Public Industrial Enterprises July 1986

iDUSTY 5AND FRNAMCE SER1ESVOLUNY`_ E IAF17

Public IndustrialEnterprises July 1986 Determinants of Performance Public Disclosure Authorized

Mahmood Ali Ayub and Sven Olaf Hegstad Public Disclosure Authorized Public Disclosure Authorized

F,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 Public Disclosure Authorized Public Industrial Enterprises

Determinants of Performance - ii -

Industry and Finance Series

Volume 17

This series is produced by the Industry Department of the World Bank to disseminate ongoing work done by the department and to stimulate further discussions on the issues. The series will include reports on individual sectors in industry, as well as studies on global aspects of world industry, problems of industrial strategy and policy, and issues in industrial finance and financial development.

Already published are the following:

*Volume 1. Structural Changes in World Industry: A Quantitative Analysis of Recent Developments

*Volume 2. Energy Efficiency and Fuel Substitution in the Cement Industry with Emphasis on Developing Countries

*Volume 3. Industrial Restructuring: Issues and Experiences in Selected Developed Economies

*Volume 4. Energy Efficiency in the Steel Industry with Emphasis on Developing Countries

*Volume 5. World Sulphur Survey

*Volume 6. Industrialization in Sub-Saharan Africa: Strategies and Performance

*Volume 7. Small Enterprise Development: Economic Issues from African Experience

*Volume 8. World Refinery Industry: Need for Restructuring

*Volume 9. Guidelines for Calculating Financial and Economic Rates of Return for DFC Projects (also in French and Spanish)

Volume 10. A Framework for Export Policy and Administration: Lessons from the East Asian Experience (also in Spanish)

Volume 11. Fertilizer Producer Pricing in Developing Countries: Issues and Approaches

Volume 12. Iron Ore: Global Prospects for the Industry, 1985-95

Volume 13. Tax Policy and Tax Reform in Semi-Industrial Countries

Volume 14. Interest Rate Policies in Selected Developing Countries 1970-82

Volume 15. Mobilizing Small-Scale Savings: Approaches, Costs, and Benefits

'Volume16. World Bank Lending to Small Enterprises

* Published as World Bank Technical Papers. INDUSTRYAND FINANCE SERIES VOLUME 17

PublicIndustrial Enterprises Determinants of Performance

Mahmood Ali Ayub and Sven Olaf Hegstad

This is a document published informally by the World Bank. In order that the information contained in it can be presented with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The World Bank does not accept responsibility for the views expressed herein, which are those of the author(s) and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank; they do not necessarily represent official policy of the Bank. The designations employed, the presentation of material, and any maps used in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates conceming the legal status of any country, territory, city, area, or of its authorities, or conceming the delimitation of its boundaries or national affiliation. The most recent World Bank publications are described in the annual spring and fall lists; the continuing research program is described in the annual Abstracts of Current Studies. The latest edition of each is available free of charge from the Publications Sales Unit, Department T, The World Bank, 1818 H Street, NW., Washington, D.C. 20433, U.S.A., or from the European Office of the Bank, 66 avenue d'lina, 75116 Paris, France.

Mahmood Ali Ayub is senior industry economist, and Sven Olaf Hegstad is senior industry specialist, in the Industry Department of the World Bank.

Libraryof Congress Cataloging-in-PublicationData Ayub, MahmoodAli, 1948- Public Industrial enterprises.

(Industry and finance series, ISSN 0256-2235 ; v. 17) Bibliography:p. 1. Government business enterprises. 2. Organizational effectiveness. I. Hegetad, Sven Olaf, 1945- II. Title. III. Series. HD3850.A98 1986 350.009'2 86-15825 ISBN 0-8213-0815-7 -v-

ABSTRACT

The main purpose of this study is to identify the factors, separate and apart from ownership, which influence the performanceof public industrial enterprises. Based on a review of performanceof state- owned enterprises in 13 countries, the authors isolate three main factors relating to the business and managerial environment that distinguish successful public enterprises from the poorly performing ones. These include: (i) the degree of competition that public enterprisesare exposed to; (ii) the degree of financial autonomy and accountabilityunder which public enterprisesoperate; and (iii) the extent and manner in which managerial autonomy and accountabilityare ensured.

It is impossible,and perhaps misleading, to assess statistically the individual importance of each of these factors. What is clear from the analysis is that, where these three factors exist as a package, the performance of public enterprises is significantlybetter than in those cases where most or all of these factors are non-existent.

- vii -

TABLE OF CONTENTS

CHAPTER 1. INTRODUCTION AND SUMMARY...... 1

Objectives and Scope of This Study ...... 1 Why Study Public Enterprises? ...... 2 Key Determinants of Performance ...... 4

CHAPTER 2. THE ROLE AND PERFORMANCE OF PUBLIC INDUSTRIAL ENTERPRISES .... ***..***...... 9

The Role of Public Industrial Enterprises ...... 9 Performance of Public Industrial Enterprises .... 12 Variations in Performance ...... 13 Comparing Public and Private Enterprise Performance ...... 14 Conclusion ...... * .. *..*...... ***** 17

CHAPTER 3. A COMPETITIVE ENVIRONMENT AS A DETERMINANT OF P°ERFORMANCE ...... 18

CHAPTER 4. FINANCIAL AUTONOMY AND ACCOUNTABILITY ...... 21

Overall Financial Framework ...... 22 Greater Market Discipline.o-...... 23 Financial Objectives and Dividend Policies .... 23 Clarification of Social Objectives ...... 24 Elimination of Government Subsidies ...... o..... 26 Supportive Financial Policies ...... 27 Appropriate Capital Structure Policies ...... 27 Greater Discipline and Transparency in Financial Relations too .... * ...... 28 Performance Measurement and Incentives Systems ...... 29 Conclusion ...... 30

CHAPTER 5. MANAGERIAL AUTONOMY AND ACCOUNTABILITY ...... 31

Three Roles in Enterprise Management ...... 31 Evolution of Public Enterprise Management Models ...... 33 Factors Affecting Organizational Designs ...... 39 Conclusion ...... 41 - viii -

TABLE OF CONTENTS (continued)

CHAPTER 6. MANAGERIAL SKILLS AND MORALE ...... 42

Managerial Skills .... * ...... 42 The Issue of Morale ...... 43

CHAPTER 7. COORDINATION WITHIN THE GOVERNMENT ...... 45

Holding Companies ...... 46 Focal Points ...... 48 Lessons from Experience ...... 51 Options for Organizational Design ...... 53 Abolition of Special Public Enterprise Regimes ... **. **. *.*..*...... 53 Organizationof Public Enterprisesas Ordinary Stock CorporationUnder Corporate Law ...... *..... 53 Reduction of Ministerial Control ...... 53 Creation of a Public Enterprise Intersectoral Board and Secretariat (Focal Point) ...... 54 Strengtheningof the Role of the Board of Directors ...... 54 Strengtheningof the Role of the Chief Executive Officer and Top Management ...... 54 Decentralizationof Managerial Autonomy and Accountabilityto the Plant/Subsidiary Level ...*...... **. 55

ANNEX I ORIGINS OF PUBLIC INDUSTRIAL ENTERPRISES ...... 56

Political and Ideological Considerations ...... 56 Takeover of Ailing Private Companies 58 Unwillingness or Inability of the Private Sector 59 Strategic Industries 59 Internal Growth and Diversification ...... 59 Creating Competition for the Private Sector 60

ANNEX II MANAGEMENTOF PUBLIC ENTERPRISES AND THE BOARD OF DIRECTORS ...... 61

The Role of the Board of Directors ...... 61 Policy-Making and Strategic Role 62 The Performance Evaluation Role 62 Appointment and Dismissal of the CEO ..... 62 Why Some Boards of Directors.AreNot Effective . 62 An Approach to Effective Board Operations *..... 63 Conclusion ...... 65 - ix -

TABLE OF CONTENTS (continued)

ANNEX III THE USE OF HOLDING COMPANIES ...... 67

Purpose ...... *. ... 67 Performance ...... 69 Strengths ...... 69 Buffer Against Excessive Political Interference 69 Professional Management of the Ownership Role 70 Leveraging Scarce Management Skills ...... 70 Benefits from Cooperation (Synergy) ...... 70 Weaknesses ...... ,...... *.,.* ...... 71 Effective Tool for Political Interference .... 71 Excessive Layers of Management and Slow Decision-Making ...... * . 71 Lower Motivation and Quality of Managers ...... 72 Excessive Financial Powers ...... 72 Lack of Appropriate Role ...... 72 When to Use Holding Companies ...... 73

STATISTICAL ANNEX ...... o...... 75

LIST OF TABLES

Summary Table - Factors Affecting Public Enterprise Performance ...... 8 Table 1 - Share of Public Enterprises in Non-U.S. Fortune 500 Companies ...... o 9 Table 2 - Profitability of Public Industrial Enterprises by Subsector, 1984 15 Table 3 - Comparison of Financial Profitability, 1984 16 Table 4 - Levels of Financial Control 30 Table 5 - Conventional Roles in Corporate Management 32 Table 6 - Management Models for Public Industrial Enterprises in Thirteen Countries .34 Table 7 - Establishment of Multisectoral Holding Companies ...... -.- .... ".-.-. . 46 Table 8 - Focal Point Organizations .49 TABLE OF CONTENTS (continued)

LIST OF FIGURES

Figure 1 - Public Ownership of Selected Industrial Subsectors, 1984 . .0 ...... 11 Figure 2 - Vicious Circle of Public Enterprise Sector Management ...... 22 Figure 3 - Life Cycle of Public Enterprise Sector Organizations...... 38 Figure 4 - Public Enterprises in Italy ...... 47

LIST OF ANNEX TABLES

Table A-1 - Establishment of Holding Companies ...... 67 Table A-2 - Summary of Strengths and Weaknesses ...... 73 Table A-3 - Use of Holding Companies ...... 74 Table SA-1 - Key Indicators of Major Public Industrial Enterprises, 1984 ...... -...... 75 Table SA-2 - Shares of Public IndustrialEnterprises in Selected Indicators, 1983 ...... 77 - xi -

FOREWORDAND ACKNOWLEDGEMENTS

This study was undertakenwith a view to fulfillingthe growing need in the World Bank as well as its member countries for understanding the underlying factors that influence the performanceof public industrial enterprises. To fulfill this need, country case studies of such enterprises were undertaken for thirteen countries,using a common information- gathering and analyticalapproach. The authors of these country studies were as follows:

Austria : Anne Milne Brazil : Mahmood Ayub France : Gilles Michel Ghana : Ebenezer Nyakotey India : Jayant Kalotra Israel : David Kochav Italy Sergio Ristuccia Norway : Bo Carlsson Pakistan : Istiqbal Mehdi Portugal : Sven Hegstad Sweden : Bo Carlsson Tunisia : Jan Svejnar Zambia : Charles Becker

The above country studies were used as backgroundmaterial for the present study. Where necessary, those studies were complementedby other sources of informationand the experiencesof other countries.

The authors are greatly indebted to Kemal Dervis, Harinder Kohli and Anil Sood for their support and guidance at all stages of preparation. We are grateful to Dennis Flannery and Ira Lieberman for their support in writing the annex on the board of directors and to Sujin Hur and Rebecca Sekse for their assistance with the statisticalwork, to Whitney Watriss for editorial suggestionsand to Wilson Peiris and Tony Tenorio for their patient help in typing various drafts of this study. Finally, the authors wish to acknowledge the helpful comments of Jean Grosdidierde Matons, Leroy Jones, Gobind Nankani, Roy Pepper, Alejandro Schwedhelm,Edilberto Segura, Mary Shirley, Charles Vuylsteke and George Zaidan. CHAPTER 1

INTRODUCTIONAND SUMMARY

1.01 Why do some public industrial enterprisesperform better than others? Why is it, for example, that public enterprises in Brazil achieve generally better financial and economic results than do their counterparts in Ghana or Pakistan? Within Brazil, what is it that makes the national iron ore company (CVRD) and the national petroleum company (PETROBRAS)much more successful than, say, the copper company (Caraiba Metais) or, at the extreme, the railroad company (RFFSA)? Why does India's Hindustan Machine Tools (HMT) remain dynamic and well-performing,whereas other public enterprises in the same country and even in the same subsector are far less successful?

1.02 There are two sets of factors. The first consists of country- specific cultural, social, political,macroeconomic and institutional characteristics,that explain inter-countrydifferences in enterprise performance. The other comprisesmore enterprise-specificmatters, such as the degree of autonomy, extent of both domestic and foreign competition, and the corporate and managerial environment. This set explains intra- country differences in enterprise performance. Clearly, there is overlap between the two sets of factors.

Objectives and Scope of This Study

1.03 This study addresses only those factors in the two sets that are peculiar to public firms and that can be influenced by policy-makers. Thus, although cultural and social factors can be of critical importance in determining the degree of success of public enterprises,they are outside the purview of this report. Similarly, the quality of individualchief executive officers has an important effect on the performanceof public enterprises,but the same is equally true for private firms. Likewise, a crucial determinant of performance of both public and private enterprises is the appropriatenessof the macropolicy and institutionalenvironment. Sound industrial, trade and financial policies, an institutionalenvironment that is not too encumbered by regulations,absence of severe rigidi- ties in product and factor markets, the existence of domestic competition, well-articulatedfiscal reforms and realistic restructuringplans are all extremely important determinantsof performance. While these policy issues are outside the purview of the present study, it is assumed throughout this study that governments would put in place at least the minimum package of such policy measures as a mutually complementaryand reinforcingelement to other more specific public enterprise management measures discussed in this report.

1.04 The background material for this study comprises 13 country case studies of public industrial enterprise sectors, which are availableon request. These studies were complementedby interviewswith senior managers of successful and not-so-successfulpublic enterprises. The countries included are Austria, Brazil, France, Ghana, India, Israel, - 2 -

Italy, Norway, Pakistan, Portugal, Sweden, Tunisia and Zambia. The case studies contain detailed informationon the genesis, role, performance, control mechanisms, organizationalstructures and key performance constraints of public industrial enterprises in each country. To ensure maximum consistencyand comparability,a common questionnairewas used for interviews with enterprise managers and government officials and for preparing the country reports. Where relevant, this report also draws on the authors' experienceswith other countries, especially Bolivia, Turkey and Uganda.

1.05 This study deals primarilywith public industrialenterprises. Since these entities generally produce tradable goods in markets with varying degrees of competition,they are distinct from some other public enterprises such as utilities and transport companies, which often are natural monopolies. However, most of the organizationaland management issues facing public industrial enterprisesare also relevant to other public companies even though the remedies may differ.

Why Study Public Enterprises?

1.06 In many countries, particularlythose in which a robust private sector did not exist, public industrial enterprises have played a crucial role in the developmentof the industrial sector. They have been the main sources of growth in several industrial sectors in countries such as Brazil, Egypt, Ghana, India, Pakistan, Turkey and Zambia, to name a few. They have also frequently played an important role in the fulfillment of certain social objectives that probably the state or the private sector would have found difficult to undertake. Unfortunately,however, the high growth rate and the fulfillment of social objectives were not achieved in many cases in a cost-effectiveway. So long as countries experienced rapid industrial growth and resources were relatively abundant, these inefficien- cies were tolerated and even ignored. However, the severe resource constraints and curtailment of growth in many countriesduring the last several years has made many governmentsrealize that public enterprises can become obstacles to the resolution of macroeconomic problems such as high inflation rates, large budgetary deficits, excessive foreign debt and the like.

1.07 Furthermore,the business environment for public industrial enterprises changed dramaticallybetween the 1950s and the 1980s. The fifties and early sixties were characterizedby strong and predictable growth. The main concern of producers was to solve production problems so as to be able to satisfy demand. Starting in the sixties, the market place gradually shifted to consumer-drivenmarkets. At the same time, developing countries became a significant factor in the world market for manufacturedgoods and a globalizationof the market place evolved rapidly. The increased speed of technologicalchange further fueled competitive pressures. Response to customer preferencesand enhanced motivation became increasinglyimportant, and rigidity in long-term planning was replaced with flexibility,creativity and planning for strategic options. Public enterprises frequently characterizedby bureaucratic organizationalstructures and relatively slow decision-makingprocesses were often ill-suited to respond effectively to the changing environment. At the same time, especially in the more developed countries,many of the original reasons for the creation of public industrial enterprisesdis- appeared or were diluted, such as being engines of growth for industrialization and fulfillinga lack of private sector initiative. Regardless of their size, stage of development and political preferences,many countries are, therefore, currently undertaking careful assessmentsof the role, performance and raison d'etre of public enterprises.

1.08 The response of several developing countries to their mounting economic problems has been to impose restrictivemonetary policies. In the absence of greater emphasis on fiscal constraintsand with somewhat more lax controls on public sector expansion and easier access to domestic and foreign credit, there has been some "crowding out" of private sector activities even in countries such as Brazil and Turkey (in the late 1970s) that have assigned priority to private sector development. At the same time, the result has been excessively high real interest rates and severe financial distress. A more appropriate policy response may be to give greater priority to reducing the public sector deficit in general, and reducing the financial subsidies to public enterprises in particular. This would help in relaxing somewhat the constraintson credit to private producers. Such a policy mix can elicit a greater supply response from the economy.

1.09 Another response of several governments to the relatively poor performance of public industrial enterpriseshas been to restructurethem as a complement to policy and institutionalreforms. Such restructuring has involved considerationof a wide variety of measures, including breaking up large units, merging small ones, selling assets, spinning units off, liquidating them completely,opening up a sector to private investors, and privatizing existing units partially or entirely. While privatizationis generally desirable, the extent, nature and speed with which it can be pursued varies from country to country. Even in countries such as Brazil and Turkey, whose governments have publicly endorsed privatization,the process has been slow in view of the need to build a broad politicalcon- sensus, to change entrenched positions, to inculcate a new approach and to overcome certain inherent economic constraints. Thus, in Brazil, even after six years of a widely publicized divestiture program, the assets of privatized public industrial enterprises have so far reached only 2-3 percent of the total assets of all public industrial enterprises,and only one of the 17 privatized firms or groups has more than 1,000 employees. In some countries, particularly in Africa, the thinness of the domestic capital markets, scarcity of experiencedentrepreneurs and reluctance of the private sector to acquire inefficientand loss-making state assets at mutually acceptable prices are inherent economic structural constraintson privatization. Even in higher income countries such as Chile, financial sector weaknesses can greatly complicate privatization. It is therefore realistic to assume that large portions of industry will remain in public ownership for considerabletime, even in countries committed to privatization. Given the substantialweight of public enterprises in most countries (see Chapter 2), the efficientmanagement of their operations is therefore likely to continue to have a crucial impact on industrialgrowth and macroeconomic prosperity in many countries. - 4 -

Key Determinantsof Performance

1.10 No country, developed or developing,has devised a perfect way to manage public industrial enterprises. Even if one were found, social and cultural differenceswould make its blind replicationalmost meaningless. But there are clearly better and worse ways of managing public enterprises.

1.11 Attempts at achieving a more efficient public enterprise sector must begin with an understandingof the factors inhibiting adequate performance. Are they purely inherent, arising from public ownership per se? Or, are there other important determinantsof performance,separate and apart from ownership, that can be influencedby governments?

1.12 To some extent, the poor performance of public industrial enterprises relative to their private counterparts(discussed in Chapter 2) can be explained by some inherent differencesrelating to ownership. These include: lack of a significantfinancial stake on the part of the putative owners of public enterprises (governmentofficials representingthe state); limited threat of bankruptcy and reorganizationbecause of easy recourse to government funds and political inertia in terms of lay-offs of large numbers of employees; the circuitous chain of decision-makingdevised to ensure public accountability;and the generally close political ties between trade unions and government officials and board members. However, while these inherent factors are important, they are only part of the explanation. Evidence from our study indicates that factors not strictly related to ownership per se are equally important, and this study attempts to identify these.

1.13 Our review of the performanceof state-ownedenterprises in 13 countries has highlighted three factors relating to the business and managerial environment that distinguish successful public enterprises from the poorly performing ones: the degree of competition that public enterprises are exposed to; the degree of financial autonomy and accountability under which public enterprises operate; and the extent and manner in which managerial autonomy and accountabilityare ensured. If these factors can be managed appropriately,the performance gap between public and private enterprises can be reduced significantly,and to levels that in many cases can be justified by the infrastructural,strategic or social reasons that motivated the creation of many public enterprises in the first place.

1.14 The first component of a framework for the management of public industrial enterprises is the degree of competition that these entities are exposed to. Our study discovered numerous cases in which public industrial enterprises subjected to a more demanding, competitive environmenttended to perform better. Inappropriatemacroeconomic, trade and industrial policies (maintenanceof an overvalued exchange rate, excessive reliance on quantitativeimport restrictions,high tariff protection,price controls, restrictive procurementpolicies, localizationprograms, investment licensing, regulationson entry and exit and other regulatory constraints)are the most potent constraints on competitionfor both public and private firms. While it was not the purpose of this study to examine comprehen- sively the above policies, it is worth repeating that without a set of -5- rational and consistent policies that engender a competitiveenvironment, specific measures for better management of public enterprises cannot help much. In addition, some specific ways of fostering competitionamong public enterprises are discussed in Chapter 3.

1.15 The second component is the degree of financial autonomy and accountabilityof public enterprises. This entails giving firms sufficient financial autonomy, holding public enterprise managers accountableand ensuring that they pay the government (and any other shareholders)an adequate return on capital in much the same way as private firms do. Many well-intentionedgovernments assume that financial profitabilityis not a high priority for public enterprises. If public managers are freed of domination by profit motives, it is argued, they will automaticallyconcen- trate on achieving broad social goals rather than narrowly defined financial targets. Generations of public managers have been taught that there is a stigma attached to financial profitability.

1.16 The problem with this view is that freedom from concern with profits frequentlydegenerates into a lack of concern for wasteful use of resources without a guarantee that social objectivesare being fulfilled adequately. While it does encourage public investment, it is often with poor timing and feasibility: the innumerable "white elephants"in public investments in many countries are ample testimony to this outcome. Public enterprises in certain social infrastructuresectors set up specificallyto achieve certain non-financialobjectives perhaps can afford to ignore financial objectives. But public industrial enterprises that produce tradable goods that face (or potentially face) domestic and foreign competition can only do so at great cost. All the above argues for strict financial discipline, including elimination of subsidies and enforcement of dividend policies for those enterprises that are financiallyviable, and a program of action toward such an end for those that are currently loss-makers. Such a program would include (i) a clear identificationof the sources of poor financial performanceand (ii) establishmentof a time-bound program for eliminating losses. Some guidelines for appropriatefinancial policies are provided in Chapter 4.

1.17 The third crucial factor that distinguishessuccessful public enterprises from poorly performing ones is managerial autonomy and accountability. Experience indicates that excessive control of, and interference with, the operational decisions of public managers (in investment, product mix, pricing, hiring and firing of staff, wage-setting,procurement policies, etc.) can suffocate managerial initiative and result in a loss of accountabilityand costly operationalinefficiencies. The problem of excessive operational control is not only that the process is time-consuming for an already over-stretchedbureaucracy, but also that government officials lack the necessary information,business perspectiveand confidence to make correct and expeditiousdecisions. Excessive political interferencein the operationalmatters of public enterprises can be reduced by demarcating clearly the roles and responsibilitiesof the government (ownership role), the board of directors (strategicrole) and the management of enterprises (operationalrole); by appointing professional directors with experience relevant to their tasks; by decentralizing - 6 - appropriate powers to the board of directors; and by introducinga management philosophy that is less control-orientedand procedure-boundand more concerned with judging managers on the basis of enterpriseviability and a limited set of indicators of performance. This decentralizedstructure can be applied to most countries and enterprises. Naturally, the types of powers granted and services provided by central units, and the mix between market and government control, will vary between countries such as Sweden and Zambia. However, these differing levels of powers and functions could be exercised within the same decentralizedorganizational structure. These issues are discussed in Chapter 5.

1.18 A related set of factors, examined in Chapter 6, affecting the autonomy and accountabilityof public enterprise managers is the ability of these managers to influence quality and availabilityof managerial skills, and to maintain morale in a situation where many of the frustrationsof the work force may be the result of powers outside of the public enterprise. Problems arise because of the inability of public enterprises to attract or retain highly qualified staff; high turnover rates of managers resulting from political changes or civil service rotation policies; and political considerationsin filling vacancies. While budgetary constraintsprevent governments from offering more competitive salaries to public managers, there are many other ways to increase motivation (delegatinggreater responsibility)and reducing the severe shortages of skills (well-focused training programs). The second factor--highmorale--can affect performance but it is generally the result, not the cause, of good performancein public enterprises. However, this study has provided several examples where public managers have adroitly used small, well-publicizedsuccesses to turn sagging morale and discipline around and thereby improve performance further.

1.19 Once the government has decided how much decision-makingpower to delegate, the next issue is how to organize its various components ,(ministries,specialized agencies, etc.) to ensure effective decision- making. The multiplicityof government agencies involved in decisions relating to public enterprises has frequently led to confusion,duplication and excessive control. Superimposedon this diffusion of responsibilityis the propensity of public enterprises to become lightning rods for excessive political interferenceand discontinuityin decision-makingwith every change in government.

1.20 To improve coordinationand reduce interference,governments have experimentedwith various institutionalarrangements. Two approaches, reviewed in Chapter 7, are the use of holding companies and central supervisory agencies (or focal points). Evidence from both developed and developing countries suggests that large-scale state-ownedmultisectoral holding companies have been a negative factor. Instead of being a buffer against excessive political interference,they have: become an effective tool for political domination and interference;created excessive layers of management; reduced the transparencyof performance;and resulted in slow and bureaucraticdecision-making. State-ownedholding companies are likely to be somewhat more effective when they are used to manage small- and medium- sized enterprises. Central supervisoryagencies can play an important role if: their functions include, and are limited to, the execution of the ownership role; their functions and size are carefully defined and circums- cribed; they are inhibited from interfering in the day-to-day operationsof the enterprises;and if they are staffed with highly qualified persons with relevant industrial and managerial experience. While one main character- istic of public enterprise sector management is excessive government interference, it is equally common to find a lack of effective control, a situation that creates uncertainty,lack of understandingand sometimes distrust and that frequentlyends up in excessive interference. The dual purpose of effective central supervisoryagencies is, on the one hand, to protect the powers of the board of directors and management from undue political maneuvering, and, on the other, to exercise effective control of a limited number of performancefactors.

1.21 The following summary table enumerates some of the key components of the three factors, indicated earlier, that influence the performance of public industrial enterprises. Details on these are provided in the following chapters. It is impossible,and perhaps misleading, to assess statisticallythe individual importance of each of these factors. What is clear from our analysis is that, where these factors exist as a package, the performanceof public enterprises is significantlybetter than in those cases where most or all of these factors are non-existent. SUMMARYTABIE: FACTORSAFFETING PUBLIC ENTERPRISEPERFORKANCE

PERFORMANCEENHANC ING FACTORS PERFORMAJCERETARDING FACTORS

Competitive Environment Competitive Environment

1. Exposure of public enterprises to domestic and foreign competi- 1. Excessive protection from domestic and foreign competition. tion via more outward-oriented trade and industrial policies and less regulation. 2. Assignment of monopoly powers or consolidation of several enterprises into a single one without any perceived economies of scale. 2. Encouragement of public enterprises to compete among themselves and with private companies in the same lines of business.

3. Wide publicity in meetings and in the national press of the comparative performance of different public enterprises.

4. Breaking uipof overly centralized and monolithic companies into smaller ones, where economies of scale are not important.

Financial Autonomy and Accountability Financial Autonomy and Accountability

5. Reliance on capital markets as the primary source of financing. 3. Disregard by government and/or puiblic managers for financial profitabtlity. 6. Where government debt financing ts used, market interest rates are charged. 4. Excessive reliance on government funds, frequently at subsidized rates. 7. Strict enforcement by government of return on capital criteria, including dividend payments. 5. No pressure on enterprises to make dtvidend payments or to earn an adequate return on capital. 8. Enterprises not overburdened with too many social objectives and, where necessary, are compensated for performance of social 6. Confitcts between social and financial objectives, and the pre- functLons. dominance of social objectives.

9. Reducition of operating subsidies, if necessary in a phased way, 7. Excessive reliance on debt rather than equity financtng. with ithe objective of their total elimination (except for item 8) 8. Complicated maze of transfers between enterprises acnd central 10. A proper debt/equity ratio consistent with the business risks. budgetary authorities.

11. Greater transparency in enterprise-governnent flows to ensure 9. Poor quality and/or delayed financial accounts. betteir financial accountability. 10. No accountability for financial performance. 12. Use of generally accepted accounting standards, timely publication of financial statements, audited by independent private 11. No financial stake of managers in company's financial perform auditors. Also, setting up of a simple financial performance ance. measurement system.

13. Linking of performance to a system of rewards and penalties for managers and key staff.

Managerial Autonomy and Accountability. and Coordination Managerial Autonomy and Accountability, and Coordination

14. Clear demarcation of the ownership, strategic and operating 12. Excessive control of, and tnterference with, the operations of roles,. public enterprises.

15. Appointment of professional directors, with relevant experience, 13. Politicized board of directors. and delegation to them of strategic decisions. 14. Little autonomy for public managers in daily operations. 16. Delegation of substantial managerial autonomy in day-to-day operations to public enterprise managers. 15. Multiplicity of agencies involved in coordination, control and supervision of public enterprises. 17. Delinking of enterprise recruitment, promotion and salary administration from public service regulations. 16. Existence of bureaucratic, over-centralized multisector holding companies. 18. Where necessary, establishment of well-focused sector holding companies with well-defined functions.

19. Strengthening of oversight agencies for better supervision, and setting up of public enterprise focal points for policy-making and execution.

20. Organization of public enterprises as ordinary stock corporations under corporate law. -9-

CHAPTER 2

THE ROLE AND PERFORMANCE OF PUBLIC INDUSTRIALENTERPRISES

2.01 The purpose of this chapter is to place the role and performance of public industrial enterprises in perspective. Section A deals with the weight of these entities in the economies of the sample countries. Section B assesses the performance of public industrial enterprises,both in absolute terms and in relation to that of private firms.

The Role of Public IndustrialEnterprises

2.02 Public industrial enterprises hold an importantposition in the economies of the sample countries, although there are differencesof magnitude. Of the 500 largest non-U.S. industrial corporationson the Fortune list for 1985, some 71 are public (see Table 1, StatisticalAnnex, for the list, which also includes petroleum companies). More significantly, public industrial enterprises account for 19 percent, 21 percent and 21.4 percent of the total sales, assets and employment,respectively, of the 500 firms. Equally important, over the last decade their importance, in terms of the number of public enterprises on the Fortune list and the share of employment,has increased, as indicatedby Table 1.

Table 1: SHARE OF PUBLIC ENTERPRISES IN NON-U.S. FORTUNE 500 COMPANIES

1975 (X of total) 1984a/ (X of total)

Number of Enterprises 54 (10.8) 71 (14.2) Total Sales (US$ billion) 162 (19.7) 359 (19.0) Total Assets (US$ billion) 156 (20.5) 339 (21.0) Total Employees (million) 3.0 (17.5) 3.8 (21.4)

a/ A significant portion of the increase in the number of public enterprises, their sales and assets is explained by the nationalizationof many industrial enterprises in France since 1981.

Source: Fortune, August 1976 and August 1985.

2.03 Based on data for the sample countries (the shares of all their public industrialenterprises in value added, gross fixed investment and employment are provided in Table 2 in the StatisticalAnnex), there is no indication that the average share of public ownership of industrial value added is related to the level of developmentof a country (as measured by - 10 -

GDP per capita).1/ There were various economic, political, social and historical factors (such as the nationalizationof Renault in France for collaboratingwith the Nazis during World War II and the more recent nationalizationsof many industrial companies since 1981 after the instal- lation of the socialistgovernment in that country) that have determined the size of the public sector, regardlessof the stage of economic development.2/ On the other hand, there appears to be a discernible pattern in the relative share of public ownership between various subsectors, regardlessof whether a country is developed or developing. This pattern which is shown in Figure 1, can be largely explained by such factors as the perception of some subsectorsas strategic or as characterized by economies of scale (such as steel, aerospace, telecommunications, etc.).

2.04 Much more important than the share of public industrialenterprises in total value added is their share in total investment. Put differently, the public enterprise sector in the sample countries,with the exception of Austria, is much more capital-intensivethan the private enterprise sector. For the sample countries, the average share of public industrialenterprises in capital formationwas 26 percent, more than twice the share in GDP of about 11 percent. As Sheahan indicated,"in Brazil and in India it is almost as if industrieswere divided between private and public enterprise according to their capital intensity."3/ To a large extent, this situation reflects the fact that some public enterpriseswere set up precisely because of the large-scalecapital requirementsof some subsectors (such as steel, oil refining, etc.) that the private sector was unable to finance. However, it may also reflect differences in the relative factor prices faced by public and private enterprises. In a number of the sample countries, the cost of capital to public industrialenterprises is lower than that to their private counterparts,and the availabilityof

1/ The correlation coefficientsof per capita GDP with the shares of public industrial enterprises in industrialand total GDP were only 0.05 and 0.03, respectively. Kendall's rank coefficientswere, respectively,0.14 and 0.39 (a zero value signifies a random pattern and 1.00 indicates exactly similar rank orders), results that are only significant at 1 percent and 10 percent levels of confidence, respectively.

2/ For some details on the origins and developmentof public industrial enterprisesin the sample countries, see Annex I.

3/ John B. Sheahan, "Public Enterprise in Developing Countries," in William G. Shepherd, Public Enterprise: Economic Analysis of Theory and Practice (Lexington,Ky.: Lexington Books, 1976). Figure 1: PublicOwnership of SelectedIndustrial Subsectors, 1984

PETRO CEET MOTOR MNN MZR TE EEO

TEXTILES ELECTRONICS CHEMICALS CNNICSCO

AUSTMIA 0 0 N.A. 0 * _**

BRAZIL 0 G 0 0 _ 6 6

FRANCE ___ Q _Q_ GHANA N.A. Q Q N.A. N.A.

INDIA ______

ISRAEL 0 ___ 0QN.A Q __

ITALY_.A 0 . . _ NA ._

_ A ______6 _ _ _° _ _ *__ _ SWEDEN 0

TUNISIA Q O N.A. GJ6 N.A.

ZAMBIA N.A. N.A. N.A.

Q I^U' OY 251 rPUBUC SOX PUSUC &75X PUSUC * MARC

NOTES: (i) In a number of countries, developed and developing, some public companies have substantial (but less than 50%) ownership in private hands. However, for the purpose of this report, these are considered as fully public. (ii) Fertilizers refers to nitrogenous fertilizers only. (iii) Telecommunications refers to services, not manufacturing. (iv) The entry for India's motor vehicles takes into account the recently established, publicly owned Maruthi plant. (v) The entries for Israel's motor vehicles and steel indicate majority ownership not by the private sector but by the General Federation of Labor Trade Unions (Histradut). (vi) N.A. indicates not available or negligible production. - 12 - subsidized lending from domestic monetary authoritiesor aid agencies encourages relatively capital-intensivetechniques.4/

Performanceof Public IndustrialEnterprises

2.05 Assessing the performanceof public industrialenterprises is made difficult by a lack of data for roughly the same time periods, problems of cross-countrycomparison and the choice of indicators available. Given these problems, the approach taken here was the "weight of evidence" one used by historians and jurists rather than one of testing hypotheses statistically. For the purpose of this study, it was deemed sufficient to rely on a set of conventionalindicators that together shed light on performance. These indicators include financial profitability, self-financingratios, productivitygrowth and the share of these enterprises in the overall budgetary deficit and external debt.

2.06 Financial profitability,as an indicator of public industrial enterprise performance,is far from perfect. Good financial performance may measure efficiency but may also result from market distortionsand pricing policies. For example, some of the more profitable public industrial enterprises in Ghana (including the distillery and cocoa factories located in Accra and Takoradi, respectively)owe their financial success almost wholly to monopolistic positions and to very high levels of protection from foreign competition,and they have negative value added at inter- national prices. Until recently, some 90 percent of the total sales of public industrial enterprises in Pakistan were subject to price controls, and this constraint affected the financial profitabilityof enterprisesin such subsectors as fertilizers,cement, automobile assembly and petroleum. However, financial profitabilitydoes indicate whether the public enterprise generates revenue or not and thereforewhether (and how much) it depends on the national treasury for its operations. As will be indicated later, the degree of managerial autonomy of public industrial enterprises, an important determinant of performance,depends to a large extent on the degree of independencefrom treasury financing.

2.07 The data for the sample countries indicate generally low financial rates of return for public industrial enterprises. For example, over the period 1977-83, the 15 largest Portuguese public enterprises achieved an average rate of return on capital employed of about 10 percent, compared to an average rate of inflation of about 21 percent. In Ghana, the comparable rate of return was about 18 percent, compared to an average rate of inflation of 96 percent. Similarly, in Pakistan, the return on capital employed averaged about 2 percent during the 1970s and slightly over 4

4/ While high capital intensity is characteristicof the public enterprise sectors in most sample countries,it is particularlymarked in Ghana. Two sugar factories and a state footwear factory installed conveyors to undertake tasks that are not even mechanized in the U.S. See Tony Killick, "The Role and Performance of the Public Sector in the Industrializationof Some African Developing Countries: An In-Depth Assessment," in The Changing Role of the Public Industrial Sector in Development, (Vienna: UNIDO, 1983). - 13 - percent during 1980-83, compared to an average rate of inflationover the whole period of about 12 percent.5! Consolidatedfinancial accounts for 140 Indian public enterprises indicate an average pre-tax return on capital employed for 1978-83 of about 11 percent in current terms. However, over 90 percent of these profits were in the petroleum sector. Among industrial countries, Austrian public industrial enterprises (in the OIAG Group) registered an average return on capital employed of about 3 percent during 1975-80, and since 1980 many of the firms have encounteredlosses. In Sweden, the Statsforetagenterprises kept fairly even pace with the rest of Swedish industry during 1970-76 in terms of both sales per employee and profitability,even though their return on invested capital as a group reached the 10 percent target only once (in 1974). However, losses were incurred as a group in all subsequent years until 1983.

2.08 Performance in terms of real value added growth has been somewhat better. For example, Brazilian public enterprisesregistered a generally strong performanceduring 1966-75, both in absolute terms and in comparison with industry as a whole. Real value added in large Brazilian public enterprises increased 3.7 times over the period, compared to 2.4 times in industry as a whole. Data are, however, not available to measure the extent to which this reflects a greater level of investment by public enterprises and the degree to which it represents an increase in total factor productivity. Since 1975, such value added increaseshave been significantlylower for both public enterprisesand industry in general. Similarly in Portugal, real value added per employee increasedby 83 percent during 1976-79 for the largest 13 public industrial enterprises. In subsequent years the increase was only 2 percent.

Variations in Performance

2.09 Financial performancevaries considerablyamong public enterprises in the same country and between subsectors across countries. In Israel, within the Israel Chemicals Limited (ICL) group, the most profitable subsidiary--theDead Sea Works--had an average annual return on revalued capital of 12 percent in 1980-83 and 18 percent in 1984, rates that are considered high for the heavy chemical industry. In contrast, another subsidiary in the same group--Fertilizersand Chemicals--hasbeen barely profitable, to some extent because it has concentratedon the domestic market in which fertilizer prices are controlled. In Zambia, some subsidiariesof Indeco (such as Zambia Breweries, Zambia Oxygen, National Breweries, Zamefa, Supa Baking and Chilanga Cement) are profitable,whereas Luangwa Industries,Zambezi Sawmills, Nitrogen Chemicals and Livingstone Motors have been loss-makers. In Ghana, the best performers have been GIHOC's cocoa processing factory, a liquor distillery and a fruit cannery, whereas its two sugar factories have been big loss-makers. In Brazil, PETROBRAS and CVRD have been consistentlyprofitable, together with a few smaller public enterprises (Mafersa, Celma, Tasa, etc.), whereas the iron and steel companies (Siderbras,Acesita and Siderama)have been substantial

5/ The data refer to the public industrial enterprisesunder the juris- diction of the Ministry of Production,which account for most of public industrial value added and employment. - 14 - loss-makers. In India, gross profits (before tax and interest charges) for public industrial enterprises during the early to mid-1970s averaged about 19 percent for the petroleum industry, 11 percent for medium and light engineering, 2 percent for steel and -3 percent for minerals and metals.

2.10 The same variation in performance is visible in the industrial- ized countries in our sample. In Austria, OMVand SGP have been the only consistently profitable public industrial enterprises. On the other hand, the largest contributors to losses have been the two large steel companies, Voest-Alpine and VEW. In France, public industrial enterprises can be classified into three groups based on their profitability: the heavy loss-makers (the two steel firms, Usinor and Sacilor, and, since 1983, Renault); those enterprises that break even or have light losses (mainly Thomson, the electronics and telecommunications company, and EMC, the chemical company); and the profitable firms (such as Dassault, Matra, Saint-Gobain and Roussel).

2.11 Looking at the data across subsectors, variation in profitability can be perceived significant (Table 2). The most profitable subsectors in 1984 for the largest public enterprises in the world (not just in the sample countries) were mining and petroleum, the least profitable ones, steel and automobiles. Almost without exception, steel companies have been the largest contributors to public sector deficits. Excess global capacity and low world prices have resulted in huge losses in many countries, including France, Italy, Sweden, Brazil, Austria, India and Pakistan. The fact that this subsector is considered strategic and is generally dominated by strong trade unions has made cost reduction difficult.

Comparing Public and Private Enterprise Performance

2.12 How does the performance of public industrial enterprises compare with their private counterparts in the sample countries? This question needs to be addressed with caution. Public industrial enterprises are often expected to fulfill at least some social objectives in addition to profit-making ones, and this role can affect their financial performance. Moreover, the public enterprise sector in many countries includes loss- making companies that the governments acquired for non-economic reasons. Also, the public enterprises' portfolio is frequently dominated by slow- growth basic industries even in countries such as France and the United Kingdom. Further problems can arise when too static a view is taken, that is, when, for example, comparison is made between a new inexperienced public enterprise and a well-established private one, or vice-versa. In this sense, comparisons of public and private firms' performance in this study (and most other studies) are not rigorous enough and should be considered as only indicative and illustrative.

2.13 With these caveats, the evidence collected in this study suggests that within the sample countries, and sometimes within the same industrial subsectors, public enterprises have generally performed worse than private ones in terms of financial profitability. Thus, in Israel, a study of a sample of 25 public industrial enterprises and 37 private companies (in which 14 public and 24 private ones were in the same industrial subsectors) - 15 -

Table 2: FITABILM (F PlEL INXETRIALHImS BY SUBS I, 1984

Numberof (kGss Net Net Incma Subsector Numberof Loss-Mkdrg Assets Incae as % of Firns Firm (US$ billion) (US$ billio) (kces Assets yirg a/ 5 1 14,808 1,097 7.4 Petrolenm b/ 26 2 183,356 7,029 3.8 Tobacco 2 0 874 16 1.8 Chiencals 5 2 17,895 293 1.6 Aerospace 3 0 7,751 73 0.9 Electronics 2 1 16,657 66 0.4 Steel ard Metal Mafacturirg 15 7 49,234 -1,853 -3.8 Motor Vehicles 4 2 25,949 -1,467 -5.7

Total/Average for Above 62 14 316,524 5,254 1.7

Source: Based on Table 1 of the Statistical Annex. a/ Average profitability mm substantially affectel by CVRDof Brazil. Without it, the average falls to 2.9 percent. b/ Not included is YPF of Argentina, whoselarge lossesdistort the picture considerably.If YPF vere inclided, average profitability woulddecline to 1.8 percent.

showed considerably lower financial profitability for the former than for the later. 6 / Before-tax profits for public enterprises averaged 1.6 percent of sales in 1976-78, compared to 11.6 percent in private companies. A comparison of the profitability of large public and private Brazilian firms for 1974 and 1978 indicates that the rate of return on equity for public enterprises was on the order of one-half the return earned by the private sector.

2.14 In India, which has one of the largest and most diversified public industrial sectors among the developing countries, the financial performance of public enterprises has been significantly worse than that of private enterprises in the same industry (though not necessarily in the same product line).7 / A recent study of public and private enterprises in India's fertilizer industry concluded that, while productivity performance in the most efficient public enterprises was comparable to that achieved by the best private enterprises, productivity in the public sector as a whole

6/ See Ehud Diskin, "Attitude toward Risk in Israeli Private and Public Enterprises," Ph.D. thesis, Hebrew University of Jerusalem, 1980 (unpublished).

7/ See V. Sri Ram, N. Sharma and K.K.P. Nair, Performance of Public Sector Undertakings (New Delhi: Economic and Scientific Research foundation, 1976). - 16 - was lower than in the private sector.8/ In this case, the differences in performancewere partly explained by the higher input costs in the public enterprises that resulted from their obligation to use indigenous feed- stocks, and because of the older, outmoded technology employed in parts of the public sector (public fertilizer plants were establishedas "pioneer" enterprises in the 1950s, whereas the first private firm came into existence in 1968). To the extent that in most of these countries public enterprises have access to more preferentialrates of interest in borrowing and frequently do not pay interest at all, the difference in performance between public and private firms is even greater.

2.15 In the case of France, the publicly owned Renault more than held its own with the private automobile companies in that country until 1983. However, informationprepared for this study indicates that French public companies were more severely affected by the recessionof 1980-82 than their private counterpartsand that the private sector recovered faster than the public sector. Finally, an analysis of the data for the 500 largest non-U.S. companies on the Fortune list for 1985 demonstratesthe higher profitabilityof private firms (Table 3).

Table 3: COMPARISON OF FINANCIAL PROFITABILITY,1984 (net income as % of gross assets)

Public Private Subsector Industrial Industrial Enterprises Enterprises

Mining 7.4 15.5 Petroleum 3.8 4.4 Tobacco 1.8 8.5 Chemicals 1.6 3.5 Aerospace 0.9 2.9 Electronics 0.4 3.6 Steel and Metal Manufacturing -3.8 0.7 Motor Vehicles -5.7 3.3

Weighted Average for Above 1.7 4.0

Source: "The Largest Non-U.S. Corporation,"Fortune, August 19, 1985.

8/ See M. Gupta, "ProductivityPerformance of the Public and the Private Sectors in India: A Case Study of the Fertilizer Industry," Indian Economic Review, Vol. 19 (1982). - 17 -

Conclusion

2.16 A portion of the relatively poor performance of public industrial enterprises can be explained by their continued concentrationin product lines and basic industries (such as steel, automobilesand fertilizers) that have experienced little growth in recent years, or in developingnew industries (such as aerospace) that are still in their infancy, and by the takeover of ailing private sector firms. However, it is still true that these enterprises have contributed significantlyto the macroeconomic problems of many countries. For example, in Brazil during 1980-82, the major federal public enterprises accounted for close to 50 percent of the overall deficit of the consolidatedfederal government. In India, the net transfers (current and capital) from the central government to public industrial enterprises increased from about US$890 million in 1978 to US$2,320 million in 1983. The public enterprise sector has also been responsible for a significantportion of the external debt in a number of countries.9/ Most notably, Brazil's public enterprisesaccount for over two-thirds of the country's external public debt. In Zambia, the internationaldebt of ZIMCO alone accounts for over half of the total external public debt. At the end of 1982, public enterprises accounted for about 60 percent of Portugal's external public debt of US$13.5 billion and some 90 percent of the country's short-term external liabilitiesof US$4 billion.

2.17 The response of several developing countries to their mounting economic problems has been to impose restrictivemonetary policies. In the absence of greater emphasis on fiscal constraints and with somewhat more lax controls on public sector expansion and easier access to domestic and foreign credit, there has been some "crowdingout" of private sector activities even in countries such as Brazil and Turkey (in the late 1970s) that attach great importance to private sector development. At the same time, the result has been excessively high real interest rates and severe financial distress. A more appropriatepolicy response may be to give greater priority to reducing the public sector deficit in general, and reducing the financial support to public enterprisesin particular. This would help in relaxing somewhat the constraintson credit to private producers. This policy mix can elicit a greater supply response from the economy.

2.18 Since public industrial enterprisesare such a significantpart of the industrial sector in most countries, efforts at making them more efficient have become imperative. In the following chapters, issues related to improving the performance of public industrial enterprisesare examined, using evidence gathered in this study.

9/ In some cases, of course, the external borrowing of public enterprises was not voluntary but forced on them by governments that used the creditworthinessof these enterprises to contract external loans. - 18 -

CHAPTER 3

A COMPETITIVE ENVIRONMENTAS A DETERMINANTOF PERFORMANCE

3.01 In view of the many determinants,mostly non-quantifiable,of the performance of public industrialenterprises, no rigorous statistical analysis exists of the correlation between the degree of competition to which public enterprises are exposed and their performance. The present study is no exception. However, evidence from the sample countries strongly indicates that public industrial enterprisessubject to a more demanding, competitiveenvironment, whether domestic or foreign, tend to perform better.

3.02 The post-World War II experience of French and Italian public ienterprises,particularly of Finsider in the Italian steel industry, of Renault in the French automobile industry and, to a lesser extent, of SNIAS in the French aviation industry, indicates that more open trade policies and substantialdomestic competitionhave kept public firms in industry under pressure to control costs, limit economic waste and rationalize the use of resources for non-commercialgoals. In Brazil, the success of CVRD is explained in large part by the fact that it has had to survive in the fiercely competitive internationalmarkets for iron ore. The same is true for some of the petrochemicalsubsidiaries of PETROBRAS in that country, and for OCP (phosphates)of Morocco. Israel Chemicals Ltd (ICL) is another good example where foreign competitionhas engendered superior performance. ICL would probably not have been significantlymore profitable had its ownership been private. In India, the publicly owned Richardson and Cruddas has faced severe competition in structuralsand industrialmachin- ery from the private company, Testeels. The results are evident, despite ,problemsduring the late 1970s owing to the repercussionsof the then newly acquired Muland plant and the disturbed industrial relations. In Zambia, three of the seven best financiallyperforming public industrialfirms are exporters (Zambia Oxygen, Zamefa and Chilanga Cement), while two of the five moderate performers (Kafue Textiles and Zambia Sugar) are small exporters. None of the poor performers are exporters.

3.03 The outstandingperformance of Hindustan Machine Tools (HMT) of India can be explained in large measure by the fact that, since the reces- sion of 1966-69 it has placed greater emphasis on exporting, marketing and profitability.16/ It has also encountered fierce domestic competition in watches, machine tools, tractors and lamps. Thus, even though the enterprise was established initially for import substitutionin the machine tools subsector,and even though it was forced by the government to go reluctantly into watch, tractor and lamp production,it has performed

10/ For details, see Pradip Khandwalla: The PerformanceDeterminants of Indian Public Enterprises,a study prepared for the World Bank, 1981. - 19 - remarkably well. Why? Admittedly it has enjoyed significantoperational autonomy and substantialministerial support, but so have the other less successful public enterprises,BPCL and BHPV, in the same (engineering) subsector. Domestic and foreign competition,which leave no room for complacency,appears to have been an important factor.

3.04 Of course, this is not to say that the response to the competitive environment is any less important than the environmentitself. Public companies in the same competitive environmenthave reacted differently, depending on their management. Occasionally,changes in leadership in the same enterprise have led to better use of the environment. Within the IRI complex of Italy, the weaker performance of the engineeringcompanies in the early post-war years compared to that of the steel firms can be explained almost completely in terms of differencesin the quality of management which IRI failed to rectify. The point is not, therefore, that competitive pressures are the determining factor in all cases, but rather that, given other factors, competitivepressures tend to bring out the best in public enterprise management. An environmentwhere challenges are non-existentcan suffocate the initiative of the best managers, public or private.

Fostering a CompetitiveEnvironment

3.05 Inappropriatemacroeconomic, trade and industrial policies are the most potent constraintson competitionfor both public and private firms. These policies include the maintenance of an overvalued exchange rate, reliance on quantitativerestrictions and bans, high tariff protection, price controls, domestic content legislation (especiallyin the automobiles and electronics subsectors),investment licensing, regulations on entry and exit and other regulatory constraints.

3.06 The rationalizationof these policies can improve the competitive environment considerably,but there are also other ways that are more specific to public enterprises. First, governments can make explicit attempts to nurture competition among public industrial enterprisesand between public and private firms. Evidence indicates that such competition can occur even among public enterprises that are considered strategic. A good example is the Heavy Mechanical Complex (HMC) at Taxila in Pakistan. The company faces competition in the production of road rollers and sugar mills from the privatized firm Ittefaq Foundry. It has to compete in the construction of cement plants not only with imports (allowed in with a duty of 20 percent), but also with another public enterprise,Karachi Shipyard. It also faces competition in its production of electrical towers, boilers and overhead travelling cranes not only from private firms but also from the publicly owned Pakistan Engineering Company (PECO). Similar competition, but on a much more intensive scale, exists among the various petrochemical subsidiariesof PETROBRAS in Brazil.

3.07 There is every reason to believe that Leon Festinger's *"social comparison theory" (which states that "people most strenuously seek to evaluate their performance by comparing themselves to others, not by using - 20 - absolute standards") 'V applies as well to public enterprises. Central monitoring agencies should therefore make informationon comparative performance widely available. In countrieswhere performancetargets are reviewed by such agencies, at least some of the meetings should be devoted to a comparative analysis of performancewith public managers under the same roof. The results could be widely publicized in the national press. It is surprising how many public enterprise managers interviewedin Brazil and Pakistan (two countrieswhere legislative and pressure group controls have been virtually non-existentduring most.of the last two decades) stressed their fear of unfavorable coverage in the national press.

3.08 Another way to subject public enterprises to competitivepressures is to encourage them to seek export markets. Of course, no amount of encouragementor cajoling can make them export if the policy frameworkhas a strong anti-export bias built into it, or if the institutional,infrastructural and marketing constraintsare severe. Where these biases and constraints are surmountable,a small shift of the products to export markets can have a sizable impact on the quality of management and competi- itiveness,even if the impact on the balance of payments is minimal. After all, only about 8 percent of the total sales of India's HMT are exported, blutthis level appears to be sufficient to give the company access to new management techniques and new technologies.

3.09 In industries in which economies of scale are not important, competition can also be enhanced by breaking up overly centralizedand monolithic holding companies or even large industrialenterprises. The lbreak-upof Statsforetag in Sweden into smaller sectoral holding companies has definitely increased competitionand made public enterprisesmore ,zonsciousof efficiency considerations. The proposal of the new Bolivian government to break up the Bolivian Mining Corporation (COMIBOL)into several competing mining enterprises could be a useful experiment. Occasionally, the break-up of large enterprises may be necessary but not sufficient for better performance. For example, the breaking up of FCI in the early 1970s in India resulted in improvementin performanceto a certain point only, mainly because the regulatory environmentthat hampered operations remained unchanged. This result stresses, once again, the importance of a favorable regulatory and policy environmentwithin which enterprise-specificchanges can 'beeffected.

:3.10 In short, public enterprises should, to the largest extent possi- ible,be subjected to competitivepressures. Toward this end, consideration should be given to introducingpublic enterpriseson stock exchanges, establishing joint ventures with the private sector, stimulatingexport initiatives,breaking up monopoly firms, allowing competitionamong public enterprises, and breaking up or reorganizingrigid conglomeratesinto more flexible and dynamic bodies.

11/ Quoted by Peters and Waterman, In Search of Excellence (New York: Harper and Row, 1982). - 21 -

CHAPTER 4

FINANCIAL AUTONOMY AND ACCOUNTABILITY

4.01 One of the major causes of the financial and economic problems that most large public enterprise sectors around the world currently face is the lack of a clear separation between the management and operations of enterprises on the one hand and political considerationsand processes on the other. In many cases, this results in diffuse and conflictingobjec- tives, limitationson managerial autonomy and accountabilityrequired for efficient commercialoperations, and a civil service culture where chief executive officers (CEOs) are more administratorsthan entrepreneurial businessmen. In such an environment,employment, investmentand pricing decisions are frequentlymade without due considerationto the financial consequences.

4.02 These roots of public enterprise sector problems are in many cases only the starting point of a vicious circle that reinforcesitself until a crisis is precipitated. Poor investment choices, low productivity and highly leveraged capital structures create losses, reduce a firm's ability to raise resources from regular capital markets and banks and drives managers to justify poor results with reference to a multitude of social objectives. As a result, employee morale and financial discipline are further eroded. For example, in Portugal it was found, among the 15 largest public enterprisesvisited, that the loss-makingones on average had increased salaries significantlymore than the profit-makingones. As the financial problems accelerate, the government finds itself forced to centralize decision-makingand controls further. Massive amounts of capital are injected into the public enterprise sector in desperate rescue operations; special financial measures have included the "Participate Bonds" in France, "Endowment Funds" in Italy, "Public Enterprise-Bonds"to be serviced by the government in Austria, conversion of government debt to equity in Turkey, and cancellationof interest payments in India. However, financial support alone gives only temporary relief, unless it is combined with comprehensivephysical and managerial restructuringmeasures. Unfor- tunately, in many cases a series of costly crises has to take place before effective corrective actions are taken. This vicious circle is illustrated in Figure 2.

4.03 A fundamental prerequisitefor improving public enterprise sector performance is to delegate more decision-makingpower to enterprise managers. Among the elements of autonomy that managers interviewedfor this study cherished most was financial autonomy, in the sense of substantial freedom from reliance on treasury financing. The two French steel companies (Usinor and Sacilor) have been chronic loss-makers,and as a result all major decisions are subject to prior approval by their tutelle authorities. By contrast, Saint-Gobainand Roussel, essentially profitable, have enjoyed the greatest possible managerial autonomy. Similarly in Sweden, the profitableTobacco Company has experiencedsignificantly more freedom than the loss-makingNJA (steel) company, even though both are subsidiaries of the same holding company, Statsforetag. The same conditions largely explain the greater autonomy of PETROBRAS and CVRD in Brazil - 22- Figure2: ViciousCircle of PublicEnterprise Sector Management STARTINGPOINT 1. Lackof managerial autonomy & accountablltty

2. Weakfinancial B. Further strsngthening discipline of centralcontrols and decistonmaking

3. Poor Investmentchoices, 7. Lessmanagerlal low productivity5 autonomyand excessiveleverage accountabilfty

~~~~/ 4. Flnanciallosses 6. Flnoncial lustifled wlth crisis social objectives

Deteriorating morale. further reduced financtal dlsclpllne as compared with the steel companies and the railroad company (RFFSA). Israel Chemicals Limited (ICL), one of the most autonomous public enterprises, has had no major government contributionssince it was established in 1967, depending instead on the retention of profits over the years.

4.04 What constitutes financial autonomy? What are the major financial elements of an environment conducive to good performance? How can financial discipline be improved, and how can public enterprisesbe made financially more accountable? Experience indicates two major components: (i) an overall financial framework in which to manage the public enterprise sector; and (ii) supportive financial policies. The overall financial framework is the driving force behind the establishmentof a productive managerial environment. The supportivepolicies and organizationneed to be designed to support and reinforce the framework.

Overall Financial Framework

4.05 The success of government officials in managing public enterprises depends crucially on a number of financial issues. These include: (i) the extent to which the enterprisesare exposed to greater market discipline; (ii) the importance of financial objectives, includingpayment of dividends; (iii) the clarity of social objectives;and (iv) the degree of restriction on access to subsidies from the state. - 23 -

Greater Market Discipline

4.06 Subjecting public enterprises to greater market discipline can be a potent way of improving financial performance. A combinationof more autonomy in financialmatters and greater exposure to, and control by, markets normally creates among public enterprise managers a stronger sense of responsibilityand a more genuine interest in the financialviability of the enterprise.

4.07 Access of public enterprises to capital markets is an important way of institutionalizingfinancial discipline. Independentbankers scrutinize investment and financing decisions before they extend funds. The use of bond markets, equity markets and joint ventures open the public enterprise to public scrutiny and to special audits and reporting requirements. Interviews with public industrial enterprise managers in Israel suggested that in enterprises whose minority shares are traded on the Tel Aviv Stock Exchange (such as Dead Sea Works, Chemicals and Fertilizers,and Cables of Zion), financial performance and discipline have been improved because of the greater transparencyresulting from the publication of semi-annual financial statements and their potential exposure to legal suits by minority shareholdersfor non-businesslike management.

4.08 Public enterprises should be encouraged to raise, to the extent possible, all their debt financing outside the government and the central bank. Whenever government debt financing is used, market interest rates should be charged; and joint ventures and floating of minority equity positions on stock exchanges should be encouraged.

Financial Objectives and Dividend Policies

4.09 Sound financialmanagement is a prerequisitefor exposure of public enterprises to the discipline and control of financialmarkets. Sound financial management entails specifying of financial objectives, establishing a mechanism to monitor these indicators,and holding managers accountable for the outcome. The main measures to be used include: return on assets or capital employed, return on equity, dividend pay-out ratio, maximum debt equity ratio and a pre-specifiedlevel of internal financing of large investment projects. The ratios are to a varying degree being used by public enterprise managers in the sample countries. One exception is Austria, where enterprise financial objectives are not specified explicitly but are couched in terms of "satisfactory"(undefined) profitability and cash flow. Officials of the public industrial enterprise overseeing unit in Sweden underlined that, in an effort to increase financial discipline and performanceamong public enterprises,the establishment of broad financial objectives and enforcement of a clear dividend policy are even more important in state-owned enterprises than in private firms, because of the former's inherent bias toward giving excessive priority to non-financial objectives. They further stressed that even though the dividends in many cases may be small, frequently the principle of paying them is more important than the amount. The enforcement of dividend policy is a highly visible and substantiveaction demonstratingthat the government gives high priority to financialmanagement and that state equity is - 24 - not considered free goods. It also forces management to rethink its growth strategies and ideally to focus on productivityand profitabilityimprove- ments as ways to finance the "new" financial obligationscreated by the dividend requirement.

4.10 In most countries in this study, dividend payments were either insignificantor non-existent. This pattern does not mean that dividend policy has been unimportant,but rather that it has been neglected for social reasons or because of the inability of public enterprises to pay.

4.11 What amount of dividends should an enterprise pay, and how should dividend policies be formulated? The policies will vary among enterprises depending on factors such as risk, growth potential, asset structure and inflation rates. As a starting point, the government should determine for each enterprise the required rate of after-tax return on equity capital. Countries with significantinflation will have to include revaluationof assets in the calculation of return. Total target return on equity, direct and indirect, normally is somewhat higher than the cost of debt and varies with the levels of financial and business risk. A portion of this return, limited in some countries by law to the direct return, can then be set as the dividend pay-out target. Typically, enterpriseswith low growth and low investmentspay out a majority of their after-tax profits. Conversely, enterprises with high growth and high investment requirementshave relatively low dividend pay-out ratios. Generally, dividends vary between 25 percent and 75 percent of after-tax profits. While profits may vary significantly between years, enterprises typically strive to pay stable dividends that grow at a steady rate at or above the inflation rate. This objective means that dividend pay-outs can be more than 100 percent of after-tax profits in bad years and a much smaller percentage in boom years.

Clarificationof Social Objectives

4.12 Public enterprises are frequently expected to pursue a variety of social objectives,which can include such diverse g ls as redistributing income, subsidizingparticular regions of a countryid creating or maintaining employment. The fulfillment of at least some social objectives by public enterprisesmay be necessary and, in some cases, even advisable.

4.13 The problem is not the fulfillmentof these objectives. Even private enterprises have some social objectives,such as maintaining excess employment during certain periods and providing adequate housing, education and health facilities to their employees. The problem is that the multiplicity of objectives and the absence of priorities among them lead to a situation where social objectives become an excuse for poor financial performance. In contrast, the watchword of Texas Instruments of the U.S. is "more than two objectives is no objectives." In other cases, social objectives become so predominant that they overwhelm the main objective of the public enterprise as a productive unit. Many examples of such situations exist in both developing and developed countries. Perhaps the best examples in our sa!mpleof countries are the bicycle plant at Chipata in Zambia, the Aboso Glass Factory and the Bolgatanga Meat Factory in Ghana - 25 - and the Bolan and Shahdadkot textile mills in Pakistan. Among the developed countries, a good example is the state-ownedaluminum company (ASV) in Norway, whose operations are substantiallyconstrained by the need to provide employment in a region where few alternativeemployment opportunities exist. Social objectives have also been extremely important for CDF-Chimie (heavy chemicals) and EMC (chemicals)in France. In the case of EMC, it has been estimated that the cost of rendering specific social services exceeded its overall deficit in 1984.

4.14 In using public industrial enterprises for social purposes, governments need to address two relevant issues. First, are the social objectives truly being fulfilled? Second, are there no better alternatives for achieving them?

4.15 There is evidence that the pursuit of social objectives by public enterprises can lead to inadvertentand perverse results. The case of Zambia's publicly owned agro-based industries (largely producing vegetable oil products, detergents and soap) is illustrative. Stringent price controls on refined oil and fats (only recently lifted) have led not only to large financial losses but also to low morale and a shift in the product mix away from the production of oils and fats. This result was precisely the opposite of the government's social priorities. Price controls are intended to benefit the poor, but end up subsidizinglarge industrial users, wholesalers or the upper middle class in urban areas. The State Cement Corporation of Pakistan (SCCP) has in the past sold cement in Lahore at Rs.2 per bag below the cost of production plus freight charges (from its Zeal Pak plant at Hyderabad, some 700 miles away) essentially to appease the urban elite in the city where the head office is located. Similarly excessive efforts to maintain or expand employment generally result in costly losses of efficiency and low morale.12/

4.16 In many instances, there may be preferred alternativesto the use of public enterprises as agents of social change. The objective of region- ally balanced economic growth is better achieved through an investment encouragementscheme that consists not only of income and corporate tax credits but also of direct cash grants from the treasury. The impressive industrial growth of the economicallydepressed North East region of Brazil over the last two decades was achieved with hardly any public investment. Often it is better to allow public enterprisesto operate on commercial, profit-seekinglines and then use their profits to achieve social goals such as income redistributionand employment creation.

12/ For further examples of such unintended results, see Mary Shirley, Managing State-OwnedEnterprises, World Bank Staff Working Paper No. 577 (Washington,D.C., 1983). - 26 -

4.17 Even in cases where investment decisions have been taken on social grounds, the operational decisions need not be.13/ For example, in a number of the sample countries, governments have establishedlarge inte- grated iron and steel plants for strategic and national security reasons. These non-commercialobjectives are achieved once the plant is set up and initiates operations. However, the plant can nonetheless be operated along commercial lines. It may well earn a lower financialreturn as a result of poor location or absence of comparativeadvantage, but society stands to gain by allowing the plant to earn as high a return as possible.

4.18 In a number of countries, especially France, Italy and Sweden, public enterprisesare compensatedexplicitly on a contractualbasis through the budget for undertaking social objectives.14/ This approach is feasible in terms of preventing public enterprises from being saddled with excessive social burdens and, at the same time, avoiding random departures from efficiency criteria. However, such compensatorypayments for undertaking non-commercialobjectives should be viewed as second-bestor even third-best solutions and should be used sparingly. Moreover, if the contracts'becometoo frequentand large'in relation to a public enterprise s normal activities, management may lose sight of its primary objective. Once such attitudes become entrenched, it is costly and time- consuming to change them. It is for these reasons that in Sweden, where the system of "contract offers" has been in existence for a considerable length of time, has limited its application to only a handful of cases. Even in those few cases, the contracts have been for a restricted time and subject to ex-post evaluation. The potential problems explain why Norway has shied away from such mechanisms.

Elimination of Government Subsidies

4.19 Public enterprisesmust learn to be financiallyautonomous. Easy recourse to government subsidies for operating purposes in most cases reduces the pressure on public managers to close non-viable operations, divest non-strategicbusinesses to free up capital and initiate tough cost cutting programs, all actions that normally are necessary for long-term survival in the private sector.

4.20 In cases where the enterprises are heavily dependent on subsidies, the government and management should agree on a plan for phasing them out over a limited period. This action may have to be matched with a complementaryprogram of physical and financial restructuring.

13/ This argument is developed by Leroy Jones, "The Linkage between Objectives and Control Mechanisms in the Public ManufacturingSector," in The Changing Role of the Public Industrial Sector in Development (UNIDO, Vienna, 1983).

14/ This approach was recommended by the 1968 Nora Report in France. See Groupe de Travail de Comite Interministerialdes Enterprises Publi- ques, Rapport sur les Enterprises Publiques (Paris: La Documentation Francaise, 1969). - 27 -

4.21 In principle, public enterprises should receive government funds only under the following circumstances: (i) increases in equity capital to finance viable expansion projects; (ii) increases in equity capital as part of restructuringoperations aimed at creating viable operations;and (iii) subsidies for performing non-economic social functions, as explicitly agreed between the government and an enterprise. The latter type of agreement should be used only in rare cases where the economic impact of the social functions is substantial.

4.22 Elimination of subsidies is one of the strongest actions that a government can take to substantiatea reorientationin financial policies and to force more economicallyoriented behavior on public managers.

Supportive Financial Policies

4.23 To support the decentralizedmanagement model described above, governments would need to establish and enforce clear capital structure policies and exercise greater discipline and transparencyin financial relations.

Appropriate Capital Structure Policies

4.24 Several major benefits arise from using a proper mix of debt and equity, as opposed to using mostly or only debt in financing public enterprises. Such a mix: (i) increases the enterprises'ability to borrow in financial markets; (ii) serves as a check against overexpansion; (iii) stimulates a profit orientation;and (iv) reduces financial risk and strengthens the ability to survive downturns in the economy.

4.25 (i) Increased Ability to Borrow in Financial Markets. With an appropriate level of equity capital in their balance sheets, public managers will be in a better position to negotiate financing with banks based on the strength of those balance sheets and the quality of their cash flow. Where investment and financing decisions can be implementedwithout government support, the board of directors should be given full authority to approve. Naturally, a state-ownedenterprise can seldom function precisely the same way that a private enterprise can, partly because of the inherent difference between the two types of ownership and partly because of "imperfections"in the banking system, i.e., both state and privately owned banks may tend to lend to state enterprises based on implied government guarantees. Many countries have, in spite of market imperfections,a high degree of market dynamism, that can contribute importantlyto efficient market-orientedbehavior.

4.26 (ii) Check against Overexpansion. A debt/equity target serves as a check against overexpansionby relating an expansion of assets to the rate of profitabilityand the growth of retained earnings. If the debt/ equity ratio deterioratesbelow a certain point, banks will normally not be willing to provide further financing without government guarantees. Equally important, the commitment to proper debt/equityfinancing can also serve as an efficient check against overexpansionaryinvestment decisions by public officials, as it ties the investment decision directly to the equity financing decision. - 28 -

4.27 (iii) Stimulating Profit Orientation. With a proper debt/equity ratio and satisfactoryprofitability, public enterpriseswill be able to borrow from banks and expand their operations based on their own capacity to generate retained earnings. This possibilitygives management an incentive to focus on profitability,since it provides the base for expanding enterprise operations.

4.28 (iv) Reduced Financial Risk. A further argument for using a proper mix of debt and equity financing,as opposed to only debt financing, is that it reduces the enterprises'financial risk. During downturns in the economy, enterprises financed with equity will have lower debt service obligations and will therefore be less likely to encounter acute liquidity problems. In many enterprises,the lack of a capital structure policy has resulted in overexpansion. Debt has been allowed to finance expansions and losses to such an extent that in many cases the enterprises have negative equity capital. At the same time, many governmentshave expanded their overall financial commitments to such an extent that they have very limited resources available for assisting these enterprises. In this situation, enterprisemanagements tend, for prolonged periods, to be preoccupiedwith day-to-day liabilitymanagement, with consequent reductionsin operational efficiency and often losses of market shares and competitiveness. With a proper debt/equity policy, public enterpriseswill be better equipped to avoid the above type of solvency and liquidity crisis.

Greater Discipline and Transparencyin Financial Relations

4.29 A review of the accounts of the major public enterprises in our sample indicated a complicated maze of transfers between central budgetary authorities and public enterprises. To improve financialdiscipline and transparencyin financial relations, and to be able to record "true" financial results, governments should demand tax, duty, interest, dividend and amortizationpayments from public enterprises the same way they do from private enterprises. Losses and negative cash flow would then have to be financed explicitly in a more transparentway through new equity, loans, etc. It is important that the "true" financial situation of enterprises be reflected in the dialogue between public enterprise managers and the government about the enterprises'future development.

4.30 If governments are to require sharp financialdiscipline from public enterprises,they must demonstrate discipline and effectivenessin their own internal financial management. Some guidelines to this end are: (i) governments' commitments to pay in new equity capital as part of large investment projects should duly consider the risks involved and their own capacity to finance these commitments; (ii) governments'payments to public enterprises approved by appropriate ministry(ies)should not be questioned and delayed by other ministries; (iii) the semi-annual issuance of approved payments should be more or less automatic without further bureaucratic delays; and (iv) public enterprises should not be used to finance government deficits.

4.31 Other factors key to financial autonomy and accountabilityare the disposition of profits after taxes and the dispositionof liquid - 29 - assets. Under normal circumstances,public enterprisesshould be allowed to transfer to retained earnings all profits after taxes and payment of dividends. In inflationaryenvironments, many public enterprises,even those with zero real growth, need significant amounts of retained earnings just to maintain their debt/equity ratio. Public enterpriseswith real growth potential should be given the freedom to exploit these opportunities with internally generated funds. A similar argument applies to the disposition of excess cash/liquidassets. Viable enterprisesshould be allowed to decide how to redeploy excess cash generated from operations and/or by divestiture. The rationale is that public enterprise managers who have successfullygenerated a positive cash balance normally are in a better position to decide upon its productive use than are government officials more distant from the market place.

Performance Measurement and Incentives Systems

4.32 An effective system for performancemeasurement and control should be established. It can serve the government in establishingmanage- rial accountabilityin a number of ways. It reveals whether, and the extent to which, the boards of directors of the public enterprisesare meeting their financial commitments. It emphasizes to the board and management the importance placed on financial objectives, performanceand discipline. It is a basis for rewards and punishments. And it provides the government a tool to control overall credit expansion in parts or all of the public enterprisessector.

4.33 The level of detail entering a financial system will vary from country to country and between enterpriseswithin countries. In general, the more an enterprise is controlled by competitivemarket forces and the less it is dependent on government financial support, the less detailed will the centralizeddata requirementsbe (see Table 4). In countries such as France, Norway and Sweden, it may be sufficient to collect the enterprises' formal annual reports. On the other hand in countries such as Ghana and Zambia, the governmentsmay need to request quarterly financial statements and production and personnel data, and may publish an annual summary of the performanceof all public enterprisesat the end of the year.

4.34 The performancemonitoring system would need to be linked to an incentive system with different levels of rewards including promotions, financial bonuses, fringe benefits, official and non-officialrecognition of achievement and publication of annual results. These incentives are most effective when they are used in a relatively competitiveenvironment and where a high degree of financial and managerial autonomy is granted to management, the latter factor being a substantialincentive in itself. In highly protected economies, such bonuses would need to be used with caution since they may well increase the incentive for non-optimal and inefficient expansion of operations. - 30 -

Table 4: LEVELS OF FINANCIAL CONTROL

Financially Public Enter- Financially Independent prises Independent Public Enter- Dependent on Public prises in Mono- Government Financial Control Tools Enterprises poly Markets Finance

Strategic Plan * * * Annual Operating Budget * Capital Investment Plan * * Quarterly Financial Reports * * Report on Capital Projects * Audited Financial Statements * * * Management Audits * Price/TariffAssessments *

Conclusion

4.35 To summarize, public enterprises should be allowed, to the extent possible, to operate freely within the confines of their business charters and the capacity of their balance sheets. Under these conditions,the government should concentrate its decision-makingand control functions on issues related to its ownership role, while decisions related to strategy and operations should be delegated to the board of directors and management. Governments should establish a minimum rate of dividend payments from all companies. For loss-makingfirms, a coherent and time-bound program should be established to eliminate the causes of financial loss so that, at the end of the period, such entities will be able to distribute dividends. The program would include the eliminationof subsidies and lpricecontrols, more emphasis on performancerelative to financial objectives, more appropriatedebt/equity policies, loosening of the borrowing constraints on public enterprises,and more transparencyin the financial relationshipbetween the government, public enterprisesand state-owned financial institutions. - 31 -

CHAPTER5

MANAGERIALAUTONOMY AND ACCOUNTABILITY

5.01 Two key determinantsof the performance of public industrial enterprises are the extent to which the government decentralizesdecision- making to enterprise managers, and how it manages the organizationalinter- face between it and enterprises. This chapter and the next one address the crucial issue of how much decision-makingpower should be delegated and what managerial environment can facilitate this process. Once the government has decided on the magnitude of its role in controlling public enterprises, the other issue is how it organizes its various components (ministries,specialized agencies, etc.) to enable it to exercise effective control, the subject of Chapter 7.

Three Roles in Enterprise Management

5.02 In private enterprises,there are three distinct roles pertaining to owners and managers: ownership, strategic and operating. The ownership role, performed by the owner(s), involves defining the business charter, setting the overall objectives of the enterprise, appointing and dismissing the board of directors, and approving annual accounts and dividend payments. These powers are formally executed through an annual shareholders' meeting. The strategic role is the responsibilityof the board of directors. Its major functions, discussed in detail in Annex II, are to decide upon strategic plans, monitor performanceagainst targets, and appoint, advise and dismiss the chief executive officer. Typically, the board is part-time,meeting four to eight times a year, and is composed of persons with high stature and professionalexperience relevant to the business of the firm. The operating role is performed by the chief executive officer (CEO) and his top management. Their major responsibilityis to manage and develop the enterprise in accordancewith agreed objectives (Table 5).

5.03 This separation of the three distinct roles, which came about because of more widespread ownership and because owners found it difficult to manage an increasing number of diverse enterprises,has stood the test of decades of industrial development and is prevalent in the private sectors of western industrializedcountries and increasinglyin most developing countrieswith "mixed" economies.

5.04 How do public enterprise organizationalstructures deviate from this prototype? The relatively short experience of many governmentswith the management of public enterprises,the excessive lines of control, the multitudinous (and occasionallyconflicting) objectives and tasks they are assigned, and the lack of adequate information flows between public enterprises and government, invariably lead to encroachmentby governments (as owners) on the strategic and operationalfunctions of public enterprises. This breakdown in the demarcation of the major functions is the source of much of the poor performance of public enterprises. - 32 -

Table 5: CONVENTIONALROLES IN CORPORATEMANAGEMENT

Roles and Functions Basic Characteristics

o Ownership Role (Owners) - Define enterprise charter - Long-term view - Appoint and dismiss the - Broad objectives Board of Directors - Broad, non-involved, - Approve annual accounts monitoring of performance ex post o Strategic Role (Board of Directors) - Provide strategic guidance - Medium-termview - Control performance - Specific strategiesand - Appoint, advise, reward operational targets and dismiss the chief - Close periodic monitoring of executive officer (CEO) performance vis-a-vis strategic objectives and o OperatingRole (Management) operational targets - Develop and manage the enterprise according to - Shorter-termview agreed upon objectives - Developmentof alternative strategies,plans and programs - Management of resources - Detailed day-to-daydecision- making and control of operational performance

5.05 A simple example will illustrate this breakdown. In a number of the sample countries,boards of directors are obliged to seek ministerial approval for relatively small investments,hiring and firing of staff, wage-setting, working capital decision-making,procurement policies, foreign travel, and much else. In a private enterprise, the power to approve these actions is almost always delegated by the shareholdersto the board of directors and further to the CEO. The logic is that the board of directors and the CEO will be judged on the basis of the financial performance of their decisions. The board can, therefore,exercise its control function by discussingand approving strategic choices, setting broad financial objectives and examining performancerather than the processes by which the outcomes are generated. The private sector control philosophy centers on control of viability, while the public administration'sphilo- sophy, in many instances,is characterizedmore by compliancewith rules and regulations. Given that the strategic role is mostly exercised by ministries, public enterpriseboards tend to be more involved in operating decisions, which impinge on the normal operating role of the CEO as executed in the private sector. In certain cases, it may be justified for governments to involve themselvesin certain decisions that private sector firms would handle at the board level. For example, governmentsmay get involved in the investment decision because of its macroeconomic impact on the overall fiscal deficit and concern about the quality of projects. As a result, high government officials in countries such as Ghana, India, - 33 -

Pakistan and Zambia are extensively involved in detailed investment decisions. There are several problems with this process: it is time-consuming for an already over-stretchedbureaucracy; the ministries lack the information and the business perspective to make the correct decisions; and, most important, it absolves the management of its responsibilityand accountability for enterprise performance.

5.06 When performancecannot be quantified and when an effective control structure does not exist, it becomes impossibleto exercise effective control of performance. When governments cannot control performance, they attempt to control processes. And ineffectivecontrol frequently leads to excessive interference.

Evolution of Public Enterprise ManagementModels

5.07 The sample countries included in this study provide a rich variety of organizationalstructures. The most highly decentralizedstruc- tures are those in Sweden and Norway. Intermediateones occur in France, Austria, Italy, Brazil, Portugal and Israel, while highly centralizedones are found in Ghana, India, Pakistan and Tunisia (Table 6).

5.08 A comparison of the two polar cases--Swedenand Norway on the one hand and Ghana on the other--is instructive. In Sweden, the Cabinet, the formal owner of public enterprises,has delegated this responsibilityto the Minister of Industry. Within the ministry, a Unit for State Participa- tions (Unit 5), staffed by eight professionals,actually execute the ownership role for the public industrialenterprises, which employ some 90,000 people. A similar number of staff within the Ministry of Industry in Norway perform the ownership role for public industrial enterprises there, which employ some 50,000 people.

5.09 In both countries, the ownership role is to define the charter, including broad objectives/purpose,of individual enterprises,to hire and fire members of the board of directors, to establish dividend objectives, to assure that strategic and financial plans are developed by the board of directors, to approve annual accounts and the allocation of profits, and to monitor performance. All major decision-makingregarding multi-annual plans, annual budgets, investments,financing, personnel policies, etc., are decentralizedto the board of directors of individual enterprises. The Unit for State Participationsdoes not intervene in the boards' decision- making, does not maintain a data bank of financial and/or operational enterprise data, does not keep a roster of management candidatesand does not perform any management services. Executives of Unit 5 participate as directors only on a limited number of boards. They stress that in their role as board members they take special care to act only in the best interests of the enterprise and not to overpower the board with their ownership role. Only in times of crisis and when state financial support is required by an enterprise does the Unit for State Participationsstep up its participationin decision-making. Most public enterprisesare organized as limited liability stock corporationsand are regulated by corporate law. The CEOs of Swedish public enterpriseshave the same responsibility for developmentand operations as do their counterpartsin the private Tal. 6: l W4DNtr IOS PMt FVUI D1UKtALwInInF IN WUIUd (lS R

fm S5m PR*=I lA&ILY PF1M DMI PCSIN GUM ZMIA tAZIL W L n N lXnnIT gnu Itr mid" SU Ueedt withd n 1w Beall wit e re enr8- Noneal Sectoral Vay lar enr- Ir aw't he- sata VEtacrdaee Sll epd_-h_teold- mi Scar_1l we-eret (It Ns1 Noc_tna iAliS MeM.Of TAtrY Ird. of Ilntl MM o t "I l thid (*)n the mnt uedit - ielatrin ed wnet edelrety. cra. Sectoml CMa1em nW It Is forca poiet Menistries, Itnistry Aelty within Melatrin ne is foai Felt. to fcwl Port. .. i ILvrdatr Offim Ofat Mintry of St-to tMedsty f NtMinitrie nd Sedcal INnS- Prnetdt Of of ie nedmt Meiatry at Neet cf (S eloes) Fie ln cntr. Chw illw paItiopMtlW6 Plnmn. Meltry of tul are focal Z ia the 1er- Mlenistr of PlnodeW Flaner to aetrl ftmed leetrel halit, focal pois. (30 emloyaee) Fienance. pout.. anofoath Boarlea, emt ofadd-omen erc (6- _e1a)

1ONlNIUr r rd.1 Cd CrdAl c. O rdal OD&, Bodh bot ntly Crcral *d Puelc CO CfirdAL Oa mmcial COee Crcal Code Peeic Code Both lde, loth. IrL SES ads, Oammaral (Yt OR (bde O:UL) tE or KliE N. dmant Om hDldiae NO esner lag ne- Tn wey lr No holeng No hodinQeg Bont horlem-al Neroc tE, Noppholdietd ILng No dominet hollt Om Nnejor hidLeg No skpr tl OiII Mleellompeer. -wc for ed hEdisg av. htld cqpenv. aepa.a-le.lelldgq. holding coepefie.cnie. holdig cq>penefen. cpy med cOW. ctM . cn *. slld fim. cnies. One for Oatr.ne for i1jay. 1r -Orieted holdisg.

[lA TOM e ly Ar c - 1 ely tr, Deaentralled he Daostralid Decentralio.ed Otetrald. t cntwllred. lantraeaz to Owntralzed to Cnmrallre to Deoentrald. LwtcA to rla QF 1I l edeiaie Used dedgnm g political delaim making decion -kieg teldw nOy heldlwsg ecen heoldiop nd (owvemt ed lmlfl ItE wkeeg to the kieeg to th laht otrolltc b bht C,,t,,11 INW Iy Cowernmt. aoi G/rent. Gonerewt. Parli_t. Boad of Dilenoca l Idof Diveeow mgnholdi% aqdteleLi aon I' t. r PdAhet. dr mloldg arid ed st1ug Hime Of Star potSci ltx partcption.

Uo CF 1 Vary h4je. Very he Very dhW. Very he Vey hie. SW. 1ee. 1. 1. 1. MAN Uj4e ID.

M . LUISG NSe. Pn. 1 t. Veriodr dgte hedgailetsand admts nd halet nd aegetad n a nd N.Wta d Nota Wel ad b4tsd maflEm apoes to Ithie Nprta matisly laer Perioldc prt. Mothy ept. ehy port. lthly7 Rports ltxthy 7ot 14 si, Pedoef tap ParYodil en Gte t. to sapr to Nqp- to CG OeC. to loerwent. to GoCneent. to Glnenrmet. to holdli . to Gon_at to 1saty aid to Meatr nd holdbq. holdtlP. (Slnw 1982). GesOut. Gon_om.

113 laIapaet Inependen State haltor. lptnt and St Autditor. Stare editow. Stte Aditor. State Aditor. - o - liepanedent Ieteen stAte od _ T t 1hl- hAditot. Mltors. State hditoe. hedftoea. htdtors. -m hod . two ad olsr lv des MoAxedaoet

CF PSI m2SI~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~b f ngyo (in. 5962)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~pnkp

(*) Los 198 gel dsh a Ia weit _empnile for actm nAi tratpetatre Id potli utilty tyg atArpdai ned o,sriAi - a *_re deeatry.

at 11_mt fimes for 1977. b/ wlewt tf4pr for 197. ct e_lst figf for 191. d/ Tasi P melopUt ate*dalO Is.5)twfporatdn tad t_t Is em _sporared

sunw O ntat' Ste*. Otfe Stsoly tf PlEa SPtaratical h NondSe oa L. m Idnstrlal Statlatic. - 35 - sector: they hire and fire top management,set salaries,set prices, selectsuppliers, negotiate loans, etc., and are also responsiblefor the financialperformance of their firms. They are exposedto the same market forcesthat confrontprivate corporations. Where a firm also has private shareholders,the board reflectsit. 15/

5.10 While duringthe last 15 years Swedenhad most of her industrial enterprisesorganized under one superholdingcompany, recently this company was brokendown into one holdingcompany for medium-sizedenterprises and a numberof branchoriented corporations, each owned directlyby the Ministry of Industry. This decisionwas in line with the government'sbelief in decentralizeddecision-making.

5.11 By contrast,public enterprise managers in Ghana have little autonomyand are subjectto excessivemultiple controls. Ghana'sexpe- riencewith publicindustrial enterprises started in the 1960sas a part of the programof the first republic(1957-66). During the period1975 to 1981 publicindustrial enterprises contributed about 60 percentof total industrialproduction in the country. Concernwith extremeministerial interferencein publicenterprise management and poor performancehave led to a numberof organizationalchanges during the 25-yearlifespan of the enterprises. 5.12 The Ghana IndustrialRolding Company (GIHOC) was establishedin 1967. A centralgovernment body named the StateEnterprise Commission (SEC),with varyingresponsibilities for developmentand controlof public enterprises,hasbeen establishedand reestablishedthree times: 1965,1976 and in 1981. In its currentincarnation, its role is to advisethe government on issuesrelated to performance,strategy, finance and personneland to monitorand evaluateperformance. The formalstructure of control places the ownershiprole with the sectoralminister. Lines of responsi- bilitythen flow throughthe holdingcompany's board and managementdown to the board and managementof the operatingcompanies. In practice,excessive powersand controlremain firmly placed in the ministriesbased on a systemof multiplecontrols, with the sectoralministers having the most importantpositions. For example,prices are set by the Pricesand Incomes Board,a unit under the Ministryof Financeand EconomicPlanning. The dismissalof five or more workersmust be authorizedby the Ministryof Labor (exceptin cases of provencriminal offenses). Importlicenses are allocatedby the Ministriesof Trade and Finance,while lettersof credit for approvedimport licenses are the responsibilityof the Bank of Ghana and the commercial banks. Approvalof immigrantquotas for non-Ghanaian personnel is handledby the Ghana InvestmentCenter and the Ministryof Interior. Wage contractsnegotiated between workers and managementneed the recommendationof the Pricesand IncomesBoard and the approvalof the Ministryof Finance. These examplesillustrate the web of bureaucratic controls,all of which are centralizedin the government.

15/ While it is difficultto observea causalrelationship between private participationin publiccompanies and their performance,many obser- vers believethat the professionalcompetence of the experienced privatesector individuals on the board of the Swedishsteel holding company,SSAB, has contributedsignificantly to its successful restructuringto date. - 36 -

5.13 Between these two extremes lie important examples of varying degrees of decentralization. For example, the French model until 1982 was one of double tutelage, with sectoral ministers in charge of approving operating methods and the Minister of Finance responsible for financial issues. After the nationalizationof the larger industrial groups, the principle of unity of tutelage was introduced to classify and rationalize the relationshipbetween the state and the firms. Today the French organizational model is essentiallysimilar to the Swedish and Norwegian ones in that no holding companies dominate the industrial sector, and direct lines of responsibilityflowing from the sectoral ministries to the board of directors and to enterprise management.

5.14 There are, however, two important differences exist between the French and Scandinavianmodels. First, the French system has a large number of control mechanisms centralized in the state bureaucracy. These include: financial controllersworking in the public enterprises and reporting to the Ministry of Finance; parliamentarycontrol of public enterprise activities;a Court of Accounts that keeps a check on annual statements and the use of state funds; and an Inspectorateof Finances, within the Ministry of Finance, that intervenes in specific financial issues. In Sweden and Norway, no such institutionalizedcontrols exist, reliance being placed instead on independent auditors who analyze the annual accounts of the public enterprises in the same way as those of private firms. Second, the role of the board of directors is less powerful and substantialin France than in the Scandinaviancountries. The French board makes some formal decisions, such as delegation of signature and approval of documents such as budgets and important loan applications,but without being able to modify them. Important management decisions relating to investment and financing are to a large extent made by the Chief Executive Officer without prior advice of the board.

5.15 The Italian model for managing public enterprises includes a superholding company, Instituto per la RicostruzioneIndustriale (IRI) to manage the state-ownedindustrial enterprises. This superholdingcompany embodies some 600,000 employees and hundreds of enterprises from telecommunications, airlines, TV and radio to steel, shipbuildingand automobiles, all organized under several subholding companies. The most obvious difference, compared to the Scandinaviancountries, is that the unit within the government responsible for state enterprises in Italy is organized as a separate ministry (Ministry of State Participations),is much larger (with some 300 employees) and has more bureaucraticand political responsibilities. Higher level public enterprise officials interviewedfelt that one of the major weaknesses of the Italian organizationalmodel was the far- reaching, strongly political, influence of the Ministry of State Participa- tions.

5.16 When governments take over the strategic role traditionallyplay- ed by the board of directors and get entangled in operating decisions, problems surface quickly. The board's principal functions of providing strategic guidance and effective control are generally replaced by poor or no guidance, weak or no controls, and slow and politicizeddecision-making - 37 -

Political relations become more important than productivityand financial performance. Enterprise morale declines, and recruitmentand motivation of qualified managers become difficult at best.

5.17 Many government officials involved with public enterprisemanage- ment totally underestimate the complexitiesinvolved in managing industrial firms and the importance of management. An enterprise is not only input, production functions,systems and output; foremost it is people and a business. The success of an enterprisedepends on creativity,flexibility, dedication, perseveranceand hard work, particularlyof top management, who, by their example, set the culture, direction and spirit for the rest of the employees. Government planning officials and civil servants appointed as directors and CEOs of public enterprises(in many cases) lack both the relevant knowledge, experienceand motivation to orchestratethe human and financial resources and myriad decisions to be made in creating and maintaining effective and viable enterprises.

5.18 It has become increasinglyevident that an important factor in the differences in productivityamong companies and among countries is the management and its effect on the motivation, culture and spirit of the enterprise. It follows, that to stimulate successfulpublic enterprise operations, governmentsmust hire top quality professionalentrepreneurs/ managers and by organizationaldesign shield these people from undue political and bureaucraticinterference.

5.19 Recognition of the inefficiencyof large centrally planned bureaucraciesand increasingunderstanding of the importanceof the individual manager are two complementaryforces currentlymoving many governments to take steps toward more decentralization,and in many cases to the most decentralizedorganizational mode for managing public enterprises,the transfer of ownership to the private sector.

5.20 As countriesmove from industrial infancy through growth and maturity into the post-industrialera, there is a pattern as to which organizationalstructures dominate in different periods. At the risk of oversimplification,Figure 3 illustrateshow public enterprise sector organizationshave tended to move through five increasinglydecentralized stages as the economy grows and matures. This evolution toward increasingly decentralizedmanagement models can be explained by some of the characteristicsof the process of industrialization. As industrialization increases, the complexitiesand competitivenessof the market place as well as the speed of change all accelerate. Rigid and bureaucraticmanagement models that are appropriateat earlier stages become obsolete, and more decentralizedmanagement structures become a necessity for survival.

5.21 Naturally,any one country does not necessarily implementall five types of organizations,instead most countries tend to move from highly centralized to more decentralizedorganizations. For example, India never used holding companies as a dominant organizationalsolution, but at times its ministries functioned like holding companies. Further, some countries may temporarilymove parts or all of their public enterprise - 38 -

Figure 3: LIFE CYCLEOF PE SECTORORGANIZATIONS

LEML (W OUSTRIALiVxmPimi

IIWAIIC! GRO!H MAlUITII RENEWALI RENEWAL171

PE SCTOR IN Holdin \ RKLATION- and Super- Directly SHIP TO GW PIEs, holding Owned by Joint Regulated by Companies Govt., Ventures/ Government Regulated Introduction PEs as Civil by on Stock Dept. of Service Commercial Exchanges Ministry Codes Code

STAGE1 STAGE2 STAGE3 ! STAGE4 STAGE5

1W.ANIZATIOtiALMOM. DIVISION(N FOIERS

OWNERSHIPROLE Govt. Govt. Govt. Govt. Govt./Priv. STRATHGICROLE Govt. Govt./Mgt. Holding Co. Board Board OPERATII ROLE Govt. Mgt./Govt. Mgt. Mgt. Mgt.

sector in the opposite direction, e.g., toward more centralizedorganiza- tional structures. Public enterprisesin Brazil during the sixties and much of the seventies, for example, had a level of autonomy envied by counterpart managers in most developingcountries. They were already using the Stage 5 model. Many public enterpriseswere minority-ownedby private partners and their shares were traded on Brazilian stock exchanges. However, because of overexpansionduring the seventies,government control mechanisms were significantlystrengthened after 1980. Similarly,public enterprises in Italy experienceda much higher degree of autonomy in the fifties and the early sixties than later. This shift to increasingcontrol was caused by an ideologicalshift toward a more socially-orientedeconomic policy, with increasing emphasis on centralizedplanning. As a result, most public enterprises became organized under a huge superholdingcompany, and a separate ministry for public enterpriseswas established. These periods of centralizationare relativelyfew and do not represent a - 39 - counter-argumentto the long-term trend of increasingdecentralization. Recently, the Italian superholdingcompany IRI has adopted a divestiture strategy of selling part of the equity capital of major subsidiariesto the private sector.

5.22 It may be speculated that the importanceof the public industrial enterprise sector in the overall economy will diminish in the post- industrial era. Most of the circumstancesthat motivated the creation of public enterprises in the first place, such as the lack of industrial infrastructure,nonexistence of equity markets and a limited private sector, have diminished over time. The new emerging high technologyand service-relatedindustries are typically not capital-intensive,but rather are entrepreneurialin character and ill-suited for exploitationby the bureaucratic organizationalcultures frequently found in public enterprises. The public enterprise sector is likely to decrease in relationship to GNP, and in some countries even in actual terms. A significant number of enterprises are likely to leave the public sector in the future. Privatization efforts, including joint ventures, are currently underway in a large number of countries, both developed and developing,such as the United Kingdom, Norway, Sweden, Italy, Spain, the Federal Republic of Germany, Japan, Turkey, Uganda, Togo, Cameroon, Brazil and Argentina, to mention a few.

5.23 The attraction and success of large superholdingcompanies are also passe. Increasinglythey are likely to be broken up into more decentralizedand flexible bodies better able to adapt to market changes. The Swedish experience is of particular interest. Statsforetag,the Swedish industrial holding company, after having been cut down in size during 1983/84, now seems to be moving toward an investmentcompany mode of operations, that is, it is introducing its subsidiarieson the stock exchanges and buying and selling companies to enhance the value and viability of its portfolio.

Factors Affecting OrganizationalDesigns

5.24 The above examples illustrate the substantialdifferences in the centralizationand decentralizationof powers among governments and public enterprises. What accounts for these differences in organizational designs?

5.25 Evidence from the country case studies suggests that, in addition to the perceived role and objectives of public enterprises,there are two main factors important to the degree of decentralization. They are: (i) the level of economic development in terms of the availabilityof industrial infrastructure,institutions, human skills, private sector resources,etc; and (ii) the degree of the market orientationof the product and factor (labor and capital) markets.

5.26 Level of Economic Development. Countries with low levels of industrialdevelopment and a limited private sector have opted to establish public enterprises in industries such as steel, cement, petrochemicals, etc. (For details, see Annex I.) In view of the large amounts of capital - 40 - required, complexity of the technology to be used and scarcity of managerial resources, governments have justified the centralizationof a number of functions for planning, implementingand controlling the overall development process. Countries such as Ghana, India, Pakistan and Zambia all experienced this pattern in the 1950s and 1960s. Under these circumstances, an appropriaterole for governmentswould be to define the major investment priorities in terms of industrial subsectors;to resolve the Issues related to appropriate technology; to set targets for public enterprises and monitor performance against these targets; and to promote, advise on and monitor the implementationof major training programs in management, production and new technology.

5.27 A word of caution is necessary, however. A centralized approach can compensate to only a limited extent for a lack of managerial skills and technologicalknow-how at the enterprise level. While a central ministry may be able to choose the appropriate type of engine to be produced, it cannot, by "long distance phone calls," advise plant managers and workers on how to solve a production problem. When the technology and processes are manageable, it may be possible to leverage the competence of a central management group and compensate partially for the limited skills of plant managers in order to develop and, for some time, manage a significant number of plants. However, as these plants mature and move into more complex technologiesand businesses, and as new products and markets are developed, new managerial resources and technologicalskills have to be developed at the enterprise level. At such a point, decisions ought to be gradually decentralized to the public enterprises.

5.28 Degree of Market Orientation. Where relatively robust product and factor markets exist, as in the European countries in the sample, the need for a centralized approach to enterprise management is far less. In such situations, financial performance,while not a perfect indicator of enterprise performance,is highly relevant. Public managers will be 'controlled"by employees, peers, the media, the financial sector, the commercial code, generally accepted accounting principles and the owners' long-term objectives.

5.29 Where markets are underdeveloped,it may be necessary to insti- tute, for a limited period, some organizationaland systems controls in the areas of financing, labor policy, pricing, production quantities,operating costs, marketing and performancemeasurement. Since there are risks and inherent costs involved in such controls,governments should abolish them as market controls become more effective.

5.30 It is interesting to note that, of all the countries in our sample, Austria, Sweden and Norway have had the most success during the last four decades in fulfilling ambitious social objectives such as free schools and universities,free child care and health programs, comprehen- sive retirement programs, equalizationof pay structures and redistribution of wealth. Yet these countries did not choose an ideologicalstrategy but rather a more pragmatic one of active industrial and/or private sector support strategy. For example, the Swedish Social Democrats, in power uninterruptedlybetween the early 1930s and 1976, emphasized the dynamism - 41 - and productivityof the private sector, whereas the government'srole was seen as setting a supportive incentivesenvironment. By building a highly competitive and strong private sector, the country was able to develop the necessary tax base for implementingsocial reforms.

Conclusion

5.31 How can governments avoid, or at least ameliorate, problems arising from excessive centralization? To some extent, as indicated, the increasing complexity of industrial activities is itself ensuring that public enterprises are delegated powers hitherto withheld from them. A good example is India, where the government has increasinglycome to appreciate that the excessively centralized structure of public enterprise control may have been appropriate in the 1950s and 1960s, but is hopelessly anachronisticat a time when many Indian enterprises have to compete in fiercely competitive foreign markets.

5.32 While the public enterprise sector may give governments fewer headaches in the post-industrialera, this prognosis is no comfort for most countries today. The important point of this discussion is that strategies for public enterprise organizationaldesign should take into consideration long-term trends and needs. Governments should try to avoid large, rigid, special-purposesolutions with high political exit barriers. Instead, the emphasis should be on creating flexible and market oriented mechanisms, partly in the interest of efficiency and partly to make it easier for the government to adapt the sector to changing circumstances.

5.33 There is no reason for developing countries to wait until the post-industrialarea to move into some of the more decentralizedforms of organization. Naturally, the types of powers granted and services provided by central units, and the mix between market and government controls,will vary significantlybetween countries like Sweden on the one hand and Zambia on the other. However, these varying levels of powers, services and controls could basically be exercised within the same type of decentralized organizationalstructure.

5.34 The experience of public industrial enterprises,as well as of large private enterprises,indicates that the worst repercussionsof a poor organizationalstructure can be avoided through scrupulous adherence to the following principles: (i) clear demarcation of the major roles and responsibilities among the three parties indicated above, namely, the government, the board of directors and the management of public enterprises;(ii) full delegation of authority in line with delegated responsibility;(iii) the appointmentof professionaldirectors and managing directors with experience directly relevant to their tasks; and (iv) introducinga management philosophy that is less control-orientedand less procedure-boundand more concernedwith judging managers on the basis of enterprise viability and a limited set of indicators of performance. Adequate decentralization is not, by itself, sufficient for satisfactoryperformance. However, its absence can confound the most meticulouslyworked out organizational structure. - 42 -

CHAPTER 6

MANAGERIAL SKILLS AND MORALE

6.01 In the previous chapter, the degree of managerial autonomy and accountabilitywas identified as an important determinant of public enterprise performance. The main factors influencingthe degree of such delegation of autonomy were examined. As indicated, there is a host of such factors that together determine how much decision-makingpower is actually decentralizedby governments to enterprisemanagers. This chapter highlights two such factors of particular relevance to public enterprises. They are the quality of managerial skills in public enterprises,and the ability of public enterprise managers to maintain morale. Together, these two factors facilitate government efforts to decentralizeand to enable public enterprise managers to create an environment conducive to greater productivityand innovation.

Managerial Skills

6.02 The quality of managerial staff of public enterprises in developing countries varies considerablyacross countries and even across enterprises in the same country. At one extreme are Brazil and Israel, where public enterprise managers are generally well-educatedin their fields, are disdainful of government bureaucracy and are knowledgeableof private enterprise methods. At the other extreme are Ghana, India, Pakistan and Zambia, where many public managers are closer in their attitudes to bureau- crats than to private entrepreneurs.

6.03 A number of factors explain the variations in the quantity and quality of managerial skills in developing countries. First, with few exceptions (such as Brazil until the late 1970s) public enterprise managers are paid substantiallyless than their counterpartsin the private sector. For example, in Zambia, public enterprise salary ceilings are still set with reference to civil service levels. In mid-1984, a managing director of a modern private sector Zambian firm earned on average Kwacha (K) 35,000-40,000per year. The comparable salary at an enterprise under the publicly owned Indeco was only K 15,000 (about US$10,000). The difference was sufficientlyvast that even the impressive perks given in the parastatal sector (automobile,housing allowance, and free electricity, water and telephone) did not come close to bridging the gap. In some countries, it is not only difficult to maintain managers within the public sector, but there may also be an exodus abroad. India and Pakistan in particular have suffered a substantial brain-drain to the Middle East over the last decade. One of the most severely affected has been National Refinery Limited of Pakistan, the demand for whose technical and managerial staff in the Middle East has been considerable.

6.04 Second, public enterprises in developing countries typically experience high rates of turnover of managerial staff either because of political changes or because of the rotation involved in the civil service. In India and Pakistan, for example, the chairman and managing directors are appointed for five-year terms but rarely stay more than - 43 -

three. HMT of India, earlier identified as one of the most successful Indian public industrial enterprises,was also one of the few corporations in India to have the same chief executive for almost 15 years. Equally damaging to public enterprise performance is procrastinationin filling top management vacancies. The brief period in the late 1970s during which BHPV of India was without a chief executives proved very costly to the enterprise.

6.05 Third, in a number of developing countries, public enterprises are considered dumping grounds for politiciansand retired generals. Their lack of technical and managerial know-how becomes a formidable handicap and a source of poor performance.

6.06 Fourth, there have been severe shortages of managerial skills in some sample countries, particularlyZambia and Ghana. This problem has not only created bottlenecks but has also caused excessive dependence on the chief executive officers and a few managers. To some extent, the problem has become even more acute, at least in the short run, in such countries as Zambia that encourage through regulation the increasing participationof nationals in the place of expatriates.

6.07 How can governments reduce this constraint? One of the most effective ways of nurturing a motivated and more dynamic public enterprise management is to depoliticize,to the extent possible, its hiring and firing. Sufficient managerial autonomy and accountabilitywould ensure a less bureaucraticresponse. Another interesting solution is illustratedby ICL of Israel. Senior management staff are employed under special contracts, have no tenure and can be dismissed at any time, but are also paid higher salaries than in other public enterprises.

6.08 Given the tight budgetary situations in many developing countries, raising the salaries of public managers closer to their counterparts in the private sector may not feasible. There are even those who believe that raising public managers' salaries may make the positions even more attractive for political appointments. However, if the government envisions a more dynamic and decentralizedrole for public enterprises,it would need to attract highly qualified and experiencedmanagers. To do this, it would need to make the salaries of public managers much more competitive with those of their counterpartsin the private sector.

6.09 While there are numerous training programs for technical skills in the public enterprises examined, there are far fewer for managerial ones. To gain from economies of scales in training, some countries have created training centers for groups of enterprises. For example, the Portuguese holding company for manufacturingenterprises, IPE, provides training to managers of its subsidiaries. Similarly, the Austrian holding company, OIAG, sponsors in management training for its subsidiaries. Both holding companies arrange for trainees to participate in programs abroad.

The Issue of Morale

6.10 High morale is generally the result, not the cause, of good performance in public enterprises. However, these entities would have a - 44 -

far better chance of success within a given environment if they were not saddled with an understandablywidespread (but not always fair) opinion that they are a locus of corruption,special privilege and irremediable performance. Projectinga more positive image of public enterprises could influence performance at the margin. In private U.S. companies,"the volume of contrived opportunitiesfor showering pins, buttons, badges and medals on people is staggeringat McDonald's,Tupperware, IBM and many others. They actively seek out and pursue endless excuses to give out rewards."16/

6.11 A demoralized organizationneeds success to keep up its momentum. Without it, disillusionmentcan set in rapidly. No spectacular incidents of success are needed; sometimes even minor gains rebuild confidence and restore morale. The chief executive officer needs to identify areas of activity where proper planning and determinedeffort can result in quick results. A number of examples from India's public engineeringfirms in the late 1970s illustrate this point.17/ In the case of BHPV, the company won a large morale-boostingorder for paper-convertingmachinery in the face of stiff internationalcompetition, and its top management lost no opportunity in publicizing this achievementwithin the enterprise and outside. In the case of Richardson and Cruddas, reducing the share of low-value structurals,increasing scrap sales and raising sales of points and crossings to railways were quick-returntasks whose intrinsic returns were more than just the value of the sales. The secret of success in the early stages of turnaroundsmay well lie more in such seeminglyminor but well-publicizedincidents of success that shore up confidencethan in grand, long-term strategies.

6.12 The successfulpeformance of some Israeli companies after their privatization (Tadiran,Elbit and Haifa Chemicals)was at least to some extent the result of an improved image and the increase in staff morale. Obviously, there were other factors, such as new technologyand marketing. By contrast, the sale of the government'sminority shares in ZIM (Israel's national shipping company) to Israel Corporationhas not been successful because of personal rivalries and financial irregularitiesthat have severely affected company morale.

6.13 Finally, governments,as owners of public enterprises,have every right to investigatecharges of irregularitiesand to ensure honest management. However, this action should be done in as non-politicalaway as possible. A major source of concern among Zambian public enterprise managers in 1984 was the formation of an Anti-CorruptionCommission (ACC) empowered to investigatecharges of corruption. It appears to have been to demoralized public managers, especiallylower level ones, and to have made them even more reluctant to make decisions individually. Such legal investigationsneed to be specific and to be completed rapidly to end uncertaintiesabout the future of enterprise leadership.

16/ Peters and Waterman, In Search of Excellence.

17/ See Khandwalla, Performance Determinants. - 45 -

CHAPTER 7

COORDINATIONWITHIN THE GOVERNMENT

7.01 The greater the centralizationof powers relating to the management of public enterprises,the greater is the need for attention to the organizationof decision-makingwithin the government. Lack of coordinationamong the various government entities responsible for public enterprises frequently contributes to dysfunctionalpolicy dialogue between the government and the enterprises,and this weakness in turn culminates in excessive interference. Such a lack of coordinationassumes various forms. There may be inadequate coordinationbetween investment decisions (ministriesof industry and planning) and financing decisions (ministryof finance). There may be inconsistenciesbetween pricing decisions (ministry of industry and/or commerce) and subsidy decisions (ministryof finance). In addition, conflicts may arise (and frequentlydo arise) between the financial objectives as stated by the ministry of finance and the social objectives enunciated by the sectoral ministries. This is but a small sample of the innumerable problems that arise in practice. Superimposedon these problems is, first, the propensity of public enterprises to become lightning rods for excessive political interferenceand for discontinuity in decision-makingwith every change in government,and, second, the lack of technical, financial and managerial skills required to run increasingly complex industrial enterprises.

7.02 For these reasons--namely,the need for coordination,for reduced political interference,and for greater industrial experience governments have attempted to devise organizationalarrangements that place government at arm's length from public enterprises. One solution has been to decentralize the ownership role of public enterprises to a holding company. Another has been to establish public enterprise oversight agencies and specializedgovernment bureaus, broadly referred to here as focal points.

Holding Companies

7.03 Of the thirteen countries covered in this study, six rely, or have relied on large multi-sectoralholding companies (sometimesreferred to as superholdingcompanies) to manage a significantpart of the public enterprise sector. Italy established IRI in 1933 to manage the industrial portfolio of several large banks that were rescued by the government during the Depression. In the other five countries, the use of holding companies is a more recent phenomenon. In the case of Pakistan, the experiment was short-lived,between 1972 and 1978. Table 7 summarizes this information.

7.04 The rationale for setting up a holding or multi-sectorholding company is that it can: (i) act as an effective buffer against excessive political interference;(ii) provide effective coordinationof decision- making, clear strategic guidance and improved financial discipline; - 46 -

Table 7: ESTABLISHMENTOF MULTISECTORALHOLDING COMPANIES

Year No. of Created Country Holding Company Employees

1933 Italy Istituto per la RicostuzioneIndustriale (IRI), Superholding 600,000

1965 Zambia Zambia Industrial and Mining Corp. Ltd. (ZIMCO), Superholding Industrial Development Corporation (INDECO), Subholding 20,000

1967 Ghana Ghana IndustrialHolding Corp. (GIHOC) 15,000

1970 Sweden Statsforetag 48,000 a/

1970 Austria OsterreichischeIndustrieverwaltungs Aktiengesellscheft(OIAG) 150,000

1972-78 Pakistan Board of IndustrialManagement 80,000

a/ Prior to the break-up of Statsforetag.

(iii) assist in pooling scarce managerial talents, thereby infusing industrial and enterprise-levelexperience into the enterprises;and (iv) lead to benefits of synergy, such as economies of scale in large procurement contracts as well as in establishingforeign marketing outlets, lower cost financing,etc. 18/

7.05 These arguments are all very valid, and some state-ownedholding companies have indeed performed adequately. However, the overall experience with large state-ownedmulti-sectoral holding companies has been negative. The drawbacks (and strengths) of these entities are examined in detail in Annex III. To summarizehere, instead of being buffers against excessive political interference,they can become powerful tools

18/ From the point of view of the holding company, there is also a tax advantage, in that losses in one public enterprise can be used to reduce income tax payments in profitable enterprises in the same subsector and, depending on local tax laws, even in public enterprises in other subsectors. - 47 - for political domination, interferenceand control. In some cases, the sheer size of the multi-sectoralcompanies creates unnecessary layers of management and results in slow and bureaucraticdecision-making. The layers of enterprises and governmentbodies governing public enterprises in Italy is a case in point (Figure 4).

Figure 4: PUBLIC ENTERPRISES IN ITALY

PARLIAMENT (Approval of long-term plans and use of endowment funds)

Interministerial Committee for Public Enterprises

I Ministry of State Participations

.I IRI Superholding

I 8 Large Sectoral Holding Companies

Subsidiaries/Sometimes Holding Company's

I Operating Companies

| CONSUMERS l

7.06 Holding companies can sometimes also lead to lower motivation and quality of public enterprisemanagers when important operationaldecisions become centralized at the level of the holding company. For instance, a number of frustrated senior public enterprisemanagers in Pakistan complained that the government'sefforts to decentralizedecision-making to their enterprises ended up "getting stuck" at the holding company level. Even more serious is the excessive financial power that becomes vested in - 48 - the holding companies. Their ability to transfer funds between subsidiaries can result in inefficientresource allocation,the starving of viable firms with strong long-term growth potential and the costly sustenanceof non-viable enterprises.

7.07 Sweden's experiencewith a superholdingcompany is instructive. The state holding company, StatsforetagAB, was established in 1970. Given its rather disappointingperformance, two government inquiries were initiated in 1977 and 1982. They concluded that the Statsforetaghad simply become too large and too complex to be managed effectivelyas a single entity. It was felt that the major subsidiariesof the holding company, such as LKAB (mining), SSAB (steel) and Svenska Petroleum,were so large that their financial needs were jeopardizing the existence not only of the other relatively smaller companies but also of the entire holding company. Moreover, excessive involvement of top managementwith the larger companies left insufficienttime for the smaller subsidiaries. At the same time, the sheer size of the financial transactionsand the social implica- tions of plant closures were, in many cases, such that all negotiations took place directly between the subsidiariesand the Swedish government. The management of the holding company was left with limited, if any, effective role in decision-making. As a result, Sweden has moved from the concept of a single superholdingcompany to several sectoral ones in shipbuilding,steel, textiles, apparel and the like.

7.08 Unfortunately,the experience of developing countries in the sample of countries suggests that superholdingcompanies invariably lead to greater, rather than less, politicization,to excessive centralizationof decision-makingand, therefore, to poor overall performance. In Pakistan, after the nationalizationof industries in 1972, a Board of Industrial Management (BIM) was set up as a superholdingcompany along the lines of IRI of Italy. In view of the poor record of the Board, and upon the recommendations of two high-poweredcommissions (Uquaili and Uttra), it was dissolved in 1978. Separate sectoral holding companies were set up in its place in the chemicals, fertilizer, automobile,cement, engineering, petroleum refining, steel and textile subsectors.

7.09 State-ownedholding companies are likely to be most effective when they are used to manage small- and medium-sizedenterprises and when they are not allowed to grow beyond a certain size. Further, they can also be effective when they are establishedalong sectoral lines to function as a mother company for a group of operating companies. The Israeli chemical group, ICL, is a good example of a relatively successful sectoral holding company. (Further details on the advantages and drawbacks of holding companies, and when and how they are most effective, are provided in Annex III.)

Focal Points 7.10 The use of holding companies is one way to coordinate decision- making 8nd to cope with excessive political interferenceand lack of industrial experience. Another form, prevalent in some of the sample countries, consists of public enterprise oversight agencies and specialized - 49 - government bureaus. Broadly speaking, these entities, also called focal points, are of two types: those that perform the full ownership role and those that have a more limited ownership role in the sense of being supervisoryor advisory. The functions of these units and examples are provided in Table 8.

7.11 Among the focal points playing an ownership role are those in Norway, Sweden and Austria, all of which are very small, each employing less than 10 people. They concentrate,in accordance with the decentralized management philosophy of their governments,on their limited but

Table 8: FOCAL POINT ORGANIZATIONS

A. Focal Points with an Ownership Role

Responsibilities

o To establish and amend the charter of the enterprise o To hire and fire the board of directors o To approve annual accounts and dividend payments o To hire auditors o To monitor performance

Examples

1. Norway: Public Enterprise Unit in Ministry of Industry 2. Sweden: Unit Five in Ministry of Industry 3. Austria: Ministry of Public Enterprises 4. Italy: Ministry of State Participations 5. Canada: Overseeing Agency for Crown Corporations

B. Focal Points with a Supervisory/AdvisoryRole

Responsibilities

o To define financial and operationalobjectives o To review multi-annualplans and budgets o To monitor performance on a monthly or quarterlybasis o To provide technical assistance and training

Examples

1. Brazil : Special Secretariat for the Control of State Enterprises (SEST) within Ministry of Planning 2. India : Bureau of Public Enterprises (BPE) 3. Pakistan : Experts Advisory Cell (EAC) within the Ministry of Production 4. Israel : The Government Company Authority (GCA) within the Ministry of Finance 5. Ghana : The State Enterprise Commission (SEC) - 50 - important ownership role. Italy's Ministry of State Participations,on the other hand, with some 300 employees, has been criticized for being excessively large, bureaucraticand political.

7.12 Focal points with supervisoryfunctions can play a valuable role in obtaining timely and relevant informationon public industrialenterprises, a prerequisitefor effectivelyassessing performance.19/ A relatively successful focal point with supervisoryresponsibility has been SEST in Brazil. Its experience has shown that clarity in the role and objectives of such entities can make them useful agents of change in the direction of greater efficiency for public enterprises. In its six years of existence, SEST has been able to bring greater accountabilityto public enterpriseswithout getting involved in day-to-day operations,even though some of its operationalfunctions (such as the recommendingof ceilings and targets for certain categories of expenditures)have the potential for leading to excessive interferencein the operational functions of public enterprises. Through its continuous monitoring of the key economic variables, and with the full political support of the government, it has been able to bring the priorities of even powerful enterprises such as PETROBRAS and CVRD more in line with the exigencies of the overall economy. And yet these results were achieved by focusing mainly on broad aggregates and ensuring compliance. SEST's small but professionallyvery competent staff of less than 50 professionals,deliberately kept to that level by the government, inhibited it from excessive interferencein the day-to-day affairs of the enterprises.

7.13 Strong political support and real operationalpowers are the cornerstonesof a successful supervisory focal point. Some of the key operational powers of SEST are:

(i) To recommend ceilings and targets for all major categories of revenues and expenditures, foreign debt, imports and fuel consumption of public enterprises;

(ii) To recommend distributionof public enterprise profits and dividends;

(iii) To recommend new equity capital increases for public enterprises; and

(iv) To recommend creation and acquisitionof new entities as well as the liquidationand divestiture of old ones.

Last, but not least, its recommendationsare made directly to the President of the Republic for final approval.

7.14 The contrast of SEST with SEC of Ghana is instructive. The precise role of SEC has never been very clear, and its political support

19/ For example, prior to the establishmentof SEST in Brazil in 1979, even information on the number of public enterprises in existence was not readily available. - 51 - has wavered. The agency's work is frustrated by delayed reporting by the public enterprises; sometimes the reports are so late as to be virtually useless. On the other hand, public enterprise managers interviewedconsider the agency an unnecessarybottleneck in their operations.

7.15 In terms of effectiveness,Israel's Government Company Authority lies between SEST of Brazil and SEC of Ghana. The Authority, a unit within the Ministry of Finance, does not have any real powers beyond reviewing budgets and reports. The influenceof the Authority has, therefore, depended mainly on the extent to which the Minister of Finance has wished to use it as a professionalunit, and on the caliber and qualificationsof the director of the unit. The Authority was fairly effective in the 1960s and early 1970s when it was headed by an energetic director. In later years, the Authority lost some of its influence because of less support from the Ministers of Finance, who appointed either weaker directors or politically motivated ones. The Barak Commission,appointed by the Ministers of Finance and Justice in 1968 to prepare a draft law for public enterprises, made a number of recommendationsto strengthen the powers of the Authority. These recommendations,which were not implemented,were: vi) to attach the Authority to the chairman of the Committee of Economic Ministers (rather than making it a unit within the Ministry of Finance); and (ii) to give the Authority the power to nominate one director on the board of each public enterprise. In retrospect,these were reasonable proposals in the right direction. However, they may not have been enough to elevate the Authority's effectivenessand importance to the level of SEST. To achieve that position, some of the financial powers normally belonging to the Ministry of Finance would have to have been transferredto the Authority.

Lessons from Experience

7.16 The experience of both developing and developed countries suggests that superholdingcompanies are generally unwieldy,unresponsive to the rapid changes that typify industrial sector activity,and excessively prone to political interference. A somewhat more attractive organizational arrangement,though still a second-bestsolution, is a well-focused sectoral holding company. Such a sectoral entity can have several advantages in addition to acting as a buffer against excessivepolitical interference. Research and developmentexpenditures too large for individual enterprises can be shared among several of them. Economies of scale can arise in negotiating large-scaleprocurement contracts,in establishing foreign market outlets and in pooling scarce managerial talents. They have also been useful in arranging domestic and external financingthat individual enterprises could not have obtained on their own. Restructuringof an industrial subsector is also easier in some cases if the ownership of participating firms is under the same umbrella. The establishmentof the steel holding company, SSAB, in Sweden in 1977 has been a key to the successful restructuringthere. However, it needs to be emphasized that holding companies should be sectoral to achieve the synergiesenumerated. The performance of holding companies that cut across sectors (such as the PIDC in Pakistan) is generally poor. It also needs to be stressed that the performance of even sectoral holding companies depends to a considerable extent on the existence of active domestic and foreign competition, well-developed financial markets and freedom of entry for new firms. - 52 -

7.17 The main objectives of focal points would be to: (i) lead the public enterprise sector reform process; (ii) develop and maintain high quality board of directors for public enterprises;(iii) coordinate government decision-makingand protect public enterprise boards from excessive government interference;and (iv) monitor public enterprise performance. The establishmentof an ill-conceivedand poorly constituted focal point can become an additional layer of bureaucracythat works against a more decentralizedmanagement structure. However, it can be an effective catalytic agent for change and a promoter of more effective operations if it adheres to the above mentioned objectives.

7.18 Evidence from the sample countries also indicates that it is desirable to strengthen oversight agencies with some elements of the ownership role in addition to their supervisory functions. This is done in Norway, Sweden, Austria and Italy. Also, the success of SEST depends, to some extent, on the fact that it has been granted several powers that normally belong to the ownership role. Second, particularlyin the case of developing countries, there are strong reasons for elevating the ownership role of oversight agencies beyond the realm of one individual ministry, 'thatis, to have it report to an intersectoralcommittee, to organize it as a quasi-governmentalunit, and/or to attach it to the Prime Minister's/ President'sOffice. This type of solution has several advantages:

(i) it increases the status and importance of the unit to levels commensuratewith the size, importance and complexitiesattached to the public enterprise sector;

(ii) it creates a unit with a direct link to the country's political leadership and which can concentrateall its attention on public enterprise issues without being distracted by other ministerial tasks;

(iii) it will be easier to create a less political and more enterprise/industry-orientedculture in a unit operating "half outside" the normal ministries.

(iv) it may, in certain situations,be a more politicallyacceptable solution, since no existing functional ministry will gain power over the others.

(v) it avoids undue concentrationof power in one functional ministry.

7.19 Naturally, what is a good solution for Brazil may not be a good solution for Ghana, India or Sweden and above conclusionswill not apply to all situations. For example, in the case of Sweden and Norway, the ownership and supervisory roles are merged into one unit. However, owing to the very decentralizedmanagement structure used in these countries, most supervisory functions are performed by the market place and by boards of directors. This reduces the importance of the issue of where to organize the focal point. Further, some countries may choose to establish two or more focal points for different types of public enterprises, for example, one for manufacturing enterprises and one for public utilities. - 53 -

Options for OrganizationalDesign

7.20 Establishmentof an institutionaland policy environmentcondu- cive to good public enterprise performanceis a challengingtask. Under- standably, solutions will vary among countries depending on factors such as the level of economic development,availability of free markets, political orientation and ideologicalcurrents. However, there is a set of options and a basic organizationalstructure, the general thrust and direction of which applies to most countries. Basically what is proposed is an institutional and policy framework which stresses the decentralizationof decision-makingpowers and responsibilityto the board of directors and management, and the strengtheningof the capabilitiesof these two bodies. Naturally, the types of powers granted and services provided by central government units, the mix between market and government controls, and the role, compositionand location of focal points will vary significantly across countries. However, these variations could be exercised within the same organizationalstructure and policy framework as suggested below.

7.21 (i) Abolition of Special Public Enterprise Regimes: Special public enterprise laws and regulation, including personnel regimes, should be eliminated or limited to essential provisions outlining the general framework for the organizationand management of public enterprises. This approach avoids unnecessary details that unduly tie the hands of governments and enterprise management. Employees of the enterprises should be governed by the laws and contracts applicable to private sector enterprises.

7.22 (ii) Organizationof Public Enterprises as Ordinary Stock Cor- porations under Corporate Law: Most public enterprises are of a different nature and have different histories. For this reason, the government as shareholder should have the flexibility to adjust the charter of each one according to its nature, task and even managerial ability. However, there needs to be a consistent decentralizedorganizational logic, as described by commercial law. That is, the shareholderswill appoint the board of directors by voting according to their respective shareholdings. The board will, in turn, choose among its members a chairman of the board and will appoint and dismiss the general manager and assistant general managers. All employees will be hired and fired by management.

7.23 (iii) Reduction of Ministerial Control: One logical and indis- pensable corollary to restoring the normal function of the various parts of the corporate mechanism (shareholders,board, management) as provided by corporate law is to separate the ownership role from the sectoral role and to make sure that sectoral ministries concern themselves only with the analysis, formulationand implementationof macroeconomicand sectoral policy matters. This system excludes direct operational ties with ownership and its undue involvement in both business policies and day-to-day management. In many countries, the sectoral ministries need to strengthen their capabilitiesto manage sectoral policies and operationalresponsi- bilities. Thus, in most cases, the "*loss"of direct authority over public enterprises is not likely to create any severe employment problem. - 54 -

7.24 (iv) Creation of a Public Enterprise IntersectoralBoard and Secretariat (Focal Point): The ownership role of the government could be delegated to a Public Enterprise IntersectoralBoard (PEIB) or similar body. In addition, the PEIB could be charged with taking a leadership role in improving the institutionaland managerial environmentof public 'enterprises.The PEIB would ideally be composed of representativesfrom the leading ministries, as well as a number of industry leaders from private and public enterprises. The PEIB would typicallymeet only four to six times a year and would operate only as a decision-makingbody. One of the board's first tasks would be to constitute its own administrativearm, e.g., a permanentPublic Enterprise Secretariat.

7.25 Whereas the board's role would be policy-making,the secretariat would act as administrator,coordinator and executing agency of the board. The secretariatcould be organized as a quasi-governmentalunit that would allow management to compete through market-levelsalaries for highly qualified employees. While the secretariatwould report to its board, it Ls important that it have access to the prime minister's or president's office or to a ministry, either formally or informally. It should have a small and highly qualified staff, The general manager of the secretariat should have substantialexperience as a senior manager of a large private enterprise and preferably similar experience of a public one. The success of the secretariatwill depend on its ability to create and maintain a high quality board of directors and strong financial discipline,while at the same time avoiding excessive reporting systems and interferencein the day-to-day operations of public enterprises.

7.26 (v) Strengtheningof the Role of the Board of Directors: Because the government frequently lacks the time or the capacity to provide a large number of very different and complex public enterpriseswith strategic guidance and effective performance evaluation, this task must be delegated one level below the government, that is, to the board of directors.

7.27 The government as shareholder should confer to the board of directors all the powers and responsibilitiesthat such boards have in any regular commercial corporation. The board will represent the government's interest so far as it maximizes the health and long-term viability of the enterprise. Board members should be composed of leaders from industry, trade, academia and government. (See Annex II for further details on the effective use of boards.) While care must be taken to select highly qualified board members and to grant a high degree of autonomy, it is equally important that the board delegate full operating responsibility to management, not interfere in day-to-day operations and concentrate its energies on the more limited but equally important strategic role, i.e., on the provision of guidance and the monitoring of performance.

7.28 (vi) Strengtheningof the Role of the Chief Executive Officer (CEO) and Top Management: The government should create an environment for CEOs that: (i) rewards initiative and monitors performance(output) rather than regulates actions and control costs (input); (ii) states explicit financial objectives and rewards financial discipline; (iii) eliminates subsidies to cover operating losses; and (iv) delegates full operational - 55 - authority to the CEO. For example, the CEO should, within budgetary limits, have the power to appoint the top management team, subject only to board approval; to hire and fire employees and to make investment and expenditure decisions.

7.29 (vii) Decentralizationof Managerial Autonomy and Accountability to the Plant/SubsidiaryLevel: When the government decides to decentralize more autonomy and accountabilityto public enterprise management, care must be taken to assure that this process spreads throughout each one. For example, should there be sectoral holding companies, the government should ensure that the decentralizationprocess does not stop at the level of the holding company board, but that the board and management of operating enterprises also benefit from the process. Organizationalcultures resist change, and special attention must therefore be given to the management of decentralization.

7.30 The above list of recommendationsfor organizationaland institutional changes, by no means exhaustive, indicates the type of measures which if adopted can, at the very least, alleviate some of the glaring inefficienciesof public industrial enterprises. These measures, comple- mented by the nurturing of a competitiveenvironment, an appropriate industrial, trade, financial and regulatory policy framework and the existence of substantialmanagerial and financial autonomy and accountability, constitute a package of reforms that can improve the performance of public industrial enterprises considerably. - 56 -

ANNEX I

ORIGINS OF PUBLIC INDUSTRIALENTERPRISES

Since public industrialenterprises produce items that can be manufactured by private enterprises,which in some countries and sectors are the predominantproducers, the question arises: what led to the birth and development of public enterprises in such activities? Were the factors purely political and ideological,or were they primarily economic?

Based on evidence from the sample of countries,the most important factors in the genesis and developmentof public industrial enterprises have been: political/ideologicalconsiderations; the takeover of ailing private companies;an unwillingnessor inabilityof private companies to undertake certain large industrial investments;the maintenance of control over strategic sectors; internal growth and diversification; and efforts to foster greater competition.

Political and IdeologicalConsiderations

While purely political and/or ideologicalfactors are far less important in the creation of public industrial enterprises than for public enterprises in general, there have been specific examples of this rationale, for various periods, in France, Austria, Portugal, India, Pakistan, Zambia and Ghana.

Although a dirigiste traditionhas existed in France ever since the days of Colbert in the seventeenth century, it was only after World War II that genuine political considerationsbecame important in setting up public industrial enterprises. The turning point was General Charles de Gaulle's major speech on October IL,1944, at Lille, when he announced the takeover by the state of the management of all large wealth. A "morality" argument subsequentlyled to the public appropriationof firms whose owners had collaboratedwith the Nazis during the war. This was the case with Renault and the two companies that today comprise SNIAS. Similarly in Austria, the Western European natiLonwith the largest state-ownedsector, most of the nationalizationsin the immediate post-war period were political,with enterprisesbeing taken over from their Nazi owners. An example is the nationalizationof the Hermann Goering steel plant established in Linz, which is now part of the largest industrialenterprise in the country, Voest-AlpineAG. Other political nationalizationsin post-war Austria were the takeover of nearly all mining operations (iron ore, lignite, lead, zinc and oil), steel, aluminum, non-ferrousmetals, fertilizers and oil refining products, as well as some of the larger machinery and electricalequipment industries.

More recent examples of ideologicalfactors leading to the setting up of public industrial enterprises in Western Europe can be found in Portugal since 1974 and in France since 1981. In Portugal, the Decree Law of 1976--the legal framework for statutory public enterprises--wasa product of the revolutionof 1974 that overthrew Europe's oldest dictator- ship. In France, the role of the public sector in industry has increased - 57 - dramatically since the election in 1981 of the Socialist Party (with the Communist Party as a junior coalition partner). Most of the banking sector, two major financial companies and the five largest industrial groups were fully nationalized. The government also acquired majority ownership in five other industrial groups (three of which were foreign- controlled) and two main steel groups. These nationalizationswere clear parts of an overall political program, rather than the result of economic, financial or technical considerations.

Among the developing countries in the sample, there have been periods when political and/or ideological factors were predominant. The post-independenceleadership of India believed that socialist development necessitatedthe creation of strong and viable public enterprises. Even in Pakistan, where economic developmenthas traditionallybeen more private- oriented, there was a period (1972-77)of rapid increase in the size of the public industrial enterprise sector on purely ideologicalgrounds. In Ghana, particularlyduring the First Republic (1957-66),an important political considerationwas the breaking up of the economic power of foreign companies. In Zambia, the unilateral declarationof independence by Rhodesia (now Zimbabwe) in 1965 and the subsequent border clashes with that country contributed to a strong sense of urgency for industrialization, which led to the setting up of a number of public industrial enterprises under Indeco's management. Subsequently,the Mulungushi Declaration of 1968 marked a watershed in the growth of the country's public industrial enterprises.1/ The Zambezi Sawmills were taken over in 1968, and two large copper companieswere nationalizedin 1969-70.

While political and ideologicalconsiderations were important in some countries at some points, nationalizationshould not be confused with the creation of public industrialenterprises. Indeed, in many cases, public industrial enterpriseswere set up precisely to loosen somewhat the direct control of the state over certain industrial activities in the public sector. Thus, in Brazil, administrativereforms under Decree-Law 200 of 1967 made it attractive for units within the general government to be restructuredas public enterprises.2 / This option explains, to a large extent, the fact that the most spectacular growth of public industrial enterprises in Brazil occurred after a conservative,market-oriented military regime came to power in 1964. No less than about one-third of Brazil's federal public enterprises were establishedduring 1970-75. Similarly, in Israel, the Government Companies Law (Section 4a) requires public enterprises to behave in the same way as any private firm. This requirement explains why, although the Labor Government has been the leading force in successive Israeli coalitiongovernments (1948-77 and again since 1984), the establishmentand management of public industrial

1/ The Declaration refers to a speech delivered by PresidentKenneth Kaunda to the National Council of the United National Independence Party (UNIP); it has formed the basis of the economic philosophy of Zambia.

2/ For details on these reforms, see Helio Beltrao, "A Revolucao Silenciosa,"Visao, August 31, 1976. - 58 - enterprises have always reflected pragmatic considerations. Likewise, in France and Sweden, governments have explicitly stated that public enterprises should operate in business-likefashion. 3/

Takeover of Ailing Private Companies

In a number of cases, governments have decided to bail out large private industrialfirms (and sometimes whole industries)particularly in the automobile, steel, textiles and shipbuildingsubsectors to preserve employment or for some similar reason. Much European industry nationalized during the 1930s was in response to such considerations. However, there are also more recent examples in the sample countries. In France, the government'smassive rescue operations of the two largest steel companies, Usinor and Sacilor in 1978 and Renault's acquisitionof Berliet (Citroen's ailing truck subsidiary),are good examples. In Italy, EGAM, the state holding company set up for mining and metallurgy, acquired over 40 previously private companies in such diverse fields as shipping, insurance and textiles machinery.4 / State-ownedcommercial banks in Italy have frequently complained about being saddled with large rescue operations such as those of the two major chemical groups, SIR and Liquid-chemica. How- ever, sometimes they have had no choice: Credit Italiano, Banco Commerciale Italiana and Banco di Roma, the three large banks, are all controlled by IRI. Similar rescues of failing firms also occurred in Austria: in 1980, Landberbank took over Eumig, the camera and projector manufacturer.

Sweden's shipbuildingindustry, among the world's largest, came into state hands piece by piece for similar reasons, with the culmination in the establishmentof the state-ownedSvenska Varv in 1977, and the takeover of the last shipyard to remain in private hands, Kockums, in 1978. Similarly, the steel industry, traditionallyone of Sweden's most competitive, came under state control in 1977 with the establishmentof Swedish Steel, into which the loss-making steel companies were merged. Other take- overs involved Sweden's two large clothing and textile concerns and the two largest pulp and paper manufacturers. What is ironic about such expansion of state-ownershipin Sweden is that it occurred when the non-socialist parties were in power, parties that were elected on a platform of rolling back Swedish socialism. Faced with a widespread economic crisis in industry, the Swedish conservativesnationalized more property in 4 years than the Social Democrats had in 44 years.

Cases of takeover of ailing companies among the developing countries also exist, although they have been less dramatic. Several small companies have been nationalized in India and Pakistan for this reason. In Brazil, where until the last five years the economy has been expanding

3/ For example, in France, the introductionto the Law on Nationalizations of February 11, 1982 (OfficialGazette) states: "Nationalization should in no case mean statism."

4/ EGAM was later liquidated and its assets passed on to IRI and ENI. - 59 - rapidly, the only major takeover was in the mid-1950s; it involved three steel plants (Usiminas,Cofavi and Cosipa). In Ghana, a number of private companies were taken over in 1978-79 (such as the Dutch Company R.T. Briscoe), but the reasons were mainly alleged business malpractice such as tax evasion and violation of current regulations.

Unwillingness or Inability of the Private Sector

Many large-scale,capital-intensive and resource-basedprojects require investment and technology beyond the capability or willingness of existing national private industrialists. The extent of the risk involved in such projects and their long gestation periods sometimesexclude domestic private participation,and many countries have restrictionson foreign investment. The state therefore emerges as the sole entrepreneur. The emergence of public industrial enterprises for such reasons has been common in most developing countries, especially India, Pakistan, Ghana and Zambia. In Brazil as well, the setting up of the giant PETROBRAS and CVRD companies was dictated primarily by the need for capital on a scale too large for the domestic private sector, especially when foreign investment in these subsectors had been rejected on politicalgrounds. In Israel, the setting up of Dead Sea Works, which extracts and processes potash, and Israel Chemicals Limited (ICL) was the result of the large investment requirements. Even in developed countries,governments have set up public industrial enterpriseswhere the risks were perceived as too high by the private sector. Examples exist in such high technology industries as electronics,microprocessors, biotechnology, robotics, energy-saving equipment and telecommunications.

Strategic Industries

Many governments prefer to exercise control over certain key industrial subsectors they want to use to direct the economic and social development of the country. Such "commandingheights" arguments have been important in India and also in Pakistan, Ghana and Zambia in certain periods. Evidence from the sample countries indicates that, regardless of the size and stage of development of a country, state ownership is much higher in such strategic subsectors as steel and telecommunicationsas compared, for example, to textiles. In the same category are those public industrial enterpriseswith links to defense, such as the Israel Aircraft Industries and Brazil's state-owned producer of small aircraft (Embraer) and the Brazilian nuclear company (Nuclebras).

Internal Growth and Diversification

This reason for increasing the role of public industrial enterprises has been termed "creeping nationalization"by Edouard Bonnefous, the former president of the French Senatets finance commission. From a political point of view, internal growth and diversificationhave been less visible and less controversialreasons for expanding state ownership. For - 60 - example, while the number of French state-owned companies actually declined from 170 to 130 over the 1970s, their subsidiariesmultiplied from 266 to 650. In particular,Elf-Aquitaine, the French state oil company, has used its oil profits to acquire companies (such as the U.S. natural resources giant, Texas Gulf) and to diversify into chemicals, biotechnology,pharma- ceuticals, metals, plastics, solar energy and other fields.5/ In the case of Italy, a state company, GEPI, was set up in 1971 to develop a national capacity in electronics,aerospace and nuclear energy. Within four years, it had 54 operating firms for a total of 25,000 employees. In Austria, Chemie Linz, the chemical arm of OIAG, similarly moved increasinglyout of fertilizers and into synthetics and basic chemicals during the 1970s.

Among developing countries, the best examples of growth through diversificationare found in Brazil. The vast majority of public industrial enterprises set up in that country after 1967 in the mining and manufacturing sectors owe their existence to the creation of subsidiaries, occasionallythird- and fourth-generationones. The experience of PETROBRAS and CVRD are instructive. In the mid-1960s, PETROBRAS was a single domestic company whose primary activity was exploration. Today, it controls, or is a dominant partner in, over 70 other companies. Similarly, in the case of CVRD, the enterprise began its operations primarily in the mining and shipping of iron ore.. Then in the 1970s, it moved aggressively through its subsidiariesand minority-ownedaffiliates into bauxite, alumina, aluminum, manganese, phosphates, fertilizer,pulp and paper and titanium.6/

Creating Competition for the Private Sector

In a very few cases, public enterpriseshave been set up to prod the private sector through greater competition. One such example was the Israeli government'slaunching of a state-owned insurance company to compete with private insurance firms.7/ Another example was the setting up of the Ghana National Trading Corporation to compete with foreign trading firms and thus break their oligopolisticpower. While few public firms have been established for this motive, the degree of competition between public and private enterprises, and even of public industrial enterprises among themselves, can be an important factor in the performance of public enterprises.

5/ See Edouard Bonnefous, "Senat Rapport d'Informationsur le Controle des Enterprises Publiques en 1977," (Filiales et prises de participation des enterprises publiques),No. 379 (Paris: Senate of France, 1977).

6/ In its diversificationefforts, the enterprise overextendeditself. For that reason, together with a decline in internationaliron-ore prices in 1977, it was forced by its financial crisis to sell its subsidiariesin bauxite, aluminum, cellulose and manganese to the private sector.

7/ See "Yuval Israel Insurance Company," Tel Aviv University, Faculty of Management case study. - 61 -

ANNEX II

MANAGEMENT OF PUBLIC ENTERPRISESAND THE BOARD OF DIRECTORS

The board of directors is one of the three organizationalpillars of effectivemanagement of public enterprises. In one capacity, it serves as a crucial "buffer" between the ownership role of the government and the operating role of the chief executive officer (CEO) and senior executives of the enterprise. The ownership role, performed by the owner(s), involves defining the business charter, setting the objectives of the enterprise, appointing and dismissing the board of directors, and approvingannual accounts and dividend payments. These powers are formally executed through the annual shareholders'meeting. The operating role is performedby the CEO and the top management. Their major responsibilityis to manage and develop the enterprise in accordance with agreed objectives. The strategic role is the responsibilityof the board of directors. Its major functions are to decide upon strategic plans, to monitor performanceagainst targets, and to appoint and dismiss the CEO. Typically the board is part-time, meeting 6-12 times a year, and is composed of persons with high stature and professionalexperience relevant to the business of the firm.

The role of the board is examined here from the perspectiveof ideology, autonomy and accountability,roles and responsibilities,respon- siveness to market forces, problems common to existing boards of public enterprises,pitfalls to avoid, and ways to constitute an effectiveboard.

The Role of the Board of Directors

The board has three important roles vis-a-vis management: (1) approval of policy and strategy; (2) performance evaluation;and (3) appointmentand dismissal of the chief executive officer.

Policy-Makingand Strategic Role

The board's function here is to provide strategic guidance to the CEO and management by participatingin the formulationand evaluation the enterprise'sstrategic plan and by critically examining the enterprise's future direction. Major policy decisions taken by the board are normally logical extensions of the strategic plan. These include, inter alia, dividend policy, medium- and long-term borrowing over and above stated amounts, short-term borrowing in excess of normal trade requirements,major capital expendituresexceeding certain defined limits, issuance of shares, closure or sale of business units or major assets of the business, acquisitions,joint ventures or major technical licensing agreements,new or revised compensationand pension programs, and salary increases and bonuses for the CEO and senior executives. In making major policy decisions, the board serves as a check on the operating power of the CEO. Generally, however, these policy decisions are based on the recommendations of management, for which the board should receive adequate documentation and analysis in advance of the board meeting at which the decision is to be made. - 62 -

Given the complexity of large-scalepublic enterprisesand the number of diverse policy areas upon which the board must make policy decisions, it is not surprising that, increasingly,the boards have delegated their authority to committees ander them, such as, for example, audit, finance and strategic planning committees. Perhaps of greatest importance is the executive committee, which meets with the CEO to deal with relevant board matters in the interim between Board meetings. The committees allow the board to divide their responsibilitiesand work effectively,to call on members of management responsible for a specific area under board review and to call on outside expertise as required.

The PerformanceEvaluation Role

By conducting regularly scheduled, periodic meetings, the board can evaluate management's ability to develop and implement strategic plans as well as scrutinize its performance relative to stated objectives. Normally, it does so by a review and approval of strategic plans and annual budgets, and analysis and discussion of regular and special reports submitted by management prior to board meetings. Moreover, the board, as an independentverifier of management's performance,should also review the audited financial statements of the enterprise as well as other audit findings, as prepared by independentpublic accountants,who are normally appointed or reappointedby the shareholders'meeting on an annual basis. The board's evaluator's role is, therefore, primarily oriented to a review of management'seffectiveness in running the day-to-day business. This role clearly implies that management is to run the enterprise on an ongoing basis and that the board should abstain from interventionin day-to-day matters.

Appointment and Dismissal of the CEO

The board should normally be empowered to hire and fire the CEO as well as approve the appointment of key executives of the enterprise as recommendedby the CEO. This power should take the appointmentof the CEO out of the realm of pure politics and ensure selection based on merit and prior qualifications. Moreover, this power is the major leverage the Board has to deal with poor operating performance by a CEO and other management.

Why Some Boards of Directors Are Not Effective

The cause of ineffectiveboards is usually problems of design and/or implementation. The concept is a sound one that has stood the test of time and which, if properly employed, seems to be the best method of guiding a commercial enterprise. Difficultiesarise when the role of the board is misunderstoodor abuses are permitted in its operations. If the convening authority of the board expects it to approve the most detailed actions of management, it will fail. On the other hand, if the Board must seek approval of the convening authority before taking action, it will fail. The owner of the enterprise must be certain that the Board has full authority to carry out its role and that the board will, in return, permit management to have the requisite authority to carry out its role. - 63 -

The single most damaging abuse of these principlesis the appointment of board members who are improperlymotivated or incompetent. In the private sector, we note cases in which family members or social acquaintancesof the controllingshareholder are placed on a board or cases in which members are selected because they are certain to vote along with the majority. In the public enterprisesector, we find Board memberships dispensed as political favors or given to individualswho can be counted on to maintain the status quo.

When governments take over the strategicrole traditionally played by the board of directors and get entangled in operating decisions, problems surface quickly. The board's principal functions of providing strategic guidance and effective control are generally replaced by inadequate guidance, weak or no controls, and slow and politicizeddecision- making. Political relations become more important than productivityand financial performance. Enterprisemorale declines, and recruitmentand motivation of qualified managers becomes difficult at best. The overlap of the political and commercialspheres has had disastrousresults in the public industrial enterprises of many developing countries.

Too often, boards are regarded as a necessary evil by enterprise management. On the one hand, the board may intervene in day-to-dayopera- tions, usurping management'spower and perogatives. In addition,ministerial representativeson the board may hamper the operating effectivenessof the enterpriseby imposing political demands on management. On the other hand, with a politically powerful CEO, the board may collectivelybecome a 'rubber stamp" for management action. Neither extreme is desirable. Perhaps the most desirable situation is a state of creative tension between management and the board, with the balance of power roughly in equilibrium.

An Approach to Effective Board Operations

As public enterpriseshave grown in size and complexityand the external environmentsurrounding them has grown more uncertain, the ideology of socially oriented state ownership that has led to appointment of representativesof governmentministries as board members, together with the political appointmentof management representatives,is being increasinglyreappraised in light of the need to orient the enterprises toward the market. While the board must be representativeof the owner, the government, it can be staffed with members from outside government who offer a diversity of skills, independentviewpoints and experiences that can increase the effectivenessof the Board. Moreover, the appointment of independent,outside directors,with fixed renewable tenures, can offer the enterprise continuity of representationbeyond transient changes in government and, thereby, depoliticizethe board to a certain extent. That is not to imply, however, that the board should not be responsive to the overall requirementsof the government and the dictates of the law, or regulations governing public enterprises. To the contrary, these matters are one of the primary fiduciary responsibilitiesof board members. But the role should be objectivelyresponsive, based on defined criteria and not on the changing whims or dictates of the government or the ministries - 64 - represented on the board. In other words, in order to operate efficiently and effectively,the Board must be able to buffer or shelter the public enterprise from the political expediency of the moment.

Moreover, board members should be protectiveof the larger social good as well as the wider interests of the enterprisewith respect to such issues as, for example, environmentalimpact, service to consumers,market responsivenessto customers who may be in the private sector and dependent on the enterprise for goods and services, and avoidance of corrupt practices.

As such, the board of a public enterpriseshould be accountable to its owner, the government,based to the fullest extent possible on objective standards,which may include quantitativegoals, i.e., financial performance targets, and qualitative ones, normally related to its larger social role in society, as well as on compliancewith clearly established laws, rules and regulations. In this context, what constitutes an effective board? It should contain the following elements:

o Size: 6-15 members, depending on the size and complexity of the enterprise.

o Board Representation. The board should comprise members with professionalexperience relevant to the business of the enterprise. Chief eiecutive officers of industrial firms would be ideal candidates for the board of a public firm. Other suitable candidateswould be distinguishedsenior members of the banking, legal, accountingor educationalworlds. The professional standing of board members should be higher than or at least equal to that of the CEO of the enterprise. Normally, government officials,who directly represent the owner should not be on the board in order to avoid undue interferenceof the owner in the strategic role of the board and to avoid reducing the accountabilityof the board. Exceptions to this rule should certainly be permitted in the case of highly qualified government officials whose ability to make important contributionsto the board is unquestioned. The chairman of the board should preferablybe distinct from the CEO except in those cases where the CEO is the best suited candidate for the chairmanship. Above all, appointmentsto the board should be based on merit and the potential contributionof the member to the effective operation of the enterprise. Board members should have fixed, renewable tenures on a staggered basis so to provide continuityand avoid replacementof board 'memberswith changes in government.

o Board Committees. Some of the policy work of the board should be assigned to appropriate committees,with the assistanceof members of management and outside specialistsas necessary. Board committees avoid overloading the board with too many detailed matters. - 65 -

o SubsidiaryBoards. Boards of subsidiary firms should actively participatein providing direction and regional sensitivityto main operating subsidiaries,affiliates and joint ventures. Subsidiary boards will enhance the process of decentralizationin large public enterprises and further attune the enterprise to market conditions.

o Frequency of Meetings. The board should normally meet 6-10 times a year, with a calendaT or notice of planned meetings provided well in advance. Ideally the meetings should be rotated among the primary operating locations of the enterprise so that the Board is able to visit operationalsites and have occasion to meet with line management.

O Board Agenda. A clearly defined agenda should be prepared for each board meeting, which normally should consider such items as approval of previous board minutes, review of the financial results for the period as compared to the budget, analysis of reports on operating performance,decision-making on key policy issues, and handling of other items, including committee reports and specific problems.

o Board Materials. All documents to be discussed by the board should normally be posted to members one week in advance of meetings. Board members should not be expected to make decisions at meetings without supportinganalysis that is, the say so of management is not a sufficientbasis, nor should they be expected to digest critical material at the meeting.

O Minutes. Minutes of board meetings should be prepared and mailed out within a set period following the meetings.

O Compensationof Board Members. To attract experienced,capable, board participation,directors' fees should be paid to public enterprise board members commensuratewith the fees paid to private sector board members in the country. Normally, board members should be expected to commit 1-2 additional days per meeting to the enterprise over and above the actual day of the meeting.

Conclusion

The exact design of the role and functions of the board of director will clearly vary from country to country depending on the cultural and business environment. However, in most countries, the board can be an effective instrument for change in, and developmentof, public enterprises. All too often, however, it is treated in a proforma way by management, or else it is a politically charged vehicle for government interventionin key public sector enterprises. Neither approach is desirable, particularlyin light of the uncertain and difficultworld economic conditions. Public enterprises must be held accountable for the same level of efficiency as the private sector. This standard can only be - 66 - achieved with a professionalboard able to evaluate management's performance,make key policy decisions in a timely manner and guide the enterprise strategically. As such, the board must be accountableto the owner, the government, for results according to defined objectives, but not be controlledby government'spolitical whims. The board is an important buffer between government and management, but, more importantly,it must function in its own right as one of the three organizationalpillars of a public enterprise.

A board of directors is made up of highly charged, usually egotistical,human beings. Therefore, it is subject to all of the short- comings that befall any human endeavor. The board cannot be expected to solve all the problems faced by an enterprise,but when properly designed and constituted,it is the best arrangement for guiding a firm toward its mandated goals. Conversely,when a board is ill-conceivedor poorly constituted,it will be inadequate at best and obstructionistat worst. - 67 -

ANNEX III

THE USE OF HOLDING COMPANIES

Purpose

Of the 13 countries covered in this study, 6 rely on large multi-sectoralholding companies to manage a significantpart of the public enterprise sector. Italy establishedIRI in 1933 to manage the industrial portfolio of several large banks that were rescued by the government after the economic crash. In the other five countries, the use of holding companies is a more recent phenomenon: it was only in the late sixties and early seventies that strong holding companies started to play a dominant role (see Annex Table A-1).

Some of the purposes behind the establishmentof holding companies were: (i) to protect operating companies from undue political interference;(ii) to establish a professionalbody that on behalf of the government, could exercise the ownership role, e.g., provide effective strategic guidance and constructivecontrol; and (iii) coordinate decision-makingwithin the government as it relates to public enterprises. These are the more clearly stated objectives for establishingholding companies.

Table A-1: ESTABLISHMENTOF HOLDING COMPANIES

Year of No. of Estab. Country Holding Company Employees

1933 Italy Istituto per la RicostuzioneIndustriale (IRI), superholding 600,000

1965 Zambia Zambia Industrial and Mining Corp. Ltd. (ZIMCO), superholding Industrial Development Corporation (INDECO), subholding 20,000

1967 Ghana Ghana IndustrialHolding Corp. (GIHOC)

1970 Sweden Statsforetag 48,000a/

1970 Austria OsterreichischeIndustrieverwaltungs Aktiengesellscheft(OIAG) 150,000

1972 Pakistan Board of Industrial Management 80,000

a/ Prior to the break-up of Statsforetag. - 68 -

It is probable that the choice of a holding company to serve the above purposes was heavily influenced by a number of broad global trends that took place during the sixties and early seventies:

(i) Industrial organizationand economic thinking were characterized by technology, long-term planning and economies of scale, influenced by the peaking of the industrial era and demand-driven markets after World War II.

(ii) European countries encouraged by rapid economic growth during the fifties and early sixties impatientlywanted to move further with aggressive social and economic reforms. With increased knowledge of technology,economics and advanced planning, the belief was that economic growth could be further enhanced by increasing state interventionin the market place. "Why should we accept downturns in the business cycle that do not exist in East-bloc countries?"was a commonly asked question. Many European countries therefore embarked upon a more "activitist"industrial policy to promote direct and indirect investmentsin target industries and geographic locations, to bail out large, sick private firms, to reduce unemployment,to manage the business cycle, to further build social reform programs, etc.

(iii) African countries, once liberalizedfrom their capitalistic colonial masters, naturally looked for somewhat less capitalistic solutions in choosing an economic strategy. The planned economy approach to economic development,which included a large element of state-ownedenterprises, seemed to be a viable alternativeor complement,given the Soviet Union's relatively sharp economic progress during the two first decades after World War II.

With all these trends taking place the creation of large holding or superholdingcompanies to tackle the management problems of public enterprises surely looked like an effective solution to many countries.

Unfortunately,most of the underlying forces that supported those above trends changed dramaticallyduring the late sixties and early seventies. For example, the demand-drivenmarkets created by the rebuilding of Europe and Japan leveled out for the first time in the sixties; at the same time, developing countries became a significantfactor in the world market for manufacturedgoods; the increased speed of technologicalchange further fueled competitivepressures; the oil shocks put downward pressure on global growth; the western world entered the post-industrialarea; economies of scale became less of a competitivefactor; exploitationof human knowledge and motivation became increasinglyimportant; rigidity in long-term planning was replaced with flexibility,creativity and strategic options; the centralisticstrategy to planned economic development proved bankrupt;and decentralized,pluralistic and market-orientedapproaches to economic development took a lead role. Given all these new trends changes how have the holding companies performed? - 69 -

Performance

The financial performance of the holding companies has been very poor. Huge losses and extreme accumulation of debt have caused most of them into severe financial problems, with serious macro-economic consequences for the country in question. The only exception is Pakistan's holding company, whose performance is more difficult to judge because some 90 percent of its production is sold in markets with cost-plus price controls.

The reason for the poor financial performancecannot be blamed only on the holding company structure as such, but more importantlyon the host of decisions taken during the seventies that flew in the face of changing trends and market realities. On average, public enterpriseshave performed poorly whether they have been managed by holding companies or not (see Chapter 2). To compare the performanceof public enterprise with or without holding companies is impossible,given the many factors influ- encing performance. However, it is equally clear that holding companies have not been able to improve the performanceof public enterprise sectors in comparison with how they would have faired without holding companies. Of course, there is always the exception to the general rules. ENI, the Italian petroleum holding company, experiencedvery successful growth and profitabilityduring the fifties and early sixties. However, today circumstanceshave changed. ENI is much bigger and more bureaucratic, while the market place is more complex and competitive forces are stronger and faster.

What, then, are the strengths and weaknesses of the holding companies? These are discussed next.

Strengths

The major strengths of the holding company concept for management of public enterprises can be grouped into four broad categories:

Buffer against Excessive Political Interference

The creation of a holding company between public enterprisesand sectoral ministries can be one effectiveway of delinking the political and the commercial spheres. To achieve this objective, care must be taken to: (i) depoliticizethe selection of board members and managers of both the holding company and its subsidiaries;(ii) hire directors and managers with professionaland relevant experience;and (iii) duly decentralizethe ownership role of the holding company with respect to subsidiaries. Countries such as Austria and Norway have written into their public enterprise laws that certain government officials, such as ministers, secretariesof state and employees of the sectoral ministry cannot serve on the board of directors of public enterprises. However, in most countries even in Austria, the selection process is still heavily politicized. In Zambia the president of the country was the chairman of the board of the superholdingcompany, Zimco, for many years. In Austria, all the four major political parties get their proportionateshare of board members in - 70 -

the superholdingas well as in the eight subholding companies. Thus, depolitizationof the management structure of a holding company is at best difficult. Another important way to buffer against undue political [nterferencecan be to organize the subsidiariesof the holding company under private commercial law and to avoid any special statutory laws that extend the civil service culture of government to public enterprises.

ProfessionalManagement of the Ownership Role

Holding companies should be establishedoutside the civil service culture of the government and be staffed with people with professional managerial skills. This approach puts them in a much better position, as compared with the ministries, to provide effective strategic guidance and constructiveperformance evaluation. This structure normally will facilitatea more rational dialogue about social and financial objectives and improved financial discipline and systems. In fact, most of the holding companies have relatively sophisticatedfinancial planning and control systems. Unfortunately,enterprises are run not by systems but rather by people, and skills and motivationalfactors in most cases far outweigh the importance of systems.

Leveraging Scarce Management Skills

Especially in developing countries in the early phases of development,there may be situations where public enterprises can benefit from the pooling of certain scarce managerial talents into a central holding company. This central resource then guides and monitors project design and development,organization, staffing and performance. Scarce central skills can be leveraged to a large extent where plant level complexity is low. However, one has to be very cautious about how far this central resource can be leveraged. As markets and plant level complexity gradually increase and plant managers acquire appropriateskills, the central holding company resources must be decentralized,a transition that is frequentlydifficult because of vested interests and resistance to change.

Benefits from Cooperation (Synergy)

The benefits of cooperationare likely to be the greates between enterpriseswithin the same subsector, as for example, petrochemicalsand chemicals, iron ore and steel, forestry and pulp and paper. Research and development projects and upstream production processes too large for individual public enterprisescan be shared among several of them. Economies of scale can be an advantage in negotiatinglarge-scale procurement contracts as well in establishingforeign marketing outlets. Restructuringof an industrial subsector is in some cases easier if participatingfirms are under the same ownership umbrella. For example, the sectoral steel holding company, Swedish Steel AB, established in the mid-seventies,has been very successful in restructuringand increasing the competitivenessof the Swedish commercial steel business. Lower cost financing is another benefit frequently claimed by holding company managers. This result is likely when the holding company portfolio - 71 - consists primarily of small- and medium-sizedenterprises. With larger ones, the freedom and responsibilityto arrange their own financing is a powerful incentive to find new, creative, efficientand low-cost sources of financing in ways a central holding company frequentlywould not identify because it is further removed from the action. Last, the tax advantage offered by a holding company from of organizationcan be substantial. Losses in one public enterprise can be used to reduce income taxes in profitable ones in the same subsector and, depending on local tax laws, even in those in other subsectors.

With so many factors in favor of holding companies, why is it their performancerecord is so weak? Following is a discussion of some of the weaknesses of the holding company concept.

Weaknesses

Effective Tool for Political Interference

In a political environment,holding companies can be turned by politicians into effective tools. If the boards of directors and management of superholdingand sectoral holding companies are selected primarily along political lines there is a great risk that decision-making will become excessivelypoliticized. Similarly, when large amounts of state financingare provided to the public enterprise sector through the holding company for expansionaryand/or "bail-out"purposes, resource allocation tends to be politicized and ineffective,e.g., investment choices are poor, non-viable firms are bailed out, excessive employment in the face of mounting losses, etc. In these situations, the superholding company is an effective tool only for carrying out politicalwills.

Clearly, if IRI, the Italian superholdingcompany, employing some 600,000 people, had been organized into, say, 30 independent firms, it would have been more difficult for the government to exercise excessive political influence in their affairs. The Norwegian government contemp- lated establishinga state holding company in 1977, but after a study rejected the idea. It appears the main reason was that a holding company would give the politicianstoo powerful an instrument.

Excessive Layers of Management, and Slow Decision-Making

With the acceleratingpace of change in markets, products and technology,enterprises must have the resources and power to make quick, in many cases instant, decisions, if they are to compete effectivelyand survive in the market place. Superholdingand holding companies represent in most cases unnecessarylayers of management whose negative effects on enterprise performanceby far exceed the positive ones. Clearly, a private company whose major policy decisions must be approved only by its board of directors will have a decisive competitiveadvantage compared with the operating companies at the bottom of the IRI hierarchy. However, the problem here is not only the speed of decision-makingbut, equally important, the cost of bureaucraticpolicies and the motivationalfactors determining the quality of decision-making. - 72 -

Lower Motivation and Quality of Managers

Enterprises that have independenceand demonstratedeffectiveness are normally better able to recruit and keep high-calibermanagers than are enterprisesmore dependent on others for their decision-making. Independ- ent public enterprisesare fully responsible for all major business functions, such as obtaining financing from the banking system, maintaining their own internationalsales organizationand developing personnel policies. There is no holding company to bail them out if things go wrong. These enterprisesbecome masters of their own future. Such characteristics appeal to and attract higher quality managers. Since the long-term success or failure of most enterprises is closely linked to the quality of management motivationalfactors attracting and maintaininghigh-quality leaders are imperative for success.

Excessive Financial Powers.

The financial policies of holding companies can be more or less centralized. The most dangerous approach is a highly centralizedset of financial policies according to which subsidiariesare required to transfer all profits and cash balances to the holding company's accounts and where the holding company management has full power to allocate these funds. In this situation,profits and cash balances of viable firms are likely to subsidize non-viable operations. Some of the major problems of this approach are: (i) resource allocation is inefficient,as the holding company management,being removed from the market place, is less qualified to put excess cash balances into productive uses than are managers of successful firms; (ii) most viable firms must retain most of their after-tax earnings if they are to sustain long-term growth; (iii) non-viable enterprisesare maintaiLnedat great cost long after they reach their terminal stage; and (iv) a Lack of financial autonomy at the enterprise level has a negative effect on the motivation and quality of managers. A much more decentralizedapproach is desirable in most cases. Direct lending and borrowing between the holding company and its subsidiariesshould be handled at arms-length.

Lack of AppropriateRole.

The circumstancesthat influence the design of organizations clhangeover time, and organizationsneed to changed accordingly. However, b,ecauseof vested interests, resistance to change and rigidity in systems, it frequently takes a crisis to force needed organizationalreform. This was the case with the Swedish holding company, Statsforetag,and some of its major subsidiariesin steel, pulp and paper, and shipbuilding. In the late seventies, Statsforetag'ssubsidiaries in these subsectors had to go through a major restructuringaffecting the employment of thousands of people and costing several billion dollars. The size of the related financial requirementsand the social implicationsof plant closures were such that all communicationand negotiation soon took place directly between the management and boards of directors of the subsidiariesand the Ministry of Industry. Holding company management had a very limited, if - 73 -

any, role to play in the decision-making. In a country more formalistic than Sweden, all decision-makingwould probably have been slowed or could, in some cases, have even been stopped by holding company executives clinging to their formal powers.

In the end, the Swedish government decided to transfer the largest enterprises in Statsforetag'sportfolio to the Ministry of Industry. Further motive behind this action was to avoid having holding company management allocate 90 percent of its time to a few large money- losing firms and neglecting its responsibilitiestoward the medium- and small-sized firms where it would be more able to play an effective role.

When to Use Holding Companies

In trying to decide when and how to use holding companies each situationmust be evaluated on its own merit. The relevance of the different strengths and weaknesses discussed above will vary with each situation. There are, however, a number of more general lesson that can be drawn.

The popularity of large holding/superholdingcompanies to manage public enterpriseshas long since reached its peak and is now at a low ebb. The major reasons are that: (i) the broad trends that motivated the establishmentof large holding structures have changed dramaticallyand now favor more decentralizedand flexible management structures;(ii) the performanceof most large holding companies has been weak, and most are in crisis; and (iii) when weighing the strengths and weaknesses of large holding structures (see Annex Table A-2), the negative factors tend to outweigh the positive ones by far.

Table A-2: SUMMARY OF STRENGTHS AND WEAKNESSES

Strengths Weaknesses o Initial Buffer Against Excessive o Lack of AppropriateRole Political Interference o ExcessiveFinancial o Coordinationand Professional Powers Management of the Ownership Role o Effect Tool for Political o LeveragingScarce Interference Management Skills o ExcessiveLayers of o Benefits from Cooperation Management, Bureaucraticand (Synergy) Slow Decision-making o Negative Effect on Motivation and Quality of Managers - 74 -

The major reasons for establishinga holding company were in many cases to create a "buffer against excessive political interference"and to establish "professionalmanagement of the ownership role." These objectives can, in most cases, be better achieved by establishinga small public enterprise unit to exercise the ownership role, by decentralizing the financial and strategic powers to a board of directors of the enterprise,and by legally organizing the public enterprisesunder the commercial code and eliminatingthe statutory codes promoting a civil service culture (see the recommendationsin Chapter 3).

"Leveragingscarce management skills" and "benefits from cooperation (synergy)"are the two main objectives that will continue to motivate the use of holding companies also in the future. However, the scope of these holding companies is likely to be more limited and focused. It is primarily in the case of medium-sizedenterprises that both developing and developed countries can get the most "value added" from holding companies. In this instance, the holding company can provide certain technical and managerial skills, improve financial discipline,stimulate beneficial cooperationamong enterprisesand provide broader opportunitiesfor management developmentand career planning. Large enterprisesshould normally have all the required managerial and technical skills within themselves,in which case the holding company has very little if any positive value to contribute;rather, its "value added" is likely to be negative. Small enterprisesshould be privatized, since they require an entrepreneurial management style rarely found in the public enterprises. The above discussion is summarized in Annex Table A-3.

Large enterpriseswithin certain sectors, such as petrochemicals, steel and manufacturing,may benefit from being organized under a sectoral holding company. However, in these circumstances,the holding company takes on the character of corporate management for a group of operating companies. This type of "operating"holding company is not considered in Table 4.

Table A-3: USE OF HOLDING COMPANIES

MARKET SITUATION

COMPETITIVE MONOPOLY

Private SIZE SMALL Privatize Holding

OF MEDIUM Holding Holding

PEs Directly Owned Directly Owned LARGE by Government by Government -75-

STATISTICAL ANNEX TableS&-1: Key Indicators for Major Public In&strial Enterprises, 1984

SAME ASSEIS Mr INIUOE EfRRSE aWIXY - DDM SUBSFclcR (US $ millim)

A. Public Ehterprises frmm Sanple Coxtries

1. Eati Nazionale Idrocaburi(EI) Italy 25,798 23,989 (50) 130,897 Petroletn 2. IRI Italy 23,354 NA NA 504,915 Metal mfg;shipbldg,etc. 3. ElfA4quitaine France 20,662 16,895 743 76,219 Petroleam 4. Ccompa.ieFrancaise des Petroles France 18,159 11,011 149 44,981 Petroleem 5. Petrobras Brazil 17,087 13,797 633 58,507 Petrolemt 6. Rimilt France 12,227 10,727 (1,436) 213,725 mbtor vehicles & parts 7. Indian Oil India 8,856 2,594 80 29,757 Petroleem 8. Copapie Gererale d-Electricite France 8,480 9,372 68 161,900 Electromic, appliances 9. Montedison Italy 7,044 6,098 (47) 71,215 Chanicals 10. Saint-Gobain France 7,015 5,903 59 125,228 [a] B1t. mtls;netal prod. 11. Tthoen France 6,620 7,285 (2) 111,900 [a] Electrunic, appliances 12. Rlune-oulenc France 5,856 4,487 227 79,230 [a] lumicals 13. Sacilor France 4,334 4,433 (920) 63,583 Metal mfg-steel 14. Pechiney France 4,064 3,884 62 48,230 Metal mfg-aliunin,steel 15. V1M-Alpine Austria 3,780 5,329 27 69,988 Metal mfg-steel;etc. 16. Oil & Natural Gas Cmissiom India 3,367 5,931 781 37,016 Petroleam 17. Steel Authority of India India 3,013 6,892 (208) 205,236 Metal mfg-steel 18. Aerospatiale France 2,875 4,627 38 35,000 Aerospace 19. cmV Austria 2,834 1,463 12 7,297 Petrolemi 20. EFl4 Italy 2,547 1,941 (324) 40,400 Metal mfg-steel 21. Zambia Industrial & Mining (ZI}CO) Zaibia 2,272 3,593 (12) 131,390 Mining-copper 22. Petroleos de PErtugal (PEIEROGAL) Portugal 2,160 2,722 4 6,910 Petrolen 23. Ctenie Linz Austria 1,822 402 7 7,847 Chenicals 24. Canlhia Vale do Rio Doce (CVRD) Brazil 1,685 4,863 915 20,299 Mining-irmn ore 25. Charbonnagesde France France 1,677 2,990 (24) 50,978 C'bmicals/inining 26. SNBXXA France 1,628 2,345 7 25,475 Aircraft engines 27. Matra France 1,571 1,829 8 28,000 Aerospace 28. Enterprise Miniere et (:iimique France 1,570 780 7 12,761 Mining-copper 29. Bull France 1,555 651 56 26,453 Cacuters,office equip. 30. SSAB(Svnskt Stal) Sweden 1,445 1,247 24 15,202 Metal mfg-steel 31. Statsforetag Crkxp 9weden 1,428 1,349 7 24,901 Industrial,trans. equip 32. Bharat HeavyElectricals India 1,191 2,096 36 73,180 Industrial,trans. equip 33. UI5MAS Brazil 960 1,920 59 14,606 Metal mfg-steel 34. Austria Tabakwerke Austria 925 319 9 2,217 Tobacco 35. Pakistan State Oil Pakistan 912 119 8 1,972 Petroleum 36. Siderurgica Nacional Brazil 897 3,805 (166) 27,208 Metal ifg-steel

Subtotal 211,670 177,688 837 2,584,623

[a] Also includes certain subsidiaries oined 50%or less, either fully or on a prorated basis. -76-

Tble SA-1: Key Indicators for YMjor Public ltAtrial Ehterprises, 1984 (citinued)

MS ASSS NEr UCOI

INTEVUSE omum - - -4?mrlfslBG (US $ million)

B. Other Public- hterprise

1. Pb (Petroles M icms) !mico 19,405 37,433 7 175,420 Petroleun 2. VolknagI GerMy 16,035 10,438 86 238,353 Notor vehicles 3. Kwait Petrolem Kuwit 14,997 16,692 958 14,640 Petrolem 4. Petrolsos de V1emzela Venezuela 13,598 17,109 2,239 43,553 Petrolau 5. Yaciaientos Petamliferas (YI) Argentina 6,880 9,110 (3,593) 33,728 Petrolem 6. Chinese Petrolem Taini 5,943 4,323 531 21,702 Petrolan 7. Neste Finland 5,364 2,434 12 8,422 Petrolen 8. British Steel Britain 5,008 5,221 (382) 78,750 Metal mfg-steel 9. Britibh leylard Britain 4,543 2,581 108 80,478 Meter vehicles 10. Statoil NPrWy 4,364 5,057 144 4,855 Petrolem 11. NDrsk Hydro NM-Way 4,354 3,918 130 19,825 Chemicals 12. E tWE Spain 4,305 1,649 20 5,422 Petroleum;chemicals 13. lsrkiye Petrolleri Thrkey 4,180 3,182 354 16,504 Petroleun 14. Petro-Csada Cmada 3,626 6,851 194 6,697 Petrolan 15. Saigitter Germmy 3,423 2,615 (153) 45,920 Metal mfg;shipbldg 16. Cockerill Sare Belgiun 2,991 2,392 73 19,159 Metal mfg-steel 17. Saarberg,rke Germany 2,455 1,560 (29) 31,030 Petroleu;mining-coal 18. berizh Oil Britain 2,298 1,182 56 21,474 Petrolem;auto parts 19. Isoor S. Africa 2,262 3,137 59 62,600 Metal mfg-steel 20. VWAG Gerniny 2,097 1,897 38 20,979 Metal mfg-steel 21. Bipresa Combhiana de Petrol Colcibia 1,907 1,540 28 10,403 Petroleum 22. Rolls-Royce Britain 1,882 1,295 27 40,900 Aerospace 23. Tabacalera Spain 1,658 555 7 8,614 Tobacco 24. caUlnw-QRuz Chile 1,535 3,570 144 25,339 Mining-copper 25. EIresa Nacional del Petroleo Chile 1,416 995 60 3,531 Petroleum 26. British Shipbuilers Britain 1,323 882 (346) 48,500 Shipbldg;trans.equip 27. Petroperu Peru 1,310 783 9 9,821 Petrolazn 28. Philippe Nstional Oil Philippines 1,240 1,447 42 9,316 Petrolam 29. Eiso,-Qtzeit Finland 1,149 1,629 12 13,402 Paper & Vood products 30. ENBSID Spain 1,115 1,852 (153) 21,012 Metal mfg-steel 31. SEAT Spain 1,087 2,203 (225) 23,610 MDtorvehicles 32. ChxnaSteel Taiim 1,026 2,669 111 7,620 Metal mfg-steel 33. Valmet Finland 1,000 756 15 16,258 Indh,trial,transp equip 34. Hispanoil Spain 997 253 37 262 Petrolem 35. GCAMINES Zaire 971 2,002 43 35,846 Mining-copper

Subtotal (Other Public Baterprises) 147,744 161,212 663 1,223,945 Subtotal (Ssple Contries) 211,670 177,688 837 2,584,623

Orsnd Total 359,414 338,900 1,500 3,808,568

Source: "Ibe largest NI-U.S. Corporatioms Ranked by Sales." Fortume (August 19, 1985) - 77 -

Table SA-2: Shares of Public Inistrial Enterprises in Selected Indicators, 1982 (percentaes)

Share of IrKiustrial Total Inrdustrial Total Irkdustrial Total Industry CDP GDP Investiurnt Investnent Emplyment EaplWnent in GDP

Austria 29 11 19 14 25 9 39

Brazil 31 11 33 20 14 3 35

France a/ 22 7 22 16 16 8 32

Ghana b/ 48 3 85 55 63 13 7

India 35 9 75 22 23 3 26

Israel 33 9 50 12 13 4 27

Italy 13 5 22 12 15 6 40

Pakistan 26 7 50 13 22 3 27

Portugal 28 11 70 18 14 5 40

Sweden 26 8 46 16 17 5 31

Tunisia 56 20 69 36 55 25 36

Zanbia 80 30 95 80 85 10 38

Average for all countries 35.6 10.9 53.0 26.2 30.2 7.8 31.5

Average for developed countries c/ 22.5 7.7 27.2 14.5 18.2 7.0 35.5

Average for developixg countries 42.1 12.5 65.9 32.0 36.1 8.2 29.5

Source: Staff estimates; World Bank, World DevelopmentReport 1985, (NewYork: Oxford University Press, MNUDD) a/ Includirg enterprises nationalized in 1982. x xlnirg joint ventures. c/ Austria, France, Italy and Sweden. DISTRIBUTORS OF WORLD BANK PUBLICATIONS ALGERIA FILAND KORE REUUC OF SPAIN O&e d Pinkie Akatw_ini Kiaupp Pan Kora BookCorporation Mundi-Prena Ub. S.A. Ur _vwsim PO. Sox 1i P0. O.x 101. Kwangwhamun Cto 37 1. plc centrasdei Aknoua SF4 1 Seoul 2800tMIMdM Alp, Heldik 10 KUWAff SRI LANKA AND THE ARGENTINA FRANCE MWBMLI Carlos Hirch. SEL WorddBank Publicao P.O. Box 548 Lae Hous Sookshop GaWa Gum 66 Avenue d'lea AAS. .pO. Bomx244 Florida165, 4th FloorOk. 46346575116 Park MA LA YS0 Si Ch pa A. 1333DIAtmOS w GER,MAN FEDERAL Bookhop, Limited Gadinb Mawatha AUSTRALIA, PJAUANEW REPULIC OF P.O. Box lt27, alanPantai Oaa Co0U bo 2 GUNA. FL. SOLOMON UNO-Ve1* KualaLumpur SWEDEN SLANDS.AND VANUATU D-S30 Bonni F i IniovLiine De r Snmnck b OTC3 ABCEFrites Kungl. Ovx W Docuent Delver GREEC SanLorenzo 153-I1.Col. del HvokhAtd Syo y N0S.G 001 MEMRD vafie Deleg.Benito Juarez Attn: Mr. EkldeSegerback Sydney.NSW 2001 24,Ippodarmou Street 03100Mexico City Ree~stn1.Box 16356 AUSTRIA Athens..116355-0 Stkol Geroldand Co. MOROCCO For Subecnptionordens A-GeroWind HONG KONG, MACAU MEMRS Wennn-Wiliams AD GrAben31 Asia 2000Ltd. 2 RueMoliee Raine Box3004 Graben31 6 Fl.. 146Princ EdwardRoad, W. Casablanca S-10425 Stockholm SAHRAIN Kowloon UI.ANDS SWIERLAND MEMR 2o1K MedicalBooks Europe, DV (MDE) Lbrr Paot P0OBox 22103 HUNGARY Noorderwal38. 6 RueGrenus ManamaTown 3U7 Kultura 72418L Lochem Caseposl 381 BANGLADESH PO.Box t NEWZEAUND CHt211Genevell Asitlcr dSoietyvnrie 39 uaes 2R. Hill and Son. Ltd. TANZANIA G.P.O. BoxSOO q(l WINDIA PrivateBas Oxford Universiy Pr GPO.h USSPublshers' Diributo Ltd. New Ma*et P0. Box5299 Post8ox7015 Auckland Dar esSaan BELGIUM New Delhi110002 NIGERIA THAILAND PubWcatioNde NationsUnis 10First Main Road UniversityPre Limited CentralDe ntStore 1060dri Roi 202Gandhi Nagar ThreeCrowns Building Jericho 306Salom Road 1l60 Brussels Bangalore560009 PrivateMail US 50f Bangkok BRAZIL ApeeasyChambers. P.O. box 736 Ibaan TRINIDAD TOBAGO Pubikacoes Tecnia 5 WallaceStreet TRWAI*NIDADtudiAGO InnyteacionaisLtda. Bombay400001 NORWAY SystematisStudie Unit Rua Peixoto Goride. 209 S81-., ChowTingheeL ne Tanum Karl Johan, A S 55 Etern Main Road 01409Sao Paulo. SP Calcutta 700016 P. O. Box 1177Sentrum Cure" t CANADA 71188,1(CA). Swarup Nagar' 0idW Ii Lo Diffuseur Kampu 2002 PAISTAN SiA C.P.46,1501 Ampere Swreet Sivaga oad Mirza BookAgency Socete Turusierne*de Difusion Bomh8vi6e. Quebe Nang& 65, Shahirah-e-QUAd.e-Azam 5 Avenuede Carhae j4B SE6 Nun-600034 P.O.Box No. 729 Tunis C4HSIMdrsL0E3 Lahore3 TURKEY Editonal Renacimiento INDONSLA, PANAMA Ham Kitapevi A.S. Miralnones354Li Cultura 49 sk Cuesi Satag )1. Sam Ratulangi37 Panamena. S.A. Bevoglu-Istanbul Jakarta Pust Av. 7,E 16 COLOMBIA P.O.box 181 Pnm oeIUGANDA Boao EnlaceLtda. IRELAND Uganda BokP.O.s ApartadoAero 34270 TDC Publishers PERU SOA Box7145 " D.E. 12 North Fredenck Street Editorial Desarrolo SA Kampala Apartado Aereo4430 Dublin I Lptdo MUND ARABEMIRAXES Cali Valle ISAEuift MEMRB CaSTAVAe ISRAELes t PHILQINES P.0. Box6097 Librena Trejos IJe sem Post Budding National BookStore Shariah Calle 11-13 P.O. Box81 701 Rizal Avenue UNID KINGDOM AND Av. Fernandex Guell Roe lenkd 91WO Metro M NORTHERN IRELAND 5 nJose rtALY PORTUGAL Microinfo Ltd. c&mEDIVouaIRECnnbbwa a LivrariaPortugal P.0. Box3 Centred'Edition et de Difousn SPA S RuaDo Carmo 70-74 Alton, HampshireGU 34 2PG AfliCa (CEDA) Via Lamarinora45 1200Lisbon England 04 B.P.541 CaselaPostale 552 SAUDI ARABIA VENEZUELA Abidan 04 Plateau 50U1Florence larkrBook Store Libreia dl Eate CYPRUS JAPAN P O. Box3196 Aptdo. 60.337 MEM8 EasternBook Service Riyadh 11471 Caracas 1060-A P.O.Box 2098 37-3.Hongo 3-Chome, SINGAPORE,TAIWAN, WESTERNSAMOA Nicosia Bunkyo-kuI11 SURMA WesleyBookshop DENMAIt3C Tokyo InformationPubliations PO. Box207 SamuundsLitteratur JORDAN Private,Ltd. Apia RomoernsAlge n1 MEMRD 024061stFl., Pe-Fu Industrl ZIMBABWE DK-1970Coperhge V PO. Box 3143 Bl, 24 New Industral Ro TextbookSa Pvt. Ltd. EGYan ARAB UBLIC OF Jl Singapore Box3799 Al Ahra Am SOLTITAFIUCA Harare Al CaGmStet KENI Oxford Unrivesity Press Cairo AfcBaook Sewvice(E.A.) Ltd. Southern Afria P.O. Box45245 CaPeTown 1000 NaMCbw The World Bank Headquarters European Office Tokyo Office U 1818 H Street, N.W. 66, avenue d'1ena Kokusai Building Washington, D.C. 20433, U.S.A. 75 1 16 Paris, France 1-1 Marunouchi 3-chome Telephone: (202) 477-1234 Telephone: (1) 47.23.54.21 Chiyoda-ku, Tokyo 100, Japan Telex: WUI 64145 WORLDBANK Telex: 842-620628 Telephone: (03) 214-5001 RCA 248423WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTONDC

ISSN 0256-2235 ISBN 0-8213-0815-7