Department of Economics
Doctoral Thesis
“Decentralization and Sources of Interprovincial Growth; Dependency and Disparities: A Case study of Pakistan” by Tahir Sadiq
Registration # DECON01141001
Supervisor Prof. Dr. M AslamChaudhary HoD, Economics (UoL)
i
“Decentralization and Sources of Interprovincial Growth; Dependency and Disparities:
A Case study of Pakistan”
By:
Tahir Sadiq
Registration # DECON01141001
This dissertation is submitted to Department of Economics, The University of Lahore, Pakistan in partial fulfillment of the requirement for the completion of Ph.D. in Economics.
Approved by:
External Examiners
i) Dr. Shabbir Ahmed
ii) Dr. Ahmed
Internal Examiner/ Advisor
Chairman (committee)
Head of Department
ii
Author’s Declaration
I, Mr. Tahir Sadiq, hereby state that my PhD thesis titled:
“Decentralization and Sources of Interprovincial Growth; Dependency and Disparities:
A Case study of Pakistan” is my own work and has not been submitted previously by me for taking any degree from The
University of Lahore or anywhere else in the country/world.
At any time if my statement is found to be incorrect even after I graduate from the university
(UOL) has the right to withdraw my PhD degree.
Tahir Sadiq Registration # DECON01141001 PhD candidate
iii
Plagiarism Undertaking
I solemnly declare that research work presented in the thesis titled:
“Decentralization and Sources of Interprovincial Growth; Dependency and Disparities:
A Case study of Pakistan” is solely my research work with no significant contribution from any other person. Small contribution/help wherever taken has been duly acknowledged and that complete thesis has been written by me.
I understand the zero tolerance policy of the HEC and The University of Lahore towards plagiarism. Therefore, I as an Author of the above titled thesis declare that no portion of my thesis has been plagiarized and any material used as reference is properly referred/cited.
I undertake that if I am found guilty of any formal plagiarism in the above titled thesis even after award of PhD degree, the university reserves the rights to withdraw / revoke my PhD degree and that HEC and the University has the right to publish my name on the HEC/University Website on which names of students are placed who submitted plagiarized thesis. I declair that plagiarism in this thesis is as per HEC/UOL policy.
Signature:______
Tahir Sadiq Registration # DECON01141001 PhD candidate
iv
Dedication
To my Wife and my Son
Muhammad
v
Acknowledgment
to bestow the ﷺ First of all, thanks to Almighty Allah Who sent the Holy Prophet knowledge to humanity. I would like to express my special thanks of gratitude to my supervisor Prof. Dr. M. Aslam Chaudhary who provided a continuous support and guidance to complete my PhD.
One is motivational force and one supporting force, in my thesis motivational force is my wife and supportive strength has been presented by one and only Prof. Dr. M. Aslam Chaudhary, as without his guidance it would not have been possible for me to complete this research study. He is really a mentor and great asset for the university especially and to the country. Lastly, I would like to thanks Mr. Wasim Saleem Phd student at PIDE for helping me in this long journey to complete this task. Last but not the least, a special heartiest tribute to my parents, wife, children and other family members, for suffering during my busy schedule of research days. Thank God it is over now.
Tahir Sadiq
Registration # DECON01141001
vi
Accroynames ADE Average Daily Employment
CPI Consumer Price Index
DFID Department for International Development
ECM Error Correction Model
EG Engle Granger
GDP Gross Domestic Product
GOP Government of Punjab
GRP Gross Regional Product
HCI Human Capital Index
HDI Human Development Index
HIES Household Integrated Economic Survey
HYV High Yield Variety
IPP Institute of Public Policy
IPR Institute of Policy Reforms
KPK Khyber Pakhtunkhwa
NEPRA National Electric Power Regulatory Authority
NFC National Finance Commission
NIHDI Non-income human development index
PGDP Provincial Gross domestic product
vii
PIDE Pakistan Institute of Development Economics
PILDAT Pakistan Institute of Legislative development and Transparency
PPAF Pakistan Poverty Alleviation Fund
RIA Regional Income Accounts
SPDC Social Policy & Development Centre
VA Value Added
viii
Abstract Regional (provincial) growth inequalities and deprivation among provinces of Pakistan is one of the hot issues. It is a general perception that small provinces are neglected in the development process, while main focus of development policies was on large provinces. This study the first of its kind, which analyzes the relationship among the growth rates of provinces in Pakistan. It also explores sources of regional inequalities among provinces. In this context, the role of monetary and fiscal polices, are also explored. To understand, the nature of economic relationships and regional pattern of growth among provinces, the study Applied Engle Granger (EG) cointegration technique. The use of EG enables to analyze inter-provincial growth rates on one to one basis. The study has estimated 144 co-integrating equations to explore the dynamic relationship between GDP, agriculture, industry and services sectors of Pakistan and in its four provinces. Empirical evidences are drawn within sectoral relationship and among the four provinces of Pakistan.
The results reveal that GDP growth rate of Punjab and Sindh provinces complement each other, in the development process. However, it revealed that the deprived provinces of Balochistan and KPK growth rates depend upon Sindh and Punjab. The sectoral cointegration analysis indicated that linkages across sectors and provinces are weak; among poor provinces. The findings confirm that there is a significant sectoral relationship and integration between relatively developed provinces of Punjab and Sindh. Their growth benefited each other, across provinces. This may be due to the fact that planning in Pakistan, before the 18th Amendment in the constitution, was centralized, and polices were focused on sectoral development rather than regional development; while regional income inequalities were ignored which led to provincial deprivation and inequalities. Further, the results of the study also supported that the growth of agriculture and industry, in each province, significantly depends on the growth of the services sector of the same province, as well as, on the service sectors of other provinces.
The study has also estimated the impact of monetary and fiscal Policy on inter provincial growth differentials. For this, purpose panel data was utilized by applying fixed effect model. The analysis indicated that fiscal policy helped to reduce the regional growth disparities in Pakistan, whereas monetary policies enrage the situation of increasing income inequalities in the provinces.
The study indicted high variations in the level of financial development across provinces, which may be one of the reasons for less development in the poor provinces like Balochistan and KPK. Lastly, an important finding of the study is that the incidences of terrorism and electricity load shedding, have augmented the regional growth disparities among provinces and in Pakistan. The shortfall in electricity has most affected the large scale manufacturing sector of Sindh, whereas
ix
the terrorism mostly affects KPK and Balochistan, which further fueled deprivation of the neglected province.
The study suggests that planning and policy making should use from bottom up approach to address the above cited emerging economic issues. Provinces be empowered to have lion’s share in financial exchequer and economic decision making. The central planning body had members from each province who had responsibility to keep check on regional equity and disparities. The very concept of the institution has lost, since government servants (CSP officers) are serving on these positions now and they hardly play their role in this respect. It is recommended that the central planning body (Planning Commission) and Central bank should play their role in curtailing regional income inequalities and deprivations. There is an earnest need to integrate all provinces in to the main framework of economic development, on equity basis. Besides, although there is a right step of NFC award, which provides more resources to provinces but still there is a need to strengthen it more in terms of more economic power to provinces if they have to be integrated. Special development programs are needed to integrate relatively underdeveloped provinces of Balochistan and KPK; if at all regional equity has to be maintained and these provinces have to bring out of the sense of deprivation.
x
Table of Contents Author’s Declaration ...... iii Plagiarism Undertaking ...... iv Dedication ...... v Acknowledgment ...... vi Accroynames ...... vii Abstract ...... ix Table of Contents ...... xi List of Tables ...... xvii CHAPTER1 ...... 1 Introduction and Statement of the Problem ...... 1 1.1 Introduction ...... 1 Table 1.1: Province wise GDP Per Capita at constant factor cost of 2005-06 (Rs.) ...... 5 Table 1.2: Province-wise GDP Growth Rate from 1973 to 2013 (Percent) ...... 5 1.2 Statement of Problem ...... 8 1.3 Research Questions ...... 9 1.4 Objectives of the Study ...... 10 1.5 Novelty of the study ...... 11 1.6 Organization of the Study ...... 12 CHAPTER 2 ...... 13 Literature Review ...... 13 2.1 Theoretical foundations of Economics Growth ...... 13 2.2 The Demand Switching Policies and Economic Growth...... 15 2.3 Empirical Analysis of Growth and Macro Econometric Modeling ...... 17 2.3.1 Empirical Determinates of Economic Growth...... 17 2.3.2 Macroeconomic Modeling in Pakistan ...... 19 2.4 Regional Growth Theories ...... 23 Figure 2.1: Schematic Diagram for Regional Growth determinants...... 26 2.5 The level of Regional Disparities in Pakistan ...... 28 Table 2.1: Pakistan: Infrastructure by Provinces ...... 33 xi
Table 2.2: Annual Growth rates of the provincial Economies, in different time periods (%) ...... 34 Table 2.3: Estimated Provincial Gross Domestic Product (PGDP) of provinces by sectors 1999-2000 to 2014-15 ...... 37 2.6 Role of Human and Physical Capital in Economic Development in Pakistan ...... 37 2.7 Role of Human and Capital infrastructure in economic development ...... 42 2.8 Terrorism and economic development ...... 44 2.9 Employment, Output, Economic growth and Productivity ...... 50 Table 2.4: Elasticity: Results of the Models ...... 55 2.10 Research Gap ...... 58 Chapter 3 ...... 61 Data Source and Methodology ...... 61 3.1 Sources of Data and Methodology ...... 61 Table 3.1: Province-wise Indicators and Data Sources ...... 62 3.2 Money Supply by Province ...... 62 3.3 Gross Fixed Capital Formation by Province ...... 63 3.3.1 Private Sector ...... 63 3.3.2 Public Sector...... 64 3.3.3 The Government Sector ...... 64 3.4 Final Private Consumption Expenditure ...... 65 3.5 Variables Used for the Study ...... 66 Table 3.2: Variables Used in the Study ...... 67 3.6 Strategy for Provincial Integration by Using Sub-Sectoral Causality Analysis ...... 68 3.7 Lag Selection Criteria ...... 70 3.8 Unit Root Test ...... 71 3.9 Co-integration Test ...... 71 Table 3.3 Sector-wise Co-Integration Equation ...... 73 3.10 Growth Equation ...... 73 3.11 General Model for the Study ...... 74 3.12 Impacts of Fiscal and Monetary Policies on Provincial Economic Growth Rates ...... 75 3.13 Impact of Human, Social and Physical Capital on Provincial Growth ...... 76 3.14 Estimation of Capital Output-ratio and Employment Elasticity ...... 77
xii
3.15 Impact of Electricity Load-shedding and Terrorism on Economic Growth of a Province ...... 77 3.16 Modeling Technique ...... 79 3.16.1 Fixed Effects Model ...... 79 3.16.2 The Random Effects Model ...... 81 Chapter: 4...... 84 Intra-Provincial Economic Linkages, ...... 84 Economic Fluctuations and their Adjustments ...... 84 4.1 Sectoral Development in Pakistan ...... 85 4.2 Industrial Development ...... 86 4.3 The Services Sector ...... 88 4.4 The State of Provincial Integration in the National economy ...... 90 4.5 Profile of the Provinces ...... 91 4.5.1 The Province of Punjab ...... 91 4.5.2 The Province of Sindh ...... 92 4.5.3 Khyber Pakhtunkhwa (KPK) province ...... 93 4.5.4 Balochistan Province ...... 93 4.6 Cointegration and Error Correction Model (ECM) ...... 94 4.7 The National Economy of Pakistan: Granger Causality and Cointegration Analysis ...... 95 4.7.1 Economic Linkages of Punjab: Empirical Evidences of Cointegration and Error Correction Model ...... 97 4.7.2 GDP Growth of Punjab and its Linkages...... 97 4.7.3 Punjab: The Linkages of Agriculture Sector and Adjustment of the Sector ...... 97 4.7.4 Punjab Province: Industrial Development and Adjustment ...... 100 Table 4.1: Co-integration Results for Punjab ...... 101 4.7.5 Punjab Province: Services Development and Adjustment ...... 104 4.8 Sindh: Cointegration and Result of ECM ...... 105 4.8.1 GDP Growth of Sindh and its Linkages ...... 106 4.8.2 Sindh: The Integration of Agriculture ...... 106 4.8.3 Industrial Development and Linkages of Sindh ...... 108 4.8.4 Sindh: The Integration of the Services Development ...... 111 4.9 KPK Province: Empirical Results: Cointegration and ECM ...... 113
xiii
4.9.1 Economic Profile of KPK ...... 113 4.9.2 GDP Growth of KPK and its Linkages ...... 113 Table 4.3 A: Cointegration Results for KPK ...... 114 Table 4.3 B: Cointegration Results for KPK ...... 115 4.9.3 KPK Province: Co-integration of Agriculture and Its Linkages ...... 115 4.9.4 KPK Province: Industrial Growth and Its Linkages ...... 117 4.9.4 KPK Province: Services Sector and Its Linkages...... 120 4.10 Empirical Findings for Balochistan: Cointegration and ECM...... 122 4.10.1 GDP Growth of Balochistan and its Linkages ...... 122 4.10.2 Balochistan Province: Empirical Evidences; Linkages of Agriculture Sector ...... 123 Table 4.4: Cointegration Results for Balochistan ...... 124 4.10.3 Industrial Sector of Balochistan: Linkages and Dependency ...... 126 Table 4.5: Cointegration Results for Balochistan ...... 127 4.10.4 Services Sector of Balochistan: Integration and Linkages ...... 128 Table 4.6: Cointegration Results for Balochistan ...... 130 4.11. Summary of the Findings ...... 132 Annexures Tables: Chapter 4 ...... 138 Chapter 5 ...... 151 Impact of Monetary and Fiscal Policies on ...... 151 Inter Provincial Growth Disparities ...... 151 Table 5.1: Growth differential across provinces and across sectors ...... 152 5.1 The Model: Determinants of Economic Growth ...... 154 5.2 Regional Growth Disparities ...... 157 Table 5.2: Fixed effect coefficients of the Growth models ...... 159 5.3 Provincial Current Expenditure ...... 160 Table 5.3: Per capita current provincial expenditures 1990-2015 ...... 162 5.4 Provincial Development Expenditure ...... 162 Table 5.4 Per Capita provincial development expenditures from 1990-2015 ...... 164 5.5 Federal Expenditures ...... 164 5.6 The Impact of Money Supply growth and Financial Market ...... 168 5.7 Summary / Conclusions of Fiscal and Monetary determinants of growth ...... 170
xiv
Appendix 5.1 ...... 173 Chapter 6 ...... 174 Impacts of Economic Shocks on Inter-ProvincialGrowth Disparities ...... 174 6.1 Emergence of Terrorism ...... 175 6.2 Electricity (Energy) Shortage & Load Shedding...... 175 6.3 The Role of Physical and Social Infrastructure ...... 176 6.3.1 Terrorism and GDP Growth of Provinces ...... 177 Table 6.1: Losses of terrorist Attacks ...... 178 Table 6.2: Cost of terrorism ...... 179 6.4 Impact of Terrorism on Regional Disparity ...... 179 6.4.1 The model and Empirical Findings ...... 180 Table 6.3: Results of model of Terrorism and Growth ...... 180 6.4.2 Terrorism: Fixed effect coefficients of the provincial Growth ...... 181 Table 6.4: Fixed effect coefficients of the provincial growth with terrorism ...... 181 6.5 Electricity Load Shedding and Industrial Value added, by Provinces ...... 183 Table 6.5: Sectoral contribution towards GDP growth ...... 184 Table 6.7: GDP per unit of Electricity by province in Pakistan Year 2014-15 ...... 184 6.5.1 Impact of Energy Crisis ...... 185 Table 6.7: Hours of Outages and Cost of Load-shedding by Province ...... 186 Table 6.8: Fixed Effect Model Results with and Without Load-shedding Dummy ...... 187 6.5.2 Fixed Effect Co-efficient and Their Impacts ...... 189 6.5.3 Impact of Electricity Shortage ...... 189 Table 6.9: Fixed Effect Coefficients of Provincial Manufacturing Growth andProvincial Industrial Electricity Consumption ...... 190 6.6 Changes in the Manufacturing Share in National GDP ...... 192 Table 6.10: Province-wise Share in National Manufacturing Sector ...... 192 6.7 Infrastructure Development and GDP Growth of Provinces ...... 194 6.7.1Role of Physical and Social Infrastructure in Economic Growth: The Modeled Empirical Evidences ...... 197 6.8 Regional Employment Elasticity: Provincial Estimates Introduction ...... 200 6.8.1 Estimation and Discussion of results ...... 200
xv
Table 6.11: Results of employment elasticity across provinces, 1990-2015 ...... 200 Table 6.12: provincial employment structure of labour force ...... 204 6.9 Conclusions ...... 206 Chapter 7 ...... 211 Conclusions and Policy Implications ...... 211 References ...... 222 Annexures ...... 234
xvi
List of Tables Table 1.1: Province wise GDP Per Capita at constant factor cost of 2005-06 (Rs.) ...... 5 Table 1.2: Province-wise GDP Growth Rate from 1973 to 2013 (Percent) ...... 5 Table 2.1: Pakistan: Infrastructure by Provinces ...... 33 Table 2.2: Annual Growth rates of the provincial Economies, in different time periods ...... 34 Table 2.3: Estimated Provincial Gross Domestic Product (PGDP) of provinces by sectors 1999-2000 to 2014-15 ...... 37 Table 2.4: Elasticity: Results of the Models ...... 55 Table 3.1: Province-wise Indicators and Data Sources ...... 62 Table 3.2: Variables Used in the Study...... 67 Table 3.3 Sector-wise Co-integration Equation ...... 73 Table 4.1: Co-integration Results for Punjab ...... 101 Table 4.2: Cointegration Results for Sindh ...... 107 Table 4.3 A: Cointegration Results for KPK ...... 114 Table 4.3 B: Cointegration Results for KPK ...... 115 Table 4.4: Cointegration Results for Balochistan ...... 124 Table 4.5: Cointegration Results for Balochistan ...... 127 Table 4.6: Cointegration Results for Balochistan ...... 130 Table 5.1: Growth differential across provinces and across sectors ...... 152 Table 5.2: Fixed effect coefficients of the Growth models ...... 159 Table 5.3: Per capita current provincial expenditures 1990-2015 ...... 162 Table 5.4 Per Capita provincial development expenditures from 1990-2015...... 164 Table 6.1: Losses of terrorist Attacks ...... 178 Table 6.2: Cost of terrorism ...... 179 Table 6.3: Results of model of Terrorism and Growth ...... 180 Table 6.4: Fixed effect coefficients of the provincial growth with terrorism ...... 181 Table 6.5: Sectoral contribution towards GDP growth ...... 184 Table 6.7: GDP per unit of Electricity by province in Pakistan Year 2014-15 ...... 184 Table 6.7: Hours of Outages and Cost of Load-shedding by Province ...... 186 Table 6.8: Fixed Effect Model Results with and Without Load-shedding Dummy ...... 187 Table 6.9: Fixed Effect Coefficients of Provincial Manufacturing Growth and Provincial Industrial Electricity Consumption ...... 190 Table 6.10: Province-wise Share in National Manufacturing Sector ...... 192 Table 6.11: Results of employment elasticity across provinces, 1990-2015 ...... 200 Table 6.12: provincial employment structure of labour force ...... 204
xvii
CHAPTER1
Introduction and Statement of the Problem
1.1 Introduction The pace of economic growth, as well as, regional development are widely analyzed phenomena.
One of the most desirable outcome of economic growth is that masses of a nation equitable benefit from the fruit of economic growth; the desired goal of trickle-down effect will take place and ensured, otherwise, economic growth without equitable growth may lead to deprivation for some regions which creates political unrest too. Regional income inequalities have serious implications for economic development of a country, country creates political instability, deprivation and social unrest. It is generally understood that that separation of Bangladesh was regional income inequality and deprivation between two parts of the country i.e West and East
Pakistan. Presently, the small provinces also have the feeling of deprivation due to the regional income inequalities, particularly in Balochistan1 In order to identify the level of regional income inequalities, regional income accounts (RIA) are a very helpful tool. Bangaliwala (1995),
Bangali and Sadaqat (2005), Arby (2008) and Arby and Rasheed (2010), GOP (2008), SPDC
(2014), IPR (2015), Nadia (2017) and Pasha (2018) have attempted to decompose the national
GDP into provincial GDP. Data is not compiled on provincial level in Pakistan. Data is hardly available at district level or at division level. All the above-mentioned studies only tried to decompose the national GDP into provincial GDP, but none of the above studies tried to analyze the relationship and linkages between regional growth and their implications. As the devolution
1see for details Chaudhry .M. A. Agriculture Development and policies with special reference to Balochistan. (1989).
1
process took place, the role of provinces became crucial in economic development. In Pakistan first time provinces were given reasonable financial autonomy in Musharaf era i.e in 2000s.
National Finance Award provided resasonable funds at the disposal of provinces. So far financial and decision powers were concentrate at Central; Federal level. The 7th National Finance
Commission (NFC) award in the late 2009 significantly increased the financial shares of provincial governments in national tax revenues. After the 18th Amendment to the constitution,
17 ministries were devolved to the provinces. Thus, the revenues, as well as, responsibilities of the provinces increased. For efficient provincial planning there is an increasing need to have up- to-date information on provincial economy, as well as, their emerging economic issues like disparities. Governments and institutions for planning and policy purposes need to know the development level and standard of living of the people at regional level. Regional (Provincial level) income estimation is the most useful barometer to measure the economic growth and development of a region. Per capita income in a region is an indicator of standard of living of people. Thus, it is important to construct Gross Provincial Product (GPP) on a regular basis.
This helps provincial governments to identify sectors where they have a comparative advantage over other provinces and areas that need to be focus to accelerate growth and as well as to integrate them into the national economy. It is also important to see the extent to which macroeconomic policies like monetary, fiscal and trade policies affect different regions. Do these policies affect all regions equitability there are discriminatory at provincial level? Pakistan is basically agriculture-based country. Its over 20 percent of GDP come from agriculture, which is mostly concentrated around Indus Basin (Punhab and Sindh), where heavy investment was made by the federal government, which could be a source of regional disparities for more details see
2
Chaudhry. M.A (1990) and (1994). If so, it could be a source of inequalities which need to be addressed.
The rate of economic growth and growth differentails between countries is one of the main objectives of macroeconomic policies. An ample body of literature indicated regarding determents of economic growth at a specific country level and at a cross country level. Solow,
(1956), Swan, (1956), Ramsey (1928), Cass (1965), and Koopmans (1965), Diamond (1965), P.
Romer (1990), Grossman and Helpman (1991a), and Aghion and Howitt (1992) develop models to analyze the growth differential and sources of growth among countries; to formulate efficient policies and guide lines for improving income and to accelerate growth region and country level.
Another important question that economist tried to analyze is the growth differential within a country. There is a need to maintain a balanced growth among different sectors and regions
(provinces) for political and economic stability. A region is a continuous area with a clear-cut spatial location and borders. Regional (Provincial) accounts can be defined as a mirror image of the national accounts. Each region is treated as a separate economic entity. These accounts provide important indicators for assessing regional economic diversity and the relative importance of different economic entities at regional level. It can also be used to understand regional disparities and distribution of income amongst the various units of an economy and region. Therefore, it provides a very useful information to accelerate growth as well as calk out policies to address regional economic issues.
In Pakistan annual estimates of national income are prepared by Pakistan Bureau of Statistics
(PBS) a federal agency. These accounts are presented in aggregated form and are prepared for the whole economy but ignore regional development and data for the same is not released.
3
Shares of different sectors in GDP are also given for the economy as a whole. The economic territory of Pakistan is subdivided into the four provinces; Punjab, Sindh, Khyber Pakhtunkhwa
(KPK) and Balochistan. In addition to this, Federally Administrated Tribal Areas (FATA), Gilgit and Swat territory are also under the management of Pakistan; but these are small terrorities.
These provinces are governed by provincial governments and have unitary authorities.
Since the national GDP growth is the accumulation of provincial GDP growth, so any change in the GDP growth performance of a province will affect national GDP growth. But hardly any such information is available. The general overview of the provincial economies reveals the fact that Punjab is dominated by agriculture and industry, Sindh has relatively higher share of
Industry and also of agriculture. Balochistan has same share of Mining and Quarrying and KPK has ame contribution in mining, forestry, and electricity production, but these provinces remained relatively poor, insignificantly of poor political management. Punjab is the largest province in terms of population and economy size, followed by Sindh, KPK and Balochistan.
This indicates that all the regions are very different in terms of economic development; level of activities is also different. While analyzing the provincial growth of the economy conventionally, top to bottom approach was applied in Pakistan. most of the economic and financial powers were concentrated to centre. Most of the studies took macro indicators and tried to address the economic and social issues of the Pakistan but ignored regional (provincial) economic and economic issues. As data on many social and economic indicators at the provinces was not available. It is also important to see the impact of macro variables on the regional (provincial) growth. The non-availability of data at provincial level was a major bottleneck to carryout regional studies. The pioneering work on aggregating the provincial GDP was done by
Bangaliwala (1995). The table 1.3 below provides the provincial economic growth rates since 4
1973. For details see Chaudhey. A. M. (1989), it may be noted that these figures are not based upon any primary data, rather complied from national data.
The table 1.1 gives the GDP per Capita at constant factor cost of 2005-06, Sindh has. the highest
GDP per capita followed by Punjab, Balochistan and KPK respectively. It is also important to note that KPK has the highest per capita average annual growth rate followed by Punjab, Sindh and Balochistan.
Table 1.1: Province wise GDP Per Capita at constant factor cost of 2005-06 (Rs.) Years Punjab Sindh) KPK Balochistan
(GDP of (GDP of Sindh (GDP of KPK (GDP of Balochistan Punjab / Pop of / Pop of Sindh / Pop of KPK) / Pop of Balochsitan) Punjab)
1990 33,006 46,238 25,960 43,858 1995 37,386 48,550 29,697 50,277 2000 41,038 51,064 31,134 47,974 2005 46,655 58,957 35,063 52,381 2010 50,205 64,925 41,357 47,734 2015 55,581 68,550 47,256 48,989 Average annual 2.74 1.93 3.28 0.47 Growth (1990-2015) Source: Calculated by author by using SPDC (2005), IPR (2015) and Pakistan bureau of Statistics
Table 1.2: Province-wise GDP Growth Rate from 1973 to 2013 (Percent) Punjab Sindh NWFP Balochistan Pakistan 1973-1978 3.88 3.04 2.76 3.90 3.48 1978-1989 5.84 6.89 6.36 4.55 6.16 1989-2000 4.92 3.69 4.38 4.13 4.43 2000-2008 4.87 6.21 5.63 3.34 5.23 2008-2013 3.02 2.08 4.97 1.88 2.86 1973-2013 4.75 4.76 5.01 3.75 4.71 Source: Calculated by author by using SPDC (2005), IPR(2015) and Pakistan bureau of Statistics
The table 1.2 shows that in the high growth period of 1978 to 1989 and 2000 to 2008, Sindh has higher growth, as compare to all other provinces. In the low growth periods, Punjab manages to perform better. This pattern in the growth performance of the provinces highlight the structural 5
and policy specific issues, which has created regional (provincial) income inequalities and disparities among relatively poor provinces. Balochistan hardly ever match with national growth; a source of disparity and result of centralized planning Chaudhry, M. A. (1989) calculated that the per hector output of Balochistan was 4.0 time higher than Punjab, 2.6 times higher than
Sindh. The author further discussed that however it as mainly due to small agriculture sector, which remained underdeveloped and ignored by the centre. The result is that its ample resources remained untapped i.e over 2 million hectors are not developed, minerals are not extracted, as a result the province presented a picture poverty and deprivation.
From a high GDP growth of 3.9 percent during 1973-78, the GDP of Balochistan decreased to
1.88 percent during 2008-2013. As stated above, agriculture is dominated sector in Balochistan, the virgin land was hardly brought into and still small population lives in grazing land. Over time, the modern inputs like HYV and use of fertilizer and irrigation remained poor in the province. As a result, agriculture growth fell. To uplift the province, there was need to introduce green revolution, which was not done2. Some of the major issues, related to the provinces, are highlighted by ADB (2005) as Balochistan is the largest province of Pakistan in terms of area; which covers 44 percent of the area but is having 5 percent of the country’s population. The main strength of the province is its natural resources, agriculture and 750 KM coastal highway, over two virgin lands, which are yet to be exploited. There is a lack of any economic policy related to investment in the province especially, in minerals and fisheries. It is however ignored Province in National Development Plans. Hardly any effort was made to bring this province in the national stream. It is very poor in terms of social and economic indicators for details see Pakistan
Economic Survey, Government of Pakistan (different issues).
2see for more details Chaudhry. M. A (1987) Phd. Thesis Temple University USA. 6
Pakistan Poverty Alleviation Fund (PPAF: 2013) Another important finding of many of development programs implemented in Balochistan in past is that in the community driven projects; in most of the cases the 1st and 2nd tier is highly unsustainable in the long run. So there is strong need to mobilise recourses in such a manner so that priorities should be set out and focused time to time programs for the improvement of social sector are to be implemented. The linkage of the programs to skill development and community works are to be ensured inorder to benefit the poor household of the province. The report states that the human development indicators are poor as gender disparity is high in the provinces, female literacy is hardly 16 percent compared to 32 percent at national level.3 So, Balochistan requires a great deal of policy attention towards its economic and social deprivation. For this purpose, research is needed which may dig out such sources and provide basis for addressing regional issues. Such literature is not present. This may be very reason that this study is undertaken.
The GDP growth of NWFP was also low during 1973-1978, as reported in the above table 1.3 as
2.76 percent. It increased to 4.97 percent in 2008-2013. Although, the province has faced several challenges in its economic and social development but eventually the KPK province has made some progress to catch up with other provinces. But it still need to be integrated into the pace of development at national level. The World Bank (2005) reported that key challenges to NWFP included poverty as during 1990-91, 37 percent of the population was living below the poverty line, whereas during 2001-02 it was 35 percent. Besides, the province failed to develop industrial base, which could address unemployment and integration of the economy.
It is important to explore the linkages between GDP growth of provinces by sector in Pakistan. it will help us to understand the growth interdependence among provinces. This is the least
3See for detailsPakistan Poverty Alleviation Fund (PPAF) (2013) 7
explored area in Pakistan. It is not explored that up to what extent provincial growth depends on their own endowments and growth of other provinces. It is also unknown that how macro policies effect provincial growth in Pakistan. The general perception is that macro policies effects growth and development of large province (Punjab & Sindh) more as compare to small provinces (KPK and Balochistan). Because, Punjab and Sindh have a large share in population so, get’s a larger share of revenues and federal expenditures thus contribute to growth of these provinces. Small province (KPK and Balochistan), lack in basic services and development due to unequal distribution of resources at federal level. This needevidence-based research, and this study will answer most of the above-mentioned issues. Which are rarely explored in existing literature and it will initiate a new debate upon regional integration in Pakistan.
1.2 Statement of Problem The provincial growth disparities have serious implication for the country It may lead to economic and social disparities which may ultimately translate to disintegration of the country.
Therefore, it is important to investigate sources of such disparities and provide guidelines for for economic policies which may be focused on equitable growth at region (provinces).
Pertaining to Pakistan, there is hardly any literature on regional growth growth disparities in
Pakistan. One of the obvious reasons for this is the lack of availability of data and neglect of the issuse issue by the federal authorities. This has happened also because economic planning was in the Federal domain, which was not focused on equitable development across regions. The approach now requires ‘bottom up’ whereby Provincial equitable growth is focused and provinces are provinces are enabled by approaching financial decisions making powers. To do so there is a need to investigate and highlight sources and relationships. This requires research on
8
provincial economies, the identification of forward and backward linkages at provincial level, inter and intra provincial linkages, as well as, among sectors.
The study will investigate and contribute in the literature by answering the following research problems.
1.3 Research Questions The study will attempt to answer several questions pending so far, to be explored, for regional development in Pakistan. The main focus will be on:
Is the growth of provinces in Pakistan is co-integrated with each other?
Whether overall growth rate of one province complement the growth of rate of other province?
Are there different determinants of regional growth, at provincial level?
Do the of Federal Fiscal and Monetary policies created economic differentials between
Provinces?
Did electricity load-shedding has affected provinces differently i.e for industrial activities across provinces?
Has terrorism differently affected the economic growth among provinces?
To chalk out policy implications for provincial growth and development, based upon the empirical results pertaining to the above questions, will be the ultimate goal to explore above cited questions.
9
1.4 Objectives of the Study The specific objectives of the study are given below.
I. Overall Provincial GDP growth linkages will be explored, in terms of whether GDP
growth of a province, in short run and in long run, spill over and affect GDP growth of
another province.
II. The economic growth linkages and growth and sectoral interdepence are also studied by
this research.
III. Study will point out and analyze the determinants of economic growth at provincial level
and the role of Fiscal and Monetary variables on the regional (province) growth in
Pakistan will also be studied.
IV. In the light of the outcomes (iii) above the study will also analyze the overall growth
differentials among provinces to provide policy framework for regional integration.
V. The impact of recent emerging issues of electricity load-shedding, will be explored
particularly, on the industrial activities of provinces.
VI. Study will also estimate economic impacts of terrorism on the economic growth of all
provinces. It will be highlighted whether, it has equally affected the provinces or created
inequalities.
VII. Based upon empirical evidences, policy recommendations will be provided.
10
1.5 Novelty of the study It is a first study which is focused on provincial growth inequalities. More specific focus on how regional growth contributes to overall economic growth by exploring grass root to top central level approach for development. There is hardly any comprehensive study which had been carried out carried out so far, pertaining to Pakistan’s economy. The study will explore the integration among provinces, within province at sectoral and within provinces. The spillover effects of sectoral growth across provinces, which have never been explored so far. To analyze the sectoral causality within provinces and across provinces; the study has first time estimated 88 pair-wise Granger Causality equations for Agriculture Sector, Manufacturing Sector and
Services Sector. So, total 289 pair-wise Granger Causality equations have been estimated by the study to analyze causal relationships between GDP growth rates and sectoral growth rates of the provinces. The study has also estimated the long run and the short run adjustment paths of GDP growth rates and sectoral growth rates among provinces, which have never been explored. For this, the study used 144 cointegrating equations between GDP growth rates and sectoral growth rates of the provinces.
An important contribution is the identification of centralize policies role which has contributed to regional inequalities. The need for equitable distribution of the fruits of economic growth is highlighted on the basis of empirical evidences. The impacrs of new emrging issues like shortage of electricity and its impact on industrial sectors has been quantified. Moreover, the impact of terrorism the growth of each province which has not been identified so far, Thus, as ststed above is analyzed in terms of its impact on economics of provinces which may have been a source to increase regional inequalities. The study has several new contributions into existing literature, as
11
it initiated new debate on regional inequalities; based upon key macroeconomic factors at provincial level.
It will provide base to such analysis, at sectoral level and national level has not been carry out so far; a gap in the literature which is being fulfilled through this study. To formulate new policies by the planners.
Most importantly it will help to address equitable distribution of the fruits of backward areas into the main framework of economic development. Thus, by achieving equitable growth and benefit will reduce deprivation and political tensions among provinces. which will be the contribution of the study.
1.6 Organization of the Study Hereafter, the organization of the study is as follow: chapter 2 consists of literature review.
Chapter 3 outlines the methodology used for the analysis. In chapter 4, the findings are analyzed and policy guide lines are explored. Chapter 5 and chapter 6 will explore the growth differentials among provinces. Chapter 7 provides conclusions of the study and policy guide lines are also provided in this chapter.
12
CHAPTER 2
Literature Review
2.1 Theoretical foundations of Economics Growth The economic growth is the most analyzed phenomena in macroeconomic analysis. Many theories have been developed to understand the sources of economic growth and growth differential among different countries, and still it is one of the top subject for research. To analyze the growth differentials, at regional level, there is a need to study and understand the growth process of country. This section will review the existing literature on growth theories which will setup the stage by identify the important determinates of economic growth.
Adam Smith to Marshall, all economists believe that higher rate of economic growth is mainly due to capital accumulation (Roll (1938), Hahn and Matthew (1964), Madison (1991)). The rationale was that more and efficient use of resources raises productivity of workers, which is the main source of economic growth. So, to promote investment capital there is a need improve rates of saving. Harrod (1939) and Domar (1947a, 1947b) used Keynesian model and assumed a constant rate of saving and capital output ratio in deriving growth model. In their model, the rate of growth of output related to the rate of saving and capital output ratio. Solow, (1956) and
Swan, (1956), Ramsey (1928), Cass (1965), and Koopmans (1965) and Diamond (1965) further develop the models to analyze the cross country economic growth differences. They assume that capital, labor, and knowledge are inputs, and these are combined to produce output in a country.
It also assumes that labor and knowledge grow at a constant rate and saving is exogenous. The models show that saving rate has a level effect on output but not growth effect. The underlying assumption of above models that all factors of production and saving rate is same for all areas in a country which may not be true. These assumptions may be true at country level and may differ 13
across regions, but the assumption of homogenous factors of production across all regions of a country and constant saving rate for all regions, may not be applicable for regions. Secondly, the assumption of free mobility of factors of production within a country may also be not true because some areas or sectors within country are suitable for certain type of activities and those activities requires certain amount of some factors of production, so it is not possible to shift all factors of production if return from other activities has increased. It is more true for developing countries like Pakistan, where monopolies are there in production and different level of human resource development in regions is not same, for example 2 percent literacy (women) in
Balochistan cannot match with Punjab where female literacy is more than 4 percent. These resources are hardly mobilized across regions or provinces.
The following studies showed that agriculture is a classic example for such situation, if one area is suitable for cropping then some capital of that is not moveable to other area for some other activities. Theories of basic technological change are most important for understanding why the world as a whole-and, more specifically, the economies at the technological frontier can grow in the long run? But these theories have less to do with the determination of relative rates of growth across economies.
Above all models show that the capital’s earnings reflect its contribution to output, and capital accumulation does not account for a large part of either long run growth or cross-country income differences. And the only determinant of income in the models other than capital is the
“effectiveness of labor”, whose behavior is taken as exogenous. Romer. P (1990), Grossman and
Helpman (1991a), and Aghionand Howitt (1992) analyzed one question, how knowledge is produced in an economy and how its behavior contributes in the long run. The above models argued that technological progress generated by the discovery of new ideas was the only way to
14
avoid diminishing returns in the long run. In these models, the purposive behavior that underlay innovations hinged on the prospect of monopoly profits, which provided individual incentives to carry out costly research. Many studies showed that the model for research and development has an edge on other models of growth in explaining cross country growth differences. The growth of knowledge appears to be the central reason that output and standards of living are so much higher today than in previous centuries.
Based on above theoretical researches it can state that regional growth may differ because of different rate of technical progress with in regions, the growth of capital stock may vary between regions, the growth of the labour force may vary between regions.
2.2 The Demand Switching Policies and Economic Growth Ajisafe and Folorunso (2002) analyzed the role of demand switching policies on the economic growth. The multiplier effect is used to estimate the impact of fiscal and monetary policies on the economy. In macroeconomic policy framework fiscal and monetary policies are of huge essence.
Relative policy importance for macroeconomic stability has been subjected to debate for a long period of time. The debate starts with Keynesian framework and monetarist proposition and it never ends.
Keynesian view support exogeneity of government expenditure that take part in economic growth. Cyrus and Elias (2014) verified fiscal dominance in economic growth. While Wagner
(1890) claimed reverse causation among these variables imply endogeneity of fiscal policy also proved by Ansari, Gordon and Akuamoah (1997). Neoclassical tradition about crowding out effect on output challenged Keynesian proposition Spencer and Yohe (1970). Monetarists support monetary policy dominance in output and inflation determination Scarth (2014).
Monetarism believes in interest insensitivity of money demand implies that changes in aggregate
15
demand directly lead by money supply which alter nominal output. It means the economic growth can be generated through forceful push of positive monetary shock. Contemporary macroeconomic theories lay emphasis in both disciplines to achieve macroeconomic stability.
Mundell (1971) emphasized on both policy relevance where monetary policy should deal with inflation dynamics and external matters while fiscal stance should determine supply side of economy aims to safeguard internal stability. Furthermore, modern Keynesians believe that slope of both the IS and LM schedules are in normal range where both monetary and fiscal policies are effective Scarth (2014).
Exogenous changes in role of trade i.e. imports and exports that are components of aggregate demand are also likely to alter equilibrium income. Furthermore, openness of economy changes the autonomous expenditures’ multiplier value and thus, vulnerability of the economy to both foreign and domestic changes in autonomous expenditures Scarth (2014).
These models bring rest of the world into analysis via external sector and they try to model the impact of external sector via exchange rate and difference between domestic and foreign interest rate Mankiw (2012). But at regional level these two variables are same for all the intra country regions, so may be these variables are not relevant for determining regional growth difference at intra country level. It seems that more important variables at intra country level are differences among labor and capital productivity, social and physical infrastructure and spatial difference which will allow some regions to take more benefits from a similar shock.
16
2.3 Empirical Analysis of Growth and Macro Econometric Modeling 2.3.1 Empirical Determinates of Economic Growth The major concern of an empirical analysis is to quantify the qualitative theoretical analysis to make it useful for formulation of economic policy. Madison (1991) Mankiw, Romer and
Weale (1992), Barrow and Sala-i-Martin (1995), Temple (1999), McMohan and Squire (2003) provide empirical evidence on factors influencing the rates of economic growth across countries.
Barro (1996) attempted to analyze empirically the growth theories and he showed that level of schooling, life expectancy, lower fertility, government consumption, rule of law, Inflation,
Political freedom and the terms of trade are major determinates of economic growth. Interesting to note that contradicting to initial model this study showed that saving and capital accumulation are not the major contributors to economic growth.
Graff (1995) analyzed the role of human capital in explaining economic growth at cross country level. He found that that the accumulation of physical capital, human capital and technological progress are important determinants of economic growth. Jenkins (1995), for the UK economy, confirmed the finding that the investment in human capital instigates to increase productivity.
Similarly, Asteriouand Agiomirgianakis (2001), for the Greece economy explored the relationship between formal education and gross domestic product. They found significant relationship between the two and further depicts that the causality runs through education variables to economic growth.
Jerzmanowski (2016) shows that in case of USA, deregulation boosts growth by accelerating both Total Factor Productivity (TFP) growth and the accumulation of physical capital, without having any impact on human capital and the effects of deregulation are largely independent of states initial level of development; both rich and poor states grow faster after deregulation.
17
Fortunato (2015) showed that the success of democratic institutions is closely related to the educational attainment of the population, which is necessary for long term economic growth.
Iqbal and Zahid (1998) showed that primary education, physical capital and trade openness has positive effects on economic growth of Pakistan and budget deficit and external debt has a negative effect on economic growth of Pakistan. Whereas, Khan (2006) showed that macroeconomic stability, foreign direct investment and financial sector development plays an important role in increasing total factor productivity and education is not an important determinant of TFP in case of Pakistan. This contradicts the finding of Iqbal and Zahid’s finding.
Mahmood (1992) concluded that net foreign private investment has significant effect on the growth of real gross national product, while disbursements of external grants and loans, domestic savings and exports have a statistically insignificant impact on real Gross National Product. Iqbal
& Malik (1993) showed that real interest rate, remittances, exports and expected rate of inflation are important determinants of savings at macro level.
Rahman and Salahuddin (2010) showed that financial sector is important for short and long term economic growth in case of Pakistan. Further, it shows that FDI, human capital and stock market inequality has positive effect on economic growth. Ullah and et.al (2014) shows that real domestic investment, foreign investment, export, remittances and literacy rate have significant effects on GDP growth.
Jangraiz (2012) uses health, education and R&D as a measure of human capital. It showed that human capital has significant effect on institutions and physical capital formation. Azam and
Khattak (2009) showed that domestic investment, FDI, human capital and trade openness has a significant effect on economic growth of Pakistan. Jamil and Ahmed (2010) showed that
18
electricity consumption has a positive effect on GDP and biggest source of electricity consumption is private expenditure on electricity consumption.
All above cities studied highlighted the determinants of growth especially, at country level.
However, the regional dynamics or the determinants of provincial growth of Pakistan have rarely been explored in literature. The regional growth differentials and what explains the development of any particular region is hardly present. Moreover, the development of the regions; according to their population densities or geographical locations are not given much importance. Lastly is growth of regions or provinces in Pakistan converging or diverging from each other is not shown on the above studies. The idea of sustainable or indigenous growth is difficult to achieve from deriving results at country specific level, while ignoring the regional development of that a particular country. So, the less availability of literature on regional performances of Pakistan calls for intense research in this area.
2.3.2 Macroeconomic Modeling in Pakistan Economists constructed macroeconomic models for identifying source of growth at country level. This section reviews the different macro models that were constructed to analyze the performance of Pakistan economy. This will enable us to identify the key determinates of growth and other macro variables at national level.
PIDE (2011) has developed a model for macroeconomic modeling of Pakistan, the model covers
Production block, aggregate demand block, fiscal block, foreign trade block and monetary and price block. Model consists of 21 equations out of which 13 are behavioral and rest are identities.
In production block they included agriculture, manufacturing and services.
Agriculture sector production is assumed to be a function of labor force engaged in agriculture, disbursement of credit to agriculture sector and availability of water. They assume that inclusion 19
of agriculture credit will capture the use of all inputs (fertilizer, use of tractor and pesticides) and they also include another variable road length as a proxy for infrastructure development.
The manufacturing sector includes small -scale and large-scale industries, construction, electricity and gas sub-sectors. Furthermore, export-processing industries are also included in this sector. The production in manufacturing is depend on, capital stock, labour force employed in the manufacturing sector, credit disbursed to manufacturing sector, infrastructure, import of machinery and equipments and use of domestic raw material. The services sector value added is taken as function of aggregate demand in real term. The aggregate output (GDP) is defined as the sum of the value added of agriculture, manufacturing and services sectors.
The aggregate demand for goods and services is the sum of consumption, investment, government expenditures and the trade balance. The consumption sub-sector is disaggregated into private consumption and government consumption. The specification of real private consumption function is based on an optimizing model of life -cycle behavior. The main variables explaining the real private consumption are the real disposable income, real interest rate, real money balances. The real disposable income is equal to GDP minus direct tax revenues, indirect tax revenues, worker’s remittances and credit to private sector divided by consumer price index. The real government consumption depends on the development expenditure relative to GDP.
Aggregate investment is disaggregated into private investment, government investment and increase in stocks. The private investment depends on real income, real interest rate, Ratio of private sector credit to GDP and Government Investment. Government investment is measured by the expenditures on capital construction such as infrastructure and innovations. Government investment serves as fiscal policy instruments and is assumed to be exogenously determined.
20
The fiscal sector constitutes government revenue and government expenditures. In this sector the budget deficit is resulted when government spending exceeds government revenues. The total government revenue is the sum of direct taxes, indirect taxes and non-tax revenue. The model is work in the following way:
(i) Production affects consumption, exports, imports, government revenues, government expenditures, which in turn affects the domestic price level. (ii) Credit to private sector affects private investment which influences output level through the channel of capital stock. (iii) Public investment influences private investment, which in turn affects output level. (iv) Foreign price affect domestic price level, which in turn affect the prices of raw material. (v) Domestic price level is also affected by real and monetary variables. (vi) Real effective exchange rate determines the volume of imports, which in turn, affect private investment. (vii) Private investment affect real output, which effect government revenues and expenditures and hence budget deficits. (viii)
Disequilibrium between aggregate demand and aggregate supply also affects the domestic price level. All variables except interest rate are in logarithms, the data has been taken from different sources; mainly from the Handbook of Pakistan’s Economy 2005, Bulletin of the State Bank of
Pakistan (Various Issues), Pakistan Economic Survey (various issues), Federal Bureau of
Statistics and the International Financial Statistics (IFS) of the International Monetary Fund.
The study has used single-equation based co integration approach (Engle-Granger two -step procedure) to estimate the model. The performance of all estimated equations will then be evaluated using mean absolute percentage error (MAPE) and Theil’s inequality coefficient (U) and slight misspecification will be tolerated if the forecasting ability of the equations is good.
The Engle-Granger two-step co integration procedure is used to derive the long-run and short - run elasticities for the period 1972-2009. The test of significance of each estimated equation
21
seems to validate the model. The estimated long-run parameters are used to perform simulation experiments to determine the model’s ability to track historical data and to assess the behavior of the key macroeconomic variables in response to the changes in selected exogenous variables.
The results indicate that the majority of macroeconomic variables follow an increasing trend over the period of simulation, 2009-2013.
Institute of Public Policy (IPP) (2010), developed a structural simultaneous equations macro econometric model for Pakistan using the Keynesian open macroeconomic framework in a medium-term setting. The model is estimated for the period 1980-81 to 2008-09, using the OLS technique. The model attempts to capture the effect especially of fiscal, monetary, and exchange rate policies. The model contains 18 equations: 11 behavioral equations and 7 identities in six blocks. Including exogenous and policy variables, there are 40 variables in the model.
The real private investment depends upon the real income, the real rate of interest, which is given exogenously; and the relative domestic price of the imported capital goods multiplied by the nominal exchange rate, defined as the unit value index of capital imports divided by the domestic price level; Public investment, is taken as another explanatory variable to see whether it crowds- in or crowds-out private investment.
The expenditure on exports of goods and services is depend on real world income and world income and the relative competitiveness of the Pakistani goods in the international market is measured as the domestic unit value index of exports divided by the domestic price level.
The expenditure on imports of goods and services depend on real income, real interest rate and relative domestic price of imported goods.
The domestic price level is a function of the ratio of the money supply and GDP, the domestic unit value index of imports multiplied by the nominal exchange rate.
22
Government revenues depend on income level of the country and value of imports and total public expenditures is sum of government consumption expenditures and public investment.
Monetary policy has two tools, namely, changes in money supply and the interest rate. In this model, the money supply is modeled as behaviorally determined while the interest rate is taken as exogenously set.
The change in money supply is function of the fiscal deficit, the real interest rate and the real income level. The current account (as percentage of GDP) is measured through an identity which states that the current account (as percentage of GDP) is equal to ratio of the sum of the net exports and net factor income from abroad to the nominal GDP level in the economy.
The employment demand equation is dependent on the real income and the real wage rate. The level of poverty depends on per capita income and relative price of food inflation.
Ex-post simulation of the model produces satisfactory results. Forecasting with the model over the next three years reveals the potential tradeoff between inflation and growth. It turns out that fiscal policy involving a jump in public investment produces better results than an expansionary monetary policy or adjustment in the exchange rate.
2.4 Regional Growth Theories Capello (2011) argues that space influences the way an economic system works. It is a source of economic advantages (or disadvantages) such as high (or low) endowments of production factors. It also generates geographical advantages, like the easy (or difficult) accessibility of an area, and a high (or low) endowment of raw materials. Two large group of theories have been developed i) location theory, it deals with the economic mechanisms that distribute activities by location. Location theory involves investigation into the location choices of firms and households, it uses the concepts of externalities and agglomeration economies to highlight the
23
issues of disparities and distribution of benefits. ii) Regional growth (and development) theory, it focuses on spatial aspects of economic growth and the territorial distribution of income.
Haris (2008) the growth theories are a tool to analyze growth determinants from the supply-side took a homogenous Cobb-Douglas aggregate production function with constant or diminishing returns-to-scale, and differentiate with respect to time. This ignores the regional disparities that may arise because of difference in technology with respect to region, and growth rate of factor of production may also vary from region to region. The assumption regarding constant growth of labor may be true at country level but at intra country level the two areas may have different growth rate of population and different composition of population. Many studies showed that the population growth is lower than the population growth at rural part. Similarly, showed that the saving rate also varies with the regional. The growth theories discussed in previous section also made assumptions that there is perfect competition in all markets so that factors of production are paid equal to their marginal products; price are fixed for the firms, factor of production are completely mobile throughout country and this factor mobility leads to factor price equalization in all parts of the country. These models further assume that there is instantaneous distribution of new knowledge or technology shock throughout the economy, so that all agents have access to the new technology. This implies that distance does not play any role in the model.
The assumption of capital and labour mobility in the neoclassical models showed that there can be no systematic long-run differences in the growth rate of factors across regions. Thus, it is likely that differences in total factor productivity levels are the most important source of any long-run growth differences. If technological change is at least partly determined by an endogenous process as in Romer (1990), and the initial level of knowledge vary from region to region this implies that the ability of a particular region to adopt the technological advancement
24
is different from other which leads to systematic difference in the growth of the both regions.
Such models have been developed in international trade and known as technological-gap models.
The prediction from these models is that in the long-run there should not be disparities between regional growth rates. However; lagging regions have a lot of catching up to do and the incentive to invest may vary between regions. Further, these models show that the regions with high catch- up effect observed higher level of growth in total factor productivity. Technology-gap’ models suggest, firstly, that differences in TFP are likely to be the main driver of persistent regional growth disparities; secondly, that leading regions have higher TFP because of their greater stock of knowledge and human capital; and, thirdly, that much of the technological development that occurs in lagging regions is through diffusion of existing technology rather than the development of new products and processes. Hence, the following conclusion is generally explored by
Armstrong and Taylor (2000) The new trade theory Krugman (1980); Krugman and Venables
(1990) and new economic geography models Krugman (1991) Krugman and Venables (1995);
Baldwin et. al (2003) are developed to answer the geographic clustering of industries. New economic geography models are concentrates on labour mobility sees additional workers migrating to those regions where clusters have formed, in order to benefit from strong home- market effects. Zhao (2007) shows that in the long run economic growth of a country is not only because of by more factors of production and improvement in technology but by also depends on the expansion in spatial structure. Spatial economic structure means the existence of externalities in the growth process. Growth is a process of spatial accumulation and organization: characterized by a geographically distributed and deepened structure of production.
The one of important questions is which regions are likely to grow fastest if factors are fully mobile? In the classical model, areas with a high capital to labour ratio will have high wages and
25
low yield on investment. This means that labour and capital are predicted to move in opposite directions. Low wage regions attract capital but loose labour (capital to labour ratio expanding) and high wage regions attract labour but loose capital (capital to labour ratio declining). This implies that output per worker (productivity) is predicted to grow faster in low wage regions than high wage regions. The other variable is technical progress. The expectation is that this would be highly mobile between regions. It is anticipated that regions with low technology would gain rapid productivity improvements by exploiting the technical gap between themselves and the high-tech areas. Technical knowledge will increase over time and its rate of growth depends on the number of workers in the knowledge industry & the existing stock of knowledge. The neo- classical model predicts that over the long-run disparities in per capita income will disappear, because capital will flow from high to low wage regions and labour in the opposite direction until such time as the returns are equal and there is no incentive to move. Further, poor regions will benefit from technological catch-up4.
Figure 2.1: Schematic Diagram for Regional Growth determinants.
Source: Armstrong and Taylor (2000)
4This paragraph has been taken from RELOCE Lecture 3a Regional Growth - the Neo-classical perspective
23/02/10. 26
Armstrong and Taylor (2000) discussed the regional output growth as a result of three important variables. In above diagram three block (I) Growth of capital stock (II) Growth of labor force
(III) Technical progress is discussed by authors that explain the output growth of any particular region. Whenever, there is an increase in the rate of return of the capital stock of a region relative to other, it will eventually lead to net inflows of capital to that region thus, increasing the capital stock of that particular region. The other channel of the growth in capital stock is increase in saving rate of the region, which will increase the investment of the residents and will eventually increase the capital growth of the region.
While, discussing the labour force growth block the authors stated that the birth or death rate of any region will affect its population growth thus, eventually effecting the growth of labour force of that region. The other channel is if regional wages of any particular region is higher than the other regions it will result in; inmigration of population from other regions thus, resultantly increase in the labour force growth of that particular region.
While, discussing the technical progress the authors reported the investment in research and development and further the increase in literacy rate of a region will increase its technical progress. The other channel is, if a region receives technical knowledge inflow from other regions it will increase its technical progress.
So, finally authors found that all the three blocks such as, growth in capital stock, growth of labour force and technical progress will eventually increase the growth in output of the region.
27
2.5 The level of Regional Disparities in Pakistan This section reviews various studies specific to Pakistan those analyze the level and source of regional disparities in Pakistan.
Hussain (1993) discussed that during 1960s the disparities between East and West Pakistan was very high which led to the separation of East Pakistan as Bangladesh in 1971, so in 1970s and
1980s, the issue of regional disparity has gained a high priority for the government. Over the period of the time there has been a lot changes in the income of provinces but at the same time the inequality has also been increased both in interprovincial and intraprovincial further this inequality has led to the increase in poverty within regions and across regions. Most of the studies found that inequalities are correlated with growth, as whenever the growth takes place it benefits more to the regions that are more developed then to the regions that are less developed.
The author quoted different studies that highlights the fact that regions which has high infrastructure development have attained high per capita income over time as compared others.
In 1959-60 Karachi accounted for 39 percent of the value-added industry followed by Lahore and Faisalabad, hence the total value-added industry in these three districts accounts for 60 percent. But with passage of time when the peripheries of these districts developed the industrial activities also expended to these peripheries as Karachi still accounts for 35 percent of value added and the central Punjab that includes Sheikhupura and northern Punjab that includes
Jhelum accounts for 19 percent of the industries whereas in Sindh only growth has been taken place in Dadu only.
Further the author discussed the incidence of poverty in provinces of Pakistan and reported that incidence of poverty is highest in Punjab and lowest KPK, as 31 percent of rural population in
Punjab live below poverty line, 27 percent in Balochistan, 18 percent in Sindh and 18 percent in 28
KPK, whereas in urban sector still Punjab has the highest poverty of 25 percent followed by 23 percent Balochistan, 14 percent KPK and 10 percent in Sindh, so in urban sector Sindh has the lowest poverty incidence among provinces. So presently government has to focus on equitable growth and focus on planning that allocate resources in an equitable manner in agriculture, industry and irrigation, so regional dimensions has to be taken care of while planning for regional equitable growth. While making regional planning the linkage of regions especially the less developed regions are to be taken into account for any Fiscal or Monetary measures that are to be adopted, further the infrastructural development of the regions is needed and lastly any policy like tax incentives should take dynamics of poverty and inequality into account before implementing the particular policy.
Jamal (2015) estimated the spatial disparities in socio economic development of Pakistan and found that in urban areas, the per capita income of urban Sindh was highest and urban
Balochistan was lowest among provinces, but with rural per capita income the case is little different as the rural per capita income of KPK and Balochistan was slightly higher than Sindh and further the overall annual per capita income estimated by author was Rs. 45000, in urban sector it was Rs. 55000 and rural sector it was Rs. 39000. The inequality calculated by author shows that in KPK and Balochistan the magnitude of the inequality is relatively low as compared to other large provinces. The highest magnitude of the inequality is in urban Sindh while the lowest is in rural Balochistan. The multidimensional disparities between provinces show that the overall Gini coefficient is 0.57, which is quite high among districts of Pakistan. Whereas the Gini coefficient of Balochistan is 0.63 and for Punjab it is 0.35 so almost half of the Balochistan, but the Gini coefficient of KPK and Sindh is almost equal showing that in rural and urban sector
Sindh there is high degree of inequality. Further the Gini coefficient of housing and health shows 29
that there is high disparity between districts due to housing and health as the Gini coefficient of housing is 0.76 and for health it is 0.67 among regions. The study concluded that as in development ranking Karachi also dominated for development but after adjusting the inequalities at the district level it was found that the rank of Karachi has been decreased, so government has to take into account the regional disparities before forming any policy and especially the districts dynamics of the regions are needed to be taken care of.
Arif, G. M and Fraooq, S (2011) pointed out that historically Pakistan has experienced high fluctuations in poverty as for 2000 to 2006, country experienced a decline in poverty but later due to low economic growth and high inflation the poverty has increased in 2010 as compared to
2001. Further the authors reported that at provincial level there is high variation in poverty estimates as rural Sindh has shown a decline of poverty from 48.3 percent in 2000-01 to 28.9 percent in 2004-05, so the ranking of rural Sindh has been reversed across provinces as in 2000-
01 rural Sindh was the poorest region across provinces but in 2004-05 it is not. The major reason discussed by the authors is the exceptional performance of agriculture sector of Sindh in 2004-
05. The decline in rural poverty of Punjab was marginal and for KPK and Balochistan it was also quite small. Further the urban poverty of Sindh has also declined drastically from 20.7 percent in
2000-01 to 13.8 percent in 2004-05, whereas the poverty of other provinces declined but at as low rate. So the major decline in national poverty estimates for the time period of 2004-05 was due to poverty reduction in Sindh and Balochistan. The inequalities reported by authors show that between the period of 2000-01 to 2004-05 the overall consumption inequalities has been increased more in urban areas than in rural areas the rich and poor gap has increased. It has also been observed that during the period of 2000-01 to 2004-05 when the poverty declined the inequality has increased which shows that the overall growth has benefited rich more to poor. 30
However, the non-income inequality groups both in rural and urban shows that from the period
2000-01 to 2004-05 the inequality has been decreased.
For the East and Southeast Asian countries, the authors found that since there is as reduction in poverty but inequality remained the issue for all countries except Malaysia. In the first half of
2000 decade between 2000 to 2005 the overall poverty decreased but the consumption and income inequality has been increased in Pakistan, mainly this rise of inequality depict in the rise in inequality of Urban areas that show the uneven growth during the first half of 2000 decade.
The similar situation observed in 1992-93 to 1998-99 when the poverty remained same especially in urban areas but inequality had increased and rural inequality increased but poverty decreased so from 1993 to 2001 there exist a negative relationship between poverty and inequality.
While analyzing the relationship of growth, poverty and inequality, the authors argued that the relationship between these variables is complex and multidimensional and this also holds for
Pakistan. In 1960s the growth rate was high as 6.8 percent per annum but both poverty and income inequality increased in the period and mainly the growth of these time periods was due to foreign aid and exchange rates were overvalued. During 1970s the growth declined to 4.8 percent per annum but urban poverty and inequality declined in the same period. The 1980s experienced averaged growth rate of 6.5 percent, in 1980s the growth of agriculture was slow as averaged
3.36 percent but industry grew at 8 percent and high growth of services and foreign assistance the rural and urban poverty declined from 30.36 percent in 1980 to 22.1 percent in 1988 but inequality increased mildly, finally in the period of 1990s due to instability in govern net both poverty and inequality rose but the relationship of poverty and inequality from 1990 to 1999 was
31
opposite. During the first six years of 2000 decade as government had launched PRSP the growth was about 6.2 percent in first five years i.e 2000 to 2005 during this period the high growth and reduction in poverty has been achieved but inequality remained there. In the table below shown by the authors that for physical and social infrastructure there is high degree of inequality between provinces. KPK and Balochistan has poor availability of infrastructure even in Punjab and Sindh the rural sector of Sindh and southern Punjab has high inequality in physical and social infrastructure with northern and central Punjab, so regions with low infrastructure availability will grow less as compared to regions with good infrastructure availability thus growth will be unequally distributed among regions. So, the equal distribution of resources are to be ensured to get sustained growth as the idea of inclusive growth states that equal access to opportunities should be provided to all segments of the society.
Nazir, M and Yasin, H. M (2011) analysed the economic growth and regional convergence in
Pakistan, the authors found that in Pakistan the regional disparities are not only due to difference in culture or demography but much of the regional disparities lies in the diversities of social and economic development among regions. During the last half century most of the development focused on large cities like Karachi so it resulted in high migration to big cities for employment thus created disparities among provinces and increased the problems of poverty and inequality.
So it is important to analyze the convergence of growth process between regions thus to minimize the economic problems. The authors used the data set from 1979 to 2005 in a panel form to find out the absolute and conditional convergence among regions. The variables used are the per capital household income, literacy and saving rates, enrollment rates, population growth, infant mortality rate, dependence ratio and cured birth rates. The time period is divided into six
32
intervals to test which span allows the regional convergence the intervals are 1979-84, 1984-
1988,1988-1993,1993-1997,1997-2001,2001-2005.
Table 2.1: Pakistan: Infrastructure by Provinces Physical Infrastructure Social infrastructure
Distance to metal Soling Piped Education Health Province road < 1 Electricity street Drain water ins ins Punjab 80 47 66 58 9 34 30.5 Sindh 67 10 30 23 7 37.5 27.3 KPK 38 34 29 19 20 33.3 24.5 Balochistan 20 12 8 7 9 22.3 11.3 Source: Arif, G. M and Fraooq, S (2011)
The authors found that during the period of 1979-1988 the results of per capita income showed convergences among provinces at a rate of 7.4 percent per annum and the results were significant, for the period of 1988-1998 the sign of the coefficient is negative but insignificant, for 1998-2005 the results were opposite as the signs were positive and insignificant thus implies a weak signal of divergence this is due to the fact that during this time period the poverty and inequality has been increased. Only the period of 1979-1988 showed convergence is due to the fact that economic performance was better during the period as growth rate was high and inflation was low and high increase in workers’ remittances increased the living standards of the people.
The analysis was then extended to rural and urban disparities it was found by the authors that in overall time period the results are significant at 10 percent and sign is also negative, but in interval the rural income was insignificant showing no signs of convergence, this may be due to the fluctuations in agriculture productivity reported by authors. Similarly, the urban income shows that in overall time period from 1979-2005 the signs are significant ad negative but for the
33
intervals the coefficient of urban income turned to be insignificant. So, in the entire time period the implied speed of convergence for rural sector is 2 percent and urban income convergence is
2.6 percent, so the authors found that rural and urban economies are not like to converge as they follow their independent growth paths, the reason is as per capita growth may not be the only variable that explains the complex growth process of convergence. In this analysis the coefficient of savings turned out to be low as 0.02 percent but the coefficient of population is high as 0.32 percent and is negative showing increase in population will decrease the growth of real income.
The speed convergence from the lagged income is 8 percent and most of the regions are near to steady state and lastly the difference in the regional re due to the difference in factors that determine steady state.
The authors finally concluded that for the entire time period of 1979-2005; except for 1979-88 there is no sign of absolute convergence, not the signs of regional convergence is observed and further the signs of divergences are observed during 1998-2005. So absolute, convergence shows increase in disparities among region; whereas conditional convergence shows that in rural urban sector there are signs of regional convergence.
The growth of provinces and their regional differences was discussed by Pasha (2015) the author has calculated annual growth rates of the provinces in different time periods as shown in below table 2.2.
Table 2.2: Annual Growth rates of the provincial Economies, in different time periods (%) Province 1999-2000 to 2008-09 2013-14 and 1999-2000 to 2014- 2007-08 to 2012- 2014-15 15 13 Punjab 4.8 2.9 4.4 4.1 Sindh 6.1 1.9 3.4 4.3 Khyber-Pakhtunkhwa 5.4 4.9 5.1 5.2 Balochistan 3 1.7 2.7 2.5 Pakistan 5.6 2.8 4.1 4.4 Source: Pasha (2015), IPR report.
34
The author argued that during Musharraf period the economic growth of Sindh was highest as around 6 percent which was more than the growth of national economy, but since afterward in the preceding years the growth rate of Sindh has been declined significantly as it was 2 percent from 2008-09 to 2012-13 but from last two years it has shown some improvement, whereas the growth of Punjab during Musharraf period was below than the national economy growth but since then it has shown improvement. Surprisingly the KPK has maintained its growth rate around 5 percent in last fifteen years and Balochistan turned out to be struggling economy as its growth rate did not exceed 3 [percent in last fifteen years, as result the people of Balochistan is now suffering from great deprivation and exclusion in today time period.
Pasha (2015) extended the regional disparity analysis to the sectoral, level in each province and found that agriculture has been the dominated sector in Punjab, with its share around 24 percent in provincial economy and this is more than 20 percent of the overall economy, but due to loss in dynamism, this sector grew at the rate of 4.5 in the decade of 1990s. The industrial sector has shown a rapid growth in Punjab especially in Musharraf era, but since afterward due sever power shortages this sector has affected badly, the services sector has also grown at higher rate but due to fall in commodity producing sectors this sector has also lacked dynamism. The economy of
Sindh has shown dynamism in Musharraf period as the growth rate of industry was 10 percent and currently the industry of Sindh has its highest share in industrial growth of Provincial Gross domestic product (PGDP), but after 2008 due to law and order situation the economic momentum has decreased from 6 percent to 2 percent after 2008. The agriculture sector of Sindh has also not performed well as its average annual growth was less than 2 percent and the growth of services sector has also fallen to 3 percent due to lack of security problems. The economy of
KPK has grown sharply mainly due to high inflows of remittances as 20 percent of household 35
income in KPK is from remittances which is higher than Punjab as it is 10 percent in Punjab and less than 3 percent in Sindh and Balochistan, so due to high inflow of remittances KPK has grown at rate of 15 percent in recent years. The services sector of the province has also shown high growth and mainly it is remittances led services growth as services account for 60 percent of the provincial economy. The growth of Balochistan has remained slow mainly due to insurgencies and the military actions taken in the provinces the investment and per capita income has decreased sharply, but due to the high growth in fruits and vegetables and the 7th NFC award the economy of Balochistan has shown some growth.
The author finally concluded that during Musharraf period the inter provincial inequality has increased but during the democratic regimes the inequality has been decreased and remained same in last fifteen years, but the intra provincial inequality results shows that in Punjab the intra provincial inequality is highest as compared to other provinces after Punjab comes Sindh as intra provincial inequality is also higher in Sindh. Whereas in KPK the inequality is less among household and further in Balochistan the inequality is also less.
The author computed a table of sectoral share of each province indifferent time periods which clearly shows the dominance of Punjab in agriculture, the highest contribution of industry in
Sindh and further the services share of Punjab and Sindh compared to other two provinces. The table clearly shows the sectoral contribution of each sector in all provinces at different time periods.
36
Table 2.3: Estimated Provincial Gross Domestic Product (PGDP) of provinces by sectors 1999-2000 to 2014-15
Agriculture (% Share) Industry (% Share) Services (% Share)
Province 1999- 2007- 2013- 2014- 1999- 2007- 2013- 2014- 1999- 2007- 2013- 2014-15 2000 08 14 15 2000 08 14 15 2000 08 14
Punjab 31.61 25.37 24.45 24.07 13.3 15.46 15.48 15.35 55.09 59.17 60.06 60.58
Sindh 22.48 16.23 16.16 16.06 24.48 32.16 29.3 29.32 53.04 51.15 54.54 54.62
KP 25.32 18.57 17.25 16.87 17.54 22.87 23.24 22.89 57.14 58.57 59.52 60.24
Balochistan 27.83 25.78 28.86 28.97 26.29 26.92 26.31 26.61 45.88 47.28 44.81 44.4
Source: Pasha 2015. IPR Report.
2.6 Role of Human and Physical Capital in Economic Development in Pakistan The province wise growth of human capital accumulation is calculated by Aftab.Z and Sabir.M
(2006) the authors calculated the human capital index in three-time periods as from 1982-83 to
1989-90, 1990-91 to 1999-2000 and 2001-2004. The study found that during the period of 1982-
83 to 1989-90 the human capital index has grown at a rate of 3.4 percent annually with higher growth in Punjab as 3.3 percent and Sindh with 4.8 percent. In the second period from 1991-
2000 the overall growth has been decreased mainly due to slow down of growth in Sind as 1.5 percent per annum, but post 2000 the human capital growth of Sindh has increased to 6.3 percent per annum thus the overall growth of HCI was 4 percent for the period 2000-2004. In overall human capital index Sindh contributes 24 percent while the contribution of Punjab is 63 percent but the performance of Punjab is relatively stable between 3.2 percent to 3.3 percent per annum and is not subject to drastic changes. The other two provinces NWFP and Balochistan have small share in overall HCI as both account for 13 percent in overall HCI further the growth rate shows that NWFP grew at 2.6 percent over 22 years and Balochistan grew at 3.4 percent over 22 years
37
The authors decomposed the analysis of human capital growth at sectoral level and found that in services sector during 1982-83 to 1989-90 Sindh had the highest growth of human capital in services sector as annual compounded growth rate of Sindh during the period was 9.8 percent followed by Punjab as 6 percent, NWFP by 3.8 percent and Balochistan by 0.9 percent.
However, in the decade of 1990s the human capital growth rates among provinces changed as during 1990s the highest growth rate of human capital in services was observed in Balochistan, followed by NWFP and Punjab but post 1999 Sindh regained its momentum and the growth in
2000-2004 was highest in Balochistan as 11.1 percent followed by Sindh and NWFP as 6.3 percent and than Punjab as 3.8 percent. A Similar, pattern has been observed in human capital index of manufacturing across provinces as from the time period 1982-83 to 1989-90 the human capital growth in manufacturing of Sindh was highest at 6.7 percent and for Punjab it was 1.5 percent and NWFP and Balochistan has shown a deaccumulation of capital for the same period, but in the second period from 1990-91 to 1999-00 NWFP has shown little improvement with average annual growth of 5 percent while Punjab grew at 3 percent, Sindh at 0.3 percent and
Balochistan at minus 3 percent. While during the period of 2000-2004 average growth of Sindh was 10.9 percent, Punjab grew at 4.9 percent the rate of growth of NWFP at 3.2 percent and the growth of Balochistan was 20.6 percent. In agriculture mainly human capital accumulation growth has been observed in Punjab which is an obvious fact, whereas NWFP and Balochistan being the small contributor to overall agriculture output has shown an increase in human capital accumulation in agriculture sector but Sindh has shown a decline for the period from 1983-2000 and a slight improvement after 2000. The authors concluded that in 1980s Punjab and Sindh were the main contributors for the human capital accumulation especially in manufacturing and services industries, but in 1990s the growth of human capital in both provinces has been
38
decreased whereas NWFP and Balochistangrew at fast pace in this decade, but since 2000 Sindh has regained its growth momentum and has shown significant improvement.
Javed. M and et.al (2011) the authors attempted to calculate the impact of physical and social infrastructure on the total factor productivity of Punjab using multivariate Cobb-Douglous production function for the time of 1970-2005. The study found that the investment of public sector on physical social infrastructure have a great impact on the TFP of the province as investment on social infrastructure if increases by 1 percent it will increase TFP by 0.13 percent and investment on physical infrastructure such as roads. Irrigation and electrification if increased by one percent will increase the TFP by 0.24 percent and finally an increase in the investment on agriculture by 1 percent will increase the TFP by 0.21 percent. So, the importance on investment on agriculture and public investment on agriculture cannot be denied as the authors concluded that growth in agriculture will reduce poverty by four times more than the non-agriculture sectors. The increase in investment in agriculture will increase the TFP of agriculture and reduces the poverty and finally improves the socio-economic conditions of the large segment of the population. Physical infrastructure plays vital role in enchaining productivity whereas the role of social infrastructure can also not be denied as education will increase the efficiency of farming of the rural community.
Sasaki. K and et.al (2003) analyzed the role of public and private capital on regional disparities in Pakistan’s economy. The authors divided the regions in two categories as relatively advanced regions as Region 1 and relatively backward regions as Region 2. The study found that in per capita income a significant disparity is present among all provinces of Pakistan for the estimated time period of 1972 to 1997. Further the regional disparity in labor productivity is also increasing since 1982. The analysis of advanced and backward regions showed that almost 90 percent of the
39
private capital is being allocated to relatively advanced regions and further the economic activities are also concentrated to these regions but public capital is concentrated in both advanced and backward regions. The public capital is used more efficiently in advanced regions than in backward regions. Since the private capital is less in Region 2 and further the efficiency of public capital is also less in Regions 2 the authors concluded that the less availability of capital in Region 2 has pushed the labor of Region 2 to use capital efficiently hence the Region 2 possess the quality or having the potential to increase income and production if share capital is increased in Region 2. The technology gap analysis also revealed the same fact between two regions the use of technology and the availability of technology is subject to great disparity between regions. If the allocation of private and public capital stock is to be changed it will have large impacts on the efficiency of national economy and regional disparities.
Afridi (2016) estimated the impacts of human and physical capital on economic growth of
Pakistan. The author used the data from 1972 to 2013 of Gross domestic product, primary enrollments, infant mortality rates, birth rates and physical capital as fixed capital formation. The author used ARDL and VECM models, the results of ARDL technique shows that an increase in birth rates by 1 percent will increase the GDP by 6.14 percent, similarly a 1 percent increase in infant mortality rate will decrease the GDP by 27 percent and lastly a 1 percent increase in physical capital will increase the GDP by 7 percent. The error correction model showed that the short run adjustment of any disequilibrium in these variables will be adjustedat a rate of 41.93 percent in one-time period. Thus finally the authors concluded that there is a positive relationship of physical and human capital on economic growth both in short run and long run.
Khan & Rehmann (2012) has calculated human capital level by province in Pakistan and found significant differences in the state of human capital at inter and intra provincial level. They also
40
found that human capital accumulation rate also varies from province to province. Pasha and
Naeem (1999) finds that the low level of social indicators in the country is a consequence of poor initial conditions or has there been deterioration due to relatively low rate of improvement over time? The study concludes that Pakistan is a case of a country which not only started with low level of human endowment but the situation has been exacerbated by the low level of improvement in it over time.
Farooq. F. and et.al (2012) studied the role of human capital on economic growth of Pakistan.
The authors applied OLS technique to the data from 1972 to 2010. The study found that there exists a relationship between economic growth and human capital formation. Further the results showed that education enrollment, Gini coefficient, gross domestic fixed capital formation have positive and significant impacts on economic growth and head count ratio and investment growth have the negative impacts on economic growth. The authors are of the view that due to lack of investment in Pakistan the investment growth turned out to be negative and further mostly investment expenditures in Pakistan by public sector is carried out on non productive sources, so due to the above discussion investment may have negative impacts on economic growth.
However, the main idea underlying the study was that human capital has a significant role in determining national income and this is supported by the empirical evidences.
Briscoe. G and Wilson. R (2004) analyzed the impact of human capital on economic growth, the authors disused that the European Union (EU) statistics shows that in 15 member states the increase in education and training increase the economic growth of these states. The authors discussed that various studies reported the social rate of return from 6 percent to 12 percent, further the vocational training programs too have strong impacts on labour productivity. The performance of any organization depends upon the quality of the human resources. There are
41
different ways of enhancing human productivity but most importantly are the role of leadership and enhancement of the skills of labour force. Many of the econometric models have been developed to capture the investment in human capital and finally analyze the role of human capital in overall economic growth, broadly the evidence suggests that a one percent increase in school enrollment will increase the per capita GDP by one percent to 3 percent and further the additional year in secondary schooling will further increase the per capita GDP. The results in literature vary with country to country but the positive impacts of education on per capita GDP are found in almost all models. The intangible investment in shape of research and development that improves the skills and productivity of the labour is also been observed in many studies across countries. The authors also discussed that technology in the shape of externality and spillover effects can increase the labour productivity of the firm and thus finally increase the overall economic performance of the country.
2.7 Role of Human and Capital infrastructure in economic development The impacts of public investment on infrastructure is analyzed by Ahmed.S and et.al (2012) the authors applied dynamic CGE modeling to analyzed the macro micro impacts of investment of public sector in infrastructure on economic growth. The study made distinction between households and firms having open access to credit and saving instruments and those who do not have access to these instruments, the public investment on infrastructure in the above study has been taken in two different methods as public investment on infrastructure financed by taxes and secondly financed by foreign borrowings. The results showed that public taxes or borrowing to finance investment on infrastructure have similar results in long run as increase in macroeconomic performance and reduction in poverty. Further, the study found that in a short span of time for one year, the effect of tax, decreases the industrial output as industrial
42
production bears a high amount of taxes, thus in the short period of time financing from taxes will decrease economic growth. On the contrary if public investment is financed by borrowing it will have a kind of Dutch disease as exports will decrease in the first period. The long run growth of real GDP under tax will be 1.01 percent and under borrowing it is 1.29 percent. The household expenditures under tax regime will increase by 0.94 percent and under borrowing they will increase by 1.2 percent. The decrease in consumption and investment expenditures under tax regime in the first period will be compensated by increase in supply of output in the long run.
Lastly the poverty will decrease by 0.31 percent under tax regime and under borrowing the poverty will decrease by 0.4 percent, however the inequality will only improve at a very slower rate under both regimes. Finally, the authors concluded that the public investment on infrastructure either financed by taxes or by borrowing there impacts highly depends on the policies of the government, that where this investment is made either it will improve the physical infrastructure that is needed for the economy or it is spent on the infrastructure that have less impacts on economic growth.
Khan (2005) estimated the role of human capital in economic development across 72 low and middle income developing countries from 1980-2002. The variables used in the study were real per capita GDP, investment as percentage of GDP, initial income, inflation, institutional quality, average years of schooling, gross secondary enrollment, literacy rate and life expectance at birth.
The study found that investment to GDP ratio, initial income, inflation and institutional quality all variables are highly significant and have expected signs across all countries which means that there is a strong relationship of economic growth with sound macroeconomic policies, institutional quality, law and order and absence of corruption. Further the results also confirmed that higher level of educational attainment and better health facilities will also lead to more 43
economic growth. The size of the coefficients show that rise in investment have strong impacts on economic growth further health and education coefficients also shows that there is a sizeable impact of both variables on economic growth across countries. The cross countries the performance of Pakistan is not satisfactory as China, Malaysia, Singapore and South Korea are having annual real per capita growth is between 3.5 percent to 8 percent whereas in Pakistan it is
2 percent from 1980-2002, so Pakistan is clearly behind many countries and is not a tiger. To achieve higher growth, the study suggested that level of investment should be increased and further the quality of institutions should be improved, as investment in Pakistan is around 16 percent to 17 percent of GDP from past few years whereas in fast growing counties it is 25 percent of GDP. So 5 to 6 percentage points investment is required to increase real GDP by 1 percentage point than Pakistan can become closer to Sri Lanka. During 1980-2002 Pakistan has an average of 5.6 on institutional quality, so this has to improve by government in order to achieve high growth rates. Lastly Pakistan has to invest in education and increase the average year of schooling so that along with physical infrastructure human infrastructure will also be improved and the significant growth targets will be achieved.
2.8 Terrorism and economic development Gaibulloevk (2008) estimated the impact of terrorism on 42 Asian countries from 1970-2004 in panel estimation. The author estimated three equations using OLS as in first equation the author introduced a variable of political violence as determinant of economic growth and in second and third equation the author estimated the impact of political violence on reducing investment and increase in government expenditures on security. The study found that transnational terrorism and internal conflicts have negative impact on growth of all countries but external conflicts turned out to be insignificant in the study. Moreover, the author built different models to analyze
44
the dimensions of terrorism and found that in model 1 coefficient of transnational terrorism is significant and is -0.015 it means any additional terrorist attack one million people lowers the
GDP per capita by 1.5 percent in that given year, when political variable is included in model 5 the coefficient of terrorism turned out to be -0.02 which is not a significant difference from the model 1. So it means with and without democracy the results are similar. The impact of terrorism on investment and government spending revealed that the increase in internal conflicts decrease investment by 0.73 percentage points and external conflicts will decrease investment by 0.66 percentage points, but both the coefficients of internal and external conflicts are insignificant in model 5 it means whenever there is a shock the investment cannot be withdrawn immediately in a short span of time thus investment is insensitive to conflicts whereas it captures the impact of conflicts and lower future investment as the size of the coefficients of internal and external conflicts are large. Government usually increase its spending in bad time periods and decrease its spending in good time periods, the results show that if there is an increase in terrorist attacks it will increase government spending as in transnational terrorism one unit increase in terrorist attacks of one million people will increase the government spending by 1.5 percent, external conflicts increase government spending by 1.4 percent and internal conflicts increase government spending by 1 percent, but the results of government spending are robust. So the author finally concluded that the terrorist attacks though effect growth of the economies and at the same time decrease investment and result in increasing government spending.
The war against terrorism and their repercussions are discussed by Rabbi (2012), the author argued that Pakistan has received certain benefits and has also incurred losses due to the war against terror. The major benefits Pakistan received are removal of sanctions that are imposed after nuclear test of 1998 and military coup of 1999, further high amounts of grants and aids are 45
given to Pakistan that helped the country to improve its economic conditions as Bush administration allocated $1billion package for control on borders, refuges assistance and poverty alleviation. Further the US loan to Pakistan of $1 billion was rescheduled and also the loan from other international institutions which was $ 38 billion from World Banks, IMF and ADB was rescheduled. Pakistan also got assistance during earthquake of 2005, flood crises of 2009 and in many development projects, as US had provided $4.75 billion aid to Pakistan from 2001 to 2008.
So Pakistan became one of the major recipient of US foreign aid during 2001. On the contrary
Pakistan has paid a high cost of this war against terror and in spite of the high assistance from
US the economic conditions of Pakistan did not improved significantly as during 2001-2006 inflation increased from 4.4 percent to 7.9 percent, the external debt rose drastically from $32.46 billion in 2003 to $50.14 billion in 2009, this is because most of the US aid directly goes to military accounts and is not spent to improve economic conditions. Due to high spending on war against terror Pakistan was unable allocate much of the resources to education, health, irrigation and road networks thus resulting in the deterioration of these indicators over the period has been observed in Pakistan. The government policies were inconsistent due to which food inflation rose, domestic and foreign investment decreased and further the Ethnic and social violence has deteriorated economic conditions of the country. The author quoted that in 2009 Shoukat Tarin advisor to Prime Minister told to media that Pakistan has yet paid $35 billion as cost for terrorism and it is increasing every day. So due to political instability, militancy and terrorism
Pakistan was unable to improve its economic conditions in spite of the aid from US. The war on terrorism has also put socio cultural repercussions on Pakistan’s economy as in the past allying with US, Pakistan has received the enmity of Russia and India and the second alliance with US resulted in millions of refuges coming to homeland of Pakistan that created social problems like
46
sectarianism, arms, drugs and terrorist extremism that includes suicidal attacks. The author finally concluded that though Pakistan has received benefits from war on terror but the economy has to pay a very high cost for this war, so a careful policy options are needed to be decided by the policymakers in order to bring the economy out of this turmoil.
Michael. S (2007) argued the same repercussions of terrorism on Pakistan’s economy as since after 9/11 Pakistan participated with other countries against terrorism as a front line state. For this the economy of Pakistan has suffered a lot as the internal security of the country weakens, number of terrorist attacks has been increased in Pakistan especially in major cities like Karachi and Federally administrated tribal areas (FATA). So both economic and social costs paid by the country in fighting against terror is very high and further not only recent war against terror but in the past during 1987/88 the end of Afghan war lot of groups turned back to Pakistan which resulted in increase of sectarianism and extremism that affected the society and economy of the country badly.
Khan (2013) analyzed the social, political and economic effects of war on terror, as the author argued that on social aspect during 2009 the incidence of terrorism has increased in Pakistan mostly these attacks were in KPK and FATA as 559 terrorist attacks in FATA killed 644 people and in KPK 1137 terrorist attacks killed 1438 people. During 2010, 459 terrorist attacks in KPK killed 836 people and the data of 2012 showed that total 1577 terrorist attacks killed 2050 people of the country, thus the high attacks have crippled the economy and destroyed the infrastructure and property of the country drastically. The unemployment has also increased from 3.12 million in 2009-10 to 3.40 million in 20010-11 further in KPK the unemployment increased from 0.10 million in 2009-10 to 0.12 million in 2010-11. Militants also attacked educational institutions as
47
in 2011, 79 schools were attacked by militants in KPK and 56 schools in FATA and 401 schools were targeted in Swat valley in 2010-11. The KPK education department reported that militants destroyed 758 schools in the province including 640 in Malakand division. The Pakistan peace institute reported that 2.7 million people in KPK and 3.5 million people in FATA were displaced due to military operations, largest displacement took place in South Waziristan where 428000 people displaced due to military operations, further high amount of people displaced from
Orakzai Agency, Khyber Agency and Malakand division due to military operations. All these people suffer from heavy losses from leaving their homes, job places and assets for which they suffered from physiological trauma.
The political costs reported by author include political instability as the local social. Political and judicial system has been destroyed by militants especially in Malakand division and FATA, as jirga system of FATA has been severally damaged by militants due to repeated attacks. The
North Alliance formed in Kabul had most of their linkages with India thus has created lot of difficulties for Pakistan especially creating political disruptions and challenging for every political solution suggested by Pakistan to Afghan government for war against terrorism
The economic cost incurred by Pakistan are reported by authors, as reduction in foreign investment, output, exports and taxes and the credit risk of the country has been increased. The indirect costs include increase in unemployment, loss of tourism, displacement of people and lastly terrorist attacks on the development projects initiated by government. The total cost on terrorism from 2001-02 to 2009-10 is $67.9 billion. The cost of war on exports has increased from $1.4 billion in 2001-02 to $2.90 billion in 2010-11 and the cost of foreign investment has increased from $0.15 billion in 2001-02 to $2.10 billion in 2010-11. Whereas Pakistan has only
48
received $15 billion in shape of Coalition support fund (CSF) but has incurred $68 billion. So only 14 percent of the losses to Pakistan’s economy had been reimbursed. The FDI as percentage of GDP has decreased from 3.9 percent in 2003 to 0.62 percent in 2011. So the authors concluded that in fighting against terrorism Pakistan has born high losses in social, political and economic fronts whereas the compensation was far less than the losses and finally the burden of war on terror has crippled the economy and sound measures with availability of resources are required to bring the country out of this turmoil.
Khan. Z and Farooq. M (2014) analyzed impact of terrorism on key economic indicators of the country. The authors found that the growth rate decreased from 3.5 percent in 1997-98 to 2 percent in 2001-02, mainly this low growth was due to direct involvement of Pakistan in war against terror, global recession after 9/11 and lastly increase in terrorist activities in the country, finally high prices of oil during these years is also responsible for the low growth. The increase of terrorism in the country have also affected the flow of FDI, since 1980s Pakistan is attracting
FDI by different incentives that includes tax holidays, infrastructural availability less duties on imports of machinery and other fiscal and monetary incentives. Due to these measures the flow of FDI in 1995-96 crossed $1 billion, but later due to 9/11 attacks, global financial crises and increase in terrorism the FDI flows decreased to $322 million in 2000-01. The other important impact of terrorism is on fiscal and budgetary resources, to counter terrorism Pakistan has increased its budgetary resources for military and law and order agencies as before war on terror the law and order expense in 2001 was 109 percent which increased to 45.1 percent in 2005-06.
KPK being the most effected provinces has spent huge amount to counter terrorism as in 2007-
08 Rs. 6.5 billion is spent to counter terrorism in KPK and Rs. 18 billion was spent in 2009-10.
49
This situation has worsened the position of government receipts and expenditures, the deficit after war increased from Rs. 164,900 million in 2000-01 to Rs. 928,497 million in 2009-10.
The authors further discussed that the tourism industry has also been effected by terrorism as
Pakistan has to face a reduction of 10.3 percent tourist arrival to Pakistan in 2001, since then there is a slight improvement but again in 2007 the visitors to Pakistan reduced by 6.4 percent mainly this reduction was from Europe and USA. The other important impact of war on terror is the increase in inflation especially the hike of food prices that has worsened the poverty problems of Pakistan, during the four and half years of present government the minimum cost of food basket has increased drastically by 79 percent, overall inflation has increased from 10.3 percent in 2007-08 to 21 percent in 2008-09. During 2003 the civilian deaths were 74 percent and of militants it was 13.2 percent, further in 2011 and 2012 civilian fatalities increased to 43.4 percent and 47.8 percent respectively. The total displaced people in 2010 were around 2 million, these IDPs have placed an additional burden on the economy and further the issues of law and order arose especially in KPK. The author finally concluded that terrorism has severally damaged the economy of Pakistan as growth rates has been declined, investment has decreased, budgetary deficit has increased due to high spending on war against terror and lastly the displaced people have also been a burden to the economy, so Pakistan needs good support from internal and external resources to counter the negative effects of terrorism on the economy.
2.9 Employment, Output, Economic growth and Productivity The total factor productivity in crop and livestock in Pakistan is estimated by Mahboob.H and
Ellahi. M. (2013) the empirical results of average annual growth rate of total factor productivity are calculated by Tornqvist indices and the study is divided into two time periods 1980-81 to
1994-95 and 1995-96 to 2009-10. The authors found that the total productivity of Balochistan is 50
four times higher than all other provinces in entire time period the total factor productivity of
Punjab, Sindh and KPK remained same as one percent per annum. The combined results of crop and livestock total factor productivity in KPK is 1.27 percent per annum which is higher than
Punjab and Sindh showing that KPK is more efficient in both crop and livestock productivity as compared to Punjab and Sindh, whereas the total factor productivity of Balochistan is 4.01 percent per annum which is even higher then KPK and other two provinces showing the higher efficiency of Balochistan over other provinces mainly this is due to the additional canal water provided to Balochistan in beginning of 1990s that gave boost to crop output. Within two time periods the TFP was higher in 1980-81 to 1994-95 in all provinces with Balochistan as exceptionally higher 7.25 percent per annum. In 1995-96 to 2009-10 the TFP decrease as -2.67 percent average annually and further during the same time period KPK showed a decrease in
TFP by -0.25 percent per annum. The combined total factor productivity of Punjab has decreased from 0.87 percent to 0.41 percent in these two-time periods but for Sindh the TFP has been increased from 0.64 percent to 0.94 percent annually. Most likely the reduction in the TFP of
Punjab, KPK and Balochistan in second time period is due to reduction in production of beef and mutton. The authors finally concluded that irrigation turend out t be the strong variable for enhancing productivity as in Punjab effective irrigation, road infrastructure and nutrition befitted
Punjab more as compared to other provinces, further the literacy and medical facilities benefited spatially to the TFP across provinces. Regions like Punjab where well-developed activities in crops and livestock are undertaken shows improvement in TFP of the province. The other provinces where the research and extension of crops have not yet been introduced as in Punjab, needs to be introduced so that TFP of those provinces will increase further and lastly a great deal of attention is needed to pay on the livestock industry as it is a rising industry across provinces.
51
Ahsan. H. and Ahmed. A. (2011) discussed the impacts of services sector on Pakistan’s economy. The authors argued that the structure of the Pakistan’s economy has changed from few decades as the share of agriculture has been decreased from 43.6 percent of GDP in 1960-61 to
21.5 percent of GDP in 2009-10, the share of industry has been increased from 15.6 of GDP percent in 1960-61 to 25.2 percent of GDP in 2009-10 and lastly the share of services sector has been increased sharply from 39 percent of GDP in 1960-61 to 53.3 percent of GDP in 2009-10.
The economy of Pakistan has been transforming from agriculture to services sector and the rapid growth of services sector is due to the development of financial institutions in Pakistan in recent past. The growth rate of finance and insurance from 1975 to 2010 is 6.8 percent. Further employment is also increasing in services sector and people are moving from agriculture to services sector. The revenue contribution of services sector is 26 percent which is quite high from 1 percent of agriculture sectors. The share of services in exports has been to 20.77 percent in 2009-10 and in imports it was 17.99 percent in 2009-10. The inflows of FDI in services sector has been decreased from $1951.1 million in 2008-09 to $767 million in 2009-10, mainly this decline is due to sever energy crisis in the economy and further deterioration in law and order situation in the country and global financial crisis. The data on FDI shows that the growth of net inflows of FDI during 2002 to 2010 was 18.3 percent and from 2002 to 2010 the growth of FDI in services sector was 28 percent higher than 14.3 percent in goods sector. The share of service sector in total employment has also been increased from 27 percent in 1973 to 34.5 percent in
2009. While analyzing the wage rate the authors found that in services sector 35 percent of the employees are getting Rs.4000 and in goods sector 32 percent of the employees are receiving Rs.
4000. So the authors finally concluded that services sector is a rapid growing sector of the
52
economy as the employment generation opportunities and growth in services sector is faster then goods sector.
The employment and output are discussed by Majid. N. (2000) the author argued that from the time period of 1983 to 1995 the self-employed labor is 44 percent of employed labour force from which 46 percent belongs to rural and 37 percent belongs to urban areas, further this percentage has decreased in urban sector and increased in rural sector. Mostly the self-employed workers belong to agriculture sector and are not literate. The author calculated that almost 4 million of the workers are underemployed in Pakistan in 1994-95, underemployment is more in rural areas than in urban areas, as in 1987-88 the rural underemployment was 2.8 times more than urban underemployment and it is almost doubled in 1993-94. The author argued that in 1970s the productivity increased due to the growth in output and employment. In 1980s the productivity was higher but employment rate was low so the growth rate of output was higher, but in 1990s the productivity decreased but employment remained there and the output growth so decreased.
The employment elasticity was higher in 1970s then decreased in 1980s but further increased in
1990s. The author suggested that the employment strategy of Pakistan should be designed in a way that it should eliminate gap between labour force employed, underemployed and unemployed. So the strategies may include the growth of manufacturing sector, increase of labour force absorption in agriculture sector and introduction of new sub sectors within small enterprises. This will revive the economy in such a manner that downward pressure on productivity in agriculture will be reduced and positive focus on employment generation should be placed in manufacturing and small scale enterprises.
53
Khan. J. (2005) calculated the employment elasticity of small scale manufacturing sector. The author divided small scale industries in two groups first industry have 1 to 9 workers and second industry having 10 to 19 workers, the data has been used on 275 observations for the time period of 1976 to 1986. The author calculated employment elasticity with respect to value of output, capital labor ratio and wages. The results show that the wage elasticity is negative in low employing industries as compared to high employing industries, so with increase in employment generation wage sensitivity will decrease and output sensitivity will increase. Across all models the employment elasticity with respect to wage is less then capital output ratio and value of output, so it means that in small scale manufacturing units’ employment has a negative relationship with wages positive relationship between capital output ratio and value of output.
Further the employment elasticity writhe respect to output is greater than employment elasticity with respect to capital out ratio and wages. Further the author found that as wages increases the employment will fall but decrease in employment is more in small scale as compared to large scale industries. The elasticity with capital output ratio shows that in large units the increase in capital decrease the demand for labour faster then small scale units and finally the employment elasticity of both small and large units is positive with output showing that an increase in output will increase the labour demand and further the increase in demand for labour in large units is higher than small units. The elasticity results quoted by author are as follows:
54
Table 2.4: Elasticity: Results of the Models Identical Models Wage elasticity Capital ratio Value of elasticity Product elasticity
First model (1 to 9) -0.47 -0.24 0.66
Second model (10 to 19) -0.37 -0.29 0.69
Third model (1 to 19) -0.42 -0.43 0.72
Source: Khan. J(2005)
Haider. A. (2010) analyzed the sectoral employment generation of Pakistan in pooled analysis and Feasible Generalized Least Square method (FGLS) is used. The author found that in agriculture sector the employment elasticity is negative with respect to GDP. As one percent growth of output in agriculture sector will have the negative effects on employment growth of agriculture by 7.6 percent, so employment demand in agriculture will decrease as output grows.
On the other side the elasticity of employment demand for all non-agriculture sectors with respect to output is positive, so increase in the output of non-agriculture sector will generate employment more especially construction, manufacturing and mining sectors. The author then calculated the threshold levels of the sectors that will absorb labour. The threshold level of mining and manufacturing was highest from 1974-2008, so GDP in these sectors should increase by 19 percent in order to generate employment, but in actual the GDP growth of manufacturing was 6 percent in 1974-2008 which is far less than the threshold level. Mining and manufacturing employ 13 percent of the labour force and the contribution of these sectors to national GDP is 22 percent and from 1974-2008 these two sectors combined grew at 5.7 percent. The jobless growth of GDP in Pakistan can be a result of jobless growth in mining and manufacturing sectors. So the
55
author finally concluded that as agriculture is the dominating sector of Pakistan but is not contributing towards employment generation, whereas the non-agriculture sectors can generate more employment among them mining and manufacturing are the key sectors that needed the attention of the government in order to achieve their threshold level of growth which will result in generation of employment in these sectors.
Hull. K. (2009) discussed the relationship of economic growth, employment and poverty reduction in a framework of three stages, as in first stage the decomposed impacts of economic growth on job creation are analyzed to understand that either employment and productivity are associated with the reduction in poverty or not, in second stage sectoral analysis is carried out and finally the broad policy instruments are analyzed in stage three. The author discussed that in understanding about employment creation GDP per capita should be decomposed into three parts as productivity changes, change in employment rate and demographic changes. The per worker output growth can be calculated from growth in value added by per worker in two-time periods and growth is attributed to change in employment and is calculated from value added of per worker. The author analyzed the data of largest Central American state Nicaragua from 2001-
2005 and found that value added growth during the period was 12 percent. The results also showed that in three quarters 74 percent of the growth is linked to structure of the population, so it means if there is a decrease in number of dependents per working age person it will generate a growth of 5.3 percent, further the employment will also be increased and lastly the decrease in productivity will decrease the growth rate. So, it is important to understand at first step that whether the growth is associated with employment or productivity rather than simply calculating the elasticity of employment.
56
The author further discussed that the second step is to find out that either growth is helping in reducing poverty or not. So, if employment intensive growth or productivity increased growth reduces poverty then the relationship of employment intensive growth and productivity growth with poverty will be negative. Based on the empirical data the author found that there exists a relationship between employment intensive growth, productivity growth and poverty reduction.
During the sectoral analysis the author found that employment intensive growth reduces poverty in manufacturing, construction and mining sectors but in agriculture sector the employment intensive growth increases poverty. However, the productivity increase growth reduces poverty in agriculture sector. So, this second step will help to find the reduction of poverty either related to employment intensive growth or productivity intensive growth. The third step is the broad policy making which emphasized certain facts that minimum wage legislation turned to be significant in employment intensive growth but openness of trade is highly significant, education enhance the skills and improves the productivity, so in productivity growth education has an important role to play.
Aisha Ghaus, Pasha and Ghaus (1996) examined the difference in the social development of each province study confirms that there exists a strong correlation between levels of social and economic development. Urbanization, provincial administrative development and geographical and economic significance appear to be the sources of regional variation. Balochistan appeared to have the lowest level of social development followed by Sindh, N.W.F.P and Punjab. Uddin
(2007) observed that the demand for social services is expanding rapidly, mainly owing to high population growth and rapid urbanization. Siddiqui (2008) shows that the government provision of social services affects human capabilities significantly, she alsofinds that aggregate statistics at the national level hides region specific reasons of poverty and inequalities. The variations in 57
the indicators across the provinces are an indicative of regional disparities in quality of life in terms of income, health and education. Pasha and Hassan (1982) showed that almost 15 per cent of the population of Punjab and 27 per cent of Sindh lives in the underdeveloped area. Jamal and
Salman (1988) also indicated that there is lot of difference in the level of development at intra and inter provincial level in Pakistan. Both studies have indicated that, significant differences in demographic, institutional, social sectors and economic base of each province is the main reason for inter provincial disparities. The above studies clearly indicate following things first, the level of development varies from province to province. Second, the source of regional disparities is because of different economic, social and geographical base.
2.10 Research Gap In macroeconomic analysis economic growth plays a pivotal role; either the fluctuations are short run or long run, they are still the source of great attention to macroeconomists. Many of the economists believe that while discussing the growth fluctuations of any economy, the regional growth patterns are also to be analyzed. Moreover, the multidimensional approach of macroeconomics suggests that the human capital development is also an important indicator, which should be incorporated while analyzing growth patterns at country level or regional level.
Most of the macroeconomic models developed placed a lot attention to the accumulation of capital that will increase the output of the country, but they ignore the role of labour force accumulation in output growth. Innovations plays important role in development, as the growth of knowledge play an important role in the output growth and in improvement in living standards. So, based on above theoretical researches it can state that regional growth may differ because of different rate of technical progress with in regions, the growth of capital stock may
58
vary between regions, the growth of the labour force and technical innovations may vary between regions, thus affecting their growth patterns.
Barro (1996) attempted to analyze empirically the growth theories and he showed that level of schooling, life expectancy, lower fertility, government consumption, rule of law, Inflation,
Political freedom and the terms of trade are major determinates of economic growth. Graff
(1995) analyzed the role of human capital in explaining economic growth at cross country level.
He found that that the accumulation of physical capital, human capital and technological progress are important determinants of economic growth. Jenkins (1995), for the UK economy, confirmed the finding that the investment in human capital instigates to increase productivity.
Similarly, Asteriouand Agiomirgianakis (2001), for the Greece economy explored the relationship between formal education and gross domestic product. They found significant relationship between the two and further depicts that the causality runs through education variables to economic growth.
Literature pertaining to Pakistan is mostly available at country level. Most of the models developed historically by economists show the economic growth of the country with different macroeconomic indicators. Many of the studies pertaining to human development have been performed in the literature related Pakistan. Either, economic growth; in short run or in long run, is mostly discussed in literature at country level. However, very few studies focus on the regional development of the country. Chaudhry. A (1989) and Chaudhry. A (1999), Hussain (1993),
Jamal (2015), Nazir, M and Yasin, H. M (2011), Aisha Ghaus, Pasha and Ghaus (1996) and many other researches discussed the regional economy of Pakistan.
All the studies incorporated certain variables on which data was available or calculated by authors to explain regional performances. But, hardly any study focused on growth differentials
59
among regions; moreover, the integration of provinces at sectoral and subsectoral level and the convergence of growth patterns among provinces are rarely discussed in the available body of literature on Pakistan’s economy. The emerging issues like electricity load shedding, terrorrim and increasing disparities in human and physical development, among provinces is hardly performed by the researchers. The very obvious reason for this was the lack of availability of data on regional income accounts, as pioneering work on regional income accounts was done by
Bengali (1993). So, the current study has the advantage of using that data and chalk out growth differences among provinces of Pakistan. The recent issues related to provincial development will also be discussed in the present study.
The study will explore the growth linkages of provinces within sectors, across sectors, within provinces and across provinces. It will give us the relative importance of any sector contributing to growth of that province and either; this particular sector is integrated to the other sectors of the same province and other provinces or not. Thus, the present study will explore the idea of sustainable or indigenous development; in context of provincial development applies to Pakistan or not. The study will also analyze the growth differentials among provinces of Pakistan especially in the context of the effectiveness of fiscal or monetary policy in explain growth differentials among provinces. Lastly present study will also explore if complete decentralization policies are opted, important question is either they work in context of Pakistan or not.
The above stated issues are the reason to carry out the present study as historically the integration of provinces is hardly discussed. So, the present study will focus on regional differences and finally how they can be resolved in such a manner that the inclusiveness of growth can be ensured in Pakistan and lastly how provinces can reap benefits from national growth of the country.
60
Chapter 3
Data Source and Methodology
3.1 Sources of Data and Methodology
In this chapter, sources of data and the methodology to be utilized are described; moreover, how the objectives of the study are analyzed econometrically, is explained in this chapter. In Pakistan, data at provincial level do not exist; therefore, new data sources are developed.
The most difficult task is to gather information for all relevant and required variables at regional level. New data has been compiled for provinces. Still comprehensive data for provinces is not available. Researchers and international organizations tried to estimate the province-wise GDP for Pakistan and the data on all indicators pertaining to socio-economic variables is seen fly or still not available.
Table 3.1, below, presents the list of important variables, indicators and data sources for provinces pertaining to private consumption, investment level and money supply etc. Whenever data is not available, such gaps are filled by using appropriate methodology to carryout research on issues for which regional information is not available.
61
Table 3.1: Province-wise Indicators and Data Sources Variable Allocators* Data Source Province wise GDP SPDC and IPR Province wise Household Household Integrated Economic Consumption Survey (HIES), by Pakistan Bureau of Statistics (PBS). Province Wise Private Gross Fixed Share in bank Credit and State Bank of Pakistan, Capital Formation Advances publication year (SBP) Province wise Public Gross Fixed Share of Provincial ADP and Fiscal Operations, Provincial Capital Formation Allocation of Federal PSDP and Federal Budgetary by origin and Population Documents Province wise employment Employment in numbers Labour Force Survey, PBS Remittances Share in Foreign HIES,PBS Remittances Province wise social indicators Pakistan Social and Living Standards Measurement Survey (PSLM), PBS. Province wise Money Supply Share in bank Credit and SBP Advances Province wise electricity Energy Yearbook and NEPRA Consumption and Generation By ministry of what *where data is available at aggregated level, decompose data by province by using relevant indicators. 3.2 Money Supply by Province Khan and Senhadji (2001), Mubarik and Hussain (2005), Li (2006), Patrick, H. T. (1966), and
Qayyum, Abdul (2006) have used money supply as a determinant of growth of an economy.
I have estimated money supply of province by share in credit and advances. The share of each province, in total credit and total advances, is also estimated. Such data is available at State Bank of Pakistan (SBP) website and in the Handbook of Pakistan Economy. Equation below gives the formula used for decomposition of money supply.
CD AD i.t i.t MSi.t MSn.t * CDN.t AD N.t
MS = Total Money Supply in province i at time t year. i.t
MS = Total National Money Supply in a t year. n.t
CD = Credit deposit by province i in t year. i.t 62
AD = Advances by province i in year t. i.t
CD = Credit deposit by nationally in year t. N.t
AD = Advances nationally in year t. N .t
The share of each province, in total credits and deposits, has been estimated by dividing credits and deposits by province i with total credit and deposit national in the same year and multiply it with national money supply. It will provide us the money supply by province i.
3.3 Gross Fixed Capital Formation by Province Mujahid, N., Amin, A., & Khattak, S. W. (2014); Shahzad, F. (2015); Ghani, E., & Din, M. U.
(2006) and Xu, Z. (2000) have used gross fixed capital formation as a proxy for capital or investment and analyzed its impact on growth. gross fixed capital formation by province has been estimated by using different allocators. The total gross fixed capital formation (GFCF) is the sum of GFCF by private sector, public sector and general government. I have used relevant indicator for each activity separately. The following equation provides such estimates.
GFCF, in above equation, is the sum of private sector (GFCFp), public sector (GFCFpublic) and government sector (GFCFgovt). The methodology for decomposition is given below:
3.3.1 Private Sector The GFCF for private sector for each province is estimated by using credit distributed to private sector for each province by commercial banks. The estimate of equation below provides the formula used for GFCF for private sector
63
Gross fixed capital formation by private sector in province i at time t.
Gross fixed capital formation by private sector, nationally at time t.
CD = Credit distributed to private sector in province i at time t. p.i.t CD = Credit distributed to private sector nationally at time t. p.n.t In the above equation, the total credits to private sector by bank in province i are divided with total credits distributed to private sector nationally. This ratio is multiplied with GFCF by private sector nationally which will give the GFCF by private sector for province i (GFCFp,i,t).
3.3.2 Public Sector The GFCF for public sector is decomposed by using following equation.
=Gross fixed capital formation by public sector in province i at time t.
= Gross fixed capital formation by public sector nationally at time t.
= Credit distributed to public sector in province i at time t.
= Credit distributed to public sector nationally at time t.
I have divided the total credits to public sector enterprises by bank in province i with total credits distributed to public sector enterprises nationally. Therefore, multiply this ratio with GFCF by public sector enterprises nationally which will give us the GFCF by public sector enterprises by province i.
3.3.3 The Government Sector The GFCF for government is decomposed by using following equation. 64
= Gross fixed capital formation by general government in province i at time t.
= Gross fixed capital formation by general government, nationally at time t.
= Annual Development Budget by province i at time t.
= Total National Annual Development Budget at time t.
I have divided the Annual Development Program (ADP) of a province i by total development expenditure nationally, then multiply this ratio with GFCF by government which will provide the
GFCF by government of province i.
3.4 Final Private Consumption Expenditure The study has also decomposed the final private consumption expenditures for each province.
The Household Integrated Economic Survey (HIES) at national level provides the household wise expenditures for each province in Pakistan. Equations, below, is the methodology, used to decompose the final private consumption expenditures by province in Pakistan.
PC EX.i.t PCEi.t PCEN.t * PCEX.N.t
PCEX.P.t PCEX.U .i.t * NU .i.t PCEX.R.i.t * N R.i.t
PCEX.N.t PCEX.U .N.t * NU .N.t PCEX.R.N.t * N R.N.t
PCEi.t = total final consumption by province i in year t.
PCE N .t = total final consumption nationally in year t.
PC EX.i.t = private consumption expenditures in province i in year t.
PC EX.N .t = private consumption expenditures nationally in year t.
PC EX.U .i.t = private consumption expenditures in urban areas of province i in year t. 65
PC EX.R.i.t = private consumption expenditures in rural areas of province i in year t.
PC EX.U .N.t = private consumption expenditures in urban areas nationally in year t.
PC EX.R.N.t = private consumption expenditures in rural areas nationally in year t.
NU .N .t = urban population nationally in year t.
N = rural population nationally in year t. R.N.t
NU .i.t = urban population of province i in year t.
N = rural population of province i in year t R.i.t To compute total consumption expenditures by province, I have estimated the share of province
‘i’ in national household consumption. Firstly, the aggregate consumption expenditures of province ‘i’ is divided by the national aggregate consumption expenditures and thereafter this ratio is multiplied by national final consumption expenditure to estimate the total consumption expenditures by province.
3.5 Variables Used for the Study The study requires information at provincial level in Pakistan. The table below provides the definition of each variable along with the source of data.
66
Table 3.2: Variables Used in the Study Variable Definition Sources EMP Number of person employed by province Labour Force Survey (LFS) NR Federal Net Revenue (billion Rs) Ministry of Finance, Government of Pakistan (MOF) GDP Constant factor cost of 2005-06 SPDC (2005), (IPR) (2015) Pop Population millions of people Economic Survey of Pakistan Elec Electricity consumption Million kwh National Transmission and Despatch Company (NTDC) Pakistan. LR Adult literacy rate of age 15 and above PSLMS FGD Federal Government Development Expenditure MOF (Rs. Billion) PDE Provincial Development Expenditures MOF TR Provincial Tax Revenue (Rs. Billion) MOF NTR Provincial Not Tax Revenue (Rs. Billion) MOF ToR Provincial total revenue (Rs. Billion) MOF Road Road length (kms) Economic Survey of Pakistan LET 000’s Kms NTDC SSC Million Kwh NTDC Rem Remittances Million USD HIES, PBS and Economic survey of Pakistan Cons Private Final Consumption Expenditure Million HIES, PBS and Economic survey of Rs. Pakistan Save Savings Million Rs HIES, PBS GFCF Gross Fixed Capital Formation million Rs Economic Survey of Pakistan MS M2 Money Supply SBP Beds Number of hospital beds Provincial development statistics PCEX Provincial current expenditures MOF FCE Federal Consumption expenditures MOF HCI Human Capital Index SCH Number of Schools Education Statistics Kills Total number of people killed in terrorist South Asia Terrorism portal activities HOS Total number hospitals Provincial development statistics
67
3.6 Strategy for Provincial Integration by Using Sub-Sectoral Causality Analysis A relation shift for causalities is given below:
To test either the growth rate of province i complements the growth of province j, the study uses panel causality test and test that either the output of sector k in province i causes output of sector h in province j.
The simple model which tests the causal relationship between outputs of sector k in province i and output of sector h in province j presented by Granger (1969) is as follows:
m1 m2
Xi,k,t = ∑ aq X i,k,t – n + ∑ bq Y j,h,t – n + εt ……………………………………….. (1)
n = 1 n = 1
m3 m4
Yj,h,t = ∑ cq X i,k,t – n + ∑ dq Y j,h,t – n + ηt ……………………………………….. (2)
n = 1 n = 1
Here the error terms, εt and ηt are not correlated and the mean of E [εt ,ηt]=0. The variable ms show the lag lengths. In the above equations, the direction of causality runs from Y to X if bq is not equal to zero. Similarly, the direction of causality runs from X to Y if cq is not equal to zero.
Further, there is presence of bi directional causality if both bq and cq are not equal to zero and there is no causality between X and Y if both bq and cq are equal to zero.
In the present study, Xi,k,t shows output of sector k in province i at time period t and and Yj,h,t shows output of sector h in province j at time period t.
The null and alternate hypotheses for Xi,k,t causes Yj,h,t are as follows:
68
Hypothesis: 1
Ho: The GDP growth rate of province i, of sector k in time t (Xi,k,t)does not Granger Cause the
GDP growth rate of province j, of sector h in time t (Yj,h,t)
H1: The GDP growth rate of province i, of sector k in time t (Xi,k,t )does Granger Cause the GDP growth rate of province j, of sector h in time t (Yj,h,t)
Hypothesis: 2
For the equation 2, null and alternate hypotheses are as follows:
Ho: The GDP growth rate of province j, of sector h in time t (Yj,h,t )does not Granger Cause the
GDP growth rate of province i, of sector k in time t (Xi,k,t)
H1: The GDP growth rate of province j, of sector h in time t (Yj,h,t )does not Granger Cause the
GDP growth rate of province i, of sector k in time t (Xi,k,t)
The one primary objective of the present study is to develop a regional linkages model for strong regional integration model. To achieve the above objective, following analysis should be performed:
(i) Inter and intra-provincial causality test
(ii) Inter and intra-provincial co-integration test
The above two analyses will allow us to analyze the sectoral linkages within provinces and across provinces. I have used Granger Causality Test for estimating the causality between variables. If both variables are integrated order one, I (1), and there is a co-integrating relationship between them, Granger Causality Test could be based on the following Error
Correction Models (ECMs).
69
….. (A)
…… (B)
Equation (A) is estimated for testing within sector and across provinces causality and equation
(B) is estimated for across sector and across provinces causality. The analysis has been done at sectoral level, i.e., three sectors; Agriculture, Industry and Services. We have also done analysis at GDP growth rate as well. Total 240 causality equations have been estimated.
Granger Causality provides us the short run relationship between variables. To find the long run relationship between variables, the study tested co-integration between variables. The study used pairwise co-integration between variables. To do this, we have used Engle Granger Co- integration test and then we also calculate the error correction term for understanding short term dynamics. For this purpose, we have used province-wise and sector-wise per-capita GDP growth from 1973 to 2015. I have identified the causality at different significant level. The causality is strong if p-value is less than 5 per cent, causality is moderate if p-value is greater than 5 per cent but less than 10 per cent and causality is weak if p-value is greater than 10 but less than 20 per cent. This allows us to identify the strength of relationship within provinces.
Strong Causation is referred in the present study at 5 percent level of significance.
Moderate Causation is referred in the present study at 10 percent level of significance.
Weak Causation is referred in the present study at 20 percent level of significance.
3.7 Lag Selection Criteria
One of the most important steps in time series analysis is the selection of the optimal lag. Many criteria have developed for optimal lag selection but the most renowned is AIC, SC and HQ
70
criteria. I have also used these criteria for the optimal lag selection. The AIC, SC and HQ show that oner period lag is appropriate for the causalities and cointegeration estimationss. Therefore, the same is applied to draw evidinces.
3.8 Unit Root Test
Firstly, is to test the stationary assumption of the series the Augmented Dickey-Fuller (ADF) unit root test is used for examining the stationary of the data set. All varaibles are inorder I(1). The
ADF is commonly used test for stationarity, which is applied for estimation. The following equations shows the simple form of AR(1) process used to test the stationarity.
The Null Hypothesis is , aginst the alternative hypothesis ; is tested.
The biggest advantage of ADF that it is mostly used in case of Pakistani series and it is easy to apply. We use ADF if is IID if it is not then we have to use other test. The result shows that in our case is IID.
3.9 Co-integration Test
A time series is said to be integrated of order one if it generates the stationary series of Yt . The theory developed by Granger (1969) and elaborated by the Engle and Granger (1987). If two non-stationary series generate a stationary series, then they are co-integrated. The co-integration explains the long term relationship. Engle & Granger gave a two-step procedure to test the co- integration.
Step 1 is to estimate the long run equation and estimate the error term or residuals. If errors are stationary at level, then we say that both series are co-integrated. 71
…………….. (1)
The µt is estimated by following equation.
µt = GDPi.t – (α + β1GDPj.t) ……………… (2)
Step 2 is to estimate the error correction term. Error correction model is used to find out the short run adjustment between variables. It helps us to understand short run adjustment in equilibrium between variables.
…………………… (3)
The value of α in the equation 3 gives the short run error adjustment term. The value of µt is estimated from equation 1 and plugged in equation 3. I have used Engle and Granger co- integration because it can be useful for simple co-integration. If both time variables are found to be I (1), then there may exist a long run relationship between these variables. E&G Co- integration test will help us to test the existence of long run relationship between any pair of variables. Co-integration is a powerful theory because it allows us to understand and test the existence of equilibrium among two or more variables each of which is individually non- stationary. Important to note that, all series GDP of province i, Agriculture of province i,
Industry of province i, and services of province i, have unit root and integrated of I (1). These satisfied the first condition of co-integration. So, we can use co-integration to find the long run relationship between all variables. The next task was to run multiple co-integration or individual co-integration but I preferred to use pair-wise co-integration between sectoral growth rates of provinces. The objective is to analyze the relationship between province and sectors that has allowed us to estimate the partial impact of activity i in a province on the activity i in other province and activity j of the same province. Using multiple co-integrations means reduction in the degree of freedom which is number of observations and number of estimated parameters. As 72
the number of estimated parameters increase, the degree of freedom will decrease. So, we have used E&G to maximize degree of freedom.
The table 3.3, below, provides sector-wise total number of co-integration equations estimated.
For each province and each sector, we have estimated 11 equations to completely understand the dynamics of a particular sector of a province.
Table 3.3 Sector-wise Co-Integration Equation Number of Pair wise Cointegration equations Agriculture 44 Industry 44 Serivces 44 GDP 12 Total 144 3.10 Growth Equation The study uses panel data for above analysis given in chapter 1. We have four provinces (cross sections) and data from 1990 to 2015. Since the GDP growth estimates are developed by different studies from 1973 to 2015 as explained above. Therefore, the Granger causality test and co-integration analysis is performed in the present study for this time period. While exploring the growth determinants and disparities between provinces and further addressing the issues like infrastructural development and growth of provinces, terrorism and growth, load shedding and growth, employment elasticity and capital output-ratio requires the availability of data at provincial level on all variables explained above. As the data on these variables is not present so for the growth equations, the study used data from 1990 to 2015 in panel form and fixed effect model on the panel data from 1990 to 2015 is used to avoid econometric problems related to less number of observations.
Following Engen & Skinner (1992), the author extended theroatical model and applied to analyze the impact of fiscal and monetary policies, on the GDP growth of the country. The 73
model is similar to above cited model to analyze the impact of fiscal policy on the GDP growth of developing countries.5
Equation 1 below is the standard equation used by Solow (1996) and many studies which have extended to incorporate other determinate of economic growth.
……………………. (1)
Where, at time t; is the output in an economy, is efficiency of the production function of a country, is the amount of labor available in a country, is the amount of capital available in a country. The equation 2 is the more general which incorporates more determinates of output used by various studies:
……………………. (2)
Where, at time t; is the output in an economy, is the vector of supply side determinates of output an economy, is the vector of demand side and fiscal determinates of output in an economy, is the vector of monetary determinates of output in an economy, is the vector of foreign sector determinates of output in an economy, is the vector of financial sector determinates of output in an economy, is the vector of institutional determinates of output in an economy.
3.11 General Model for the Study After identifying the conventional growth determinants from the theory, the first model will estimate the growth determinants of Provinces. There are two ways of estimating the conventional model, first we estimate equation for each province separately and then do coefficient comparisons. But the disadvantage of this method is that we are not able to analyze
5For detailes please see Engen & Skinner (1992) “Fiscal Policy and Economic Growth, National Bureau of Economic Research, Cambridge, MA. 74
the different impact of Federal Policies which are same for all regions, like tax rate, policy rate, etc. on different regions.
The above problem can be resolved by using a panel data analysis. Then equation 2 can be written as follows:
Yit = f (Xit, Git, Mit, FTit)……………………. (3)
Where, at time t; is the output for ith province, is the vector of supply side determinates of output in an economy, is the vector of demand side and fiscal determinates for ith province,
is the vector of monetary determinates for ith province, is the vector of foreign sector determinates for ith province.
In the above equation 3, variables of financial sector and institutional has been dropped as they were present in equation 2. The reason is that equation 2 identifies the determinants of growth which different studies have been opted but as in the present study model 1 is using data on provinces within time period across cross-sections for all provinces. So, the non-availability of information regarding institutional and financial data at provincial level might disturb the model, hence the basic model has eliminated these variables.
3.12 Impacts of Fiscal and Monetary Policies on Provincial Economic Growth Rates The following causalities are tested to see inter-provincial linkages and dependencies.
Do fiscal and monetary policies benefit equally to all provinces or not?
To test the impacts of fiscal and monetary policies, we have estimated the equation 4. This enables us the estimation and analysis of differential impacts of fiscal and monetary policies on provincial growth.
…………………………….. (4)
75
Is the vector of controlled variables. is the vector for province-wise fiscal policy variable, is the federal fiscal policy variables vector and is province-wise monetary policy variables vector. will allow us to estimate the impact of different variables on different provinces.
Firstly, I decompose total expenditures into current and development expenditures for provincial government and federal government separately and analyze the impacts of both on provincial growth.
3.13 Impact of Human, Social and Physical Capital on Provincial Growth Mujahid, N., Amin, A., and Khattak, S.W. (2014), and Imran, M., and Niazi, J. (2011) have analyzed the impact of human and physical capital in case of Pakistan economy. We have analyzed this question and estimated a separate equation because in the presence of development expenditures estimating social and physical infrastructure variables leads to multi co-linearity problem. To analyze the impact of social and physical infrastructure on economic growth of a province, we have used province-wise indicator for social development and infrastructure development and estimate equation 9, given below:
…………………… (9)
is the vector of controlled variables. is the social infrastructure variable which is estimated by province and is the physical infrastructure variable estimated for each province.
Anwar (2012) has estimated the province-wise human capital index. We have used following formula to calculate province-wise social infrastructure index:
76
3.14 Estimation of Capital Output-ratio and Employment Elasticity This study has also calculated capital output-ratio and employment elasticity for each province.
The equation below is the formula for estimation of employment elasticity for each province.
The above formula will provide province-wise employment elasticity.
To estimate capital output-ratio, I have used the formula derived from the Solow model.
Equation below gives the capital output-ratio.
Where is the saving rate, is the population growth rate, is the GDP growth rate and is the depreciation rate. The above equation is the capital output-ratio derived from Solow model.
According to above equation; dividing the saving rate in each province by sum of population growth, GDP growth and depreciation rate in each province.
3.15 Impact of Electricity Load-shedding and Terrorism on Economic Growth of a Province Romer (1986) and Barro & Salai Martin (1991) developed model to analyze economic performenance of the countries; focusing on whether the developing counties are likely to catch up with developed world. This is also known as convergence hyposthesis. In this respect the following model is applied in a general form:
A negative value of shows convergence. Öcal and Yildirim (2010) has extended the above model to incorporate the impact of terrorism incidents on the GDP growth of Turkey. I have 77
applied the same model to analyze the impact of terrorism activiites on the GDP growth of Provinces of Pakistan6.
Padda, I. and et al., (2015), I have constructed a special model to analyze the impact of terrorism on growth of each province. I have estimated the following model to analyze the impact of terrorism activities on economic growth.
‘Kills’ is the number of people killed in terrorist activities in each province. The AR term in the model is to introduce to control the problem of auto correlation and impact of all other factors that may affect the GDP.
The next issue estimated the impact of electricity load-shedding equal impact to all provinces.
So, to do this, I have estimated the impact of load-shedding in each province’s large scale manufacturing.
IPP (2013) has modelled the behavior of a firm in the presence of frequent and persistent electricity outages. The firm is assumed to be operating in a competitive environment given the smallness of its size and pursues profit maximization. If electricity outages are seen as, more or less, permanent in nature then the optimal size of the firm is lower than in the absence of outages. In particular, there is a tendency to shed some labor. Under electricity load-shedding firm initially experiences electricity outages which reduce production. After electricity load- shedding output and without electricity load-shedding give the gap between price and marginal cost. If the gap is larger the bigger will be the outage. If marginal cost is too high, then the firm makes no adjustment.
The following equation has been estimated for this analysis.
6Öcal, N. and Jülide Yildirim (2010),Regional effects of terrorism on economic growth in Turkey: A geographically weighted regression approach, Journal of Peace Research, Vol. 47, No. 4 (july 2010), pp. 477-489 78
= Per-capita growth rate of large scale manufacturing in province i at time t
= Electricity consumed by industrial sector of province i at time t
= Per-capita Growth of other sectors
= Electricity load-shedding dummy variable since 2009-2015 = 1 otherwise 0
If Di is significant and negative, then load-shedding has a significant negative impact on growth rate of manufacturing sector of a province. ECI and GOS have used as control variables to capture the impact of other variables.
3.16 Modeling Technique As explained above that we have data for four provinces from 1990 to 2015 which implies that we will use panel data estimation technique rather than time series analysis. The biggest advantage of panel data is that it allowed us to capture the province specific effects and differential impact of a particular variable on province.
3.16.1 Fixed Effects Model It may be noted that so far there is hardly any comprehensive study on Pakistan which may have estimated such important determinants.
The objective of the above analysis is to analyze the determinants of growth of a province and the factors that explain the variation in growth process of a province. The fixed effect model will allow controlling for the average differences across province in any observable or unobservable predictors, such as differences in quality and sophistication. When you have data that fall into categories like provinces and cities, we shall normally want to control for characteristics of those categories that might affect the dependent variable.
79
Fixed effect (FE) model allows for differences in the intercept among cross section units. FE model is a reasonable approach when we are confident that the difference between cross section units can be viewed as parametric shifts of the regression function. That is, it is based on the assumption that differences across cross sections can be captured through intercept.
The FE model is represented by the equation:
Yit i X it it
Where i = 1, 2,3,4 and t = 1, 2, 25 i refers to cross section (or the group) and t refers to year.
Note: The component or intercept i in this model differs across cross-section units but is time- invariant. Assumptions of the fixed effect model [Wooldridge (1999)] are:
1. E it / X it 0
2. No multicollinearity exists among the explanatory variables. That is, there are no perfect relationships among the explanatory variables.
2 3. V it / X it V it for all t = 1,2,…,T and all i = 1,2,…,n.
4. Cov it js / X it 0 for all t ≠ s (errors are uncorrelated over time).
2 5. Conditional on X it , the it are independent and identically distributed as Normal (0, ).
6. EX it i 0
Under these assumptions, the fixed effect estimator of β is the best linear unbiased estimator.
Assumption 4 implies that the errors are serially uncorrelated.
80
This model could be estimated by the least squares dummy variables method. This is a classical regression model.
3.16.2 The Random Effects Model In contrast to FE model, the random effects (RE) model views individual specific constant terms as randomly distributed across cross-sectional units. This would be appropriate if we believed that sample cross-sectional units were drawn from a large population.
The RE model is represented by the equation:
Yit X it i it Where i = 1, 2, 3, 4; t = 1, 2, …, 25 Note: The component μi is the random disturbance characterizing the ith observation and is constant through time. The RE formulation assumes that intercept term has random element whereas the FE formulation does not. This is the first difference between the two estimators.
Assumptions of the RE model are:
1. Ei 0
2 2. V i 3. E(μi μj) = 0 for all i ≠ j This model could be rewritten as
Yit X it it
Where it i it Then for this error component model
2 2 2 4. E it
2 5. Eitis for t ≠ s
81
In addition to the above assumption and the assumptions in the FE model, we also require another assumption (in place of the 6th assumption)
6. EX it i 0 It can be noted that μi– i.e., the cross-section specific error term in RE model - is uncorrelated with Xit whereas αi – cross-section specific term in the FE model - is correlated with the Xit.
This is the second difference between the two estimators.
The estimation in case of the RE model is done using the feasible generalized least squares
(FGLS) method as necessitated by its error variance structure.
The two estimators – FE and RE - have different properties depending on the correlation between group specific effects and the regressors.
I. If the effects are uncorrelated with the explanatory variables, the RE estimator is consistent and efficient. The FE estimator is consistent but not efficient;
II. If the effects are correlated with the explanatory variables, the FE estimator is consistent and efficient but the RE estimator is now inconsistent.
This difference is applied to construct the Hausman test which helps deciding the choice between the FE and the RE estimator. The null hypothesis under the Hausman test is given below:
H = Effects and explanatory variables are uncorrelated.
The test statistic is defined as
1 H RE FE FE RE RE FE Where RE and FE refer to random effects and fixed effects estimates respectively and Σ stands for the estimated variance-covariance matrix. Under null hypothesis, the test statistic follows asymptotically a chi-square distribution with k degrees of freedom; where k is the number of coefficients compared. If test statistic is greater than ‘critical value’ of the chi-square distribution 82
at appropriate degrees of freedom, we reject the null hypothesis Johnston and Di Nardo, (1997).
On the basis of the Hausman test, this study adopts the Fixed Effect model.
We have Generalize Least squares method with weighted by white error term to estimate the parameters of fixed effect models.
83
Chapter: 4
Intra-Provincial Economic Linkages,
Economic Fluctuations and their Adjustments This chapter is focused to explore the level of integration between provincial GDP growth rates of the provinces of Pakistan; which is known as ‘bottom to top’ approach in research. Such research and analysis of provincial economies has not been done for Pakistan. There are three main sectors of the country i.e. Agriculture growth, Manufacturing growth and Services growth.
Thus to analyse the dynamics of provincial GDP integration among provinces, it is necessary that the GDP of all the provinces is decomposed into the above cited three sectors. Moreover, the interprovincial and intra-provincial sectoral relationships of these components of provincial GDP provides us an in depth insights of the integration of GDP at provincial level. The long run and short run integration and the interdependence of provincial GDP growth rates, as well as sectoral growth between provinces will help to analyze the spill over effects of growth among provinces.
Granger Causality test and Engle Granger Cointegration test is applied to draw empirical evidences, as discussed in methodology. The Granger Causality test shows the flow of causation between sectors, as this technique is only used to understand the cause and effect theory, but this technique is unable to identify short run adjustment paths and long run integration of series. To understand the short run adjustment path and long run integration between variables the present study has used Engle Granger cointegration technique. By exploring sectoral linkages between provinces will let us know the extent of integration and linkages among provincial GDP of provinces and between sectors of provincial GDP among provinces. The study will analyze how 84
an integrated growth among provinces can be achieved and to maintain sustained growth of national economy. It will also help to understand whether periphery regions are integrating with national economy or not. In other words, it has been explored whether convergence among provinces is taking place or not.
To explore the growth linkages among provinces, it is important to analyse the dynamics provincial economy. Thus, this chapter is divided into two sections, to explore the provincial
GDP growth and sectoral growth causalities and integration among provinces of Pakistan.
Section one, will discuss the Cointegration analysis by applying Error Correction model between provincial GDP growths and sectoral growths; across sectors and across provinces. Section two of the Chapter will explore presence of causal relationships between provincial GDP growth and sectoral growths; within sectors, across sectors and across provinces.
4.1 Sectoral Development in Pakistan The economy of Pakistan is semi-industrialized and mostly the centres of growth are along Indus
River and few major urban areas. The growth poles of Pakistan are mostly in Karachi and in urban centres in Punjab along with few other developed areas of the country. The major industrial activities were developed in the 1960’s. Thereafter, hardly much focus was on industrial development. Different five years plans are developed that focused on the medium term and long term growth targets, but the poor or irregular implementation of these plans, results in non achievement of the growth targets set by government in these plans. It has also been observed that the social indicators have performed poor than macroeconomic indicators; the very reason for this is the lack of policies and funds dedicated to the social development in
Pakistan. Many of the researches highlights; that without social or human development it will be difficult for any country to achieve sustainable growth patterns. Moreover, much of the research 85
in the area of macroeconomics, also indicate that for sustainable growth the regions of the country should grow in an integrated manner.
Usman. M (2016) analysed the data of 1990-2014 and tested the hypotheses that either agriculture have a significant contribution to national GDP growth or not and further the role of major crops, other crops and live stocks to the development of agriculture and their significant contribution to national GDP are explored by the author. The author finally concluded that the role of major crops, other crops and livestock are important for the agricultural development of the economy and further they will increase the national GDP of the economy also.7 The low growth of agriculture in provinces is also due to high federalism as discussed by Pakistan
Institute of Legislative development and Transparency (PILDAT) (2015) highlighted the main issues related to agriculture development of Pakistan. The report emphasized the fact that there are two type of challenges pertaining to agricultural output of Pakistan, one type of issues are related to availability of water and the other type are non water issues.8
4.2 Industrial Development Industrial development has an important role to play in the economic growth of any economy.
Most of the developed countries had taken the root of transformation from simple agricultural base, to high tech industrial base and has finally transformed the major part of their GDP in services sector growth; Whereas the case of developing countries are little different. There are socio economic issues related to the developing countries in transforming their economies from simple agriculture to high value added manufacturing industries and finally to the services sector growth. The leading issues include the low rate of literacy, especially the human development
7. See for details Usman. M (2016). 8. See; for details Pakistan Institute of Legislative development and Transparency (PILDAT) (2015). 86
and business climate that are responsible for the non development of industrial base in developing countries.
Industry basically takes inputs from both the agriculture and services sector, so for any country to achieve a high growth trend in manufacturing. The manufacturing sector should have backward linkages with the agriculture and forward linkages with the services sector. Without such linkages the industrial growth would be unsustainable and the industry will not experience constant path of growth. Innovations in manufacturing sector have now become significantly important; especially in the global markets only those economies have higher shares that are innovative and use the domestic resources efficiently. The foreign investment also give boost to the industrial development but this can be possible if domestic markets are conducive for investment and the foreign investors are secure in their investment projects.
Most of the industries of Pakistan are low value added and non-competitive in international markets. One of the key reasons is that; the industry of Pakistan is not attaining sustainable pattern of growth. It will be difficult for the local investors to adopt new technology as the illiteracy across all provinces is high; moreover, there is lack of vocational training programs in the country. So in most of the cases the industrialist of Pakistan could not be able to find proper human resource to deal with the new technological development across the globe. The other major problems with industrial development related to Pakistan are as most of the industries in are agro based and are still low value added. So during years when agriculture growth is low the manufacturers have to import the raw material from abroad that decreases its competiveness. It will also be difficult for industry to create forward and backward linkages and the spill over effects to other sectors of the economy. The reasons include lack of financial availability, insurances and other services that are absent in many provinces across country.
87
4.3 The Services Sector The services sector is a source of achieving high economic growth of the economies. The importance of services sector to any economy’s growth is mainly associated with two important factors. Firstly, services sector provides high value added products even more than the commodity producing sectors, so due to the high value addition of the sector to the economy’s output the economy can achieve high growth rates. Secondly, services sector provides forward and backward linkages with the commodity producing sectors of the economy, thus enabling the commodity producing sectors to achieve high growth in their own sectors and subsequently contribute to the overall growth of the economy. Many of the developed nations has transformed most of their economies from the growth of commodity producing sectors to the services producing sector and enjoying high rates of growth. Another important reason for the high development of the services sector is that it does not require huge land for cultivation like in agriculture sector or it does not require huge amount of physical capital involves like in the development of manufacturing sector. So the economies that are lack of geographical area or physical capital intensity finds easier to establish themselves in services growth. It is true that many of the developed nations have achieved high rates of services sector growth and the other developing nations are following the footprints of developed nations. However, the development of services sector is highly dependent upon the development of commodity producing sectors and human development because eventuality it is the commodity producing sectors or skilled labour that facilitates the growth of services sector. It eventually helps the services sector to develop backward and forward linkages within the economy. Without the development of sound agriculture base, industrial base and human development it will be difficult for any economy to achieve growth in services sector.
88
Pakistan during past few years have realised the importance of service sector and now major share in GDP of Pakistan comes from the services growth. There are few studies that highlighted the importance of services sector in the economy as highlighted by Word Bank in its report
World Bank (2016) The services sector contributes 59 percent to the Pakistan GDP in 2016, it has become a main engine of growth. The services sector grew at a rate of 5.7 percent in 2016, which is highest of last decade. The growth of services sector was basically supported by whole sale, retail trade, finance, insurance and government services sub sectors. The whole sale and retail subsector increased to 4.6 percent in 2016 from 2.6 percent in 2015. Finance and insurance increased to 7.8 percent in 2016 compared to 6.5 percent in 2015, as banks enjoy high pre tax profits. The growth of general government services increased to 11.1 percent in 2016 from 4.8 percent in 2015. The communication industry has increased to 10.9 percent in 2016 compared to
2.6 percent in 2015.
The ministry of commerce Government of Pakistan report further elaborated that that the share of services sector in Pakistan has grown sharply in past few years as from 56 percent of GDP in
2005 it increased to 57.7 percent of GDP in 2013-14. Major sectors include; financial services, insurance, transport, retail, wholesale and storages. The share of services in global economy is
0.06 percent in 2013-14 mainly the low share is due to non tradable services and barriers to services exports. Thus, the government should ensure high growth of services exports by facilitating services sector development in the economy. The increase in services sector largely depends upon the growth of commodity producing sectors. So the government should ensure the sustainable growth of commodity producing sectors and build regulatory institutions,
89
international engagements and harmonisation in policies to increase further the share of services sector in the economy9.
The main contribution of high growth of services during 2010-11 was from public administration and defence which increased by 13.2 percent and social services that increased by 7.1 percent.
Moreover the research and education, business services, consultancy services, information technology and computers also made significant contribution towards the growth of services sector.10
4.4 The State of Provincial Integration in the National economy In case of Pakistan, all its provinces are not developed equitably. The small provinces feel deprived and political tension has been developed overtime. The centre has hardly made any significant effort to bring the deprived areas into the main framework of the economy. In the 6th
Five-year development Plan, a special chapter was added to improved deprived regions, however, it was hardly implemented. The attention of the central government has been focused on the development of Punjab and Sindh, whereas KPK and Balochistan were not given the required attention of the government. The introduction of Green Revolution significantly developed agriculture in Punjab and Sindh; mainly focused on the Indus Basin. More than ninety percent industries are located in Sindh (Karachi) and Punjab. Moreover, the provincial integration between Punjab, Sindh, KPK and Balochistan was never been the priority of any government. If the growth of all provinces were taken care of and further the growth should be linked to each other among provinces, it could have developed the country on equitable basis,
9.For details see; Report (http://www.commerce.gov.pk/wp- content/uploads/2015/03/Introduction_services_sector.pdf.)
10See for details (Report on Trade in services sector in Pakistan, M/o Commerce, GOP.).
90
which must have led to remove deprivation of small provinces. Inter-provincial tension is mainly due to inequitable economic growth in the provinces. Hardly any comprehensive study exists which may have identified the regional inequalities and deprivations on scientific basis. As if one province grows it will increase the growth of other province, sustainable growth patterns could have achieved by the country. Thus, due to the lack of provincial integration, the growth of the country is low and unsustainable.
In order to achieve an integrated growth patterns among provinces, it is important to explore the sectoral growth linkages among provinces. Ikram 92009) pointed out that while discussing the issues of sustainable and integrated growth among regions, it is important that study should address issues of employment, poverty or deprivation and delivery of services in these regions.
More importantly the economic disparities or divergences in human resources between regions are needed to be estimated correctly. As in the past between East and West Pakistan it was the economic disparities that resulted in the disintegration of the country. So in current time period if the regional disparities are no addressed properly or policies related to regional divergences are to be formulated. It may lead to severe consequences.
4.5 Profile of the Provinces 4.5.1 The Province of Punjab Punjab being the largest province is the main focus of the government as strengthening of the institutions, fiscal operations and up gradation of social services are basically the main agenda of the provincial government. Its share in Population is around 53 per cent. The GDP growth of
Punjab decreased from 3.88 percent during 1973-78 to 3.02 percent during 2008-13. The share of agriculture has declined from 27.9 per cent in 1990 to 23.8 per cent in 2015. The share of
91
industry has also decreased from 24 per cent to 15.4 per cent and share of services has increased from 48 per cent to 61 per cent in above mentioned period.
Agriculture is one of the main sector contributing high to the provincial GDP, but since from last decade due lack of research and development, the labour force participation in this sector has been declined. Moreover, due to structural changes agriculture growth could be improved but ironically it is not being done. The others sectors of the provincial economy include manufacturing, transport, retail and other services. As in 2004 two third of the labour force was engaged to these sectors, private sectors play an important role in this as almost, 90 percent of the labor force is engaged with private sector in these sub sectors. The small and medium businesses in the province have not utilized their full potential, as due to lack of electricity and support from the public sector especially the availability of funds. Thus, improved public sector infrastructure and effective devolution plane is needed in this province to fully exploit its growth potential.
4.5.2 The Province of Sindh Sindh is the second largest province of Pakistan in terms of population and in terms of its share in national economy. The total population has increased from 24.4 million in 1990 to 46 million in 2015 that shows an average growth rate of 2.6 per cent per annum. The employment has increased from 6.7 million to 13 million during the same period with an average growth rate of
2.7 per cent per annum. The employment growth is slightly higher than growth of population.
The real GDP growth rate is 4.2 percent, on an average. The shares of agriculture and industry have declined and share of services sector is increasing.
92
4.5.3 Khyber Pakhtunkhwa (KPK) province The KPK is the third largest province of Pakistan in terms of population and in terms of its share in national economy. The total population has increased from 14 million in 1990 to 26 million in
2015 that shows an average growth rate of 2.4 per cent per annum. The employment has increased from 3.3 million to 5.9 million during the same period with an average growth rate of
2.4 per cent per annum. The employment growth is equal to the growth of population. The real
GDP at constant factor cost of 2005-06 has increased from 0.4 trillion in 1990 to 1.2 trillion in
2015 and growth rate of 4.9 per cent on an average. The share of agriculture has declined from
26.3 per cent in 1990 to 15.7 per cent in 2015. The share of industry has increased from 20.7 per cent to 28.6 per cent and the share of services has increased from 52.9 per cent to 55.8 per cent in above mentioned period. This section presents the cointegration and error correction model results for KP.
4.5.4 Balochistan Province The Balochistan is the largest province of Pakistan in terms of area and the smallest in terms of population and the share in national economy. The total population of the province has increased from 5.4 million in 1995 to 9.9 million in 2015 that shows an average growth rate of 2.5 per cent per annum. The employment has increased from 1.1 million to 2.9 million during the same period with an average growth rate of 4.0 per cent per annum. The employment growth is greater than the growth of population. The real GDP at constant factor cost of 2005-06 has increased from 236 billion in 1990 to 487 billion in 2015 and growth rate of 2.9 per cent on an average.
The share of agriculture has declined from 33 per cent in 1990 to 27.6 per cent in 2015. The share of industry has increased from 22 per cent to 25 per cent and the share of services has increased from 44.8 per cent to 47.4 per cent in above mentioned period.
93
4.6 Cointegration and Error Correction Model (ECM) This section will focus on the inter linkages and intra linkages of GDP growth rates of provincial economies and inter linkages and intra linkages of sectoral growth relationships of provinces; i.e agriculture, manufacturing and services sectors across provinces. The model applied for this purpose is given below. All the tables in this chapter are based on the estimates of the following equations.
The equations tested for cointegration relationships are:
Step 1 is to estimate the long run equation and estimate the error term or residuals. If errors are stationary at level, then we say that both series are co-integrated.
……………………….1
Y represents the industrial value added of a province and X represents the value added of agriculture, industry and services, where i and j represent provinces: Punjab, Sindh, KP and Balochistan.
The µt is estimated by following equation.
µt= Yi.t – (α + βXj.t) ………………2
Step 2 is to estimate the error correction term; error correction model is used to find out the short run adjustment between variables; it helps us to understand short run adjustment in equilibrium between variables.
……………………….3
The value of α in the equation 3 gives the short run error adjustment term. The value of µt is estimated from equation 1 and plug that in equation 3.
There is hardly any on comprehensive regional (provincial) study that may have explored economic linkages between provinces of Pakistan and between the sectors of provinces. For policy formulation regarding regional development and sectoral growth, it is important that these
94
linkages are explored; so that sustainable and equitable economic development across provinces is ensured. This is the very reason that the present study will, firstly, apply the Cointegration and
Error Correction Model to draw empirical evidences of GDP growth linkages between provinces and sectoral growths among provinces.
4.7 The National Economy of Pakistan: Granger Causality and Cointegration Analysis
Husain (2014) investigated the causality between, GDP exports and imports growth of Pakistan by using bi-variate autoregressive processes from 1976 to 2011.11 Sterletis, A and Afxentious, P.
(2000) investigated causality between exports, imports and economic growth of fifty countries from 1970-1993.12 Khatoon. S. and Aisha. Z (2009) discussed casual relationship of federal revenues and expenditures for the period 1972 to 2007. The author found that there exists causality from federal expenditures to federal revenues, whereas the total revenues do not
Granger cause expenditures. This may be the cause of high budget deficits, as government first spend and then collect revenues. Rehman. J and et.al (causal nexus between economic growth, exports and external debt (PIDE) estimated causal linkages between GDP growth, exports and
11 The author found that there exists bidirectional causality between GDP and exports of Pakistan and causality from Growth to exports is more significant. Moreover there is a weak direction of causality found by author from imports to exports and from imports to GDP for Pakistan. 12The authors found that including the Asian tigers the export led growth of neo-classical theory is not supported, as there was absence of relationship especially; Granger causality from exports to growth in most of the countries. The similar results were found with imports to growth as in most of the developing countries the casual relationship from imports to growth is not significant. The authors finally concluded that main drivers of the growth are domestic resources. The structural transformation or institutional developments come more utilisation of dynamic domestic resources, rather than relying on trade with other countries. International trade is not strategically indispensable for development of a country. 95
debt servicing of Pakistan for the period of 1970-2007. 13 Riaz and et.al (2014) estimated causal link between inflation, inflation uncertainty and economic growth of Pakistan for the period of
1972-2012.14 Fatima. H. and et.al (2012 confirmed) explored the short run and long run linkages among these variables through cointegration and causality test was performed on education,
GDP, poverty and physical capita from 1972-2009.15 Zaheer. R and Siddiqui. Z (2017) analysed the relationships between per capita GDP, CPI, GDP, government and private consumption for the time period of 1980-2016.16
All of the studies above used country level data on Pakistan’s economy and explore the causality and cointegration between macroeconomic indicators of Pakistan. There is hardly any study that identified linkages between provincial GDP growth rates and sectoral growths of provinces.
Thus literature is non-existed, which may have focused on regional linkages and development in
Pakistan. It is one of the major reasons that present study explored the growth linkages of provinces, which will identify growth relationships in each region (province).
13The authors found that exports are not determining the growth in Pakistan economy. Rather it is the economic growth of Pakistan that determines exports. The authors found that from debt servicing to economic growth with bivariate and trivate system of equations, there exit a unidirectional causality from debt servicing to GDP. 14The authors found that there exists unidirectional causality from inflation to economic growth of Pakistan and the sign is negative. So, high inflation means low economic growth. The inflation uncertainty affects the economy positively and the sign is also positive, but there is no causal relationship between the variables. The authors finally found that Granger causality runs from inflation to economic growth, whereas, from growth to inflation, inflation uncertainty to growth and growth to inflation uncertainty and finally from inflation uncertainty to inflation no significant causal relationship is found. 15The authors confirmed that education helps in reducing poverty and improving socio-economic factors in Pakistan. The study found that in education, poverty, economic growth and physical capital there exist a significant long runrelationship. Lastly the study found that there exist a bidirectional causality between poverty and growth, between poverty and education and there exist causality from physical capital to poverty, education and physical capital. 16The study found that 36 percent of the economy is not registered. The study analysed the impacts of government and private consumption and CPI upon GDP of Pakistan, mainly the study found that private and government spending has positive impact on GDP growth of Pakistan. 96
4.7.1 Economic Linkages of Punjab: Empirical Evidences of Cointegration and Error Correction Model (ECM)
Punjab is the largest provinces in terms of population and its share in Population is around 53 per cent. The real GDP at constant factor cost of 2005-06 has increased from Rs. 2 trillion in 1990 to
Rs. 5.8 trillion in 2015 and growth rate was 4.4 percent; on an average. The share of agriculture has declined from 27.9 per cent in 1990 to 23.8 per cent in 2015. The share of industry has also decreased from 24 per cent to 15.4 percent. The share of services has increased from 48 per cent to 61 per cent in above mentioned period, respectively. The main objective of this chapter is to explain the growth phenomena of Punjab. This has been done by analysing the contribution of other provinces and sectors growth in Punjab’s growth. Second, empirical evidence pertaining to impact of different factors on growth of Punjab will be analyzed.
4.7.2 GDP Growth of Punjab and its Linkages
Overall, the results show that GDP growth of Punjab is only integrated with GDP growth of
Sindh. The GDP growth of Punjab significantly affects GDP growth of Sindh in long run and short run and the error correction term in above table below 4.1 that 24 percent short run disequilibrium in GDP growth of Sindh will be adjustedby GDP growth of Punjab. This shows that the GDP growth rate of Punjab contributes significantly into the GDP growth of Sindh.
These two provinces are interlinked in their development.
4.7.3 Punjab: The Linkages of Agriculture Sector and Adjustment of the Sector
Agriculture sector is the largest sector of the Punjab economy. The share of Punjab in national economy is above 60 per cent. The results of ECM reveal that a significant long run relationship exists between agriculture sector’s growth of Punjab, the agriculture sector’s growth of Sindh 97
and KPK (Table: 4.1). This relation also exists between manufacturing sector growth of Punjab and services sector growth of Punjab with Sindh and Balochistan. The short run and long run parameters between agriculture growth of Punjab and agriculture growth of Sindh and KPK are significant at one percent and 10 percent level of significance respectively. The error correction term shows that 31 percent disequilibrium in agriculture growth of Punjab is adjustedby agriculture growth of Sindh and 19 percent disequilibrium in agriculture growth of Punjab is adjusted by agriculture growth of KPK. The agriculture of Punjab is integrated with manufacturing sector of Punjab and Balochistan in the short run and long run. About 25 percent short run disequilibrium in agriculture growth of Punjab is adjustedby manufacturing growth of
Punjab and 11 percent disequilibrium in agriculture growth of Punjab is adjusted by manufacturing growth of Balochistan. The disequilibrium of manufacturing sector of Punjab is adjusted faster than manufacturing sector of Balochistan. There exists strong presence of linkages between agriculture and manufacturing sector of Punjab.
The agriculture sector of Punjab is also integrated with services sector of Punjab, Sindh and
Balochistan. The error correction term shows that 40 percent of disequilibrium in agriculture growth of Punjab, in short run, is adjusted by services growth of Punjab, further 25 percent disequilibrium in agriculture growth of Punjab, in short run, is adjusted by services growth of
Sindh and 30 percent disequilibrium in agriculture growth of Punjab, in short run, is adjusted by services growth of Balochistan. The dynamics of agriculture growth of Punjab shows that the agriculture growth of Punjab is significantly affected by the manufacturing and services growth of Punjab and the speed of adjustment shows that any disequilibrium in agriculture growth of
Punjab will be adjusted by services growth of Punjab at higher rate than the manufacturing growth of Punjab. The case with Sindh reveals that agriculture and services growth of Sindh
98
significantly contribute to the agriculture growth of Punjab and the speed of convergence to equilibrium from agriculture of Sindh is higher than the services of Sindh. The results of agriculture growth of Punjab with small provinces show that agriculture of KPK also contribute into the agriculture of Punjab. The agriculture growth of Punjab is also affected by manufacturing and services growth of Balochistan. In this case, the speed of convergence to equilibrium from services growth of Balochistan is higher than manufacturing growth of
Balochistan. It may be noted that Balochistan province is a small province in terms of its economy. In brief, the analysis of agriculture sector of Punjab reveals that within agriculture sector there are relationships present between agriculture growth of Punjab and agriculture growth of Sindh and KPK, so within sector across provinces there are linkages present in agriculture growth of Punjab. The results of manufacturing sector reveals that only in Punjab, the backward linkage of manufacturing is present with agriculture growth of Punjab. Whereas, the services sector of Punjab itself and Sindh and Balochistan have developed significant backward linkages with agriculture growth of Punjab.
The dependence of agriculture growth of Punjab on the agriculture growth of Sindh and KPK is mainly due to flow of inputs and industrialisation in Punjab and Sindh, which are dependent on each other. The flow of food items from Punjab to KPK and flow of energy inputs from KPK to
Punjab etc has led to each other’s dependence and linkages. So, any of the disturbances in agriculture output of Punjab will also be linked with agriculture of KPK and Sindh; as both provinces shares their boundaries with Punjab and Sindh. Besides, due differences in weather conditions, as well as, cropping pattern agricultural produces flow among these provinces.
Punjab still relies heavily on agriculture, so any disturbance in the production of agriculture disrupts the GDP growth of Punjab. The disturbances in agriculture production of Punjab was 99
discussed by IPR (2015), the growth performance of the provinces and the sectoral contribution in national value added is estimated. The study finds that agriculture dominates in Punjab with
24 percent compared to national growth of agriculture as 20 percent. The agriculture sector has developed strong forward and backward linkages and has pivotal role in the provincial development.17
The agriculture sector has developed backward linkages with other provinces, as stated above.
The agriculture growth of Punjab has failed to develop forward linkages with manufacturing and services growth of other provinces. Whereas, the backward linkages of manufacturing growth of
Punjab and services growth of other provinces have been developed with agriculture growth of
Punjab. Moreover, within sector the short run disequilibrium in agriculture growth of Punjab is adjusted by agriculture growth of Sindh and KPK, where the rate of adjustment with Sindh is faster; may be due to similar cropping pattern and adjacent province, as well as Sindh being the second largest province.
4.7.4 Punjab Province: Industrial Development and Adjustment
The table 4.1 shows that manufacturing sector growth of Sindh and Balochistan contributes significantly into the manufacturing sector growth of Punjab. The error correction coefficients of manufacturing growth of Sindh show that 15 percent disequilibrium in short run in manufacturing sector of Punjab is adjusted by manufacturing growth of Sindh; further 11 percent disequilibrium in short run in manufacturing growth of Punjab is adjusted by manufacturing growth of Balochistan. Further, the growth of services sector of Punjab and Sindh contribute into the growth of agriculture sector of Punjab. The error correction model tells that 23 percent short
17The report states that due to lack of policy attention by government and lack of dynamism in this sector agriculture of Punjab grew around 4.5 percent in 90s and since then it is growing at a rate of 2 percent for details see IPR (2015) 100
run disequilibrium in manufacturing growth of Punjab is adjusted by services growth of Punjab and 23 percent short run disequilibrium in manufacturing growth of Punjab is adjusted by services growth of Sindh.
Table 4.1: Co-integration Results for Punjab Equ- Dependent Independent Long run Error correction ation parameters parameters Agriculture Sector 1 Agriculture growth of Agriculture growth of Sindh 0.32 -0.31 Punjab (0.0004)* (0.009)* 2 Agriculture growth of Agriculture growth of KPK 0.19 -0.19 Punjab (0.07)*** (0.07)*** 3 Agriculture growth of Manufacturing growth of 0.134 -0.25 Punjab Punjab (0.014)* (0.02)** 4 Agriculture growth of Services growth of Punjab 0.233 -0.399 Punjab (0.003)** (0.002)** 5 Agriculture growth of Services growth of Sindh 0.17 -0.25 Punjab (0.017)** (0.02)** 6 Agriculture growth of Services growth of 0.224 -0.295 Punjab Balochistan (0.022)** (0007)* Industrial Sector 7 Manufacturing growth of Manufacturing growth of 0.18 -0.149 Punjab Sindh (0.0002)* (0.003)* 8 Manufacturing growth of Manufacturing growth of 0.061 -0.107 Punjab Balochistan (0.0000)* (0.0007)* 9 Manufacturing growth of Services growth of Punjab 0.28 -0.233 Punjab (0.008)* (0.0135)* 10 Manufacturing growth of Services growth of Sindh 0.362 -0.234 Punjab (0.005)* (0.022)** Services Sector 11 Services growth of Punjab Manufacturing growth of 0.083 -0.0716 Sindh (0.011)* (0.049)** 12 Services growth of Punjab Manufacturing growth of 0.028 -0.038 KPK (0.041)** (0.085)*** 13 Services growth of Punjab Manufacturing growth of 0.0344 -0.0628 Balochistan (0.0288)** (0.0069)*** 14 Services growth of Punjab Services growth of Sindh 0.265 -0.1299 (0.0036)* (0.126) GDP Growth 15 GDP growth of Punjab GDP growth of Sindh 0.229 -0.1613 (0.0006)* (0.024)** *significant at 1 percent level of significance, ** significant at 5 percent level of significance, *** significant at 10 percent level of significance. The value of α in the equation 3 gives the short run error adjustment term. The value of µt is estimated from equation 1 and plug that in equation (3)/Model: 1. 2. µt= Yi.t – (α + βXj.t) 3.
101
Mostly the manufacturing sector of Punjab is based on relatively less capital intensity; as compared to Sindh. However, Punjab is focusing on the development of high value added products for its industrial growth; the dynamics of manufacturing growth in Punjab is discussed by Ahsan Ullah .and Iqbal. J. M (2016). They estimated the growth of manufacturing industries at national and provincial level for the period of 1969 to 2005. To estimate the industrial growth, the authors used two structural variables value added (VA) and average daily employment
(ADE). The average daily employment in Punjab had been increased to 158 percent during the period of 1969 to 2005. There is a decrease in growth rate during the period of 1990 to 2000, as annual growth rate decreased from 2.77 percent in 1990 to 2.42 percent in 2000 and it may be due to crisis in the country during 2001.18
Punjab Industrial Sector Plan, Government of Punjab (GOP) (2014) discussed the low growth of industries in Punjab and developed a policy formwork to increase the industrial base of Punjab and achieve the target of 7 percent growth by 2018 in Gross Regional Product (GRP). The Gross regional Product shows continuous low growth of industrial share in Punjab whereas the share of services increased to 55 percent in GRP of Punjab by 2010-11, agriculture contributes 24 percent and industry contributes 21 percent.19 The industrial products of Punjab are different from
Sindh, only similarities exist in agro based industries in both provinces. Moreover, most of the industrial products of Sindh are exported in international markets rather and also sold in Punjab and vice versa. The speed of adjustment from manufacturing growth of Sindh appears to be low towards Punjab; the very reason may be that most of output from Karachi is exported abroad.
18 there is a continuous expansion of average daily employment recorded in Sindh, as during 1969-2005 the employment expanded to 71.64 percent. The manufacturing of KPK is considerably small in size as compared to Sindh and Punjab. During 2005-06 the industrial employment in KPK was only 6.7 percent of Pakistan. The size of manufacturing is very limited in Balochistan. During the year of 2005 the manufacturing employment was only 2.06 of manufacturing employment of Pakistan for details see Ahsan Ullah .and Iqbal. J . M (2016). 19For details see Punjab Industrial Sector Plan, Government of Punjab (GOP) (2014) 102
Balochistan has very narrow industrial base most of the products from industrial sector of
Balochistan are provided in terms of raw material to the industrial base of Punjab like oil, gas, minerals and metals etc. Thus, the low base of industry in Balochistan does not make significant contribution towards the growth of industry in Punjab. Lahore Chamber of Commerce and
Industry (2016) indicated that the share of Punjab to the national value added of large scale manufacturing decreased slightly from 49.3 percent in 2010-11 and further to 48.9 percent in
2014-15. Punjab is more dominated by small scale manufacturing units as its share in national value added of small scale units is 71 percent. The share of Punjab to national value added has increased from 58 percent in 2010-11 to 75 in 2014-15.
However, with services sector, the case is different as the growth of services sector of Punjab and
Sindh contribute into the growth of manufacturing sector of Punjab. The error correction term is also high as it is 23 percent from Sindh and same percent from Punjab. The high adjustment of services sector growth of Punjab and Sindh to the manufacturing growth of Punjab is due to the development of financial institutions and industrial base; in these two provinces. Mostly the financial institutions and other services are concentrated to these two large provinces. Labour for many services is provided by Punjab to Sindh. The growth of manufacturing sector highly depends upon the growth of services sector; due to its strong forward linkages. Thus, within these two large provinces the rate of adjustment from services growth to manufacturing growth of Punjab is relatively high compared to the manufacturing sector growth; as discussed earlier.
The manufacturing sector of Punjab is an important contributor to the national manufacturing output. However, due to the issues of; energy, crisis, lack of skilled labour and lack of value added manufacturing units in Punjab, the manufacturing growth of Punjab has developed fewer linkages across sectors and across provinces of the country. Moreover, the high dependency of 103
manufacturing growth of Punjab on Sindh is due to these above explained reasons. The government of Punjab is trying to improve investment climate of Punjab and took certain steps.
Hence, one can hope that in future the manufacturing performance of the Punjab may be improved.
4.7.5 Punjab Province: Services Development and Adjustment
The services sector is the largest sector of Pakistan and the Punjab economy. The manufacturing sector of all remaining provinces and services sector of Sindh contributes into the growth of services sector of Punjab. The error correction model shows that 7.2 percent short run disequilibrium in services growth of Punjab is adjusted by manufacturing growth of Sindh, 3.8 percent short run disequilibrium in services growth of Punjab is adjusted by manufacturing growth of KPK and 6.3 percent short run disequilibrium in services growth of Punjab is adjusted by manufacturing growth of Balochistan. The very reason for such linkage is that labour force for KPK found their destiny in Punjab.
It is evident from the above analysis that there are not very strong linkages of services sector of
Punjab with commodity producing sectors, especially, with agriculture of other provinces.
Moreover, within sector, the services growth of Punjab is highly affected by services growth of
Sindh, whereas from services growth of other provinces, the links are missing. Government of
Punjab is trying to improve these linkages and developed a policy framework to improve the service delivery in the province by developing both; human infrastructure and other services institutions as discussed in a report Planning and Development Department, Government of
Punjab (GOP: 2009) formulated a report on economic development of Punjab and accelerating services delivery. The report stated some of the key features of the provincial economy of Punjab and how provincial government is trying to remove the bottlenecks for sustainable development 104
of the province. The demography of Punjab is very important and could give huge benefits to the province. The working population is high in Punjab, as compared to other provinces. This was made possible through providing better infrastructure facilities and training in education and skill development programs.20
The manufacturing sector of all provinces and services sector of Sindh contributes into the growth of services sector of Punjab. The error correction model shows that though the rate of adjustment from manufacturing growth of other provinces is low. It may be due to the fact that manufacturing sectors of other provinces have not yet developed forward linkages with services sector of Punjab. Secondly most of the capital intensive manufacturing units are located in
Punjab and Sindh, which requires huge amount of services for their growth.
KPK and Balochistan are relatively small provinces and they have to develop their services sector base. Infact, the analysis above revealed that these two small provinces are one step back in their manufacturing growth as compared to large provinces. So it may be one of the important reasons that services growth of KPK and Balochistan has low contribution towards the manufacturing development of Punjab.
4.8 Sindh: Cointegration and Result of ECM Sindh is the second largest province of Pakistan in terms of population and in terms of its share in national economy. The total population has increased from 24.4 million in 1990 to 46 million in 2015 that shows an average growth rate of 2.6 per cent per annum. The employment has increased from 6.7 million to 13 million during the same period with an average growth rate of
2.7 per cent per annum. The employment growth is slightly higher than growth of population.
20For details see Planning and Development Department, Government of Punjab (GOP) (2009).
105
The real GDP growth rate was 4.2 per cent, on an average. The shares of agriculture and industry have declined and share of services sector has increased over time.
4.8.1 GDP Growth of Sindh and its Linkages The GDP growth of Sindh is significantly affected by the GDP growth of Punjab in the long run and short run. The error correction term shows that 24 percent short run disequilibrium in GDP growth of Sindh will be adjusted by GDP growth of Punjab (table4.1). Moreover, the GDP growth of Punjab is also effected by GDP growth of Sindh, the error correction term shows that
16 percent short run disequilibrium in GDP growth of Punjab will be adjusted by GDP growth of
Sindh (table4.1). Further, the GDP growth of Sindh also significantly depends on the GDP growth of KPK although the parameter value is small, but significant. The important finding emerges from this analysis that Punjab and Sindh both complement each other’s growth. The good performance of Punjab also has a positive impact on the growth of Sindh. (KPK and
Balochistan)
4.8.2 Sindh: The Integration of Agriculture The table 4.2 shows cointegration results and the short run and long run parameter values of agriculture growth of Sindh with across provinces and across sectors. The result reveals the agriculture sector of Sindh is integrated with agriculture sector, manufacturing sector and services sector of Punjab, this implies that the growth of Punjab is vital for the agriculture sector of Sindh. Further, the agriculture sector of Sindh also depends on the manufacturing and services sectors of Balochistan. It is important to note that the services sector of Sindh is also vital for the
Agriculture growth of Sindh. It shows that the Agriculture growth of Sindh is benefiting from the growth of other provinces more than the growth of its own sectors. It may be due to integration of agriculture, industry and financial sector.
106
Table 4.2: Cointegration Results for Sindh
Equ- Dependent Independent Long run Error Ations parameters correction parameters Agriculture Sector 1 Agriculture growth of Agriculture growth of Punjab 0.546 -0.39 Sindh (0.0001)* (0.0021)* 2 Agriculture growth of Manufacturing growth of Punjab 0.129 -0.14 Sindh (0.036)** (0.4)** 3 Agriculture growth of Manufacturing growth of 0.035 -0.115 Sindh Balochistan (0.077)*** (0.076)*** 4 Agriculture growth of Services growth of Punjab 0.2167 -0.346 Sindh (0.0047)* (0.0009)* 5 Agriculture growth of Services growth of Sindh 0.195 -0.259 Sindh (0.0127)* (0.0085)* 6 Agriculture growth of Services growth of KPK 0.143 -0.21 Sindh (0.035)** (0.03)** 7 Agriculture growth of Services growth of Balochistan 0.31 -0.34 Sindh (0.009)* (0.005)* Industrial Sector 8 Manufacturing Manufacturing growth of Punjab 0.25 -0.28 growth of Sindh (0.029)** (0.019)** 9 Manufacturing Manufacturing growth of KPK 0.055 -0.134 growth of Sindh (0.10)*** (0.059)*** 10 Manufacturing Manufacturing growth of 0.056 -0.179 growth of Sindh Balochistan (0.1)*** (0.025)** 11 Manufacturing Services growth of Punjab 0.29 -0.265 growth of Sindh (0.018)** (0.018)** 12 Manufacturing Services growth of Sindh 0.356 -0.289 growth of Sindh (0.024)** (0.013)** Services Sector 13 Services growth of Manufacturing growth of Punjab 0.217 -0.253 Sindh (0.002)* (0.011)* 14 Services growth of Manufacturing growth of Sindh 0.16 -0.155 Sindh (0.0000)* (0.0017)* 15 Services growth of Manufacturing growth of KPK 0.027 -0.0498 Sindh (0.095)* (0.10)*** 16 Services growth of Services growth of Punjab 0.34 -0.26 Sindh (0.0006)* (0.026)** GDP Growth 17 GDP growth of Sindh GDP growth of Punjab 0.33 -0.24 (0.01)* (0.04)** 18 GDP growth of Sindh GDP growth of KPK 0.043 -0.04 (0.66) (0.69)
*significant at 1 percent level of significance, ** significant at 5 percent level of significance, *** significant at 10 percent level of significance. 107
The error correction term shows that 39 percent, 14 percent and 25 percent short run disequilibrium in agriculture growth of Sindh is adjusted by agriculture growth, manufacturing growth and services sector growth of Punjab, respectively. About 12 percent and 34 percent disequilibrium in agriculture growth of Sindh, in short run, will be adjusted by manufacturing and services sectors of Balochistan. Further, the result shows that 26 percent short disequilibrium in agriculture sector of Sindh will be adjusted by services growth of Sindh, 21 percent short disequilibrium in agriculture sector of Sindh will be adjusted by services growth of KPK.
It is evident from the above analysis that like in Punjab, the agriculture sector of Sindh has failed to create forward linkages with manufacturing and services sectors of the provincial economies.
However, the backward linkages of the manufacturing and especially, the services sector have been developed towards the agriculture growth of Sindh. Whereas within agriculture sector it is only the growth of agriculture in Punjab that effects agriculture growth of Sindh and the possible reason is common cropping patterns of both the economies as discussed earlier in case of Punjab.
4.8.3 Industrial Development and Linkages of Sindh The manufacturing sector of Sindh is co-integrated with the manufacturing and services sector of
Punjab, manufacturing sectors of KPK and Balochistan and services sector of Sindh. The error correction model illustrates that 28 percent and 26 percent disequilibrium in manufacturing sector Sindh will be adjusted by manufacturing growth and services sector growth of Punjab, 13 percent disequilibrium in manufacturing sector Sindh will be adjusted by manufacturing growth of KPK and 18 percent disequilibrium in manufacturing sector Sindh will be adjusted by manufacturing growth of Balochistan. The error correction coefficients show that 29 percent disequilibrium in manufacturing sector of Sindh is adjusted by services sector of Sindh.
108
The result reveals that the speed of adjustment towards any disequilibrium in manufacturing sector of Sindh is fast with the manufacturing sector of Punjab, whereas the speed of adjustment towards any disequilibrium in manufacturing sector of Sindh is slow from the manufacturing sector of KPK and Balochistan, which is as expected. These are the small provinces and industrialisation is very limited there.
The industrial development of Pakistan was highly concentrated to Sindh in the beginning. This is one of the reasons that over the period of time the industrial sector of Sindh has developed its linkages with the manufacturing growth of other provinces. The high growth of manufacturing in
Sindh has affected the province badly in terms of high urbanisation problems. Secondly, most of the industry in Sindh is present in urban sector thus the rural and urban disparities have also emerged to higher extent in Sindh. Thirdly, within the urban sector there are disparities as most of the industry in Sindh is present in Karachi. The share of Karachi in provincial economy was
54 percent and mainly it was due to port and high investments in major industrial sector and sub sectors like food, tobacco and textiles, this eventually led to the polarisation of Karachi within provinces during 70s. So the city is suffering from huge problems in its social indicators as to cater the huge number of population, secondly the disparities within the urban sector of Sindh have also exploded. The industrial development of Sindh was discussed by Husain. I (2014) discussed that the annual population growth of Sindh during 1961-71 was 4.6 percent. During
1972 Sindh had 21.5 percent of total population of Pakistan and Karachi is having 3.6 million people, which was one fourth of the provincial population. The labour force was 4.1 million and area under crop was 3.6 million hectors during early 70s.21
21The province have low rate of literacy which is one of the serious bottleneck for provincial development. The agriculture sector contributes 17 percent to Gross Provincial Product (GPP), manufacturing contributes 36 percent 109
Sindh State of Environment and Development, IUCN, (2004) Sindh contributes high in textiles as 60 percent of the country’s textile industry is located in Sindh. Recently due to deterioration in law and order situation and political crisis, many of the industrial units are closed by the investors of Sindh. The structural adjustment and privatisation under WTO regime have also effected the growth of industry badly.22
Mehmood. K and Ahmed.R (2012) discuss the Spatio-temporal change in Sindh. The authors used the data of 1981 to 1998 for analysis. The authors found that during 1981-98 the economy of Sindh has developed rapidly but inequitably. As most of the urban centres were the focus of development, the rural sector districts received less of the fruits of development. The authors that policy planners should focus on the development of rural sector of Sindh and further growth should be equitably distributed between districts of the provinces. It will not only reduce disparity but also serves the needs of inclusive growth in the province.23
For the development of intraprovincial linkages in manufacturing growth of Sindh it is important to establish significant amount of linkage of manufacturing growth of KPK and Balochistan with
Sindh. This could be done by establishing similar level of industrial base in the provinces and secondly the key inputs to manufacturing of Sindh are to be provided by manufacturing units of
KPK and Balochistan. The results of services growth to the manufacturing growth of Sindh are similar as they were in case of manufacturing of Punjab. Since the concentration of financial institutions are large in Sindh and Punjab.
to GPP and services contribute 47 percent to GPP. Sindh is still having the large industrial base compared to other provinces see for details Husain .I (2014). 22see for details Sindh State of Environment and Development, IUCN, (2004) 23see for details Mehmood. K and Ahmed.R (2012) 110
Finally, the result reveals that the speed of adjustment towards any disequilibrium in manufacturing sector of Sindh is fast with the manufacturing sector of Punjab. The speed of adjustment towards any disequilibrium in manufacturing sector of Sindh is slow from the manufacturing sector of KPK and Balochistan. In case of services growth, the speed of adjustment is fast for Sindh, as compared to Punjab.
The above analysis indicates the growth in manufacturing units of Sindh has made significant contribution to the manufacturing output of the country. Moreover, the manufacturing growth of
Sindh has performed better than Punjab, as it has developed most of its linkages with the manufacturing growth of Punjab and other small provinces such as KPK and Balochistan. Hence the small provinces KPK and Balochistan depend highly in their own manufacturing growth on the manufacturing growth of Sindh.
4.8.4 Sindh: The Integration of the Services Development The long run and short run parameters of services sector growth of Sindh is also given in table
4.2. Services sector growth of Sindh and manufacturing sector growth of Punjab, Sindh and KPK are significant, thus showing a strong relationship. The error correction model shows that 25 percent short run disequilibrium in service growth of Sindh will be adjusted by manufacturing growth of Punjab. Similarly, 15 percent short run disequilibrium in service growth of Sindh will be adjusted by manufacturing growth of Sindh and 5 percent short run disequilibrium in service growth of Sindh will be adjusted by manufacturing growth of KPK. So with manufacturing growth of Punjab the speed of adjustment towards equilibrium in services growth of Sindh is fast, followed by manufacturing growth of Sindh and with manufacturing growth of KPK the speed of adjustment towards equilibrium in services growth of Sindh is sluggish. The error correction model also shows that 26 percent short run disequilibrium in services growth of Sindh 111
is adjusted by services growth of Punjab. This shows that services growth of Punjab has a significant impact on the services growth of Sindh. Any short run deviation from equilibrium in services growth of Sindh is quickly adjusted by services growth of Punjab, which shows integration between these two provinces.
The services sector of Sindh though has created some of the backward linkages with the commodity producing sector of the provinces especially with the manufacturing sector of the province of Punjab. The respectable development of services sector and its linkages with manufacturing across provinces, particularly with Punjab is due to the high concentration of services growth in these provinces. Secondly, being the second largest province Sindh has attracted human resources, from rest of the country. The largest city is Karachi and port are the source of such attractions.
The above analysis indicates that Sindh has improved its services sector significantly; especially towards the manufacturing industries. So the linkages of manufacturing growth with services growth across provinces are highly integrated. Moreover, the provincial government has also developed its human infrastructure, as the human indicators show that for human development the provincial government of Sindh has implemented plans in certain areas and achieved good goals. Still it is far away to turn its human resources into human capital. Its literacy rate is hardly
60% and mass unemployment, as well as, urban ghetto areas are calling for a special program to improve the economic conditions in Sindh. There are disparities in human services, especially, between rural and urban sector of Sindh. Thus, the provincial government has to follow more focused plans towards the improvement of social indicators in such a manner that improvement
112
should not be polarized or create disparities; rather the improvement in social indicators should take into account all segments of the rural and urban population of the province.
4.9 KPK Province: Empirical Results: Cointegration and ECM
4.9.1 Economic Profile of KPK The KPK is the third largest province of Pakistan, in terms of population and its share in national economy. The total population has increased from 14 million in 1990 to 26 million in 2015 that shows an average growth rate of 2.4 per cent per annum. The employment has increased from
3.3 million to 5.9 million during the same period. The real GDP at constant factor cost of 2005-
06 has increased from PKRs 0.4 trillion in 1990 to PKRs 1.2 trillion in 2015; growth rate of 4.9 per cent on an average per annum. The share of agriculture has declined from 26.3 percent in
1990 to 15.7 per cent in 2015. The share of industry has increased from 20.7 percent to 28.6 percent and the share of services has increased from 52.9 percent to 55.8 percent in above mentioned period. This section presents the cointegration and error correction model results for
KPK. The equation 1 to 25 in the table 4.3 give the long run and short run parameters for KPK province.
4.9.2 GDP Growth of KPK and its Linkages Overall, the result shows that the GDP growth of KPK is dependent on the GDP growth of Punjab and
Sindh. The Long run parameters and error correction terms are significant. The ECM shows 35 percent and 25 percent variation in short run equilibrium of GDP growth of KPK is adjusted by the Sindh and
Punjab, respectively (T-4.3).
113
Table 4.3 A: Cointegration Results for KPK Equ- Dependent Independent Long run Error ation# parameters correction parameters Agriculture Sector 1 Agriculture growth of KPK Agriculture growth of 0.338 -0.386 Punjab (0.0016)* (0.0018)* 2 Agriculture growth of KPK Agriculture growth of 0.312 -0.44 Sindh (0.0000)* (0.0003)* 3 Agriculture growth of KPK Manufacturing growth of 0.15 -0.3004 Punjab (0.006)* (0.0081)* 4 Agriculture growth of KPK Manufacturing growth of 0.1008 -0.19 Sindh (0.0184)** (0.018)* 5 Agriculture growth of KPK Services growth of 0.217 -0.374 Punjab (0.006)* (0.001)* 6 Agriculture growth of KPK Services growth of Sindh 0.214 -0.316 (0.0014)* (0.0027)* 7 Agriculture growth of KPK Services growth of KPK 0.199 -0.35 (0.001)* (0.0012)* 8 Agriculture growth of KPK Services growth of 0.326 -0.33 Balochistan (0.006)* (0.0049)* Industrial Sector 9 Manufacturing growth of Agriculture growth of 0.64 -0.261 KPK Punjab (0.009)* (0.011)* 10 Manufacturing growth of Agriculture growth of 0.34 -0.181 KPK Sindh (0.069)*** (0.045)** 11 Manufacturing growth of Agriculture growth of 0.556 -0.24 KPK KPK (0.03)** (0.016)** 12 Manufacturing growth of Manufacturing growth of 0.354 -0.273 KPK Punjab (0.02)** (0.018)** 13 Manufacturing growth of Manufacturing growth of 0.379 -0.274 KPK Sindh (0.03)** (0.02)** 14 Manufacturing growth of Manufacturing growth of 0.147 -0.256 KPK Balochistan (0.006)* (0.002)* 15 Manufacturing growth of Services growth of 0.48 0.294 KPK Punjab (0.008)* (0.008)* 16 Manufacturing growth of Services growth of Sindh 0.45 -0.26 KPK (0.019)** (0.0183)** 17 Manufacturing growth of Services growth of KPK 0.35 -0.25 KPK (0.046)** (0.025)** 18 Manufacturing growth of Services growth of 0.504 -0.281 KPK Balochistan (0.013)* (0.002)*
114
Table 4.3 B: Cointegration Results for KPK Services Sector 19 Services growth of KPK Manufacturing growth of 0.151 -0.211 Punjab (0.0061)* (0.004)* 20 Services growth of KPK Manufacturing growth of 0.116 -0.143 Sindh (0.002)* (0.001)* 21 Services growth of KPK Manufacturing growth of 0.048 -0.073 KPK (0.0067)* (0.0169)** 22 Services growth of KPK Services growth of 0.165 -0.128 Punjab (0.021)** (0.094)*** 23 Services growth of KPK Services growth of Sindh 0.188 -0.21 (0.023)** (0.009)* GDP Growth 24 GDP growth of KPK GDP growth of Punjab 0.31 -0.25 (0.003)* (0.03)** 25 GDP growth of KPK GDP growth of Sindh 0.31 -0.35 (0.002)* (0.0009)* *significant at 1 percent level of significance, ** significant at 5 percent level of significance, *** significant at 10 percent level of significance
4.9.3 KPK Province: Co-integration of Agriculture and Its Linkages The agriculture sector growth of KPK is integrated with agriculture sector growth of Punjab and
Sindh. Table 4.3 shows the long run and short run values of agriculture sector of KPK with other provinces. The results reveal that there is a significant long run and short run relationship between agriculture growth of KPK and agriculture growth of Punjab and Sindh. The error correction coefficient shows that 39 percent short run disequilibrium in agriculture growth of
KPK is adjusted by agriculture growth of Punjab, 44 percent short run disequilibrium in agriculture growth of KPK is adjusted by agriculture growth of Sindh (Table 4.3).
Further, the agriculture sector growth of KPK is integrated with the manufacturing sector growth of Punjab, Sindh and KPK. There is a significant long run and short run relationship between agriculture growth of KPK and manufacturing growth of Punjab and Sindh. The error correction model shows that 30 percent disequilibrium in agriculture growth of KPK in short run is adjusted 115
by manufacturing of Punjab and 19 percent short run disequilibrium in agriculture growth of
KPK is adjusted by manufacturing sector of Sindh.
The agriculture growth of KPK is integrated with the services sector growth of Punjab, Sindh,
KPK and Balochistan. There is a significant long run and short run relationship between agriculture growth of KPK and services growth of all provinces. The error correction model shows that 37 percent short run disequilibrium in agriculture sector of KPK is adjusted by services sector of Punjab, 32 percent short run disequilibrium in agriculture sector of KPK is adjusted by services sector of Sindh, 35 percent short run disequilibrium in agriculture sector of
KPK is adjusted by services sector of KPK and 33 percent short run disequilibrium in agriculture sector of KPK is adjusted by services sector of Balochistan. The results show that the services sector growth effects the growth of agriculture more than the industry.
The agriculture growth of KPK is significantly affected by agriculture growth of Punjab and
Sindh and further from the manufacturing growth of Punjab and Sindh. Moreover, the agriculture growth of KPK is significantly affected by services growth of all provinces. This shows that agriculture sector growth of KPK is very low and is highly dependent on the other sectors of other provinces.
The agriculture growth of KPK has not developed its forward linkages with manufacturing and services sector of other provinces. Contrarily the performance of agriculture sector is so low in
KPK that the growth of agriculture in the province largely depends upon the agriculture growth of Punjab and Sindh. Moreover, in manufacturing it is the same issue that the manufacturing of
Punjab and Sindh has created backward linkages with agriculture growth of KPK. However, in services the issue is entirely different; the growth of agriculture in KPK is largely dependent on
116
the services of all three provinces and KPK itself thus, highlighting the fact that services growth of all provinces has created backward linkages with agricultural development of KPK.
The above analysis reveals the poor performance of agriculture sector of KPK and its high dependence on the other sectors across provinces. It also highlights the vulnerability of agriculture growth of KPK, as it depends highly on the growth of agriculture, manufacturing and services growth of large provinces. So to provide a sustainable pattern of agriculture growth in
KPK policies for the development of the sector within the province are to be worked out and implemented properly with time.
4.9.4 KPK Province: Industrial Growth and Its Linkages The short run and long run parameters of manufacturing growth of KPK across sectors and across provinces have been estimated. The results are reported in the above table (4.3). The long run and short run parameters of manufacturing growth of KPK and agriculture growth of Punjab,
Sindh and KPK are significant showing a strong relationship of manufacturing growth of KPK and agriculture growth of Punjab, Sindh and KPK. The error correction term shows that 26 percent short run disequilibrium in manufacturing sector of KPK is adjusted by agriculture sector of Punjab, 18 percent short run disequilibrium in manufacturing sector of KPK is adjusted by agriculture sector of Sindh and 24 percent short run disequilibrium in manufacturing sector of
KPK is adjusted by agriculture sector of KPK. These results highlight the importance of agriculture sector towards manufacturing sector growth of KPK, further it also shows that the manufacturing sector KPK is highly dependent on the agriculture sector of other provinces.
Further the long run and short run parameters of manufacturing growth of KPK and manufacturing growth of Punjab Sindh and Balochistan are also significant showing a strong
117
relationship between manufacturing growth of KPK and manufacturing growth of Punjab Sindh and Balochistan. The error correction term shows that 27 percent short run disequilibrium in manufacturing sector of KPK is adjusted by manufacturing growth of Punjab, 27 percent short run disequilibrium in manufacturing sector of KPK is adjusted by manufacturing growth of
Sindh and 26 percent short run disequilibrium in manufacturing sector of KPK is adjusted by manufacturing growth of Balochistan. The manufacturing growth of KPK highlights the fact that manufacturing growth of KPK is also highly dependent on the manufacturing growth of other provinces and if any shock happened to manufacturing growth of KPK it will be adjusted quickly by the manufacturing growth of other provinces.
The long run and short run parameters of manufacturing growth of KPK and services growth of
Punjab, Sindh, KPK and Balochistan shows significant relationship between manufacturing growth of KPK and services growth of Punjab, Sindh, KPK and Balochistan. The error correction term shows that 29 percent disequilibrium in short run in manufacturing sector of
KPK will be adjusted by services growth of Punjab, 26 percent disequilibrium in short run in manufacturing sector of KPK will be adjusted by services growth of Sindh, 25 percent disequilibrium in short run in manufacturing sector of KPK will be adjusted by services growth of KPK and 28 percent disequilibrium in short run in manufacturing sector of KPK will be adjusted by services growth of Balochistan. The manufacturing sector of KPK is highly integrated with the services growth of other provinces and KPK itself.
The manufacturing growth of KPK was hardly a priority of the government, although the province has comparative advantage over other provinces in minerals and metals. Most of the industry in KPK is still small sized or medium sized and is low value added. The high
118
dependency of manufacturing growth of KPK is now being recognized by the federal and provincial government especially and certain policies are formulated to develop industry in KPK as discussed in Industrial Policy KPK (2016). The province is enriched of natural resources that can attract high amount of investment. Despite of all this, the level of investment is very low in the province as only there are only 12000 small, medium and large industrial units in the province24.
The above analysis indicates that mostly industrial units of KPK are small or medium sized. So the manufacturing growth of KPK highlights the fact that manufacturing growth of KPK is highly dependent on the manufacturing growth of other provinces especially the large provinces
Punjab ad Sindh. If any shock happened to manufacturing growth of KPK it will be adjusted quickly by the manufacturing growth of other provinces. The high dependency of manufacturing growth of KPK to the manufacturing growth of other provinces shows two important facts.
Firstly, the industrial base of KPK itself is very low and suffers from huge problems in its establishment as discussed preciously. Secondly, the established manufacturing units in KPK take high amount of inputs for their growth from the manufacturing units of other provinces.
Thus the vulnerability of manufacturing growth of KPK is high in case of its high dependence on the manufacturing growth of other provinces. The error correction term shows that the speed of adjustment from services growth of all provinces is high towards the short term disequilibrium in manufacturing growth of KPK. The manufacturing sector of KPK is highly dependent on the services growth of large provinces.
24see for details Industrial Policy KPK (2016)
119
Hence, if the true potential of the province, according to its comparative advantage over other provinces, is explored and manufacturing units including the large scale manufacturing units, especially in hydro electricity production are developed, it could increase the growth of the province and eventually the integration level of the manufacturing growth of KPK with other provinces.
4.9.4 KPK Province: Services Sector and Its Linkages The co-integration results of services sector growth of KPK across sectors and across provinces show that the services sector growth of KPK is integrated with manufacturing sector growth of
Punjab, Sindh and KPK. Further, the services sector growth of KPK is only integrated with agriculture sector of KPK. The services sector growth of KPK further is integrated with services sector growth of Punjab and Sindh.
The long and short run parameters of services sector growth of KPK are integrated with manufacturing sector growth of Punjab, Sindh and KPK as the relation in long run relationship.
The error correction model shows that 21 percent disequilibrium in services sector growth of
KPK is adjusted by manufacturing growth of Punjab, and 14 percent disequilibrium in services sector growth of KPK is adjusted by manufacturing growth of Sindh and 7.3 percent disequilibrium in services sector growth of KPK is adjusted by its own manufacturing growth of
KPK. This implies that the manufacturing sector of Punjab adjust the disequilibrium in services sector of KPK at a higher speed followed by Sindh, however the speed of adjustment of services sector disequilibrium in KPK from manufacturing sector of KPK is sluggish.
The long run and short run parameters of services sector growth of KPK with services sector growth of Punjab and Sindh are significant showing a strong relationship. The error correction
120
model shows that 13 percent short run disequilibrium in services sector of KPK is adjusted by services sector growth of Punjab and 21 percent short run disequilibrium in services sector of
KPK is adjusted by services sector growth of Sindh. The research reports indicate the important facts regarding the services development in KPK. The development of the services sector in
KPK is analyzed by Department for International Development (DIFD) (2005) stated that the economic report of NWFP and the service delivery to the province. The report argues that the extension services and research services have failed to deliver in the agriculture sector growth of
KPK. The new technological services rarely benefit the traditional setup of agriculture system.
The small farmers particularly ignored from the availability of these services. The existing institutions remained poor in terms of their role in development.25
It was found that in services sector growth of KPK; the service growth of KPK depend on the manufacturing sector growth of large provinces. The services sector development of KPK depends on manufacturing growth of KPK but the rate of adjustment from manufacturing growth of KPK to services growth of KPK is low. Thus, the manufacturing sector of large provinces
Punjab and Sindh have developed strong forward linkages with services sector growth of KPK.
However, the manufacturing growth of KPK has created low level of forward linkages with services growth of KPK. Moreover, within the services sector the linkages of KPK have been developed with large provinces Punjab and Sindh rather than KPK itself or with Balochistan.
This highlights the important fact that services growth of KPK largely depends on the manufacturing and services growth of the large provinces i.e Punjab and Sindh.
25See for details Department for International Development (DIFD) (2005) 121
The above analysis shows that services sector development is very low in KPK, due to which the province has not developed significant amount of relationships with other sectors of the economy. If the services sector of the KPK is developed according to the potential of the province, it will eventually benefit the provincial growth and national growth.
4.10 Empirical Findings for Balochistan: Cointegration and ECM The Balochistan is the largest province of Pakistan, in terms of area and the smallest in terms of population and its share in national economy. The total population of the province has increased from 5.4 million in 1995 to 9.9 million in 2015, which shows an average growth rate of 2.5 per cent per annum. The employment has increased from 1.1 million to 2.9 million during the same period with an average growth rate of 4.0 per cent per annum. The real GDP at constant factor cost of 2005-06 has increased from PRs 236 billion in 1990 to PRs 487 billion in 2015; growth rate of 2.9 percent, on an average. The share of agriculture has declined from 33 percent in 1990 to 27.6 per cent in 2015. The share of industry of Balochistan in GDP has increased from 22 per cent to 25 percent and the share of services has increased from 44.8 per cent to 47.4 per cent in above mentioned period.
4.10.1 GDP Growth of Balochistan and its Linkages The long run and short run dynamics of GDP growth of Balochistan. It is important to note that like KPK the GDP growth of Balochistan is also depends on the GDP growth of Punjab and
Sindh. The GDP growth of Punjab affect Balochistan significantly in short run and long run as
22 percent short run disequilibrium in GDP growth of Balochistan is adjusted by GDP growth of
Punjab shown in below table 4.6. The GDP growth of Sindh affect Balochistan significantly in long run and short run as 21 percent disequilibrium in GDP growth of Balochistan in short run will be adjusted by GDP growth of Sindh shown in below table 4.6.
122
4.10.2 Balochistan Province: Empirical Evidences; Linkages of Agriculture Sector
The empirical results show that the agricultures sector growth of Balochistan is integrated with the agriculture sector growth of all provinces. Table 4.4 shows that there is a significant long run and short run relationship between agriculture growths of Balochistan with the agriculture growth of all other provinces. The error correction model shows that 44 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by agriculture growth of Sindh, 25 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by agriculture growth of KPK and 29 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by agriculture growth of Punjab.
The results indicated a significant long run and short run relationship between agriculture growth of Balochistan and manufacturing growth of Punjab, Sindh, KPK and Balochistan (T-4.4). The error correction model shows that 24 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by manufacturing growth of Punjab, 18 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by manufacturing growth of Sindh, 14 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by manufacturing growth of KPK and 15 percent disequilibrium in agriculture growth of
Balochistan in short run is adjusted by manufacturing growth of Balochistan.
123
Table 4.4: Cointegration Results for Balochistan Dependent Independent Long run Error parameters correction parameters Equation Agriculture Sector 1 Agriculture growth of Agriculture growth of 0.248 -0.44 Balochistan Sindh (0.015)** (0.009)* 2 Agriculture growth of Agriculture growth of 0.241 -0.248 Balochistan KPK (0.046)** (0.014)* 3 Agriculture growth of Agriculture growth of 0.339 -0.29 Balochistan Punjab (0.0093)* (0.011)* 4 Agriculture growth of Manufacturing growth of 0.133 -0.236 Balochistan Punjab (0.017)** (0.02)** 5 Agriculture growth of Manufacturing growth of 0.096 -0.184 Balochistan Sindh (0.06)** (0.017)** 6 Agriculture growth of Manufacturing growth of 0.068 -0.143 Balochistan KPK (0.044)** (0.068)*** 7 Agriculture growth of Manufacturing growth of 0.043 -0.146 Balochistan Boluchistan (0.049)** (0.049)** 8 Agriculture growth of Services growth of 0.161 -0.239 Balochistan Punjab (0.019)** (0.021)** 9 Agriculture growth of Services growth of KPK 0.139 -0.212 Balochistan (0.033)** (0.029)** 10 Agriculture growth of Services growth of 0.262 -0.295 Balochistan Balochistan (0.015)** (0.007)* 11 Agriculture growth of Services growth of Sindh 0.173 -0.269 Balochistan (0.019)** (0.007)* *significant at 1 percent level of significance, ** significant at 5 percent level of significance, *** significant at 10 percent level of significance.
The table 4.4 above shows that the agriculture sector growth of Balochistan is integrated with services sector growth of Balochistan itself and services growth of all other three provinces.
There exists a significant long run and short run relationship between agriculture growth of
Balochistan and services growth of Punjab, Sindh, KPK and Balochistan. The error correction model shows that 24 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by services growth of Punjab, 21 percent disequilibrium in agriculture growth of
Balochistan in short run is adjusted by services growth of KPK, 29 percent disequilibrium in
124
agriculture growth of Balochistan in short run is adjusted by services growth of Balochistan and
27 percent disequilibrium in agriculture growth of Balochistan in short run is adjusted by services growth of Sindh.
Being the small province, the agriculture growth of Balochistan is highly dependent upon the agriculture growth, manufacturing growth and services growth of other provinces. The high dependency of Balochistan on different sectors of other provinces is discussed as by the World
Bank (2005). Besides, Thapa. B. G. and Baloch. A. M. (2017), the authors discussed that the decentralised policies did not work properly in Balochistan, the extension policies and practises work differently in case of Balochistan, as they followed the traditional top to bottom approach in practise that benefited less to the province. Most of the agricultural development policies of
Pakistan are devoted towards main exports crops like cotton, sugarcane, rice, wheat and cereals.
It mostly benefits Punjab and Sindh, whereas the province, like Balochistan where dates and palm have comparative advantage are not the main focus of the extension services of ministry of agriculture.
The above analysis revealed that there is a lot of potential in the agriculture growth of
Balochistan, provided if this hidden potential is properly explored by the government. As if fruit processing units and crop specific zones are established in the province, it will give huge benefits to the agriculture growth of the province and can increase the exports of the country in terms of fruits also. If government manage irrigation, water logging and salinity issues properly and efficiently it will eventually benefit the agriculture growth of provincial economy. All this will reduce the high dependency of agriculture growth in Balochistan over other provinces.
125
4.10.3 Industrial Sector of Balochistan: Linkages and Dependency The results of long run and short run parameters of manufacturing sector growth of Balochistan and agriculture growth of Sindh, KPK and Punjab are given in Table 4.5 below shows almost all long run and short run parameters are significant, showing a strong relationship of the manufacturing sector’s growth of Balochistan and agriculture growth of Sindh, KPK and Punjab.
The error correction term show that 17 percent short run disequilibrium in manufacturing sector of Balochistan is adjusted by agriculture growth of Sindh, 15 percent short run disequilibrium in manufacturing sector of Balochistan is adjusted by agriculture growth of KPK and 15 percent short run disequilibrium in manufacturing sector of Balochistan is adjusted by agriculture growth of Punjab. The results show that the short run disequilibrium in manufacturing sector of
Balochistan is adjusted by agriculture growth of other provinces but the speed of adjustment is slow.
The significance of short run parameter shows that in short run disequilibrium between both the series will be adjusted at higher rate. The error correction term shows that 28 percent short run disequilibrium in manufacturing growth of Balochistan is adjusted by manufacturing growth
Punjab and 22 percent short run disequilibrium in manufacturing sector of Balochistan is adjusted by manufacturing growth of Sindh. The manufacturing sectors and agriculture sectors of both large provinces are significantly affecting the manufacturing growth of Balochistan.
However, the speed of convergence towards equilibrium in manufacturing sector of Balochistan is fast with manufacturing growth of large provinces than the agriculture growth of these provinces. It is important to note that the services sector of provinces does not affect the manufacturing growth of Balochistan. This is maybe due to the fact that the industrial base of
Balochistan is very small.
126
Table 4.5: Cointegration Results for Balochistan Dependent Independent Long run Error parameters correction parameters Equations Industrial Sector 1 Manufacturing growth of Agriculture growth of 0.44 -0.168 Balochistan Sindh (0.09)*** (0.049)** 2 Manufacturing growth of Agriculture growth of 0.515 -0.155 Balochistan KPK (0.09)*** (0.05)** 3 Manufacturing growth of Agriculture growth of 0.707 -0.149 Balochistan Punjab (0.046)** (0.095)*** 4 Manufacturing growth of Manufacturing growth of 0.7001 -0.289 Balochistan Punjab (0.010)* (0.0096)* 5 Manufacturing growth of Manufacturing growth of 0.7745 -0.235 Balochistan Sindh (0.009)* (0.044)** *significant at 1 percent level of significance, ** significant at 5 percent level of significance, *** significant at 10 percent level of significance.
The manufacturing growth of large provinces significantly affect the short term disequilibrium to the manufacturing of Balochistan. The obvious reason is the lack of industry in Balochistan especially the large scale industry. Since mostly manufacturing units in Balochistan are small scale so a short term shock to these units are adjusted by the manufacturing support of the large provinces.
Most importantly the manufacturing of Balochistan is not affected by services growth of any of the province. It is basically due to two important factors. Firstly, the services sector of small provinces KPK and Balochistan are not well established thus having no contribution to the manufacturing growth of Balochistan. Secondly, the services sector of large provinces Punjab and Sindh are established in such a manner as they help to those industries from where the rate of return is higher. Since in Balochistan as discussed earlier the industries are small scale or cottage industries so due to their low rate of return the services sector of larger provinces rarely contributes towards the growth of manufacturing sector of Balochistan.
127
The above analysis indicates that the potential of Balochistan in manufacturing sector should by explored. If it need to be done seriously by the provincial and federal government the manufacturing of Balochistan could contribute significantly to the overall growth of the country and the provincial growth of the country.
4.10.4 Services Sector of Balochistan: Integration and Linkages The results in table 4.6 below show that the services sector growth of Balochistan is integrated with the manufacturing growth of Punjab, Sindh and KPK and services sector growth of
Balochistan. The services sector growth of Balochistan is integrated with the agriculture growth of Balochistan itself and with the agriculture growth of all other three provinces. The services sector growth of Balochistan is integrated with services sector growth of Punjab, Sindh and
KPK.
The long run and short run parameters of services growth of Balochistan with agriculture growth of Punjab, Sindh and KPK are significant thus showing their integrated growth. The error correction model shows that 36 percent short run disequilibrium in services growth of
Balochistan is adjusted by agriculture growth of Punjab, 36 percent short run disequilibrium in services growth of Balochistan is adjusted by agriculture growth of Sindh and 32 percent short run disequilibrium in services growth of Balochistan is adjusted by agriculture growth of KPK. It means that services sector of Balochistan is significantly and strongly affected by the agriculture sectors growth of other provinces and the short term adjustment to disequilibrium in services sector of Balochistan is speedily adjusted by the agriculture sectors of other provinces.
The long run and short run parameters of services growth of Balochistan with manufacturing across sectors and across provinces. The long run and short run parameters of services growth of
Balochistan with manufacturing growth of Punjab and Sindh are significant showing a strong
128
relationship between services growth of Balochistan and manufacturing growth of Punjab and
Sindh. However, the parameters services growth of Balochistan with manufacturing growth of
KPK are insignificant but there exits integration between both series. The error correction model states that 34 percent short run disequilibrium in services sector of Balochistan is adjusted by manufacturing sector of Punjab, 20 percent short run disequilibrium in services sector of
Balochistan is adjusted by manufacturing sector of Sindh. So similar to agriculture sector the manufacturing sector of Punjab and Sindh significantly affects the services sector of Balochistan and the speed of convergence towards equilibrium in services sector of Balochistan is fast with manufacturing sector of Punjab followed by Sindh.
The long and short run parameters of services growth of Balochistan and services growth of
Punjab, Sindh and KPK are significant showing a strong relationship between services growth of
Balochistan and services growth of Punjab, Sindh and KPK. The error correction model states that 42 percent short run disequilibrium in servicers sector of Balochistan is adjusted by services sector growth of Punjab, 32 percent short run disequilibrium in servicers sector of Balochistan is adjusted by services sector growth of Sindh and 34 percent short run disequilibrium in servicers sector of Balochistan is adjusted by services sector growth of KPK. It means even the services sector of Balochistan is significantly affected by the services sectors of other provinces as the speed of adjustment towards equilibrium in services sector of Balochistan is much fast with services sector of Punjab followed by KPK and Sindh.
129
Table 4.6: Cointegration Results for Balochistan Dependent Independent Long run Error parameters correction parameters Equations Services Sector 17 Services growth of Agriculture growth of 0.325 -0.359 Balochistan Punjab (0.049)** (0.015)** 18 Services growth of Agriculture growth of 0.293 -0.363 Balochistan Sindh (0.0075)* (0.0035)* 19 Services growth of Agriculture growth of 0.41 -0.318 Balochistan KPK (0.0002)* (0.0019)* 20 Services growth of Manufacturing growth of 0.1819 -0.34 Balochistan Punjab (0.0092)* (0.006)* 21 Services growth of Manufacturing growth of 0.119 -0.2 Balochistan Sindh (0.018)** (0.015)** 22 Services growth of Services growth of 0.278 -0.423 Balochistan Punjab (0.004)* (0.003)* 23 Services growth of Services growth of Sindh 0.25 -0.324 Balochistan (0.008)* (0.009)* 24 Services growth of Services growth of KPK 0.188 -0.336 Balochistan (0.006)* (0.0018)* GDP Growth 25 GDP growth of Balochistan GDP growth of Punjab 0.17 -0.22 (0.04)** (0.03)** 26 GDP growth of Balochistan GDP growth of Sindh 0.148 -0.21 (0.03)** (0.02)** *significant at 1 percent level of significance, ** significant at 5 percent level of significance, *** significant at 10 percent level of significance.
The above analysis indicates the high dependency of services sector growth of Balochistan on commodity producing sectors and services sector of other provinces. Moreover, the high dependency of services sector growth of Balochistan on agriculture growth of other provinces reveals the poor development of services sector in the province. The other important finding is that the agriculture sector of Punjab, Sindh and KPK has developed forward linkages with the services sector of Balochistan. The long run and short run parameters of services growth of
Balochistan with manufacturing across sectors and across provinces indicate that the manufacturing growth of Punjab and Sindh have significant and strong relationship with services growth of Balochistan. The long and short run parameters of services growth of Balochistan and
130
services growth of Punjab, Sindh and KPK are significant showing a strong relationship between services growth of Balochistan and services growth of Punjab, Sindh and KPK. It means even the services sector of Balochistan is significantly affected by the services sectors of other provinces.
The above analysis shows that both physical and social services are not developed properly in
Balochistan. As a result of which the growth of services sector is almost negligible and its contribution to national growth is very low and the dependency of services growth in Balochistan is high on sectoral growths of large provinces Punjab and Sindh. This means that inorder to have a significant growth and development of relationships in services sector of Balochistan across sectors and across provinces. The federal and provincial governments have to adopt focused and well implemented policies to improve the services growth of the province.
4.11 Inter relationships between Provinces
The GDP growth of the country mainly depends upon the GDP growth of large provinces Punjab and Sindh. The GDP growth of the provinces reveals that mainly the large provinces, Punjab and
Sindh are integrated with each other. Moreover, the growth of the small provinces depends on the growth of large provinces. The low growth of GPD of small provinces was mainly attributed to the low development of these provinces, as the studies reported regarding KPK and
Balochistan, highlight the key facts that both the provinces have huge potential. The high growth could be achieved through sustainable and integrated pattern of growth of these provinces, provided they are properly focused by the policy planers of the federal and provincial governments.
131
The low growth of the large provinces and the country as whole was mainly due to high inconsistencies in all sectors especially manufacturing sector over the period of time, moreover economic growth was focused by the policy planners rather than, inclusive growth. This eventually led to the unsustainable growth of large provinces and country as whole.
The above analysis highlighted that the potential of small provinces KPK and Balochistan is not been still fully exploited by the government. This is the very reason, that these small provinces sill depends heavily on the growth of large provinces Punjab and Sindh. However, the provincial economies of large provinces are not performing well and especially after the liberalisation of trade these provincial economies are facing high challenges to compete in international markets.
Hence, to have better integration in GDP growth of the provinces, the small provinces should be developed in such a manner as they should significantly contribute to the overall growth of the country. The commodity producing sectors of large provinces are to be focused properly, inorder to compete internationally and develop linkages with small provinces. So as the growth of large the provinces should be integrated with growth of small provinces. Hence both small and large provinces should grow in an integrated manner to achieve a sustainable growth pattern of the country.
4.11. Summary of the Findings This chapter was focused on exploring cointegration relationships among sectors and provinces
GDP, as well as, indicating the inter-linkages between provincial economics of Pakistan.
Moreover, the speed of adjustment was also observed for the same linkages, when these provincial economies were in crises. After analyzing short run and long run dynamics of agriculture, manufacturing and services sectors across provinces and across sectors, it is important to analyze the dynamics of agriculture, manufacturing and services sector of one
132
province, with other provinces across sectors. It will provide basics for how provinces are integrated to each other; in different sectors.
In this chapter, first the GDP growth causalities and linkages through Granger causality test and
Engle Granger Cointegration test. Firstly, the growth causalities and linkages are explored with
GDP growth of four provinces of the country. Secondly, the growth causalities and linkages are explored at sectoral level of each province26. Thereafter, Cointegration and Error Correction
Model (ECM) was estimated between provincial GDP growth and sectoral growth rates of the provinces. The results of causality test of GDP growth of provinces revealed that there are 25 equations, which were estimated and 9 equations showed the presence of causal relationships27.
The cointegration analysis revealed that the large two provinces i.e. Punjab and Sindh, are integrated with each other and the speed of short term adjustment of GDP growth of Sindh is fast than Punjab, the very reason may be due to its global linkages, port and major industrial hub, as well as, head office of financial market and trade centre of the country i.e. export and imports destiny. Secondly, both the large provinces are integrated with small provinces in such a manner that if any disequilibrium occurs, it was adjusted by their linkages with the provinces of Punjab or Sindh. In other words, the small provinces are dependent upon large provinces. These small provinces hardly function to recover it from any economic crises. It may be the very reason that these provinces remained relatively more underdeveloped and hardly any effective policies were introduced by the federation to integrate the periphery areas of Pakistan. These findings have important bearing for policy makers and think tank, to introduce certain measures to bring these small provinces into the main framework of the economy; rather than let them lag behind other
26 . For casualty tests and cause effect see tables 4.1-4.5 (appendix). These tables are not discussing but explored to understand cointegration results. 27 See appendix summary tables 4.1- 4.5. 133
provinces. The high dependence of small provinces of KPK and Balochistan, on large provinces
Punjab and Sindh, need to be addressed so that the population of these provinces do not feel deprived and equitable distribution of the growth fruits are ensured. It will help the whole country to develop on equitable basis and these economies will be integrated into the main framework.
The empirical analysis at sectoral level indicated that the commodity producing sectors of large provinces have significant linkage with each other provinces and these are affecting the growth of small provinces. Similar, results were found for the services sector; rather this linkage was even stronger than the above cited sectors. It may be the reason that a bulk of labor force move to
Sindh (Karachi) and Punjab for jobs search. Hence, the sectoral growth analysis also revealed the same facts as found for overall GDP growth of the provinces; that large provinces are integrated highly with each other and with small provinces, which are dependent on the large provinces.
This can be seen from the cointegration results for sectoral level estimates between provinces.
The analysis of the sectoral level is divided in three types: as co-integration within sector across provinces, cointegration within province across sectors and cointegration across sector and across provinces (see chapter appendix table: 4.6).
The cointegration within sectors across provinces shows the linkages across provinces within a particular sector. According to above table there are 25 co-integrating equations out of 36. It implies that the growth in sectors agriculture or manufacturing or services of any province; either province Punjab or Sindh or KPK or Balochistan affect the growth of the same sector in other provinces. This confirms that the macroeconomic variables’ linkages within provinces. A study by IPR (2015) provides the share of various sectors in the provincial GDP, which indicated that the composition of the provincial economies is quite similar, so the likelihood is higher that if a
134
macro policy designed for large province is also implemented in small provinces, they can also grow equally. Other studies also indicated that sectoral growth showed that macro variables are important in explaining variation, as compared to geographical factors.(Shahbaz, M., Ahmad, K
2008)28.
The cointegration results for within provinces and across sectors highlight the forward and backward linkages within provinces. It is important to note that the forward and backward linkages with small provinces are weak. The results show that out of 24 pair wise co-integrated equations, 11 linkage equations turned up as a significant. Tiwari, A. and S. Kg (2011) shows that the output of commodity producing sector will affect the growth of services sector, but this study finds that a provincial level this relationship does not hold.
The present study has estimated co-integrated equations to test the pair wise cointegration across province and across sector. The results show that out of 72 co-integrating equations, 40 equations were significant; in case of Agriculture and Manufacturing; Out of 16 cointegration equations there were 9 equations which turned up significant. However, in the case of agriculture, dependence on manufacturing, 6 out of 16 equations have significant long run and short run relationship.
It is important to note that manufacturing sector of Punjab and Sindh does not depend on the agricultural growth anymore; it was used to do so in the past. Thus, such a structural change and dependent need to be noted. The larger industry set up is related to textile and textile family products. The inputs of these industries are now either imported or these industries have been
28 , Also see Chaudhary, A. (2008), Ajmair, M. and K. Hussain (2017), Ahmad, K., et al. (2012) and ILYAS, M., AHMAD, H., AFZAL, M., & MAHMOOD, T. (2010); shows that macro variables significantly determine the growth of a sector and overall GDP growth. 135
moved out of country too29. This shows that the backward linkages between manufacturing and agriculture do not exist for Punjab and Sindh. However, the manufacturing and agriculture of
Punjab and Sindh significantly affected by the services growth of these large provinces. The results also indicated that there exists backward linkage of services sector of Punjab and Sindh; with agriculture and manufacturing growth of these provinces. However, forward linkages are mostly present from manufacturing to services growth of theses provinces. Such linkage does not exist from agriculture growth to services growth of these provinces.
In case of small provinces, KPK and Balochistan, agriculture growth has not developed significant forward linkages with manufacturing sectors within provinces and across provinces.
The manufacturing sector analysis show that there are some forward linkages present from manufacturing to services growth in these provinces. The backward linkages of services growth to manufacturing growth are there. However, such linkages for agriculture growth of these provinces are mostly absent. The main reason is the low development of services and manufacturing activities in these two small provinces are evident. By addressing the lack of integration of small province may help to improve the economies of the deprived areas.
In brief it may said, firstly, there exit linkages and integration within and across sectors of large provinces; especially for manufacturing and services sectors. However, due to low base and week agriculture sector, most of these links are absent between agriculture and manufacturing growth small provinces. Secondly, the services sector has developed strong forward and backward linkages in these large provinces. The very reason for this is the high concentration of services sector development in these large provinces, particularly, financial and international linkages of the Sindh province. Thirdly, the small provinces are far behind, as they do not have
29 . See for details Chaudhary m. Aslam & Aslam A. (2018), to be published yet. 136
significant and strong linkages within and across commodity producing sectors. They remain periphery areas and relatively underdeveloped regions. Fourthly, the small provinces KPK and
Balochistan, depend highly on the growth of large provinces, in overall growth (GDP) and in sectoral growth. It shows the vulnerability of small provinces towards large provinces. It calls for improvements in the regional economies of small provinces so that these can be integrated into the main stream of development. It may be the very reason that these provinces feel deprived and criticise larger provinces30. No wonder that the small provinces remained relatively poor and were also not integrated with large provinces, as well as, with the national economy. The regional development of the deprived areas of Pakistan remains a challenge for the policy makers and planners in Pakistan. This study has first time pointed out areas and need for linkages of sectors and provinces to address the regional development issue in Pakistan.
30 . The provincial composition of agriculture, industry and services is calculated by IPR (2015), which indicated that the composition of agriculture sector is high in Punjab and Balochistan; in their respective provincial economies. The provinces of Balochistan has agriculture the only main sector which hardly contribute 3% to national exchequer; still very small. The services sector composition is highest in provincial economies of Punjab and KPK; followed by Sindh and Balochistan. So, if the full potential of these provincial economies are exploited, it will help them to contribute significantly to overall national growth and develop linkages within sectors of their own provincial economies and across sectors of large provinces. For example, mineral development in Balochistan can uplift the province and it can also help other provinces too. CPAC is likely to improve the situation, which a significant effort to uplift the periphery areas.
137
Annexures Tables: Chapter 4 Table: 4.1 Granger Causality analysis of GDP growth, Agriculture growth, Manufacturing growth and Services growth; within sector, across Provinces Punjab Equ. Flow of causation 1 Unidirectional causality exists from GDP growth of Punjab to GDP growth of Sindh 2 Unidirectional causality exists from manufacturing of Punjab to services of Punjab 3 Unidirectional causality exits from manufacturing of Punjab to services of Sindh 4 Unidirectional causality exists from manufacturing of Punjab to services of KPK 5 Unidirectional causality exists from services of Punjab to manufacturing of Sindh 6 Bidirectional causality exits between services of Punjab and services of Sindh 7 Bidirectional causality exits between services of Punjab and services of KPK Sindh Flow of causation 8 Bidirectional causality between GDP growth of Sindh to GDP growth of KPK 9 Unidirectional causality exists from agriculture of Sindh to agriculture of KPK 10 Unidirectional causality exist from agriculture of Sindh to agriculture of Balochistan 11 Unidirectional causality exist from agriculture of Sindh to services of Punjab 12 Bidirectional causality exits between manufacturing of Sindh and Manufg. of KPK 13 Bidirectional causality exits between Manufg. of Sindh and Manufg. of Balochistan 14 Bidirectional causality exists between manufacturing of Sindh to services of Sindh 15 Unidirectional causality exists from manufacturing of Sindh to services of KPK 16 Unidirectional causality exists from services of Sindh to agriculture of Balochistan 17 Unidirectional causality exits from services of Sindh to services of KPK 18 Unidirectional causality exits from services of Sindh to services of Balochistan 19 Bidirectional causality exits between services of KPK and services of Balochistan Continue...... KPK Flow of causation 20 Bidirectional causality between GDP growth of KPK to GDP growth of Sindh 21 Unidirectional causality exists from manufacturing of KPK to agriculture of Sindh 22 Unidirectional causality exists from manufg. of KPK to manufg. of Balochistan Balochistan Flow of causation 23 Unidirectional causality exists from GDP growth of Balochistan to GDP growth of Punjab 24 Unidirectional causality exist from agriculture of Balochistan to agriculture of Punjab 25 Unidirectional causality exits from agriculture of Balochistan to manufacturing of KPK 26 Unidirectional causality exists from agriculture of Balochistan to manufag. of Balochistan 27 Unidirectional causality exists from manufg. of Balochistan to manufg. of Punjab 28 Unidirectional causality exists from manufacturing of Balochistan to services of Punjab 29 Unidirectional causality exists from manufacturing of Balochistan to services of KPK 30 Unidirectional causality exists from services of Balochistan to manufacturing of Sindh 31 Unidirectional causality exist from services of Balochistan to agriculture of KPK 32 Unidirectional causality exists from services of Balochistan to manufacturing of KPK 33 Bidirectional causality exits between services of Balochistan and services of KPK Source: Author’s estimate and for detail results see Annexure 4-A to 4-G, of chapter4.
138
Table : 4. Composition of Provincial Economies 2014-15 ( Share in Provincial GDP, %)
Province/sector Punjab Sindh KPK Balochistan Agriculture 24.07 16.06 16.87 28.97 Industry 15.35 29.32 22.89 26.61 Services 60.58 54.62 60.24 44.42 100 100 100 100 Source: IPR(2015) Table: 4.2 Summary of Granger Causality Equations of GDP growth of Provinces Sector GDP Growth Total Equations Causal Relationships among Equations Granger Causality test between GDP Growth across provinces 25 9 Source: the table is calculated from the estimates by the author. Table: 4.3. Summary of Granger Causality Equations of Agriculture sector; within sector, across sectors and across provinces Sector Agriculture Sector Total Equations Causal Relationships among Equations Granger Causality test within agriculture Sector across provinces 24 6 Granger Causality test of agriculture Sector across; across sectors, across provinces 64 6 Total 88 12 Source: the table is calculated from the estimates by the auth Table: 4.4. Summary of Granger causality equations of Manufacturing sector; within sector, across sector and across provinces Sector Manufacturing Sector Total Equations Causal Relationships among Equations Granger Causality test within manufacturing Sector across provinces 24 12 Granger Causality test of manufacturing Sector across; across sectors, across provinces 64 17 Total 88 29 Source: the table is calculated from the estimates by the author.
Table:4.5. Summary of Granger causality equations of Services sector; within sector, across sector and across provinces
Sector Services Sector Total Equations Causal Relationships among Equations Granger Causality test within services Sector across provinces 24 18 Granger Causality test of services Sector across; across sectors, across provinces 64 14 Total 88 32 Source: the table is calculated from the estimates by the author.
Table: 4.6. Summary of Cointegration equations Estimated Sector Significant Parameters Total Cointegration Cointegration within Sector across provinces 25 36 Cointegration within Province across Sectors 11 24 Across Sectors Across Provinces 42 72 Total equations 78 132 Source: the table is calculated from the estimates by the author.
139
ANNEXURE 4-A Table: 4.1 Granger Causality analysis of GDP Growth between Provinces GDP growth of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 GDP growth of Punjab GDP growth of Sindh Unidirectional causality from GDP growth of Punjab to GDP growth of Sindh Equation 2 GDP growth of Sindh GDP growth of Punjab No causation Equation 3 GDP growth of Punjab GDP growth of KPK No causation Equation 4 GDP growth of KPK GDP growth of Punjab Equation 5 GDP growth of Punjab GDP growth of No causation Balochistan Equation 6 GDP growth of GDP growth of Punjab Unidirectional causality from GDP growth of Balochistan to GDP Balochistan growth of Punjab GDP growth of Sindh Independent variable Dependent variable Flow of causation Equation 7 GDP growth of Sindh GDP growth of Punjab No causation Equation 8 GDP growth of Punjab GDP growth of Sindh Unidirectional causality from GDP growth of Punjab to GDP growth of Sindh Equation 9 GDP growth of Sindh GDP growth of KPK Bidirectional causality from GDP growth of Sindh to GDP growth of Equation GDP growth of KPK GDP growth of Sindh KPK 10 Equation GDP growth of Sindh GDP growth of No causation 11 Balochistan Equation GDP growth of GDP growth of Sindh No causation 12 Balochistan GDP growth of KPK Independent variable Dependent variable Flow of causation Equation GDP growth of KPK GDP growth of Punjab No causation 13 Equation GDP growth of Punjab GDP growth of KPK No causation 14 Equation GDP growth of KPK GDP growth of Sindh Bidirectional causality exits between GDP growth of KPK to GDP 15 growth of Sindh Equation GDP growth of Sindh GDP growth of KPK 16 Equation GDP growth of KPK GDP growth of 17 Balochistan No causation Equation GDP growth of GDP growth of KPK 18 Balochistan No causation GDP growth of Balochistan Independent variable Dependent variable Flow of causation Equation GDP growth of GDP growth of Punjab Unidirectional causality from GDP growth of Balochistan to GDP 19 Balochistan growth of Punjab Equation GDP growth of Punjab GDP growth of No causation 20 Balochistan Equation GDP growth of GDP growth of Sindh No causation 21 Balochistan Equation GDP growth of Sindh GDP growth of No causation 22 Balochistan Equation GDP growth of GDP growth of KPK No causation 23 Balochistan Equation GDP growth of KPK GDP growth of No causation 24 Balochistan Equation Agriculture growth of Services growth of 25 Punjab KPK No causation
Source: the table is draw from the estimates by the author: see Annexure 1-A
140
ANNEXURE 4-B
Table: 4.4 Granger Causality analysis of Agriculture Sector; within sector, across Provinces Agriculture Sector of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 Agriculture growth of Punjab Agriculture growth of Sindh No causation Equation 2 Agriculture growth of Sindh Agriculture growth of Punjab No causation Equation 3 Agriculture growth of Punjab Agriculture growth of KPK No causation
Equation 4 Agriculture growth of KPK Agriculture growth of Punjab No causation Equation 5 Agriculture growth of Punjab Agriculture growth of Balochistan No causation Equation 6 Agriculture growth of Balochistan Agriculture growth of Punjab Unidirectional causality from agriculture of Balochistan to agriculture of Punjab Agriculture Sector of Sindh Independent variable Dependent variable Flow of causation From To Equation 7 Agriculture growth of Sindh Agriculture growth of Punjab No causation
Equation 8 Agriculture growth of Punjab Agriculture growth of Sindh No causation
Equation 9 Agriculture growth of Sindh Agriculture growth of KPK Unidirectional causality from agriculture of Sindh to agriculture of KPK Equation 10 Agriculture growth of KPK Agriculture growth of Sindh No causation Equation 11 Agriculture growth of Sindh Agriculture growth of Balochistan Unidirectional causality from agriculture of Sindh to agriculture of Balochistan Equation 12 Agriculture growth of Balochistan Agriculture growth of Sindh No Causation
ANNEXURE 4-B Table: 4.4 Granger Causality analysis of Agriculture Sector; within sector, across Provinces
Agriculture Sector of KPK Independent variable Dependent variable Flow of causation From To Equation 13 Agriculture growth of KPK Agriculture growth of Punjab No causation Equation 14 Agriculture growth of Punjab Agriculture growth of KPK No causation Equation 15 Agriculture growth of KPK Agriculture growth of Sindh No causation
Equation 16 Agriculture growth of Sindh Agriculture growth of KPK Unidirectional causality from agriculture of Sindh to agriculture of KPK Equation 17 Agriculture growth of KPK Agriculture growth of Balochistan No causation Equation 18 Agriculture growth of Balochistan Agriculture growth of KPK No causation Agriculture Sector of Balochistan Independent variable Dependent variable Flow of causation From To Equation 19 Agriculture growth of Balochistan Agriculture growth of Punjab Unidirectional causality from agriculture of Balochistan to agriculture of Punjab Equation 20 Agriculture growth of Punjab Agriculture growth of Balochistan No causation Equation 21 Agriculture growth of Balochistan Agriculture growth of Sindh No causation
Equation 22 Agriculture growth of Sindh Agriculture growth of Balochistan Unidirectional causality from agriculture of Sindh to agriculture of Balochistan Equation 23 Agriculture growth of Balochistan Agriculture growth of KPK No causation Equation 24 Agriculture growth of KPK Agriculture growth of Balochistan No causation Source: the table is draw from the estimates by the author: see Annexure 1-A
141
ANNEXURE 4-C Table: 4.6 Granger Causality analysis of Agriculture Sector; across sectors, across Provinces Agriculture Sector of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 Agriculture growth of Punjab Manufacturing growth of Punjab No causation Equation 2 Manufacturing growth of Punjab Agriculture growth of Punjab No causation Equation 3 Agriculture growth of Punjab Manufacturing growth No causation of Sindh Equation 4 Manufacturing growth of Sindh Agriculture growth of No causation Punjab Equation 5 Agriculture growth of Punjab Manufacturing growth No causation of KPK Equation 6 Manufacturing growth of KPK Agriculture growth of No causation Punjab Equation 7 Agriculture growth of Punjab Manufacturing growth No causation of Balochistan Equation 8 Manufacturing growth of Balochistan Agriculture growth of No causation Punjab Equation 9 Agriculture growth of Punjab Services growth of No causation Punjab Equation 10 Services growth of Punjab Agriculture growth of No causation Punjab Equation 11 Agriculture growth of Punjab Services growth of No causation Sindh Equation 12 Services growth of Sindh Agriculture growth of No causation Punjab Equation 13 Agriculture growth of Punjab Services growth of KPK No causation Equation 14 Services growth of KPK Agriculture growth of No causation Punjab Equation 15 Agriculture growth of Punjab Services growth of No causation Balochistan Equation 16 Services growth of Balochistan Agriculture growth of No causation Punjab Agriculture Sector of Sindh Independent variable Dependent variable Flow of causation From To Equation 17 Agriculture growth of Sindh Manufacturing growth of Punjab No causation Equation 18 Manufacturing growth of Punjab Agriculture growth of Sindh No causation Equation 19 Agriculture growth of Sindh Manufacturing growth No causation of Sindh Equation 20 Manufacturing growth of Sindh Agriculture growth of No causation Sindh Equation 21 Agriculture growth of Sindh Manufacturing growth No causation of KPK Equation 22 Manufacturing growth of KPK Agriculture growth of Unidirectional causality from manufacturing of Sindh KPK to agriculture of Sindh Equation 23 Agriculture growth of Sindh Manufacturing growth No causation of Balochistan Equation 24 Manufacturing growth of Balochistan Agriculture growth of No causation Sindh Equation 25 Agriculture growth of Sindh Services growth of Unidirectional causality from agriculture of Punjab Sindh to services of Punjab Equation 26 Services growth of Punjab Agriculture growth of No causation Sindh Equation 27 Agriculture growth of Sindh Services growth of No causation Sindh Equation 28 Services growth of Sindh Agriculture growth of No causation Sindh Equation 29 Agriculture growth of Sindh Services growth of KPK No causation
142
Equation 30 Services growth of KPK Agriculture growth of No causation Sindh Equation 31 Agriculture growth of Sindh Services growth of No causation Balochistan Equation 32 Services growth of Balochistan Agriculture growth of No causation Sindh Agriculture Sector of KPK Independent variable Dependent variable Flow of causation From To Equation 33 Agriculture growth of KPK Manufacturing growth of Punjab No causation Equation 34 Manufacturing growth of Punjab Agriculture growth of KPK No causation Equation 35 Agriculture growth of KPK Manufacturing growth No causation of Sindh Equation 36 Manufacturing growth of Sindh Agriculture growth of No causation KPK Equation 37 Agriculture growth of KPK Manufacturing growth No causation of KPK Equation 38 Manufacturing growth of KPK Agriculture growth of No causation KPK Equation 39 Agriculture growth of KPK Manufacturing growth No causation of Balochistan Equation 40 Manufacturing growth of Balochistan Agriculture growth of No causation KPK Equation 41 Agriculture growth of KPK Services growth of No causation Punjab Equation 42 Services growth of Punjab Agriculture growth of No causation KPK Equation 43 Agriculture growth of KPK Services growth of No causation Sindh Equation 44 Services growth of Sindh Agriculture growth of No causation KPK Equation 45 Agriculture growth of KPK Services growth of KPK No causation Equation 46 Services growth of KPK Agriculture growth of No causation KPK Equation 47 Agriculture growth of KPK Services growth of No causation Balochistan Equation 48 Services growth of Balochistan Agriculture growth of Unidirectional causality from services of KPK Balochistan to agriculture of KPK Agriculture Sector of Balochistan Independent variable Dependent variable Flow of causation From To Equation 49 Agriculture growth of Balochistan Manufacturing growth of Punjab No causation Equation 50 Manufacturing growth of Punjab Agriculture growth of Balochistan No causation Equation 51 Agriculture growth of Balochistan Manufacturing growth No causation of Sindh Equation 52 Manufacturing growth of Sindh Agriculture growth of No causation Balochistan Equation 53 Agriculture growth of Balochistan Manufacturing growth Unidirectional causality from agriculture of of KPK Balochistan to manufacturing of KPK Equation 54 Manufacturing growth of KPK Agriculture growth of No causation Balochistan Equation 55 Agriculture growth of Balochistan Manufacturing growth Unidirectional causality from agriculture of of Balochistan Balochistan to manufacturing of Balochistan Equation 56 Manufacturing growth of Balochistan Agriculture growth of No causation Balochistan Equation 57 Agriculture growth of Balochistan Services growth of No causation Punjab Equation 58 Services growth of Punjab Agriculture growth of No causation Balochistan Equation 59 Agriculture growth of Balochistan Services growth of No causation Sindh Equation 60 Services growth of Sindh Agriculture growth of Unidirectional causality from services of Sindh Balochistan to agriculture of Balochistan 143
Equation 61 Agriculture growth of Balochistan Services growth of KPK No causation Equation 62 Services growth of KPK Agriculture growth of No causation Balochistan Equation 63 Agriculture growth of Balochistan Services growth of No causation Balochistan Equation 64 Services growth of Balochistan Agriculture growth of No causation Balochistan Source: the table is draw from the estimates by the author: see Annexure 1-A ANNEXURE 4-D Table : 4.8. Granger Causality analysis of Manufacturing Sector; within sector, across provinces Manufacturing Sector of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 Manufacturing growth of Manufacturing growth of Punjab Sindh No causation
Equation 2 Manufacturing growth of Manufacturing growth of Sindh Punjab No causation
Equation 3 Manufacturing growth of Manufacturing growth of No causation Punjab KPK
Equation 4 Manufacturing growth of Manufacturing growth of No causation KPK Punjab Equation 5 Manufacturing growth of Manufacturing growth of No causation Punjab Balochistan Equation 6 Manufacturing growth of Manufacturing growth of Unidirectional causality from manufacturing of Balochistan to Balochistan Punjab manufacturing of Punjab Manufacturing Sector of Sindh Independent variable Dependent variable Flow of causation From To Equation 7 Manufacturing growth of Manufacturing growth of No causation Sindh Punjab Equation 8 Manufacturing growth of Manufacturing growth of Punjab Sindh No causation
Equation 9 Manufacturing growth of Manufacturing growth of Bidirectional causality exits between manufacturing of Sindh Sindh KPK and manufacturing of KPK Equation 10 Manufacturing growth of Manufacturing growth of KPK Sindh Equation 11 Manufacturing growth of Manufacturing growth of Bidirectional causality exits between manufacturing of Sindh Sindh Balochistan and manufacturing of Balochistan Equation 12 Manufacturing growth of Manufacturing growth of Balochistan Sindh Manufacturing Sector of KPK Independent variable Dependent variable Flow of causation From To Equation 13 Manufacturing growth of Manufacturing growth of KPK Punjab No causation
Equation 14 Manufacturing growth of Manufacturing growth of Punjab KPK No causation
Equation 15 Manufacturing growth of Manufacturing growth of Bidirectional causality exits between manufacturing of Sindh KPK Sindh and manufacturing of KPK
Equation 16 Manufacturing growth of Manufacturing growth of Sindh KPK Equation 17 Manufacturing growth of Manufacturing growth of Unidirectional causality from manufacturing of KPK to KPK Balochistan manufacturing of Balochistan Equation 18 Manufacturing growth of Manufacturing growth of No causation Balochistan KPK Manufacturing Sector of Balochistan Independent variable Dependent variable Flow of causation From To Equation 19 Manufacturing growth of Manufacturing growth of Balochistan Punjab Unidirectional causality from manufacturing of Balochistan to 144
manufacturing of Punjab Equation 20 Manufacturing growth of Manufacturing growth of Punjab Balochistan No causation
Equation 21 Manufacturing growth of Manufacturing growth of Bidirectional causality exits between manufacturing of Sindh Balochistan Sindh and manufacturing of Balochistan
Equation 22 Manufacturing growth of Manufacturing growth of Sindh Balochistan
Equation 23 Manufacturing growth of Manufacturing growth of No causation Balochistan KPK
Equation 24 Manufacturing growth of Manufacturing growth of Unidirectional causality from manufacturing of KPK to KPK Balochistan manufacturing of Balochistan
Source: the table is draw from the estimates by the author: see Annexure 1-A
ANNEXURE 4-E Table: 4.9. Granger Causality analysis of Manufacturing Sector; across sectors, across provinces Manufacturing Sector of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 Manufacturing growth of Punjab Agriculture growth of Punjab No causation Equation 2 Agriculture growth of Punjab Manufacturing growth of Punjab No causation Equation 3 Manufacturing growth of Punjab Agriculture growth of No causation Sindh Equation 4 Agriculture growth of Sindh Manufacturing growth of No causation Punjab Equation 5 Manufacturing growth of Punjab Agriculture growth of No causation KPK Equation 6 Agriculture growth of KPK Manufacturing growth of No causation Punjab Equation 7 Manufacturing growth of Punjab Agriculture growth of No causation Balochistan Equation 8 Agriculture growth of Balochistan Manufacturing growth of No causation Punjab Equation 9 Manufacturing growth of Punjab Services growth of Punjab Unidirectional causality from manufacturing of Punjab to services of Punjab Equation 10 Services growth of Punjab Manufacturing growth of No causation Punjab Equation 11 Manufacturing growth of Punjab Services growth of Sindh Unidirectional causality from manufacturing of Punjab to services of Sindh Equation 12 Services growth of Sindh Manufacturing growth of No causation Punjab Equation 13 Manufacturing growth of Punjab Services growth of KPK Unidirectional causality from manufacturing of Punjab to services of KPK Equation 14 Services growth of KPK Manufacturing growth of No causation Punjab Equation 15 Manufacturing growth of Punjab Services growth of No causation Balochistan Equation 16 Services growth of Balochistan Manufacturing growth of No causation Punjab Manufacturing Sector of Sindh Independent variable Dependent variable Flow of causation From To Equation 17 Manufacturing growth of Sindh Agriculture growth of Punjab No causation Equation 18 Agriculture growth of Punjab Manufacturing growth of Sindh No causation Equation 19 Manufacturing growth of Sindh Agriculture growth of No causation Sindh Equation 20 Agriculture growth of Sindh Manufacturing growth of No causation Sindh Equation 21 Manufacturing growth of Sindh Agriculture growth of No causation 145
KPK Equation 22 Agriculture growth of KPK Manufacturing growth of No causation Sindh Equation 23 Manufacturing growth of Sindh Agriculture growth of No causation Balochistan Equation 24 Agriculture growth of Balochistan Manufacturing growth of No causation Sindh Equation 25 Manufacturing growth of Sindh Services growth of Punjab No causation
Equation 26 Services growth of Punjab Manufacturing growth of Unidirectional causality from services of Punjab to Sindh manufacturing of Sindh Equation 27 Manufacturing growth of Sindh Services growth of Sindh Bidirectional causality exists between manufacturing of Sindh to services of Sindh Equation 28 Services growth of Sindh Manufacturing growth of Sindh Equation 29 Manufacturing growth of Sindh Services growth of KPK Unidirectional causality from manufacturing of Sindh to services of KPK Equation 30 Services growth of KPK Manufacturing growth of No causation Sindh Equation 31 Manufacturing growth of Sindh Services growth of No causation Balochistan Equation 32 Services growth of Balochistan Manufacturing growth of Unidirectional causality from services of Sindh Balochistan to manufacturing of Sindh Manufacturing Sector of KPK Independent variable Dependent variable Flow of causation From To Equation 33 Manufacturing growth of KPK Agriculture growth of Punjab No causation Equation 34 Agriculture growth of Punjab Manufacturing growth of KPK No causation Equation 35 Manufacturing growth of KPK Agriculture growth of Unidirectional causality from manufacturing of Sindh KPK to agriculture of Sindh Equation 36 Agriculture growth of Sindh Manufacturing growth of No causation KPK Equation 37 Manufacturing growth of KPK Agriculture growth of No causation KPK Equation 38 Agriculture growth of KPK Manufacturing growth of No causation KPK Equation 39 Manufacturing growth of KPK Agriculture growth of No causation Balochistan Equation 40 Agriculture growth of Balochistan Manufacturing growth of Unidirectional causality from agriculture of KPK Balochistan to manufacturing of KPK Equation 41 Manufacturing growth of KPK Services growth of Punjab No causation
Equation 42 Services growth of Punjab Manufacturing growth of No causation KPK Equation 43 Manufacturing growth of KPK Services growth of Sindh No causation
Equation 44 Services growth of Sindh Manufacturing growth of No causation KPK Equation 45 Manufacturing growth of KPK Services growth of KPK No causation Equation 46 Services growth of KPK Manufacturing growth of No causation KPK Equation 47 Manufacturing growth of KPK Services growth of No causation Balochistan Equation 48 Services growth of Balochistan Manufacturing growth of Unidirectional causality from services of KPK Balochistan to manufacturing of KPK Manufacturing Sector of Balochistan Independent variable Dependent variable Flow of causation From To Equation 49 Manufacturing growth of Balochistan Agriculture growth of Punjab No causation Equation 50 Agriculture growth of Punjab Manufacturing growth of Balochistan No causation Equation 51 Manufacturing growth of Balochistan Agriculture growth of No causation Sindh Equation 52 Agriculture growth of Sindh Manufacturing growth of No causation 146
Balochistan Equation 53 Manufacturing growth of Balochistan Agriculture growth of No causation KPK Equation 54 Agriculture growth of KPK Manufacturing growth of No causation Balochistan Equation 55 Manufacturing growth of Balochistan Agriculture growth of No causation Balochistan Equation 56 Agriculture growth of Balochistan Manufacturing growth of Unidirectional causality from agriculture of Balochistan Balochistan to manufacturing of Balochistan Equation 57 Manufacturing growth of Balochistan Services growth of Punjab Unidirectional causality from manufacturing of Balochistan to services of Punjab Equation 58 Services growth of Punjab Manufacturing growth of No causation Balochistan Equation 59 Manufacturing growth of Balochistan Services growth of Sindh No causation
Equation 60 Services growth of Sindh Manufacturing growth of No causation Balochistan Equation 61 Manufacturing growth of Balochistan Services growth of KPK Unidirectional causality from manufacturing of Punjab to services of KPK Equation 62 Services growth of KPK Manufacturing growth of No causation Balochistan Equation 63 Manufacturing growth of Balochistan Services growth of No causation Balochistan Equation 64 Services growth of Balochistan Manufacturing growth of No causation Balochistan Source: the table is draw from the estimates by the author: see Annexure 1-A
ANNEXURE 4-F Table: 4.12. Granger Causality analysis of Services Sector; within sector, across provinces Services Sector of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 Services growth of Punjab Services growth of Sindh Bidirectional causality exits between services of Punjab and services of Sindh Equation 2 Services growth of Sindh Services growth of Punjab
Equation 3 Services growth of Punjab Services growth of KPK Bidirectional causality exits between services of Punjab and services of KPK Equation 4 Services growth of KPK Services growth of Punjab Equation 5 Services growth of Punjab Services growth of No causation Balochistan Equation 6 Services growth of Services growth of Punjab No causation Balochistan Services Sector of Sindh From To Independent variable Dependent variable Flow of causation Equation 7 Services growth of Sindh Services growth of Punjab Bidirectional causality exits between services of Sindh and Equation 8 Services growth of Punjab Services growth of Sindh services of Punjab
Equation 9 Services growth of Sindh Services growth of KPK Unidirectional causality exits from services of Sindh to services of KPK
Equation 10 Services growth of KPK Services growth of Sindh No causation Equation 11 Services growth of Sindh Services growth of Unidirectional causality exits from services of Sindh to services Balochistan of Balochistan Equation 12 Services growth of Services growth of Sindh No causation Balochistan Services Sector of KPK Independent variable Dependent variable Flow of causation From To Equation 13 Services growth of KPK Services growth of Punjab Bidirectional causality exits between services of Punjab and Equation 14 Services growth of Punjab Services growth of KPK services of KPK
Equation 15 Services growth of KPK Services growth of Sindh No causation
147
Equation 16 Services growth of Sindh Services growth of KPK Unidirectional causality from services of Sindh to services of KPK Equation 17 Services growth of KPK Services growth of Bidirectional causality exits between services of KPK and Balochistan services of Balochistan Equation 18 Services growth of Services growth of KPK Balochistan Services Sector of Balochistan Independent variable Dependent variable Flow of causation From To Equation 19 Services growth of Services growth of Punjab No causation Balochistan
Equation 20 Services growth of Punjab Services growth of Balochistan No causation Equation 21 Services growth of Services growth of Sindh No causation Balochistan
Equation 22 Services growth of Sindh Services growth of Unidirectional causality from services of Sindh to services of Balochistan Balochistan Equation 23 Services growth of Services growth of KPK Bidirectional causality exits between services of Balochistan and Balochistan services of KPK
Equation 24 Services growth of KPK Services growth of Balochistan Source: the table is draw from the estimates by the author: see Annexure 1-A
ANNEXURE 4-G
Table: 4.13. Granger Causality analysis of Services Sector; across sectors, across provinces Services Sector of Punjab Independent variable Dependent variable Flow of causation From To Equation 1 Services growth of Punjab Agriculture growth of Punjab No causation Equation 2 Agriculture growth of Punjab Services growth of Punjab No causation Equation 3 Services growth of Punjab Agriculture growth of No causation Sindh Equation 4 Agriculture growth of Sindh Services growth of Punjab Unidirectional causality from agriculture of Sindh to services of Punjab Equation 5 Services growth of Punjab Agriculture growth of No causation KPK Equation 6 Agriculture growth of KPK Services growth of Punjab No causation Equation 7 Services growth of Punjab Agriculture growth of No causation Balochistan Equation 8 Agriculture growth of Balochistan Services growth of Punjab No causation Equation 9 Services growth of Punjab Manufacturing growth of No causation Punjab Equation 10 Manufacturing growth of Punjab Services growth of Punjab Unidirectional causality from manufacturing of Punjab to services of Punjab Equation 11 Services growth of Punjab Manufacturing growth of Unidirectional causality from services of Punjab to Sindh manufacturing of Sindh Equation 12 Manufacturing growth of Sindh Services growth of Punjab No causation Equation 13 Services growth of Punjab Manufacturing growth of No causation KPK Equation 14 Manufacturing growth of KPK Services growth of Punjab No causation Equation 15 Services growth of Punjab Manufacturing growth of No causation Balochistan Equation 16 Manufacturing growth of Balochistan Services growth of Punjab Unidirectional causality from manufacturing of Balochistan to services of Punjab Services Sector of Sindh Independent variable Dependent variable Flow of causation From To Equation 17 Services growth of Sindh Agriculture growth of Punjab No causation Equation 18 Agriculture growth of Punjab Services growth of Sindh 148
No causation Equation 19 Services growth of Sindh Agriculture growth of No causation Sindh Equation 20 Agriculture growth of Sindh Services growth of Sindh No causation Equation 21 Services growth of Sindh Agriculture growth of No causation KPK Equation 22 Agriculture growth of KPK Services growth of Sindh No causation Equation 23 Services growth of Sindh Agriculture growth of Unidirectional causality from services of Sindh to Balochistan agriculture of Balochistan Equation 24 Agriculture growth of Balochistan Services growth of Sindh No causation Equation 25 Services growth of Sindh Manufacturing growth of No causation Punjab Equation 26 Manufacturing growth of Punjab Services growth of Sindh Unidirectional causality from manufacturing of Punjab to services of Sindh Equation 27 Services growth of Sindh Manufacturing growth of Bidirectional causality exists between services of Sindh Sindh and manufacturing of Sindh Equation 28 Manufacturing growth of Sindh Services growth of Sindh Equation 29 Services growth of Sindh Manufacturing growth of No causation KPK Equation 30 Manufacturing growth of KPK Services growth of Sindh No causation Equation 31 Services growth of Sindh Manufacturing growth of No causation Balochistan Equation 32 Manufacturing growth of Balochistan Services growth of Sindh No causation Services Sector of KPK Independent variable Dependent variable Flow of causation From To Equation 33 Services growth of KPK Agriculture growth of Punjab No causation Equation 34 Agriculture growth of Punjab Services growth of KPK No causation Equation 35 Services growth of KPK Agriculture growth of No causation Sindh Equation 36 Agriculture growth of Sindh Services growth of KPK No causation Equation 37 Services growth of KPK Agriculture growth of No causation KPK Equation 38 Agriculture growth of KPK Services growth of KPK No causation Equation 39 Services growth of KPK Agriculture growth of No causation Balochistan Equation 40 Agriculture growth of Balochistan Services growth of KPK No causation Equation 41 Services growth of KPK Manufacturing growth of No causation Punjab Equation 42 Manufacturing growth of Punjab Services growth of KPK Unidirectional causality from manufacturing of Punjab to services of KPK Equation 43 Services growth of KPK Manufacturing growth of No causation Sindh Equation 44 Manufacturing growth of Sindh Services growth of KPK Unidirectional causality from manufacturing of Sindh to services of KPK Equation 45 Services growth of KPK Manufacturing growth of No causation KPK Equation 46 Manufacturing growth of KPK Services growth of KPK No causation Equation 47 Services growth of KPK Manufacturing growth of No causation Balochistan Equation 48 Manufacturing growth of Balochistan Services growth of KPK Unidirectional causality from manufacturing of Balochistan to services of KPK Services Sector of Balochistan Independent variable Dependent variable Flow of causation From To Equation 49 Services growth of Balochistan Agriculture growth of Punjab No causation Equation 50 Agriculture growth of Punjab Services growth of Balochistan No causation Equation 51 Services growth of Balochistan Agriculture growth of No causation Sindh Equation 52 Agriculture growth of Sindh Services growth of No causation Balochistan Equation 53 Services growth of Balochistan Agriculture growth of Unidirectional causality from services of
149
KPK Balochistan to agriculture of KPK Equation 54 Agriculture growth of KPK Services growth of No causation Balochistan Equation 55 Services growth of Balochistan Agriculture growth of No causation Balochistan Equation 56 Agriculture growth of Balochistan Services growth of No causation Balochistan Equation 57 Services growth of Balochistan Manufacturing growth of No causation Punjab Equation 58 Manufacturing growth of Punjab Services growth of No causation Balochistan Equation 59 Services growth of Balochistan Manufacturing growth of Unidirectional causality from services of Sindh Balochistan to manufacturing of Sindh Equation 60 Manufacturing growth of Sindh Services growth of No causation Balochistan Equation 61 Services growth of Balochistan Manufacturing growth of Unidirectional causality from services of KPK Balochistan to manufacturing of KPK Equation 62 Manufacturing growth of KPK Services growth of No causation Balochistan Equation 63 Services growth of Balochistan Manufacturing growth of No causation Balochistan Equation 64 Manufacturing growth of Balochistan Services growth of No causation Balochistan Source: the table is draw from the estimates by the author: see Annexure 1-A
150
Chapter 5
Impact of Monetary and Fiscal Policies on
Inter Provincial Growth Disparities The last chapter was focused to analyze co-integration of sectors within provinces and with the national economy. In this respect, the speed of adjustment of provincial economies and sectors were also identified. It was learned that major development focus was on the larger two provinces i.e. Punjab and Sindh. The small provinces were deprived to benefit in the development planning and process. The economies of the provinces and the national economy have gone through swings in the economic growth. As a result, some provinces suffered more than that of other provinces which has led to the emergence of deprivation and social tension among provinces, particularly among the small provinces periphery territories like FATA. The emerging issues like shortage of electricity and other energy inputs, concentration of federal policies on the large province, funding green revolution along the Indus Basin (in Punjab and
Sindh), emerging unemployment, double digit inflation and inequalities in the province of public goods as well as social infrastructure has led to regional economic issues. It is important to find out and analyze such sources of inequalities so that effective policies may be introduce for sustainable and equitable economic growth in all the regions of the country. There is hardly any comprehensive study which may have been focused to identify such sources of inequalities.
Therefore, this chapter will be focused to estimate the growth differentials between provinces and estimating the factors that contributes into the regional disparities in Pakistan. This chapter is divided in two sections; section one will analyze the impact of fiscal and monetary policies on 151
growth rates and growth differentials of the provinces. Section two consists upon the cross cutting edges that includes; i) the impact of terrorism on provincial growth rates and emerging growth disparities, ii) impact of electricity load shedding on the provincial growth rates and emerging growth disparities, iii) impacts of physical and social infrastructural development on provincial growth rates, iv) estimation of provincial capital output ratios and lastly, v) calculating the employment elasticity of the provinces to point out employment generation.
The table 5.1 below provides the growth differential between provinces by sectors. The value closer to one indicates no regional differences among provinces. The table reveals the extent of regional difference in the growth of provinces. The level of growth in Balochistan is very low in all sectors as compare to other provinces. The highest growth differential is for the industrial sector. The province is deprived from industrial development too.
Table 5.1: Growth differential across provinces and across sectors GDP (Provincial) Agriculture Industry Services Punjab / Sindh 1.00 0.92 1.4 1.12 Punjab / KPK 0.95 1.08 1.2 1.01 Punjab / Balochistan 1.27 0.98 0.6 1.47 Sindh / KPK 0.95 1.17 0.9 0.90 Sindh / Balochistan 1.27 1.06 0.5 1.32 KPK / Balochistan 1.34 0.91 0.5 1.46 Source: Author’s Calculation.
The larger provinces Sindh and Punjab were growing almost at a similar growth rates. The small province lagged behind Punjab and Sindh. The KPK province also experienced similar growth pattern.
The study has used panel data analysis technique to estimate the regional differences among provinces. It is important to find out whether inequalities of the GDP growth of provinces are fiscal or monetary phenomenon. Besides, it will be pointed out how the difference in terms of 152
fiscal and monetary policies, explains the GDP growth differences among provinces.31 The impact of Fiscal and monetary policies on GDP growth of provinces is estimated by using panel data from 1990 to 2015 and Fixed effect model is applied where cross section terms include GDP growth of provinces w.r.t time, which are regressed in a panel form on fiscal variables; monetary variables and control variables. The significance of the model is that it allows the intercept to vary across cross sectional terms and time period is fixed. It will provide highlights how the inclusion of a monetary and fiscal variables explains the difference of intercept across cross sections.
The fiscal variables include federal and provincial governments current and development expenditures. However, the monetary variables include money supply at provincial level and the control variable includes provincial consumption expenditures and provincial private investment expenditures. The investment at provincial level includes private investment, public investment and general public investment. The rationale of using only provincial private investment as the variable for investment is public investment and general public investment generally includes the developmental expenditures to the provinces so the impact of public and general public investment is already captured in provincial development expenditures.
The expenditure side model is built to capture the fiscal and monetary impacts on provincial
GDP growth rates. The important reason is that these expenditures affect the GDP growth of provinces.
The chapter is organized as under. Section 5.1 analyzes the growth determinants for Pakistan and their impact on growth differential among provinces. Section 5.2 provides evidences for the
31 It may be noted that the fiscal and monetary policies are formulated by the federal/central government. The outcome will indicate whether the financial decision of the federal government have affected the inequalities among provinces. 153
impact of terrorism32 and electricity load shedding on the GDP growth of the provinces. Finally, section 5.3 is focused to identify the impact of social capital on the provincial growth differential.
5.1 The Model: Determinants of Economic Growth An econometric model has been estimated to check robustness33 of variables. The robustness results are reported in table 5.2. It provides determinants of provincial GDP which have never done before. Fixed effect model is estimated to draw empirical evidences. Final estimated model and its results are discussed below.
The model is developed to analyze the impact of Fiscal and Monetary policies on regional disparities. Before analyzing the regional disparities, the significance and robustness of different parameters are checked. To do this, I used specific to general model approach; one by one variables are tested (see appendix 5.1), until the general model that includes possible fiscal and monetary variables. The variables that study has used are mostly used by various studies for analyzing the growth determinants for the economy of Pakistan34. The equations tested in fixed effect model are given in appendix 5.1:
The equation (6) below has been estimated by using fixed effect model, which has already been explained in methodology chapter. The study has used specific to general approach by adding new variable at different level (see appendix 5.1). This enables us to test the robustness of the variables. The models will give us the impact of fiscal variables (provincial and federal governments) along with some control variables on the GDP growth differentials of provinces.
32 It may be noted that Pakistan has been a major country affected by the terrorism due to war in Afghanistan. Its two provinces, Balochistan and KPK were the major victims for more than a decade. It badly affected Pakistan’s economy, particularly the above cited two provinces. Pakistan has to fight for terrorism for a year with Talibans to establish law and order in the country. 33 The regression provides better results than OLS since its results are based upon weighted results and outliers don’t affect the results. The model is provided in appendix 1 (5.1) 34 For details, see review of literature 154
The complete model 6 given below captures the impact of monetary variables as the provincial money supply is included to analyze the impact of monetary policy on provincial GDP growth rates, but before analyzing the regional disparities, I have tested the significance and stability of different parameters. To do this, I have used specific to general model approach that means, thus
I first applied a very simply model; with few variables and introduced other variables one by one until I comleted a general model that includes fiscal and monetary variables. The varibles that study has used are mostly the same as applied by various studies for analyzing the growth in
Pakistan; a simple model may be defined as following:
1
2
I have estimated the above equations by using fixed effect model, which has already explained in methodology chapter. The study has use specific to general approach by adding new variable at each level. This enable to test the robustness of the variables. The model 1 includes only the impact of provincial consumption expenditures on the GDP growth of provinces. Model 2 along with provincial consumption expenditures provincial private investment. The difference in the fixed effect coefficients highlights the ability of a province to reap benefit from the provincial investment. Model 3 incorporates the provincial current expenditures. Model 4 is estimated by adding provincial development expenditures. Model 5 is estimated by including the federal total expenditures. These models will give us the impact of fiscal variables (provincial and federal) along with some control variables on the GDP growth differentials of provinces. The model 6 is
155
capture the impact of monetary variables. The provincial money supply is included to analyze the impact of monetary policy on provincial GDP growth rates. The following Table 5.2 below shows the results for all the models.
Table 5.2: Results of the Growth models
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 C -0.030 -0.04 -0.05 -0.05 -0.06 -0.05 D(LOG(CONS)) 0.415 0.41 0.37 0.38 0.34 0.23 (0.000)* (0.000)* (0.000)* (0.000)* (0.000)* (0.000)* D(LOG(INV)) 0.14 0.13 0.13 0.13 0.092 (0.000)* (0.000)* (0.000)* (0.000)* (0.005)* D(LOG(PCE)) 0.11 0.11 0.08 (0.003)* (0.003)* (0.034)** D(LOG(PDE)) 0.03 0.03 (0.031)** (0.037)** D(LOG(FCE+FGD)) 0.15 0.103 (0.002)* (0.022)** D(LOG(PCE+PDE)) 0.078 (0.015)** D(LOG(MS)) 0.33 (0.000)* R-squared 0.362 0.439 0.488 0.513 0.56 0.62 No. of observations 104 104 104 104 104 104 Source: Author’s estimates
The table above 5.2 shows empirical results from model 1 to model 5. All variables are significant except provincial current expenditures in model 535. The first two models explain the impact of provincial growth of consumption expenditures and private investment on GDP growth. The result shows that 0.41 percent and 0.14 percent of GDP growth of provinces is explained by provincial growth of consumption expenditures and private investment, respectively. The model 3, includes growth of provincial current expenditures, the parameters are significant, the coefficient value is 0.11 percent.
35 It may be case that federal expenditures are in larger quantity and these expenditures may determine the direction of the economies. 156
Model 2 captures the impact of fiscal and monetary policy together. The total expenditures of provincial and federal government have been taken, as explanatory variables, and further the provincial growth of money supply is included in model 2. The model shows that all parameters are significant. When provincial growth of money supply is added into the model, the growth of provincial consumption expenditures has been reduced from 0.34 percent in model 5 to 0.23 percent in model 2. The provincial growth of money supply affects GDP growth of provinces by
0.33 percent. It can be seen that GDP growth of provinces are monetary phenomenon rather than fiscal.
5.2 Regional Growth Disparities Fiscal and monetary policies are adopted to attain a sustainable growth. Both the policies have short term and long-term impact on the growth of the economy. It is a long debate between Neo
Classical economists and Neo Keynesian economist, as first school of thought believe that intervention by the government into the economy help to close the demand and supply gap. The neo classical believe in market forces which automatically adjusted the imbalances by the market forces. Contrarily, the Neo Keynesian believe that the gap between demand and supply will meet in long run but in short run government has to intervene into the economy to correct the gap; either through demand driven policies or through supply driven policies. So, the concept of fiscal or monetary interventions is primarily given by Keynesian school of thought.
The economy of Pakistan since from start is having huge gaps in import and exports, saving and investment, budget deficits and eventually the external and internal debt burden has increased tremendously.36 Historically in most of the time periods, the economy of Pakistan had either adopted fiscal or monetary policies to correct the gaps and improve the overall economic
36 These issues are common phenomena for developing economies since there exists hardly competitive economies. For such issues in detail, see Chaudhary M.A and Aslam A (2018), forthcoming. 157
performance of the country. So, while analyzing the growth determinants and differentials of provincial economies of Pakistan, it is important to analyze the fact that either it is the fiscal policies or the monetary policies which may have reduced or increased disparities among provinces.37
The fixed effect coefficients of the above models are shown in below table 5.2. The values of the coefficients show that from Model 1 to Model 4, the fixed coefficients are relatively stable.
When fiscal expenditures are included in model 5 the coefficient values relatively show some improvements and the disparities among provinces are slightly reduced. However, in model 6 when the monetary variable is included the coefficient values changes significantly and the disparities among provinces increased. The large provinces Punjab and Sindh benefited more than the small provinces KPK and Balochistan, due to the inclusion of monetary variable. The final model 6 shows that GDP growth of the provinces are monetary phenomenon rather than fiscal and the growth differentials between large and small provinces increased due to the monetary factors. The economy of small province like Baluchsitan is hardly integrated into the financial markets. The monetary institutions are also not widely spread in the province.
37 The mega projects like dams and canals, as well as support for manufacturing sector was provided to the large provinces by establishing tax free industrial zones. However, several facilities were not available for small provinces. 158
Table 5.2: Fixed effect coefficients of the Growth models Provinces Fixed Effect Coefficients Model Model Model Model Model Model 1 2 3 4 5 6 Balochistan -0.011 -0.004 -0.005 -0.004 -0.004 -0.019 KP 0.008 0.008 0.008 0.007 0.007 -0.008 Punjab 0.003 -0.001 -0.001 0.000 -0.001 0.014 Sindh 0.000 -0.002 -0.003 -0.003 -0.003 0.013 Difference Punjab-Sindh 0.0022 0.0008 0.0022 0.0022 0.0019 0.0007
Difference Punjab-KP -0.0048 -0.0092 -0.0086 -0.008 -0.0077 0.0213
Difference Punjab- Balochistan 0.0134 0.0025 0.0043 0.0038 0.0031 0.0323 Sindh-KP -0.007 -0.01 -0.0108 -0.0102 -0.0096 0.0206
Sindh-Balochistan 0.0112 0.0017 0.0021 0.0016 0.0012 0.0316
Kp-Balochistan 0.0183 0.0117 0.0129 0.0118 0.0108 0.0111
Source: Author’s estimate
As explained in methodology, cross sectional dummy variables in the fixed effect model allows to estimate the impact of particular variables across cross sections. The table above (5.2) shows the fixed effect coefficients for each model. The difference between two coefficients shows the difference in growth in the presence of particular set of variables. Further, the change in the coefficients by introducing a new variable in the model gives the impact of that particular variable on the GDP growth of that particular province.
Each column in the above table shows the impact of a particular variable on the regional growth differential. The model is used as the base model and it may be said that it indicates growth differential among the provinces. The result shows that Punjab has a higher growth rate as
159
compare to Sindh and Balochistan, because the difference between fixed effect coefficient is positive. However, the growth of Punjab is less than KPK; as the values of coefficients from model 1 to 5 are negative and less than the growth of KPK. The final model i.e model 2 shows the coefficient differentials are positive showing that growth of Punjab was higher than KPK province.
5.3 Provincial Current Expenditure The model 3 is the first model that capture the impact of provincial current expenditure on the growth of the province.
Punjab and Sindh
According to model 2 the difference between Punjab and Sindh growth is 0.0008 percent this implies that the growth of provincial private investment expenditures affected positively more to the GDP growth of Sindh than GDP growth of Punjab. The provincial current expenditures added in model 3 the differential in GDP growth increased to 0.0022 percent.
Punjab and KPK
The GDP growth disparities between Punjab and KPK show that in basic model 2 growth differences was -0.0092 percent showing that GDP growth of KPK more than the GDP growth of
Punjab if we have same level of investment in Punjab and KPK.
When we have added the growth of provincial current expenditures model 3 the GDP growth differential reduced to -0.0086 percent in model 3 as compared to -0.0092 percent in model 2.
This shows that the provincial current expenditure has reduced the growth differential among
Punjab and KPK.
Punjab and Balochistan
160
The GDP growth difference of Punjab and Balochistan in basic model 2 show that the GDP growth difference between Punjab and Balochistan was 0.0025 percent. The model 3 shows that the provincial current expenditures have increase the growth disparities between Punjab and
Balochistan. The growth differential has slightly raised to 0.0043 percent. This means provincial current expenditures effect GDP growth of Punjab more than the GDP growth of Balochistan.
Sindh and KPK
The GDP growth difference of Sindh and KPK was -0.01 in model 2 percent indicating that growth of provincial private investment expenditures is affecting GDP growth of KPK more than
Sindh. Further, when growth of provincial current expenditure is added in model 3 the disparity further increases to -0.0108 percent indicating that provincial current expenditures are effecting more significantly to the GDP growth of KPK than Sindh.
Sindh and Balochistan
The GDP growth differentials of Sindh and Balochistan is 0.0017 percent in model 2 and by adding growth of provincial current expenditures it has increased and become 0.0021 percent in model 3. It means that provincial growth of current expenditures increases GDP growth of Sindh over Balochistan.
Balochistan and KPK
The GDP growth difference of KPK and Balochistan in model 2 is 0.0117 percent and when growth of provincial current expenditure is added in the model 3 the GDP growth disparities increased to 0.0129 percent indicating that growth of provincial current expenditures is effecting
GDP growth of KPK more than Balochistan.
The table 5.3 gives the per capita expenditures by province from 1990 to 2015. It is important to note that the Punjab has the lowest per capita expenditures and Balochistan has the highest per
161
capita expenditures both in 1990 and 2015. This shows that the province with the higher level of current expenditure is not having that much GDP growth rate this shows that the structural differences the efficiency in utilizing the expenditure is very important for the growth.
Table 5.3: Per capita current provincial expenditures 1990-2015 (Rs.)
1990 2015 Average annual Growth Punjab 421 (4th)* 6,374 (4th) 11.5 (3rd) Sindh 530 (3rd) 8,335 (3rd) 11.6 (2nd) KP 661 (2nd) 8,598 (2nd) 10.8 (4th) Balochistan 826 (1st) 13,325 (1st) 11.8 (1st) *ranking on the basis of per capita expenditures and rate of growth. Source: Handbook of Pakistan Economy and Author’s estimate 5.4 Provincial Development Expenditure The model 4 in the table 5.2 shows the impact of provincial development expenditures on the regional growth differential among provinces.
Punjab and Sindh
The growth differential between Punjab and Sindh in model 3 remained unchanged by adding the provincial development expenditure into model 4. So, development expenditure does not change the growth differential among Punjab and Sindh.
Punjab and KPK
The GDP growth disparities between Punjab and KPK in model 3 was -0.0086 percent and by adding the provincial development expenditure it has decreased to -0.008 percent. This shows that the provincial development expenditure has reduced the growth differential among Punjab and KP.
Punjab and Balochistan
The GDP growth difference of Punjab and Balochistan in model 3was 0.0043 percent and when growth of provincial development expenditure is added in model 4 the GDP growth disparities
162
reduced to 0.0038 percent. This means provincial development expenditure effect GDP growth of Punjab more than the GDP growth of Balochistan.
Sindh and KPK
The GDP growth difference of Sindh and KPK was -0.0108 percent in model 3 and after adding provincial development expenditures the difference between GDP growth of two provinces has decreased to -0.0102 percent in model 4.
Sindh and Balochistan
The GDP growth differentials of Sindh and Balochistan in model 3 was 0.0021 percent and it has become 0.0016 in model 4 by adding provincial development expenditures.
Balochistan and KPK
The GDP growth difference of KPK and Balochistan in model 3 was 0.0129 percent and it has become 0.0118 in model 4. This indicates that growth of provincial development expenditures is affecting GDP growth of KPK more than Balochistan.
The above analysis shows that the provincial development expenditure has decrease the regional difference significantly and the main reason is the higher development expenditures by small provinces like Balochistan and KPK. The table 5.4 given below the per capita development expenditures by four provinces from 1990-2015. The growth in per-capita development expenditure is highest in Sindh followed by KPK, Punjab and Balochistan for the time period of
1990-2015.
163
Table 5.4 Per Capita provincial development expenditures from 1990-2015 (Rs.)
1990 2015 Growth Rate(%) Punjab 122 2,019 11.9 Sindh 129 2,892 13.2 KP 225 4,072 12.3 Balochistan 436 5,126 10.4 Source: Handbook of Pakistan Economy.
5.5 Federal Expenditures After looking at the impact of provincial expenditure now we are looking at the federal expenditures and their impact on regional disparities in Pakistan. Model 5 in tables 5.1 and 5.2 give the parameters and fixed effect coefficients for the federal Expenditure.
The model 4 in the table 5.2 shows the impact of provincial development expenditures on the regional growth differential among provinces.
Punjab and Sindh
The growth differential between Punjab and Sindh in model 4 was 0.0022 percent and by adding federal expenditure the growth differential decreased to 0.0019 percent in model 5. This shows that the federal development expenditures has reduced the growth differential between Punjab and Sindh.
Punjab and KPK
The GDP growth disparities between Punjab and KPK in model 4, is -0.0080 percent. The federal expenditure has reduced the GDP growth disparities between Punjab and KP to -0.0077 percent in model 5.
Punjab and Balochistan
164
The GDP growth difference of Punjab and Balochistan in model 4, is 0.0038 per cent and it has decreased to 0.0031 percent in model 5. So the federal expenditure has reduced the regional difference between Punjab and Balochistan.
Sindh and KPK
The GDP growth difference of Sindh and KPK was -0.0102 percent in model 4 and it has decreased to -0.0096 percent in model 5.
Sindh and Balochistan
The GDP growth differentials of Sindh and Balochistan in model 4 was 0.0016 percent and it has decreased to 0.0012 percent in model 5.
Balochistan and KPK
The GDP growth difference of KPK and Balochistan in model 4 was 0.0118 percent and it has decreased to 0.0108 percent in model 5. The growth differential among Balochistan and KP has decreased due to federal government expenditures.
The above analysis reveals that the federal government expenditures reduced the regional growth differential between provinces. The findings by Khan, R. and B. Jabeen Hashmi (2015) also confirm that federal government expenditures are very important for reducing the income inequality among household
Punjab and Balochistan
The GDP growth difference of Punjab and Balochistan in basic model 2 show that the GDP growth difference between Punjab and Balochistan was 0.0025 percent. The model 3 shows that the provincial current expenditures has increase the growth disparities between Punjab and
Balochistan. The growth differential has slightly increased to 0.0043 percent. This means
165
provincial current expenditures effect GDP growth of Punjab more than the GDP growth of
Balochistan.
Sindh and KPK
The GDP growth difference of Sindh and KPK was -0.01 in model 2, indicating that growth of provincial private investment expenditures is affecting GDP growth of KPK more than Sindh.
Further, when growth of provincial current expenditure is added in model 3, the disparity further increases to -0.0108 percent; indicating that provincial current expenditures are affecting more significantly to the GDP growth of KPK than Sindh.
Sindh and Balochistan
The GDP growth differentials of Sindh and Balochistan is 0.0017 percent in model 2 and by adding growth of provincial current expenditures, it has increased to 0.0021 percent in model 3.
It means that provincial growth of current expenditures increases GDP growth of Sindh more as compared to Balochistan.
Balochistan and KPK
The GDP growth difference of KPK and Balochistan in model 2 is 0.0117 percent and when growth of provincial current expenditure is added in the model 3 the GDP growth disparities increased to 0.0129 percent indicating that growth of provincial current expenditures is effecting
GDP growth of KPK more than Balochistan.
The major chunk of development expenditure is with the federal government and it also implement mega projects. Therefore, it is important to analyze the federal expenditures and their impact on regional disparities in Pakistan. Model 5 in tables 5.4 and 5.3 provides the results of fixed effect coefficients for the federal Expenditures.
Impact of Federal Public Expenditure
166
The growth differential between Punjab and Sindh in model 4 was 0.0022 percent and by adding federal expenditure the growth differential decreased to 0.0019 percent, in model 5. This shows that the federal development expenditures have reduced the growth differential between Punjab and Sindh, which are the largest provinces.
The GDP growth disparities between Punjab and KPK in model 4, is -0.0080 percent. The addition of federal expenditure has reduced the GDP growth disparities between Punjab and KP to -0.0077 percent in model 5.
The GDP growth difference of Punjab and Balochistan in model 4, is 0.0038 per cent and it has slightly decreased to 0.0031 percent in model 5. So the federal expenditure has slightly reduced the regional difference between Punjab and Balochistan.
The GDP growth difference of Sindh and KPK was -0.0102 percent in model 4 and it has slightly decreased to -0.0096 percent in model 5.
The GDP growth differentials of Sindh and Balochistan in model 4 was 0.0016 percent and it has also slightly decreased to 0.0012 percent, in model 5.
The GDP growth difference of KPK and Balochistan in model 4 was 0.0118 percent and it has also slightly decreased to 0.0108 percent, in model 5. The growth differential among Balochistan and KPK has decreased due to federal government expenditures, but this affect was not very significant.
The above analysis reveals that the federal government expenditures slightly reduced the regional growth differential between provinces. The findings by Khan, R. and B. Jabeen Hashmi
(2015) also confirm that federal government expenditures are very important for reducing the
167
income inequality among household. The evidences that indicate that federal public expenditures did not lead to disparities among provinces; rather it helped to slightly reduce it.
5.6 The Impact of Money Supply growth and Financial Market The financial market and money supply may have impact on regional growth difference. The empirical evidences indicated that the money supply has a significant impact on the GDP growth, but its impact on regional difference needs to be explored. The model 6 in the table 5.2 provides the impact of money supply on regional growth difference in Pakistan. The details are given below
The growth differential between Punjab and Sindh in model 5 was 0.0019 percent. The growth differential has further decreased to 0.0007 percent, when money supply was added to the model.
This shows the money supply effects the growth of Punjab more as compare to growth of Sindh; the growth differentials have been reduced significantly between the two large provinces.
As well as Punjab and KPK is concerned, the GDP growth disparities between Punjab and KPK in model 5 was, -0.0077 percent, but when money supply is added to the model, the GDP growth difference between Punjab and KPK has increased to 0.0213 percent, in model 6, showing that monetary policy benefited GDP growth of Punjab more than KPK. The drastic positive impact of monetary policy upon growth of Punjab has increased the growth differentials at higher rate between Punjab and KPK. It was pointed out that Punjab being the larger province, it has relatively developed financial market, which turn out to be contribute more to its growth. As a result, the smaller province (KPK) was affected and regional disparities increase between these provinces.
The GDP growth difference of Punjab and Balochistan in model 5 was, 0.0031 percent and it increased to 0.0323 percent in model 6; after including money supply variable. That again shows
168
that money supply benefited Punjab more than Balochistan. The case of monetary policy with
Punjab and Balochistan is similar to that of Punjab and KPK. The monetary policy has significantly increased the growth of Punjab, which resulted into the increasing of growth differentials between Punjab and Balochistan. In nutshell, larger province (Punjab) and small province KPK and Balochistan happen to increase inequalities in growth due to different level of money supply and development of financial markets.
The GDP growth difference of Sindh and KPK was -0.0096 percent in model 5 and it has increased to 0.0206 percent in model 6. This shows that the monetary policy benefited Sindh more than KPK. The growth differentials between Sindh and KPK increased due to monetary phenomenon.
The GDP growth differentials of Sindh and Balochistan in model 5 was 0.0012 percent and it increased to 0.0316 percent in model 6. Such increase in growth differentials reveal the impact of monetary policy on provincial growth of Sindh as compared to Balochistan. The above cited evidences indicated that growth differential increased between Sindh, KPK and Balochistan.
The GDP growth difference of KPK and Balochistan in model 5 was 0.0108 percent and it almost remained same i.e. to 0.0111 percent in model 6. Thus monetary phenomenon was not very affective in the small provinces.
The above analysis indicated that the monetary policy has increased the growth differential among provinces. As already pointed out that that the fiscal policy did not increase the growth inequalities between provinces in Pakistan. However, the monetary policy has widened the growth difference between provinces; particularly the small provinces did not benefit from it.as a result growth differential increased among small provinces and larger provinces.
Development of Financial Market and Banking in Provinces 169
The following table 5.5 below shows the variation in the financial development in the Provinces.
It is important to note that there are significant variations in deposits and advances between provinces. The distribution of advances is more skewed towards large provinces as compare to deposits. Moreover, the growth of deposits and advances are high in Punjab and Sindh as compared to, KPK and Balochistan. This justify earlier findings regarding the impact of monetary policy and financial market on growth of provinces.it also indicates that informing financial markets in small province could benefit them, and it could also help to reduce growth differential among provinces.
Table 5.5: Share of provinces in deposits and advances (%). Deposits Advances Punjab 53.1 55.5 Sindh 36.0 43.1 KP 8.5 1.2 Balochistan 2.3 0.2 Source: Handbook of Pakistan Economy SBP and estimated by author
In brief, the above analysis reveals that the fiscal policy did not contribute to regional difference in terms of growth among provinces in Pakistan, but the monetary variables has lead to increase in the regional growth differences in Pakistan. The main reason is the financial development across provinces.
The summary of the fiscal development and monetary development on either increasing or decreasing the growth differentials among regions (table 5.1 appendix).
5.7 Summary / Conclusions of Fiscal and Monetary determinants of growth The empirical evidence indicated the fiscal development could slightly help to reduce growth differentials among provinces. The fiscal expenditure policies adopted by federal government almost equitably affected all provinces.
170
However fiscal expenditure policies opted by the provincial governments could reduce growth differentials among provinces; if these are made a formulated carefully. The provincial expenditures are by respective provincial governments has increased growth differentials between Punjab, Sindh and small provinces. The increase in growth differentials between Punjab and Sindh appear to be the result of an increase in provincial expenditures. Punjab is the largest province among all provinces, so to carry out provincial affairs high amount of the provincial expenditures are spent on current expenditures. These current expenditures do not contribute significant to the growth of the provincial economy. Therefore, its impact towards growth differential was not in this respect. The coefficients of provincial development expenditures indicated that by adding provincial development expenditures the parameter values from Punjab to Sindh remained unchanged. There is also a need to improve development expenditures for all provinces, particularly, small provinces need to bring in to the main framework of development.
The monetary variables show entirely different picture as shown by fiscal variables. The inclusion of monetary variable in the model, it indicated a reduction growth differential between
Punjab and Sindh. However, the growth differentials between Punjab and two small provinces
KPK and Balochistan have increased. Moreover, the growth differentials between Sindh and other two small provinces KPK and Balochistan have also increased. Even the growth differentials between the two small provinces have increased too. Thus, monetary variables explained more about growth differential among provinces.
This analysis revealed two important findings. Firstly, fiscal policies did not contribute growth differentials between provincial economies. However, monetary policies increased growth differentials between large provincial economies, and also between large and small provinces.
171
The above findings help to improve fiscal and monetary policies for provincial economies. The governments should adopt fiscal measures more to reduce provincial growth inequalities.
Moreover, effective use of both fiscal policy and monetary policy can help to improve growth differential among provinces. It has also been pointed out that financial developments in Punjab and Sindh have led to improve the economies of large provinces. So to achieve overall high growth and for equitable distribution of the fruits of economic growth it is important to improve more fiscal measure for all provinces. However monetary policy as it is centrally controlled and easily managed in comparison to fiscal policy, actually shows its benefits to small provinces. To reduce these growth differentials between large and small provinces, in most of the cases government may opt for policy mix; along with monetary and fiscal policy.
172
Appendix 5.1 Table 5.1: Impact of Fiscal and Monetary variables on the regional growth differential
Provincial Federal Money Supply Expenditures Expenditures (Current Development) Difference Punjab-Sindh Increase Decrease Decrease Difference Punjab-KP Decrease Decrease increase Difference Punjab- Decrease Decrease Increase Balochistan Sindh-KPK Decrease Decrease Increase Sindh-Balochistan Decrease Decrease Increase KPK-Balochistan Decrease Decrease Increase
Source: compiled by the Author
173
Chapter 6
Impacts of Economic Shocks on Inter-ProvincialGrowth Disparities The economy of Pakistan and its provinces (regions) have gone through severe economic shocks; due to international terrorism, energy shocks, unemployment, law and order situation, poor institutions, corruption, and ultimately poor outcome of social sectors development and insignificant inclusive growth.38 In this chapter, the impact of terrorism, electricity load shedding and physical and social infrastructure on provincial growth differentials will be analyzed. There is no study which may have focused on these issues. In the first section, an analysis of the impacts of terrorism on economic growth of provincial economies is analyzed. The second part will focus on the impact of electricity load shedding on the economic growth of provinces. In the third part, the role of physical and social infrastructure will be explored for the provincial economies. In the light of above, growth differentials were also identified. It will highlight up to what extent terrorism, electricity load shedding and physical, as well as, social infrastructure have created growth disparities among provinces. The very reason to estimate electricity load shedding and terrorism separately, on the growth of provincial economies, is that; both issues have recently emerged in the last one decade, so it is important to explore that, which issue either electricity load shedding or terrorism has effected how much to the provincial economies. These events may be an additional source of growth disparities among provinces. So far, there is hardly any comprehensive study which may have been focused to analyze the above losses. New
38 For details, see Zulfiqar K. (2017),Ph. D Thesis, Punjab University, and Zulfiqar K. and Chaudhary M. Aslam (2017), PEASR. 174
findings in this respect will provide policy guidelines to cope with the new emerging regional inequalities in Pakistan.
6.1 Emergence of Terrorism Terrorism is one of the important factors which have significantly affected the growth of all provincial and national economies.39 Over the last one decade, the terrorist attacks have increased in almost all provinces.40 It has not only affected the domestic investment but also resulted in the decrease in foreign investment, as well as, it has become a challenge to maintain law and order. In Pakistan, over 50 thousand civilians and army persons died due to these attacks.
The first part of this section will estimate the impact of terrorism and its intensity on the growth of provincial economies. Moreover, it will also analyze the growth differentials between provinces that to what extent terrorism have contributed in increasing growth differentials among provinces.
6.2 Electricity (Energy) Shortage & Load Shedding The economy of Pakistan is facing serve energy crisis from the past one decade. It has crippled the economy, particularly manufacturing sector suffered due to shortage of energy input. On average, every day, the nation has to suffer from electricity shortage and its load shedding. The growth rates of agriculture, manufacturing and services sectors are severely affected by the non availability of electricity. The manufacturing sector of Pakistan has lost its competitiveness in the international markets; due to high costs of inputs, which are mainly the result of shortage of
39 As per one estimated, $100 Billion losses occurred due to terrorism. For further details, see the estimates of this study. 40 It may be noted that terrorism in Pakistan emerged as a result of Pakistan’s support for US army in Afghanistan, after 9/11 incidence. 175
electricity and its high charges. The large provinces of Punjab and Sindh have better industrial base, as compared to KPK and Balochistan. So the impact of shortfall in electricity is highly captured by low production of medium and large scale industrial units in these provinces. On the other hand, production activities and businesses were badly affected by the terrorist attacks in
KPK and Balochistan. Thus, the output losses due to terrorism and shortfall of electricity have not only affected the provincial growth rates but it also increased the growth differentials among provinces. So the extent to which electricity load shedding has affected the provincial growth rates and created disparities among provinces is analyzed in the second part of this chapter.
6.3 The Role of Physical and Social Infrastructure The physical and social infrastructure plays key role in economic development of any economy.
The state of social infrastructure is not very good in Pakistan; as compared to physical infrastructure. One of the important reasons for poor such conditions is the allocation of meager budget to development expenditures; as compared to non-development expenditures of the federal and provincial governments. Therefore, the performance of social indicators is very poor in all provinces of the country.41 Moreover, some provinces have received more investments in physical infrastructure development, as compared to other provinces i.e. small provinces were neglected in this respect. The KPK province is trying hard to develop its physical infrastructure and the province has performed better than other provinces but it suffered the most from terrorist attacks. The province of Balochistan is far behind than the large provinces in infrastructural development. So to capture the impact of physical and social infrastructure on the provincial
41 For example, literacy rate is below 60% and health facilities are also very poor. The health and education expenditures hardly exceeded 2.5% of GDP. Half of the population does not have sanitary facilities and even one third of the population does not get safe drinking water. For more details, see Pakistan Economic Survey 2016-17. 176
growth differentials is analyzed in this chapter. In addition to above, the role of monetary and fiscal policies will also be analyzed.
6.3.1 Terrorism and GDP Growth of Provinces Pakistan economic survey (2010-11) reported that paying a heavy cost both in direct and indirect terms for the terrorism, as the direct and indirect cost o economy has been raised from $2.669 billion in 2001-2002 to $13.6 in 2009-10.42 Furthermore, the investment to GDP ratio has also fell drastically; as from 22.5 percent of GDP in 2006-07 to 13.4 percent of GDP in 2010-11. It is still not over 15% of GDP (2016-17). As a result, economic growth also slowed down during the period. Pakistan is also suffering from this problem from decade or so. The repeated terrorist attacks on Pakistan have severely affected its economic growth. Pakistan is still supporting around 3 million refugees from Afghanistan, due to terrorism. Pakistan needs strong measures and resources to improve the economic crisis i.e. also to control terrorism for rehabilitation of refugees and investment in country. Pakistan has incurred high direct and indirect costs for the terrorist attacks. The special section in Pakistan Economic survey (2015-16) reported that
Pakistan has adopted a 20 points National Action Plan that includes different preventive measures to control terrorism in the country. Moreover, Pakistan completed Zarb-e-Azab mission against terrorism without any discrimination. The mission Zarb-e-Azb has achieved most of its targets in reducing terrorism in the country and improving the overall national security situation. However, recovery of the economy is still to be seen. Pakistan spent billions of rupees to meet military and other security expenditures, which significantly reduced
42 The terrorist attacks and law and order deterioration was emerged as a result of 9/11 attack of USA and Afghanistan. Pakistan supported US attacks which turned Taliban against Pakistan. It severely affected Pakistan’s economy, which is analyzed in the next section. 177
development expenditures in Pakistan. The below table 6.1 highlights the losses due to terrorism in different sectors of the economy during 2014 and 2015.
Table 6.1: Losses of terrorist Attacks (US$ Billions) Years Years S. No Sectors 2014-15 2015-16* Total 1 Exports 1.08 0.80 1.88 2 Compensation to affects 0.04 0.01 0.05 3 Physical infrastructure 0.12 0.07 0.19 4 Foreign investment 4.56 2.04 6.60 5 Privatization 0.01 0.00 0.01 6 Industrial output 0.02 0.01 0.03 7 Tax collection 2.94 2.32 5.26 8 Cost of Uncertainty 0.03 0.01 0.04 9 Expenditures over run 0.40 0.28 0.68 10 Others 0.04 0.02 0.06 Total losses 9.24 5.56 14.80 *Estimated on the basis of 9 months actual data (July-March) Source: Pakistan Economic survey, 2015-16
The table 6.1 above indicates that most of the macroeconomic indicators deteriorated in 2015, as compared to 2014 due to terrorism. The exports fell from $ 1.08 billion in 2014 to 0.80 $ billion in 2015. The foreign investment decreased from $ 4.56 billion in 2014 to $ 2.04 billion in 2015.
The industrial output also decreased during this period. The economy of Pakistan, in 2014, has suffered losses of $ 9.24 billion in different sectors of the economy; due to terrorist attacks.
These losses were around$5.56 in 2015. However, within two years 2014 and 2015, the economy of Pakistan lost around $15 billion, which is a huge loss for the economy. The Pakistan
Economic Survey 2015-16 reported that Pakistan has incurred a direct and indirect cost of
$118.31 billion, after 9/11 event, which is Rs. 9869.16 billion in the last 14 years; due to terrorist attack. The USA has paid partially for the cost of military operation. But still, economic losses have to be borne by Pakistan, which had badly affected Pakistan’s economy and development. It
178
is in addition to over 40,000 lives lost in this war. The table 6.2 shows the total cost incurred by
Pakistan due to terrorism since 2001-02 to 2015-16
Table 6.2: Cost of terrorism to Pakistan
Years Billion $ Billion Rs. % Change 2001-02 2.67 163.90 2002-03 2.75 160.80 3.0 2003-04 2.93 168.80 6.7 2004-05 3.41 202.40 16.3 2005-06 3.99 238.60 16.9 2006-07 4.67 283.20 17.2 2007-08 6.94 434.10 48.6 2008-09 9.18 720.60 32.3 2009-10 13.56 1136.40 47.7 2010-11 23.77 2037.33 75.3 2011-12 11.98 1052.77 - 49.6 2012-13 9.97 964.24 - 16.8 2013-14 7.70 791.52 - 22.8 2014-15 9.24 936.30 20.0 2015-16* 5.55 578.20 - 39.9 Total 118.32 9869.16 *Estimated on the basis of 9 months actual data (July-March) Source: Pakistan Economic survey, 2015-16, GoP, Ministry of Finance.
The above table 6.2 indicates that terrorism has imposed heavy cost on the economy of Pakistan. Pakistan has faced repeatedly terrorist attacks in major cities of all provinces, for more than a decade. Pakistan has paid total cost of $ 118.32 billion between 2001 and 2015. It is almost around one year’s GDP of Pakistan. Moreover, Pakistan’s economy is in crisis since 2000-01, and has not rehabilitated yet.
6.4 Impact of Terrorism on Regional Disparity The following section is focused to analysis impact of terrorism and electricity load shedding on the provincial growth differentials. The analysis will highlight intensity of the issues and related policy measures will be identified for improving upon the emerging issues. Firstly, the impact of terrorism on GDP growth differentials, between provinces is analyzed. Secondly, the impacts electricity load shedding, on industrial value added of the provinces, is highlighted.
179
6.4.1 The model and Empirical Findings The following model has been estimated to estimate impacts of terrorism on the economy of Pakistan.
Killsi.t is the number of people killed in terrorist activities in each province and lnGDPi.t is the log of GDP at constant factor cost of 2005-06 of each province. The AR term in the model is to introduce to control the problem of auto correlation and impact of all other factors that may affect the GDP.
The results of the first model of terrorism and provincial growth differentials are reported below:
Table 6.3: Results of model of Terrorism and Growth Independent variable Coefficient Stand. error t- stat. p-value Constant 17.075 3.067 5.567 0.0000 Growth of number of person -0.00002 0.000006 -2.2593 0.0265** killed in terrorist attacks (G Kills) AR (1) 0.987 0.0125 79.097 0.0000 R-squared 0.9985 No. Of =88 observations *Significant at 5% or less. Source: Estimated by the author, also see Annexure 2B The above table 6.3 shows the results of the growth of number of person killed in terrorist attacks, which are used as a proxy of terrorism. The value of the coefficient is significant at less than 5 percent level of significance. The sign is also as per expectation i.e. negative. Since log of
GDP is a non-stationary series, so autoregressive (AR) process is introduced, which make the log of GDP series stationary and captures variations in GDP due to other factors. The results show that terrorism significantly affected the GDP growth of Provinces.
180
A study by Padda I., and et.al (2015) also analyzed the impact of terrorism on economic growth of Pakistan. The authors used the data from 1981 to 2012,per capita GDP growth was regressed on population growth rate.43 The study found that terrorism is significantly affecting the economy of Pakistan and the sign of coefficient is negative, showing its negative impacts to economic growth. The other costs associated with terrorism are infrastructural loss, reduction in foreign investment, capital flight and low public revenues and reduction in development expenditures. The study showed that one percent increase in terrorist incidents reduces the GDP per capita by 0.39 percent. Thus, this study confirms and supports the findings of earlier study that is significantly affected the provinces of Pakistan.
6.4.2 Terrorism: Fixed effect coefficients of the provincial Growth
The impact of terrorism on GDP growth and further how terrorism has increased disparities between provinces is given in table 6.4 below.
Table 6.4: Fixed effect coefficients of the provincial growth with terrorism Provinces Fixed effect Coefficients Balochistan -2.18 KPK -0.07 Punjab 1.36 Sindh 0.75 Difference Punjab-Sindh 0.61 Difference Punjab-KPK 1.43 Difference Punjab-Balochistan 3.54 Sindh-KPK 0.82 Sindh- Balochistan 2.93 KPK- Balochistan 2.11 Source: Estimated by the author, also see Annexure 2B for details.
43 The variables regressed were human capital, trade openness, and foreign aid. 181
The table 6.4 shows the fixed effect coefficients of the model. In the above table, difference between two coefficients shows the difference in growth in the presence of particular set of variables. The fixed effect coefficients show that terrorism has affected the GDP growth of
Balochistan more than all other provinces, as fixed effect coefficient is -2.18. After Balochistan terrorism has affected GDP growth of KPK more than Punjab and Sindh. Moreover, it has affected Sindh more than Punjab, as fixed effect coefficient of KPK is -0.07 for Sindh is 0.75 and for Punjab is 1.36. It may be said due to terrorism; regional disparities have increased in
Pakistan.
In short, it may be stated that terrorism, which is continuously effecting for more than a decade has slowed down the growth of all provinces. However, Balochistan and KPK were the more victims of terrorism, than other provinces. Thus, it has increased further growth inequalities between provinces. Already neglected small provinces were also more victims of terrorism, as compared to large provinces. The small provinces are more vulnerable to terrorist attacks for two important reasons. Firstly, the GDP growth of small provinces in low in overall national economy, so if any terrorist attack happens in small provinces, it will have large impacts on their growth and development. Secondly, the small provinces, KPK and Balochistan, are more easy prey for terrorist due to their geographical locations.44 Since KPK has large border with FATA, so immigrants from FATA are difficult to control by provincial government of KPK. Similarly,
Balochistan shares its boundaries with Afghanistan, therefore refugees from Afghanistan were difficult to monitor due to having same culture and ethnic identity as well as limited provincial government resources of small provinces was also an obstacle to control terrorism. Thirdly, the
44 These provinces were on the border of Afghanistan, as well as, border population had their strong links with Afghans. Therefore, it was easy to attack on them. Moreover, the flow of refugees was also towards these small provinces too. 182
lack of law and order situation and security agencies control over terrorism were weak in KPK and Balochistan, as compared to Punjab and Sindh.
The military agencies were working in KPK and Balochistan but still it is like a proxy war between Pakistan army and militants, which still continues. It has become a continuous threat for
Pakistan.
6.5 Electricity Load Shedding and Industrial Value added, by Provinces Energy is a key and important input for all manufacturing activities. Energy shortage has emerged as a major issue for Pakistan’s economy. The progress of nations and their survival is limited by ensuring continuous supply of needed energy. Without efficient supply of energy, it will be rather impossible for a country to be competitive or to succeed in the current era. Thus, while analyzing for growth of any country or region it is important to identify its energy demand and supply. Pakistan is suffering from severe energy crisis, especially in electricity and gas, since
2006. These crises started getting deeper and crippled the economy, as a result, the growth rates of all sectors was affected due to energy shortage. Electricity and gas is rationed for industry and for other consumption. On average, the shortage of energy is so much that there is electricity load shedding for more than 15 hours per day.45 Heavy investment is needed to combat this issue.
Presently, government is working both on demand management policies and also trying to improve its supply. Under CPEC, around 10,000KV of electricity has been added to thenational grid but the energy shortage has not yet been met.
45 The electricity load shedding was over twelve hours in urban areas and around 18 hours in the rural areas. It not only affected small and large industrial production but also affected agricultural sector badly. 183
Table 6.5: Sectoral contribution towards GDP growth Sector 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 P Agriculture 0.05 0.43 0.79 0.57 0.53 0.53 -0.04 Industry 0.71 0.95 0.54 0.13 0.92 0.99 1.4 Manufacturing 0.19 0.34 0.28 0.61 0.76 0.53 0.68 Services 1.81 2.24 2.51 2.95 2.6 2.52 3.35 Real GDP (FC) 2.58 3.62 3.84 3.65 4.05 4.04 4.71 Source: Pakistan Economic Survey 2015-16
The below table 6.5 shows the percentage contribution by provincial GDP towards national. It also shows GDP percentage of electricity consumed by the provinces in years 2014-15.
Table 6.7: GDP per unit of Electricity by province in Pakistan Year 2014-15 GDP (%) Electricity Contribution consumption GDP per unit of to National GDP (Rs Million) (million kwh) Electricity (Rs.) GDP) Provinces Price Punjab 5,771,318 47,805 120.7 54
Sindh 3,152,458 17,688 178.2 30
KPK 1,220,903 8,698 140.4 11
Balochistan 487,045 3,922 124.2 5
Source: IPR (2015), NEPRA state of industry report, 2016
The table 6.5 indicates that Punjab and Sindh have the largest contribution towards the national
GDP and their consumption of electricity is also high, as compared to KPK and Balochistan.
Since Punjab contributes 54 percent to the national GDP and Sindh contributes 30 percent to the total GDP, therefore, the electricity consumed by these two large provinces is also high; as 61 percent for Punjab and 23 percent for Sindh. It may one of the reasons that, in the present study, the fixed effect coefficients of Punjab and Sindh are almost equally effected by electricity load shedding. If these two large provinces are provided with less electricity availability, it will eventually decrease their share in national GDP. The small provinces KPK contributes 11 percent to national GDP and it consumes 11 percent of the electricity. It indicates that in KPK,
184
still the demand for electricity availability is relatively less than that of Punjab and Sindh.
Secondly, of the industrial units in KPK are of small and medium sized that requires less of the electricity input, but eventually contributes less to the national GDP.
In the case of Balochistan, being smaller province, it only contributes 5 percent to the national
GDP and it consumes 5 percent of the total available electricity. In Balochistan, large manufacturing units are very few, mostly industries are cottage or small scale industries that do not require much of the electricity. This is the important reason as, why contribution of
Balochistan’s GDP is so low in national GDP. Moreover, the fixed effect coefficient, in case
Balochistan for electricity load shedding, turned to be positive in our present study. It means the electricity consumption is very less in industry output of the province and it further validate that the industry of Balochistan is so small that it failed to capture the negative effects of electricity load shedding.
6.5.1 Impact of Energy Crisis After have analyzed the importance of commodity producing sectors and their share in overall
GDP, it is now important to have the analysis that whether the recent energy crisis have affected these sectors and to what extent.
Pakistan is suffering from severe energy crises; especially in electricity and gas. Since from
2006, these crises started getting deep had crippled the economy as the growth rates of all sectors decreased due to energy shortage. The shortage of energy is so much that there is electricity load-shedding for more than 16 hours per day. IPP (2012) estimated a cost of load-shedding to the economy is 7 per cent of the GDP and more than 1 trillion Rs. This study also shows that not only the cost of load-shedding to the economy is high but it also varies from province to province. Table 6.7 below gives the cost of load-shedding to manufacturing sector of Pakistan.
185
One important thing to note that Sindh has the lost number of hours of outages but the cost of outages is very high. This is due to the fact that industry in Sindh is more electricity intensive as compare to other provinces. One other thing is that not only hours of outage but the cost of outages also varies from province to province. So, it is important to analyze the impact of load- shedding on the growth of the province. Does the load-shedding effect a particular province more than other provinces? In this study, the effects of electricity load-shedding on the manufacturing sector of each province are explored by applying the fixed effect model using panel data from 1988 to 2015.
Table 6.7: Hours of Outages and Cost of Load-shedding by Province Hours of Outages cost of load shedding % of value added Punjab 2666 11.9 Sindh 2200 14.9 KPK 2777 11.8 Balochistan 3733 11.8 Source: IPP (2012)
The present study has analyzed two models; model 1 and model 2, as model 1, will provide evidences for the impact of electricity consumption on manufacturing value added of the provinces with a controlled variable of per-capita growth of other sectors. The model 2 shows the impact of electricity consumption on manufacturing value added of the provinces with electricity load-shedding dummy from 2006 and with a controlled variable of per-capita growth of other sectors. The rationale of introducing dummy of electricity load-shedding from 2006 is that since the crises of electricity outages worsened from 200646.
…
= Per-capita growth rate of large scale manufacturing in province j at time t.
46 https://defence.pk/pdf/threads/countrywide-load-shedding-may-start-from-today.3335/ 186
= Electricity consumed by industrial sector i of province j at time t
= Per-capita growth of other sectors
= Load-shedding dummy variable since 2007-2015 = 1 for electricity load-shedding, for other years = 0
The regressand variable is the per-capita growth rate of manufacturing sector and the regressors are the per-capita industrial consumption of the electricity, per-capita growth of other sectors and dummy for electricity load shedding since 2007. The model one is without LS and model two is with load shedding dummy variable (LS). The results are reported below (table 6.8).
Table 6.8: Fixed Effect Model Results with and Without Load-shedding Dummy Model 1 Model 2 Independent variable Coefficient p-value Coefficient p-value Constant -0.0059 0.728 0.0142 0.489 Electricity consumed by industrial sector 0.5* 0.012 0.4612* 0.02 Per capita Growth of other sectors 2.24* 0.0000 2.257* 0.0000 Loadshedding Dummy -0.053** 0.082 Source: calculated by the author, see Annexure 2D*Significant at less than 5% **Significant at less than 10%
In the above models, the per-capita growth of manufacturing sector was regressed upon electricity consumed by industrial sector and per-capita growth of other sectors of the economy estimated as the GDP minus value added by the LSM. The model 1is without dummy for electricity load-shedding and model 2 is with load-shedding dummy variable.
The positive and significant value of per-capita growth of other sectors indicates that the growth of the other sectors of the economy complements the growth of LSM47. Ajmair, M. (2014) also finds the same result.
The second variable used is the electricity consumption. The impact captured by electricity consumption of industrial sector is also strong and significant, thus, highlighting the fact that for
47Ajmair, M. (2014). Impact of Industrial Sector on GDP (Pakistan Case). European Journal of Contemporary Economics and Management May. 187
higher growth of manufacturing sector, the availability of electricity is essential. Finally, the positive coefficient value of electricity consumption by industrial sector indicates that increase in electricity consumption by industrial sector of provinces will increase the manufacturing growthof the provinces and vice versa.Shahbaz, M., & Lean, H. H. (2012) also show that electricity consumption has a positive impact on GDP growth in case of Pakistan.48
To capture the impact of electricity shortage, a dummy variable from 2007 onwards, for load- shedding is introduced in model 2.
The introduction of electricity load-shedding as dummy variable has provided important information. The results show that electricity is an important and significant input for manufacturing. In model 2, the parameter value of electricity consumption by industry and per- capita growth of other sectors is significant at less than 5 percent level of significance. The electricity load-shedding dummy is also significant and its sign is also correct as per explanation and significant at less than 10 percent level of significance. The dummy variable indicates that load-shedding is negative and significant showing that load-shedding has negatively affected the growth of industrial sector; as one percent increase in load-shedding will decrease the provincial per-capita manufacturing growth by 0.05 percent. It is the very reason that shortage of electricity has significantly decreased the manufacturing sector’s growth. Hence, the positive sign of electricity consumption in industrial output in model 2 also reveals the same fact that if electricity consumption in industrial output is increased by provinces, it will eventually increase the manufacturing growth of provinces. Moreover, the negative sign of electricity dummy variable indicates that the electricity crisis has negative and has significant impact on the growth of manufacturing units of the provinces.
48Shahbaz, M., & Lean, H. H. (2012). The dynamics of electricity consumption and economic growth: A revisit study of their causality in Pakistan. Energy, 39(1), 146-153. 188
6.5.2 Fixed Effect Co-efficient and Their Impacts The fixed effect coefficients of the above two models are reported in below table 6.9.
The fixed effect coefficients of with and without dummy show that electricity load-shedding has damaged the manufacturing growth of the provinces. The manufacturing growth differentials between provinces have also increased due to electricity shortfalls. The fixed effect coefficients
(model 1) show that electricity load-shedding has affected negatively to the growth of manufacturing sector, all provinces except Balochistan. It may be due to the fact that manufacturing base is very small in Balochistan and mostly industries are small scale or medium scale that does not require much of the electricity consumption. However, the manufacturing units of large provinces i.e., Punjab and Sindh are affected by electricity load-shedding; as fixed effect coefficients of both large provinces are same i.e., -0.01624. The manufacturing growth of
KPK is also affected by electricity load-shedding but less than the large provinces; as the fixed effect coefficient of KPK is -0.00558. Interestingly, the fixed effect coefficient of manufacturing growth of Balochistan is 0.038067. The positive sign does not indicate that electricity load- shedding has affected the provincial manufacturing growth. Moreover, mostly the manufacturing units are small in Balochistan and cottage unit requires more labor than electricity.
6.5.3 Impact of Electricity Shortage The manufacturing growth of Punjab and Sindh are equally negatively affected by electricity shortage (Table 6.9). Also, the manufacturing growth differential of Punjab and KPK is -0.0107 percent. Showing that manufacturing growth of both the provinces are affected by electricity shortage but the impact of electricity crisis upon manufacturing growth of Punjab is higher than
KPK provinces.
189
Table 6.9: Fixed Effect Coefficients of Provincial Manufacturing Growth andProvincial Industrial Electricity Consumption Model 1: Model 2 : (With electricity load shedding (Without electricity Dummy) load shedding) Balochistan (B) 0.038067 0.038785 KPK -0.00558 -0.0058 Punjab (P) -0.01624 -0.01638 Sindh (S) -0.01624 -0.01661 Differences Punjab – Sindh 0.0000 0.0002 KPK -0.0107 -0.0106 Punjab– Balochistan -0.0543 -0.0552 Sindh – KPK -0.0107 -0.0108 Sindh – Balochistan -0.0543 -0.0554 KPK – Balochistan -0.0436 -0.0446 Source: Estimated by the author, Annexure 2D
The manufacturing growth differential, due to electricity shortage between Punjab and
Balochistan is much higher i.e., -0.0543 percent. It shows that manufacturing growth of Punjab has more affected of electricity shortage as compared to Balochistan. The higher manufacturing growth differential between two provinces is also due to the fact discussed earlier that the manufacturing units of Balochistan have not been affected much due to electricity load-shedding.
The manufacturing growth differential, due to electricity load-shedding, between Sindh and KPK is -0.0107 percent. It shows that, as compared to KPK, the manufacturing growth of Sindh is more affected by electricity load-shedding. The obvious fact is that in Sindh, the concentration of capital intensive large manufacturing units is more than others. The large or medium scale firms are less established in KPK, as compared to Sindh. It is one of the important reason that manufacturing growth of Sindh has affected more than the manufacturing growth of KPK.So, the manufacturing growth differentials between Sindh and KPK are higher due to electricity load- shedding.
190
In the case of manufacturing growth differentials between Sindh and Balochistan, it is similar to that of Punjab and Balochistan. The growth differential in manufacturing growth of Sindh and
Balochistan, due to electricity load-shedding, is -0.0543 percent. It shows the same fact that electricity load shedding has affected the growth of Sindh significantly. Due to low industrial base of manufacturing sector in Balochistan, and due to cottage industries of small scale industries in Balochistan shows big differential. In other words, it shows increased regional impact of inequality in this regard, among provinces.
The electricity load-shedding has also decreased the growth of manufacturing units in KPK. The manufacturing growth differential of KPK and Balochistan is -0.0436 percent, it is quite high showing that electricity load-shedding have more affected the manufacturing growth of
KPKmore than Balochistan which must have been added to inequalities among provinces.
Interestingly, the manufacturing growth difference between KPK and Balochistan is less than manufacturing growth difference between Punjab and Balochistan, and Sindh and Balochistan.
The main reason for this is the high concentration of manufacturing units in Sindh and Punjab as compared to KPK. The overall evidences indicate that electricity load-shedding has affected differentially to the provinces. As a result, it leads to income inequalities among provinces.
However, in this case, small provinces were not much affected from electricity of load-shedding.
Model 2: Electricity Load-shedding and Manufacturing Growth Differences between
Provinces
The fixed effect coefficients of model 2 do not show much of the variation as compared to model one results. Only a slight variation in fixed effect coefficients between the manufacturing growth rates of the provinces as indicated by model 2. The fixed effect coefficient of manufacturing growth of Punjab in model 2 is -0.01638 whereas for Sindh, the value of fixed effect coefficient
191
is -0.01661. Thus, showing that with the introduction of electricity dummy, the manufacturing growth of Sindh slightly decrease more as compared to a decrease in manufacturing growth of
Punjab. The fixed effect coefficients of manufacturing growth of KPK is also slightly decreased to -0.0058, in model 2, as compared to -0.00558 in model one. The fixed effect coefficient of manufacturing growth of Balochistan increases slightly to 0.038785 in model 2 as compared to
0.038067 in model one. The rationale for positive coefficient value of manufacturing growth of
Balochistan has been discussed earlier for model one.
6.6 Changes in the Manufacturing Share in National GDP The share of manufacturing sector towards national value added of manufacturing sector has changed over time. It may be said that the above discussed factors along with other contributed towards such changes.49
Table 6.10: Province-wise Share in National Manufacturing Sector Punjab Sindh KPK Balochistan Pakistan 1990-91 46.5 47.1 4 1.6 100 2000-01 44 45 10 1.5 100 2007-08 38 48 11 2.3 100 2010-11 44 42 13 2.1 100 2015-16 44 41 12 2.1 100 Source: SPDC 2005 and IPR(2015)
The above table 6.10 shows that the value added of manufacturing share of Punjab in national manufacturing in 1990-91 was 46 percent which was almost equal to Sindh. It decreased to 44% by 2015-16. However, Sindh having the same share in 1990-91, decreased its share to 41% i.e., a decrease of almost 6%. In Sindh, almost over 50% industries were alone in Karachi. But recently, due to terrorism, Karachi affected the most. Therefore, many industries were closed or
49 It may be interesting to explore further, which factors may have led to changes in the sources of manufacturing growth among provinces, as well as, for value added manufacturing differentials among provinces. 192
moved out of Karachi.50This also shows the fact that why the regional differences, in manufacturing sector growth rates, between Punjab and Sindh are slightly different. The value- addedmanufacturing in Sindh has been decreased over the period but still remained second higher as compared to other provinces. By 2015-16, the share of Punjab was 44 percent of GDP slightly higher than Sindh which was 41 percent.
In the case of KPK, the value-added share of manufacturing sector of KPK to national manufacturing sectors was 4 percent in 1990-91 and it increased to 12 percent 2015-16. Since
KPK has shown a growth in the manufacturing sector but still its contribution is far less than
Punjab and Sindh. In 2015-16, manufacturing share of KPK to overall manufacturing output is
12 percent. It may be noted that energy crisis was also far less in KPK, as compared to Punjab and Sindh. However, KPK was the most victim of terrorism.
The results reveal that the manufacturing sector of Balochistanis the least effected province and the reason is that the value added of manufacturing sector of Balochistan to national manufacturing growth of Balochistan is 1.6 percent in 1990-91 and it has only increased to 2.1 percent in 2015-16, thus, the contribution of manufacturing sector of Balochistan is very small to the national manufacturing sector. It highlights the fact that the manufacturing base of
Balochistan is very small and mostly manufacturing units are small, as compared to other provinces. So, the impacts of load-shedding on industrial output are not very significant.
However, KPK is the most victim of terrorism.
In brief, the regional growth differentials between Punjab and Sindh manufacturing growth are more significant due to load-shedding, as compared to KPK and Balochistan. Between two large provinces i.e., Punjab and Sindh, the differential in manufacturing growth is less but between
50 It may have been one of the major sector which contributed to decrease the manufacturing share in national GDP. 193
two large provinces and two small provinces i.e., KPK and Balochistan, the manufacturing growth differential is very high. This shows that between large and small provinces since the industrial base is high in large provinces. So, it is affected much more negatively due to electricity load shedding, as compared to small provinces.
6.7 Infrastructure Development and GDP Growth of Provinces Infrastructure plays an important role in the growth of an economy. The availability of social and physical infrastructure attracts investment and reduces cost too. Infrastructure development is a prime concern of the government, as in most economies, it is the government that make public development expenditures which generally includes the development of road networks, telecommunications, basic facilities that includes health, education, water and sanitation.Pakistan is suffering from low development of physical infrastructure in most of the provinces.Thus, inequality of physical infrastructure developmentcan affect the provincial growth rates and the national economy. As a result, inequality in basic infrastructural development connected to inequality among provinces. Naizi, J. and Imran, M. (2011) estimated the impact of infrastructure on economic development of Pakistan using data for 1975-2010. Their focus was first, the impact of infrastructure on total factor productivity and secondly, the impact of infrastructure in economic growth. The results showed that human capital and public capital stock had large and significant impact on the total factor productivity. In the second model, the availability of physical infrastructure such as roads, telecommunication, electricity generation and availability of water for agriculture showed significant impact on economic growth. The authors concluded that, as compared to other infrastructure in Pakistan, the development of road networks do not contribute significant benefits. Resources need to be more invested in power generation and communication, availability of water. Lastly, the authors concluded that presently
194
only 0.65 percent of GDP is spent on electricity generation thus in order to meet the demand, this share of investment needs to be increased to 1.5 percent of GDP. However, there is no study which may have been focused on the role of infrastructure in provinces. There is no need to focus on the same at regional level (provinces), so that the neglected area may be brought in to the main frame of the economy.
The authors Naizi, J. and Imran, M. (2011) have calculated the percentage growth of infrastructure in last four decades:(see appendix table 6.1)
It may be noted that rather than increase in electricity generation to meet the growing need of the economy; its generation decreased by (-26.4%) in 2010-11. This trend continued thereafter too.
Same is the case for per-capita availability of water supply for which the growth rate was -1.5%
(1970’s), which increased to 2.3% in 2000’s. There was a slight improvement of only 0.5% in
2010-11. The road length increased over the period of 1970’s to 2011. The number of telephones grew rapidly since 1970 but its growth fell by -5.4% in 2010-11. However, mobile phones increased around 5% in 2010-11. In addition to above, another important factor is human capital, which plays an important role in economic development of any economy. There is an ample body of literature which explains the importance of human capital and its role in overall economic development. Besides, technology also plays additional role in increasing productivity.
In Pakistan, the human resources have not been deployed much. Public policies ignored social sector development, the school enrollment ratios, and accesses to higher education are low in
Pakistan. The low development of human capital in Pakistan was disused in a cross-country study of low and middle-income countries by Khan, M. (2005). The study revealed that the average per-capita growth of Pakistan from 1980-2002 is broadly similar to South Asian countries but Malaysia, South Korea and other Asian countries have invested much more for
195
human development as compared to Pakistan. The study concluded that Pakistan has to invest more on institutional development and human capital to increase domestic and foreign private investment.
Recently, the low development of human and physical capital has been realized by the government and different projects have been initiated to improve the state of social and physical infrastructure. The role of human capital in economic development has been discussed by
Rehman, H. and Khan, M. (2012). It has been stated that human capital is a stock of competences, knowledge and traits embodied in workforce. The countries which have achieved high growth have invested significantly in education and health; as accessibility to higher education in South Korea is 98 %, and in Finland it is 94 %. However, in India, it is 15 % and in
Pakistan, it is 7.8%.51The authors reported that Pakistan spends only 1.7 percent of the GDP on education. Besides, hardly 0.22 percent of GDP is spent on higher education, which is meager allocation. What can be achieved from this meager investment in these sectors?
The overall and provincialdisparities in human capitalwere also discussed by Rehman, H. and
Khan, M. (2012). The authors estimated the human capital development in different provinces of
Pakistan and for overall Pakistan from 1979-2007. Further, the analyses were also extended to rural and urban disparities for human capital development in different provinces. The authors estimated the Human Capital Index (HCI) for provinces and overall Pakistan for the time period of 1979-2007. The results reported by authors showed that between Punjab and Sindh the human development trends were same as of the whole Pakistan from 1979-2007; the overall HCI was
0.41 in 1979 and it increased to 0.57 in 2007. The authors argued that the HCI of Balochistan remained low throughout the study period. However, the overall trend of HCI of the provinces
51 It is for the age group of 13 to 17 years of citizens. 196
was mixed throughout the time period from 1979-2007. The HCI for rural areas was 0.34 in 1979 and it increased to 0.53 in 2007. Since 1996, all rural sectors of the provinces showed upward trend, whereas the trend of rural Balochistan and Sindh was stagnant. Rural sector of Punjab and
KPK showed upward trend after 1996, so divergences among provinces increased, during that period; especially in rural sectors of the provinces. The HCI of urban regions of the provinces increased from 0.51 in 1979 to 0.68 in 2007. The human development of urban Sindh was high throughout the time period. Moreover, similar to the rural sector, the urban sector of the provinces showed upward trend till 1998 and since then, the divergences increased especially the urban KPK and Balochistan showed downward trend. So, the authors concluded that different provinces of Pakistan have different socio-cultural and political basis. It was stated that if these inequalities persist for longer time period, especially in health and education then it may cause serious problems for the economy, as a whole, and for the provincial economies too. So, in order to, analyze whether the government of Pakistan has successfully ensured the availability of social and physical infrastructure at regional level, fixed effect model of growthconsisting upon explaining variables, social and physical infrastructure has been estimated. It will allow analyzing the role of social and physical infrastructure in provincial development. Furthermore, the growth differentials of provinces will be analyzed; based upon empirical evidences pertaining to social and physical infrastructure.
6.7.1Role of Physical and Social Infrastructure in Economic Growth: The Modeled Empirical Evidences The above table 6.11 also shows the regional differentials between GDP growth of provinces; due to physical and social capital. The results of the fixed effect coefficients reveal that the physical and social infrastructure has increased the growth of Punjab more than Sindh. So the
197
growth differential between Punjab and Sindh is 0.0056 percent indicating the two important facts. Firstly, the sate of physical and social development is good and higher in Punjab as compared to Sindh. Secondly, the physical and social infrastructural development of Sindh is contributing less to the provincial economic growth of Sindh, in comparison to Punjab.
Particularly, the rural infrastructure in Sindh is in a dismal condition which may have shown up its impacts.
The growth differentials due to physical and social infrastructure between Punjab and KPK is
0.0028 percent. This shows that growth of Punjab is higher than KPK due to physical and social infrastructural development. Interestingly the growth difference of Punjab and KPK is less than
Punjab and Sindh, it shows that the physical and social infrastructure of KPK is contributing more to the growth of KPK as compared to Sindh.
The growth difference due to physical and social infrastructure development in Punjab and
Balochistan is 0.0245. This difference is quite high as compared between Punjab and Sindh,
Punjab and KPK. The very reason for this is the poor availability of social and physical infrastructure in Balochistan, which is not contributing towards the growth of the province. Even the fixed effect coefficient value of Balochistan is negative i.e -0.0162, showing the poor or no contribution of social and physical infrastructure developments to the province.
The growth difference between Sindh and KPK due to physical and social infrastructure, is -
0.0028. It shows that physical and social infrastructure has benefited more to the growth of KPK as compared to Sindh. The negative sign shows the high growth of KPK over Sindh due to availability of physical and social infrastructure. The growth differential of Sindh and
Balochistan due to physical and social infrastructure is 0.0189. It shows the high growth of Sindh
198
over Balochistan, as availability of physical and social infrastructure has benefited more to the growth of Sindh over Balochistan.
The growth difference due to physical and social infrastructure between KPK and Balochistan is
0.0217. This shows that growth of KPK has benefited more over growth of Balochistan due to the availability of social and physical infrastructure.
The above results show that Punjab and KPK have achieved high growth due to development of physical and human infrastructure. Whereas, Sindh and Balochistan are progressing slowly due to poor infrastructure.52
In the above analysis reveal that the physical and social capital has benefited more to the GDP growth of Punjab, as compared to other provinces, followed by KPK, Sindh and Balochistan.
Interestingly the contribution of social and physical capital to the GDP growth of KPK is more than Sindh. Moreover, the results of estimated model show that both physical and social capital contributes significantly to the growth of provincial economies. Therefore, the importance of the availability of physical and social infrastructure cannot be denied. However, it has been observed that the infrastructural development has contributed less to the growth of Sindh, whereas it has negatively contributed to Balochistan showing the poor availability and deterioration of physical and social infrastructure in the province.
The disparities in human development index of Punjab are discussed by Chaudhry. A and
Qasim.M., (2014) and found determinants of Human Development Index (HDI) and Non income human development index (NIHDI) of 35 districts of Punjab for the year 2010-11, to find out the disparities between districts of Punjab. The authors found that social infrastructure,
52 It may be noted that literacy among rural areas is even less than 5%, which indicates poor literacy rate. 199
industrialization and remittances have positive impacts on HDI and NIHDI but population density has positive impacts on HDI. Further, the authors found that to reduce disparities the infrastructure should be increased, as it has positive effect on HDI and NIHDI.
Thus, it may be concluded that the availability of physical and social capital for growth process of the provincial economies is important, however, disparities in the availability of infrastructure between provinces will have consequences on their growth. It can further increase the growth differentials among provinces.
6.8 Regional Employment Elasticity: Provincial Estimates Introduction This part is focused to illustrate the employment elasticity of the provinces. The elasticity of employment greater than one means that employment is elastic and if the value is less than one, it reflects that employment is inelastic, further the values closer to one shows higher degree of elasticity and values closer to zero shows less degree of elasticity.
6.8.1 Estimation and Discussion of results The employment elasticity measures how the employment will change with change in output.
The employment elasticity of the provinces is calculated by dividing average growth rate of labour force employed with average GDP growth rate.
Table 6.11: Results of employment elasticity across provinces, 1990-2015 Avg. GDP Avg. growth Employment Unemploy Growth of rate of Labor Elasticity ment rate provinces (%)* Punjab 4.5 2.3 0.5 6.29 Sindh 4.2 2.7 0.6 4.66 KP 4.7 2.0 0.4 7.71 Balochistan 3.2 3.6 1.1 3.92 Source: Calculated by the author. Pakistan Economic Survey, Different Issues. * 2014-15. The above table 6.11 shows the employment elasticity of the provinces, it can be seen that employment elasticity of Balochistan is 1.1 which means if GDP growth of Balochistan increases 200
by one percent it will increase employment growth of Balochistan by 1.1 percent. So employment in Balochistan is highly elastic as compared to other provinces. The reason maybe that the province has ample ideal resources to explore. The employment elasticity of Sindh is 0.6, it means if GDP growth of Sindh increases by one percent it will increase employment growth of
Sindh by 0.6 percent. The employment elasticity of Punjab is 0.5, it means if GDP growth of
Punjab increases by one percent it will increase employment growth of Punjab by 0.5 percent.
The employment elasticity of KPK is 0.4, it means if GDP growth of KPK increases by one percent it will increase employment growth of KPK by 0.4 percent. Provinces with higher employment elasticity shows that with increase in GDP growth labour absorption rate will increase thus more jobs will be created. The KPK with low employment elasticity shows that increase in GDP growth do not absorb the growing labour growth of KPK, thus, resultantly the unemployment rate will be higher. In Punjab again with low employment elasticity compared to
Sindh and Balochistan, the labour absorption will be less. In Sindh with low employment elasticity compared to Balochistan the labour absorption will be less and finally Balochistan with high employment elasticity compared to other provinces the, labour absorption will be higher.
It is also evident from the last column that unemployment rates in 2014-15, were lower in provinces, where employment elasticity is higher. In Balochistan where employment elasticity is higher than all other provinces, the unemployment rate is 3.92 percent, which is less than all other provinces. Furthermore Sindh has high employment elasticity than Punjab and KPK, therefore, unemployment rate in Sindh is 4.66 percent, less than Punjab and KPK. Between
Punjab and KPK, employment elasticity is higher in Punjab, thus unemployment rate of Punjab is 6.29 percent which is less than unemployment rate of KPK i.e 7.71 percent. So provinces with high employment elasticity have lower unemployment rates. So far there is hardly any study
201
which may have pointed out provincial employment elasticities. Based upon these elasticities, public policy may be chalked out by province and at national level to address regional issue of unemployment53.
The sectoral labour absorption of provinces is reported in Pakistan Economic Survey 2014-15, shows that the provinces with higher employment elasticities absorb most of the provincial labour force and has less unemployment in the province. The above result show that the employment elasticity of Balochistan is higher, as compared to other provinces i.e. 1.1, thus the unemployment rate is less in the provinces. The major sectors that absorb most of the working population of Balochistan in comparison to other provinces are reported in Pakistan Economic
Survey 2014-15, as 43.43 percent of the labour force is engaged in agriculture sector, which is high than other provinces, except Punjab. The mining and quarrying sector of Balochistan absorbs 1.12 percent of its labour force, which is significantly high than other provinces. The whole sale and retail trade sector of Balochistan absorbs 16.64 percent of its labour, which is again high than other provinces. The public administration and defense absorbs 5.96 percent of provincial labour force, which again is significantly higher than other provinces. So the above analysis reveals that; firstly, the employment elasticity is higher in Balochistan as compared to other provinces. Secondly, the sectoral absorption capacity of Balochistan in major sectors of the provincial economy is quite high than other provinces. The employment elasticity of Sindh comes after Balochistan and is higher than Punjab and KPK i.e 0.6. The labour absorption of
Sindh in the provincial sectors of Sindh’s economy are reported by Pakistan Economic Survey
2014-15, as 39.28 percent of the labour force in Sindh is absorbed by agriculture sector, which is
53 .Different studies have indicated much higher level of unemployment in Pakistan. However, these studied were mainly concentrated at national level. For details see Zulifqar I and Chaudhary M. Aslam (2018), forthcoming, Pakistan Economic and Social Review. This study indicated that unemployment was double digits in Pakistan. 202
higher than KPK and less than Punjab. The manufacturing sector of Sindh absorbs 15.17 percent of its working labour force; it is higher than all other provinces except Punjab. The whole sale and retail sales absorbs 15.82 percent of the provincial working labour force, which is higher than labour absorption by Punjab and KPK. Moreover, the Public administration and defense absorbs 3.16 percent of the labour force, which again is higher than Punjab and KPK. So it is found that if employment elasticity of Sindh is higher than Punjab and KPK; firstly, its unemployment rate is less than both of the provinces. Secondly, the labour absorption of the province in major sectors of the provincial economy is also high in comparison to Punjab and
KPK. Thus, highlighting the fact that if elasticity is higher the unemployment will be less and labour absorption or employment generation is high.
The elasticity results of Punjab show that, the employment elasticity of Punjab is 0.5 which is higher than KPK but less than Balochistan and Sindh. The employment rate of Punjab is also higher than Sindh and Balochistan, but lower than KPK. The labour absorption of the province shows that most of its labour is engaged in agriculture as 44.70 percent and in manufacturing as
16.47 percent. The labour force absorbed by these sectors of the provincial economy is higher than other provinces. However, the other sectors show relatively less labour absorption by the province in comparison to other sectors. The mining and quarrying absorbs 0.07 percent of the labour force which is less than all other provinces. Moreover, the transport and storages absorbs
4.13 percent of the labour force, Public administration and defense absorbs 1.80 percent of labour force and health and social work absorbs 1.21 percent of the labour force; all these sectors of the Punjab’s economy absorbs less of the labour force in comparison to other provinces.
203
So the above analysis show that since the employment elasticity of Punjab is less than Sindh and
Balochistan, thus at the sometime its employment generation capacity other than agriculture and manufacturing sectors is far less in other sectors as compared to Sindh and Balochistan.
Lastly, KPK has the lowest employment elasticity of 0.4, so the unemployment rate is higher as
7.71 percent in the provinces. The provinces absorb most of its labour force in agriculture sector which is 34.56 percent and it is quite less than other provinces. The manufacturing absorbs 11.26 percent of provincial working labour force, whole sale and retail sector absorbs 15.51 of the labour force, construction absorbs 12.46 percent of the labour force and transport and storage absorbs 8.19 percent of the labour force of the provincial economy. However, the labour absorption capacity of these sectors of the provincial economy is quite high, but still less than the labour absorption capacities of these sectors in other provinces. So the above analysis reveals that; firstly, the employment elasticity of KPK is less than other provinces. Secondly the employment generation capacity of the KPK in major sectors is also less as compared to other provinces. The labour absorption in major sectors of all provinces is shown in table below.
Table 6.12: provincial employment structure of labour force. (Percentage)
Major industry division Pakistan Punjab Sindh KPK Balochistan Total 100 100 100 100 100 Agriculture and allied 42.27 44.70 39.28 34.56 43.43 Mining and Quarrying 0.16 0.07 0.11 0.28 1.12 Manufacturing 15.33 16.47 15.17 11.26 11.34 Construction 7.31 6.59 7.00 12.46 6.53 Whole sale and retail trade 14.64 13.85 15.82 15.51 16.64 Transport and storage 5.00 4.13 5.74 8.19 5.14 Public administration and 2.44 1.80 3.16 2.69 5.96 defense Education 3.85 3.48 3.77 6.13 3.83 Human health and social 1.28 1.21 1.15 2.03 1.24 work Source: Pakistan Economic Survey 2014-15
204
The above table 6.12 shows the percentage of total labour force employed by major sectors of the economy. It can be seen that, in overall economy, almost 42.27 percent of the labour force is employed in agriculture sector, whereas manufacturing absorbs 15.33 percent of labour and whole sale and retail sale absorbs 14.64 percent of total labour force. The other sectors that absorb most of the labour force are construction which absorbs 7.31 percent of labour force mining and quarrying absorbs 0.16 percent, transport storage absorbs 5.00 percent, public administration 2.44 percent, education 3.85 percent and human health 1.28 percent.
So agriculture, manufacturing and whole sale & retail trade are the major labour absorbing sectors of the national economy. However, in provincial economies most of the labour force is absorbed by agriculture sector followed by manufacturing, whole sale and retail trade and construction. The transport and storage, Public administration and defense, education and health absorb small portion of the overall labour force of the national and as well as provincial economies.
In brief, the above analysis indicates that in two major provinces Punjab and Sindh, employment elasticity of Sindh is higher than Punjab. Between two small provinces KPK and Balochistan, employment elasticity of Balochistan is higher than KPK. Finally, among all provinces employment elasticity of Balochistan is higher than all other provinces followed by Sindh,
Punjab and KPK. Thus, it may be concluded that the low unemployment rate of Balochistan is due to its high employment elasticity. Most of the labour force in Balochistan is engaged in agriculture which is 43.43 higher than all provinces. the labour absorption capacity of other sectors such as whole sale and retail sector, manufacturing, construction and mining and
Quarrying is high, indicating that if these sectors are focused by government could generate high level of employment in the provincial and national economy.
205
The employment elasticity of Punjab as compared to Sindh is less, thus the unemployment rate of Punjab is higher than Sindh. Sindh mainly, absorbs most of the labour force in mining and
Quarrying, construction, whole sale and retail, transport storage, public administration and education as compared to Punjab. Finally, KPK has the lowest employment elasticity from all provinces and high unemployment rate mainly it is because of the less absorption in of labour in agriculture and manufacturing sectors as compared to other provinces. Although in some sectors like mining and Quarrying, construction, whole sale and retail, transport storage and education absorbs most of the labour force of KPK but since the less absorption of labour force in two major sectors agriculture and manufacturing cannot offset the loss to less jobs for labour force in
KPK. As well as employment generation is concerned, Pakistan is still an agriculture based country. Agriculture, being the largest sector can still absorb significant number of labor force.
Balochistan and Sindh still have millions of virgin land which could be brought under cultivation and surplus labor may be absorbed in agriculture. Secondly, manufacturing and education sectors are to be developed in all provinces, as these sectors can absorb more labour force, if needed investment is made in these sectors. Thirdly, the mining and quarrying in KPK and Balochistan is needed to be developed which can generate more employment opportunities in these provinces.
6.9 Conclusions This chapter analyzed the determinants and disparities of growth between provinces. The chapter is divided into two sections; as section one analyzed the growth differentials between provinces due to fiscal and monetary policies. Section two discussed the cross cutting edges; it includes the growth disparities between provinces due to terrorism, electricity load shedding, physical and social infrastructure, and employment elasticities of the provinces.
206
The empirical evidences indicated that, between provinces, the growth differentials increased due to monetary policies introduced by the federal government. The fiscal policy can have decreased such disparities by focusing on small provinces which are deprived of mega project, undertaken by the federal government. In other words, there is a need to increase development expenditures for provision of public goods like infrastructure development, as well as, improvements in agriculture and industry sectors. It is a responsibility of the Planning commission of Pakistan to keep an eye on regional disparities; it is the very reason that members of the Planning commission were to be appointed on provincial bases. These memberships have been taken over by the bureaucracy, as a result, the very basic role of the members has been lost, which led to regional disparities. Thus, these institutions need to be revives to their original spirit. The increase in growth differentials are largely found between large provinces and small provinces i.e. KPK and Balochistan. The deprived provinces need to be integrated in to the main stream of development. Whereas, the provincial expenditure policies reveal that the growth disparities between Punjab and Sindh increases due to high provincial current expenditures of
Punjab, as compared to Sindh. In other cases, between provinces the provincial expenditure policies reduce growth differentials among provinces.
The results of section two indicated that between provinces the growth disparities have been increased due to terrorism. The main victim of terrorism are small provinces i.e. Balochistan followed by KPK. The growth differentials due to terrorism have increased significantly between large and small provinces. Moreover, the growth differentials due to electricity load shedding have also increased such differentials between provinces. As load shedding has severally damaged, the provincial economies of Punjab and Sindh almost equally followed by KPK. So between the large provinces growth differentials are negligibly present, due to electricity load
207
shedding. The case with Balochistan showed interesting results, as the impact of electricity on manufacturing growth of Balochistan is not much; being a small province, having very small industrial base.
The growth differentials due to physical and social infrastructure between provinces have also increased drastically, especially in the case of Balochistan. The impact of availability of physical and social infrastructure on growth of Balochistan is negative, thus, showing that the state of availability of physical and social infrastructure is poor that it has affected the provincial economy negatively. Moreover, the growth differentials between large provinces, Punjab and
Sindh are also high, due to the fact that the physical and social infrastructure has affected growth of Punjab more than Sindh. Besides, the availability of infrastructure in KPK has benefited more to the provincial growth, as compared to Sindh. Hence it is Punjab followed by KPK and Sindh that has positively affected by availability of physical and social infrastructure. So, eventually between provinces the growth differentials are high due to the availability of physical and social infrastructure.
Finally, the results of employment elasticity show that in Balochistan the employment elasticity is quite high as 1.1and the provinces have also lowest unemployment rate too. It is followed by
Sindh, Punjab and KPK. The lower the elasticity, the higher will be unemployment rate, as in case of KPK, the unemployment rate is highest, as 7.71 percent, as compared to other provinces.
It also highlights an interesting fact that the employment elasticity is much higher in Balochistan even more than large provinces, Punjab and Sindh. So, if untapped natural resources like virgin land and minerals of the Balochistan are explored, it will not only contribute highly to the employment generation of the provincial economy but it will also contribute to bring the deprived province at par with national economic growth too. All above indicate that there are
208
regional income and growth inequalities issues, which need to be addresses. Particularly, the small provinces need to integrate with the national economy. It will help to remove the sense of deprivation by the provinces, as well as, strengthen the nation too.
209
Appendix
Table 1: Decade wise Growth Rate of Different Types of Infrastructure
(percentage) 20 199 00 201 1970s* 1980s 0s s 0-11 Per Capita GDP growth 2.7 2.9 1.8 2.8 0.3 - Per Capita Electricity Generation (Grwh)** 6.1 7.4 2.3 1.9 26.4 - Per Capita Water Availability (MAF) -1.5 -1 -1.2 2.3 0.5 Length of Roads (Kilometers) 4 6.3 3.9 0.5 -0.5 Telecommunications Number of Telephone Lines (per 1000 people) 7.8 11.1 11.3 0.2 -5.4 45.7 72. - Mobile Phones availability (per 1000 people) *** 3 4.9 Source: Naizi. J and Imran.M (2011). http://www.pide.org.pk/pdf/PDR/2011/Volume4/355-364.pdf . * From 1975-76 to 1979-1980 ** ?????
210
Chapter 7
Conclusions and Policy Implications This study is focused to explore the development differentials at provincial (regional) level in
Pakistan. There is hardly any such study which may have been focused to such avenue; the reason being non-availability of regional data. As a result, half of the province was not integrated with national economy which led to deprivation in the periphery provinces. For this purpose, econometric techniques were applied to draw empirical evidence. In this respect, the process and quantum of GDP (Provincial) and per capita income differentials and inequalities among provinces were identified. The economic development activities were mainly concentrated in
Punjab and Sindh; the largest provinces of Pakistan. The small provinces, KPK and Balochistan, did not benefit much from fiscal and development planning, although nine developments (Five year) plans were implementing in Pakistan. Moreover, regional sustainable economic development was not visible across provinces; such outcome has led to regional thinking among provinces that larger provinces are growing on the cost of other provinces, which is creating tension among provinces. It is the very reason that National Financial Award was revised and the federal allocation trends was raised for deprived provinces, so that these can be integrated in to national economy.
In withstanding the above, to understand the provincial economy the sectorial growth analysis of the provinces across sectors and across provinces is also carried out. It helped to identify the dynamics of the sectorial growth patterns among provinces. The study has also estimated the provincial growth differentials due to fiscal and monetary policy, as most of the provincial
211
governments in most of the time argued that federal fiscal policies usually benefits growth of larger provinces; rather than small provinces. Moreover, the provincial growth differentials accrued due to terrorism, electricity load shedding and development of physical and social infrastructure is also analyzed.
The results of GDP growth of the provinces show important findings. The large provinces,
Punjab and Sindh have developed integration with each other and adjust the short term disequilibrium at higher speed. However, the small provinces, KPK and Balochistan are still dependent highly on the GDP growth of large provinces. If any short term shock occurs to these small provincial economies it is adjusted by large provinces in short span of time, as compared to small provinces. It shows that mostly the policies of GDP growth are biased towards the development of large provinces, ignoring the important facts as highlighted above that even small provinces, if taken care properly by government could have improved their economic activities, as a result they can be integrated in the main framework of the economic development.
Since GDP growth is the composition of agriculture growth, manufacturing growth and services growth. So to explore the growth process, especially by the small provinces, sectorial linkages of these provinces with each other and with large provinces were analyzed. Thus the present study has analyzed the sectoral dynamics of provincial development and their relationships with each other too.
The analysis of agriculture sector growth; within sectors, across sectors and across provinces, reveals that the agriculture sector growth of all provinces is not integrated with each other’s. The major findings of the above analysis of agriculture sector indicate that within sectors, the growth of agriculture has integrated between large provinces Punjab and Sindh. Moreover, the adjustment to short term disequilibrium in agriculture sector of any of the large province is rapid
212
which help to overcome shocks to correct by agriculture growth of the provinces. In case of small provinces KPK and Balochistan, the integration within agriculture sector is present from large provinces Punjab and Sindh to small provinces KPK and Balochistan. In case of KPK the short term adjustment in disequilibrium in agriculture output of KPK is high dependent upon
Punjab and Sindh, and similar is the case with agriculture growth of Balochistan. The agricultural policies which were mainly focused on the agricultural development of large provinces (Indus-Basin) benefited these provinces the most. In case of small provinces within agricultural development the small provinces generally depend upon the agriculture growth of large provinces. No wonder, after seventy years of development efforts, the Balochistan is still waiting to explore its over two million virgin land and it hardly contribute tow percent towards national earnings.
The across sector and across provinces analysis indicated that, in case of large provinces, the agriculture sector is dependent on growth of manufacturing sector. The most likely reason may be the agro based industry of these two large provinces that take high inputs from the agriculture output; from these two large provinces. Interestingly, the services growth of almost all the provinces have significant role towards the development of agriculture sector across sectors and across provinces. It reveals that the services sector has created backward linkages with agriculture sectors of the provinces. Moreover, the backward linkages of manufacturing sector are also present in large provinces. The backward linkages of manufacturing sector are also present from large provinces to the agriculture development of small provinces. Since agriculture and manufacturer sector of the small provinces are not much developed, relatively, therefore people move across provinces to find jobs which helped to integrate services sectors.
213
The study has highlighted the strength of relationship of manufacturing sector growth; across sectors and across provinces too. The manufacturing growth of small provinces KPK and
Balochistan are significantly affected by agriculture growth of large provinces. The high dependency of small provinces in their manufacturing growth reveals that firstly, the agriculture output in these small provinces is poor and do not meet the demands of provincial manufacturing sector. Secondly, it may be the case that most of the industrial inputs in the form of raw material are not produced locally so the manufacturing growth of these two small provinces take agriculture inputs from large provinces Punjab and Sindh, which again shows their dependency.
The point to be made is that these provinces do have potential, but due to the concentration of federal development plans towards Indus-Basin these small remained deprived.54 The manufacturing sector growth of Sindh significantly affects the manufacturing growth of Punjab,
KPK and Balochistan. Further, the manufacturing growth of Punjab significantly affects the manufacturing growth of Sindh, KPK and Balochistan. The manufacturing sector growth of
Balochistan affects significantly the manufacturing growth of Punjab, Sindh and KPK. This effect may come from the manufacturing units of oil, gas and minerals extraction established in
Balochistan though at a small level, which may act, as providing the key inputs to the manufacturing units of Punjab and Sindh. Lastly the manufacturing sector growth of KPK significantly affects the manufacturing growth of Sindh only.
The services sector growth of Punjab significantly affects the manufacturing sector growth of
Punjab and Sindh. The services sector of Sindh affects significantly the manufacturing sector of
Punjab, Sindh and KPK. The services sector growth of KPK and Balochistan significantly the manufacturing sector growth of KPK.
54For more details see Chaudhary M.A. (1989), Chaudary M.A. and Aslam A.2018,UOL,Eeconomics Department. Lahore. 214
In brief, the manufacturing sector has developed forward linkages with services growth of almost all provinces. Secondly, the agriculture sector has failed to develop forward linkages with services growth of provinces accept Balochistan. Thirdly, within services sector the linkages are there across provinces. Fourthly, the services growth of large provinces complements each other in services sector growth. Fifthly, the large provinces Punjab and Sindh have developed forward linkages from manufacturing growth to services growth. Sixthly, among small provinces, it is only the manufacturing growth of KPK that has developed forward linkages with services growth of other provinces. However, the speed of adjustment from manufacturing growth of
KPK to disequilibrium in services growth of other provinces is very week. The services growth of Balochistan is still largely dependent on the agriculture, manufacturing and services growth of other provinces.
In nutshell, the large provinces, Punjab and Sindh having high concentration of services and industry are enjoying the benefits of services sector development and have developed linkages of this sector with other sectors of the economies. Whereas the two small provinces, KPK and
Balochistan are still far behind and are largely dependent on the services growth of large provinces; for their sectoral growth.
The Role of Fiscal and Monetary policies, Infrastructure, terrorism and Energy Shortage
The study has analyzed six models using panel data from 1990 to 2015 i.e. 25 years’ data has been used with fixed effect modeling technique, where cross sections are the GDP growth of four provinces along with different determinants of growth. One of the important objectives of these models is to understand that either the growth differential between provinces is fiscal phenomenon or monetary phenomenon. The change in the coefficients of fixed effect model of these estimate indicate that the significance and importance of the particular variable when
215
included in the model. The study also analyzed the impact of social and physical infrastructure, terrorism and electricity load shedding on the provincial growth differentials. The major findings are abridged below.
The study explored an impact of consumption expenditures, in the model are, and then private investment expenditures is included in model 2. To examine the change in the structure of the economies, provincial current expenditures is included in model 3. In model 4 provincial development expenditures is included. In model 5 federal total expenditures are included and finally in model 6, money supply is included. The inclusion of all these variables revealed the fact that how structural parameters of the particular economy changes with these variables. The results indicated that, with provincial expenditures the growth disparities increase between large provinces Punjab and Sindh. This increase is basically due to the high current expenditures of
Punjab, over Sindh. However, the growth disparities between large provinces Punjab & Sindh and small provinces between KPK and Balochistan have decreased with the inclusion of provincial expenditures (model 4). The growth differentials (model 5) show that the provincial growth disparities decrease between all provinces due to fiscal expenditures. So it indicates that fiscal expenditures help in reducing growth differentials between provinces; irrespective of the size of the province. The monetary variable inclusion (model 6) revealed different results. The results of the model 6, shows that the monetary policies have reduced the growth differentials between large provinces i.e. Punjab and Sindh. The important reason for this is the large development of financial sector in these two provinces, which benefits both provinces. However, the growth differentials between large and small provinces have increased significantly, due to the inclusion of monetary variables.
216
The analysis highlights two important findings; firstly, if the overall national growth is attained through the growth of large provinces, then monetary policies will work better as compared to fiscal. Still differential increased, overtime, among large and small provinces. Secondly, if growth is to be achieved in a sustainable and balanced manner fiscal policies will have to be integrated with monetary policies.
The provincial analysis of terrorism indicated that it has affected all provinces but its strong and significant impacts are on the GDP growth of Balochistan; as fixed effect coefficient of
Balochistan is -2.18, Balochistan shares its long border with Afghanistan and most of terrorist attacks were from Afghanistan, due to which it is easy to attack Balochistan. Similarly, KPK comes after Balochistan, as fixed effect coefficient for KPK is -0.07 which means after
Balochistan terrorism has badly affected the GDP growth of KPK. It may also be due to the same fact, as indicated above, that KPK has a border with Afghanistan too. Thus, the level of insurgences in KPK is also high. Balochistan and KPK are small economies, as compared to
Punjab and Sindh and the intensity of terrorism shocks are greater in these provinces. The provinces got affected the most in this cold war in Afghanistan. The growth differentials of large provinces Punjab and Sindh is 0.61 percent, thus between two large provinces, the growth disparities are less, as compared to small provinces. Moreover, between two small provinces
KPK and Balochistan the growth differential is 2.11 percent, which is quite high.
In short, terrorism has affected significantly to the growth process of Balochistan followed by
KPK, Sindh and Punjab.it may have contributed to increase economic differentials among provinces, particularly, between large and small provinces.
As well as, the impact of energy crisis is concerned; it has also severely damaged the economy.
Almost, all sectors have suffered due to shortage of electricity with its major impacts on
217
industrial sector. Sindh and Punjab are relatively having more industrial activities, as compared to small provinces. The fixed effect coefficient for Sindh is -0. 0166.After Sindh mainly the impact of electricity load shedding is on the manufacturing growth of Punjab; and its fixed effect coefficient is -0.0164. The KPK is still in developing stage and its industrial base. Thus the impact of electricity load shedding, on manufacturing growth of KPK, is less than Sindh and
Punjab as the fixed effect coefficient of KPK is -0.0058. Finally, Balochistan has very poor industrial base and its coefficient is only 0.038 which indicates a small impact of electricity load shedding on its industrial activities.
The GDP growth disparities, due to social and physical infrastructure, between two large provinces of Punjab and Sindh not very significantly. However, at the development of Physical and social infrastructure is better in Punjab, as compared to Sindh. Interestingly, the GDP growth of KPK has shown a higher significance which indicates that social and physical infrastructure is better in this province. The fixed effect coefficient of KPK is lower than of Punjab but higher than all other provinces.
The GDP growth disparities between large provinces and KPK are relatively less than
Balochistan. The GDP growth disparities between Sindh and KPK indicate that such disparities are not significant. The GDP growth, due to availability of social and physical infrastructure, in
KPK is better as compared to Sindh. The GDP growth disparity between Sindh and Balochistan indicated that Balochistan has less benefited.in other words Balochistan is far behind in the development of infrastructure which has affected its growth.it is due to the larger territory of
Balochistan.
The GDP growth disparities between two small provinces KPK and Balochistan are almost the same as the GDP disparity of Punjab and Balochistan. This is due to the fact that the GDP
218
growth of KPK is more than the GDP growth of Balochistan, due to better social and physical infrastructure there.
Policy Recommendations
In order to reap the benefits from being the agro based economy it is important that a sustainable growth of agriculture should be ensured in large provinces i.e. Punjab and Sindh by both federal and provincial governments. The agriculture sector of large provinces need to be integrated with manufacturing output across provinces. This will help in creating the forward linkages of agriculture sector with manufacturing growth across provinces. For this purpose, there is a need that small provinces should be integrated through fiscal measure and investment in the small provinces, particularly through development expenditure.
The public policies need to be integrated and linked agriculture growth of Punjab and Sindh with the manufacturing of other provinces; especially with small provinces. This can only be done having a sustainable growth in the agriculture sector. The virgin land in Balochistan needs to be converted to cultivation by special development plan to integrate the deprived regions. For
Balochistan, investment in mineral development will significantly help to up lift the province and integrate it with national economy.
The relationship of agriculture growth with services growth of all provinces needs to be further strengthened by improving services sector for financial institution, agriculture and manufacturing activities. It can be done by chalking out industrial and agricultural policy by equitable investment in all provinces.
The manufacturing sector of Punjab and Sindh are interlinked. However, there is to strength it further to develop efficient forward linkages with service sector. Investment through fiscal policy
219
for establishment of industrial zone in small province will help to improve these sectors which will reduce regional inequalities.
Most of the manufacturing units in Punjab and Sindh are agro based such industrial development is insignificant in small provinces. Therefore, special development progress is needed to develop infant industries in these provinces which can’t be done without public sector’s support; since infrastructure is also very poor in Balochistan.
Pakistan is a developing country where instituions are not very much developed. Therefore, special attention need to be paid to develop, rather centralized policies are continued. There is a need to decentralize decision making and devolution of power, which could help to integrate neglected area. The provinces need not be dependent on center for investment.
The provincial economies of the two small provinces need to have special development program since their infrastructure and economy have significantly affected by terrorism activities. Besides these provinces were already less developed, as compared to large provinces, the emergence of new losses ha further created income inequalities among regions. Regional integrated program for industry, agriculture and infrastructure is needed to integrate themselves with the large provinces. It will help to curtail regional inequality.
In addition to above, the shortage of energy has seriously damaged all provincial economies. The generated has high cost. Thus there is a need to develop cheap sources of electricity generation, without such development, in industrial activities cannot be competitive. There are significant efforts made in this respect, through CPEC program, which has curtailed the electricity load shedding. The short term, medium term and long term framework of the electricity production policies should be developed and followed strictly.
220
Pakistan has good resources of coal, so policies to explore this resource are to be formed and implemented efficiently, to handle the electricity crisis, it will reduce such dependency upon oil, which is expensive and it will also save foreign exchange.
The availability of physical and social infrastructure is the first foundation to initiate development activities: such infrastructure is still very poor, particularly, in Balochistan. It is the very reason that the province has very poor economic conditions. The development budget outlays especially in developing physical and social infrastructure across all provinces by federal and provincial governments need to be increased, particularly, special attention need to be paid to develop Balochistan. The practice of the government to cut the development budget, while meeting other expenditures or reduce budget deficit need to be avoided and priority need to be assigned to developed deprived regions as well as, to reduce regional inequalities.
Lastly, in most of the cases physical and social infrastructure is present in rural and urban sectors of main three provinces. However, its quality is so poor that even half of the population do not have safe drinking water, the access is confined to limited affluent class of the society. Besides such priorities it leads to disparities between regions and created disparities among provinces.
221
References ADB (2005). Technical Assistance Islamic Republic of Pakistan: Balochistan Economic Report. Asian Development Bank, Project Number: 39003 Afridi, A. H. (2016). Human Capital and Economic Growth of Pakistan. Business & Economic Review, 8(1), 77-86. Afxentiou, P., & Serletis, A. (2000). Output growth and variability of export and import growth: international evidence from granger causality tests. The Developing Economies, 38(2), 141-163. Afzal, M., Malik, M. E., Begum, I., Sarwar, K., & Fatima, H. (2012). Relationship among education, poverty and economic growth in Pakistan: an econometric analysis. Journal of Elementary Education, 22(1), 23-45. Afzal, M., Rehman, H., & Rehman, J. (2008). Causal nexus between economic growth, export and external debt servicing: The case of Pakistan. Retrieved December, 14, 2009. Aghion, Philippe, and Howitt, Peter. (1992). A Model of Growth through Creative Destruction.Econometrica 60 (March): 323–351. Ahmad, Z., Iqbal, F., & Mehmood, S. (2016). Relationship between Sectors Shares and Economic Growth in Pakistan: A Time Series Modeling Approach. Journal of Statistics, 23(1). Ahmed, A., & Ahsan, H. (2011). Contribution of services sector in the economy of Pakistan. Pakistan Institute Of Development Economics Islamabad, 79. Aisha, Z., & Khatoon, S. (2009). Government expenditure and tax revenue, causality and cointegration: The experience of Pakistan (1972-2007). The Pakistan Development Review, 951- 959. Akbari, S. A. H., Riazuddin, R., & Choudhry, M. K. (1993). Growth of Manufacturing Employment in Pakistan: A Comparative Analysis of Punjab and Sindh (Preliminary Results)[with Comments]. The Pakistan Development Review, 32(4), 1267-1277. Akram, N., & Gulzar, A. (2013). Climate change and economic growth: An empirical analysis of Pakistan. Pakistan Journal of Applied Economics, 23(1), 31-54. Ali, S., Asghar, M., Kalroo, R. A., Anjum, S., & Ayaz, M. (2014). Manufacturing Sector Employment and Multidimensional Poverty in Pakistan: A Case Study of Punjab Province.
222
Ali, S., Farooq, F., & Chaudhry, I. (2012). Human Capital Formation And Economic Growth In Pakistan. Pakistan Journal of Social Sciences (PJSS), 32(1), 229-240. Aisha Ghaus P., Hafiz A. Pasha, and Asma Z., (2010), Fiscal Equalization Among Provinces in the NFC Awards, Pakistan Development Review, winter.
Amjad, R., Arif, G. M., & Mustafa, U. (2008). Does the labor market structure explain differences in poverty in rural Punjab? Aghion, Philippe, and Howitt, Peter. (1992). A Model of Growth through Creative Destruction. Econometrica 60 (March): 323–351.
Arby, M. F. (2008). Some Issues in the National Income Accounts of Pakistan (Rebasing, Quarterly and Provincial Accounts and Growth Accounting). PhD Thesis, Pakistan Institute of Development Economics, Islamabad, Pakistan. Arby, M. F. and Rasheed, M. A. (2010). Estimating Gross Provincial Accounts Of Sindh. Journal Pakistan Business Review, volume 12,pp 539-587. Armstrong, H.W., Taylor, J. (2000), Regional Economics and Policy (3rd Ed). Oxford: Blackwell Aslam, A., & Zulfiqar, K. (2016). Policy Framework for Inclusive Growth: A Case Study of Selected Asian Countries. Policy, 12, 21-40. Asteriou, D.,Agiomirgianakis, G.M. (2001). Human capital and economic growth: time series evidence from Greece, Journal of Policy Modeling, 2001, Vol. 23, No 5, p. 481-489 Azam, M. &Khattak (2009). Empirical Analysis of The Determinants Of Economic Growth In Pakistan, 1971-2005. Sarhad J. Agric. Vol.25, No.2, 2009 Badar, H., Ghafoor, A., & Adil, S. A. (2007). Factors affecting agricultural production of Punjab (Pakistan). Pak. J. Agri. Sci, 44(3). Baer, W. (1964). Regional Inequality and Economic Growth in Brazil, Economic Development and Cultural Change, 12(3), 268-285. Retrieved from http://www.jstor.org/stable/1152261 Balassa, B. (1978). Exports and economic growth: further evidence. Journal of development Economics, 5(2), 181-189. Baldwin, R. E., & Martin, P. (2004). Agglomeration and regional growth. Handbook of regional and urban economics, 4, 2671-2711.
223
Baloch, M. A., & Thapa, G. B. (2017). Review of the agricultural extension modes and services with the focus to Balochistan, Pakistan. Journal of the Saudi Society of Agricultural Sciences. doi:https://doi.org/10.1016/j.jssas.2017.05.001 Belton, F., Haizheng, L., & Min Qiang Zhao (2010), Human capital, economic growth, and regional inequality in China. Journal of Development Economics, Volume 92, Issue 2, July 2010, Pages 215-231. Bengali, K. and Sadaqat,M. (2005). Regional Accounts of Pakistan: Methodology and Estimates 1973-2000 (Working Paper No.5). Reterived from Social Policy and Development Centre website: http://www.spdc-pak.com/publications/Working%20Papers/WP-05.pdf Bengaliwala, K. (1995). Temporal and Regional Decomposition of National Accounts of Pakistan. PhD Thesis, University of Karachi, Pakistan. Capello, R. (2011). Location, regional growth and local development theories. Aestimum(58), 1. Cass, David. (1965). “Optimum Growth in an Aggregative Model of Capital Accumulation.” Review of Economic Studies 32 (July): 233–240. Chaudhary M. Aslam, (1989), Agricultural Development and Public Policies in Pakistan, Izhar Sons, Lahore.
Chaudhary M. Aslam, (1989), Modeling Industrial Growth and Agglomeration Economies in the Manufacturing Sector of Pakistan. The Pakistan Development Review, (1989-a).
Chaudhary M. Aslam, (1990), Economic Growth and Regional Disparity in Production Activities in Pakistan), Pakistan Economic & Social Review.
Chaudhary M. Aslam, (2019), Economic Management and Emerging Issues in Pakistan, forthcoming, HEC, GOP, Islamabad.
Chaudhary M. Aslam, And Saeed A. (1990), Economic Growth and Regional Disparity in Production Activities in Pakistan), Pakistan Economic & Social Review, " (1990). Chaudhary. M.A. and Hamid. A. (1999). Human Resource development and Management in Pakistan, (1999), published by Ferozsons (Pvt) Limited, Lahore, Pakistan. Clark, D. (2010). Regional Growth the Neo-Classical perspective: lecture notes. Dawkins, C. J. (2003). Regional development theory: conceptual foundations, classic works, and recent developments. CPL bibliography, 18(2), 131-172. Diamond, Peter A. 1965. “National Debt in a Neoclassical Growth Model.” American Economic Review 55 (December): 1126–1150.
224
Din, M. U., Ghani, E., & Mahmood, T. (2007). Technical efficiency of Pakistan's manufacturing sector: a stochastic frontier and data envelopment analysis. The Pakistan Development Review, 1-18. Easterly, W. (2001). The political economy of growth without development: A case study of Pakistan. Paper for the Analytical Narratives of Growth Project, Kennedy School of Government, Harvard University. Ellahi, H. M. (2013). Overtime Growth in Crop and Livestock Productivity in Pakistan’s Provincial Context. Global Journal of Science Frontier Research, 13(10). Escobal, J., & Torero, M. (2000). Does geography explain differences in economic growth in Perú: Bid (Banco Interamericano de Desarrollo). Farooq, M., & Khan, Z. Impact of Terrorism on Foreign Direct Investment and Key Indicators of Development in Pakistan. Gaibulloev, K., & Sandler, T. (2009). The impact of terrorism and conflicts on growth in Asia. Economics & Politics, 21(3), 359-383. Ghani, E., & Din, M. U. (2006). The impact of public investment on economic growth in Pakistan. The Pakistan Development Review, 87-98. Government of Balochsitan, 'Blochistan Development Statistics, Various Issues' Balochsitan Bureau ofStatistics, Quetta Government of KP, 'KP Development Statistics, Various Issues' KP Bureau ofStatistics, Peshawar. Government of Pakistan, 'Labour Force Survey, Various Issues', Bureau of Statistics, Government of Pakistan, 'Pakistan Economic Survey, Various Issues', Finance Division, Economic Advisor's Wing, Islamabad. Government of Pakistan,'Household Integrated Economic Survey, Various Reports', Pakistan,Bureau of Statistics, Islamabad. Government of Punjab, 'Punjab Development Statistics, Various Issues' Punjab Bureau of Statistics, Lahore Government of Sindh, 'Sindh Development Statistics, Various Issues' Sindh Bureau ofStatistics, Karachi. Granger, C. (1969). Investigating Causal Relations by Econometric Models and Cross-spectral Methods. Econometrica, 37(3), 424-438. doi:10.2307/1912791 225
Grossman, Gene M., and Helpman, Elhanan. 1991a. Innovation and Growth in the Global Economy. Cambridge, MA: MIT Press. Hahn, F. H., & Matthews, R. C. (1964). The theory of economic growth: a survey. The Economic Journal, 74(296), 779-902. Haider, A. Sectoral Analysis of Employment Demand (Jobless growth) in Pakistan. Hull, K. (2009). Understanding the relationship between economic growth, employment and poverty reduction. Promoting pro-poor growth: Employment, 69-94. Hull, K. (2009). Understanding the relationship between economic growth, employment and poverty reduction. Promoting pro-poor growth: Employment, 69-94. Hull, K. (2009). Understanding the relationship between economic growth, employment and poverty reduction. Promoting pro-poor growth: Employment, 69-94. Husain, I. (2004). Economy of Pakistan: Past, present and future. In Keynote address at the conference on Islamizaton and the Pakistani economy held at the Woodrow Wilson Center, Washington, DC. Husain, I. (2010). Pakistan's Growth Experience. IBA Business Review, 5(2). Hussain, A. (1993). Regional Economic Disparity in Pakistan and a framework for Regional Policy. Paper presented at the Wilton Park Conference at Wiston House, Sussex, England. Hussain, M. (2005), “Inflation and Growth: Estimation of Threshold Point for Pakistan”, Pakistan Business Review, Vol. 7, No. 3 Hussain, M. (2005), “Inflation and Growth: Estimation of Threshold Point for Pakistan”, Pakistan Business Review, Vol. 7, No. 3 Hussain, M. A. (2014). Economic Growth, Exports and Imports in Pakistan: Granger Causality Analysis. Economic Growth in Pakistan, 13. Hussain, S. S., & Ahmed, V. (2012). Experiments with industrial policy: The case of Pakistan. Sustainable Development Policy Institute. Hussain, S. T., Khan, U., Malik, K. Z., & Faheem, A. (2012). The constraints to industry in Punjab, Pakistan. The Lahore journal of economics, 17, 135. Hussain, S., & Malik, S. (2011). Inflation and economic growth: Evidence from Pakistan. International Journal of Economics and Finance, 3(5), 262. Hyder, S., Akram, N., & Padda, I. U. H. (2015). Impact of terrorism on economic development in Pakistan. Pakistan Business Review, 839. 226
Ikram, K. (2009). Economic Development-A view from the Provinces. CREB, Lahore School of Economics, Lahore. Impact of War in Afghanistan and Ensuing Terrorism on Pakistan’s Economy. (2014-15). Islamabad: Government of Pakistan. Imran, M., & Niazi, J. (2011). Infrastructure and growth. The Pakistan Development Review, 355-364. Iqbal, Z., & Malik, N. (1993). Institutional Variations in Saving Behaviour in Pakistan, The Pakistan Development Review, 32(4), 1293-1311. Retrieved from http://www.jstor.org/stable/41259735 Iqbal, Z., & Zahid, G. (1998). Macroeconomic Determinants of Economic Growth in Pakistan, The Pakistan Development Review, 37(2), 125-148. Retrieved from http://www.jstor.org/stable/41260096, Islamabad. Jenkins, H. (1995). Education and production in the United Kingdom, Oxford: Nuffield College, (Economics discussion paper, No 101) Johnston, J., & DiNardo, J. (1997). Econometric methods (4th ed.). New York: McGraw-Hill. MLA Citation. Kalim, R. (2001). Capital Intensity in The Large-Scale Manufacturing of Pakistan. Pakistan Economic and Social Review, Volume XXXIX(No. 2), 135-151. Kemal, A. R. (2006). Key Issues in Industrial Growth in Pakistan. Lahore Journal of Economics, 11. Khan, J. I. (2005). Intra-Model Employment Elasticities (A Case Study of Pakistan’s Small– Scale Manufacturing Sector). Khan, Jangraiz (2012) The Role of Human Capital In Economic Growth Of Pakistan (1971- 2008). PhD thesis, University of Peshawar, Peshawar. Khan, M. A., & Rehman, H. U. (2012). Regional Disparities In Human Capital: The Case of Pakistan. Pakistan Economic and Social Review, Volume 50(1), 57-69. Khan, M. S. Human Capital and Economic Growth in Pakistan. The Pakistan Development Review, 44 : 4(455–478). Khan, M. T. The Social, Political and Economic Effects of the War on Terror: Pakistan 2009 To 2011. ISSRA PAPERS, 65.
227
Khan, M., & Sasaki, K. (2003). Regional disparity in Pakistan’s economy: Regional econometric analysis of causes and remedies. Khan, M.S. and S.A. Senhadji (2001), “Threshold Effects in the Relationship between Inflation and Growth”, IMF Staff Papers, Vol. 48, No. 1 Koopmans, Tjalling C. (1965). On the Concept of Optimal Economic Growth. In The Economic Approach to Development Planning. Amsterdam: Elsevier. Li, M (2006). Inflation and Economic Growth: Threshold Effects and Transmission Mechanisms. University of Alberta, Working papers. Majid, N. (2000). Pakistan: Employment, output and productivity. Mankiw, N. G. (2014). Principles of macroeconomics: Cengage Learning. Michael, S. (2007). Terrorism a socio-economic and political phenomenon with special reference to Pakistan. Journal of management and social sciences, 3(1), 35-46. Mubarik, Y. A. (2005), “Inflation and Growth: An Estimate of the Threshold Level of Inflation in Pakistan”, State Bank of Pakistan Research Bulletin, Vol. 1, No. 1 Mujahid, N., Amin, A., & Khattak, S. W. (2014). Human Capital Investment and Physical Capital Nexus (A Path to Economic Growth of the Country): A Case Study of Pakistan 1980- 2010. Putaj Humanities & Social Sciences,21(2). Mundell, R. (1965), “Growth, Stability and Inflationary Finance,” Journal of Political Economy, Vol. 73 Nadeem, N., Mushtaq, K., & Dawson, P. J. (2013). Impact of Public Sector Investment on TFP in Agriculture in Punjab, Pakistan. Pakistan Journal of Social Sciences (PJSS), 33(1). Nadeem, N., Mushtaq, K., & Javed, M. I. (2011). Impact of Social and Physical Infrastructure on Agricultural Productivity in Punjab, Pakistan-A Production Function Approach. Pakistan Journal of life and social sciences, 9(2), 1-6. Nadia, T. (2017). The Size and Growth of the Economy of Lahore. Lahore Chamber of Commerce and Industry, Lahore. Naeem Ur Rehman Khattak, Iftikhar Ahmad, & Khan, J. (2010). Fiscal Decentralisation in Pakistan. The Pakistan Development Review, 49:4 (Part II). Pasha, H. A. (2015). Growth of The Provincial Economies. Institute of Policy Reforms, Lahore. Pahsa. H.A. (2018). Growth and Inequality in Pakistan. Friedrich-Ebert-Stiftung (FES), Pakistan.
228
Patrick, H. T. (1966). Financial development and economic growth in underdeveloped countries. Economic development and Cultural change,14(2), 174-189. PILDAT (2006). Dynamics of Federalism in Pakistan: Current Challenges and Future Directions. PILDAT, Islamabad. PILDAT (2015). Background Paper: Promotion of Agriculture in Pakistan. PILDAT, Islamabad. Qasim, M., & Chaudhary, A. R. Determinants of Human Development Disparities: A Cross District Analysis of Punjab, Pakistan. National College of Business Administration & Economics, Lahore. Qayyum, Abdul. "Money, inflation, and growth in Pakistan." The Pakistan Development Review (2006): 203-212. Qureshi, S. K., & Arif, G. M. (2001). Profile of poverty in Pakistan, 1998-99: Pakistan institute of development economics (PIDE). Rabbi, F. (2012). War against Terrorism and its Repercussions for Pakistan. Pakistan Journal of History and Culture, 33(2). Ramsey, F. P. (1928). A Mathematical Theory of Saving,Economic Journal 38 (December): 543–559. Romer, Paul M. (1990). “Endogenous Technological Change,” Journal of Political Economy Sabir, M., & Aftab, Z. (2006). Province-wise growth patterns in human capital accumulation. The Pakistan Development Review, 873-890. Safi, G. M., Gadiwala, M. S., Burke, F., Azam, M., & Baqa, M. F. (2014). Agricultural Productivity in Balochistan Province of Pakistan A Geographical Analysis. Journal of Basic & Applied Sciences, 10, 292. Sandilah, M. N., & Yasin, H. M. (2011). Economic growth and regional convergence: the case of Pakistan. The Pakistan Development Review, 333-353. Sarwar, K., Afzal, M., Shafiq, M., & Rehman, H. (2013). Institutions and economic growth in South Asia. Journal of Quality and Technology Management, 9(2), 1-23. Scarth, W. M. (1988). Macroeconomics: An introduction to advanced methods: Harcourt Brace Jovanovich Incorporated. Schaffer, W. (2014). Regional Models of income determination: simple economic-base theory. Regional Impact Models. Georgia Institute of Technology, accessed April, 18.
229
Shabbir, T., and A. Mahmood (1992) The Effects of Foreign Private Investment on Economic Growth in Pakistan, The Pakistan Development Review 31:4 831–841. Shahzad, F. (2015). Role of human capital on economic growth: A case study of Pakistan. International Journal of Accounting and Economics Studies, 3(1), 20-24. Siddiqui z and Zaheer R (2017). Regional Integration and Economic Growth: A Convergence Analysis for Pakistan. J Glob Econ 5: 255. doi:10.4172/2375-4389.1000255 Siddiqui, R., Samad, G., Nasir, M., & Jalil, H. H. (2012). The impact of climate change on major agricultural crops: evidence from Punjab, Pakistan. The Pakistan Development Review, 261-274. Sindh Strategy for Sustainable Development, IUCN, 2007. Sindh Programme Office. xxvii+265 pp Siraj, S., & Bengali, K. (2007). Estimation of the Harrod-Domar Growth Equation: Pakistan’s Case. Journal of Independent Studies and Research, 5(2), 46-51. Solow, Robert M. (1956). “A Contribution to the Theory of Economic Growth,” Quarterly Journal of Economics 70 (February): 65–94. State Bank of Pakistan, 'Handbook of Pakistan Economy, 2015. Swan, T. W. (1956). “Economic Growth and Capital Accumulation,” Economic Record 32 (November): 334–361 Tariq, M. (2003). Causes of Industrial Failure and Its Implications in NWFP. Capacity Building for Science and Technology, 42. Ullah, S., Khan, S. &Ullah, F. (2014). “Assessment of key determinants for economic growth in Pakistan”, Journal of Theoretical and Applied Economics, Volume XXI (2014), No. 9(598), pp. 103-114. Usman, M. (2016). Contribution of agriculture sector in the GDP growth rate of Pakistan. Journal of Global Economics. Wilson, R. A., & Briscoe, G. (2005). The impact of human capital on economic growth: a review. World Bank (2005). Pakistan North West Frontier Province Economic Report Accelerating Growth and Improving Public Service Delivery in the NWFP: The Way Forward. World Bank, Report No. 32764-PK Xu, Z. (2000). Financial development, investment, and economic growth. Economic Inquiry, 38(2), 331-344. 230
231
Comments by James Fackler One Possible critiques: perhaps I missed it, but I did not find discussion ……. That can inform this study. Answer is given on page 71 the study has used data from 1973 to 2015 for cointegration analysis. The data is available annually without any missing observations.
232
233
Annexures
( See Vol. II)
234