than determined. In 2008–09, only 60 per cent of the subsidy booked by the utilities was released to them. In many cases, subsidies are not released in a timely manner. 312 | Indian Infrastructure: Evolving Perspectives

35000 105% 30000 94% 100% 95% 25000 89% 84% 90% 20000 85% 15000 29665 80% Rs crore 75% 10000 19518 70% 12233 13590 5000 62% 65% 0 60% 2005–06 2006–07 2007–08 2008–09 Subsidy booked by utilities Figure 16.14: Subsidies booked by distribution utilities are rising but payment by state governments is inadequate Sources: Power Finance Corporation Ltd, IDFC analysis Rajasthan and AP have exhibited the maximum increase in subsidy between 2005–06 and 2008–09 (see Figure 16.15). While both states subsidize several categories of consumers, AP has introduced free power for agriculture since FY 2006–07. Not surprisingly, the payment of subsidy by the state governments to the utilities has been falling drastically in these two states. It is pertinent to note that the Government of Andhra Pradesh paid a subsidy of Rs 2,146 crore during 2009–10 against the total committed subsidy of Rs 3,486 crores to the utilities.14 The subsidy committed for 2010–11 has increased further to Rs 3,652 crores.15 On the whole, except Rajasthan, AP and Jharkhand, state governments have been paying the full amount of subsidy due to utilities (see Figure 16.16). 8000 7655 7980 2005-06 2008-09 6000

4000 2602 2637 1832 1581 2000 1436 1268 1629 1533 944 1179 915 1080 1178 1289 361 363 0

b

Punja

Gujarat

Haryana

Rajasthan

Jharkhand

Tamil Nadu

Uttar Pradesh

Andhra Pradesh

Madhya Pradesh Figure 16.15: Top ten states exhibiting the maximum increase in subsidy booked (in Rs crore) Source: Power Finance Corporation Ltd, IDFC analysis Power Distribution: Being Driven to Insolvency | 313

100% 100% 100% 100% 98% 96% 94% 86% 80% 84% 62% 66% 60% 58% Rajasthan 40% 38% Karnataka 37% Jharkhand 20% Andhra Pradesh 14% Tamil Nadu 0% 7% 2005–06 2006–07 2007–08 2008–09 Figure 16.16: States not paying full amount of subsidy to utilities Source: Power Finance Corporation Ltd, IDFC analysis

What are the specific reasons for increasing losses in the worst performing states? As seen from Table 16.1, the states that have exhibited the most deterioration in financial health are Uttar Pradesh (UP), Punjab, Karnataka, Haryana, MP, Rajasthan, TN and AP. Analysis of the underlying causes helps identify three clear reasons behind such deterioration. Poor governance In case of Andhra Pradesh, increasing dependence of the utilities on subsidy and non-payment of this subsidy by the state government is the reason behind the worsening financial health of utilities. The State Government has been providing free power to agriculture from FY 2006–07. This implies that as per the provisions of EA 03, the state government has to provide subsidies to the utilities in lieu of provision of free power. Therefore, given the dependence on subsidies, losses without subsidy have spiraled between FY 2005–06 and FY 2008–09. Further, while losses on subsidy-booked basis have been fairly low for the utilities (see Table 16.1), losses on subsidy-received basis have shot up between FY 2005–06 and FY 2008–09, thereby indicating that the state government has not been able to provide the requisite subsidies to the utilities (see Figure 16.15 also). In the states of TN, Rajasthan, Haryana and UP, poor governance on the part of utilities and state governments by way of not seeking/allowing tariff revision (see Table 16.7) has been responsible for the increasing financial insolvency of utilities. This does not imply that inefficiency of utilities has no role to play in their weak financial health. But the root cause of the problem in these states is the failure of governance. In Tamil Nadu, post the issue of the first tariff order in March 2003, the Tamil Nadu Electricity Board (TNEB) approached the Tamil Nadu Electricity Regulatory 314 | Indian Infrastructure: Evolving Perspectives

Commission for revision of retail tariff only in July 2010; this despite manifold increase in TNEB’s input cost and accumulating revenue deficit since 2003–04. Similarly, utilities in Rajasthan have not approached the ERC for a tariff revision since FY 2004–05. The Rajasthan Electricity Regulatory Commission (RERC) in all tariff orders since then has left the revenue gap uncovered. RERC, in the orders passed on approval of Annual Revenue Requirement of utilities, has observed that the revenue gap can be bridged or reduced by measures such as tariff increases, power purchase adjustment in retail supply tariff, subsidy from state government and by higher reduction of AT&C losses than targeted. It has urged the utilities to file tariff petitions for bridging the revenue gap. A note of the Government of Rajasthan dated December 200916 on the state’s power sector clearly brings out that utilities need the permission of the state government to file tariff revisions petitions. The remedial measures contemplated by the state government to salvage the deteriorating financial position of the utilities include permission to utilities to file tariff petitions for the year 2009–10, allowing utilities to file Power Purchase Fuel Cost Adjustment (PPFCA) formula before RERC, allowing utilities to recover Rs 300 crore as the power purchase and fuel cost adjustment for the quarter ended December 2007 and March 2008 as approved suo moto by RERC, and requesting RERC to suo moto allow recovery of PPFCA for the year 2008–09. The state government estimates that the last measure alone would provide a relief of at least Rs 1000 crore for FY 2008–09 and more than Rs 1250 crore for 2009–10 to the utilities. As is the case with RERC, the Haryana Electricity Regulatory Commission (HERC) too has not increased retail tariff in Haryana from FY 2005–06 to FY 2009–10 because the two utilities in the state have neither given a tariff proposal nor suggested any other mechanism to deal with their revenue gaps. Consequently, HERC has addressed the revenue deficit each year by considering additional revenue resulting from further reduction in loss level, additional government subsidy, and creation of regulatory asset. In FY 2008–09, HERC has not been able to address the revenue deficit fully and therefore left the same untreated. In UP, the state has not seen tariff hikes commensurate with the increasing cost. Therefore, the revenue deficit of utilities has been increasing every year. While there were no tariff revisions between FY 2005–06 to FY 2007–08 (since the utilities did not propose any revisions), only partial tariff revision has taken place in FY 2008–09. The revenue gap of utilities, as proposed by the utilities themselves, is being met through committed government subsidy, additional government subsidy and institutional finances against government repayment guarantee (see Table 16.11). The latter being akin to working capital loans, the UP Electricity Regulatory Commission (UPERC) in its orders has warned that such management of financial Power Distribution: Being Driven to Insolvency | 315 resources grossly against the prudent financial practice and continuation of such a practice would lead the utilities into a vicious debt trap.

Table 16.11: Funding of revenue gap of utilities in Uttar Pradesh FY 2006–07 FY 2007–08 FY 2008–09 Total expenses of utilities 13,428 16,260 17,535 Revenue from prevailing tariffs 9,992 11,424 13,218 Revenue gap 3,436 4,836 4,317 Funded through: Tariff revision - - 1,839 Government subsidy 1,012 1,822 1,532 Additional government subsidy 500 Short term loans 1,151 2,307 Others (power purchase cost savings/efficiency improvement) 773 707 946 Source: Compiled from orders issued by Uttar Pradesh Electricity Regulatory Commission Notes: Totals may not match due to rounding off; No orders were issued in relation to the Annual Revenue Requirement and Tariff Determination for FY 2005–06 because of inordinate delays in submissions related to the same by the utilities.

The problem of governance in UP extends beyond just tariff revision. As observed by the UPERC in its tariff orders, despite the reform process in the state power sector starting as early as 1999, structural arrangements of the sector have not been resolved. The sector is characterized by the lack of accountability, institutional strengthening, institutional capacity and autonomy in management. Direct intervention of State Government continues in minutest administrative, technical and commercial matters. Finally, adherence to the legislative mandate of the sector has almost grounded to a halt.

Inefficiency of utilities In Punjab and MP, the main reason for the declining financial health of utilities is inefficiency. In both states, utilities have not been able to achieve the distribution loss reduction targets laid down by the State Government17 /ERCs and the actual distribution losses have generally been much higher than the targets. Consequently, while arriving at the power purchase quantum for the utilities during truing up and for subsequent years, the ERCs have considered the distribution loss levels mandated by them. Further, they have disallowed power purchase expenses on account of high T&D loss (see Figures 16.17 and 16.18). 316 | Indian Infrastructure: Evolving Perspectives

28

27 26 25 25 24 24 24 24 24 24 23 23 23 22 22 22 21 Proposed by PSEB 20 21 Approved & trued up by PSERC 20 Actual as per PSEB 20 20 18 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.17: Distribution losses in Punjab (%) Source: Compiled from orders issued by Punjab State Electricity Regulatory Commission

45 43 43 41 41 42 40 39 40 37 39 37 38 37 35 36 34 33 35 31 33 30 29 30 27 29 27 25 2006–07 2007–08 2008–09 Poorv MP - Actual losses Poorv MP - State govt’s target Paschim MP - Actual losses Paschim MP - State govt’s target Madhya MP - Actual losses Madhya MP - State govt’s target Figure 16.18: Distribution losses in Madhya Pradesh (%) Sources: Compiled from orders issued by Madhya Pradesh Electricity Regulatory Commission, Annual Revenue Requirement Filings by Utilities for FY 2009–10, Annual Accounts for Paschim MP discom for FY 2006–07, Forum of Regulators/Crisil Infrastructure Advisory’s report on Assessment of Reasons for Financial Viability of Utilities It would be useful to note that in case of Punjab, distribution loss levels have always been a contentious issue, with their being vast differences in the targets proposed by the Punjab State Electricity board (PSEB) and that approved by the Punjab State Power Distribution: Being Driven to Insolvency | 317

Electricity Regulatory Commission (PSERC). The main reason for this is the changing approach and norm for ascertaining agriculture consumption, which is largely unmetered, and the debate over the reasonability of distribution loss reduction trajectory proposed by PSEB and set by PSERC. Another issue related to inefficiency of PSEB is its employee costs. PSERC, in its orders, has repeatedly observed that the employee cost of PSEB is one of the highest in the country and has recommended that that PSEB takes effective steps to contain this cost. While PSEB has cited several measures such as freezing fresh recruitment, complete ban on creation of new posts, withdrawal of compassionate appointments to dependants of deceased employees, introducing special schemes for employees to avail of long leave for self-employment for controlling employee cost, PSERC has argued that the PSEB has not taken a holistic view of its manpower requirements keeping in view norms issued by PSERC and productivity levels. There are therefore, significant differences in the actual employee costs of PSEB and that allowed by PSERC (see Figure 16.19). The Appellate Tribunal for Electricity (ATE)18 has supported the PSERC view’s and observed that PSEB’s initiatives for reducing employee cost are not forceful and have remained ineffective. It has further observed that the employee cost of PSERC would remain capped until performance parameters improve.

2500 Proposed by PSEB Approved by PSERC

Actual as per PSEB Trued up by PSERC 2202

2042 2000

1768

1751

1631

1627

1558

1541

1500 1379

1275 1275

1000

500

0 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09 Figure 16.19: Employee costs of PSEB (Rs crore) Source: Compiled from orders issued by Punjab State Electricity Regulatory Commission Note: In truing up employee costs, PSERC has allowed annual increases at the rate of WPI on the previous year’s trued-up costs. Therefore, approved and trued up costs differ in some years. While inefficiency of utilities should not be ignored and ERCs are right in penalizing errant utilities by not allowing costs on account of inefficiency to be passed on to consumers in the form of higher tariffs, the fact remains that this only adds to the 318 | Indian Infrastructure: Evolving Perspectives vicious cycle of poor financial health of utilities which have few incentives to reduce financial losses in the poor governance framework of the sector. It would be useful to note that Maharashtra, too, has seen some issues of inefficiency, with the ERC not allowing the Maharashtra State Electricity Distribution Company Ltd to recover higher than approved operation and maintenance (O&M) and higher interest costs (due to high short-term loans) through tariffs.19

Extraneous circumstances In Karnataka, the multi-year tariff (MYT) orders for the utilities for the period 2007–08 to 2009–10 was delayed due to a direction from the ATE to the Karnataka ERC (KERC) not to pass the tariff orders for the utilities till the ATE passes the final orders in case of an appeal filed by the Karnataka Power Transmission Corporation Limited challenging a past Tariff Order issued by the KERC. Consequently, the MYT orders for utilities were issued in January 2008 and became applicable from February 1, 2008. As a result, costs incurred for FY 2007–08 could not be fully recovered during the same financial year and tariffs could not be revised for FY 2008–09. The delay in issue of tariff order due to a pending appeal was therefore responsible for the financial deterioration of utilities in FY 2008–09.

How are utilities financing their revenue deficits? The revenue deficit of utilities is being financed through state government subsidies, short term borrowings from banks, and delays in payments to be made for power purchase. A snapshot of this financing is indicated in Table 16.12. It would be useful to point out that the quantum of change in factors such as bank loans would not necessarily match up to losses not met through subsidies. This is because bank loans are also used to finance long-term capital expenditure by utilities. Absence of information prevents the disaggregation of these loans by time periods to understand the extent of short-term borrowing by utilities. However, there is evidence to suggest that a large part of the increase in bank loans may be on account of short-term borrowings by utilities to finance their revenue deficits. This analysis is presented later in this section. Table 16.12: Means of financing the revenue deficits of utilities (indicative) (Rs crore) 2005–06 2008–09 Financial losses without subsidy 20,914 50,585 Direct subsidy 12,233 29,665 Outstanding loans from state governments* 37,328 28,682 Loans from banks/FIs/bonds* 55,318 1,15,285 Sources: Power Finance Corporation Ltd, IDFC analysis Power Distribution: Being Driven to Insolvency | 319

Notes: *For the state of Uttar Pradesh, Uttar Pradesh Power Corporation Ltd has been included. UPPCL is the state government owned company that holds 100 per cent shares of the distribution utilities in Uttar Pradesh. UPPCL also borrows from financial institutions on behalf of these utilities.

Reliance on state government financing As mentioned earlier, direct subsidies provided by state governments have more than doubled between 2005–06 and 2008–09. In addition to these subsidies, state government support for utilities is also available in the form of loans, guarantees, and equity investments. In recent years, loans from state governments to utilities have declined for two reasons. The first is that new loans from state government to utilities have reduced seemingly because of the poor financial position of some of the state governments themselves. In fact, this is also an important reason for the non-payment of subsidies to utilities. The second is that state governments have sometimes adjusted overdue loans to utilities against subsidies due. For example, in Punjab, the state government had recalled its overdue loans of Rs 1,362 crore in February 2008 and adjusted the same against balance unpaid subsidy for 2007–08.20 State governments also give cash subventions to utilities to meet the revenue deficits. Examples include Rajasthan (see Table 16.13). Further, they extend guarantees to financial institutions to support the borrowings of utilities because of the precarious financial position of these utilities. The Thirteenth Finance Commission (TFC) observes that the overall outstanding guarantees extended by states to utilities as on March 31, 2008 amounted to Rs 88,385 crore. The TFC further observes that equity investments amounted to Rs 71,268 crore on March 31, 2008 and have not been earning financial returns for the state governments, barring isolated instances.

Financing through short-term borrowings Utilities have been resorting to increased levels of short-term borrowing to cover their revenue deficit, meet repayment obligations against past loans, and meet the increased costs of power purchase. It is relevant to point out that given the already precarious financial position of many utilities, the short-term loans from banks are obtained against guarantees by state governments. Table 16.13 provides information on the states that have exhibited high increases in loans from banks/FIs/bonds. There is evidence to indicate that a large part of these loans is going towards meeting the revenue deficit of utilities rather than towards capital expenditure (see Table 16.13). Several ERCs have highlighted this in the orders issued by them. It would be useful to point out that the interest on the short- term loans taken to meet revenue deficits are not allowed as a pass-through in tariff 320 | Indian Infrastructure: Evolving Perspectives 21 23 determined that the Punjab State 22 Electricity Board has been diverting capital funds for revenue purposes. The ERC, has based its findings on the Board’s audited accounts and determined cumulative diversion of Rs 3828 crore in 2006–07 which reduced to observes, “Such an approach is not only against the regulatory principles but also against consumer interests as it would lead to increase in costs by way of interest and repayment obligation on such loans. The Commission views a management of financial resources grossly against the prudent practice and warns that continuation of such a would lead the power utilities of UP into a vicious debt trap. However, the Commission has allowed these institutional loans as subsidy from GoUP and the debt servicing of such loans is to be directly funded by the GoUP through budgetary provisions.” The Punjab ERC has in its tariff orders Rs 2,625 crore in 2008–09. of their revenue gap through short term loans from banks. The UP ERC that approved by the Haryana ERC in 2008–09. This variation is primarily on account of interest burden short-term loans taken by the distribution utilities for meeting the revenue deficit. guarantees to utilities from to distribution government Regulatory Commissions banks/FIs/bonds power sector Outstanding loans Outstanding Observations from orders issued by Electricity 2005–06 2008–09 2008–09

Table 16.13: Outstanding bank loans and government guarantees in select states (Rs crore) Punjab 5175 13518 2,909 Uttar Pradesh* 7478 13713 15,794 As indicated in Table 16.11, utilities UP have been meeting a large portion KarnatakaHaryana 1431 1523 3103 7088 1,208 2,708 There is a large variation in the actual interest cost claimed by utilities and Power Distribution: Being Driven to Insolvency | 321 25 has observed that since the distribution utilities are government- 24 of the Government Rajasthan states that there was incremental 26 The revenue deficit of distribution utilities is partly met by the cash subvention received from the State Government as agreed in Financial Restructuring Plan. The balance gap as deferred revenue subvention receivable from the State Government in its books as per Financial Restructuring Plan approved. However, utilities have resorted to short-term borrowings to meet this unmet revenue deficit. The interest on these borrowings for the utilities was as follows: The MPERC keep the distribution companies financially afloat. Ajmer – Rs 145 crore in 2007–08 and 169 2008–09 Jaipur – Rs 121 crore in 2007–08 and 91 2008–09 Jodhpur – Rs 109 crore in 2007–08 and Rs120 2008–09 A note short-term borrowing of Rs 3189 crore during 2007–08 and 3807 during 2008–09 by the distribution utilities in state to meet higher power purchase cost from bilateral, UI and energy exchanges. deficit over and above the subsidy provided to some consumer categories, devolves on the state exchequer by way of bail out or increased subvention to guarantees to utilities from to distribution government Regulatory Commissions banks/FIs/bonds power sector Outstanding loans Outstanding Observations from orders issued by Electricity 2005–06 2008–09 2008–09

Table 16.13: Outstanding bank loans and government guarantees in select states (Rs crore) (contd...) Rajasthan 6202 20316 22,262 MadhyaPradesh 651 4215 3,368 owned companies, therefore, eventually the additional burden of revenue 322 | Indian Infrastructure: Evolving Perspectives ies. that the TNEB is borrowing funds to meet 27 In lieu of these dues, the utilities are borrowing funds 28 repayment obligations, payment of interest on borrowings and also to meet the revenue expenditure in view of consistent deficit for the past years. It has further observed that there been no major generation capacity addition by TNEB for the last ten years. This implies that loans capital expenditure should be on the lower side. pointed out that the Government of AP has to pay nearly Rs 10,000 crore as dues to these utilities. from banks. In its response, the Central AP distribution utility has confirmed that it has raised short-term loans on behalf of the state government against the subsidy receivable. guarantees to Outstanding bank loans and government guarantees in select states (Rs crore) (contd...) : utilities from to distribution government Regulatory Commissions banks/FIs/bonds power sector Outstanding loans Outstanding Observations from orders issued by Electricity 2005–06 2008–09 2008–09

Table 16.13 Power Finance Corporation Ltd, Mercados Report for Thirteenth Commission, Orders issued by appropriate Electricity *Includes Uttar Pradesh Power Corporation Ltd (UPPCL). UPPCL is the state-government-owned company that holds 100 per cent : Andhra 3543 9003 10,257 As part of the regulatory process for approval Annual Revenue Tamil Nadu 9304 21502 2,694 The TN ERC has observed Pradesh Requirement of distribution utilities for 2011–12, some stakeholders have shares of the distribution utilities in Uttar Pradesh. UPPCL also borrows from financial institutions on behalf these utilit Regulatory Commissions, IDFC analysis. Sources: Notes Power Distribution: Being Driven to Insolvency | 323 by ERCs (see Table 16.14). This is because these loans are not related to particular assets on the part of the utilities, and tariff regulations prevent ERCs from allowing interest costs on short-term loans beyond the normative working capital interest. This is leading the utilities into a debt trap.

Table 16.14: Interest expenses disallowed by ERCs primarily on account of short-term loans taken by distribution utilities (Rs crore)

2005–06 2006–07 2007–08 2008–09 Haryana 23 60 178 365 Punjab 25 54 254 465 Maharashtra* 163 11 143 Rajasthan 313 353 NA NA Uttar Pradesh 80.41 – 505 Sources: Forum of Regulators/Crisil Infrastructure Advisory’s Report on Assessment of Reasons for Financial Viability of Utilities. Notes: *For MSEDCL; NA – Not available.

Delays in payments for power purchase Utilities have been delaying payments on account of power purchase. While there is no direct evidence of such delays, a reasonable sense of such delays can be gauged from the payments outstanding to state-owned generating companies, transmission companies and trading companies (see Figure 16.20). About 75 per cent of the increase in outstanding payments is accounted for by UP. But outstanding payments have also doubled in states such as Gujarat, Madhya Pradesh, Maharashtra and Haryana.

45000 41700 20000 2005–06 2008–09 40000 33771 15000 35000 10000 30000 26958 24609 5000 25000 0 20000 15000 10000

Gujarat

Haryana

Rajasthan

5000 Karnataka

0 Maharashtra

Uttar Pradesh 2005–06 2006–07 2007–08 2008–09

Madhya Pradesh

Jammu & Kashmir Figure 16.20: Debtors for sale/transmission of power for state-owned generation, trading and transmission companies (Rs crore) Sources: Power Finance Corporation Ltd, IDFC analysis 324 | Indian Infrastructure: Evolving Perspectives

How are the other states performing better? States such as Chhattisgarh, Gujarat, HP, Kerala, West Bengal and Orissa have either exhibited stable financial performance or have reduced their financial losses. Kerala, West Bengal and HP have negligible agriculture consumption (see Table 16.4). Kerala and HP have a favourable power purchase mix, with 50 per cent of Kerala’s installed capacity and 75 per cent of HP’s installed capacity being hydro.29 Further, the SEB in HP also has a substantial amount of surplus power after accounting for the state’s power consumption. Thus, favourable consumer mix and demand supply position plays an important role in the better performance of SEBs/utilities in these states. Chhattisgarh too has low agriculture consumption (at 11 per cent in 2008–09). These states also have lower power purchase costs; with these costs accounting for 58 per cent to 67 per cent of total costs. With Chhattisgarh being coal-rich, thermal power is available cheap. This power is sold at market rates, adding to the revenues of the SEB. Thus, despite the inefficiency in terms of high and rising AT&C losses, the state has been amongst the better performers. What is clear from the above is that though some SEBs/Utilities may be inefficient (in terms of AT&C losses or expenditure patterns); factors such as favourable consumer mix, favourable demand supply position, and abundance of resources help these SEBs/Utilities overcome the losses incurred due to their inefficiency. Skewed tariffs (see Table 16.9) further work to the advantage of utilities in these states. Gujarat appears as an exception and has been amongst the better performers because of the substantial reduction in AT&C losses, tariff revisions30, restriction of annual subsidies, and timely payment of subsidies.

Where are financial losses headed? It is important to understand the future course of losses because the finances of many state governments may not be able to bear the burden of these losses and banks may not have an unending appetite to finance them at the guarantee of such state governments. Further, if the losses persist and keep increasing, banks may find themselves at risk as the debt servicing ability of utilities would be under serious pressure. The rising losses are also a concern, given the expected surge in power generation capacity in the country in the next few years; from 143877 MW at the end of 2009–10 to 240963 MW at the end of 2013–14 (see Figure 16.21). The share of the private sector would increase from 11 per cent to 28 per cent of the total capacity during this period. Unless the solvency of the distribution business is restored with urgency, it is unlikely that the utilities will be able to sustain payments for this additional capacity. Back of the envelope computations indicate that losses of utilities would range between Rs 93,000 crore to Rs 1,50,000 crore in 2012–13. Power Distribution: Being Driven to Insolvency | 325

These losses and the interest burden of short-term borrowings for meeting them would together account for 1 per cent to 1.6 per cent of the country’s GDP in this year (see Table 16.15).

Table 16.15: Estimated financial losses of utilities in 2012–13 (Rs crore)

2008–09 2012–13 Business Higher Higher Tariffs Higher AT&C as usual annual AT&C match loss reduction increase loss costs & tariffs in tariffs reduction match costs AT&C loss (%) 28.44 24.44 24.44 20.44 24.44 20.44 Average tariff 3.66 4.62 5.36 4.62 4.98 4.98 Average cost of supply 3.40 4.98 4.98 4.98 4.98 4.98 Average revenue realized (without subsidy) 2.62 3.49 4.05 3.68 3.76 3.96 Loss per unit (without subsidy) 0.78 1.49 0.93 1.30 1.22 1.02 Commercial losses 148829 92941 130303 121900 101950 Interest for short- term loans 17859 11153 15636 14628 12234 Total losses from distribution sector 166688 104094 145939 136529 114183 Ratio of commercial losses to GDP 1.6% 1.0% 1.4% 1.3% 1.1% Source: IDFC analysis General assumptions – Increase in average cost of supply: 10 per cent p.a. – Units input in 2012–13: 1001970 MU, computed by taking a 10 per cent increase in energy input over 2011–12 (estimated using energy availability projected in 17th EPS and transmission losses of 4 per cent) – GDP in 2012–13 at market prices: Rs 9348857 crore as projected by the GOI’s High Powered Expert Committee for Estimating the Investment Requirements for Urban Infrastructure Services – RoI for short term loans: 12 per cent Case-specific assumptions – Business as usual: Tariff increase of 6 per cent p.a., reduction in AT&C losses of 1 percentage point p.a. 326 | Indian Infrastructure: Evolving Perspectives

– Higher annual increase in tariffs: Tariff increase of 10 per cent, reduction in AT&C losses of 1 percentage point p.a. – Higher AT&C loss reduction: Tariff increase of 6 per cent p.a., reduction in AT&C losses of 2 percentage point p.a. – Tariffs keep pace with costs: Tariffs equal average cost of supply, reduction in AT&C losses of 1 percentage point p.a. – Higher AT&C loss reduction and tariffs keep pace with costs: Tariffs equal average cost of supply, reduction in AT&C losses of 2 percentage point p.a. 300000

250000 67535 200000 25371 30125 150000 16194 100000 177898 156818 173428 127683 50000

0 2009–10 2011–12(T) 2011–12(E) 2013–14

Public sector Private sector Figure 16.21: Private sector will lead future capacity addition in India (MW) Sources: Planning Commission, Central Electricity Authority, Infraline, IDFC analysis

Financial viability of the sector requires increase in tariffs and reduction in AT&C loss levels In the wake of the above analysis, turning around the distribution sector is possible by three measures: 1. increase in tariffs, 2. reduction in AT&C losses, and 3. reduced power purchase costs. Back-of-the-envelope calculations under these three scenarios indicate the following:

Increase in tariffs If only tariffs were to be increased to eliminate subsidies and bridge the entire revenue gap for the distribution business at the 2008–09 level, the all-India average tariff would need to increase by 30 per cent from the 2008–09 level.

Reduction in AT&C losses The R-APDRP targets the reduction of AT&C losses to 15 per cent in urban and in high-density areas by July 2013. At this level of AT&C loss (for the country), ARR without subsidy in 2008–09 would have increased to Rs 3.20/kWh, thereby reducing Power Distribution: Being Driven to Insolvency | 327 the revenue gap to Rs 0.20/kWh. A tariff hike of 6 per cent would then suffice to eliminate this revenue gap.

Reduction in power purchase costs Given the ACoS of Rs 3.40/kWh in 2008–09, it can be estimated that power purchase costs amount to Rs 3.03/kWh. A reduction in this cost can prove crucial for improving the financial viability of the utilities. Trends in bids for Ultra Mega Power Projects (UMPPs) indicate that power can be made available at competitive and lower rates than is currently the case (see Table 16.16). Assuming a scenario where power can be procured at the tariff for the Krishnapatnam UMPP (the highest tariff amongst the UMPPs) and normative transmission charges are Rs 0.22/kWh, the ACoS would amount to Rs 2.92. This would reduce the revenue gap by over 60 per cent to Rs 0.30/kWh. A tariff hike of 10 per cent would then suffice to eliminate this revenue gap.

Table 16.16: Tariff trends in UMPP bids

Name of UMPP Levelised tariff (Rs/kWh) Mundra, Gujarat 2.264 Sasan, Madhya Pradesh 1.196 Krishnapatnam, Andhra Pradesh 2.333 Tilaiya, Jharkhand 1.770 Source: Central Electricity Authority

CONCLUSION Contrary to popular belief, AT&C losses are not the raison d’etre for the worsening financial health of the power distribution sector. Overall AT&C loss levels, though high in absolute terms, have shown improvement over time. Analysis indicates that the sector has been unable to cope with the significant increases in power purchase costs on account of inadequate tariffs. Power purchase costs have increased primarily due to expensive short-term power purchase. But tariff revision has not kept pace with this increase. In some states, tariffs have not been revised for several years. In others, poor information base of utilities, non-availability of audited accounts, inefficiency of utilities, absence of truing-up, and inadequate governance has marred tariff revisions. Sometimes, cost recovery has been postponed to future years by resorting to the creation of ‘regulatory assets’. Deeper analysis indicates that eight states have shown the worse trends as far as rising commercial losses are concerned. These are UP, Punjab, Karnataka, Haryana, MP, Rajasthan, TN and AP. Poor governance and inefficiency of utilities are the 328 | Indian Infrastructure: Evolving Perspectives main reasons for the rising losses in these states. The rising losses are being financed by state governments by way of subsidies and by the financial system in the country. But subsidy payments by many state governments are inadequate and irregular. The situation is such that utilities are caught in the vicious cycle of increasing costs, rising revenue deficits, rising short-term borrowings to meet these deficits, and consequent increase in costs that are not allowed to be recovered through tariffs. As a result, utilities have been delaying payments against power purchase and now there is evidence of defaults by some utilities for such payments.31 More recently, utilities have started resorting to higher load shedding to avoid the burden of extra power purchase costs32 and to prevent their financial position from vitiating further. If urgent steps are not taken to drastically improve the governance of the sector and improve the efficiency of utilities, the sector will head towards insolvency and future capacity addition plans may get jeopardized. State governments are clearly not in a financial position to support the sector in the long run. A bigger and grave concern is the impact of this situation on the financial system of the country, which would get affected not only because of its exposure to the distribution sector but also because of the debt servicing by generation project developers whose cash flows are dependent on payments from distribution utilities.

Manisha Gulati Power Distribution: Being Driven to Insolvency | 329

ANNEXURE List of distribution utilities referred to in the note

Andhra Pradesh Central AP: Andhra Pradesh Central Power Distribution Company Ltd Eastern AP: Andhra Pradesh Eastern Power Distribution Company Ltd Northern AP: Andhra Pradesh Northern Power Distribution Company Ltd Southern AP: Andhra Pradesh Southern Power Distribution Company Ltd Assam Central Assam: Central Assam Electricity Distribution Co. Ltd Lower Assam: Lower Assam Electricity Distribution Co. Ltd Upper Assam: Upper Assam Electricity Distribution Co. Ltd Arunachal Pradesh Arunachal Pradesh: Department of Power, Arunachal Pradesh Bihar Bihar: Bihar State Electricity Board Chhattisgarh Chhattisgarh: Chhattisgarh State Electricity Board

Delhi North Delhi Power Limited (NDPL) BSES Rajdhani Power Limited (BRPL) BSES Yamuna Power Limited (BYPL) Gujarat Dakshin Gujarat: Dakshin Gujarat Vij Co. Ltd Madhya Gujarat: Madhya Gujarat Vij Co. Ltd Paschim Gujarat: Paschim Gujarat Vij Co. Ltd Uttar Gujarat: Uttar Gujarat Vij Co. Ltd Torrent Ahmedabad: Torrent Power, Ahmedabad Torrent Surat: Torrent Power, Surat Haryana Dakshin Haryana: Dakshin Haryana Bijli Vitran Nigam Ltd Uttar Haryana: Uttar Haryana Bijli Vitran Nigam Ltd 330 | Indian Infrastructure: Evolving Perspectives

Himachal Pradesh Himachal Pradesh: Himachal Pradesh State Electricity Board Karnataka Bangalore: Bangalore Electricity Supply Company Ltd Chamundeshwari: Chamundeshwari Electricity Supply Company Ltd Gulbarga: Gulbarga Electricity Supply Company Ltd Hubli: Hubli Electricity Supply Company Ltd Mangalore: Mangalore Electricity Supply Company Ltd Jharkhand Jharkhand: Jharkhand State Electricity Board Kerala Kerala: Kerala State Electricity Board Madhya Pradesh Madhya MP: MP Madhya Kshetra Vidyut Vitran Co. Ltd Paschim MP: MP Paschim Kshetra Vidyut Vitran Co. Ltd Purv MP: MP Purv Kshetra Vidyut Vitran Co. Ltd Maharashtra BEST: Brihanmumbai Electric Supply & Transport Undertaking MSEDCL: Maharashtra State Electricity Distribution Co. Ltd Reliance Mumbai: Reliance Infrastructure Limited – Distribution Business Manipur Manipur: Electricity Department, Manipur Meghalaya Meghalaya: Meghalaya State Electricity Board Nagaland Nagaland: Department of Power, Nagaland Orissa Central Orissa: Central Electricity Supply Utility of Orissa Ltd Northern Orissa: Northern Electricity Supply Company of Orissa Ltd Southern Orissa: Southern Electricity Supply Company of Orissa Ltd Western Orissa: Western Electricity Supply Company of Orissa Ltd Power Distribution: Being Driven to Insolvency | 331

Punjab Punjab: Punjab State Electricity Board

Rajasthan Ajmer: Ajmer Vidyut Vitran Nigam Ltd Jaipur: Jaipur Vidyut Vitran Nigam Ltd Jodhpur: Jodhpur Vidyut Vitran Nigam Ltd

Sikkim Sikkim: Energy & Power Department, Government of Sikkim

Tamil Nadu Tamil Nadu: Tamil Nadu Electricity Board

Tripura Tripura: Tripura State Electricity Corporation Ltd

Uttar Pradesh Dakshin UP: Dakshinanchal Vidyut Vittran Nigam Ltd Madhya UP: Madhyanchal Vidyut Vitran Nigam Ltd Poorv UP: Poorvanchal Vidyut Vitran Nigam Ltd Paschim UP: Pashchimanchal Vidyut Vitran Nigam Ltd

Uttarakhand Uttarakhand: Uttrakhand Power Corporation Ltd

West Bengal WBSEDCL: West Bengal State Electricity Distribution Company Ltd

NOTES 1. AT&C losses reflect the technical losses incurred in transmission and distribution of electricity as well as the commercial losses arising out of theft and deficiencies in billing and collection. 2. For the purpose of this note, distribution utilities refer to utilities selling power directly to consumers and include State Electricity Boards (SEBs), State Power Departments and distribution companies (Utilities). 3. Projected by the Thirteenth Finance Commission at 2008 tariffs 4. Cash profit is defined as profit after tax + depreciation + miscellaneous expenses written off + deferred tax. The definition of cash losses needs to be understood accordingly. 332 | Indian Infrastructure: Evolving Perspectives

5. Discussions with experts indicate that the trend in loss reduction would remain the same even with the availability of more reliable information. 6. 24 per cent in 2006–07 and 22.87 per cent in 2008–09 7. This is corroborated by the Tamil Nadu Electricity Regulatory Commission’s opinion in the 2010–11 tariff order that poor metering of agriculture consumption leads to gross underestimation of the total capacity of agricultural electricity connections in the state. Therefore, the state government subsidy towards agricultural consumption determined on the basis of this capacity is vastly inadequate to cover the actual expenditure incurred by the Tamil Nadu Electricity Board. 8. Tariff order issued by Madhya Pradesh Electricity Regulatory Commission for FY 2010–11 9. For more information, read the Report on Implementation of Accelerated Power Development and Reforms Programme (APDRP), Ninth Report, Standing Committee on Energy (2005–06), Fourteenth Lok Sabha, available at http://164.100.24.208/ls/ CommitteeR/Energy/9rep.pdf 10. There has been news of re-tendering of projects with winning bidders being relegated to the background or being dismissed. Contractual issues have led to some major and experienced companies not being able to bid. Bid values are being questioned as is the ability of some IT companies to execute projects in such values. The time frame of 18 months is proving to be unrealistic as the utilities are not ready or equipped (by way of expertise) to absorb and drive the project. Further, the pace of decision making is not in line with the tight timelines of the program. 11. UI is a mechanism developed to improve grid efficiency and grid discipline by imposing charges on those who defer from their scheduled power generation or drawal. 12. Available at http://www.mop.rajasthan.gov.in/downloadpdf/noteonpoewrsector.pdf, accessed on 18 January 2011 13. This data pertains to August 2009 and has been used to give a sense of the price differential between domestic and imported coal. (Source: Planning Commission) 14. Tariff order issued by Tamil Nadu Electricity Regulatory Commission for Tamil Nadu Electricity Board for 2010–11 15. Tariff order issued by Tamil Nadu Electricity Regulatory Commission for Tamil Nadu Electricity Board for 2010–11 16. Available at http://www.mop.rajasthan.gov.in/downloadpdf/noteonpoewrsector.pdf, accessed on January 18, 2011 17. In MP, the state government laid down the distribution loss reduction trajectory for the utilities for the period FY 2006–07 to FY 2010–11 in December 2006. 18. Judgement dated May 26, 2006 as referred to in PSERC’s Tariff Orders for FY 2007–08 and FY 2008–09 Power Distribution: Being Driven to Insolvency | 333

19. Forum of Regulators/Crisil Infrastructure Advisory’s Report on Assessment of Reasons for Financial Viability of Utilities 20. Tariff order issued by the Punjab State Electricity Regulatory Commission for the Punjab State Electricity Board for 2009–10. 21. Tariff order for the distribution utilities for 2008–09 22. Tariff orders issued for the Punjab State Electricity Board for 2008–09 and 2010–11 23. Forum of Regulators/Crisil Infrastructure Advisory’s Report on Assessment of Reasons for Financial Viability of Utilities 24. Tariff order issued for the distribution utilities for 2010–11 25. Order on Multi-year Aggregate Revenue Requirement for the year 2007–08 and 2008–09 for the distribution utilities 26. Available at http://www.mop.rajasthan.gov.in/downloadpdf/noteonpoewrsector.pdf, accessed on January 18, 2011 27. Tariff order for Tamil Nadu Electricity Board for 2010–11 28. Available at http://www.apcentralpower.com/ARR/Replies%20ARR%20filings% 20FY%202011–12/Replies% 20to%20Objections%20on%20ARR%20filings%20FY% 202011–12.doc accessed on April 7, 2011 29. As in December 2010; Source: Power Scenario at a Glance, Jan 2011, Central Electricity Authority 30. This does not imply that the other states recording better performance have not seen tariff revisions. Tariff revisions have taken place as and when necessary. However, in case of states such as HP, Kerala and Chhattisgarh which have revenue surplus or have very low revenue deficits, tariff revision is not the prime concern. 31. The MP ERC in the tariff order issued for distribution utilities in MP for 2009–10 has observed that the losses incurred by utilities have not only wiped out their entire equity capital but have also forced them to default against payments due from them. 32. Transcript of Power Finance Corporation Limited’s Investors’ Conference Call held on January 17, 2011 and available at http://www.pfcindia.comPFCTranscript_17012011.pdf 334 | Indian Infrastructure: Evolving Perspectives

INDIA SOLAR POLICY: Elements Casting Shadow 17 on Harnessing the Potential November 2011

1. INTRODUCTION India, chugging along its self-defined high growth trajectory to achieve inclusive and sustainable development, is faced with the multiple challenges of ensuring energy security, overcoming energy poverty and defining a low-carbon development path. To power its economic growth, the country has traditionally depended primarily on conventional fuel which is also likely to remain the mainstay for future development. However, despite sitting on large coal reserves India has been recently facing large supply disruptions, and domestic coal production has failed to keep pace with the rapid build-up in demand. To add to this, the conventional thermal energy market is going through convulsions the world over. Faced with critical challenges, India has to look out for and harness alternative energy sources to secure its development objectives.

Blessed with 300 sunny days a year and receiving an average hourly radiation of 200 MW/sq km, the country is well placed to overcome its key challenges by harnessing the enormous solar potential. Around 12.5 per cent of the land mass, or 413,000 sq km, could be used for harnessing solar energy.1 Recognising this, the Government of India (GOI) included solar energy as a key mission under the National Action Plan on Climate Change and formally launched the Jawaharlal Nehru National Solar Mission (JNNSM) in 2010. The JNNSM is a major initiative of the GOI to ramp up its solar power generation capacities in a phased manner and seeks to provide impetus to the development of a huge solar market in India. The mission has set out to achieve 1000 MW of grid-connected solar projects at 33 kV and above (with equal share of solar photovoltaic or SPV and concentrated solar thermal or CST), 100 MW of rooftop and small solar projects, and 200 MW of off-grid projects by India Solar Policy | 335

2013 (Phase I); 4000 MW (and 10,000 MW on the uptick) grid-connected and 1000 MW off-grid projects by 2017 (Phase II); and 20,000 MW grid-connected and 2000 MW off-grid projects by 2022 (Phase III). The JNNSM aims to achieve 20 million solar lighting systems for rural areas and increase the solar thermal collector area to 15 million sq metres by 2017 and 20 million sq metres by 2022. To kickstart Phase I, the solar policy (in JNNSM) put forth a mechanism of ‘bundling’ relatively expensive solar power from grid-connected projects, selected on pre- defined criteria under JNNSM, with equivalent capacity of power from the unallocated quota of the GOI generated at NTPC coal-based stations, which is relatively cheaper. This ‘bundled power’ would be sold by NTPC Vidyut Vyapar Nigam Ltd (NVVNL) to the distribution utilities at Central Electricity Regulatory Commission (CERC) determined prices. The Mission provided for NVVNL to procure the solar power by entering into a power purchase agreement (PPA) with the selected solar power developers (SPDs) connected to the grid at a voltage level of 33 kV and above before March 2013. The solar procurement price initially proposed for Phase I was Rs 17.91/unit. Further, considering that some of the grid- connected solar projects were already at an advanced stage of development, the guidelines for migration of projects from their respective existing arrangements to the ones envisaged under JNNSM were also issued. The projects to be selected under this scheme provide for deployment of both solar PV technology projects and solar thermal technology projects in a ratio of 50:50, in MW terms. Subsequently, guidelines for selection of new grid-connected solar power projects (SPV and CST) were issued in July 2010 by the Ministry of New and Renewable Energy (MNRE). Further, the JNNSM provides for promotion of manufacturing and development of solar technology by mandating domestic content in projects in Phase I with the objective of establishing India as a global leader in solar energy, and also provides for promotion of off-grid and small solar applications with the objective of scaling up solar deployment and meeting energy requirements in rural and remote areas. In addition to the impetus provided at the central government level through JNNSM, several states like Gujarat, Rajasthan and Karnataka have come out with their own solar policies, providing for preferential tariffs and other deployment support (which include provisioning of infrastructure, wasteland for development, evacuation infrastructure and solar parks). At the close of Batch I of Phase I, 37 projects totalling 620 MW (listed in Appendix 1) were allotted under the JNNSM (comprising 30 SPV projects totalling 150 MW and seven CST projects totalling 370 MW), 84 MWs of projects brought under the fold of the migration scheme, about 716 MWs of solar projects allotted under the Gujarat Policy (Appendix 2) and several others in various states under state policies. 336 | Indian Infrastructure: Evolving Perspectives

At a time when project selection for the first batch of JNNSM Phase I has got over, as also the bidding for the second batch of Phase I, it is prudent to take stock of the developments and assess the effectiveness of the solar policy for achieving the strategic objectives. The questions to ask are: • Does the JNNSM inspire confidence in facilitating speedy deployment of solar capacity and meeting its desired objectives? • Is the present solar policy really aligned with the strategic vision of the government? This note explores these questions in the light of the developments of the first batch of JNNSM Phase I and the provisions of the solar policies in India and the experiences in other countries.

2. ISSUES AFFECTING EFFECTIVE HARNESSING OF SOLAR POTENTIAL On the face of it, with the large allotment of capacities under the JNNSM, it appears that the mission is off to a successful start and India has indeed put together a comprehensive solar policy, a major improvement over the previous guidelines, to make the country the mecca of solar development and deployment. However, as the India Solar programme moves from policy to implementation, several issues and concerns have surfaced, proving to be major hurdles for serious solar power developers and lenders. In fact, several projects allotted may not get deployed. Financing and bankability of solar projects under the JNNSM is emerging to be a major concern, arising from several issues in the policy or inadequacies therein. Thus, despite an apparently improved policy and support provided, solar power development remains expensive and risky for developers and lenders. These issues of concern are discussed here: a. Auctioning or reverse bidding Initially, with the announcement of the JNNSM, solar projects were offered a feed- in-tariff (FiT) or preferential tariff of Rs 17.91/unit to SPV projects and Rs 15.31/ unit to CST projects shortlisted in Batch I. However, considering the overwhelming response received for the SPV capacity offered, GOI chose to replace the FiT with a reverse bidding or auctioning mechanism where the fixed capacity of SPV offered would be allotted based on least price offered by SPDs and not first come, first serve basis. The auctioning process resulted in allotment of the first 150 MW SPV projects and 470 MWs CST projects, with 37 SPDs emerging as winners comprising 30 SPV projects and 7 CST projects. The auctioning led to huge discounts over the initial FiT, where the weighted average of the quoted tariffs for SPV was Rs 12.16/unit and CST was Rs 11.41/unit implying an average of 32 per cent and 25 per cent India Solar Policy | 337 respectively of the CERC-declared FiTs. However, it may be noted that some of the gains due to auctioning would be offset by higher tariffs paid to the 84 MW projects qualified under the migration scheme as compared to the rates offered to these projects by state distribution companies (discoms). Considering the huge savings over the economic life of the projects, the move towards auctioning does appear to be a great success. However, such huge discount in tariffs over CERC-determined FiT and the ultimate selection of the SPDs have raised a number of questions regarding the financial feasibility of the projects, hindrances in financial closure and hence timely completion of the projects. The CERC- determined tariffs included reasonable returns for the SPDs, but offering such discounted tariffs in the face of high financing costs may result in lenders finding the returns unattractive for a risky business in an immature market. One could also suspect that to protect the margins developers may resort to using substandard equipment which could result in sub-optimal performance. Also, the winning list of potential developers has companies with questionable credibility in the solar business. The potential developers include a wool yarn maker, an animation company, auto dealer and pipes supplier. In view of this, it is to be seen how many of the projects achieve financial closure and are commissioned on time. It is indeed the case that very few projects would be able to get non-recourse financing. Although the present policy does include penal provisions imposed on SPDs for abandoning projects or underperformance, lenders may prefer to wait and watch developments as they evolve given no precedence of the effectiveness of penal provisions and the various risks they are exposed to. This is not only going to affect the deployment of projects in Batch I but also the next tranche of bidding. At the same time, the cost benefits of the auctioning mechanism should not be ignored. Strengthening the prequalification criteria of bidders to include prior arrangement with a credible EPC partner and due diligence report on project viability accompanying bids would somewhat mitigate the risks perceived by the lenders. Unfortunately, the guidelines released by the ministry for Batch II of Phase I does not strengthen the pre-qualification criteria of the bidders and retains the same principles of allocation, except for extending the timeline for financial closure from 180 days to 210 days from the signing of the power purchase agreement (PPA). The present policy also does not provide for any protection from interest-rate risks. At a time when real interest rates are high, a 25-year PPA carries with it the risks of high tariff regime and discoms imposing pressure on the SPD to revisit tariffs when interest rates come down. Also, there is no way consumers can get the benefit of declining interest once the PPA is sealed at a higher interest rate. 338 | Indian Infrastructure: Evolving Perspectives b. Information on solar resource incidence and performance of technologies The performance of solar projects and returns thereof are highly dependent on the incidence of solar radiation. Quality solar radiation data with very high degree of accuracy and high level of confidence is an essential prerequisite for choice of technology, project development and viability of the project. Information on Direct Normal Incidence (DNI) of solar radiation is required for CST and Global Horizontal Incidence (GHI) is required for SPVs. Currently, good quality annual information with fairly high degree of accuracy is available, which is good for framing broad policies. However, moving towards planning for projects requires seasonal and monthly information, while project development, assessing project feasibility and designing projects require monthly, daily and hourly data with high level of confidence. Typical meteorological year data may not be always appropriate as they are mostly city based and not for remote areas where the project sites are located. Ground measurement of meteorological monthly-daily-hourly data at the regional level as well as within a few kilometres of proposed sites need to be collected over a statistically acceptable time period to provide developers and lenders the confidence desired. Inaccuracy in solar resource estimation affects expected future returns of the project. Since the information is backward looking, lenders prefer to be conservative with resources and thus prefer higher probability or ‘P’ levels (where P50 is average and P90 is high level of confidence but more conservative). In view of this, although the solar policy is well founded, planning for phased targets and identification of projects, let alone designing of projects, is definitely not on a firm footing. The fact that equal weights are given to large grid-connected SPV and CST in the absence of adequate information on solar resources on regional and site- specific locations for monthly, daily and hourly variations, raises questions about the effectiveness of the policy. Also, the risks of project selection and design are very high for lenders to feel comfortable. This resource issue brings up the next concern about the right choice of technology and size. The annual information shows that few regions in India have high DNI suitable for CST, but GHI levels are high and spread over several regions and thus more suitable for SPV. It may be also noted that SPV technology is very flexible in the context of any topography of land, but CST has to be necessarily set up on flat topography to ensure high levels of efficiency. Also, land requirement is large for grid-connected SPV projects. Thus, in the absence of adequate information on the correlation between the proposed sites and the technology selected by the SPDs, the lenders may find the risks very high and often not commensurate with the returns. Also, the emphasis on grid-connected SPVs in JNNSM compared to smaller off- grid and decentralised SPV applications, as well as the equal emphasis on SPV and CST capacity additions, appear to have weak analytical bases. India Solar Policy | 339

There is an urgency to build up in a transparent manner a repository of detailed information on solar radiation which, in turn, could be used to substantiate feasibility and hence bankability of projects. Although the ministry is putting in place 50 monitoring stations on the ground for the measurement of solar resources, more such on-ground information needs to be rapidly built up and widely disseminated for fast development and deployment of projects. c. Technology risk As discussed above, technology choice and design is strongly correlated with the quantum, type and variability of solar resource. Besides, solar technologies, be they grid-connected SPV or CST, are at early stages of deployment and development in India, and therefore carry higher risks on applicability and performance. In India, crystalline silicon technology accounts for most of the market and, currently, the market share of thin-film technology, though fast increasing, is very small (about 10 per cent only). Thin-film technology has not reached the efficiency level of crystalline solar cells, but could bring down the costs of production considerably. This is the risk of technology obsolescence in an emerging and rapidly developing technology segment. CSTs are currently not economically viable the world over and there is very limited information on their performance in India. Lenders are not comfortable about extending non-recourse finance because of higher levels of project construction and operating risks. The risks could be mitigated by building up information on the performance of technologies in the light of the resources available. The effectiveness and appropriateness of the JNNSM rests to a great extent on this information. d. PPA under JNNSM and financial state of discoms The PPA for the bundled scheme under JNNSM between the NVVNL and SPDs is a major payment security concern for lenders. This risk arises from the fact that NVVNL as a trader passes on risks of non-receipt of revenues from discoms for bundled power sales to the SPDs. Poor and worsening financial situation of the state discoms increases the credit risks for the lenders. Further, few discoms would honour their obligations under PPA for twenty years if costs of solar power comes down in the next five years. Given that payment default by state discoms is common, lenders do not find PPAs entirely credible and bankable. Recognising the importance of this concern of lenders, the ministry introduced an additional payment security scheme for grid-connected solar projects under JNNSM. The government has approved the Payment Security Scheme to facilitate financial closure of projects under Phase I of the JNNSM by extending Gross Budgetary Support (GBS) amounting to Rs 486 crore to the ministry in the event of defaults in 340 | Indian Infrastructure: Evolving Perspectives payment by the state discoms to NVVNL. The core component of the Payment Security Scheme (PSS) is to create Solar Payment Security Account (SPSA) financed from GBS to the ministry for availability of adequate funds to address all possible payment related risks in case of defaults by discoms for the bundled power. The PPAs have payment security mechanism for recovering payments through Letter of Credit (LOC), an escrow mechanism, and subsequently sale to a third party or even power exchange pending bilateral negotiation with the third party. The Payment Security Scheme will be implemented by the ministry with the provision of NVVNL opening the SPSA for this purpose and draw funds as per mechanism/provisions of the scheme. The funds for each year shall be allocated by MNRE into SPSA. As per estimates, the Rs 486-crore fund requirement is scheduled for Phase I as:

Table 17.1: Projected deployment of funds in SPSA

Fund deployment Incremental fund Total fund pattern deployment (crore) capacity (crore) 1 Jul 11 1.0 1.0 1 Jan 12 1.0 2.0 1 Jul 12 32.85 34.25 1 Jan13 23.47 58.32 1 Jul 13 58.32 116.64 1 Jan 14 126.39 243.03 1 Jul 14 243.02 466.05 Source: Implementation of a Payment Security Scheme (PSS) for Grid-connected Solar Power projects under Phase I of Jawaharlal Nehru National Solar Mission during the year 2011–12, MNRE, GOI, New Delhi, India, 2011 Introduction of this scheme has considerably mitigated the payment risks perceived by the lenders, but the uncertainty associated with the prospects of the projects beyond the first phase of JNNSM and the extremely poor financial situation of the discoms continue to pose bankability concerns for the lenders. e. Policy risk beyond Phase I of JNNSM A prime risk perceived by the developers and lenders is the policy uncertainty related to the bundling scheme beyond 2014. The project life is about 25 years, but the bundled scheme as well as the payment security is applicable for projects in Phase I, leaving the future of projects beyond 2014 uncertain. Further, in a situation where the power sector is reeling under acute power crisis, rising power supply deficits and coal supply disruptions, the availability of unallocated power from NTPC coal- India Solar Policy | 341 based plants is itself doubtful. Even if such unallocated quota exists and is allocated for the bundling scheme, it remains uncertain whether such power would be cheap enough if imported coal at very high prices is used. In fact, considering the phenomenal rise in imported coal prices in the global market and also due to the regulatory changes in coal exporting countries (viz. Indonesia and Australia), generation costs for entirely imported coal-based power in India would probably be as expensive as solar power (i.e. Rs 10 to Rs 15/unit). This also raises the issue that grid parity for solar projects may be reached earlier than that perceived in JNNSM. All of these question the validity of the development trajectory mapped in JNNSM and that of the associated provisions. f. Regulatory risk The Renewable Purchase Obligations (RPOs) provided for in the Electricity Act 2003, the solar-specific RPOs introduced under it, and targets specified through amendment of the National Tariff Policy, are considered a major push for grid- connected solar in JNNSM. The Renewable Energy Certificates (REC) are considered an important financing mechanism and incentive for SPDs. Although the market for REC has great potential, effective policing of RPOs is required as a prerequisite. Enforcing RPOs is an issue considering that most discoms are state-owned and it becomes a self-imposed penalty for non-compliance. Also, under circumstances that costs of solar technology and projects are coming down rapidly, the regulators would accordingly be revising the floor and forbearance prices downwards. In fact, only recently, the CERC has revised the floor and forbearance prices of solar REC from Rs 12,000/MWh to Rs 9300/MWh and Rs 17,000/MWh to Rs 13,400/MWh respectively. This regulatory uncertainty raises concern about the financial viability of the projects. It will be a while before the costs stabilise and the market for REC really picks up in any meaningful manner. The complexity arises from poor financial situation of the state discoms and their ability to absorb or pass through high cost of renewable energy. The solar policy needs to take cognizance of these issues associated with RPO and REC while specifying the targets and quantifying the benefits. g. Infrastructure constraint Solar project sites are mostly located in remote areas and wastelands, which are often not well connected and are lacking in infrastructure. Under the JNNSM, developers are required to put in place the required infrastructure themselves and also acquire land, secure water connection, get all clearances and put in place the evacuation infrastructure. This is not only costly and challenging for any developer, but also extremely time-consuming, and adds to the risks of delay in completion of 342 | Indian Infrastructure: Evolving Perspectives projects. Often, detailed information about project sites in remote locations is not available to the developer at the time of bidding. Creating a database with detailed information on the topography and project site, creating a land bank for projects, facilitating or getting all clearances for the developer, providing the necessary infrastructure in project sites and support in evacuation infrastructure would mitigate substantial risks and go a long way in successful deployment of solar projects. Also, financial support from the Clean Energy Fund could be provided to SPDs for infrastructure build-up, including evacuation infrastructure, to facilitate solar project development and deployment. h. Other concerns The JNNSM capacity development and deployment trajectory fails to take cognizance of the fact that finance companies are bound by exposure limits for the power sector, and renewable energy, including solar, is treated as part of the exposure to power. This is a major obstacle faced by the lending community and solar power deployment would face a major financing crunch unless the Reserve Bank of India (RBI) declares the solar sector a priority sector for lending or renewable energy is treated separately from the power sector.

3. INCONSISTENCIES WITH STRATEGIC VISION a. Equal weight to grid-connected SPV & CST projects Mandating development of SPV and CST in the 50:50 ratio and thus imposing technology reservation would curb efficient market allocation and cost reduction. If the ultimate objective of the India Solar Mission is to create an environment for competitive solar energy penetration in the country, then such reservations on technology would not be aligned with the strategic vision. JNNSM, in its attempt to encourage both the technologies, is dictating the technology choice rather than allowing the market to select the most efficient and cost-effective technology for Indian conditions. SPV is an established technology for large and small size projects, and the cost is coming down rapidly, while globally the use of large-scale CST for the generation of electricity is still a niche strategy. CST so far has been successfully implemented in only a few locations worldwide. It is indeed the case that northwestern India is among the list of world regions showing the best solar resource, but the CST technology is not yet mature. The land and water requirement of CST is also very high as compared to SPV. Rajasthan, which has high DNI levels suited to CST, has been reported to have a ‘critical’ water supply status, but still about 86 per cent of Phase I CSP projects are located in this state. It is common knowledge that cost competitiveness of solar energy could be achieved by major breakthroughs in India Solar Policy | 343 technology. This can only be achieved by a focused approach on Research and Development (R&D). Reservation in technology deployment is unlikely to have any major impact on this in India. On the other hand, if all the technologies are set to compete against each other, the most appropriate technology will be chosen in a cost-effective manner. Thus, auctioning without technology reservation could be adopted for award of projects. b. Emphasis on large MW size grid-connected SPV projects The emphasis in the JNNSM on developing MW-size grid-connected SPV projects, instead of primarily focusing on off-grid and decentralised small solar applications, appears to be lacking in strategic vision. Planning for MW-scale CST is still justified, where appropriate, but disproportionate focus on deploying MW-scale grid- connected SPV is definitely questionable. If, in the Indian context, the smaller decentralised and off-grid applications are more appropriate, then the subsidy or financial support should be directed towards the same. The German policy, which is often hailed as progressive and has played a significant role in Germany having the largest cumulative solar capacity in the world, has been primarily promoting smaller size solar applications. In fact, the country offers higher FiTs for smaller installations (e.g. rooftop PV), while the lowest FiTs apply to free-field installations. Germany, despite having much poorer solar radiation than India, went ahead and promoted development and deployment of smaller SPV applications, while India has focused on large MW-size projects despite having high GHI levels in most regions of India. Considering the large number of unelectrified households, the large power deficit situation, high supply losses, poor grid connectivity and poor financial status of the discoms, it would be more appropriate to promote decentralised solar applications. Solar resources being widely spread may have to be promoted in rural and remote areas on a small scale rather than centralised scale, more particularly keeping in view the problems of remote areas and weak and inefficient grids. With power from conventional fuels becoming more expensive, solar power is likely to achieve grid parity sooner than expected. This would make solar power at the decentralised or off-grid level more affordable. Smaller SPV applications could cover remote and rural electrification, street lighting, home lighting systems, integrated solar projects and hybrid stand-alone solar projects in places where grid supply is poor or absent. The total requirement for solar off- grid and decentralised applications could be very large, and this would provide enough positive impetus to the global solar industry. Large-size grid integrated projects may not give the desired multilevel benefits as compared to small/medium size projects. Thus, keeping in view rural development, rural employment, demand- side management, energy security, land availability, etc. the major focus should be 344 | Indian Infrastructure: Evolving Perspectives on channelising subsidy towards development and deployment of small solar applications so that the rural unelectrified may have access to clean energy at an affordable price. c. Mandating domestic content in solar technology The JNNSM guidelines specify that in the case of solar PV technology, all deployment under the scheme should use a module manufactured in India in Batch I, and in Batch II even the cells used in deployment should be domestically manufactured. In the case of CST, it specifies that 30 per cent of the total project cost be utilised for domestic equipment. This proposal may appear to protect the interests of Indian manufacturers and financiers, and give a major thrust to domestic manufacturing accordingly facilitating continuous reduction in cost of solar power. Further, it may also appear that given India’s significance as an emerging high growth market, this proposal would encourage international players to invest in manufacturing in the country. This, in turn, would help build up sustainable manufacturing capacity within the country, thereby creating jobs and significant opportunities for lenders and investors. However, the question arises whether this is consistent with the comparative advantage of India and aligned with its other policy objectives.

The fact is that India lacks a robust manufacturing base for solar components and systems for solar systems. Currently, it does not have any infrastructure for raw material production (polysilicon) and is entirely dependent on imports for the same. The bulk of the SPV industry is dependent on imports of critical raw materials and components—including silicon wafers. The entire SPV value chain shows that India does not have the comparative advantage up to the wafer-making stage. To reach the levels of wafer production, it has to produce polysilicon which involves the reduction stage using coke and conventional energy. India does not have enough coke for its steel industry and has to depend on imports. Availability of power is itself a major constraint and power is getting more expensive. Also, production of silicon, wafers and modules is capital- and not labour-intensive. Further, the bulk of the SPV cost is accounted for by PV modules, while the remaining comprises balance of system and construction. Thus, India clearly does not have the comparative advantage in production of SPV and has little ability to ramp up production in a big way, influence breakthrough innovations and bring about significant cost reductions just on the basis of a domestic solar market. Reduction in SPV costs depend on the size of the global PV market and, most importantly, on research breakthroughs. The domestic SPV market is unlikely to have any major impact on the global PV market, as the Indian market is small compared to the world market. The world market in Europe is, however, going through demand India Solar Policy | 345 reduction in the wake of the financial crisis. It may be noted that even the Chinese solar manufacturing industry (see Box 17.1), which has grown many fold riding on the export market in Europe, is now looking inward to the domestic market, and to export markets in the Asian region. The Indian semi-conductor industry had been receiving government support for a very long time and if it really had the comparative advantage it would have done well with wafer production and solar cell manufacturing from the waste of wafer manufacturing like the Chinese did. Indian solar cell and module manufacturers are overwhelmingly in favour of domestic content rules as they realize that it is difficult for them to compete with the much larger and lower-cost Chinese companies, moreso when their export market in Europe is going through convulsions and a slowdown. The United States has strongly opposed India’s local content requirements specified in JNNSM as it is creating an export hurdle to their solar companies like Sunpower and First Solar. It is noteworthy that the Indian installers and developers have also opposed this local content requirement as this results in reduced choice for suppliers and higher costs. There is no denying that developing manufacturing capability strengthens security of supply, but there could be other ways of encouraging manufacturing and at the same time ensuring that the benefits of incentives extended to them are not reaped by other countries. The strategic vision of India should be to encourage high efficiency and low cost immediate delivery of power on a globally competitive basis to benefit the common man and consumers, and to discourage monopolies and increase in subsidy burden, as also restrictive trade practices. In this regard, reservation of domestic content in modules and cells is unlikely to meet the desired objectives and may even result in no interest or ability on the part of domestic manufacturers to reduce costs and address possibilities of time overruns in the absence of adequate manufacturing capability. Creating domestic demand may not be necessary to encourage manufacturing. In fact, in the solar space, manufacturing has received encouragement for a long time—even well before the focus shifted towards solar deployment. Thus, instead of linking domestic content in deployment, the government should continue to provide financial support wherever the comparative advantage exists in the entire value chain of SPV and CST, and also increase the spend on R&D. Germany and many other countries have spent large amounts on R&D for a long time. Capital and other subsidies could also be extended to manufacturing as long as the same is passed on to the SPD through the equipment supplied to them and not passed on in exports in other countries. 346 | Indian Infrastructure: Evolving Perspectives

4. HAVE STATE SOLAR POLICIES DONE BETTER? Besides the GOI initiative with JNNSM, several states have formulated their own solar policies to attract investments in the solar space in their states. With the preferential tariff abandoned under the JNNSM, and replaced by the reverse bidding scheme for selection of SPDs, developers have turned to state-level policies. Most states have introduced their own FiT, which is kept between the CERC-specified tariff and the average bid tariff under JNNSM, to attract investments in the state. Gujarat, Rajasthan and Karnataka have already come up with their own solar policies, and several other states like Andhra Pradesh and Maharashtra are in the process of formulating their own policies. Gujarat was the first to introduce a solar policy in 2009, even before JNNSM was introduced by GOI. It was hailed by SPDs as a progressive solar policy, and with the Government of Gujarat demonstrating its commitment to development of renewable energy, the state over and above JNNSM has allotted 716 MW of solar capacity against an initially-declared target of 500 MW and signed PPAs for this capacity with 34 SPDs. The share of SPV and CST in the 716 MW capacity is 365 MW and 351 MW respectively (see Appendix 2). Many developers have also expressed interest in setting up manufacturing capacity in the state which expects to bring in a massive investment of about Rs 12,000 crore over the next few years. The Gujarat model offers procuring solar power from developers at a fixed tariff, a la the German and the Spanish model (see Box 17.1). However, unlike the German model, the focus is not so much on developing smaller off-grid and rooftop capacities but the emphasis is more on grid-supplied capacity addition. The state also facilitates land acquisition, clearances and provision for evacuation and other infrastructure. The state has also identified land banks for solar power development from wasteland areas available in the state. This would not only provide for productive use of wasteland areas but also bring in development in these areas. Taking into consideration the rapid development in solar technology bringing down costs and taking note of the need for preferential tariffs in the initial years till grid parity is achieved, Gujarat offers CERC-approved higher tariff for the first 12 years and a much lower tariff for the remaining 13 years. The applicable tariff as recently approved by CERC is for SPV Rs 15/kWh for the first 12 years and Rs 5/kWh subsequently, and for CST it is Rs 11/kWh for the 12 years and Rs 4/kWh subsequently. Rajasthan, which is also planning to develop more than 500 MW over the next few years, over and above JNNSM, on the other hand, has adopted an approach similar to the JNNSM. The policy offers all possible routes for the SPDs, from grid-connected to off-grid, and has also provided for bundling arrangements and auctioning routes for the state to procure power. Rajasthan, with its huge solar potential, is in any case India Solar Policy | 347 a very attractive destination for investors. The state has also already strengthened the transmission network, more specifically in regions where solar potential is relatively high. Karnataka introduced the policy with the intent to harness more than 200 MW of solar capacity, in addition to JNNSM, and has proposed to adopt the competitive bidding route for procurement of solar power at a discounted tariff. It is expected that soon most of the states would go for the auctioning route, as the cost burden of fixed FiT is quite high as more capacities are added and consumers are provided with subsidised power because of political considerations. The experience of Spain is a case in point, where recent reduction in FiT consequent to the financial crisis has seen a major downsizing of solar capacity addition. In these states with high solar potential, substantial capacity addition may also see annual capping of additional solar capacity if transmission capacity fails to keep pace. Notwithstanding aggressive policies introduced and large numbers of PPAs and capacity allotted, projects in these states may actually get delayed due to many reasons. Several projects may be postponed because of delay in financial closure as lenders and developers may still consider some of the risks much higher than the returns in a high-financing cost regime. The main concern with the states is the financial weakness of the state-owned discoms with high losses and surviving on subsidy from state governments. The other key risk faced by the SPDs is inadequacy of information on solar resources of the level of detail required for effective technology choice, planning and design of projects. In fact, many developers may decide to pay an annual penalty of several lakhs of rupees for delaying their projects, and wait for costs to drop so that they can get better returns. The risks are many and though state policies have been an improvement over JNNSM in some states, it is unlikely that large solar capacity additions would actually fructify over the next couple of years. With the intent of mitigating the risks perceived by the developers and lenders, and reducing costs of financing as well as facilitating availability of finance, Gujarat and Rajasthan have been developing solar parks, other states like Andhra Pradesh are to follow suit. The solar park concept is similar to that of industrial parks and involves the state government identifying areas where several MW-size plants amounting to more than 1000 MW capacity of solar generation can be established. The state government would provide to the solar park the necessary infrastructure, regulatory and other governmental support, including special fiscal and financial incentives as well as facilitating clearances. The costs incurred by the state government would be subsequently recovered from the developers in the park. Solar parks could be developed to accommodate both generators as well as solar manufacturers. The 500 MW Charanaka Solar Park in Patan District of Gujarat is being developed, and it is believed that all the projects in the park are on track. Similarly, Rajasthan is in 348 | Indian Infrastructure: Evolving Perspectives the process of developing a solar park which could accommodate more than 1000 MW. The park in Gujarat is being developed by the GPCL on around 2400 hectares of wasteland. It is expected that the solar park would decrease the cost of solar power generation due to economies of scale, accelerated development, fiscal benefits and reduced cost of finance because of reduced risks. The Planning Commission has approved Rs 210 crore worth of central assistance to the Gujarat Park and the Asian Development Bank (ADB) has approved a soft loan of about US$100 million, which includes development of a smart grid for evacuation of power. The two key challenges of availability of finance and financing cost for solar projects have been mitigated to a great extent with the development of solar parks, which appear to be a solution to scaling up MW-size solar projects in India. Discussions are on for the creation of a Solar Park Finance Vehicle and other financial tools, supported by credit enhancement mechanisms by domestic and international governments.

Box 17.1: Solar power: International experience

Germany: German solar policy’s objectives – both explicit and implicit – are among the most aggressive in the world. Germany’s planned phase-out of its substantial carbon- free nuclear generation plants, as well as its vulnerability to political uncertainty of natural gas supplies from Russia, had exacerbated its energy security concerns: a development that accelerated its support for renewable energy. On the domestic value addition front, solar PV has been a major part of Germany’s export-oriented economic development approach. Germany fixes the price for solar power in the form of feed-in tariffs (FiTs) over 20 years. Higher FiTs are offered for smaller size systems, usually rooftop, while the least are offered to ground-mounted systems greater than one megawatt. The quantity is somewhat controlled by imposing a strict annual degression rate, which is a percentage reduction in FiTs based on the quantity or solar capacity installed during the previous year. Due to the recent drop in solar PV prices, Germany reduced its FiTs twice during 2010. Even then, the annual installed solar PV capacity exceeded 7,400 MWs, equal to about a third of solar capacity addition expected under JNNSM in the coming decade. This translates to a large financial commitment for Germany’s electricity consumers over the next 20 years. Spain: Spain set the price for solar procurement by offering FiTs for 25 years. However, unlike Germany, it also fixed the quantity in the form of a cap, to limit the financial impact on its utilities. To circumvent the issue of project selection, as noted earlier, the Spanish government decided to accept all projects till one year after 85 per cent of the annual cap was met. When the Spanish government increased its FiTs for PV by 75 per cent in 2007 to provide a boost to its solar sector, 2,661 MW of PV were installed, exceeding the annual cap of 1,200 MW two times over. The additional capacity of 1,461 MW meant a large unexpected financial commitment of a net present value of several billion euros over the next 25 years. Further, the Spanish government had and India Solar Policy | 349

continues to keep electricity consumer tariffs low and reimburses utilities for the deficit by paying through the national budget, i.e. taxpayer monies. Spain was one of the worst hit countries during the financial crisis with a high budget deficit. Although the deficit was not all due to support for renewable energy, the government could not keep offering high FiTs for solar energy generation (Craig 2009). In September 2008, it slashed the FiTs by 23 per cent. The Spanish PV market collapsed with only 70 MW of installed capacity being added in 2009. Further, the Spanish government is even considering retroactive cuts to FiTs for existing projects, a move that breaches contracts and provides considerable uncertainty to the Spanish solar sector. California, USA: In December 2010, the California Public Utilities Commission in the United States introduced the Renewable Auction Mechanism to procure renewable energy projects of less than 20 MW, which mainly include solar. Under this mechanism, the required installed capacity will be fixed and projects selected based on least cost rather than first come, first served basis at a set FiT (CPUC 2010). The programme aims to use standard terms and conditions to lower transactional costs and provide contractual transparency needed for effective financing. China: China enacted its landmark Renewable Energy Law in 2005, which gave high priority to the development and utilisation of renewable energy. This led to a big push in renewable energy deployment, especially in the wind sector where China now has the largest wind deployment (approximately 45 GW) in the world. However, solar capacity additions have been relatively small until recently. The total installed solar capacity by the end of 2010 was approximately 900 MW, with more than half of this capacity (520 MW) coming in 2010 alone. Until 2009, the main push for solar PV in China was in off-grid installations for remote rural communities, the result of Brightness Rural Electrification and Township Electrification programmes that started almost a decade ago. In 2009, just prior to the Copenhagen talks, China launched its most ambitious solar deployment program, the Golden Sun initiative, to create some domestic demand for its solar manufacturers in anticipation of the declining international PV demand during the early days of the financial crisis. The programme aims to instal approximately 642 MW of grid-connected and off-grid solar PV at a cost of approximately US$3 billion over the next three years. However, the annual demand is an order of magnitude smaller than China’s PV cell manufacturing capacity. China has shown phenomenal growth in production, increasing its PV manufacturing capacity eighty-fold in the last five years; it was the largest manufacturer in 2010, producing approximately 13,000 MW, or 48 per cent of the global capacity. The Chinese solar energy industry began in the mid-1980s, when semiconductor companies started manufacturing solar cells with waste raw material from wafer production. By 2000, the domestic industry could fulfil the modest Chinese domestic market demand, although there were very little exports. Since 2005, China has focused on supplying solar PV equipment to Western countries such as Germany, Spain, and the US, where demand was buoyed by generous 350 | Indian Infrastructure: Evolving Perspectives

purchase support for PV deployment. The Chinese solar industry started developing a comprehensive supply chain, including the manufacture of polysilicon material, ingots, wafers, cells, and modules. This growth in the solar PV industry was concurrent with the Chinese government’s push after 2000 to develop a comprehensive semiconductor industry from chip design to production and testing. The Chinese government’s pro-export currency policy arguably played a major role in its export-oriented growth. This currency policy (used by Japan in the 1980s and Korea in the 1990s) pegged the Chinese currency to the US dollar, thus preventing it from appreciating against the same. The Chinese government also offers tax incentives and low-cost credit and financing from state-controlled banks to its solar industries, advantages enjoyed by other Chinese manufacturing sectors as well. Chinese manufacturers have also benefited from low labour costs, subsidised electricity rates, and close proximity to raw material suppliers. In terms of R&D support and strategic goals, the Chinese government has identified energy technologies such as hydrogen fuel cells, energy efficiency, clean coal, and renewable energy as focuses of the National High-Tech Development Plan (863 program), while making utility-scale renewable energy development central to the National Basic Research Program (973 program). It approved US$585 million jointly for the 863 and 973 programs in 2008. China’s recent purchase support policy initiatives do show promise, but it might be hard to raise domestic demand to match its manufacturing capacity, since the relatively high costs will be borne by the electricity consumers and the state exchequer via the National Renewable Energy Fund. Source: Ranjit Deshmukh, Ranjit Bharvirkar, Ashwin Gambhir and Amol Phadke. 2011. ‘Analysis of International Policies in the Solar Electricity Sector: Lessons for India’, Prayas Energy Group, Pune, India, and Lawrence Berkeley National Laboratory, CA, USA; India, July 2011. Ranjit Deshmukh, Ashwin Gambhir and Girish Sant. 2011. ‘India’s Solar Mission: Procurement and Auctions’, Economic & Political Weekly, 46 (28), July 9, 2011.

5. CONCLUDING COMMENTS The India Solar Mission is a significant improvement over all previous solar policies formulated by the government, and governments, both at the centre and in the states, are serious about kickstarting the development of solar energy and harnessing the huge potential that exists in India. Several countries, including Germany, that have added huge solar capacities do not have as much potential as India does, and thus there is huge investment opportunity in the solar space in India. However, to attract serious investors, India has to get its act right and put in place a conducive policy aligned with the country’s strategic vision and objectives. There are some elements in the present policy acting as hurdles in the way of mitigating the risk India Solar Policy | 351 perception, and some of the provisions may not be entirely aligned with the strategic objectives. However, drawing from international experiences and the lessons from the initiatives in India so far, the enormous potential for solar energy in the country can be unleashed by overcoming prevailing hindrances. The key challenge to developing solar projects is availability of finance and high financing costs primarily arising from the high risk associated with viability of projects and the weak financial status of the state-owned discoms. Although JNNSM attempts to address these, there are clear gaps in the way solar policy has been rolled out so far. The move towards the auctioning approach is indeed useful from the perspective of bringing down the costs at which solar energy is offered by SPDs, but a precondition for successful implementation of the same would be to strengthen the prequalifying criteria of the potential bidders. Ensuring firm agreements with credible EPC and detailed project viability reports should be included as prequalifying criteria to prevent participation of non-serious players. Also, the bid guarantee bonds should be deterrent enough to any non-serious participant. The most important element for mitigating multiple risks is to put in place adequate and quality information on solar resources and performance of technologies in a transparent manner. A lot of attention and investment should be channelised towards creating such a knowledge repository on a priority basis to mitigate risks perceived by the developer and the investor community. The targets specified in policies and plans should be based on a strong analytical foundation to send out credible signals. The off-taker risk arising from the high loss and poor financial situation of the procurers is common to both conventional and non-conventional power generators. In the case of solar energy, the risks are higher because energy is more expensive and hence risk exposure is higher. The GOI, as additional comfort to the prevailing payment security mechanisms in PPA, has provided for the creation of a solar payment security account from the gross budgetary support. Further, financial support extended by international development banks (like ADB) and international governmental support in the form of risk guarantee mechanisms and soft loans would bring additional comfort to lenders and developers. However, the real difference can be seen when state governments take necessary steps to improve the financial situation of the discoms. To rapidly scale up solar projects and prevent implementation delays, the government should ensure that delays do not arise with regard to clearances, land acquisition, and inadequacies in infrastructure. A Clean Energy Fund and other similar funds could be used to support and finance evacuation and other infrastructure essential for developing projects. 352 | Indian Infrastructure: Evolving Perspectives

Technology reservation in any form may not be optimal for allocation of resources. Thus, providing equal focus on SPV and CST has little justification when CST has not achieved much success and SPV has made significant strides in bringing down cost of solar energy. Ideally, for MW-scale projects, auctioning without technology reservation should be the right approach to ensure harnessing the most efficient technology in a cost-effective manner. Also, mandating domestic content for solar projects is sub-optimal. Creating a domestic market for the manufacturing sector, which is not large enough to bring about reduction in costs and boost technology R&D, may actually lead to adverse market situations and converse incentives. Moreover, incentives to the manufacturing sector should be extended to those aspects of the value chain where there is comparative advantage. As the deployment of solar projects picks up, the stimulus will automatically flow to the manufacturing segments. At this stage of development in the country, when there is more than adequate capacity globally to source technology at economical terms, it makes little sense to mandate domestic component in project development. Finally, excessive emphasis on MW-size projects may not be appropriate for faster and greater penetration of solar technology in India. Further, given the priorities of the government to eradicate energy poverty and provide energy access to all, and given the regional spread of solar resource availability in India, it would be more useful to promote smaller size off-grid and rooftop projects. This does not in any way imply that MW-size projects should not be promoted, but the focus should shift more towards the smaller-sized projects. This would not only unleash a large untapped demand and huge potential for SPV technology but in the process create a huge market for SPV technology. The German model where higher FiT is offered to smaller projects compared to larger projects could be looked into for its applicability to India. Such focus on smaller-sized projects would also help channel subsidy better, release subsidies due to migration from diesel-based captive generation and mitigate the large offtaker risks prevailing. As the grid gets extended, many of these smaller installations would be able to also provide energy to the grid, and with well-planned caps on in-firm capacity addition the problem of congestion and excessive financial liability could also be addressed. It is therefore clear that the gaps in the prevailing policy are few but quite critical for effectively unleashing the solar potential in India. The solutions are widely recognised and need to be prioritised so as not to lose the momentum created by JNNSM and state initiatives, and ensure that India emerges as a global destination for investment in solar energy. Sambit Basu India Solar Policy | 353

APPENDIX 1

Table 17.2: List of projects selected under migration scheme of JNNSM

Sr. Name of applicant State Capacity Whether no. of the solar PV plant or solar (MW) thermal 1 Maharashtra State Power Generation Co. Limited, (MAHAGENCO), Mumbai Maharashtra 4 Solar-PV 2 Clover Solar Pvt Ltd, Mumbai Maharashtra 2 Solar-PV 3 Videocon Industries Ltd, Mumbai Maharashtra 5 Solar-PV 4 Enterprises Business Solutions, USA Punjab 5 Solar-PV 5 Azure Power (Punjab) Pvt Ltd, Amritsar Punjab 2 Solar-PV 6 Acme Tele Power Limited, Gurgaon Rajasthan 10 Solar- Thermal 7 Comet Power Pvt Ltd, Mumbai Rajasthan 5 Solar-PV 8 Refex Refrigerants Limited, Chennai Rajasthan 5 Solar-PV 9 Aston Field Solar (Rajasthan) Pvt Ltd Rajasthan 5 Solar-PV 10 Dalmia Solar Power Limited, New Delhi Rajasthan 10 Solar- Thermal 11 Entegra Ltd, Ansal Bhawan, New Delhi Rajasthan 10 Solar- Thermal 12 Entegra Ltd, Ansal Bhawan, New Delhi Rajasthan 1 Solar-PV 13 AES Solar Energy Pvt Ltd, Gurgaon, Haryana Rajasthan 5 Solar-PV 14 Moser Baer Photo Voltaic Ltd, New Delhi Rajasthan 5 Solar-PV 15 OPG Energy Pvt Ltd, Chennai Tamil Nadu Rajasthan 5 Solar-PV 16 Swiss Park Vanijya Pvt Ltd Rajasthan 5 Solar-PV Total 84 Source: MNRE (2010). ‘List of Project Developers Qualified for Migration to Jawaharlal Nehru National Solar Mission’, Government of India, New Delhi, India. NVVNL signed MoU with the above 16 project developers to set up to 84 MW capacity solar power projects under migration scheme, comprising 54 MW capacity through SPV and balance 30 MW through STP. 354 | Indian Infrastructure: Evolving Perspectives Pradesh Rajasthan Rajasthan Anantapur Dist.: AP Pradesh District: Jodhpur, Tehsil:Phaldom Village:Amla 5 Jodhpur District, Phalodi Tehsil, Rawre Village type (MW) Gurgaon PV 5 Jaisalmer, Pokhran Rajasthan city project cap Table 17.3: List of projects selected under JNNSM for bundling scheme idder name Bidder’s Solar Proj Location State Saisudhir Energy Limited Hyderabad PV 5 T. Veerapuram, Rayadurg Taluk, Andhra B Camelot Enterprises Private Limited (Project Company : Firestone Trading Pvt Ltd.)Khaya Solar Projects Private LimitedDDE Renewable Energy LimitedElectromech Maritech Pvt Ltd GurgaonFinehope Allied Energy Private Limited PVVasavi Solar Power Pvt Ltd New Delhi FaridabadKarnataka Power Corporation Limited PV 5 PVNewton Solar Private Limited Pune BangaloreGreentech Power Private Limited 5 5Tehshil: Naguar, Vill: Mundwa Dist: Naguar, PVSaidham Overseas Private Limited Rajasthan PV HyderabadMahindra Solar One Private Limited Dist: Naguar, Tehshil: Khinvsar, Vill: Bhojas Mumbai Dist: Naguar, Tehshil: Khinvsar, Vill: Bhojas 5 Rajasthan Azure Powe (Rajasthan) Pvt Ltd PV Rajasthan Delhi New Ahmedabad 5 Gurgaon PVRithwik Projects Private Limited PV Mumbai PV Mandya, Malavalli, Belakavadi 5 PV Naguar, Tehshil: Khinvsar, Vill: Bhojas Dist: 5 5 PV 5 Rajasthan New Delhi Naguar, Khinvsar,Bhojas 5 Hyderabad Naguar, Tehshil: Khinvsar, Vill: Bhojas Dist: PV 5 Maharashtra Seamless Limited Kalhe Jodhpur, Phalodi,BAP PV Rajasthan Viraj Renewables Energy Private Limited Naguar, Khinvsar, Bhojas Karnataka 5 MumbaiNorthwest Energy Private Limited 5SunEdison Energy India Private Limited Naguar, Jayal, Kathali PV Electrical Manufacturing Co. Ltd Chennai Anantapur, Kadiri, Kutagulla MumbaiAlex Spectrum Radiation Private Limited Rajasthan Kolkata PV PV Kolkata Rajasthan 5 Rajasthan PV 5 PV Andhra Bikaner, Kolayat, Deh 5 Village-Kantia, District-Nagaur Rajasthan 5 Gajner, Kolkayat, Bikaner Maharashtra Allahabad, Naini Rajasthan Rajasthan Rajasthan Uttar Pradesh India Solar Policy | 355 Gujarat Tamil Nadu Vilalge Dwarka, Mojap Thummala Village Pradesh Tuticorin District, Kombukaranatham type (MW) Thermal 20 Jamnagar, Thermal 100 Bikaner, Kolayat, Ladkan Rajasthan city project cap Chennai PV 5 Kolkata PV 5 Jaisalmer, Pokhran, LanwaHyderabad ThermalNavi Mumbai 50 Anantapur, Pamidi, Virannapalle Rajasthan Pradesh Table 17.3: List of projects selected under JNNSM for bundling scheme (contd...) MNRE (2011). ‘JNNSM Phase-I Selected Projects List’, Government of India, New Delhi, India : idder name Bidder’s Solar Proj Location State B Godavari Power and Ispat Limited (Project Company: Godavari Green Indian Oil Corporation LimitedCoastal Projects LimitedWelspun Solar AP Private Limited New DelhiCCCL Infrastructure Limited PVAlex Solar Private Limited New Delhi PVPunj Lloyd Infrastructure Ltd 5 HyderabadAmrit Animation Pvt Ltd PV(Project Company : Amrit Energy Pvt Ltd) 5 Barmer, village MarudiOswal Wollen Mills Limited 5Precision Technik Private Limited Anantapur District, Amadgur Mandal, Kolkata GurgaonLanco Infratech Limited Chitradurga, Molakalmur, Murudi PV(Project Company: Diwakar Solar Projects PV Andhra Private Limited) Kolkata LudhianaKVK Energy Ventures Private Limited 5 5 PVMegha Engineering and Infrastructure Ltd PV Karnataka Hyderabad Rajasthan (Project Company: MEIL Green Power Ltd) Bap, Phalodi Jodhpur, Khurda, Khurda ThermalRajasthan Sun Technique Energy Pvt Ltd 100 5 5Aurum Renewable Energy Private Limited Jaisalmer, Nachana-1, Chinnu Mumbai Jaisalmer, Pokhran, Nokh Jodhpur, Phalodi,NatisaraEnergy Limited) HyderabadCorporate Ispat Alloys Limited Thermal 100Source Rajasthan Rajasthan Jaisalmer, Nachna, Chinnu Rajasthan Rajasthan Orissa Mumbai Thermal 50 Raipur Jaisalmer, Pokhran, Nokh Rajasthan Thermal 50 Jaisalmer, Parewar Rajasthan Rajasthan Andhra 356 | Indian Infrastructure: Evolving Perspectives

APPENDIX 2

Table 17.4: Allotment of solar capacities in Gujarat

Sr. no. Name of company Allotment, MW 1 AES Solar Company Pvt Ltd, USA 15 2 Astonfield Solar (Gujarat), USA 25 3 Azure Power Ltd, USA 15 4 Common Wealth Business Technologies, UK 10 5 Dreisatz GmbH, Germany 25 6 Environmental Systems Pvt Ltd, Mumbai 5 7 Euro Solar Ltd, Bhachau 5 8 JSW Energy, Mumbai 5 9 KRIBHCO, Surat 5 10 Lanco Solar Pvt Ltd, Hyderabad 35 11 Mi GmbH, Germany 25 12 Millennium Synergy Ltd, Bangalore 10 13 Moser Baer Ltd, NOIDA 15 14 PLG Power Ltd, Nasik 40 15 Precious Energy Ltd (Moser Baer), New Delhi 15 16 Solar Semiconductor Pvt Ltd, Hyderabad 20 17 Solitaire Energies Ltd (Moser Baer), New Delhi 15 18 Sunkon Energy Pvt Ltd, Surat 10 19 Tathith Energies USA 5 20 Top Sun Energy Ltd, Gandhinagar 5 21 Torrent Power Ltd, Ahmedabad 25 22 Unity Power Ltd (Videocon Group), Aurangabad 5 23 Waree Energies Ltd, Surat 20 24 Zeba Solar, Portugal 10 Grand total of photovoltaic solar thermal projects 365

India Solar Policy | 357

Table 17.4: Allotment of solar capacities in Gujarat (contd...)

Sr. no. Name of company Allotment, MW 1 ACME Telepower Ltd, Gurgaon 46 2 Adani Power Ltd, Ahmedabad 40 3 Cargo Motors, Delhi 25 4 Electrotherm Ltd, Ahmedabad 40 5 Abengoa Ltd, Spain 40 6 IDFC Delhi 10 7 KG Design Services Pvt Ltd, Coimbatore 10 8 Sun Borne Energy Technologies Gujarat L. 50 9 NTPC, New Delhi 50 10 Welspun Urja Ltd, Ahmedabad 40 Grand total of solar thermal 351 Grand total (photovoltaic + thermal) 716 Source: Government of Gujarat (2009). ‘Capacity allotment for solar power project development in Gujarat’, Gujarat, India. 358 | Indian Infrastructure: Evolving Perspectives

REFERENCES 1. ADB. 2011. Presentations from Solar Workshop & Training organized by ADB & NERL, New Delhi, India. 2. Ahuja, Dushyant. 2010. ‘Financing Solar Projects in India’, Energetica India, May/June 2010. 3. Arora, D. S., Sarah Busche, Shannon Cowlin, Tobias Engelmeier, Hanna Jaritz, Anelia Milbrandt and Shannon Wang. 2010. ‘Indian Renewable Energy Status Report: Background Report for DIREC 2010’, NREL, REN21, gtz, IRADe, India, October 2010 4. Deshmukh, Ranjit, Ranjit Bharvirkar, Ashwin Gambhir and Amol Phadke. 2011. ‘Analysis of International Policies in the Solar Electricity Sector: Lessons for India’, Prayas Energy Group, Pune, India, and Lawrence Berkeley National Laboratory, CA, USA; India, July 2011. 5. Deshmukh, Ranjit, Ashwin Gambhir and Girish Sant. 2011. ‘India’s Solar Mission: Procurement and Auctions’, Economic & Political Weekly, 46(28) July 9, 2011. 6. GERC. 2010. ‘Determination of Tariff for Procurement of Power by the Distribution Licensees and others from Solar Energy Projects’, Order No. 2 of 2010, Ahmedabad, Gujarat, India. 7. Government of Gujarat. 2009. ‘Capacity allotment for solar power project development in Gujarat’, Gujarat, India. 8. Government of Gujarat. 2009. Energy and Petrochemicals Department, ‘Solar Power Policy’, Gandhinagar, Gujarat, India. 9. Government of Rajasthan. 2010. Energy Department, ‘Rajasthan Solar Policy’, Rajasthan, India. 10. Green World Investor. 2010. ‘Can ADB Rescue India’s JNNSM with Loan Guarantees and Equity Investment?’, www.greenworldinvestor.com, 14 Dec 2010. 11. Green World Investor. 2010. ‘USA Opposes India’s Solar Energy Domestic Content Requirements’, www.greenworldinvestor.com, 15 Dec 2010. 12. Karnataka Government Secretariat. 2011. ‘Karnataka Solar Policy (2011–16)’, Bangalore, Karnataka, India. 13. MNRE. 2009. ‘Jawaharlal Nehru National Solar Mission: Towards Building SOLAR INDIA’, Government of India, New Delhi, India. http://india.gov.in/ allimpfrms/alldocs/15657.pdf India Solar Policy | 359

14. MNRE. 2010. ‘Guidelines for Migration of Existing Under Development Grid- connected Solar Projects from Existing Arrangements to the Jawaharlal Nehru National Solar Mission (JNNSM)’, Government of India, New Delhi, India. 15. MNRE. 2010. ‘List of Project Developers Qualified for Migration to Jawaharlal Nehru National Solar Mission’, Government of India, New Delhi, India. 16. MNRE. 2011. ‘Implementation of a Payment Security Scheme (PSS) for Grid- connected Solar Power projects under Phase I of Jawaharlal Nehru National Solar Mission (JNNSM) during the year 2011-12 Government of India’, New Delhi, India. 17. MNRE. 2011. ‘Jawaharlal Nehru National Solar Mission: Building Solar India— Guidelines for Selection of New Grid connected Solar Power Projects Batch II’, Government of India, New Delhi, India. 18. MNRE. 2011. ‘JNNSM Phase-I Selected Projects List’, Government of India, New Delhi, India. 19. MNRE. 2011. ‘Payment Security Mechanism for Grid-connected Solar Power Projects under Phase 1 of JNNSM’, Press Note, Government of India, New Delhi, India. 20. Peddada, S. Rao. 2010. ‘National Solar Mission & Solar Technology Deployment in India’, Presentation, New Delhi, India, October 2010. 21. Prabhu, Raj. 2011. ‘Focus on India’s Solar Policy Framework’, www.pv-magazine.com, 06 April 2011. 22. Raghavan, Shuba V, Anshu Bharadwaj, Anupam A Thatte, Santosh Harish, Kaveri K Iychettira, Rajalakshmi Perumal and Ganesh Nayak (2010). ‘Harnessing Solar Energy: Options for India’, CSTEP, Bangalore, India, 2010. 23. Stuart, Becky. 2011. ‘India Solar Industry off to a Successful 2011, but Bankability Issues Exist’, www.pv-magazine.com, 20 April 2011.

NOTE 1. Arora, D. S., Sarah Busche, Shannon Cowlin, Tobias Engelmeier, Hanna Jaritz, Anelia Milbrandt and Shannon Wang, ‘Indian Renewable Energy Status Report: Background Report for DIREC 2010’, NREL, REN21, gtz, IRADe, India, October 2010, p 37.

Telecom Sector Reform | 361

TELECOM SECTOR REFORM: Restructuring 18 Telecommunications as if the Future Mattered

December 1998

1. INTRODUCTION Although telecommunications was an early starter in attracting private sector participation,1 it seems to be caught in the same quagmire as the other early starter, power. Few projects have achieved financial closure. Investment in the sector has been below expectations, and there are strident and continuing calls to renegotiate contractual awards, supported by financial institutions that have sunk funds into projects with apparently over-optimistic expectations. Presently, the Telecom Regulatory Authority of India (TRAI), and the government are trying to develop an acceptable restructuring plan for the sector.

There are three basic options that seem to be under consideration, which are discussed below. They are a soft bailout of the existing operators (Rollover); a middle option that seeks to partially soften terms by instituting a revenue sharing system in place of the current license fees (‘Muddle’ path); and no bailout at all, except tariff realignments by TRAI (Hardball). All these plans, except perhaps Hardball, are designed more to relieve the problems of the existing operators and financial institutions, rather than put in place arrangements to provide efficient telecommunication services to the Indian user. Recently, however, the Prime Minister’s Advisory Group has reportedly proposed a radical restructuring plan that seeks to create a unified telecommunications market open to operators in all segments (Clean Slate). This note tries to evaluate these restructuring options against the technological trends in the telecommunications industry and their implications for the economics of the sector. 364 | Indian Infrastructure: Evolving Perspectives

2. OPTIONS FOR RESTRUCTURING Option 1: Rollover The most lenient (from the point of view of the financial institutions and the existing license holders) proposal on the table is to extend the license period of the operators and accept a moratorium on the payment of license fees to the government.2 The argument for this stems from the negative effect of opaque government policies and the assumption that bankrupt telecom companies automatically imply a non-viable and non-financeable telecom sector. While there have been problems with auction design, with obtaining clearances for spectrum rights and right of way, with interconnection to the Department of Telecommunications (DoT), and a host of other matters, it is debatable whether all of them together can be held collectively responsible for the current state of affairs. The second presumption ignores the distinction between individual companies and the sector as a whole. A number of airline companies have gone bankrupt in the US, which continues to have a vibrant and competitive airline industry. The fact that some operators would make unwarranted profits from a generalised sectoral relief is a relatively minor objection to this approach. Effect on other sectors: The major danger is the acceptance of renegotiation and alteration of original contract terms. Apart from being legally suspect, in a country where not only telecom, but the entire infrastructure sector is being opened up, the precedence for renegotiation set by this action would vitiate the environment for contracts and concessions in all other infrastructure sectors.3 This route should be avoided to prevent exacerbating already extant moral hazards. Option 2: ‘Muddle’ path The middle path advocates a revenue sharing approach instead of fixed license fees, which would make the government a partner in the commercial risks of market evolution. If anything, this is an absolute antithesis of the privatisation initiative. It is patently incongruous to speak of corporatising and privatising DoT and in the same breath ask the government to behave as a partner to existing private firms. Effect on market expansion: Philosophy aside, this distorts incentives dramatically. If for each additional rupee generated in revenue, an operator is to pay a portion to the government, then the incentive to grow the market is considerably reduced. The operator will restrict his market expansion to a point where his portion of the marginal revenue equals marginal cost.4 An example of a similar disincentive is the current metro cellular licence fee structure (a fixed amount per subscriber), which impels companies to increase the usage per user, rather than increasing the number of users, and is a clear deterrent to attracting low-volume users into the market, limiting the use of installed infrastructure. Telecom Sector Reform | 365

The current operators should wake up to the negative effect of revenue sharing arrangements on harvesting network externalities, as competition from other areas of telecommunications begins to affect their market share. The vested interest in a smaller market size that is fostered by the revenue sharing arrangements will also create an interest group that will try to resist the expansion of the sector, driven by the forces of technological convergence. The middle path may then end up in a very messy muddle. Government as insurer: It is obvious that if the government were to take a fixed fee that is equal to its revenue share, the operator would be driven to expand his market even more. It would seem the only reason the operators are asking for a revenue sharing system is the insurance value of such an arrangement against market risk.5 Should the government really be getting into the business of insuring operators against commercial market risk, when all the discussion is about increasing private sector participation in the insurance industry? There may be a case for this where the effect of commercial factors like price is limited, and demand is subject to extraneous risks but this is hardly the case in telecommunications, which is an extremely market-driven sector. Option 3: Hardball: This option insists on maintaining the contractually agreed license fee structure, and any relief that may be forthcoming would depend solely on the tariff and other decisions of the regulatory authority, TRAI. Given the revenue implications of the above two options and the fiscal pressure India is under today, there is understandable resistance to any option that will decrease revenue flows to the government. This, of course, is not a good reason to prevent action that may lead to better and wider provision of telecom service to the Indian user, but as argued above, the other two options do not have that advantage. Moreover, even within this option, there is scope for substantial relief to operators, and consequently financial institutions, through tariff changes by the regulatory authority, TRAI. Indeed their recent consultation paper suggests that the financial viability of existing operators is among its major concerns. Effect on investment: The other group who wish to stick to this option are those who believe that renegotiating contracts would have serious implications, by increasing the moral hazard in future contract negotiations in telecom and all other sectors. Their major fear is the effect of such a stance on the level of investment in the sector and consequently the level of service available to the final user. This fear is perhaps exaggerated. The present lack of investment is at least as much due to the presence of unresolved uncertainty as due to any lack of financial viability. In any case, given that the existing investment is, to a considerable extent, sunk, future investment will depend on adequate cash flows from the sector. Infusion of capital by new equity partners and buyout of existing equity by parties better able to run 366 | Indian Infrastructure: Evolving Perspectives the business can be expected to ease liquidity constraints on existing firms because of past mistakes. A certain amount of rescheduling and write-down of extant debt commitments cannot be ruled out, but this should not be a concern, unless the implied increase in NPAs affects the overall safety of the financial sector. Digital convergence: Even if the amount of investment is not a problem, the composition of investment may very well be. Adopting the hardball option freezes the existing structure of the industry. The current licenses are simply for the right to provide a service, entrenching exclusivity and perpetuating an archaic monopolistic structure in a sector that is becoming increasingly competitive. The basic driving force of this growing competition in what was once thought to be a natural monopoly is the increasing versatility with which services can be provided, based on the digitisation of all signal transfer technology. As the manner in which signals are transferred from one location to another becomes common, it is possible for a service provider in one segment of telecommunication, say network television services, to perform the functions of another, say, the local phone company. Efforts to maintain barriers across such segments will eventually be overwhelmed by technology.6 There is a need to take action now and restructure the sector in line with its evolving technology and associated economics.

3. STARTING OVER Clean slate: The Prime Minister’s Advisory Council has reportedly suggested a complete makeover of the sector, with open competition rules across different segments of the telecommunications industry. Unfortunately, perhaps as a hangover from the old debate, they also suggest revenue sharing as a means to deal with the stranded costs of existing concessions, but that issue can be dealt with in other ways, as suggested below. Where does the present path lead? What prevents a restructuring such as the one suggested above? The fear is that such a move will end up delaying the development of the sector, ignoring the fact that trying to sort out the current mess has already taken enough time. The critical point to note is that any development of the sector on the present path is very likely to lead to a distorted and inefficient sector, by the very nature of the current dispensation. In addition, the restarting time for the sector should be low, given the amount of information released as a result of the travails of the existing operators and the growing clarity on technology, regulatory environment and the possible extent of competition from government-owned bodies like MTNL and VSNL. If anything, the environment for private sector participation may now be more propitious. Reach or revenue? It would seem that the first issue that a new telecom policy should address is whether to increase teledensity or to continue using telecom as a revenue generator. Usually, in a public service activity, there is a dichotomy between the provision Telecom Sector Reform | 367 of universal service and the profits made from the activity. Mandates to increase reach therefore result in lower profits to the operator and consequently, lower revenue to the licensing authority. But, in telecom, this picture is no longer as clear. First, the very nature of the technology of minimal marginal cost and lumpy capacity dictates that the capacity be used to its utmost, once installed. The nature of any subsidy can then be limited to reducing the cost of capacity required to serve target customers. The amount of support required will be reduced to the extent that this capacity can also be used to serve other ‘profitable’ customers.7 In addition, there are benefits from network externalities as reach expands. ‘Reach’ is therefore something that is inherent in the technology and economics of the telecommunication industry. It is not something that needs to be mandated. Instead, the government needs to use this characteristic to its advantage, by leveraging connectivity between different networks. Impact on technology: Given India’s fiscal situation and the revenue raising possibilities of telecom licenses, it is difficult to abjure the revenue maximising option, even in light of the experience of the past few years. Indeed, few countries have been able to resist this allure. There is however increasing doubt about the willingness of firms to pay simply for the right to provide telephone services in a technologically evolving environment that threatens the monopolistic nature of the right. A path that chooses to auction such rights, such as India’s, would also imply that the government would be under pressure to prevent technological advancement to protect the right granted to the operators, with possibly disastrous consequences for India’s burgeoning information technology sector. Revenue out of thin air: Developments in communication technology have two broad trends: versatility, mentioned earlier, and mobility. Mobility requires use of the radio frequency spectrum. This is a scarce resource with competing uses, and now increasingly subject to international standardisation by the International Telecommunications Union (ITU). It is possible and indeed appropriate to raise revenue by auctioning rights to the use of this spectrum, as has been done successfully in the United States and Australia, through simultaneous menu ascending auctions.8 Given the fact that the valuation of these spectrum rights is likely to undergo substantial change over time, as the Indian economy develops, it is advisable to award such usage rights for relatively short periods, e.g. 15 years, without restricting the service that they can be used for. For this, it is necessary to strengthen the office of the Wireless Adviser and develop a spectrum allocation plan after considering ITU’s recommendations before embarking on an auction of spectrum rights. Conventional wireline (whether copper or fibre-optic) telephone services do not involve the use of scarce common resources, except, on occasion, the right of way,9 and should be de-licensed and open to anyone who can provide the service, subject to quality standards. 368 | Indian Infrastructure: Evolving Perspectives

Dealing with existing licensees: The removal of restrictions on service provision implies that existing licensees can potentially gain by being able to provide a wider variety of services, thus increasing the value of the license, in addition to conferring a first mover advantage. Recent remarks from foreign equity holders in telecom consortia, e.g. Telestra, seem to indicate that this will, to a large extent, mitigate any need for compensation. Allowing existing licensees to bid for the new concessions could further reduce their resistance. Their commitments for future license payments, made under expectations of less competition, could be converted into re-saleable bidding points for the proposed spectrum auctions.10 Operators would still be liable for any license payments not converted and any unsold and unused bidding points. This will ensure that the government does not have any additional financial commitment due to the restructuring, and also, at a minimum, protect the existing revenue projections from the sector, which would be doubtful if the current system were to continue. Wiring: As mentioned earlier, the development of fixed-line networks will continue to suffer from problems related to acquisition of right of way. This needs to be addressed to enable fibre-optic cable networks to supplement the ultimately limited capacity of the radio spectrum. This is especially needed in high population and data density areas such as urban areas and business districts, where wireless technology is unable to provide the required channel capacity, given the volume of calls. The high and versatile data carrying capacity of fibre-optic networks means that they will also be ideal network resources to be re-sold to multiple service providers, who can be cable operators, broadcasters, telephone operators, internet service providers, or any other company that needs to send digital signals into the connected units. Interconnection: This is essential to promote and leverage the technological trend towards versatility. The new telecommunications policy needs to establish clear open access and interconnection guidelines. Along with opening up access of one telecom segment to operators in other segments and vice versa, this would lead to development of a data communications market that would optimise utilisation of both installed wireline and wireless transmission capacity. The presence of many operators and a technological and regulatory environment that permits the interconnection of one network to another would also lead to a quantum expansion in reach. For example, it is conceivable that a satellite phone operator like Iridium would resell its spare satellite capacity, which is a sunk cost, to operators who will use it to provide trunk connectivity to local rural networks, at a fraction of the cost needed for dedicated wireline trunk connectivity. Extensive interconnection would also lead to increased benefits from internalising network externalities. It is not necessary to trade-off reach in order to raise revenue.11 Telecom Sector Reform | 369

4. CONCLUSION The government is on the verge of announcing a new telecom policy. The earlier policy was designed in a world where telecommunications services were expected to be provided by monopolistic entities that faced little competition. However, efforts to involve the private sector under that regime have led to non-viable contracts, which today ask to be restructured. The opportunity exists now to remedy our earlier mistakes and develop a forward-looking policy that acknowledges and leverages the technological developments and economic characteristics of the sector in a manner that will serve the Indian user in the best possible manner. We can use it to build a sector that will grow with the country and make India’s information technology sector a part of the global communications revolution. It will establish a sector that is at the cutting-edge of technology, as judged by its effect on industrial competitiveness, and also provide affordable connectivity to our vast millions. Indeed, it is only through such technology that we can provide such connectivity. Or we can continue with our current fragmented, costly and city-centric company- and institution-oriented approach, which will lead to a sector that will need restructuring once again, as technological developments overwhelm our attempts to rescue current companies and institutions. We should not commit the same mistake yet another time. It is time for the government to de-license the provision of wireline telecommunications and allow operators in one segment to provide services in other segments. It should auction spectrum rights to ensure the best use of a scarce resource. Existing licensees can be allowed to convert future license fee commitments into re- saleable bidding points for these auctions, which will at least preserve expected government revenues. In preparation, the government will need to develop a spectrum allocation plan, mandate interconnection requirements for all segments of the telecommunications sector, and facilitate right of way for fibre-optic networks.

NOTES 1. Metro cellular licenses were sold in 1994; other cellular areas in 1995 and basic service rights were sold in 1996. Revenue maximisation appears to have been the primary objective of the government in auctioning licenses for the provision of telecom services. A sealed bid auction implied that all winners were subject to the ‘winner’s curse’, and bidders for the second license were required to match the winning bid. This left no natural alternative to step in, in case of default by the winning bidder. 2. A remedy tailored to the extent of difficulty would imply individualised packages for each operator. This will involve very intensive and burdensome financial monitoring, or some form of windfall profits tax. It would also imply that a policy decision has to be taken on an appropriate profit level for telecom operators, similar to a ‘guaranteed’ 370 | Indian Infrastructure: Evolving Perspectives

return. Otherwise, some operators would receive unwarranted benefits from a generalised relief. 3. It is interesting to note that the basic-case law in this area appears to be Ramanna Shetty versus International Airports Authority of India (IAAI), involving the award of a restaurant concession, utterly unrelated to the telecom sector. It is often stated that the telecom industry cannot afford delayed and protracted litigation. This has to be judged against implications of opening up renegotiations, not just for telecom, but also for other sectors as well. 4. This is a well known result in land-tenancy theory, where extensive comparisons were made in the sixties and seventies among crop sharing, cost cum crop sharing and fixed- rent land tenancy systems. 5. The incentives to adhere to such a revenue sharing system for the private sector will decline with decrease in uncertainty and the growth of the sector. As the amount that has to be given to the government increases, the incentives to under-report will rise. In effect, the government will on all likelihood be unable to reap the full benefits of the ‘partnership’, as private parties will clamour for release from the burden of government payments in the future. 6. The growth of callback services in regions like Latin America, to arbitrage differences in phone tariffs is a simple case in point. Teleintar, the Argentine international telephony operator, estimates that it lost about 30 per cent of its market share to such services. 7. Even more strongly, if the capacity in question is required to serve other customers, then no subsidy is indicated. For example, services to rural areas along the fibre-optic cable route or relay station route (depending on the technology used) linking major urban areas service provider can be provided at no additional cost. Indeed the extra revenue from rural users, if they are within the provider’s license area is an additional benefit. Also see note 11. 8. The use of simultaneous ascending menu auctions in Australia and the US is generally believed to have led to a realistic revelation of underlying valuations for the spectrum in a transparent manner. The ascending auction was designed to reveal information private to specific bidders, and minimise regret, while the menu structure allowed bidders to aggregate regions based on their business complementarity, rather than an imposed division. 9. Currently, the electricity and cable (mostly in urban areas) are the only two industries that have wired access into homes. Of these, the cable operators often usurp right of way without appropriate payment. 10. While the government shares some responsibility for the current chaos in the telecom sector, blame must also be attached to the inefficient execution of business plans by the operators and faulty business projections made by them and their financiers, and their subsequent effort to avoid the consequences of these actions. This has stunted growth in this sector over the last few years. Such behaviour should be penalised in order to maintain commercial discipline. One way of doing this would be to award only a fraction of the license fee commitments as bidding points to the existing operators. Telecom Sector Reform | 371

11. If universal service provision is not occurring at a rapid enough pace, it can be supplemented through a system of minimum subsidy bidding for the provision of a defined level of service to a specific area. These subsidies can be funded out of general revenues or from a Universal Telecom Service Fund established for the purpose. Leaving universal service obligations with the dominant government-owned operator (especially if it remains in hands) would furnish no incentive to explore cost- minimising solutions to the provision of such service and provide a continuing excuse for underperformance. Interestingly, Oftel, the British telecom regulator has determined that the benefits of universal service provision, in terms of locking in potentially profitable customers, better access to first-time users and wider brand recognition were worth more than the estimated cost incurred by British Telecom in providing the service, around £45m–£80m per annum. 372 | Indian Infrastructure: Evolving Perspectives

TRANSITIONING FROM ADMINISTRATIVE ALLOCATION OF SPECTRUM 19 TO A MARKET-BASED APPROACH February 2004

1. INTRODUCTION The Indian telecommunications sector has seen much growth and turmoil in the recent past.1 The number of mobile subscribers grew from 0.3 million in the first quarter of 1997 to 29 million in December 2003, of which 7 million were CDMA mobile subscribers.2 In November 2003, the government issued guidelines for a unified access services licence (UASL) regime, where the basic service (BS) licence and the cellular mobile telephony service (CMTS) licences would come under a single licence regime in the future. Existing licensees can opt to continue offering services under their old licence, though many have chosen to migrate.3 The government has also permitted mergers and acquisitions (M&A) of licensees within a service area, and the merged entity is now permitted to retain the spectrum originally allocated to the merging entities up to certain limits, but ‘the spectrum utilisation charges beyond 10 + 10 MHz for GSM-based system and 5 + 5 MHz for CDMA/ETDMA-based systems shall be prescribed separately’.4 Recently, the Telecom Regulatory Authority of India (TRAI) has been entrusted with the regulation of cable broadcasters—ushering in signs of regulatory convergence. These developments appear to presage a situation where the service licence would have no entry restrictions beyond some minimal level of pre-qualification. In this scenario, usage rights to the spectrum acts like an entry restriction. The multiple uses of the spectrum in a convergent environment and the increasing relevance of industry consolidation require that serious consideration be given to separating the rights to use the spectrum from service licences and thinking about alternative methods of allocating spectrum. This paper explores how just such a transition from the existing situation can be accomplished. While this note focuses on telecommunication services, the principles apply equally to all other services that use radio frequency, e.g. FM radio,5 paging and trunk radio. From Allocation of Spectrum to a Market-based Approach | 373

2. EXISTING PRACTICE 2.1 Allocation and fees for spectrum usage The Wireless Planning and Coordination wing (WPC) currently assigns frequencies to CMTS licensees from the designated bands prescribed in National Frequency Allocation Plan–2000 (NFAP–2000). A cumulative maximum of up to 4.4 MHz + 4.4 MHz for CMTS licensees is assured, but based on usage, justification and availability, additional spectrum up to 1.8 MHz + 1.8 MHz, making a total of 6.2 MHz + 6.2 MHz, ‘may be considered for assignment, on case-by-case basis, on payment of additional licence fee’.6 The CMTS licensees pay spectrum charges of 2% of the adjusted gross revenue (AGR) up to 4.4 MHz and 3% of AGR for spectrum up to 6.2 MHz. The charge rises to 4% of AGR for allocation beyond 6.2 MHz + 6.2 MHz, which shall be given if the subscriber base is more than 5 lakh. ‘This spectrum charge of 4% of AGR would also cover allocation of further spectrum, which may become possible to allocate in the future, subject to availability, to add up to a total spectrum allocation not exceeding 10 MHz + 10 MHz per operator in a service area. Such additional allocation could be considered only after a suitable subscriber base as may be prescribed, is reached.’7 Thus, the expenditure on spectrum increases only proportionately to revenue for spectrum allocations beyond 6.2 + 6.2 MHz. In addition, a major part of the one-time entry fee paid by the licensee, based on a bidding process, can also be considered as an up front payment for spectrum. Similarly, in the erstwhile basic services licence, an additional revenue share of 2% of annual gross revenue earned from WLL subscribers would be levied as spectrum charge for allocation of 5 + 5 MHz in the 824–844 MHz band paired with the 869–889 MHz. As in the cellular case, this includes royalty for spectrum as well as the licence fee for the base station and subscriber terminal (handheld or fixed). The same principle is followed for spectrum charges in the 1880–1900 MHz band for the microcellular technology-based system. Further, royalty for the use of spectrum for point-to-point links and access links (other than cellular service spectrum) are separately payable as per WPC guidelines. The authorisation of frequencies for setting up microwave links by cellular operators and issue of licences is also separately dealt with by the WPC as per existing rules.

2.2 Revenue sharing—an old practice The practice of charging a share of revenue as user fee for scarce resources is not new. Indeed, it can be said to be similar to the practice of sharecropping, whereby the landlord allows a peasant to cultivate his land in return for a share of the produce. The literature analysing this phenomenon has pointed out two main features of this type of contract, namely: 374 | Indian Infrastructure: Evolving Perspectives

(a) its risk-sharing capability, whereby the peasant is protected in case there is a crop failure, but does not enjoy the full benefits of his investment in case of a bumper crop; and (b) the dilution of incentives to exert effort and invest in the land precisely because he does not enjoy the full benefits of his investment. The use of revenue share for charging for spectrum is a relatively straightforward extension of this type of contract. But, even though revenue sharing linked spectrum fee to service provision, it was unclear, until recently, whether spectrum was an integral part of the licence (see Box 19.1).

Box 19.1: The spectrum is finally attached to the licence

If the licence were to be sold by an existing service provider to another service provider in the same service area (as is now permitted), would the buyer have access to 12.4 MHz + 12.4 MHz of spectrum8 or would she/he have to give up the seller’s allocation? This issue has only been clarified in the recent guidelines (issued on 21 February 2004) for merger of licences in a service area. The current status is that the spectrum will go with the licence but the amount held by a merged entity shall not exceed 15 MHz + 15 MHz per operator per service area for metros and category ‘A’ circles and 12.4 MHz + 12.4 MHz per operator per service area in category ‘B’ and ‘C’ circles. This will be an important instrument to rationalise the use of spectrum. A concern from the point of regulation is possible hoarding by operators. Indeed, the UK telecom regulator, Ofcom ‘considers it desirable to address acquisition of market power as well as abuse of dominance’ with reference to spectrum. The Indian guidelines are relatively generous from this viewpoint, permitting a market share of up to 67% of the combined GSM and CDMA subscriber base.

2.2.1 Effect on investment It is arguable that the benefits of risk sharing may be substantial in an emerging economy such as India, where the natural and policy variability in the business environment is high, especially for sectors such as telecom where technological change is rapid, and the ability of individual service providers to bear the risk is low compared to the national government. Moreover, since the share of revenue is relatively low, viz. 2–4% (as compared to 33% to 50% in crop sharing arrangements), the second effect, i.e. investment distortion may not be substantial. Overall, the net effect on investment may well be positive. In any event, the revenue share towards licence fee (which is by far a larger share of revenue) is expected to remain, and it would be difficult to redress any investment distortion effect by addressing spectrum charges alone. From Allocation of Spectrum to a Market-based Approach | 375

2.2.2 Effect on competition It can be argued that this type of user fee structure leads to a situation where the service provider who is using spectrum more efficiently (i.e. has more users per unit of spectrum) is charged more (because she has more revenue, she pays a higher absolute amount, given equal revenue shares) than another service provider who has fewer users per unit of spectrum. Thus, even though both service providers use the same amount of spectrum, the one with fewer users and arguably the less efficient provider pays less.9 However, another way of looking at this is to observe that this system promotes competition by facilitating entry because newer service providers are allocated the same amount of spectrum, but since they have fewer users initially, they have a certain advantage in their initial period, where they can survive with lower investments. Thus, while it may delay exit by inefficient providers, it may foster entry by new providers. The net effect on competition is therefore unclear. However, since it is unlikely that the benefits from spectrum charges alone could help keep an inefficient operator afloat, the balance may well be in favour of more competition.

2.3 What regime for spectrum under a unified licence? The questions that need to be addressed are as follows: (a) Should we delink the spectrum from service provision? (b) Should we change the way we charge for spectrum, i.e. move away from revenue share to a charge that is based on the amount of spectrum used? This is related to (a) above, for if spectrum is delinked from service provision, it would be difficult to implement a revenue sharing regime and alternative fee structures may need to be found. (c) Should we want to allow trading of spectrum bandwidth? (d) Should we have spectrum auctions? Is that the only way to have market-based charges?

3. WHY FIX WHAT IS NOT BROKEN? Should one again modify the licences to move to a spectrum fee regime that is not based on revenue share? At one level, it can be argued that since there is by now an established tradition of modifying licence conditions in India, this could be considered a relatively minor adjustment as long as the revenue implications for the service providers and therefore the government are not substantial, i.e. the financial terms of the contract are broadly similar. On the other hand, can we not wait for the existing licences to expire?10 If the revenue implications are not substantial, then why tamper with the existing system, unless it is severely distortionary? 376 | Indian Infrastructure: Evolving Perspectives

3.1 Revenue sharing and delinking The critical benefit of moving away from a revenue share regime lies in delinking the spectrum from service provision. Delinking these two would imply a fee for spectrum that is unrelated to revenue from service provision. As long as the revenue share system remains, service provision and spectrum remain linked since payment for the spectrum is based on revenues from service provision. The benefits from convergence lie in the ability to provide multiple services over a single ‘pipe’, of providing new services in hitherto unknown ways. Delinking the spectrum allows the possibility that the bandwidth may be used for services that may be more useful than what it was originally contemplated for. Its absence prevents other potentially more valuable users of the spectrum from making an offer to the existing user to use the spectrum in a mutually beneficial manner. Free entry of service providers can create additional services where none existed.11 More importantly, it will be possible for spectrum prices to reflect relative scarcity, e.g. spectrum could be very inexpensive and in relative excess supply over vast parts of rural India, thereby fostering the spread of wireless connectivity, while it would be expensive in major metros, thereby fostering the use of spectrum-saving devices. This could also be achieved if the licence areas were more finely defined, but it would be more cumbersome to implement.

3.2 Revenue sharing and demand for spectrum 3.2.1 Existing charging system can increase demand for bandwidth At first sight, the current revenue sharing regime appears to be a relatively steep charge for additional spectrum, especially for companies that are growing their revenues. The additional 1.8 MHz raises fees from 2% to 3% over the entire revenue base, i.e. a 50% increase in charges. Assuming a gross revenue of Rs 1000 crore, which is a reasonable number for a metro cellular operator today, this implies that the additional 1.8 MHz would cost Rs 10 crore (the company would have to pay Rs 30 crore instead of Rs 20 crore). More so, the company would have to pay this higher share over the entire future of its licence.12 This would need to be compared with the additional investment required in order to meet the growth in the number of subscribers with the existing spectrum. It may not be possible to technically accommodate such numbers without additional spectrum, and therefore, savings may only be to the extent to which the acquisition of additional spectrum can be deferred. However, it would appear that the current regime therefore leans towards motivating an operator to acquire more spectrum.

3.2.2 But is bandwidth scarce? The above would be especially undesirable if the amount of spectrum available in India was limited. It is not clear that this is yet the case. The amount of spectrum From Allocation of Spectrum to a Market-based Approach | 377 currently released to telecom service providers is limited compared to international allocations (see Tables 19.4 and 19.5 in Appendix, which compare allocations in India with a number of countries overseas). There also appears to be spectrum available with other non-telecom users, that can be reallocated in a mutually beneficial manner. The development of spread-spectrum and code-division methods for transmission—the next generation of both the existing GSM and CDMA are systems based on such technologies—could further reduce the demand for spectrum, though wireless data transmission may provide a counterbalancing demand. Spectrum may, however, be scarce where there is a high density of users, such as the metropolises and here the existing regime may provide inappropriate incentives to acquire more spectrum in these areas.13

3.3 Use of the allocated spectrum The existing pattern of use of spectrum indicates that there is considerable variation in the intensity of use by operators in the same circle, indicating that even if a particular operator is pressed for spectrum, the spectrum allocated for the service as a whole remains underutilised. Allocating more spectrum in this situation would not be necessary if some method could be found to transfer allocations between service providers. This indicates that there may be an opportunity for trading. Figure 19.1 shows the ratio of minimum to maximum number of users per unit of spectrum across twenty different telecom circles where more than one operator is present. In a number of circles, this seems to vary between 0.25 and 0.35, i.e. the operator with the minimum number of users per MHz has only a fourth or a third of the number of users as the operator with the maximum number of users per MHz in the same circle.14 It is conceivable that the operator who is using the spectrum less intensively would like to trade it with one who is using the spectrum more intensively, subject to the strategic considerations mentioned below. However, since operators with fewer subscribers pay relatively little for the spectrum as long as it is charged as a revenue share (a lower subscriber base implies a low AGR and a low spectrum charge), the opportunity cost of holding on to the spectrum is low. A revenue- sharing regime may therefore act as a hindrance to efficient transfer of spectrum among operators and indeed, in the extreme, force mergers between operators where only trading may have sufficed. 378 | Indian Infrastructure: Evolving Perspectives

1.00 0.93 5 0.90 0.820.84 0.80 4 0.70 0.60 0.61 0.60 3 0.50 0.45 0.40 0.35 0.38 0.35 2 0.32 0.34 0.32 0.30 0.25 0.25 0.23 0.25 Number of operators 0.190.22 0.20 0.17 1 0.10 0.06 Ratio of min to max number user per MHz 0.00 0

Delhi Bihar Kerala Punjab Orissa MumbaiChennai Gujarat HaryanaKolkata UP-East Karnataka UP-WestRajasthan Maharashtra Tamil Nadu Andhra Pradesh Madhya Pradesh Himachal Pradesh Min–max ratio West Bengal Number& A&N of operators

Figure 19.1: Variations in use of spectrum across circles and operators There are thus two reasons for moving beyond the existing system, namely: (a) Optimising the use of spectrum across different uses, i.e. where the spectrum is used for services that may be more useful than for what it was originally allocated. (b) Optimising the use of spectrum within a given use, i.e. avoiding the additional release of spectrum while spectrum allocated for the service as a whole remains underutilised. Both these benefits would need a mechanism to allow for trading of spectrum.

4. SPECTRUM TRADING 4.1 Spectrum trading is a stand-alone decision Trading in spectrum can in principle be permitted regardless of whether the spectrum is delinked from the licence. If the spectrum continues to be linked to the licence, it will be possible to have change of ownership, but not a change of use. To that extent, therefore, allowing spectrum trading is independent of delinking spectrum and licence. Trading will ease pressure on the spectrum to the extent that there is variation in spectrum usage across regions and across operators. In this scenario, the service provider who has excess spectrum in some region can lease out her/his spectrum for a specified duration to another service provider, who would then save on the capital expenditure needed to make more intensive use of her/his existing spectrum. From Allocation of Spectrum to a Market-based Approach | 379

4.2 Strategic considerations and trading It is, however, possible that commercial consideration may militate against an active trade in spectrum.1 Telecom is a business with strong network externalities, and operators may wish to hold spectrum in view of future roll-out plans. In particular, operators may not lease spectrum to competitors, who would then be able to increase their customer base and improve service quality (at a cost lower than they would have to spend otherwise), which could act as a hurdle in the future expansion of the spectrum lessor. However, spectrum trading can lead to significant activity in other industries, as in the case of the New Zealand radio broadcasting industry (see note 11). Notwithstanding the above, differing regional strategies or tactical considerations may still result in trades while necessary adjustments are made.

4.2.1 Pricing for new spectrum and trading The availability of additional spectrum to service providers who have reached certain threshold norms (as is the current practice) may also induce competing service providers to lease out unused spectrum. If she/he refuses to lease the spectrum, the competing service provider would procure the spectrum from the government anyway. By leasing the spectrum, the lessor gains the revenue that would otherwise have gone to the government. The manner of pricing adopted by the government for additional spectrum would then affect the secondary market price. 4.3 International experience with spectrum trading The extent of spectrum trading even internationally is still limited as shown in Table 19.1 below. In the US, Nextel had a difficult time aggregating spectrum across the continent. However, countries such as Australia and New Zealand have implemented spectrum trading for many years now. Spectrum trading is also permitted in countries like Guatemala! Equally importantly, many countries are now considering spectrum trading as wireless connectivity rises across the board. Both the UK and the EU have initiated consultation on the subject. Table19.1: International experience in spectrum trading Factors Australia/ United States/ UK Rest of New Zealand Canada Europe Spectrum scarcity Low High Medium Medium/low International co-ordination requirements Low Medium Medium High Regulatory restrictions Low Medium High High Political attitude to market-based mechanisms Positive Positive Positive Mixed Implementation of Fully Partially Likely to be Under spectrum trading implemented implemented introduced consideration in many countries Source: Nagpal, Amit: ‘One trade at a time: phased implementation of spectrum trading’. 380 | Indian Infrastructure: Evolving Perspectives

4.3.1 Trading and limitations on use of spectrum One key question that needs to be addressed with respect to spectrum trading is the limitations which can be put on the use of the spectrum. In considering trading, some countries are looking at limits on the type of trading, e.g. change in ownership may be permitted, but change in use may be restricted. International conventions define certain uses for specific bandwidths of spectrum, e.g. one band for cellular mobile and another for broadcasting. It is conceivable that any permitted change in the use of spectrum may need to respect these conventions, which would limit the efficiency gains from spectrum trading. Further restrictions may emerge from technical considerations of preventing interference in co-located spectrum bands from dissimilar uses with different protection technologies. Most regulators do not expect a large number of trades at this time if only change of ownership is allowed.

4.3.2 Spectrum register There will need to be significant institutional development before spectrum can be traded. However, India has recently seen the development of a modern stock trading exchange, viz. the NSE, as also a few commodity exchanges, and it can therefore be expected that the development of trading infrastructure will not be a constraint, provided a system of definition of property rights and one of maintaining a spectrum register is evolved. The spectrum register would contain information necessary to execute the trade. Table 19.6 in Appendix provides an example from Ofcom in the UK about the kind of information that the spectrum register may be expected to contain. The property rights,2 discussed in more detail below, are defined by the geographic and frequency boundaries and the price paid for using the spectrum (items highlighted in Table 19.6 in Appendix).

4.4 The building blocks for spectrum trading 4.4.1 Standard spectrum trading units In order to facilitate spectrum trading, it is useful to develop a standard measure of spectrum, like a stock holding certificate. The entire commercial spectrum to be opened for trading (both those already in use and the bands yet to be allotted) could be divided into standard spectrum units (SSUs), e.g. 20 sq km MHz as a basic unit of measurement—similar to a square metre of land, as shown in Figure 19.2. The unit need not be a cube; it can also be a cylinder as shown, where the distance is measured radially from a given geographical (latitude and longitude) grid reference. From Allocation of Spectrum to a Market-based Approach | 381

10 km 5 km radius

0.2 0.2 MHz MHz

10 km

Figure 19.2: Example of standard spectrum unit

4.4.2Dimensions of SSUs In defining SSUs, the specific dimensions of both the spectrum and geographic elements need some deliberation. The example given here relies upon the channel spacing for GSM. An existing CMTS service provider with 4.4 MHz of spectrum allocation therefore has 22 SSU per 100 sq km of service area. However, the channel spacing for CDMA is different. The spectrum dimension needs to be of sufficient fineness so that different units can be aggregated into a meaningful bandwidth for trading, e.g. with a 200 kHz unit, 8 such units can be aggregated to form 1.6 MHz, which is the channel spacing for TD-SCDMA, but the SSU in the example above cannot be used to separate 1.25 MHz, which is the CDMA channel spacing. Similarly, the geographical grid of 100 sq km (or 78.5 sq km in case of radial distances) may be inappropriate, even for urban areas. Indeed, it is expected that the size of the geographical grid would vary with population density, e.g. as in Australia (see Box 19.2), with larger grids in less densely populated areas. It is also possible to allow subdivision finer than SSUs. In the UK, the Ofcom consultation paper indicates that it is considering allowing traders to define the slices of spectrum for trading.

Box 19.2: Defining spectrum trading units in Australia

Australia sub-divided a given band into three-dimensional blocks, defined geographically by parallels of latitude and meridians of longitude and by a standard bandwidth in frequency. Boundary conditions were set in terms of interference levels, 382 | Indian Infrastructure: Evolving Perspectives

with ownership of blocks recorded in a computer database. The smallest indivisible unit of spectrum space is called the standard trading unit (STU).3 Each spectrum licence (valid for 15 years, after which it reverts to the government, presumably to be auctioned again) will consist of many STUs, i.e. indivisible cubes of spectrum space within the spectrum licence. The frequency bandwidth of STUs may vary in size depending on the spectrum band in which licences are being issued, but the area grid is constant for all bands. The Spectrum Management Authority created a Spectrum Map Grid covering the entire country, resulting in 21,998 cells. The cells are of three sizes depending on population density, from 5 minutes of arc (about 9 kilometres) on the eastern seaboard and in Adelaide, Perth and Darwin to 1 degree of arc (about 100 kilometres) in regional Australia and 3 degrees of arc (about 400 kilometres) in remote Australia. By themselves, the STUs may be too small to have significant utility, but because of their regular shape and their referential relationship with their immediate neighbours, they can be stacked vertically or horizontally with neighbouring STUs to form larger bodies of spectrum space that do have utility. Spectrum licences can be traded only in terms of whole standard trading units, or STUs. Licensees who wish to trade part of a licence can disaggregate the licence into its component STUs and sell them individually or in multiples. A spectrum licence can be traded in whole or in part, by geography (see A) or by bandwidth (see B) or by both geography and bandwidth (see C), or can be leased in whole or in part to third parties. A licensee can also look to extend the geographic coverage and/or the bandwidth of a licence by acquiring an adjacent spectrum licence from another licensee (see D). Source: Australian Communications Authority

4.4.3Transition to trading The transition of existing spectrum allocations to tradable SSUs may not pose significant challenges beyond those of defining property rights over the radio frequency, which can be accomplished by redefining the existing allocations in terms of SSUs, as described above. Service providers can then be allowed to trade their allocation based on their business perception of current and future need of spectrum. In the UK, Ofcom is proposing to transact trades by cancelling or amending the seller’s licence, and reissuing a licence to the purchaser in a six-step process, which is described in Box 19.3. This may not be necessary if the trades are in whole SSUs, where the partitioning of the spectrum assigned to the licensee is pre-defined, and the trade can be accomplished much in the manner in which a dematerialised stock is traded today.

Box 19.3: Ofcom’s proposed process for transacting a spectrum trade 1. The licensee decides what rights it wants to transfer (e.g. by an outright sale or lease). If the proposed transfer would involve a change of configuration From Allocation of Spectrum to a Market-based Approach | 383

(e.g. a partitioning of the spectrum assigned to the licensee) or a change of use, which the licensee wishes to effect in advance of the trade, the licensee can apply to Ofcom first. 2. The parties to the trade agree to the terms of the transfer. Under the trading regulations, Ofcom proposes to require that the terms of the transfer must be set out in a written contract signed by all parties. The terms of the contract may be as simple or as complex as the trade requires. In drawing up the contract, Ofcom expects that the parties will wish to conduct the appropriate due diligence and obtain the appropriate representations and warranties. It is proposed that under the trading regulations, all licence obligations (other than non-spectrum related licence conditions) must be transferred with the transferred rights (including liability for any outstanding licence fees) unless Ofcom consents otherwise. In granting such consent, Ofcom will need to be satisfied that the proposed arrangements do not affect its ability to enforce the terms of the WT Act licences. Other obligations, for example to third parties, may also be transferred. The transferor will then sign a spectrum transfer form and pass this, together with its licence documentation, to the transferee upon signing of the transfer agreement. 3. The transferee will then be responsible for sending (a) the transfer form, (b) the existing licence documentation, (c) the signed transfer agreement and (d) a competition and regulatory notification to Ofcom. 4. Ofcom will then check the documentation to ensure that the proposed transfer is consistent with (a) the spectrum registry, and (b) the trading regulations, including the requirements for competition and regulatory clearance. Should a full review of the proposed trade be required for competition clearance, this may take several weeks to complete. 5. Assuming that the proposed transfer complies with these requirements, Ofcom will then update the spectrum registry, revoke the transferor’s existing licence, issue the appropriate licence to the transferee and, where appropriate, issue a new licence to the transferor (e.g. where rights to use spectrum have been partitioned). Subject to the terms and conditions agreed between the parties, completion of the trade is likely to occur at this point. Source: ‘Spectrum Trading Consultation’, Ofcom, September 2003.

5. CHARGING SCHEMES 5.1 Spectrum charges and revenue share The current licence states that ‘Gross revenue shall be inclusive of … revenue from permissible sharing of infrastructure and any other miscellaneous revenue, without any set-off for related item of expense, etc.’ It therefore appears that under existing licence conditions, revenue from spectrum trading of the transferor would form part of gross revenue (as revenue from permissible sharing of infrastructure) and therefore be shared with the government as per the applicable regime. Conversely, 384 | Indian Infrastructure: Evolving Perspectives the payment for spectrum by the transferee may be considered equivalent to access and interconnect charges and allowed as an offset. If the revenue share percentages of the transferor and transferee were the same, these charges would offset each other. The original licensee could continue to pay for her/his spectrum allocation as a percentage of her/his AGR, while the new user would pay the original licensee for the extra spectrum. The new user would also pay more to the government as her/his AGR increases with the use of more spectrum. When the revenue share of the transferor and transferee are equal, then, as long as the AGR of the transferee increases by more than the reduction in the AGR of the transferor, the government would get more revenue without releasing more spectrum, as in Table 19.2.4 5

Table 19.2: Example of spectrum trading with revenue share payments

Transferor Transferee Items Before After Before After Adjusted gross revenue (*) Rs 500 cr. Rs 500 cr.‡ Rs 1000 cr. Rs 1200 cr. Revenue from spectrum trading n. a. Rs 10 cr. n. a. (Rs 10 cr.) AGR including trading revenue n. a. Rs 510 cr. n. a. Rs 1190 cr. Charges paid to government† Rs 15 cr. Rs 15.3 cr. Rs 30 cr. Rs 35.7 cr. Amount of spectrum owned 6.2 x 2 MHz 6.2 x 2 MHz 6.2 x 2 MHz 6.2 x 2 MHz Amount of spectrum used 6.2 x 2 MHz 5 x 2 MHz 6.2 x 2 MHz 7.4 x 2 MHz * Does not include revenue from spectrum trading † Assumes a revenue share of 3% of AGR ‡ It is assumed that the transferor does not lose revenue, as she/he did not need the traded spectrum to serve her existing subscriber base. If spectrum is traded between similar types of users, then it appears feasible to continue the same scheme of revenue share charges and allow trading. However, this would prove difficult if change of use is contemplated. In such a case, delinking the spectrum from a particular use and using a spectrum-specific charge appears much more sensible.

5.2 Charging based on amount of spectrum The proposed approach attempts to move the spectrum charging scheme from one based on revenue share to charging based on amount of spectrum being used. At the same time, it tries to minimise the financial variation between the proposed charge and the existing charge. This is in order to accomplish the delinking of spectrum and service provision in a revenue-neutral manner, minimising its effect on financial projections of service providers and the government.6 From Allocation of Spectrum to a Market-based Approach | 385

5.2.1 Charging for existing allocations in a revenue-neutral manner The current system generates total revenues for a given total amount of spectrum by various service areas (circles). It is thus possible to calculate average prices for the total bandwidth within each circle and use this to set an initial charge. To account for the possibility that the revenue to the government could rise over time as subscriber growth occurs, one can build in an escalation factor that is related to average ex-post nationwide growth of wireless telecommunication revenues.7 For example, if the current charges generate a total Rs 50 crore per year from all the telecom operators for Delhi, one can apportion it to the 25 MHz allocated to all service providers to produce a base charge of Rs 2 crore per year per MHz (or Rs 4 million per 200 kHz). Assuming that Delhi, with an area of 1500 sq km, was divided into 15 SSUs, this would imply an annual base fee of Rs 270,000 for the SSU illustrated in Figure 19.2. This base fee can then be increased by the annual growth factor for subsequent years.8

5.2.2 Differential pricing between SSUs Since the current revenue numbers are available only at the circle level, all the SSUs in a given circle will start with the same base fee, e.g. Bangalore and Gulbarga (a small town in Karnataka at considerable distance from Bangalore) have the same spectrum charge per MHz, which is clearly not reflective of spectrum usage at the two locations. Since existing circles would be divided into a number of smaller SSUs, service providers may need less spectrum in less densely populated SSUs and choose to surrender their existing holdings (which are currently at a uniform level across the entire circle) if they are not using it, rather than continue to pay for it. Existing users can be allowed a one-time option to surrender excess spectrum after the move to SSUs. As the spectrum is surrendered, the revenue loss can be added back into areas where no surrender is taking place, and the revenue from the less dense SSUs can be divided by the amount of notionally available spectrum to reduce the per unit price of spectrum in such SSUs. This method of pricing has the potential to generate an automatic differentiation between prices for spectrum in SSUs with many subscribers and those with fewer users.

5.2.3 Example: How would this work? Consider a situation where the Karnataka circle is divided into ten equal geographic units, one including Bangalore and nine comprising the rest of Karnataka. Each of them receives the 24.8 MHz that is currently allocated to the Karnataka circle. Assuming that the Karnataka circle generates Rs 826 crore in AGR, the spectrum fee is Rs 24.8 crore. Initially, this is apportioned equally to all the ten geographic units, i.e. Rs 2.48 crore per unit or Rs 2 lakh per 2 SSU (each geographic unit being 124 SSUs9 ). At this price, all but 4 MHz (20 SSUs) are surrendered in each of the 386 | Indian Infrastructure: Evolving Perspectives

9 geographic units, generating a revenue loss of Rs 18.72 crore. This is added back to the Bangalore geographic unit, which increases its spectrum fee to Rs 21.2 crore from Rs 2.48 crore, implying a spectrum fee per SSU of Rs 17 lakh10 as shown in Table 19.3. After the move to SSUs, trading in spectrum can begin with these initial prices. Being a market-based allocation system, it can be expected to generate a market-based opportunity cost for spectrum. Table 19.3: Example of revenue neutral differential pricing between SSUs Bangalore Typical GU Total geographic for rest of Karnataka unit (GU) Karnataka GUs (1 to 9) Before After Before After Before After Spectrum used per geographical unit (MHz) 24.8 24.8 24.8 4 24.8 4 Spectrum fee per geographical unit (Rs crore) 2.48 21.2 2.48 0.4 22.32 3.6 Spectrum fee per SSU in each SSU (Rs crore) 0.02 0.17 0.02 0.02 0.02 0.02 Note: ‘Before’ and ‘after’ refer to before and after the surrender of spectrum by service providers.

5.3 Charging for new allocations: should we auction? It is expected that new spectrum may be needed only in certain densely populated locations—where it may or may not be scarce. The necessity for auctions is questionable when the scarcity of the resource is yet to be established. A better understanding of the scarcity will be obtained once trading is allowed to begin. Then, in case spectrum availability is not a problem, it may not be economically inefficient to continue allocating spectrum as and when required by operators. The charging for new allocations can be done on a basis similar to the issue of new stock in a traded company. This assumes that the secondary market for spectrum will generate sufficient information in terms of prices and trades to base such an offering. If this is not the case, an auction could be resorted to or new spectrum could be allocated at the price of the spectrum already in use, which is determined as described above.

5.4 Competition issues As noted earlier, in section 4.2, service providers may use spectrum as a competitive tool, cornering spectrum in an attempt to restrict its usage by competitors or to drive up the price of traded spectrum. Such issues are within the domain of restrictive trade practices and can be addressed on a case-by-case basis either by the Competition Commission or by TRAI. From Allocation of Spectrum to a Market-based Approach | 387

6. CONCLUSION In sum, this note argues for transitioning from administrative allocation of spectrum to a market-based approach in order to optimise the use of spectrum within a given use, i.e. avoid the additional release of spectrum while spectrum allocated for the service as a whole remains underutilised, and optimise the use of spectrum across different uses, i.e. permit the spectrum to be used for services that may be more useful than for what it was originally allocated. In order to do so, it suggests that the government should: (a) Allow trading of spectrum by redefining spectrum in terms of standard spectrum units (SSUs) and creating a register of spectrum rights. The trade could be bilateral or through an organised exchange (which can be expected to emerge if volumes increase); (b) Delink the spectrum from services by moving to a fee for spectrum based on SSUs. This fee will have variation based on user density of the SSU. This fee will be benchmarked to the existing and projected revenue share under the existing system, so that the financial implications of the change from the revenue-sharing regime are limited; and (c) Allocate new spectrum on a basis similar to the issue of new stock in a traded company. The prices can be based on the prices in the secondary market for spectrum trading. If the secondary market prices are not sufficiently informative, e.g. because the number of trades in the market is small, an auction could be resorted to or if there is no scarcity, new spectrum could be allocated at the price of the spectrum already in use, which is determined as described above. 388 | Indian Infrastructure: Evolving Perspectives

APPENDIX Table 19.4: Allocation of spectrum in other countries Sr. Name of No. of GSM Frequency Average GSM Number of no. the country operators available for frequency per subscribers as GSM service operator 2x on 2001 (’000) 2x (MHz) ** (MHz) 1. Austria 4 59.6 14.9 6565.9 2. Belgium 3 81.0 27.0 7690.0 3. Czech Republic 3 49.8 16.6 6769.0 4. Denmark 4 109.6 27.4 3954.0 5. Estonia 3 51.6 17.2 651.2 6. Finland 6 70.8 11.8 4044.0 7. France 3 74.4 24.8 35922.3 8. Germany 4 80.0 20.0 56245.0 9. Greece 3 45.0 15.0 7962.0 10. Hungary 3 68.6 22.9 4968.0 11. Iceland 6 69.6 11.6 235.4 12. Ireland 3 62.4 20.8 2800.0 13. Italy 4 71.6 17.9 48698.0 14. Lithuania 3 43.4 14.5 93.2 15. Netherlands 5 105.8 21.2 11900.0 16. Poland 3 48.8 16.3 10050.0 17. Portugal 3 41.8 13.9 7977.5 18. Romania 3 32.0 10.7 3860.0 19. Spain 3 64.2 21.4 26494.2 20. Sweden 3 75.0 25.0 6867.0 21. Switzerland 3 79.6 26.5 5226.0 22. United Kingdom 4 105 26.3 47026.0 23. China 2 45.0 22.5 144812.0 24. Australia 4 30.0 7.5 11169.0 25. Hong Kong 6 84.1 14.0 5701.7 26. Indonesia 3 25.0 8.3 5303.0 27. Malaysia 5 90.0 18 7128.0 28. Philippines 3 25.0 8.3 10568.0 29. Singapore 3 37.8 12.6 2858.8 30. Taiwan 6 75.2 12.5 21633.0 31. Thailand 3 57.1 19.0 7550.0 Source: Recommendations of the TRAI on intra-circle mergers and acquisition guidelines, January 30, 2004. From Allocation of Spectrum to a Market-based Approach | 389

Table 19.5: Allocation of spectrum in Indian telecom circles Sr. Name of the No. of Frequency Max and No. of GSM no. circle operators available min GSM subscribers for GSM frequency in December service per operator 2003 (’000) GSM Basic (MHz)** (2x) 2x (MHz) 1. Delhi 4 4 30.4 10 6.2 2934.41 2. Mumbai 4 3 28.4 8 6.2 2507.23 3. Chennai 4 3 24.8 6.2 6.2 759.79 4. Kolkata 3 2 18.6 6.2 6.2 792.90 5. Maharashtra 4 3 24.8 6.2 6.2 1967.72 6. Gujarat 4 3 26.6 8 6.2 1804.89 7. Andhra Pradesh 4 3 24.8 6.2 6.2 1545.51 8. Karnataka 4 4 24.8 6.2 6.2 1493.12 9. Tamil Nadu 4 4 24.8 6.2 6.2 1236.42 10. Kerala 4 2 23 6.2 4.4 1023.10 11. Punjab 3 3 18.6 6.2 6.2 1779.84 12. Haryana 4 3 24.8 6.2 6.2 414.51 13. UP-West 3 2 18.6 6.2 6.2 859.68 14. UP-East 2 2 12.4 6.2 6.2 602.25 15. Rajasthan 3 3 18.6 6.2 6.2 437.58 16. Madhya Pradesh 4 3 24.8 6.2 6.2 692.66 17. West Bengal & A&N 2 2 10.6 6.2 4.4 243.08 18. Himachal Pradesh 3 2 16.8 6.2 4.4 136.97 19. Bihar 2 2 12.4 6.2 6.2 449.67 20. Orissa 2 2 12.4 6.2 6.2 228.91 21. Assam 1 2 6.2 6.2 42.16 22. NE 1 1 4.4 4.4 10.22 23. Jammu and Kashmir 1 6.2 6.2 28.74 Note: Circles 1 to 4 are metros, 5 to 9 are category ‘A’ service areas, 10 to 17 are category ‘B’ service areas and 18 to 23 are category ‘C’ areas. ** In addition, many Basic Licence operators offer a CDMA mobile service, for which the usual allocation is 2.5 x 2 MHz (going up to 5 x 2 MHz) Source: Recommendations of the TRAI on intra-circle mergers and acquisition guidelines, January 30, 2004, and COAI. 390 | Indian Infrastructure: Evolving Perspectives

Table 19.6: Public spectrum registry—example of contents

Data field Description 1. Name of licensee Name of the individual or enterprise holding the licence (as notified to Ofcom) 2. Contact details Postal and email addresses and telephone numbers for correspondence with the holder of that licence 3. Current use Description of the current application of the licence 4. Frequency boundaries The radio frequency range of the assignment, specified either of right in terms of: • a central frequency with channel width, e.g. 415.25MHz ± 100kHz, or • a frequency range, e.g. 415.15 to 415.35MHz 5. Geographical boundaries Specification of the geographical characteristics of the right, either: • in terms of boundaries specified as planes between grid references, or • a radial boundary a specified distance from a particular grid reference 6. Power Statement of any power restrictions on the licence, particularly for apparatus specified licences, e.g.: • equivalent isotropically radiated power (EIRP), at the specified location, or • effective power flux density at the specified boundaries 7. Guard bands Specification of the frequency range of any guard bands associated with that assignment 8. Authorised use Description of the restrictions of use of the licence, e.g.: • harmonisation restrictions, or • other restrictions defined in the licence, e.g. changes of use permissible only within the constraints of MASTS for PBR, or • other technical limitations on the nature of transmissions 9. Other obligations Statement of any non-spectrum obligations under the licence, e.g. or conditions roll-out obligations 10. Administrative The annual AIP payable by the licensee Incentive Pricing Notes on availability of information: Parties to a proposed trade may wish to conduct a due diligence process prior to the trade. In certain cases, the parties may require access to technical From Allocation of Spectrum to a Market-based Approach | 391

information held by Ofcom. In particular, parties may require information regarding licence conditions, patterns of transmissions and guideline interference levels for co-channel users, co-located users or adjacent channel users. This will also be necessary in order to undertake suitable technical assessments for change of use and reconfiguration. While some of this information will be available on the spectrum registry, much of the information is likely to be highly detailed and bespoke to particular trades. Ofcom proposes to introduce a system which will allow WT Act licensees to request such information in writing, setting out reasons for the request. Such an arrangement will also allow Ofcom to consider and respond individually to requests, and where appropriate to tailor the information provided to the needs of the applicant. In deciding whether to make the requested information available, Ofcom will need to take account of any confidentiality or security considerations and will generally expect the recipient to agree to certain terms, including with regard to confidentiality and limitation of liability. The requesting party will be expected to meet Ofcom’s costs in providing this information. Security and confidentiality considerations will restrict the information that Ofcom can make available. For example, Ofcom does not expect to be entitled to make available information about many MoD assignments. It should be noted that where trades involve companies which are publicly listed, or quoted on the Alternative Investment Market (AIM), commercially sensitive information such as the agreement of a trade may first be required to be released through a Regulatory Information Service approved by the FSA or the London Stock Exchange, as appropriate. Ofcom would then update the spectrum registry to take account of the new ownership details, only after the required announcements have been made. Source: ‘Spectrum Trading Consultation’ Ofcom, September 2003.

NOTES 1. International experience with spectrum trading has so far been limited with few trades reported in the countries where it is allowed, e.g. Australia and New Zealand. 2. The clarity in spectrum property rights that would be needed to facilitate trading will also help to increase certainty in merger and acquisition transactions between service providers. 3. Conceptually, the standard trading unit is four-dimensional, the fourth dimension being time, but the temporal dimension is usually ignored to aid visualisation and practical understanding. 4. It is interesting to consider whether gains from such trading be seen as similar to gains from property sales or similar to stock trading. This will influence the nature of the taxation regime to be applied. 5. If the operator with a higher revenue share trades her/his spectrum to an operator with a lower revenue share, and the transferor loses some revenue to the transferee, then the government may lose revenue on spectrum charges unless the gain in revenue by the 392 | Indian Infrastructure: Evolving Perspectives

transferee is sufficiently high. In the case where the transferor’s revenue share is 3 per cent and the transferee’s revenue share is 2 per cent, this would imply that the increase in revenue by the transferor over and above her gain from the transferor must be at least 50 per cent of the gain from the transferor. In addition, if the government wanted to maximise revenue and not save spectrum, it could force the operator to ask for more spectrum by banning trade, and extract a higher revenue share, at the cost of releasing more spectrum. 6. There have already been two substantive changes in the licence conditions for telecom, first, through NTP 1999 and then by the introduction of UASL. Limiting the financial implications of the transition from an administrative allocation of spectrum will limit the incentive to convert this change into an opportunity for financial jockeying. 7. Using a nationwide average implies that areas which grow faster than the average will pay less than under the current system and vice versa. This may implicitly ‘subsidise’ metros in the initial period, but thereafter may penalise them, where spectrum usage is high, as their rate of growth slows and growth in the other regions rise. 8. Such a pricing scheme is similar to Administered Incentive Prices for spectrum use, e.g. in the UK mainly for non-commercial users, which reflect the value of the spectrum rather than the costs of spectrum management. 9. The spectrum dimension of SSUs is defined as in Figure 19.2, with a frequency dimension of 200 kHz. 10. At this price, operators may surrender spectrum in Bangalore too, at which point the unit price of spectrum in Bangalore will be progressively increased (which is an admittedly perverse outcome due to revenue neutrality).