School of Economics Bahauddin Zakariya University, Multan, Pakistan

Return of Investment to Education and Health in Pakistan: An Empirical Analysis

Raima Nazar

Roll No: PHDE-12-08 Session: 2012-2016

Supervised by: Prof. Dr. Imran Sharif Chaudhry Director, School of Economics Bahauddin Zakariya University Multan, Pakistan

A dissertation submitted to the School of Economics, Bahauddin Zakariya University, Multan, in fulfillment of the requirements for the degree of Doctor of Philosophy (PhD) in Economics

School of Economics Bahauddin Zakariya University, Multan, Pakistan

Dedicated To

My Parents Who instruct me how to live diligently, and educate me to live with capability and

Always prayed for my success

and

My Husband Who was always there to help me whenever I need, whose love and support helped me to accomplish this task

Declaration

I hereby declare that this thesis has not been submitted, either in the same or different form, to this or any other university for any degree.

Raima Nazar July, 2019

iii

CERTIFICATE

We accept the work contained in this dissertation titled: “Returns of Investment to Education and Health in Pakistan: An Empirical Analysis” as conforming to the required standard for fulfillment of the degree of Doctorate of Philosophy in Economics.

Director / Supervisor

______Prof. Dr. Imran Shairf Chaudhry Director, School of Economics Bahauddin Zakariya University, Multan External Examiner

______

Acknowledgements Words are always a poor approximation of what we want to say. All praise for the Allah Almighty, the omnipresent, the Merciful, the most Gracious, who is the entire source of all knowledge and wisdom. All praises are for Holy Prophet Hazrat Muhammad (SAW) who bought the message of peace and happiness to all creatures. I am thankful to Allah who gave me courage to undertake and complete this task assigned to me.

The business of acknowledgements challenges me more than the writing of the dissertation itself because it is very difficult to acknowledge the contributions of our benefactors in just few words. There had been the contributions of so many of my benefactors and well wishers and it would be out of question for me to enlist the names of all of them here but whether their names appear here in this acknowledgment or not, I want to assure them that their gestures of goodwill have never been gone forgotten nor will they ever be (In’Sha’Allah). All the words of gratitude and veneration fall incapacitated to express my feelings for my honorable supervisor Professor Dr. Imran Sharif Chaudhry (Director School of Economics, Bahauddin Zakariya University, Multan). I am very fortunate to complete this dissertation under his great supervision and I am very thankful to him from the core of my heart for his affectionate help, constructive advice, personal interest and analytical supervision. I also express my heartfelt and sincere gratitude for his scholastic attitude, generous and skillful guidance during this research project and in the accomplishment of this manuscript. It was his very personality that he was always there in with the most prudent and heedful of advice and guidance whenever I needed and always give me proper time for discussion irrespective of his busy schedule throughout the day. He was the person who always bailed me out of all the intricacies of research with a smiling face and I owe him the most earnest and heartfelt gratitude. No doubt with the encouraging support of my supervisor it becomes possible for me to accomplish the present study in stipulated time. I owe to express my gratitude to Prof. Dr. Shah Nawaz Malik, Dr. M. Zahir Faridi, Dr. Ramzan Sheikh, Dr. Muhammad Omer Chaudhry and all my respected teachers from School of Economics for guiding and encouraging me every time. I am also indebted to Dr. Atif Akbar (Department of Statistics, BZU, Multan) for his guidance and support about problems that I faced during sampling. He always gave me proper time for discussion about all problems during data collection phase and whenever I needed. May Allah always bless them.

I gratefully acknowledge the efforts and support of my friends and senior fellows for their support and guidance especially Dr. Furrukh Bashir who was always there to help me in issues about data analysis and my PhD fellows and colleagues for their support. My sincere thanks to The Women University, Multan which Almighty Allah used to provide me with the Scholarship award under faculty development programme to study for the PhD at Bahauddin Zakariya University, Multan and to Mrs. Shehnaz gul, chairperson department of Economics at The women university Multan for her moral support.

I would like to mention a few names of my family members here who suffered a lot during my research work and I believe that my success heavily depends upon their prayers. I am thankful to my deceased father Malik Nazar Hussain Awan (May Allah Almighty takes his soul to JANNA’tul’FIRDOUS and bestows him final success for all the examinations in the life hereafter, AAMEEN). He left us alone in this world on 10th of February, 2014. He was most excited on my admission in PhD but he has not enough time to see this day of honor in my life, however, his prayers remained with me even after his departure. I really have no words to say thanks to my loving mother Faiz Fouzia, She is the reason behind my success and I am really nothing without her efforts, love and prayers, may she live long. I am also indebted to my sisters Shumaila Nazar and Amber Malik and my elder brother Mr. Shehbaz Khan for their support. Last but not the least; a special and marvelous note of thanks and heartiest gratitude to my husband Mr. Sajid Ali for all his love, support, care and guidance. Without him I am really unable to accomplish anything worthwhile. He is always a source of positive support and encouragement. I am, indeed, at loss of words to enumerate his real contribution during the whole course of my research work. I owe everything of it to him. And, in fact, it was owing to the whole of my family that today I may be able to complete the degree of Doctor of Philosophy. May Allah bless them.

Raima Nazar

July 2019

ABSTRACT

The concept of human capital recognizes that not all individuals are equal and that the quality of their performance can be improved by investing in them that makes them more productive for economic activities and economy as a whole. Therefore investment in two main components that is education and health are discussed in this study because these two are the main components to improve human capital.

Education and health are the two important components of human capital. Education is a powerful instrument in reducing poverty, enhancing earnings, economic growth, empowering people, and promoting a healthy and flexible environment and creating a competitive economy. It plays an important role in shaping the ways to become skilled and handle the complexities of economic growth by the future generations. On the other hand, health is a basic and key ingredient of human capital and an important determinant of economic growth. The main objective of this study was to examine the returns of investment to education and health in Pakistan and to evaluate their impact on the economic growth of Pakistan, using cross sectional data from district Multan and time series data from Pakistan for the period of 1972 to 2016.

To achieve this objective, the study was completed in different phases. In the first phase, a comprehensive literature review was carried out using standard sources and tools for the better understanding of theoretical and empirical aspects of the study at national and international levels. In the second phase, a sample of 850 wage earners was randomly collected from city district Multan for micro data analysis by using questionnaire. Mincer earning function (1974) was used for this purpose and returns were evaluated in the form of an increase in earnings for both education and health investment. In case of education returns to total years of education, different levels of education and quality of education are estimated and disaggregated for gender, region and marital status etc. Moreover, some determinants of individual’s earnings are also evaluated. In case of health, we discussed anthropometric measures; self-reported health status, nutritional factor and some other determinants of health and evaluated their impact on earnings of the individual.

In the second phase, secondary data was collected from the economic survey of Pakistan and World Development Indicators for the period of 1972 to 2016. A total of fourteen variables were

vii selected for the study, and three main econometric models were tested for returns of investment to education and health and one for their impact on economic growth. The collected data for these variables were analyzed using computer software E-Views version 9 for secondary data and Minitab version 16 for primary data.

Ordinary least square method was used for primary data analysis while during secondary data analysis different econometric techniques were applied to examine the stationarity of data and long run and the short run relationship between education, health and economic growth. For stationarity, long run and short run relationship Augmented Dickey Fuller (ADF) test, Johansen Cointegration, Error Correction Mechanism (ECM) and Granger causality were used respectively. The results of these tests confirmed that these models are best fitted. The findings of Augmented Dickey Fuller test show that all variable are unit root at level and stationary at first difference or co-integrated of order one. The results of Johansen cointegration and VECM confirmed that education, health and economic growth have a long run relationship whereas; Error Correction Mechanism also confirmed their short run relationship.

Theoretical background of investment in education and health and ways to find returns of investment are also discussed in detail. Then a large section is devoted to discussing policies regarding health and education, their targets and results. As district Multan is selected as sample area, therefore, a detailed profile of this district and situation of health and education services in that sector are also presented in detail in a separate section.

The results of ordinary least square show that investment in education and health has a positive and significant impact on the earnings of the individual. In case of investment in education, we came to know that higher education yields higher returns in both cases when education is treated as a continuous variable and discrete variable. Returns for investment are higher for female as compared to male similarly in urban areas returns are higher due to higher opportunities as compared to rural areas and unmarried workers got higher returns as compared to married workers due to family constraints. Moreover, we also explored that medium of instruction plays a very significant role in earnings of an individual.

In case of health investment, we explored that height as a measure of long run health investment and body mass index as a short run health investment plays a significant role in earnings of an

viii individual. As these are measures of strength and shows the childhood level of investment in health and influence of some environmental factors also. Returns of this investment are discussed along with gender differences. We also explored that investment in nutritious food, pure drinking water; clean environment, precautionary measures, exercise and availability of medical facilities all play a significant role in earning determination.

At the macro level, we explored that public investment in health is very productive and we used infant mortality rate as a proxy for health return and explored the impact of health expenditure as a percentage of GDP, availability of doctors, fertility rate and literacy rate on this outcome variable. Similarly, we explored the impact of government expenditure at the primary, secondary and tertiary level on education enrollment index and found a positive and significant relationship. This study also found that education; health and economic growth are co-integrated and have a long run relationship.

TABLE OF CONTENTS

TITLE PAGE Dedication ii Declaration iii Certificate iv Acknowledgement v Abstract vii Table of contents x List of figures xv List of tables xvi List of appendix xix List of abbreviations xx CHAPTER 1 INTRODUCTION 1.1 Background of the study 1 1.2 Statement of the problem 5 1.3 Significance of the study 9 1.4 Research questions 11 1.5 Objectives of the study 11 1.6 Limitations of the study 11 1.7 Organization of the study 12 CHAPTER 2 PROFILE OF HEALTH AND EDUCATION SECTORS IN PAKISTAN 2.1 Introduction 14 2.2 Health policies during the study period 14 2.2.1 Alma Ata declaration 1978 14 2.2.2 Health policy 1990 15 2.2.3 Health policy 1997 18 2.2.4 National health policy 2001 20 2.2.5 Health policy 2009 21 2.2.6 National health vision 2016-2025 22 2.3 Five Year plans over the study period 24 2.3.1 4th five year plan (1970-1975) 24 2.3.2 5th five year plan (1978-83) 25 2.3.3 6th five year plan (1983-88) 27 2.3.4 7th five year plan (1988-93) 29 2.3.5 8th five year plan (1993-98) 31 2.3.6 Reforms in health sector (2003) 32 2.4 History of education sector in Pakistan 34

2.5 Education policies in Pakistan during the study period 36 2.5.1 New education policy 1970 36 2.5.2 The education policy 1972-1980 38 2.5.3 National education policy 1979 39 2.5.4 National education policy 1992-2002 41 2.5.5 Education policy 1998 42 2.5.6 Education policy 2009 44 2.6 Five year plan over the study period 46 2.6.1 4th five year plan (1970-1975) 46 2.6.2 5th five year plan (1978-83) 48 2.6.3 6th five year plan (1983-88) 50 2.6.4 7th five year plan (1988-93) 50 2.6.5 8th five year plan (1993-98) 52 2.6.6 Education sector reforms (2001-2005) 53 2.6.7 National education policy (1998-2010) 56 2.6.8 Education for all (EFA) strategies (2001-2010) 57 2.6.9 The national action plan and education related MDGs 57 2.7 Snapshot of district Multan 58 2.8 Education sector of district Multan 62 2.8.1 Enrollment ratios in primary education 62 2.8.2 Net primary enrollment ratios 63 2.8.3 Literacy rate 64 2.8.4 Education budget of District Multan 65 2.9 Health sector of district Multan 67 2.9.1 Health institutions in district Multan 67 2.9.2 Number of beds in health institutions of district Multan 73 2.9.3 Provision of rural health centers in long term plan (2013-2028) 74 2.10 Conclusion 75 CHAPTER 3 LITERATURE REVIEW 3.1 Introduction 77 3.2 Human capital and history 77 3.3 Returns of investment to education 81 3.4 Returns of investment to Health 95 3.5 Relationship between education, health and economic growth 105 3.6 Conclusion 109 CHAPTER 4 THEORATICAL AND CONCEPTUAL FRAMEWORK 4.1 Introduction 110 4.2 Concept of capital 111 4.2.1 Four categories of capital 112

4.3 Concept of human capital 112 4.3.1 Uses of human capital 114 4.3.2 Sources of human capital differences 115 4.4 Comparison between physical and human capital 116 4.5 Investment in education and health 119 4.6 Producing human capital: education and training 120 4.7 Investment in education 120 4.7.1 Defining rates of return to education 121 4.7.2 The Chicago school approach 123 4.7.3 The Mincer equation 123 4.7.4 The elaborate method 124 4.7.5 Two methods of estimating returns of investment to education 124 based on the Mincer equation 4.8 Investment in health 127 4.8.1 Health, human capital and income 127 4.8.2 Measures of health human capital 128 4.9 Role of human capital in theories of growth of developing countries 131 4.10 Education, health and economic growth 132 4.11 Conclusion 134 CHAPTER 5 DATA AND METHODOLOGY 5.1 Introduction 135 5.2 Description of variables 135 5.2.1 Variables for primary data analysis 135 5.2.2 Variables for secondary data analysis 138 5.3 Model Specification 141 5.3.1 Model specification for primary data analysis 141 5.3.2 Model Specification for secondary data analysis 143 5.4 Study area and sampling technique 144 5.4.1 Questionnaire design 145 5.4.2 Pre testing 145 5.4.3 Sample size determination 145 5.5 Secondary data sources 146 5.6 Methodology 146 5.6.1 Augmented Dickey fuller test 146 5.6.2 Ordinary least square 148 5.6.3 Johansen Cointegration 149 5.6.4 Vector error correction 149 5.6.5 Granger causality 150 5.7 Conclusion 151

xii

CHAPTER 6 RESULTS AND DISCUSSIONS: AN ELEMENTARY DATA ANALYSIS 6.1 Introduction 152 6.2 Personal information of respondent 152 6.3 Qualification, experience, training and skills of respondents 156 6.4 Respondent’s household information 158 6.5 Respondent’s health information 160 6.6 Conclusion 162 CHAPTER 7 PRIMARY DATA ESTIMATION OF RETURNS OF INVESTMENT TO EDUCATION AND HEALTH: EMPIRICAL EVIDENCE FROM DISTRICT MULTAN 7.1 Introduction 163 7.2 Returns of investment to education in district Multan 163 7.2.1 Mincer earning function with working hours, family background 171 and matriculation subjects 7.2.2 Extended Mincer earning function 175 7.2.3 Mincer earning function with quality of education 177 7.3 Returns of investment to health in district Multan 179 7.4 Conclusion 184 CHAPTER 8 RETURNS OF INVESTMENT TO EDUCATION AND HEALTH IN PAKISTAN: SECONDARY DATA ANALYSIS 8.1 Introduction 186 8.2 Descriptive analysis 186 8.3 Unit root test 188 8.4 Returns of public investment to health in Pakistan 189 8.4.1 Lag length selection 189 8.4.2 Unrestricted cointegration rank test 190 8.4.3 Johansen cointegration (long run estimates) 191 8.4.4 Error correction model (Short run estimates) 193 8.4.5 Granger Causality 194 8.5 Returns of public investment to education in Pakistan 195 8.5.1 Lag length selection 195 8.5.2 Unrestricted cointegration rank test 195 8.5.3 Johansen cointegration (long run estimates) 196 8.5.4 Error correction model (Short run estimates) 198 8.5.5 Granger Causality 199 8.6 Health, education and economic growth 200

xiii

8.6.1 Lag length selection 201 8.6.2 Unrestricted cointegration rank test 201 8.6.3 Johansen cointegration (long run estimates) 202 8.6.4 Error correction model (Short run estimates) 204 8.6.5 Granger Causality 205 8.7 Conclusion 206

CHAPTER 9 SUMMARY, CONCLUSION AND RECOMMENDATIONS 9.1 Introduction 209 9.2 Summary of the study 209 9.3 Findings of the study 211 9.4 Conclusion 213 9.5 Policy recommendations 214 REFERENCES 217 APPENDICES 239

xiv

LIST OF FIGURES

FIGURE TITLE PAGE 1.1 Trend analysis of GDP growth rate and health expenditures as 5 percentage of GDP 1.2 Trend analysis of GDP growth rate and education expenditures as 6 percentage of GDP 2.1 Allocation for annual budget for city district Multan 2009/10 66 4.1 Basic tradeoff in human capital theory 121

LIST OF TABLES

TABLE TITLE PAGE 1.1 Social sector indicators of some selected countries 7 2.1 Allocation of funds for the 5th plan 27 2.2 4th plan break up for new education policy 1970 38 2.3 Total expenditures under the 4th plan (1970-1978) 47 2.4 Number of urban and rural UCs in each town 59 2.5 Ethno-linguistic division 60 2.6 Town wise population (thousand persons) 61 2.7 Age structure in district Multan 61 2.8 Net primary enrollment ratio in district Multan 63 2.9 Net primary ratios (Punjab vs. Multan) 63 2.10 Literacy rate in district Multan 64 2.11 Literacy rate: district Multan vs. Punjab 64 2.12 Number of public sector education institutions 65 2.13 Share of education in the annual development plan of city district Multan, 66 2008/09 2.14 District education budget for schools FY 2009-10 to 2011-12 (Rs.Million) 67 2.15 Public health institutions in district Multan (MDA area) 68 2.16 Public health institutions in district Multan (excluding MDA area) 69 2.17 Private health institutions in Multan city district area 70 2.18 Number of beds in health institutions of district Multan 73 2.19 Number of beds in urban area of Multan: 2011 vs. 1987 74 2.20 Provision of rural health centers in the long term plan 74 2.21 Rural health centers (RHCs) required for additional population (2013- 75 2028) 5.1 Description of primary variables 135 6.1 Religion, region, gender and marital status distribution 152 6.2 Town wise sample distribution 153 6.3 Profession wise sample distribution 153 6.4 Designation wise sample distribution 154 6.5 Share of employment wise sample distribution 154 6.6 Nature of job wise sample distribution 155 6.7 Working hours wise sample distribution 155 6.8 Terminal degree and faculty wise sample distribution 156 6.9 Terminal degree institution, country and system of examination wise 157 sample distribution 6.10 High school region, subject and medium of instruction wise sample 157 distribution 6.11 Diploma and training distribution 158 6.12 Linguistic sample distribution 158 6.13 Parent’s qualification wise sample distribution 159 6.14 Residence wise sample distribution 159 6.15 Assets and happiness wise sample distribution 159

xvi

6.16 Health status, major illness and physical disability wise sample 160 distribution 6.17 Nature of major illness and disability wise sample distribution 161 6.18 Medical facilities, Source of medication and precautionary measures wise 161 sample distribution 6.19 Healthy diet, hygienic water and clean environment wise sample 161 distribution 7.1 Mincer earning function (full sample) 164 7.2 Mincer earning function of gender 165 7.2.1 Rate of return of investment to education by gender 166 7.3 Mincer earning function of region 167 7.3.1 Rate of return of investment to education by region 167 7.4 Mincer earning function of marital status 168 7.4.1 Rate of return of investment to education by marital status 168 7.5 Rate of return to education by sector of employment 169 7.6 Rate of return to education by profession 169 7.7 Rate of return to education by faculty 170 7.8 Mincer earning function with working hours, family background and 171 matriculation subjects (full sample) 7.9 Mincer earning function with working hours, family background and 173 matriculation subjects (Gender wise sub sample) 7.10 Mincer earning function with working hours, family background and 174 matriculation subjects (region wise sub sample) 7.11 Mincer earning function with working hours, family background and 175 matriculation subjects (marital status wise sub sample) 7.12 Extended Mincer earning function 176 7.13 Mincer earning function with medium of instruction (full sample) 177 7.14 Mincer earning function (medium of instruction wise sub sample) 178 7.15 Mincer earning function with anthropometric measures of health (full 180 sample) 7.16 Mincer earning function with anthropometric measures of health (gender 181 wise sub sample) 7.17 Mincer earning function for health investment (full sample) 182 7.18 Mincer earning function with self reported health indicators (full sample) 183 7.19 Mincer earning function for some determinants of health (full sample) 184 8.1 Descriptive statistics 187 8.2 ADF test results 188 8.3 Lag order selection criterions 189 8.4 Results of trace statistics 190 8.5 Results of maximum Eigenvalue 190 8.6 Normalized cointegration coefficients 191 8.7 Error correction model 193 8.8 Results of granger causality 194 8.9 Lag order selection criterions 195 8.10 Results of trace statistics 196

xvii

8.11 Results of maximum Eigenvalue 196 8.12 Normalized cointegration coefficients 197 8.13 Error correction model 199 8.14 Results of granger causality 200 8.15 Lag order selection criterions 201 8.16 Results of trace statistics 201 8.17 Results of maximum Eigenvalue 202 8.18 Normalized cointegration coefficients 203 8.19 Error correction model 205 8.20 Results of granger causality 206

xviii

LIST OF APPENDIX

APPENDIX DESCRIPTION PAGE 1 Questionnaire 239 2 MDGs target for Pakistan 242 3 Target for the vision 2010 of health policy 1997 242 4 Detail of the allocation for sub sectors of health under 4th plan 243 5 Physical achievements of health sector during 4th five year plan 243 6 Physical targets of the fifth five year plan 244 Infrastructure and manpower development during 6th plan and 244 7 expenditure on health during 6th five year plan 8 Year wise expenditure on health sector during 7th five year plan 245 9 Physical targets of 8th five year plan 245 10 Expenditure estimates for the 8th five year plan 1993-98 246 11 Formal education institutions in Pakistan 247 12 Enrollment in formal schools in Pakistan 247 13 4th five year plan (public sector allocation for education) 248 Requirements for the recurring expenditures during 4th plan (1970- 248 14 75) 15 Financial requirements estimates for education during 5th plan 249 16 Allocation for education and manpower in 6th five year plan 250 Expenditure for public sector development allocation for education in 250 17 7th five year plan (1988-93) Expenditure allocation for education and training during 8th plan 251 18 (Rs.Million) 19 Targets for the education sector reforms 251 20 Financial requirements for ESR during 2001-05 251 Minitab results of returns of investment to education and health: 252 21 primary data evidence from district Multan E-views results of returns of investment to education and health in 261 22 Pakistan E-views results of diagnostic tests of returns of investment to 272 23 education and health in Pakistan by using secondary data

xix

LIST OF ABBREVIATION

HDI Human development Index GDP Gross domestic product GNP Gross national product MDGs Millennium development goals HIV Human immunodeficiency virus AIDS Acquired immunodeficiency syndrome WHO World health organization SAP Social action program BCG Bacilli Calmette Guerin UC Union council RHC Rural health centers RHU Rural health unit BHU Basic health unit NFBE Non formal basic education BECS Basic education community schools Bed Bachelors of education academics and science PTC Primary teaching certificate EFA Education for all MDA Multan development authority OLS Ordinary least square ADF Augmented dickey fuller ECM Error correction model VECM Vector error correction model

Chapter 1 INTRODUCTION

1.1 Background of the Study

Human capital has a significant role in the sustainable economic development of an economy. Development of human capital is necessary for the evolution of political and socio-economic conditions of an economy. Economists recognize the development of human capital as a remarkable devotion to human capital development and as a basic factor responsible for the remarkable performance of most of the underdeveloped and developed countries. Such development has been achieved by the people of those economies mainly because of talent, know-how and ability attained through good health and education. Economists emphasized that differences in socio-economic development between the underdeveloped and developed economies is not due to the physical capital, natural resources and gifts but due to the quality and quantity of human capital (Imad Ullah, 2014).

According to the classical point of view, in the economic development for the 1960s, labor, land and capital were considered as major determinants of production and the main emphasize was on increasing the amount of capital by enhancing the amount of investment at the level of 15 percent of GDP to achieve at least 5 percent growth rate. In the start of 1970s, the definition of capital was expanded by the addition of human capital. Investment in health and education with better technology will lead to higher production level and speedy economic growth. By the early 1980s, the economic reason for educational investment was thus well established (Aziz, 2005).

UNDP launched Human Development Report in 1990 prepared by Dr. Mahbub ul Haq, and it was a landmark in the theoretical framework for the development. The report defined a criterion for determining the performance of a country called Human Development Index (HDI). This index was prepared on the basis of four indicators: adult literacy rate, life expectancy at birth, gross enrollment ratio and GDP per capita. The index has been assigned a maximum value of one and a minimum value of zero, and on this base, countries are classified into Low (below 0.5), Medium (between 0.5 and 0.8) and High HDI (above 0.8). The main view of this concept was that human development was not just a name of economic growth, but it means security of

1 people in their homes, in their workplace, in their societies and in their environment. According to Dr. Mahbub ul Haq: A National Agenda: Critical Choices for Pakistan's Future (1993):

“Investment on people in Pakistan not only decreases below any decent idea of national governance; we are simply not developing the country for challenges related to the technology of the 21st century. Where do we begin in such a wilderness of human neglect? We cannot change this situation over-night. It requires a reasonable amount of investment in human capital for a long period.”

In endogenous growth theories, the positive externalities are expected from investment in human capital which will generate an increasing interest for policy makers and researchers in evaluating the role of human capital formation (especially education and health) in economic growth. Importance of attaining better outcomes of investment in human capital relies on providing basic social services like education and health for protecting better living standard to the majority of the people (Behrman and Schneider, 1993). According to the evidence of cross country analysis, the income growth and the availability of basic social services (education, health, water and sanitation) can help in reducing poverty and inequality more than increasing industrial production (Bourguignon and Morrison, 1998).

Developing of human capital by means of better opportunities of education and improving the status of health will increase the potential of productivity not only for individuals but also for the community through positive externalities. Here, the example of East Asian economies is helpful because these countries have been able to decrease the inconsistencies in the formation of human capital and hence achieved pro-poor economic growth (Deininger and Squire, 1998 and Klasen, 2002).

Health and education are two primary determinants of human capital which make a person more productive and creative. Education is a vital tool in reducing poverty, increasing incomes of people, promoting the adaptable and healthy environment, empowering the individuals and creating a competitive economy. It plays a significant role to become more skilled and treat with the complications of economic growth for future generations. Educational institutes prepare people to be able to take part in all aspects of life including economic matters (Afzal, 2011).

Education, on one side, enhances growth rate and, on the hand side, it decreases poverty. It makes a clear political and social condition that pulls investment. Educated and skilled laborers are more efficient, more respectful, and anxious to follow the business rules leading to socioeconomic growth. Education has a core and positive part in the development of a country. The prosperity of a nation relies on the condition of the economy, and the position of the economy relies on the productivity of work, which thus relies on health of labor and education (Wobst and Seebens, 2005).

Public expenditure in education is suggested to be money well spent. Investment in human capital in the form of schooling has a positive effect on economic growth at both, the national and individual levels (Duflo, 2001; OECD, 2004; Sherman, 1994). There is no question that investing in human capital as a formal education is beneficial; the pressing question is where in education investments should be made. This is an extremely important issue, particularly when policies on investments in the different levels of education are developed and implemented (Besley & Burgess, 2003; McMahon, 2000). This dissertation attempts to suggest the efficient allocation of resources to primary, secondary and tertiary education considering the specific characteristics of individuals in Multan district.

Like education health is also among the fundamental capacities that offer flexibility to human life (Sen, 1998 and 1999). Health is asserted to be an essential human right, a vital part of life and is broadly perceived that enhanced health not just brings down morbidity, mortality and fertility rate, but also increase productivity (Bloom and Canning, 2001). Better health can be viewed as a noteworthy determinant for welfare and accordingly influences neediness straightforwardly. The solid relationship between economic indicators and health proposes that health is additionally a factor of economic development of the countries. As of late effect of human capital formation (particularly status of health) is acknowledged to be an important indicator of economic development in individual economies as well as across economies and over time (Bhargava, 2001; Webber, 2002; Alderman, et al. 2003 and Muysken, 2003). In this manner, health and its presumable effect on person's prosperity and economic development got enormous significance at different levels (Frank and Mustard, 1994).

Good health improves physical or mental capabilities or productivity of action. Two-way association exists between health and economic growth. It is true that Health and some other

3 types of physical capital enhance the level of GDP per capita by enhancing the labor productivity and other resources. Some amount of this increased earning is then reinvested in human capital (Akram, 2008). During 1790 to 1980, one-third of the GDP of UK was the result of improvements in health, medical facilities and food (Fogel, 1994). World Development Report (2007) says that due to the advanced improvements in the health sector, the life expectancy at birth increased worldwide from 51 years to 65 years in less than 40 years.

Returns to education and health are considered a reward for investment in these sectors. This reward can be in the form of income or other social factors like status, honor, accommodating behavior, etc. in the case of education and better health, low mortality rates and high expectancy at birth and to some extent income are the returns of investment in health. The educational returns can be calculated in various ways like monetary, non-monetary, social or private returns.

Private educational returns are demonstrated in higher salaries or wages that are given to the employee or worker. If private educational returns are high, then more students will be encouraged to enroll in the disciplines which have high market demand. This may include richer cultural interactions and greater political awareness in the society, improvements in the welfare of individual that is not the part of measured earnings (i.e. better working environment, easy access to highly paid job and so on). Social or collective returns of education include skill development and higher levels of human capitals leading to increased economic growth (Nasir and Nazli, 2000).

According to above scenario, it is right time that the role of health and education for the economic returns of society and individual is explored and analyzed. Once health and education are treated as an investment, the instant natural question is: what are the benefits of this investment to compare it with its alternatives? Such comparison can provide help to allocate public funds to various levels of health and education services, or can explain the behavior of individual regarding demand or lack of demand for these kinds of services. An evaluation of returns to health and education can also help to estimate broad health facilities and education policies. Human capital development is crucial to economic development that is why Governments should seek to adopt such policies that are compatible with the development of human capital.

1.2 Statement of the Problem

Pakistan is facing financial crises since the partition in 1947. Due to limited resources, Pakistan has very little to spend on the programs of social services like health and education. Moreover, in 1947, Pakistan inherited an inadequate health care system comprising of only one medical college, a few doctors, with the heavy burden of infectious diseases, Tuberculosis (TB), high infant and maternal mortality rate and population having low life expectancy. Even now there are very few facilities available for the people of Pakistan especially in the rural regions due to less expenditure on education and health during last seventy years. During 1970-2010, health expenditures as the percentage of GDP remain 0.5 to 0.8 percent. During the year 2015-16, health expenditures were only 2.5 percent of GDP which were very less if compared with other countries of Asia. In Pakistan, not only the share of health spending is very low but access to basic health facilities is also very difficult. Rural health services in Pakistan are always being ignored and importance is given to medical colleges, hospitals and creative healthcare.

Figure: 1.1 Trend analysis of GDP growth rate and health expenditure as percentage of GDP

0 1975 1980 1985 1990 1995 2000 2005 2010 2015 HXGDP GDPG

Data Source: World Development indicators Published by World Bank, 2016

These trend lines show the changes in the expenditures of health as a percentage of GDP and GDP growth rate. Expenditures on health as a percentage of GDP is far below than other countries which is 2.5 percent only, and it is very alarming situation. As compared to annual GDP growth rate Pakistan is contributing very little to its health sector and there is no significant growth in health expenditures over a long period. Low public expenditures on health as percentage of GDP, lack of governance, growth oriented policies (industrial bias) and lack of

5 resource utilization during planned period and mismanagement are among the top factors that are responsible in deterioration the infrastructure and health facilities in Pakistan.

There are two main issues for the indicators of public health; first is inequality in the delivery of health care services and second is the higher average health inequality. If we compare the inequality in different regions regarding access to basic health services by gender, at the regional level (urban and rural level) and income groups constitute a further warning to the existing situation of public health.

The education sector is also facing the same problems. In 1970, the share of education in public spending was 1.1 percent. During 1980, it was 0.8 percent. In 1990, it was 2.3 percent. In fiscal year 2004-05, it was 2.2 percent of GDP. It was 2.3 percent in 2009-10, while it is only 2.6 percent in 2016 which is very low compared to other economies in the region.

In Pakistan, most of the spending on education goes to recurring expenditures. Higher education always remains the priority of government while primary and secondary education is ignored. Due to less interest of government on primary education, net primary enrollment is only 73.9 percent, and literacy rate is only 54.9 percent in 2016.

Figure: 1.2 Trend analysis of GDP growth rate and education expenditure as percentage of GDP

0 1975 1980 1985 1990 1995 2000 2005 2010 2015 GDPG EXGDP

Data Source: World Development indicators Published by World Bank, 2016

As the trend lines show that there are no sudden fluctuations in education expenditure. Currently, Pakistan is spending only 2.6 percent of GDP on Education sector which is very low as compared to other developed and developing economies. Apart from considerable fluctuations in GDP growth rate, expenditures on education are quite stagnant. Despite a considerable increase

6 in enrollment and educational institutions in the decades of 1980s and 1990s, labor market of Pakistan is still inadequate in skilled and educated manpower. This may be the consequence of low investment in education and mismatch between attained education and suitability of demand for the graduates in the job market. It leads to many economic and social problems including low earning profiles and unemployment both at macro and micro levels.

The allocation of resources during last fifteen years squeezed in the social sector. The disparity in social sector is developed and the growth over time did not deliver the outcomes for the improvement of the general people, and hence inequality and poverty increased. Without the development in the social sector, high rate of growth cannot be useful in decreasing the gaps among social class and also increasing the living standards of the people. As described in Pakistan National Human Development Report (2002), improving the health care conditions, cleanliness and prevention trickle down the recurrence of disease, decreases the treatment cost and henceforth help in raising the efficiency as well as the wage of the poor and thus prevent the miserable human conditions.

Health and education indicators of Pakistan, represent a depressing picture when compared to other Asian countries. Development of social sector, especially education and health, remains a core policy agenda during last many years but still, these indicators are far from satisfactory. Comparing social sector indicators of Pakistan with few developing and developed countries gives an unfavorable picture of human development in the country.

Table: 1.1 Social Sector indicators of some selected countries

Population Health Education Infant Net Primary per bed Fertility Literacy Country Expenditures Expenditures Mortality Enrollment HDI (beds/1000 Rate Rate as % of GDP as % of GDP Rate Rate population U.K 9.3 5.6 3 4.44 1.9 99.85 99 0.90 U.S.A 17.19 5.3 3 6.17 2.01 93.75 99 0.92 China 5.2 2.05 3.8 14.79 1.55 95.1 95 0.73 India 3.9 3.8 0.9 43.19 2.51 92.4 62.8 0.62 Pakistan 2.5 2.6 0.6 57.48 2.86 73.9 54.9 0.55 Ethiopia 4.7 4.5 6.3 55.77 5.23 85.6 39 0.44 Afghanistan 9.6 3.31 0.4 117.23 5.43 28.65 28 0.47 Bangladesh 3.7 1.92 0.6 45.67 2.45 90.7 57.7 0.57 Malaysia 3.6 4.96 1.8 13.69 2.58 98.1 93.1 0.78 (Source: The World Bank, 2016)

The high rate of fertility in Pakistan always remained a problem for every government with the addition to increasing unemployment and poverty, low health status and high inflation. The continuous rise in fertility rate leads to more population growth which remained a serious problem. For example, on average basis, total fertility rate was 2.86 in year 2016 which is very high if we compare it to U.K who has fertility rate of only 1.9 and even India (having largest population after china ) which has fertility rate of 2.51 while china has only 1.55 (The World Bank, 2016).

According to MDGs, deadline of reducing the infant mortality rate is 40 while Pakistan still has very high infant mortality rates of 57.48 per thousand live births compare to 4.44 in U.K, 6.17 for U.S.A and 13.69, 14.79, 43.19 and 45.67 respectively for Malaysia, China, India and Bangladesh While it is 55.77 for Ethiopia and 117.23 for Afghanistan. Pakistan lacks far behind regarding access to remote regions as well as coverage in terms of hospital beds, hospitals, skilled paramedical staff and doctors. As far as availability of health services is concerned availability of number of beds per one thousand persons is very critical that is only 0.6 while it is 3 in U.K and U.S.A. 3.8, 1.8, 0.9, 0.6 and 0.4 in China, Malaysia, India, Bangladesh and Afghanistan respectively.

In Case of education, indicators target primary enrollment rate to 100 percent according to millennium development goals while till 2012 this rate was 57 percent and now in 2016 it is only 73.9 percent for Pakistan while it is 99.85, 98.1, 95.1, 93.75, 92.4, 90.7 and 85.6 percent for U.K, Malaysia, China, U.S.A., India, Bangladesh and Ethiopia respectively. The target literacy rate was 88 percent while it is only 54.9 percent for Pakistan in 2016 which is very low as compared to other selected countries for reference. Pakistan falls well behind in terms of human resource development. According to human development index, the rank of Pakistan is 147th out of 185 countries even far behind than that of neighboring India which is at 131st (UNDP, 2017).

Due to lack of vocational and technical education and low levels of educational attainment, labor market in Pakistan is dominated by unskilled and less educated manpower. A reasonable rise in the enrollment rate and number of educational institutions after 1980s is not yet reflected in labor market. This might be because that most of the bachelor and masters degree programs in Pakistan are merely based on academic education without emphasizing on specific skills. The slow demand for these unskilled graduates in the job markets leads to raising the rate of

8 unemployment among the educated people. So, it becomes important to explore the role of health and education for the economic benefit of individuals and society (Nasir and Nazli, 2000).

This study is therefore undertaken to estimate the economic returns to investment in health and education in Pakistan with special reference to Multan district for micro analysis and impact of investment in education and health on economic growth of Pakistan for the macro level analysis. In Pakistan, numerous studies of the same nature have been undertaken either dealing with education or health; however, attempt on the combination of both has been an ignorant area especially at micro level. Moreover, in this study, we treated education both as a discrete and continuous variable and quality of investment in education is evaluated in detail by using primary data for Multan district. Similarly, quantitative and qualitative aspects of investment in health are estimated with great care as this is the most ignorant area in literature regarding this issue, and very few studies are available regarding returns of investment to health in Pakistan. At the same time we explored the outcomes of investment in education and health at the macro level in Pakistan and explored the relationship between economic growth, education and health investment.

1.3 Significance of the Study

The Millennium Development Goals (MDGs) were adopted by 189 member countries of United Nations in September 2000 to make considerable progress toward the alleviation of poverty and attaining human development goals by the year 2015. The eight goals are: to mitigate extreme hunger and poverty; to attain universal primary education;; to improve maternal health; to promote gender equality and empowerment of women; to reduce child mortality; to combat malaria, HIV/AIDS and other diseases; to develop a global partnership for ensuring environmental sustainability and development. The goals were then assigned special targets deemed achievable by 2015 based on the evidence of past international achievements in economic development.

Pakistan is a signatory to the Millennium Development Goals (2000-2015) of United Nations. To attain eight millennium goals, the United Nations has fixed 48 indicators and 18 targets; from which Pakistan has adopted 37 indicators and 16 targets. Three out of eight goals of MDGs emphasize on health sector with sixteen indicators and four targets. The MDGs include:

Improving Maternal Health (1 target, 5 indicators), Reducing Child Mortality (1 target, 6 indicators and Combating Malaria, HIV/AIDS and other diseases (5 indicators, 2 targets).

Though Pakistan has registered a remarkable decrease in maternal and child mortality rates since 1990 with the help of essential interventions, and to combat the main diseases like measles, malaria and HIV etc. However, slow progress in the indicators of child mortality and maternal health are major constraints in the progress of Millennium Development Goals. The MDGs deadline of reducing under-five mortality rate to 52, the infant mortality rate to 40 and maternal mortality to 140 by the year 2015 seems difficult to attain in such situation.

Goal two and three focus towards development of education, Goal two ensures that children will be able to complete a full course of primary education by the year 2015. This goal emphasizes on three main indicators; net primary enrolment ratio, literacy rate and survival/completion rate from grade 1 to 5. Target primary enrollment rate was 100 percent while target literacy rate was 88 percent in Pakistan that is not properly attained due to negligence or low investment in these sectors.

Goal three is removing gender inequality in primary and secondary education by the year 2015. The government of Pakistan tried to achieve these targets and a lot of effort and planning is required to meet the targets mentioned in Vision 2030 of Planning Commission of Pakistan for healthy and an academic environment which also promotes the thinking mind.

In the light of above discussion, it is verified that physical and human capital are two main drivers of economic growth therefore this study focus on two main components of developing human capital that is education and health as the building blocks in the light of literature and empirical work on micro and macro levels. This study also evaluated the achievement of MDGs and contribution of this plan in improving the health and education status of Pakistan. This study is planned to make an analysis of investment in education and health sector in order to make future strategies and to estimate the effect of additional investment in these sectors in Pakistan, especially in Multan district.

1.4 Research Questions

i. What is the profitability of investment in the education of individuals? Whether this profitability differs with respect to gender, geographical region and marital status etc.? ii. What is the profitability of investment in the health care of individuals? Whether anthropometric measures are good indicators of childhood investment in health? iii. Whether the investment in education and health has positive externalities? Whether these investments are beneficial for the economic growth of Pakistan or not?

1.5 Objectives of the Study

Keeping in view the significant role that education and health can play in an uplifting living standard of the people, this thesis aims at empirically evaluating the role of investment in health and education sectors in the life of individuals and in economic development of Pakistan. The thesis has an overall objective of looking at returns to investment in education and health in Pakistan and focusing on the following;

i. To explore the returns of investment in education at the micro level disaggregated by gender, marital status and some other demographic factors taking education as discrete and continuous variable. ii. To estimate the returns of investment in health care at the micro level by using different health care and anthropometric measures of health. iii. To investigate the returns at macro level of public investment in health and education in Pakistan for the period 1972-2016.

1.6 Limitations of the Study

This study is based on both primary and secondary data analysis, due to large sample area in spite of great effort human bias is unavoidable. Deficiency of permanent information and data has restricted a comprehensive investigation of the study. There is no well-defined and planned set of data; most of the accessible data are incomplete and fragmented. In case of returns of investment analysis proper data set is not available at macro level regarding earnings and related factors, therefore many problems were faced by the researcher to follow basic earning function method for estimation at macro or even at micro level in case of health returns. The available

11 data related to this study is deficient in accuracy. Various sources report different figures on the same issue. Many important variables are not taken into account due to imperfect data about those variables or because of the econometric problems.

In case of education returns various other methods can also be used for the analysis of returns like shortcut or elaborate method which is not used here due to unavailability of relevant data therefore only Mincer earning function is used. Education returns can also be disaggregated by various other dimensions and many other proxies can be used for quality of education like student teacher ratio, term length etc.

In case of returns to investment in health I used various dimensions of health returns like anthropometric measures, self-reported health status, disability and various environmental and nutritious factors. Due to self-reporting bias many respondents don’t report their health status properly therefore incident of major and minor illness is not evaluated in detail while it is an important dimension of health. No- monetary return like happiness, prestige may also be estimated if some proper dataset is available. Various interaction terms can also be used for the analysis of various relations in depth. Like interaction term of education with gender, region, marital status etc. Similarly some other proxies can also be used for macro returns. These are some issues faced during this study but these are some dimensions open for the new researchers.

1.7 Organization of the Study

The thesis includes, beside this first introductory chapter, eight other chapters. The outline of these chapters and their organization is described as follows;

Chapter one covers brief introduction of the study which includes Background of the study, Problem statement, significance, objectives and limitations of the study, and organization of the study.

Chapter two contains a retrospective study of Pakistan’s economic background and existing situation of education and earning in Pakistan, problems and challenges faced by this sector will also be discussed. Similarly, this chapter contains the history, problems and challenges faced by the health sector in Pakistan. This chapter also provides a detailed profile of Multan district and

12 the present and past condition of education and health services in this district that is randomly selected for micro level analysis.

Then the chapter three will discuss the relevant national and international literature related to the returns of investment to education and health. The theoretical and conceptual framework and empirical background relevant to the topic will be presented in chapter four.

Chapter five will provide different econometric and sampling techniques to estimate returns to education and health both at the micro and macro level. Variables used in the study and their explanation will also be presented in this chapter. It will discuss the available data sets for the estimation of model and examine the nature of the main data sets.

In chapter six an elementary data analysis will be discussed about the core variables used at micro level and then Chapter seven will provide the micro level estimates about returns of investment to education and health, by using data collected from Multan district.

In chapter eight statistical and econometric techniques will be used for estimation of the model regarding returns of investment to education and health at the macro level by using secondary data source for Pakistan and then obtained results will be discussed.

Finally, the conclusion will be summarized in chapter nine; the main findings will be narrated and will explore policy implications using the evidence revealed in this study. Some suggestions for future research will also be discussed.

Chapter 2 THE RETROSPECTIVE STUDY OF HEALTH AND EDUCATION SECTOR IN PAKISTAN 2.1 Introduction

In the first section of this chapter different health policies and five-year plans are given which are presented by the government of Pakistan to improve the health sector of Pakistan. The main purpose of these health policies and plans are to overcome the health related problems by improving the health management and providing facilities in hospitals including basic health units in the country. While in the second section history of the education sector and various policies about the development of education sector are discussed. In this chapter history of Multan district and its important features are also discussed in detail because this district is selected for primary data analysis as sample area therefore a detailed analysis of present situation of health and education sector in Multan is also presented.

2.2 Health Policies during the Study Period

After independence, the health care system of Pakistan was the legacy of British Empire. This health care system included some curative services and public health services. During 1947 to 1955 the basic problem of health care system was the replacement of medical staff in hospitals. From the year 1955, the five-year plans were introduced to improve the health and education sector and to uplift the economy. The government of Pakistan has presented different policies time to time. Some of the major health policies of Pakistan are: Health Policy 1990, Health Policy 1997, National Health Policy 2001, Health Policy 2009 and National Health Vision 2016- 2025. The above-defined health policies are based on the Alma Ata Declaration of 1978.

2.2.1 Alma Ata Declaration 1978

In September 1978, the UNICEF and WHO had jointly arranged a world health conference at the place of Alma Ata in Kazakhstan. The main objective of this conference was the historical declaration of qualitative approach in health sector in under developed countries. Pakistan has also become one of the pioneer signatories of that Alma Ata Declaration. This declaration has set the target of “Health for All” by the year 2000 (WHO, 1978).

I. Important Features of Alma Ata Declaration

Some important features of Alma Ata Declaration were as follows:

i. The conference confirmed that health services are basic human rights and the achievement of health facilities is an important social good. ii. The inequality which exists in the status of health between developing and developed countries is socially, politically and economically unacceptable. iii. Social and economic development which is based on New International Economic Order has a significant role in achievement of health facilities for all rich and poor countries. iv. The people of a country have a duty and right to participate collectively and individually in the enforcement of facilities related to health. v. It is the basic responsibility of the countries to provide the health facilities for their people. vi. The facilities which are related to primary health care should be considered necessary for health care based on scientific methods. vii. Each country should make strategies and plan of actions to sustain and launch the sector of primary health care. viii. The better level of health status under the vision “Health for All” would be attained by 2000 with the help of better and fuller use of world resources.

To promote the facilities of primary health care, the Alma Ata conference has called for effective and urgent international and national action to implement and develop health care facilities throughout the world. Alma Ata Declaration is the first declaration of health related issues which were binding on member countries including Pakistan. The government of Pakistan announced its first health policy in 1990.

2.2.2 Health Policy 1990

The decade of the 1970s has been considered as the start of serious consideration and debate regarding the formulation of health policy. However, in 1988, the government was serious to formulate a national health policy which was announced in 1990.

I. Objectives of the Health Policy 1990

The important objective of the health policy 1990 was to provide the facility of health to the whole population. This policy has recognized the significance of primary health care. Following were the main objectives of this policy:

i. Primary health care system will be available throughout the country for the people without any discrimination.

ii. Health care system will be decentralized and so management boards will be formulated for this purpose.

iii. Jurisdiction and scope of institutions related to social security will be expanded.

iv. The private health sector will be enlarged and improved by giving incentives and should be controlled by provincial governments.

v. Essential and life-saving drugs will be included in separate list.

II. Major Targets of the Health Policy 1990

Following are some major targets of this health policy:

i. Different cadres will be created of trained personnel to attend the childbirth and pregnancy in hospitals.

ii. To improve the nutritional status, necessary arrangements will be done so that the infants should have a weight at least 2.5 KG at the time of birth.

iii. Infant mortality rate should be reduced up to 50 per thousand

iv. Life expectancy at birth should be increased up to 60 years.

v. Public expenditures on health should be increased up to 5 percent of GNP and these will be distributed equally between rural and urban areas.

vi. The health insurance scheme should be introduced for the selected regions of the country.

III. Implementation of Health Policy 1990

i. Basic public services related to health were established in a country on integrated basis.

ii. The community was involved in health sector at the outside to identify their needs about health facilities.

iii. Health information reporting and medical audit system should be introduced.

iv. To give education to the people so that they will improve their lifestyle and to provide social and economic support to all health related programs

IV. Social Action Program (SAP)

After 1990, it was felt by government to correct the bad performance of the social service sector that was responsible for poor health services and poor health status. With the collaboration of World Bank and other donor agencies, government of Pakistan started the Social Action Program (SAP) in 1993-94. The first stage of SAP was completed with estimated expenditures of Rs. 106.4 billion. After that SAP-II was launched for the period of 1997-2002 with estimated expenditures of Rs.498 billion. The main focus of SAP was on improvement of primary health, primary education, sanitation, water supply and population welfare.

Although Social Action Program was successful in allocating more funds for primary health care facilities but it was unsuccessful in one of its objective to increase the non-salary budget. It was found that the ratio of non-salary to current budget was 0.30 in 1996-97 and it was also 0.30 in 1993-94. SAP was also failed to achieve the goal of reducing infant mortality rate which was set as 65 but remained at 82 in 2002. But of the achievement of SAP is that it diverted attention of government and health planners towards primary health care

V. Health Related Millennium Development Goals (MDGs)

SAP program of Pakistan was followed by the Millennium Development Goals (MDGs) which was set by United Nations and other international donor agencies in the year 2000. Pakistan was a signatory to the 8 Millennium Development Goals and 18 special targets which would be achieved by the year 2015. Six targets were directly related to the health sector. There were

17 important implications for MDGs as Pakistan had enhanced implementation and financial capacity to fulfill the targets.

There are many targets which were achieved under MDGs. IMR was reduced from 120 in 1990 to 40 in the year 2015. MMR was reduced from 550 to 140 in the year 2015. MDGs have a capacity to become a new tool to measure health performance in future. It was observed a little increase in health expenditures through 2000-01; 2001-02; and 2002-03 which were stood at 0.7%, 0.74% and 0.78% respectively. A detailed representation of these targets is given in appendix 2.

2.2.3 Health Policy 1997

The health policy of 1997 was announced along with Social Action Program. Health issues were the main focus for the both public and international organizations. The health policy 1997 set its recommendations for the next ten year period along with some challenging targets.

I. Vision 2010 of Health Policy 1997

The main goal for Pakistan is to obtain the target of ‘Health for All” through the system of health care. The platform for social and economic change was created to improve the lives of people. This was based on the idea of health and social dimensions where health was considered as a source of life quality and economic development.

The Vision 2010 was considered as a complete program of development for the health and education sector. The main purpose of Vision 2010 is to limit the burden of ill health from its all preventable causes. Despite the program of primary health care the well equipped and highly organized tertiary level health care will be available for people at affordable price. The main goal for the programs of health would be to promote a better quality of life and to ensure basic health services and to attain maximum level of national development. The important targets of Vision 2010 of health policy 1997 are given in appendix 3.

II. Objectives of the Health Policy 1997

i. The people of Pakistan will be provided with preventive, rehabilitative and curative services with effective and affordable access.

ii. Awareness should be created in the community so that they can involve in the health sector.

iii. In rural areas, family planning and reproductive health services will be expanded.

iv. Different programs related to primary health care will be started like nutrition and malaria control etc.

v. Malnutrition which is prevailing in many areas should be reduced.

III. Strategies for the Health Policy 1997

The important strategies for the Health Policy 1997 are as follows:

i. The health care system at the district level will be build up through the provision of necessary support for logistic and training and in the delivery of basic elements of primary health care. ii. Satisfactory number of staff will be ensured for capacity building of human resource in basic health care units and regional health care centers. iii. Referral system will be upgraded so that emergency, secondary and tertiary level services of health care will be equitably accessed. iv. Effective and direct involvement of community will be confirmed, and the coordination between health care and other government and non-government organizations would be insured. v. Different national health schemes would be introduced and private health sector will be encouraged as an alternative way to finance the health care. vi. All vertical programs at district level health system will be integrated into primary health care system. vii. To promote health care system, the management of health institutes will be decentralized.

IV. Priority Areas for Health Policy 1997

Improvement of health standard is the main focus of Health Policy 1997. For this purpose the priority areas are as under:

a. Concentration on Serious Health Issues

The top priority for this health policy was to start concrete efforts about some serious health issues. These health issues include diarrhea, infections, acute respiratory diseases and malnutrition in children.

b. Population Control

To control population extensive reproduction services of health will be promoted in rural and urban areas including family planning.

c. Poverty and Ill Health

Poverty is responsible for the ill health of the people, and a considerable portion of the population is living below the poverty line. So, it is very difficult for health interventions to do more for improvement of health conditions.

2.2.4 National Health Policy 2001

National health policy 2001 was presented by the government of Pakistan in June 2001. This policy initiated the health agenda which was initiated by the government. This policy had found ten strategies which can bring major improvement in health sector. This policy provided a national vision for health sector which was based on the approach of “Health for All”

The most important thing for the success of any health policy is implementation of that policy. The health policy of 2001 had outlined the plan of implementation and specific targets had been set. Time frame for each key area was set over the period of 10 years. This policy was implemented with the partnership of federal health ministry and provincial health departments.

I. Main Features of the Health Policy 2001

i. Investment in the health sector will be considered as a part of poverty reduction plan by the government.

ii. Primary and secondary health sectors are required more attention rather than only concentration on tertiary health care.

iii. Good governance can achieve quality health care services so it should be the base of health sector reforms.

II. Specific Areas of Reforms

Following specific areas of reforms had been found for the health policy 2001:

i. Reducing the prevalence of contagious diseases (i.e. Malaria, TB, EPI cluster of childhood diseases, Hepatitis-B and HIV-AIDS.

ii. Eradication of managerial or professional deficiencies in the health system at district level.

iii. Gender equity will be promoted. iv. Nutritional gaps will be bridged in the poor population. v. Urban biasedness in health sector will be corrected. vi. Necessary rules and regulations for private health sector will be introduced. vii. Mass awareness will be created in the matters of public health. viii. Effective improvements will be created in the drug sector. ix. Capacity building will be initiated for the monitoring of health policy.

The health policy 2001 has a clear idea that what should be done to improve the health services. This policy acted as a collective framework and also provided a method for implementation of plans in health sectors according to the requirements.

2.2.5 Health Policy 2009

Health is a necessary social service for families, individuals, communities and nations and they cannot achieve the required level of economic development without healthy minds. The major goal of national health policy is to remove barriers in the way of affordable and essential health services for every Person. The foundation of health policy 2009 is the idea that the health service is a right of every citizen and this can be achieved by following main principles:

I. Principles of Health Policy 2009

There are following principles for health policy 2009;

i. All the citizens will be covered by the package of health insurance. ii. It is necessary to overcome economic and social inequalities to improve the outcomes of health. iii. Quality care to everyone should be provided. iv. Good governance, transparency and promotion of meritocracy should be implemented. v. Evidence based decision making should be promoted at each level of health so that actions deriving from health policies and policy development are feasible, relevant and also socially and culturally acceptable.

II. Objectives of the Health Policy 2009

The main objectives of the health policy 2009 are following:

i. Enhancing and improving access and coverage to basic health services. ii. The burden of diseases should be reduced. iii. Underprivileged and poor people will be protected against catastrophic expenditures of health and other risk factors of health. iv. Health care system will be strengthened with special focus on utilization of resources. v. In health sector, stewardship functions will be strengthened. vi. To promote strategic planning and evidence based policy in the health sector. 2.2.6 National Health Vision 2016-2025

Provision of health care service is a responsibility of provincial government and National Health Policy 2016-2025 would be in concert with the needs of provinces, priorities and expectations. The federal government of Pakistan will facilitate and supports the provincial governments by implementing and developing their strategies, by facilitating for technical and financial resource and by providing the overall vision to ensure that necessary services related to health would be accessible to all people.

I. Vision Statement

To improve the health status of all citizens of Pakistan, particularly children and women, through affordable and responsible health care system, ready to achieve the goals of sustainable economic development and fulfill its other responsibilities of global health. II. The Principle Values of National Health Vision 2016-2025

The National Health Vision provides a unified and responsible way to confront various challenges of health by ensuring to universal coverage of health as its ultimate objective. The principle values of National Health Vision 2016-2025 are as follows;

i. Innovation and Transformation

ii. Equity and pro-poor approach

iii. Good governance

iv. Transparency and Accountability

v. Cross-sectoral synergies and integration

vi. Responsiveness

III. Objectives

Following objectives are adopted by National Health Vision 2016-2025 to improve the health of Pakistani people.

i. To create cooperation between provincial and federal governments in consolidating the development and learning from past experiences and moving towards the universal coverage of health.

ii. To provide a unified vision for improving the health by ensuring the autonomy of provinces and diversity.

iii. To facilitate the coordination of the information collection, regulation, surveillance, and research for the improved system of health.

iv. To facilitate the synchronization of commonality towards international treaties and international reporting.

v. To provide the foundation for implementing the SDGs and charity in collaboration with other sectors of the economy.

2.3 Five Years Plans over the Study Period

The mechanism of five year plans was adopted to promote the whole economy and to achieve the goal of economic development. Here, the status of health in these five-year plans will be discussed.

2.3.1 4th Five Year Plan (1970-75)

Social progress and economic growth of a country are closely related to the health of the population. The health of citizens is not only dependent services of health but also on food supply and environment.

The strategy of the 4th plan was to seek a proper balance between preventive and curative services. Rural health services had been given the highest priority. Another factor which was the main cause of hurdle in health development was the shortage of trained staff in hospitals in all categories.

I. Objectives of the 4th Plan

The main objectives of the 4th five-year plan were following:

i. To improve the working and living conditions of the citizens through the safe water supply, improved sanitation and other facilities. ii. To provide balanced nutrition to the people who are below the poverty line. iii. To improve and develop comprehensive services related to health through basic health units. iv. To provide free health services for children, infants, mothers and adolescents. v. Health establishment for the industry will be setup to fight against the diseases and to care laborers and their families.

vi. More hospitals will be established with the facility of competent specialists to deal with major diseases at district and subdivision level. vii. To ensure and provide sufficient medicines for protective health purpose. viii. To train teachers, doctors and paramedical staff of hospitals. ix. To integrate health services from the district level to each union council level.

II. Major Targets of 4th Plan

The major targets of 4th five-year plan were following;

i. Eradication of malaria

ii. Eradication of smallpox

iii. Control of tuberculosis and immunization of B.C.G

iv. Maternity and child health services

v. Leprosy control

To fulfill above objectives, the government of Pakistan had allocated Rs. 2445 million. Detail of expenses is given in appendix 4. Remarkable progress in the health sector was noticed during 4th five-year plan. The expenditures on health sector as the percentage of GDP increase from 0.47 percent in 1970 to 0.72 percent in 1978. Special importance was given to malaria eradication program. Medical colleges in Pakistan were increased from 6 to 15 and the enrollment of medical students which was 900 students annually was increased to 4000 students annually. Lady health visitors increased from 1881 to 3250 and number of nurses increased from 5400 to 9711. In 1976 Drug Act was introduced. Physical achievements during fourth five-year plan are presented in appendix 5.

2.3.2 5th Five Year Plan (1978-83)

In 1978, demographic and health condition Pakistan was formulized by low death rate and high birth rate and high rate of population growth. Health expenditures as percentage of GDP were increased to 1 percent, and it was estimated that about 50 percent population of Pakistan would be in the range of 2-mile distance from public or private health services.

I. Objectives and Targets of the 5th Plan

The main targets and objectives of 5th five-year plan were as follows;

i. To make such arrangements that primary health care unit would be available within 2-4 miles for the entire population.

ii. Crude death rate would be reduced from 14 per thousand to 10.2 per thousand.

iii. Infant mortality rate of 105 per thousand would be reduced to 79 per thousand.

iv. Life expectancy would be increased from 54 years to 60 years for men and from 53 years to 59 years for women.

II. Quantitative Targets for 5th Plan

There are following quantitative targets for 5th five-year plan;

i. To provide at least one health care unit for each union council level.

ii. 5221 more basic health units will be established, and the ratio of persons served by each health unit would be increased from 12494 persons in 1978 to 7660 persons in 1982-83.

iii. Facilities of health in urban areas of Pakistan would be modernized and improved.

iv. All hospitals situated in district and tehsil level will be provided with health specialists and modern equipment.

v. Smallpox would be maintained at zero level.

III. Physical Targets of 5th Five Year Plan

Physical targets of the 5th five year plan are presented in appendix 6 where Rs. 6600 million was allocated for these targets and this represented an increase of 160 percent from 4th five-year plan. The recurring expenditures for health sector were increased from Rs. 558 million to Rs.13895 million. The detail of different expenditures is as follows;

Table: 2.1

Allocation of Funds for the 5th Plan

S. No Programs Million Rs.

1 Preventive Programs 1400

2 Health Manpower development 900

3 Rural Health Programs 2900

4 Hospital facilities 2030

5 Research in Medical Field 74

6 Others 60

Total 7304

Source: 5th Five Year Plan

IV. Physical Achievement of the 5th Plan

During the 5th five year plan it was decided to build 4596 basic health units (BHUs) and 625 regional health units (RHUs) but at the end of 5th five-year plan only 1617 basic health units and 206 regional health centers were built. It was decided that infant mortality rate would be reduced from 105 per 1000 to 79 per 1000, but this target was not achieved and at the end of this plan infant mortality rate remained only on 100 per thousand. The plan was also failed to increase life expectancy for both male and female. In some fields, the plan showed improvement in some extent. It was target that total number of doctors should be 13512 and at the end of plan there were 10203 doctors which was75 percent of the target. In case of total number of nurses, the plan got 89 percent achievement of the target.

2.3.3 6th Five Year Plan (1983-88)

In 6th five year plan it was decided that to improve the quality of life it was necessary to improve the health service of country. It was compulsory to maintain integrated health care system in

Pakistan. It needs expansion and nationwide consolidation regarding infrastructure, staffed and equipped with motivated and trained persons. Obviously the allocation of both revenue budget and capital would be manifold. The allocation of budget for health sector was increased and for 6th five-year plan, the total budget outlay was Rs. 13 billion.

I. Objectives of the 6th Plan

There are following main objectives of 6th five-year plan;

i. The crude death rate will be reduced from 12 per thousand to 10 per thousand.

ii. The life expectancy will be increased from 55 years to 60 years.

iii. The contagious diseases will be reduced from 30 percent to negligible level.

iv. Infant mortality rate (IMR) will be reduced from 100 per thousand to 60 per thousand.

v. All newborn and children will be protected against 6 preventable diseases.

vi. Third-degree malnutrition among children will be eliminated. vii. Every mother will be provided trained birth attendants during child birth. viii. Prevention of the possible occurrence of any kind of disabilities and special care for disables.

II. Targets of 6th Plan

During the 6th five year plan following physical infrastructure were needed to meet the national requirements;

i. 2620 current facilities in basic health centers will be converted with doctor residences.

ii. 2600 new basic health centers will be constructed with attached residential area for doctors and medical staff.

iii. 355 new rural health centers will be constructed.

iv. 1715 residences of doctors will be constructed at the existing basic health unit.

v. 3500 teaching beds will be provided in the medical colleges of Pakistan and another 3500 and 1220 in district level and tehsil level hospitals respectively.

vi. Hotel accommodation will be provided for physicians, trainee registrars and house surgeons.

13 billion was allocated for all these objectives and targets during the 6th five-year plan; the details are in appendix 7.

III. Achieved Physical Targets During the 6th Plan

There are following physical targets that were achieved during the 6th five-year plan:

i. The child death ratio was reduced from 12 per thousand to 11 per thousand.

ii. The life expectancy of the people was increased to 61 years.

iii. Total 85 percent union councils of Pakistan were provided with rural health centers and basic health units.

iv. The infant mortality rate (IMR) was decreased from 98 per thousand to 80 per thousand.

v. The children up to the age of 5 were 100 percent immunized, and this immunization saved 45000 children from getting disabled and saved 100000 children from dying.

vi. Diarrhea treatment with the method of oral rehydration salts showed satisfactory progress. vii. The traditional birth attendant was increased to 30000.

2.3.4 7th Five Year Plan (1988-93)

The main purpose of 7th five-year plan was to remove the rural-urban imbalances and improving the quality of health care, free treatment of persons who were suffering from TB and to establish effective emergency services. Imbalances prevailed in health sector would be removed.

I. Objectives of the 7th Plan

The 7th five-year plan has following main objectives;

i. Infant mortality rate will be reduced from 80 per thousand to 60 per thousand.

ii. All the new born children will be protected from tetanus.

iii. The life expectancy will be increased from 61 years to 63 years.

iv. The appearance of goiter disease will be prevented in the areas where this it is prevalent.

v. All the children will be protected from the occurrence of malnutrition.

vi. To reduce the chances of communicable diseases in the society.

II. Major Policy Directions

Major policy directions for the 7th five-year plan were as follows;

i. Quality of care at all levels will be improved.

ii. School health services will be introduced nationwide.

iii. Allied and emergency services in health would be improved further.

iv. Fertility regulations will be a focal point for government in primary health care service.

v. Well trained health auxiliaries will provide outreach health services in rural areas.

vi. Necessary arrangements will be arranged to improve the nutritional status of people. vii. Health insurance will be introduced against major illness. viii. Establishment of private hospitals and clinics will be encouraged, and appropriate incentives will be given to them.

ix. Imbalances in development of health manpower would be improved.

x. Care of disables and prevention of occurrence of disability would be continued.

III. Strategies for Nationwide Health Care System

For improvement in nationwide health care system the following strategies were adopted in 7th five-year plan;

i. The emphasis of the government would remain on the curative aspects and prevention of diseases.

ii. All the rural health centers and basic health units will be improved. Special beds and labor rooms will be provided in these centers for the improvement of child health care and maternity.

iii. At least one health care center would be set up in major urban areas for the population of 25000 persons.

For all above activities the budget of Rs. 13350 million was allocated. The detail is given in appendix 8.

2.3.5 8th Five Year Plan (1993-98)

Prior to the 8th five year plan, the health care system of Pakistan is characterized by high rate of infant mortality rate, presence of diarrhea and pneumonia in children, cardiovascular disease and complications in pregnancy of women. There were some other major problems with health sector, for example, high population growth rate, lack of family planning, prevalence of communicable diseases in the society etc. So in 8th five-year plan, the government tried to overcome these critical issues in health sectors.

I. Policy Initiatives of the 8th Plan

There are following major policy implications for the 8th five year plan;

i. To bring out balance in the structure of health for both rural and urban areas and provide health services at the door step of citizens.

ii. Economic coordination council will review the policy about import, sale, manufacture and quality of medicines and drugs in Pakistan.

iii. The quality of health care system would be improved and a balance would be maintained between preventive, curative and primitive care and the removal of inequalities in health care system. Health care management will be strengthened through the process of decentralization.

iv. Cancer registries would be established in all the hospitals of the major city. Detection of cancer disease and its treatment facilities would be improved.

v. Mental health services in mental hospitals of Pakistan would be improved.

II. Landmarks of the 8th Plan

The 8th five year plan has following main achievements;

i. Major Imbalances in the health sector was removed and new nurses, pharmacists and paramedics were recruited.

ii. Up gradation of rural health centers and basic health units in the country.

iii. Up gradation of hospital management.

iv. Prevention services against major illnesses covered about 90 percent of the population.

v. 33000 new health workers will be recruited at the village level.

vi. Rs. 1000 million was allocated for the prevention of narcotics.

III. Physical Targets for 8th Plan

Some physical targets for the 8th five year plan of Pakistan are presented in appendix 9. Rs. 9100 million was allocated for all these strategies and objectives. The detail of these expenditures is provided in appendix 10. The 8th five-year plan was the last five-year plan of Pakistan and government of Pakistan has stopped the series of such plans and instead of this health policies and health reforms were started. In every year the government allocated a budget for health.

2.3.6 Reforms in Health Sector (2003)

Government of Pakistan formulated health policies in 1990, 1997, 2001 and 2009. The major objectives of these policies were to improve health sector and also bring about reforms in health sector. But due to lack of implementation, these health policies did not completely meet their required targets.

The main issues for health care sector are poor management, low utilization of health care services, unprofessional attitude of health staff and absenteeism, shortage of medicines, shortage of resources and administrative delays. Health sector reforms (HSR) was introduced in 2003 to correct above-defined problems. Basically, health sector reforms (HSR) is the initiatives and aspirations of the government for mending and remodeling the health sector of Pakistan. The main objectives of these reforms are to provide efficient, progressive and better health care system.

I. Objective of the Health Sector Reforms

The main objectives of Health Sector Reforms (2003) were as follows;

i. To overcome the health related problems by preventive, rehabilitative and curative services.

ii. To make arrangements for the participation of community through change of attitude, the creation of awareness, mobilization and organization of support.

iii. To restructure the health management and planning system and decentralization of powers in health sector.

iv. To improve the utilization of health care services in Pakistan and bridging the gap between health services and community.

v. To include the existing health care programs like malaria control, immunization, mother child health and nutrition in the primary health care system.

vi. To expand maternal child health services and reproductive health services in every area of the country including family planning in both rural and urban areas. vii. To reduce the prevalence of malnutrition so that the health status of mothers and children would be improved. viii. To achieve the partnership between private and public sectors for the provision of health services to masses.

ix. To promote the coordination and actions between different sectors.

II. Guiding Principles

There are following guiding principles for health sector reforms;

i. Training of human resource in the health sector to insure good governance.

ii. Decentralization of Powers.

iii. Integration of different health programs.

iv. Inter-sectoral collaboration.

v. Participation and responsibility of community in its promotion of health.

vi. Active participation of private sector in health

2.4 History of Education Sector in Pakistan

Different initiatives taken by the government of Pakistan for the betterment of our education have been given in this chapter. In Pakistan, different governments have made different education policies. The device of five year plans was used to consider education system of Pakistan according to the social, ideological, national and economic needs of Pakistan. Since independence, these plans have played a considerable part in the development of education. These economic plans were made to get some basic objectives i.e. improving the balance of payment, raising of GDP, increase the job opportunity, providing public services to the citizens like health, education, housing and social welfare.

Education in Pakistan is now provincial matter after the 18th amendment in constitution legislated by parliament in April 2010. Now each province enjoys more autonomy in many economic and social matters including education and health. Standards in Higher Education (SHE) and Ministry of Education and Trainings (MET) are the organizations of the federal government which provide an opportunity to provincial governments for creating synergy and exchange of information, harmony and synchronization with the cooperation of international organizations.

The public school system of Pakistan is the largest social service provider, and it consists of 12 years of academic teaching. It starts from primary level of education and ends with Higher

Secondary School Certificate or intermediate level. The pre-primary class which is also called Katchi but this level is not recognized in terms of examination or budgetary provisions. In addition to private and public schools in Pakistan, there are other institutions which are called Deeni Madaras, and they offer religious education free of cost. These madaras are financed by donations and charity from the public. The majority of children who belongs from poor family attend public schools which provide tuition free education, but these public schools have poor quality of education due to lack of learning material and absence or shortage of teachers.

a. Primary and Secondary Education

In Pakistan, there are 146185 primary schools, 42147 middle schools and 29874 secondary schools. 75 percent schools are in the government sector, and 10 percent schools are run by private sector. Remaining 15 percent schools are equally divided between Deeni Madaras and non-formal basic education. A list of formal education institutions in Pakistan is provided in appendix 11. There are 17.6 million children enrolled in primary schools in which 56 percent are boys, and 46 percent are girls. In middle schools total enrollment is 6 million with 43 percent girls and 57 percent boys while in secondary schools 2.8 million children are enrolled in which 58 percent are boys and 42 percent are females. The detail of enrollment in formal schools in Pakistan is given in appendix No 12.

b. Non-Formal Basic Education

In Pakistan, there are large numbers of Non-Formal Basic Education (NFBE) institutions. In these institutes, 2.5 million students are enrolled. There are about 13000 Basic Education Community Schools (BECS) are working in Pakistan in which 0.6 million children are enrollment. BECS are operated directly under MET&SHE and financed by federal government of Pakistan. The Allama Iqbal Open University (AIOU) also established some non formal middle schools in remote areas of Pakistan with the support of Ministry of Education. Non-formal education in Pakistan has low cost especially for those children who are out-of-school and are residing in remote areas of country.

c. Deeni Madaris (Religious Education)

Deeni Madaris are playing significant role for promoting literacy and education in Pakistan especially religious education and knowledge in Pakistan. Needy and poor children from low income families are studying in these religious institutes. Madaris are providing free of cost education. Some Madaris also teach Urdu, Mathematics and English subjects in addition to religious subjects. According to National Education Management Information System (NEMIS) there are about 13240 Madaris in Pakistan. 1.79 million students are enrolled in these Madaris. Around 58000 teachers are employed in these Madaris (NEMIS, 2014)

2.5 Education policies in Pakistan during the Study Period

In 1947, Pakistan got independence with two separate parts, one was in the northwestern and the other situated in the eastern part in the subcontinent of South Asia. These parts of Pakistan constituted the provinces of Balochistan, North West Frontier Province (Now called Khyber Pakhtunkhwa), East Bangal, Sindh and West Punjab. After partition, the Pakistan faced many problems in which health and education were most important. The country had to face the competition with rest of the world, and this was only possible if the manpower of Pakistan become healthier, educated and skilled. Various governments have framed various education policies during the last 60 years stressing the significance of education.

Pakistan faced very hard competition with other economies of the world after the partition, and the future of Pakistan was depended upon how best it would face this situation with the help of skills and knowledge of its own people. Education is considered an important and fundamental need for the development and growth of human capital. Without educated and skilled manpower no nation can achieve physical and socio-economic development. The government of Pakistan has made various policies about education time to time by suggesting different methods of giving good education to its people. The government of Pakistan has presented various education policies over the study period. These policies are as follows;

2.5.1 New Education Policy 1970

New Education Policy was started on 26th March 1970 by government of Pakistan. The motto of this policy was active process of educational development which implied periodic appraisals of

36 programs and policies and an evolutionary exercise by the government. This policy stressed upon the following main areas of education reform:

I. Areas of Reforms

The reforms of new education policy 1970 are as follows:

i. Education plays a significant contribution to the inculcation and preservation of Islamic values and national unity. ii. All educational programs would be reorganized according to the economic and social needs of the people. iii. The basic role of the education is as a source of development and social change. iv. The educational administration should be decentralized to ensure administrative and financial autonomy as well as academic freedom. Education policy of 1970 stresses upon the enrollment of primary level by the year 1980. For decentralization of education sector, it was suggested to the provinces that they should make autonomous zonal authorities of colleges, autonomous authorities of district school and education service boards.

II. Main Recommendations of New Education Policy 1970

Some of the main recommendations of the education policy 1970 were as follows;

i. Teaching at graduate, postgraduate level should be strengthened ii. For promoting research in the country, a National Research Fellowship Scheme was started to financially support the research activities by the scholars individually or by the group of scholars. iii. Curriculum and course of study should be recognized immediately according to modern standards. iv. In language departments of the universities, the modern techniques of language teaching should be adopted. The government of Pakistan should build two national institutes for the promotion of languages.

v. New colleges in country should be opened to fulfill the ever rising needs of higher education with stress on the education of science, and the facility of the college education should be expanded to all regions. vi. The service conditions and salary scales of teachers should be improved so that the quality of education could be enhanced in educational institutes. vii. For teaching staff in schools and colleges sabbatical leave system should be introduced. viii. It is recommended that law should be made by all the provinces for (a) the rehabilitation of senate in the universities (b) creation of the post of elective principal in the universities of Pakistan, (c) the provision of the withdrawal of the degrees in the universities, and (d) to make universities the center of nurseries and learning of ideas and values.

The total funds of Rs. 892 crores were allocated for this policy. The details are following;

Table: 2.2

4th Plan Break-Up for New Education Policy 1970

Recurring Capital Total

West Pakistan 305 115 420

East Pakistan 233 195 428

Centre 15 30 45

Total 553 340 893

Source: New Education Policy 1970

The declarations of the new education policy 1970 clearly match with the political precariousness in the economy. Political interruption did not permit the fourth five-year plan to be executed, and this policy was relinquished.

2.5.2 The Education Policy 1972-1980

The Education Policy of 1972-1980 gives suggestions like the New Education Policy 1970. However, the nationalization of the private educational institutions was the main theme of this

38 education policy. The implementation of nationalization process put pressure on the financial condition of Pakistan. The non development expenditures of Pakistan rose by six times.

I. Objectives of Education Policy 1972-1980

Some main objectives relating to Education Policy 1972-80 are as follows;

i. To ensure the promotion, practice and preservation of the basic philosophy of Pakistan and make it the code of national and individual life.

ii. Promoting cultural and social harmony to build up national unity. iii. Involving of youth in the programs of social services so that they can play the leadership role in the sphere of life. iv. Eliminating the illiteracy through the adult education programs. v. Designing the curriculum according to the economic and social needs. vi. Providing comprehensive opportunity of studies through the technical education in Pakistan. vii. Due autonomy and economic freedom would be provided to the educational institutes in the structure of national requirements and objectives. viii. The dignity, self-respect and sense of responsibility between the teaching staff and students would be promoted in educational institutes. ix. New universities would be established in Multan, Sukkur and Swat. x. University Grants Commission would be established so that it can act as bridge between the administration of university and government.

The Education Policy 1972-80 was executed to a specific degree. The proposal with respect to the nationalization of private institutes was implemented. Likewise, during the years of 1971-78, there was an increase in enrollment rate at all levels.

2.5.3 National Education Policy 1979

New government was established in the year 1977, and a National Conference of Education was organized by the President of Pakistan in October 1977; the reason for the meeting was to give new suggestions for another policy. The national education policy of 1970 was implemented in February 1979 in Pakistan. The significant points of this education policy were to harmonize the

39 education system according to the teachings of Islam and ideology of Pakistan. It was decided that in all educational institutes the national language of Pakistan would be used as a medium of instructions.

I. Recommendation of Education Policy 1979

Following are some main recommendations of the Education Policy of 1979;

i. Revision of curriculum and reorganizing the whole contents to harmonize with islamic thoughts. ii. Traditional Madrassa education in Pakistan would be merged with the modern education system. iii. National language (Urdu) should be used as a medium of instructions in the educational institutes. iv. Mobilizing the resources of community for the purpose of education and effective participation of common people in the education/ literacy programs. v. Linking the technical and scientific education with the production process in the economy. vi. Separate educational institutes would be established for females. vii. The educational structure which is 4-tier system would be changed to 3-tier system so in colleges the classes of xiii, xiv, xv, and xvi, four years will be considered as a higher education. viii. The minimum strength of students would be fixed for degree colleges. ix. Co-curricular activities would be encouraged in educational institutes. x. Counseling and guidance services would be provided in educational institutes. xi. Postgraduate classes would be introduced in some selected girls colleges in the provinces.

Total expenditures on National Education Policy 1979 were 10281 million, and non- development expenditures were 18617 million. In 1979, the implementation of this policy was started. The medium of instructions in government schools switched over to Urdu. In this policy, the nationalization policy in Pakistan was reversed by government and hence private sector was allowed and encouraged to open new schools in country.

2.5.4 National Education Policy 1992-2002

The gross enrollment rate of education at different levels in Pakistan has been very low since independence of Pakistan. To uplift the education in the country various policies had been formulated in different time period. The main purpose of this policy is to arrange the present education system on modern aspects. This would be achieved by promoting primary level education and improving the education quality.

I. Recommendations for the Education Policy 1992

Education Policy 1992 has following main recommendations;

i. To insure 100 percent enrollment of children in primary schools by the end of the year 2002.

ii. To increase the literacy ratio to 70 percent by the end of the year 2002.

iii. To involve the community for the promotion of basic level of education in the country. iv. Introducing a stream of vocational education at public school level and also inviting the private schools in educational programs. v. All engineering and science laboratories in colleges and universities will be equipped with latest equipment and trained staff. vi. More funds will be allocated for the books and journals in educational institutes. vii. In universities, research centers will be established for community-related problems. viii. University Act will be emended for the better management of the universities. ix. For colleges, a separate funding council will be established in each province of Pakistan.

Due to lack of commitment of policy makers, shortage of resources and political instability, this policy could not achieve the required targets.

II. Achievements of Education Policy 1992

The main achievements of this policy are following;

i. School Management Committee was established for the promotion of primary education.

ii. Mixed primary schools were introduced in all provinces of Pakistan.

iii. The qualification for recruitment of new teachers has been lowered to attract the female teachers.

iv. Different literacy programs were launched to increase the rate of literacy in selected areas of Pakistan.

v. Best Teacher Awards were given to the primary teachers in national level competition.

vi. The student unions in colleges and universities were banned in a historical decision of Supreme Court of Pakistan.

2.5.5 Education Policy 1998

The Education Policy 1998 visualized Pakistan as an ideological state. This policy states that Pakistan cannot be considered as a secular country, so philosophy of Islam forms the origin of Pakistan. It was realized that the country cannot survive without converting its entire education system on the basis of Islamic foundations. The education policy of 1998 visualized and emphasized on primary and secondary education. The policy recommended some new initiatives which are as under;

I. New Initiatives of Education Policy 1998

i. Educational Authorities will be setup in each district, and it will control the functions of education foundation.

ii. To improve the enrollment in primary schools and to avoid dismal dropout rate the education policy would implement the compulsory education act by 2004-05.

iii. Financial and technical support would be taken from international donor agencies to expand and restructured primary and secondary education.

iv. Courses related to Islamic/Shariah Law will be offered at Islamia University Bahawalpur and International Islamic University Islamabad.

v. It was decided the first time that new universities in health and sciences sector can be opened by private sector.

vi. Federal law will be instituted to ensure the high quality research output in arts, law, physical sciences and social sciences at university level. vii. Serious efforts will be needed to acquire qualified faculty in universities because by the year 2010 most of the qualified faculty will retired. viii. Islamiyat would be made compulsory from class 1st to graduation level.

ix. From class 6th to 12th Holly Quran would be taught along with the translation.

x. Income generation skills and functional literacy will be provided to the women of rural region of 15 to 25 years age group.

This policy was initiated to enhance and improve the standard and quality of female teachers of primary schools. So it was expected that there should be more female teachers in newly established schools. This policy also stressed on the decentralization of educational institutes. The policy introduced non formal education programs to improve the education sector in the country. This policy also stressed to build new libraries and provide new facilities for research purpose in campuses. It was also decided that the curriculum of bachelor and master level education should be revised.

II. Targets of Education Policy 1998

The main targets of Education policy 1998 were as follows:

i. Illiteracy from Pakistan will be eliminated through programs of informal and formal education.

ii. Pakistan’s current literacy rate should be raised up to the level of 55 percent by the year 2003 and further by 70 percent till the year 2010.

iii. The gross enrollment rate would be 90 percent by the year 2002-03.

iv. The compulsory Primary Education Act would be enforced and promulgated in a phased manner.

v. At least one model secondary school would be build up in every district of Pakistan.

vi. Private investment will be encouraged in the education sector. vii. Matching grants will be provided for establishing private educational institutes through education foundation.

2.5.6 Education Policy 2009

Education is considered as a most significant investment for economic, social and human development of Pakistan. The political, societal and government structures have the significant impact on the effectiveness of the education system. The Education Policy 2009 proposes some policy options and identifies principal challenges of the present education system. Educational interventions of education sector of Pakistan which is based on the core values of faith and religion.

I. Features of Education Policy 2009

Education Policy 2009 has following main features;

a. Policy Actions to improve Equity and Access in Education

i. Millennium Development Goals and “Education for All” shall be achieved by 2015 to improve the education system. ii. The early childhood education for the children of age group of 3-5 years shall be introduced. iii. For primary classes, age limit of students will be 6 to10 years, and free education in Pakistan will include all costs related to education. iv. Complete equity in education i.e. gender, urban-rural and geographical areas shall be promoted. v. Maximum age limit for female teachers should be relaxed for encouraging females in the education sector. vi. Plans will be prepared for expanding basic facilities in schools. vii. The students of class eleven and twelve in colleges will be shifted to schools. viii. Reducing dropout rates at all levels of education shall be the main priority. ix. Vocational and technical education will be ensured and extended at all district level.

b. Policy Actions to Improve Governance and Management in Education i. The capacity of planning and management would be enhanced at all levels of education. ii. Sectoral planning would be promoted in education. iii. A system of harmonization would be developed between the government of Pakistan and international partners for the effectiveness of aid. iv. Separate educational and academic management cadres shall be introduced with specified qualification and training requirements. v. The government shall take initiatives to encourage public-private partnership in different sectors. vi. A common framework of the curriculum will be applied for both private and public educational institutes. vii. For education sector government of Pakistan will raise the allocation of the budget by 7 percent of GDP by the year 2015.

c. Policy Actions to Improve Quality and Relevance of Education.

i. The new educational requirement for elementary teachers will be bachelors’ degree with B.Ed while the old professional certificates PTC and CT shall be phased out.

ii. Institutions and standard will be setup for the training of teachers, accreditation, arrangement and certificate procedures.

iii. The government should ensure that all the recruitments, promotions, posting and professional development of teachers should be done on pure merit.

iv. The development of curriculum should be outcome based and objective driven.

v. In education, the use of information technology shall be promoted.

vi. The wing of curriculum of provincial text book boards and ministry of education shall eliminate the gender biases from the syllabus. vii. National Merit Program shall be initiated to give awards to outstanding students. It will create an order of excellence in the country.

45 viii. The curriculum of vocational and technical education will be revised to meet the new challenges of modern world.

2.6 Five Years Plans over the Study Period

To uplift the economy of Pakistan, various governments have implemented different plans and policies to fulfill economic, national and social needs of an economy. The device of different five-year plans was used to achieve these goals. The five-year plans were made to obtain some main objectives such as creating employment opportunities, increasing national income, providing social services like health and education and improving the balance of payment. Here, those portions of these five-year plans are discussed which deal with health and education over the study period.

2.6.1 4th Five Year Plan (1970-75)

After 3rd five year plan the government felt that quality aspects of development should be changed. It was decided at the beginning of 4th five year plan to reset all the measures and priorities. In documents of 4th five year plan the objectives of the plan are written as “Socio- Economic Objectives of the 4th Five Year Plan” and National Economic Council approved this plan in 1968.

I. Main Objectives of 4th Plan

Some main objectives of 4th five year plan were as follows:

i. To keep up the speed of growth in the Pakistan and to secure the efficient and maximum utilization of physical and human resources. ii. To make Pakistan self-sufficient in almost all fields. iii. To reduce the disparity between different regions.

Rs 75000 million were allocated for the development program of 4th five year plan. The main target for this plan was to achieve the growth rate of GDP of 6.5 percent and so per capita income of people will increase from Rs 567 to Rs 675 in 1974-75. The share in total development plan for public and private sector was 65 percent and 35 percent respectively.

II. Proposed Development Targets for 4th Five Year Plan

The 4th five-year plan was designed within the framework of new targets of proposed developments which were as follows;

i. To create educated electorate and literate population.

ii. To make the education system more efficient and functional in the form of economic development.

iii. To eliminate the existing biasedness and disparity in service of education.

iv. To realize the significance and importance of quality of output in the education sector.

v. To efficient and optimum use of economic and social resources which also include physical facilities for education.

vi. To consolidate and strength the educational planning and research.

An allocation of Rs 3665 million was made for 4th five-year plan. It was 7.5 percent of total funds which was allocated for the public sector development. The details of funds for 4th five year plan for different sub-sectors are given in appendix 13 while Requirement of the Recurring Expenditure during 4th Plan (1970-75) is given in appendix 14. In below table the total expenditures of education sector are given;

Table: 2.3

Total Expenditure under the 4th Plan (1970-75)

Development RecurringExpenditures Total(Rs in Millions) Expenditures

East Pakistan 2230 2320 4550

West Pakistan 1150 3050 4200

Center 285 150 435

Total 3665 5520 9185

Source: 4th Five Year Plan

2.6.2 5th Five Year Plan (1978-83)

In 1976-77, the per capita expenditures on education and training were Rs 34 which was about 1.7 percent of the gross national product of Pakistan. At the same time, in other countries, per capita educational expenditures vary from 2.3 to 3.1 percent of gross national product. In 5th five-year plan it was decided to raise the level of per capita expenditures on education and training from Rs 34 in the year 1976 to Rs 66 in the year 1982. It was decided to pay special attention to primary education. The plan suggested allocating one-fourth part of development expenditures for primary education. It was decided to reduce the dropout rate in primary schools from 5o percent in 1976 to 37 percent in 1983.

The strategy of vocational and technical education was to make the training more relevant to practical life and to improve the skills of labor. The diploma level technical programs offered by vocational institutes will also be strengthened. In this plan, it was decided that the enrollment in engineering colleges and vocational institutes would rise. Rs 27.22 billion was allocated for the education and training. The details are given in appendix 15.

I. Objectives of 5th Five Year Plan

The main objective of 5th five-year plan was the development of universal primary education. An allocation of 24.5 percent of development funds were made for primary sector education against 11.6 percent funds in 4th five year plan. 33.9 percent funds were reserved for secondary level education against 18.4 percent during the 4th five year plan. So it can be concluded that a large part of development funds was used to promote primary and secondary level education. It was emphasized that without technological progress it was not possible to literate the population. It was also decided that those funds which were allocated for up to secondary school level will not be transferred to other development programs.

II. Physical Targets of the 5th Plan

Some physical targets of 5th five-year plan are as follows;

a. Primary Education

i. 1200 new schools will be opened.

48 ii. 24800 existing government schools will be renovated. iii. Enrollment in primary education will be increased from Rs 5.61 million to Rs 8.6 million. iv. 88000 new teachers in schools and colleges will be recruited

v. Girl’s enrollment will be expanded to 11 percent per annum.

b. Secondary Education

i. Enrollment at secondary school level will be increased by 1.35 million. ii. At this way, the education enrollment of students between the age 10 to14 years will be increased to 29 percent from existing 19 percent.

c. Colleges

i. 50 new colleges will be started in which 36 colleges will be made for males, and 14 colleges will be for females. ii. Education enrollment will be increased by 50000 in which 18000 would be in new streams. iii. 63 colleges of the intermediate level will be upgraded to degree level and at college level enrollment will be increased from 5596 to 74963.

d. Universities

i. At university level enrollment will be increased from 2100 to 2500 by the year 1982-83. ii. Centre of excellence should be made in University of Engineering and Technology (UET) Lahore for research in water resources iii. Annual output of graduates from agriculture universities will be increased from 400 to 600.

2.6.3 6th Five Year Plan (1983-88)

In 6th five-year plan Rs 7 billion was allocated for the development of primary sector education. It was decided that universal level of education will be enhanced from class 5th to class 8th then from class 8th to class 10th. It was also decided that the enrollment of secondary level education will increase at about one million. It was also decided that local bodies will be involved in management, planning and maintenance of educational facilities. To increase the level of education the private sector was also allowed to open new private schools.

Women development was the special emphasis of 6th five year plan. The government decided to create different opportunities for women in education, health, social welfare and other sectors. In 6th five year plan Rs 19850 million was allocated for the programs related to manpower and educational development. For public sector, total expenditures would increase from the existing level of Rs 6380 million to Rs 14700 million by the year 1988. Allocation of education and manpower in 6th five year plan is given in appendix 16.

I. Main Physical Targets of 6th Five Year Plan

Some of the important physical targets for the 6th five year plan were as follows;

i. Education participation rate at primary level will be increased from 48 percent in 1982-83 to the level of 75 percent by the year of 1987-88. ii. 40392 new primary schools will be opened in which 31800 will be mosque schools. iii. 3807 new middle schools will be built.1073 middle schools will be upgraded to high schools. iv. Two new engineering colleges will be opened. v. M. Phil and PhD degrees will be started in water resource science in University of Engineering and Technology, Lahore.

2.6.4 7th Five Year Plan (1988-93)

The main target of 7th five year plan was to improve the living standards and income of the people. It was proposed that these objectives can be achieved through the delivery of public services, health and education. So in 7th five year plan the allocation of funds for the public sector development were raised from 15 percent during previous plan was revised to 22 percent.

Total development expenditures were 642 billion in which public sector development expenditures were Rs 350 billion.

I. Main Objectives of 7th Plan

Some main objectives for the 7th five year plan for education and training were as follows;

i. Resource base for the education sector will be broadened.

ii. To provide universal access towards primary level education.

iii. Vocational and technical education would be substantially improved

iv. The quality of education at all levels will be improved.

v. It was decided to increase literacy rate to 40 percent by the year 1992-93.

Rs 23.11 billion were allocated for public sector educational development. It was 3.6 percent of total development expenditures and 6.6 percent expenditures related to public sector development. The details of expenditures are given in appendix 17.

II. Major targets for 7th Five Year Plan

Under 7th five year plan, following targets were set to be achieved;

i. Expenditures for primary level education would be enhanced up to 44 percent of the total expenditures of education. ii. The enrollment rate at primary level education would be increased from 64 percent to 80 percent by the end of the year 1992-93. iii. 34532 new primary schools would be built in which 20000 would be mosque schools. iv. The participation rate at middle level would be increased from existing 30 percent to 42 percent. v. Three thousand new middle schools will be built and enrollment at middle level education would be increased to 1.4 million. vi. 120 new secondary schools would be opened and the participation rate at high school level would be increased from 17 percent to 24 percent. vii. In engineering universities, the enrollment would be increased from 4000 to 5000.

51 viii. It is decided by government that engineering universities will offer only postgraduate and under-graduate level classes.

2.6.5 8th Five Year Plan (1993-98)

Education is a basic right of each person and a key element of development. Since independence of country a considerable facilities has been took place. The 8th five year plan was initiated with total budget of Rs 1700 billion to enhance the socioeconomic welfare of people. It was realized that social and physical infrastructure is important like a new investment. The key targets of 8th five year plan were to increase the GDP by 40 percent, per capita income by 22 percent and to increase the literacy rate from 35 percent to 48 percent.

Social Action Program (SAP) was started in 192-93 as a component of 8th five year plan. The important purpose of SAP was to increase productivity, encourage better facilities of education and reduce poverty. Rs 69.03 billion was reserved for education and training including Rs 39.3 billion for SAP. The detail of expenditures for education and SAP is given in appendix 18.

I. The Physical Targets for 8th Five Year Plan

a. Primary Education

i. Universal access to primary level education for all children which are lying under age group 5-9.

ii. Formulating the law of compulsory primary education for the children.

iii. Increase the education enrollment in primary schools from 12.4 million students to 17.96 million students.

iv. 4300 new mosque schools would be opened in rural areas.

v. 1700 mosque schools would be converted to primary schools.

vi. 30000 new primary schools will be opened. vii. Physical infrastructure would be improved for primary schools.

b. Secondary Education

i. Education participation rate for male students would be increased from 55 percent to 55 percent and for females from 26 percent to 30 percent.

ii. Two million additional capacities of students will be created in class 6th to 10th.

c. Technical Education

i. 30 poly-technics and mono-technics will be established for promoting technical education.

ii. 100 new technical institutes will be established.

d. Higher Education

i. To promote research in sciences The UET (University of Engineering and Technology), Taxila would be upgraded as a model university.

ii. Output of graduates from engineering universities will be enhanced from 4500 to 5500 graduates annually.

iii. In colleges, graduation will be of total 3 years instead of 2 years duration.

2.6.6 Education Sector Reforms (2001-05)

Education Sector Reforms (ESR) was initiated in 2002. It was recommended to overcome all the problems of education sector. It put great attention on different areas of education like support to physical infrastructure, support to non formal and formal education, mainstreaming of Madrasas, assessment and examination reforms. Education Sector Reforms (ESR) believed that goals of equity, access, quality and efficiency can only be obtained by giving full space and support to private sector and civil society.

I. Mission Statement

“Human resource development in Pakistan as a necessary condition for economic progress, prosperity and global peace”.

II. The Vision

The vision of Education Sector Reforms (ESR) was to provide quality education to people at their maximum level. It also emphasized to produce accountable, skilled and enlightened manpower and integrate Pakistan into the age of economic growth.

III. Strategies

Some of the main strategies of ESR were following;

i. Reforms in every sector based on equity and efficiency.

ii. Education is considered as a anchor for the strategy of economic revival.

iii. Mobilization of resources from all channels.

iv. Planning and implementation of education policy on district level.

v. Community participation and public private partnership.

vi. Compulsory primary education under plan of “Education for All” vii. Outcome based budgeting, planning and auditing.

IV. Objectives

Some main objectives of ESR were following;

i. Universalization of adult literacy and primary level education.

ii. Mainstreaming of islamic Madrasas so that the graduates of these Madrasas can get employment.

iii. Quality of education should be improved with the help of training programs, better teaching faculty, textbook curriculum and proficiency based examination system.

iv. To introduce the third stream of vocational and gender education with innovative based approach for student counseling.

V. Targets of Education Sector Reforms

The main targets of Education Sector Reforms in Pakistan for the period 2001-2005 are presented in appendix 19 while financial requirements are presented in appendix no 20 in detail. For the year 2001-04, a package of Rs 55.5 billion was prepared for ESR. But later on, the program was extended to 2005 and budget was this program was also increased to Rs 100 billion.

VI. Achievements of Education Sector Reforms

Following were some main achievements of Education Sector Reforms;

i. The ordinance about compulsory education at the primary level was successfully implemented in Sindh, Punjab, Khaiber Pakhtunkhwa and Islamabad.

ii. Different incentives were introduced for the universal primary education in country.

iii. In September 2002, the government of Pakistan launched a media plan in the name of “Education for All”.

iv. 10,000 schools in different provinces of Pakistan were rehabilitated under Education Sector Reforms.

v. 500 centers for early childhood education were established in primary schools.

vi. In all the government schools, the Parents Teacher Association was set up. vii. Accountability system was introduced to control absenteeism in schools. viii. Mixed primary schools were started to narrow the gender gap.

ix. A new forum of EFA was started in south Asia and Pakistan has coordinated the forum on the basis of quality in member countries.

x. Ministry of Education had introduced National Commission for Human Development (NCHD) to promote literacy initiatives in 16 districts of Pakistan.

xi. EFA Planning on district based was introduced in 73 districts of Sindh, Punjab, Balochistan and Azad Jammu Kashmir. EFA Plans in model districts were completed in Kasur, Chakwal and Sheikhupura.

2.6.7 National Education Policy (1998-2010)

National Education Policy of 1998-2010 emphasized that education is one of the key indicators to cultural, political, moral, and socioeconomic development of a country. The nation should take bold steps to promote education and should make revolutionary reforms. The policy about secondary education was based on social action program which includes improving management and supervisory services, improving the quality, capacity building, financial sustainability and development.

I. Major Issues and Challenges for Education

Some major issues and challenges for education for Education Policy 1998-2010 were as follows:

i. More than 5.5 million children of the primary school who are in age group 5-9 are left- outs. ii. About 45 percent students of primary level have dropped-out of school. iii. In rural areas, teacher absenteeism is a common practice. iv. At elementary level schools, the instructional supervision is very weak. v. It is estimated that one-fourth of primary school teachers are untrained. vi. It is observed that learning materials in schools have poor quality and inadequate.

II. Objectives of Education Policy 1998-2010

There are following main objectives of education policy 1998-2010:

i. 190000 new primary schools will be constructed including 57000 mosque schools. ii. 60000 primary schools will be upgraded. In 20000 existing primary schools double shift will be started. iii. 527000 additional teachers will be recruited in different government schools.

iv. Necessary arrangements would be made to raise the educational expenditures from 2.1 to 4 percent of the gross national product of Pakistan. v. A massive non formal educational program would be launched which will benefit 5.5 million people. vi. It was expected that by the year 2002-03, 90 percent of primary age group children would be in schools and gross enrollment rate will rise to 105 percent. vii. It was estimated that current literacy rate was 38.9 percent. It was proposed to uplift it to 55 percent during next five years and then 70 percent by the year 2010. viii. It was decided to start a literacy movement from each part of the country to increase the schools of Non-Formal Basic Education (NFBE) from 7000 to 82000.

2.6.8 Education for All (EFA) Strategies (2001-2010)

All provinces of Pakistan are planning and implementing strategies and projects related to education out of their own resources. The Federal Ministry of Education created an opportunity for exchange of experiences and information among the provinces. Following strategies and arrangements had been adopted for implementation of Education for All (EFA) initiatives during 2001-2010:

i. Inter-provincial Education Ministers had arranged quarterly meetings for implementation of education policies. Federal Ministry of Education had supported these meetings from its own resources. ii. Provincial and National Plans of Action had been prepared for EFA and MDGs. iii. EFA Wing was established at the Federal Ministry of Education (2001-2004) to coordinate the activities which are related to EFA. iv. National conferences, series seminars and EFA Forums on Education-For-All, and on different areas of education were organized by the Federal Ministry of Education. v. Preparation of periodic assessment and country reports of progress made by country and the provinces as a whole towards achievement of MDGs and EFA was undertaken. 2.6.9 The national plan of action and education related MDGs

Government of Pakistan formulated National Plan of Action (NPA: 2013-16) in September 2013. It was designed to fulfilled education related targets and goals. The main objective of NPA

(2013) was to achieve MDGs (Millennium Development Goals) related to education in next three years.

The NPA was consolidation of eight area and provincial plans specified with local challenges, conditions and interventions. The main target of NPA was to increase the net primary enrollment rate from 68 percent in 2012 to 91 percent in 2016. It was estimated that there were 6.7 million children who were out-of-school and the plan expected to enroll 5.1 million children (2.7 million girls and 2.4 million boys) by 2016.

Following strategies were adopted to accelerate the education related MDGs under National Plan of Action (2013-16):

i. The main focus of this strategy was to enroll out-of-school children in non-formal and formal schools through enrollment drives and motivational campaigns. ii. Additional rooms will be constructed in existing schools and extra staff will be provided in public schools. iii. Two-room formal or one-room non formal schools will be constructed in those under developed areas of Pakistan where education enrollment was low due to lack of public schools. iv. Different incentives will be provided to retain and attract students especially those who belong to disadvantage group, e.g. food for education, stipend, uniforms, etc.

2.7 Snapshot of District Multan

The importance of District Multan is characterized by the historic city of Multan city, popularly known as the city of Saints due to a large number of saints, Sufis (mystics) and religious scholars of great fame lived here at different times. Multan, historically, was located at the crossroads of east-west and north-south caravan trade routes; hence it always enjoyed importance by both local and distance trade. This advantage of location persists to the present time and become even more important, with Pakistan’s main north-south road and rail routes for both national and international trade, east-west link roads and air routes passing through Multan.

Multan District lies at the centre of South Punjab and may be considered as its heartland, given its leadership role in the industry, commerce, agriculture, education and other realms in an

58 otherwise less developed sub-region of Punjab. Notably, Multan has always remained a major centre of politics in Punjab as well as Pakistan. Khanewal District to the north, Vehari District and Lodhran District to the east, district Bahawalpur to the south, and districts Muzaffargarh, Dera Ghazi Khan and Rajanpur to the west surround district Multan.

River Chenab flows along the western boundary of the district, separating it from district Muzaffargarh. The southern boundary of district Multan lies along river Sutlej, separating it from district Bahawalpur. The water of river Sutlej has been exclusively allocated to India under the World Bank mediated Indus Water Treaty of 1960, although surplus floodwater released by India or rainwater at times flows into the otherwise dried up riverbed. In Multan, a network of canals which are fed by Chenab River flows across the district and irrigates the land.

Table: 2.4 Number of urban and rural UCs in each town

Town of Multan City Total UCs Urban UCs Rural UCs District

Sher Shah Town 24 12 12

Shah Rukan Alam 25 16 9 Town

Mumtazabad Town 24 15 9

Bosan Town 24 14 10

Shujabad Town 17 2 15

Jalalpur Pirwala 15 2 13 Town

Total 129 61 68

Source: Punjab Development Statistics 2009

The total area of Multan district is 3721 km2 with four tehsils (Multan Saddar, Multan City, Shujabad and Jalalpur Pirwala). Following local government reforms under the Musharraf regime and the promulgation of Local Government Ordinance 2001, Multan City is divided into six towns ( Shershah Town, Shah Rukne Alam Town, Bosan Town, Mumtazabad Town, Jalalpur Pirwala Town and Shujabad Town. Each town is administrated by Town Municipal Corporations

(TMAs). There are 129 Union Councils in Towns of Multan City. The names and number of rural and urban union councils (UCs) in each town is given in below table 2.4. According to the census of 1998, Saraiki is spoken by 67 percent people of Multan district. The ethno-linguistic division of the Multan district is given in table 2.5. Major castes in Multan district are Arain, Gilani, Syed Gerdezi, Qureshi, Khakwani, Kamboh, Baloch, Pathan, Rajput and Jat. The district has the extreme temperature in summer rising to 49 Centigrade and falling to 1 centigrade in winter. The amount of average rainfall in the district is 127 mm. The land area of the district is very fertile and plain. However, few regions of Tehsils Shujabad and Multan which are closed to the Chenab River are flooded during the season of Monsoon.

Table: 2.5 Ethno-Linguistic Division of the Multan District

Ethno-linguistic Ethno-linguistic % population % population group group

Saraiki 66.58% Urdu-speaking 5.68%

Punjabi 11.14% Baloch 0.07%

Haryanvi 14.59% Pashtun 0.62%

Sindhi 1.04%

Source: City District Government, Multan URL: www.multan.gov.pk

The total population of Multan district according to the estimate of Punjab Development Statistics 2009 is 3.925 million. Population by town in district Multan is given in table 2.6. Federal SAP Secretariat 2002 reports that the total population of the district is 3,116,851 persons and gender wise composition is 16,35,768 (52.49%) male and 14,81,083 (47.51%) female. The rural population is 18, 02,103 (57.8%) and urban 13, 14,748 (42.2%). According to the Multan City District Government official website population, the district’s population is 5 million. Age structure of the district as stated in district census report 1998 is presented in table 2.7.

Table: 2.6 Town- Wise population (thousand persons)

Town As per Census of 1998 Estimated as on 31-12-09

Total Rural Urban Total Rural Urban

Jalalpur 358 32 326 442 39 403 pirwala

Shah Rukn- 779 558 309 289 414 365 e-Alam

Bosan 571 285 286 664 302 362 Town

Mumtazada 560 297 263 730 398 332 Town

Shujabad 427 57 370 530 62 468 Town

Sher Shah 541 273 268 705 366 339 Town

Cantt 62 62 -- 75 75 --

Source: Punjab Development Statistics 2009

Table: 2.7 Age Structure in District Multan

65 & Under1 Under51 Under10 Under15 15 to 49 15 to 64 Above

2.3% 14.3% 30.3% 43.6% 46.2% 53.2% 3.2%

Source: Government of Pakistan, District Census Report, 1998

The population density of the district is 837.9 persons/sq. km. The average size of household is 7.1 persons/ household and growth rate is 2.73%. According to Federal SAP Secretariat Report 2002, the literacy rate in the district is 43.4%. It is higher for male at 53.3% as compared to 32.3% for female. The urban literacy rate is 60.9% and rural 29.5%.

According to the Social Policy and Development Centre (SPDC), Multan ranks 13th on the district deprivation ranking of 34 districts in the province and falls in the category of medium deprivation districts. The Multi Indicator Cluster Survey (MICS) 2003-04 ranks Multan 16th on the district deprivation ranking of 34 districts.

According to a current study of Lahore University of Management Sciences, the poverty incidence in South Punjab is 50.1 percent. There are following districts including in South Punjab: Rahimyar Khan, Bahawalpur, Bahawalnagar, Multan, Vehari, Lodhran, Khanewa, Dera Ghazi Khan, Muzaffargarh, Layyah, Rajanpur. District Multan is selected purposively for investigation in this study.

2.8 Education Sector of District Multan

Education is considered as a significant investment for the economic development of the country. Neglecting of education can cost the society and no illiterate society would achieve the political, social and economic power. Unfortunately, in Pakistan, education sector never gained significance by the government and policy makers which can be observed from the level of educational infrastructure of Pakistan. The education up to the level of higher secondary is the responsibility of district authorities. So each district makes necessary arrangements for the promotion of education in its district.

Regarding literacy rates, the Multan is ranked 20th among 36 districts of Punjab. It can be observed from the PSLM 2010-11 those 48 percent girls and 68 percent boys having age of 10 years and older are literate in Multan.

2.8.1 Enrolment Ratios in Primary Education

The enrollment ratio can be obtained by taking the total enrollments and then divide it by the total number of population. The net enrollment ratio is only that part of enrollment which belongs to the specific age group. The children’s age group belonging to primary level education is 5 to 9 years.

The net enrollment ratio of Multan district is too much low as compared to the net enrollment ratio of the province as a whole. In males, the percentage of students attending primary schools is

less than 29 percent while it is 40 percent in Punjab. In case of females, this percentage is around 27 percent and 37 percent respectively.

2.8.2 Net Primary Enrolment Ratios

The net enrollment ratio of primary education in Multan district in 1998 was 43.3 percent while it was 46.6 percent in Punjab province as a whole and Multan was ranked 20th position among the 34 districts of Punjab.

Table: 2.8 Net Primary Enrolment Ratio in district Multan

Year 1998 2005

Enrollment Ratio 43.3% 49.0%

Rank in the Province 20th 29th

Source: Pakistan MDGs Goals Report, 2006, UNDP Website.

Table: 2.9 Net Enrolment Ratios (Punjab Vs Multan)

Primary School Total Population Total Population of Net Enrolment Enrolment (Dec. 2009) 5-9 years Ratio Gender (2009 - 2010) District District District District Punjab Punjab Punjab Punjab Multan Multan Multan Multan

Females 1,865,160 44,442,150 301,970 6,759,650 83,268 2,127,199 27.57% 37.00%

Males 2,059,840 47,646,850 325,455 7,280,440 77,313 2,667,376 25.94% 39.55%

Total 3,925,000 92,089,000 627,425 14,040,090 160,581 5,006,955 26.73% 38.69%

Source: Punjab Development Statistics, 2011.

In 2005, the enrollment ratio increased to 49 percent in Multan but if we see the rank of the district regarding net enrollment ratio then it went back to 20th position among the other districts of the Punjab. This is shown in table 2.8.

2.8.3 Literacy Rate

Literacy ratio can be counted by taking the number of literates (10 years or above) and then divide by the total number of population (10 years or above). An individual can be called literate if he can write or read a short and simple statement in his daily life.

In 1998, the literacy rate of Multan was 43.4 percent while it was 46.6 percent in overall Punjab. The literacy rank of Multan among 36 districts of Punjab was 17th. Multan’s literacy rate was increased to 48.4 percent in 2005, but this rank regarding literacy went down to the 23rd in overall Punjab (Table 2.10).

Table: 2.10 Literacy Rate in District Multan

Year 1998 2005

Literacy Rate 43.4% 48.4%

Rank in the Province 17th 23rd

Source: City District Government, Multan URL: www.multan.gov.pk

Table: 2.11 Literacy Rate: District Multan Vs Punjab

Year 1998 Year 2006

Punjab Multan Punjab Multan

Female 35.1% 32.3% 35.0% 37.0%

Male 57.2% 53.3% 59.0% 59.0%

Total 46.6% 43.4% 47.7% 48.5%

Source: City District Government, Multan URL: www.multan.gov.pk

It is shown in table 2.11 that during 2006, the rate of literacy was 47.7 percent in Punjab. There has been found a significant dichotomy among female and male literacy rates. The literacy rate for males is 59 percent while it is 35 percent for females. So there is gender discrimination in literacy rate.

Table: 2.12 Number of Public Sector Education Institutions

Number Enrollment Teachers

Govt. Primary Schools Total Boys Girls Total Boys Girls Total Boys Girls

1253 478 775 176180 83550 92630 4166 1921 2245

Govt. Mosque Schools 212 ------12500 ------217 ------

Govt. Middle Schools 195 88 107 71625 32832 38793 2111 957 1154

Govt. High Schools 138 101 37 117205 75298 41907 3130 2137 993

Degree Colleges for Arts and Science Subjects (Punjab, Federal & Private 26 16 10 28060 12436 15624 893 543 350

Source: Punjab Development Statistics 2009

If we compare the literacy ratio of females from 1998 to 2006 for overall Punjab then we see that it is at a standstill. A little increase of less than 2 percent in literacy rate of males is however witnessed during the period of 1998 to 2006. The overall rate of literacy has been increased from 43.4 percent to 48.5 percent for the same period.

2.8.4 Education Budget of District Multan

The total budget outlay for Multan district in 2009/10 was Rs 7264.6 million. Of this, Rs 4514 million was reserved for non-development expenditures, and Rs 2750.5 million was reserved for development expenditures. So the budget for development is 37 percent of the total district’s budget.

Table: 2.13 Share of Education in the Annual Development Plan of City District Multan, 2008/09

Revised Allocation 2008/09 (Million Rs.) Estimated Estimated Budget Budget 2007/08 2007/08 Ongoing New Total (Million Rs.) (Million Rs.)

Literacy 1.00 0.00 0.00 4.00 4.00

Education 249.70 129.70 67.40 70.00 137.00

Total (Literacy+ 250.70 129.70 67.40 74.00 141.00 Education)

Allocation for Literacy + Education 14% 11% 12% 10% 10.80%

(% of total)

Total Budget for all sectors 1772 1183 558 742 1300 (Excluding CCB)

Source: City District Government, Annual Development Program 2008/09, Multan.

In 2008/09, the share of education and literacy in development budget was 10.8 percent. In 2011- 12, the total budget for development of Multan district excluding CCBs was Rs 1300 million but only Rs 141 million was reserved for the educational development projects.

Figure: 2.1 Allocation of Annual Budget for Multan City District 2009/10

Source: Punjab Development Statistics 2009

Table: 2.14 District Education Budget for Schools FY 2009-10 to 2011-12 (Rs.Millions)

Allocation Utilization

Financial Year Development Current Total Development Current Total

115.000 270.686 385.686 2009-10 150.0 274.25 424.25 (77%) (99%) (91%)

269.159 443.892 174.733 2010-11 348.416 374.024 722.44 (72%) (61%) (50%)

2011-12 196.671 407.850 604.521 ------

Source: City District Government, Multan.

2.9 Health Sector of District Multan

2.9.1 Health Institutions in District Multan

Health institutes of Multan district are shown in Table 2.15 and 2.16. There are 16 public hospitals, 77 BHUs, 17 city medical centers, 2 dental clinics, 8 rural health centers, 11 dispensaries. The total number of private hospitals in the City district area of Multan is 71. The main inferences of health institutions in Multan district are following:

i. Hospitals

There are sixteen public hospitals in Multan, two in the area of Sher Shah Town (Nishtar Hospital and Civil Hospital) and one is in the area of Bosan Town (Red Crescent Female Hospital). Nishtar Hospital is the largest hospital in Pakistan and provides undergraduate and postgraduate medical education in the area. Children Complex hospital and Multan Institute of Cardiology are specialized health institutions. Railway, Police and Social Welfare Hospitals are also working in City are of Multan. Jalalpur Pirwala Town and Shujaabad Town have the facility of Tehsil Headquarter Hospitals in each.

ii. City Medical Centers:

Table: 2.15 Public Health Institutions in District Multan (MDA Area)

Number of Health Institutions

Name of City Dental Town Hospitals Medical BHUs RHCs Dispensaries Others Clinics Centers

Bosan 3 1 3 0 0 0 0 Town

Sher Shah 5 7 3 0 0 0 0 Town

Shah Rukne- 3 5 2 0 2 2 0 Alam Town

Musa Pak

Shaheed 3 4 2 0 0 0 0 Town

Total 14 17 10 0 2 2 0

Source: NESPAK - May 2012

There are total seventeen Medical Centers in the city district area, and they fall under the management of Municipal Corporation. In these medical centers, two of them are women medical centers are located in the Union Council No.6 of Lahori Gate and Raheemabad. Four of the city medical centers are located in the area of Musa Pak Shaheed Town, seven are in Shah Rukne Alam Town, one in Bosan Town and five in Sher Shah Town. All these City Medical Centers are located in MDA jurisdiction.

iii. Rural Health Centers (RHCs)

In Pakistan, the Rural Health Center (RHC) is provided by the government for the population of ten thousand persons in the rural area. The rural health centers are normally built in small towns or big villages. There are 8 Rural Health Centers in Multan district in which Shershah Town,

Bosan Town and Shujabad Town have two Rural Health Centers in each. One RHC is located in Jalalpur Pirwala and one in Musa Pak (Shaheed) Town. There is no rural health center in MDA jurisdiction.

iv. Basic Health Units (BHUs)

The government of Pakistan has a policy to set at least one Basic Health Unit (BHU) in every union council. There are total seventy seven BHUs in city district. There are sixteen BHUs in Jalalpur Pirwala and nine in Sher Shah Town. Ten BHUs lie in the jurisdiction of MDA.

Table: 2.16 Public Health Institutions in District Multan (Excluding MDA Area)

Number of Health Institutions

Name of City Dental Town Hospitals Medical BHUs RHCs Dispensaries Others Clinics Centers

Bosan 0 0 12 2 0 0 1 Town

Sher Shah 0 0 6 2 0 0 Town

Shah Rukne- 0 0 10 0 0 0 1 Alam Town

Musa Pak

Shaheed 0 0 10 1 0 1 3 Town

Jalalpur

Pirwala 1 0 16 1 0 4 1 Town

Shujaabad 1 0 13 2 0 4 1 Town

Total 2 0 67 8 0 9 7

Source: NESPAK - May 2012

v. Dispensaries:

In Multan district, there are eleven government dispensaries. There are four dispensaries in Jalalpur Pirwala Town and Shujabad Town in each. Shah Rukne Alam Town has two dispensaries and one is located in the Musa Pak Shaheed Town. Both Sher Shah Town and Bosan Town has no dispensary. vi. Others

Two public dental clinics are working in Shah Rukne Alam Town. There are four Municipal Corporation Health Centers (MCHs) (three are working in Musa Pak Shaheed Town and one in Shah Rukne Alam Town) and three Government Rural Dispensaries (GRDs) (one each in Bosan Town, Jalapur Pirwala Town and Shujabad Town). Sher Shah Town and Shah Rukne Alam Town have Unani Shifakhanan in each.

Table: 2.17 Private Health Institutions in Multan City District Area

Name & Address Name & Address of Sr.No Name & Address of Sr.No. Sr.No. of Hospital Hospital . Hospital

National Hospital, Asif Saeed Cheri Fatima Hospital 1 Masoom Shah 25 49 Table Trust, Bosan Nishter Road, Multan Road, Multan Road, Multan

Al-Kareem Rahman Medical Faiz Clinic, Gulgasht Medical Center 2 26 Centre, Nishtar Road, 50 Colony, Gardezi Khanewal Road, Multan Market, Multan Multan

Kehkashan Clinic, Family Hospital Fatima Maternity 3 Masoom Shah 27 Kumharanwala, 51 Clinic, Gulgasht Road, Multan Multan Colony, Multan

Talal Hospital, 40 Fatima Medical Children Hospital, 4 Feet Road, 28 Center, Rasheedabad 52 Gardezi Market, Smejabad, Multan Multan Multan

Taiba Maternity Effat Surgical Oral Dental Care, 5 Home, 100 Feet 29 Hospital & Maternity 53 Bosan Road, Multan Road, Multan Home, Multan

Navera Hospital, Jafferi Hospital, Jinnah Poly Clinic, 6 Piraan Ghayeb 30 Piraan Ghayeb Road, 54 Nawabpur Road, Road, New Multan Multan Multan

Sameen Zafar Al-Khaliq Patients Maternity Home, Naeem Clinic, Near 7 31 Care Hospital, Nishter 55 Al-Sana Hotel, Multan Shah Rukn-e-Alam Road, Multan Colony, Multan

Life Care Medical Siyal Medical Centre and Maternity Jillani Hospital, Musa 8 Center, Katchary 32 56 Home, Near Chowk Pak Shaheed, Multan Road, Multan Eidgah, Multan

Ali Hospital, Abdullah Heart Care, Marium Medical 9 Khanewal Road, 33 57 Suraj Miani Road, Center, New Multan Multan Multan

Jinnah Medical Nadeem Medical Hussain Medical 10 Center, Nishtar 34 Centre,Chowk Kadafi, 58 centre, Nishter Road, Road, Multan Multan. Multan

Khalid Bin Waleed Al-Shifa Maternity Khursheed Rafiq 11 Hospital, Nishtar 35 Home,Masum Shah 59 Hospital,Khanewal Road, Multan Road, Multan Road, Multan

Al Huda Medical Professors Hospital, Khan Diagnostic 12 Centre, Nishtar 36 Near Nishtar Hospital, 60 Center, Gulgasht Road, Multan Multan Colony, Multan

Anum Hospital, Nafees Medicare , Mohsin Hospital, 13 Masum Shah 37 Gulgasht Colony, 61 Chowk Shah Abbas, Road, Multan Multan Vehari Road, Multan

Prime Maternity and Al-Hafiz Clinic, Fatima Medical Surgical Clinic, 14 Masum Shah 38 62 Center, Sayal Hotel Gulgasht Colony, Road, Multan Road, Multan Multan

Shan Medical Marie-Stopes Clinic, Amir Medicare, Lodhi 15 Center, New 39 63 Bosan Road, Multan Colony, Multan Multan

New Rehmat Gulgasht Hospital, Azeez Hospital, 16 Hospital,Khanewal 40 Gulgasht Colony, 64 Chowk Azeez Hotel, Road, Multan Bosan Road, Multan Multan

Aisha Clinic and Punjab Hospital, Khurshid Rafique Maternity 17 Masum Shah 41 Hospital, Khanewal 65 Home,Chungi No.14, Road, Multan Road, Multan Multan

Ashraf Naseer Gulgasht Dental Medical Centre, Hospital, Chungi Misali Hospital, Suraj 18 42 66 Shah Rukne Alam No.6 Gulgasht Miani, Multan Colony, Multan Colony, Multan

Halima Hospital Razia Iqbal Medical Jinnah Poly Clinic, 19 Complex, Nishter 43 Center, Shalimar 67 Nawab-pur Road, Road, Multan Colony, Multan Multan

Medi Care Ali Clinic,Gulgasht Zam Zam Hospital, 20 Hospital, Abdali 44 Colony, Gardezi 68 Shah Rukne Alam Road, Multan Market, Multan Colony, Multan

New Al-Rahman Skin & Gynae Clinic, Fazl-ur-Rahman Hospital, Opposite 21 45 Gulgasht Colony, 69 Hospital, Musa Pak Sindbad Hotel, Multan Shaheed, Multan Multan

Hussain Medical Dar Dental Inn, Near, Rahman Hospital, 22 Centre, Nishtar 46 Gulgasht Colony, 70 Samejabad, Multan Road, Multan Multan

Aysha Clinic & Gulzar Maternity Yousaf Medical Maternity, Shah 23 47 Hospital, Bosan Road, 71 Centre, Nishtar Road, Rukne Alam Multan Multan Colony, Multan

City Hospital, Near Life Poly Clinic, Asif Clinic, Chungi Nishat Boys High 24 Nishtar Road, 48 72 No.6, Multan. School, Bosan Road, Multan Multan

Source: Integrated Master Plan for Multan, Multan Development Authority.

2.9.2 Number of Beds in health institutions of district Multan

There are 2716 beds in the hospitals of City District Multan including hospitals of Railway, Police and Social Welfare Department. According to health standards, there must be two beds for one thousand populations. Unfortunately, the number of beds in hospitals of district Multan is too much below the health standards. In city area of Multan, the number of beds in hospitals has been increased from 1812 beds in 1987 to 2546 beds in 2011 (Table 2.19). In Multan District, the total number of beds has been increased to 40 percent from the last 24 years. On the other side, the increase in population has been observed about 42 percent. So there is a reduction in the number of beds per 1000 population from 1.51 in 1987 to 1.50 in 2011.

Table: 2.18 Number of Beds in Health Institutions of Multan District

Number of Beds in:

Institutions City Urban Area Rural Area District

Government Hospitals 1700

Pakistan Army (CMH 250 Multan)

Hospitals of Police, Railway, WAPDA 450 and Social Welfare Departments

Dispensaries 24 2,546 170 Basic Health Units 152

Rural Health Centers 140

Total Beds 2716

Population (2011) 4,060,000 1,704,000 2,356,000

Hospital Beds per 0.67 1.50 0.07 1000 People

Source: Punjab Development Statistics, 2011.

Table: 2.19 Number of Beds in Urban Area of Multan: 2011 Vs 1987

Number of Beds in:

2011 1987

Population 2546 1812

Total Beds 1704000 1200113

Beds per 1000 Persons 1.50 1.51

Source: Punjab Development Statistics, 2011, Page 288

2.9.3 Provision of Rural Health Centers (RHCs) in Long Term Plan (2013-2028)

Table: 2.20 Provision of Rural Health Centers in the Long-Term Plan

Towns Existing Current No. Total Proposed Locations RHCs Backlog Required Required (2008) for 2013- 2028

Bosan 2 1 1 2 Upgradation of BHUs to Town RHCs in Punj Koha (UC 62) and Bosan Town (UC 64)

Sher Shah 0 1 1 2 Upgradation of BHUs to Town RHCs in Khokhar (UC 92) and Lar (UC 95)

Shah 0 3 1 3 Upgradation of BHUs to Rukn-e- RHCs in Kotla Maharan (UC 73, Domra (UC 72) and Alam Loother (UC 74) Town

Musa Pak 1 2 1 3 Upgradation of BHUs to RHCs in Basti Maluk (UC Shaheed 88), Jhok Lashkaran (UC 78) Town and Khanpur Maral (UC 81)

Jalalpur 0 5 2 7 Instead of 7, only 5 need to Pirwala be provided as follows:- Upgradation of BHUs to RHCs in Bahadarpur (UC

126), Lalwah (UC123) and Jahanpur (UC120), Provide new RHCs at Karam Ali Wala (UC 122) and Bait Kaitch

Shujabad 2 3 2 5 New RHC at Rasulpur (UC Town 105), Upgrade BHUs to RHCs in Thath Ghalwan (UC 107), Raja Ram (UC 103), Gardezpur (UC 110) and Ponta (UC 97)

Source:NESPAK - May 2013

Proposals for Rural Health Centers (RHCs) in City District Area during long term plan of 2013- 2028 are given in Table 2.20 and 2.21. In Table 2.21, there are given the number of RHCs for the increased population during the long term plan for the year 2013-2028.

Table: 2.21 Rural Health Centers (RHCs) required for Additional Population (2013-2028)

Increase in Rural Population No. of RHC required for Town (2013-2028) (2013-2028)

Shah Rukne Alm Town 127,457 1

Bosan Town 123,967 1

Musa Pak Shaheed Town 128,386 1

Shujabad 206,268 2

Sher Shah Town 123,384 1

Jalapur Pirwala 178,317 2

Total City District 887,780 8

Source: NESPAK - May 2013

2.10 Conclusion

75 because these countries have been able to decrease the inconsistencies in the formation of human capital and hence achieved pro-poor economic growth (Deininger and Squire, 1998 and Klasen, 2002). In case of Pakistan it was observed that as a whole our society does not realize the significance of education. In Pakistan, from the year 1995-96, the five-year plans were abolished and yearly roll on planning was started as an alternative measure. But this new strategy also could not improve our educational targets properly. So it is a need to properly focus on education sector by increasing its budget.

Similarly different health policies and five-year plans were discussed in this chapter in detail. The purpose of each health policy and five year plan is to provide preventive and affective access to the health care services. Public expenditures on health sector were increased continuously in every next five year plan or health policy. Many health related targets were achieved during these health policies and five year plans but some targets could not achieved even we cannot achieve the targets of millennium development goals, therefore, budget allocated for health sector should be increased to improve its performance.

Different health and education policies discussed in this chapter show the intention of government about the reforms in these sectors. These reforms were spelled out properly, but the implementation of these policies had never been matched with the fine words of these policies. Allocation of funds was not enough for these policies. The assessment of all plans and policies show that every plan has acknowledged the importance of education sector in Pakistan, but if we talk about the allocation of funds for education sector, then it does not show any kind of improvement.

This chapter also provided a detail profile of Multan district including its geography, area, neighborhood, population, linguistic distribution etc. it also provides a detailed picture of education situation in Multan like number of educational institutions, enrollment rate, literacy rate etc and in case of health sector it provides detailed description of number of hospitals, beds etc. This snapshot of Multan district is very important as this district is selected as Sample area and data will be used for micro analysis of returns of investment to education and health in Pakistan.

Chapter 3 LITERATURE REVIEW

3.1 Introduction

It is a misconception that economics of education or economics of human capital is recently developed by Becker and Shultz in 1950s and 1960s. But actually, from more than 300 years, the theory of human capital existed in economic and statistical literature. In 1976, Sir William Petty had made the first attempt to calculate the nation’s human capital stock. After a century, in 1976, Adam Smith, a founder of Economics, had written a book “An Inquiry into the Nature and Causes of the Wealth of Nations” in which he defined the role of human capital. This chapter provides a deep insight into the history to evaluate the role of education and health in improving human capital and their earnings.

3.2 Human Capital and History

In current years, researchers have done a great effort in quantifying and developing the idea of human capital and its application with the combination of investment in human capital formation for the activities like education, skills, migration, on-the-job-training and health care. The concept of human capital is by no means new. I start with an endeavor to disperse the confusion that the hypothesis of human capital is the completely new field in economics. Numerous young students really believe that the economic aspect of human capital has grown quite recently, start in the 1950s or 1960s. This impression has emerged most likely due to the work of two economists who, during the 1950s give the idea of human capital on the map, noticeable for anyone to see. They really resuscitated a respected idea, demonstrated its hugeness for the clarification of currently observed phenomenon, and their introduction was compelling and rigorous. These two economists deserve our regard and adoration. Luckily, our profession has not neglected to perceive their accomplishment. They are Gary Becker and Theodore Schultz both from the University of Chicago. Their work in this field and research has initiated numerous different economists to enter this field and develop it further. Schultz connected the idea of human capital to educational economics, especially to a clarification of the expansion in the efficiency of human resources. He likewise inspected the connection between human capital and

77 economic growth. Gary Becker occupied with more specialized research in statistical and mathematical economics. He measured the rates of returns to investment which people have made in their own particular efficiency and skills in self-improvement mainly through training and schooling.

The principal estimation of the human capital stock was made in 1676 by Sir William Petty for his Political Arithmetic which was published in 1690. Petty did not utilize his estimation in the help of any substantive theories or regarding any hypothetical model for the induction of causal associations. According to him ‘labor is the father of wealth”. It should consequently be incorporated into any estimate of value on laborers. He utilized the idea of human capital in endeavors to show the power of England, the economic impacts of migration, the monetary value of human life wrecked in war, and the economic misfortune to a country occurring due to deaths (Hull, 1899). Petty evaluated the supply of human capital by promoting the wage bill at the market rate of interest; the wage bill he dictated by deducting property wage from the national wage. In spite of this constraint, his methodology gives a very close estimation for deciding the capital estimation of a country.

Adam Smith published his book ‘Inquiry into the Nature and Causes of the Wealth of Nations in 1776. He was clear about the human capital role. During his discussion about the stock of economy, he remarked on the most preconceptions of people that stock of capital is always like some physical good such as machines, factories or tools. He introduced the concept of human capital by suggesting an analogy between machines and man:

"...When any costly machine is raised, the unprecedented work to be performed by it before it is exhausted, it must be expected, will supplant the capital laid out upon it, with in any event the conventional benefits. A man taught to the detriment of much labor and time to any of those businesses which require uncommon aptitude and ability might be contrasted with one of those costly machines. The work which he figures out how to perform, it must be expected, far beyond the typical wages of common labor, will replace to him the entire cost of his education, with in any event the standard benefits of a similarly important capital. It must do this, as well, in a sensible time, respect being had to the exceptionally questionable term of human life, in an indistinguishable way from to the more certain duration of the machine" (Smith 1776).

According to Smith education is one of the effective ways of increasing the human productivity. He emphasized the development of skills through training and education. According to him the growth of an individual requires the employment of economic resources like machines. It would be misjudgment that we consider only the economic value of machines and not that of people in the capital stock. The total national income of a nation is the sum of all personal and material, i.e. of all factors of production. On the other hand, J. Mill (1909) opposed the idea of Smith (1776) and advocates that it is not possible to aggregate the personal and material goods as the value of former can be estimated by the ability to satisfy the needs of people. Equally, the wealth of nation consists of material goods which exist with the reference of individuals.

In 1883, a German statistician Ernst Engel in his book “Cost Value of Human Beings” treated the investment expenditures made for man as a productive factor. He was mainly concerned with the cost of food which was invested in the growing child. He followed the approach of Petty and modified it to allow for the limited number of years an individual is employed. He concluded that a yield value of a certain individual could not be estimated by simple way. It was only estimated by their monetary value to the society. Jean Baptiste Say (1821) asserted that abilities and skills tend to increase the productivity of workers and they should be treated as capital. This was also the view of William Roscher (1878), John Stuart Mill (1909), Henry Sidgwick (1901) and Walter Bagehot (1953).

According to Friedrich List (1928) abilities and skills of people which are an inheritance from past labor and self-restraint is considered as an important component of capital stock. He asserted that in both distribution and production the contribution of human capital to output should be considered. All that portion of the production of industry which is intrinsic to man and applicable to his support should be considered as the part of national capital (McCulloch, 1870). There is a close relationship between human and conventional capital. An investment in people brings the rate of return which is consistent with the other investments while during the probable life of an individual, a normal rate of return is estimated by the value of market interest rate (McCulloch, 1870).

Huebner (1914) evaluated the value of the stock of human capital in the USA, promoted by market rate of interest, and taking into consideration death rates as per a mortality table-to be six to eight times the estimation of the value of the conventional capital stock of a country. Eidward

A. Woods and Clarence B. Metzger (1927) utilized five systems to acquire five distinct evaluations of the human capital stock in the USA during 1920. They did this to demonstrate the expansive monetary value and significance of the country's population and to awaken the sluggish public of a country by speaking to its material enthusiasm to the necessities of conserving human life.

The analytical system of human capital has been utilized in the past for similar means for which it is now being utilized, to show the economic productivity of health investment, human migration, prevention of premature death and education. A lot of research appeared during the first quarter of the twentieth century in which the economists used the analytical framework of human capital to endeavor the monetary losses due preventable death and illness (Fisher, 1908; Forsyth, 1915; Crum, 1919; Fisk, 1921). Their theory was that death and illness are main cause of loss in wealth of human. To estimate the savings, Irving Fisher (1907) recommended that Farr’s capitalized net earning approach should be used to calculate the value of human being. He calculated the monetary value of an average person of USA by adjusting the Farr’s estimates to correct for the earnings of higher average. He also used the percentage of preventability and distribution of death to calculate the average capital value of lives which were sacrificed by preventable deaths during 1907.

The Chicago School (Mincer, 1958; Schultz, 1961 and Becker, 1964) during 20th century, concentrated on human capital as a factor for the growth of salary and also emphasized on that elements which contributed to human capital formation and training. The authors found analytical relationship between earnings function and human capital through the year of schooling and years of professional experience. Besides the Chicago School, some further theories on human capital were also finalized to estimate the impact on economic growth at the macroeconomic level. Various economists (Fabricant, 1954; Abramovitz, 1956; Solow, 1956; Solow et al. 1961; Schmookler, 1966; Denison, 1980; Benhabid and Spiegel, 1994) disaggregated the factors which had effects on the output of economic systems and develops a relationship between input changes and changes in output of the main production factors above all physical goods and labor.

Many economists have suggested that maintenance cost of human capital stock and production should be included in the statement of income in the form of investment and not as a cost.

Expenditures incurred for training, education and health, etc. should be subtracted from the total cost. The qualities of human capital cannot be ignored and should be included in national accounts in order to measure the economic development. The production capacity of tangible assets runs out with the passage of time while the capacity of human assets, e.g. professional abilities, initiative and knowledge can be transferred from one generation to the next through knowledge and skills (Lenti, 1967).

Along with physical capital social and psychological capital also plays a significant role to improve human capital. Social capital is simply a set of informal norms and rules shared among members of a group that permits cooperation among them. It is simply a connection among individuals and with the help of productive links a person can learn many skills and can enhance his own capacity. Social capital is a shared knowledge, understanding, norms, rules and expectations about pattern of interactions that groups of individuals bring to a recurrent activity (Ostrom, 2000). Psychological and moral capital is also very important to improve human capital because “who I am” is as important as “What I know” thus psychological support improves the confidence, hope and optimism of an individual and it can improve individual as well as organization performance (Luthans et al, 2003).

3.3 Returns of Investment to Education

The classical theories of human capital presented by Becker (1962) and Mincer (1974) considered that education and training are two main sources of accumulation of human capital and they have a positive and direct effect on the life time earning of a person. The coefficient of years of school, in Mincerian earning function, indicated the returns to education, i.e., with an additional year of schooling how much increase in earning takes place.

According to Shultz (1961), education and training lead to knowledge, high productivity of labor and modern production techniques which further lead to technical development. It also increases productivity by providing with necessary skills and knowledge and by molding the behavior of labor. The starting point of the education theories is the reality that educated worker can earn more than the uneducated one. Hamdani (1977) conducted the research about Pakistan and calculated social rates of return using the IRR technique. This study was concerned about the male workers of Rawalpindi district.

Weisbrod (1962) argued that the short benefits of education are those benefits which are realized at the time of education was being received. On the other hand, the long run benefits of education are those when the formal education has been completed. Benefits of education to mothers are realized at the time of education was being obtained like in terms of benefits to neighbors, role of child-care in schools, and in keeping children off the street. The benefits of education in terms of subsequent employment as well as benefits of future children were realized on later stage. He had also found that that the benefits of education occur not only at various places but also at various times. The benefits of education do not only accrue in the school district which financed the education of children or to the people of the specific area but its benefits are beyond the limit.

Hansen (1963) evaluated the money rate of return from schooling by ignoring other returns associated with schooling for simplification by taking most of the data from 1950 census of population. He considered two main cost variants that were total resource cost and private resource cost. Overall the rate of return from private resource investment was higher than total resource investment. Basically, the marginal rate of return increased sharply till the completion of elementary schooling, and then fell dramatically in the beginning of high school, then again fell in the beginning two years of college and increased thereafter. Thus there must be overinvestment in schooling as private returns to investment exceed the total returns of investment.

Johnson and Stafford (1973) explored the importance of investment in the quality of education along with the quantity of education. For this purpose, they collected data from survey research center of the University of Michigan in 1965. Along with conventional variables related to individual characteristics they collected data on the state in which the respondents grew up. For measuring the effect of quality of education they considered expenditure per pupil that was adjusted for prices in 1964. The inclusion of quality variable does not alter the effect of years of schooling on earning but the quality of education itself had a significant effect on earnings of individuals. They found that 1.4 percent of the variation in earnings was due to average per pupil expenditure. The marginal rate of return for quality of school were higher as compared to rate of returns for schooling but the marginal returns falls at higher level of the quality. The results showed an underinvestment in the quality of education.

Ziderman (1973) evaluated social and the private economic returns to investment in education in Britain by using an earning follow up sample survey of 1966 census. On the average, the private rate of returns to education was higher than social returns. The effect of taxes in reducing private rate of return was very low. The inclusion of ability adjustment does not change the order of magnitude of returns while the exclusion of teachers affects the rate of return by 1 percentage point only in the case of the doctorate. Returns to the general certificate of education at A-level were low while returns were higher for the higher national certificate (H.N.C) with and without professional qualification due to low foregone earnings. In case of social rate of returns which were lower for graduate qualifications and very high for H.N.C with and without professional qualification. The effect of excluding teachers was significant only at the doctorate level.

Belanger and Lavallee (1980) determined the rate of return to private investment in four educational programs (nursing, computer science, social work and nutrition). It was found that many graduates who hold bachelor degree and those who hold vocational degree performed exactly the same job but received different monetary rewards. The benefits of individuals in monetary terms were calculated with the help of cost-benefit analysis. It was also found that more economic returns in B.D education in social work and computer science should be strong encouragement for the students of secondary level to follow the route of bachelor’s degree. The near equality between the rates of return for both C.C.V.D and B.D nurses were strikingly conclusive.

Guisinger et al. (1984) by utilizing data from 1000 households in Rawalpindi, estimated the rate of returns to incremental investment in education and its affect on the level of employment. They followed the Mincerian earning function and explored that earning power was acquired by schooling and experience. They concluded that the rate of return to public investment in schooling was low and it increased with increase in the level of education. Moreover, they argued that main cause of earning differences was the sector of employment.

Khan et al (1985) computed private rate of returns of education at different levels through earning function. They found that the rate of returns changed positively with different level of education. They found that in Pakistan, the private education returns were found to be lower than in other under developed countries. It was also found a significant and positive association between the family background of individual and his earnings.

Haque (1992) estimated rates of return to schooling in Rawalpindi, Pakistan by using Becker- Mincer framework. He explored how the rate of return differs in different work categories. He used the years of schooling as compared to level of schooling used by Hamdani (1977). By using standard approach, he compared the results with other countries. Data was collected from the survey of households in Rawalpindi (Pakistan). He concluded that human capital model had better explained the earning behavior of the employed respondents than for the self-employed, while the returns to schooling were higher for employed persons as compared to self-employed persons.

Psachoropoulos (1994) described the uses and pitfalls of elaborated earning function and extended earning function method for calculating the returns to investment in education. He discussed a global update and various controversies regarding the estimation of returns. According to full method, the returns to investment in primary education were more than the secondary and higher-secondary education in various areas of the world and private returns were more than the social returns to education because the government gave subsidy on education and thus subsidy had increased at higher level of education. Overall, the rate of education returns was higher for females as compared to males. Social and private returns to general or academic education were higher than vocational education, because the cost of vocational education was higher. Returns also vary with faculties as law, economics and engineering had higher social returns while physics, sciences and agronomy had least returns. In case of economic sectors, private sector had higher returns as compared to public sector.

Vella and Gregory (1995) applied a new technique for estimating the rate of return to education in Australia. They employed data for male school leavers which were in age group of 15 to 26 years from Australian Longitudinal Survey. The main purpose of the study was to find the effect of education on under-achievement and over-achievement on earnings of individuals, and how job and personal characteristics of individuals affect income at different levels of income. It was found that in all educational levels over-achievement of education will decrease the level of earnings relative to the level of average earnings. It was also found that job and personal characteristics of an individual also affect the earnings substantially and differently at each level of education.

Arias and McMahon (1999) developed the dynamic rates of expected returns to education for both high school and college level education for males and females for the period of 1967-1995. They estimated the effects of changes on annual basis in real earnings and institutional costs within each age group. It was found that standard rates of return underestimated the value of returns of education at college level in the year 1995 by 5 percentage points for female graduates of college and 3 percentage points for males. It was shown that when the trend of net earnings was declined then conventional cross-sectional static rate of returns tend to overstate the actual rate of return, and when the trend of net earnings was increasing then conventional cross- sectional static rates of return tend to understate.

Nasir and Ali (1999) updated the work of previous studies on returns to education by addressing the role of private schools in returns to education. This issue was not addressed earlier due to the lack of data. An extended Mincerian type earning function was utilized which control for human capital, personal characteristics, occupation and other characteristics. For this purpose he collected data from Pakistan integrated household survey and adopted OLS technique for analysis. He concluded that private schools provide quality education and hence produce more productive workers. He estimated that workers had received 30 percent higher returns if they attended private schools in the past. Moreover, he concluded that private school attendance reduced discrimination against female workers in labor market.

Idrus and Cameron (2000) examined with the help of primary data collected from rural area Rantau, Negeri of Malaysia that whether there was any difference in the earnings of employed and self-employed persons by ignoring their personal considerations that can affect their earnings. By using Mincer earning function approach, they concluded that there was no difference in their earnings because there was no significant difference in these two sectors. Moreover, the private rate of return increased from 12.7 percent for primary education to 14.2 percent for secondary education and then showed a slight decline at the university level because of the high cost of university education. Therefore, government should focus on secondary level of education because this level showed higher returns, therefore, it will be helpful in reducing poverty in rural Malay.

Siphambe (2000) evaluated the private rate of returns to education in Botswana by employing a modified form of Mincerian human capital model (1974). He explored that earnings were

85 increased with the higher level of education. Female workers on the average earned less than male with the same level of education but this inequality fell gradually at higher level of education. Moreover, he found that workers either male or female in public sector earned more wages than their counterparts working in the private sectors because public sector can afford to pay more and workers in this sector has more human capital stock of schooling. He concluded that rates of return to private education increased with higher level of education.

Nasir and Nazli (2000) estimated the impact of different levels of education on earnings and found that in Pakistan higher level of education was linked with more earnings. They examined the role of education, school quality; literacy and numeracy skills and technical training on the people’s earnings by taking time series data from Pakistan integrated household survey (PIHS) during the year 1995-96. They explored that for each more year of education brings about 7 percent economic returns for wage earners. Similarly, the effects of numeracy skills and literacy are also found significant. Moreover, technical training and private schools also showed significant results.

Vila (2000) established causal links between education of people and specific non-monetary benefits of education (NMBs). He argued that due to positive impacts on utility, non-monetary benefits of education have attracted many economists. These non-monetary benefits cannot be captured by traditional measures of economics. There are many positive impacts of education on the lives of people in the form of more wages or lower costs and these kinds of effects have both non-monetary and monetary components. Some of these effects interacted by leading to longitudinal, overlapping investment in education and cumulative benefits. He concluded that the decisions about investment in education should be based on the complete evaluation of benefits for all the individuals and their family members as well as for the whole society.

Klazar, et al. (2001) evaluated the returns in the Czech Republic by using data on the annual sample research on wage differential in the Czech Republic. They used the method of internal rate of return for exploring the returns. The difference in earnings for both men and women decreased with the higher level of education. Moreover, they concluded that rate of return for women were less than men at the university level of education. In case of the profession, returns were very low for the people employed by the state in public service.

Psacharopoulos and Patrinos (2004) reviewed the research of international level about the returns of investment in education and found that the rate of average return to the additional year of education is 10 percent. The low and middle-income countries have highest returns of education. They found that if the supply of education increases then the returns to education decreases. The results found increasing trend in private returns to higher education. They also found that the tendency of returns to education was higher for women than men in under-developed countries which show that females have a lower base level of education than their male counterparts.

Khan and Toor (2003) examined different trends in the marginal rate of returns for different levels of education for the paid employees and analyzed that how rewards for additional investment at the specific level of education have changed over time. They followed the Mincerian earning function by collecting data from Pakistan integrated household survey. They concluded that each additional level of education does not result in higher level of economic returns. Female education was more rewarded as compared to male education. Moreover, higher returns were observed for the private sector that showed the importance of quality of education, while the quality of public sector education was consistently declined.

Jamal, Toor and Khan (2003) studied the private returns to education in Pakistan by estimating Mincerian earning functions using data from the household survey. They explored inter provincial, inter sectoral and inter regional variations in the earning and education. They concluded that highest private returns were associated with higher secondary education and returns to private education decreased while the returns to secondary and tertiary education increased. Moreover, the impact of English as a medium of instruction and private schooling on earnings were also found significant.

Junankar (2003) compared the social and private rates of return to education for indigenous Australians and non-indigenous Australians. They used data from the 1991 population and housing census Australia. They examined that an additional year of education leads to better nutrition and thus better health of individuals enhance their productivity and as a result increase their income. As a result, their working life or earning span increases and they can pay more taxes that is beneficial for the society. Moreover, due to higher education, the probability of crime falls thus the society’s cost of imprisonment falls thus social returns of education increases. They found some curious results that with an increase in education the rate of return to

87 education for non-indigenous individuals followed a u-shaped pattern while for indigenous Australians it was inverted U shaped. On the average, private returns to education for indigenous Australians were higher than social returns because private cost of education was lower than the social cost of education.

Daoud (2005) estimated the private returns to education in the Palestinian territory. 8 quarterly based labor force surveys were used in the analysis from the year 1999 to 2001. Israeli closure policy and its differential impacts on Palestinian male and female workers were investigated. It was found that female workers earned 14-15 percent less than male workers in 1999. This gender gap was narrowed during the year 2001 due to the rising of unemployment of male workers in Palestinian territory. By applying regression for male and female separately, it was found that returns to schooling were higher for female workers. The work for Israel premium decreased for men and increased for women. It was observed that for the year 1999, the premium of schooling for 13 to 15 years of schooling is about 10.6 percent and for the schooling of 16 or more years, it was 25 percent. It was also observed that when males were employed in Israel then their education did not matter to their wages. The marginal effect of one more year of schooling is 2.75 percent and 2.57 percent for the year 1999 and 2001 respectively.

Pastore and Verashchagina (2006) estimated the private returns to human capital for the Balarus. They took data from Belarusian Household Survey for Income and Expenditure (BHSIE) for the period 1996 to 2001. Every year about 5000 households were interviewed. They used Mincerian earnings equations and observed that the skill payoff of households was high in 1996, at about 10 percent per year and was considered stable. There was 5 percent return for one year of work experience. After controlling the sample size bias, this result was also remained maintain, despite heavy reduction in annual return to education by 20-30 percent. They suggested that socialist system had a robust mechanism to reward and stimulate human capital accumulation and was not different from the typical market economies.

Farooq (2011) estimated the factors which had determined the earnings of female and male workers in Pakistan. He adopted separate regression for male and female workers. Data were obtained from Pakistan Social and Living Standard Measurement Survey (PSLM) of 2004-05. He concluded that schooling was one of the significant determinants of earning for both male and female workers. Returns to investment in primary and middle-level education were observed

88 very low for both male and female workers as compared to the higher level of education. In professional education, the returns to female were higher than other categories. In terms of location, the urban sector workers earned higher returns.

Kimenyi, et al. (2006) evaluated private returns to education in Kenya and the effect of human capital externalities on wages of workers. They collected data from welfare monitoring survey of Kenya in 1994 and sample covered full time employed persons between ages 15 to 65 years. By using Mincerian earning function (1974), they examined the effect of different levels of education, experience, geographical region and sex on earnings of individuals. They examined the effect of quality of education on earning by using a proxy variable of pupil-trained teacher ratio for primary schools and concluded that earnings are high in districts where this ratio is low. They also explored that an increase in education of female worker has a positive and significant effect on earning of all men in the labor force but increase in education of male workers has a positive but insignificant effect on female earnings. Thus, education level of both sexes reinforced each other in urban market and wages increases for both sexes. Private returns to education increased with the level of education and returns to education in urban regions were higher than rural regions.

In general, in urban regions, the returns to college level education were observed lower than the returns to secondary and university level education but higher than the returns to secondary and university level education in rural regions. Moreover, they found that returns to education for females were relatively higher than the returns to male at national and regional level therefore high investment was needed for female education in Kenya. They examined positive effects of human capital externalities in case of increased earnings and productivity of workers.

Jaffry, Ghulam and Shah (2007) analyzed the return to education and inequalities in wage distribution for different levels of education in Pakistan. They employed data from labor force survey for the period 1990-2003 and adopted quantile regression framework developed by Angrist, et al. (2006). They used personal and household characteristics, employment and occupational characteristics to assess the educational inequalities. They concluded that pattern of equality was heterogeneous for different levels of education. Educational inequality was higher for the middle-level education, moreover, inequality between male and female was very high as compared to other categories.

Chen and Hamori (2009) examined returns to schooling for China. They used Instrumental Variable (IV) method and Ordinary Least Square (OLS) methodology in their analysis. They found that OLS estimates for returns to education were found lower in China with compare to other transition economies while estimates of instrumental variable were higher in China. It was also found that if we did not control endogeneity bias, then OLS estimates can underestimate the true value of the rate of returns. Finally, they found that the value of returns to education for men was slightly bigger than for women in OLS estimates. While through IV methodology, the rates of returns to education for women were higher than men. That difference increased after adjustment for selectivity biases.

Awan and Hussain (2007) explored the gender differentials in wage and returns to education in Pakistan by conducting two household surveys and following an extension of Becker and Mincer models. They explored large income gaps between educated and uneducated workers and concluded that it increased with the level of experience. Due to discrimination female workers had earned less because they acquire less experience than man. Moreover, education equality was low for students belonging to poor families because they attended government schools and did not have any access to private schools which have better quality.

Canton (2007) examined social returns to education by using macro growth regressions by taking data from a group of 31 developed and some middle income countries. Based on the macro Mincer model (1974), he explored that social returns to education lay somewhere between 11 percent to 15 percent, moreover, private and social returns were more or less of same size. He used different indicators for labor productivity as GDP per worker, GDP per capita and GDP per hour worked. There were little variations in returns across these productivity indicators. For balanced growth of the economy, these returns allowed adjustment in physical capital stock. The short term returns to education varied between 7 to 10 percent when capital stock was constant.

Qiu and Hudson (2010) evaluated the private returns to education in China for four years from 1989 to 2000 by using data sets of China Health and Nutrition Survey (CHNS) that covered more than ten years of Chinese economic reforms. The survey covered the urban areas of nine provinces such as Heilongjiang, Liaoning, Jiangsu, Henan, Shandong, , Hubei, Guangxi, Hunan and Guizhou. They used basic earning function by Mincer (1974) to evaluate the effect of education, experience, square of experience, gender, region and some other interaction terms on

90 earnings. Their findings suggested that returns increase with the education. Initially, the marginal returns showed a decline but increased thereafter to 7.1 percent in 2000. In all four years although men earned more than women but education has modified that earning gap in men and women, therefore, gender gap showed a decline. Returns to education were found higher for individuals in the private sector from the year 1989 to 1997 but later on, these higher returns to education were shifted to the state sector. Moreover, they explored that earning was decreased if an individual located or worked in an area away from Beijing and Shanghai.

Warunsiri and Mcnown (2010) estimated the returns to education for Thai workers who were born during 1946-67, by taking data from the national statistical office of Thailand and from labor force survey for 11986-2005. They used pseudo panel approach for estimating the returns of education because this technique was helpful for controlling the effects of unobtrusive factors like motivation and ability which creates downward bias in the estimates. Instrumental variable estimation was also applied to confirm the validity of pseudo panel approach. By using pseudo panel estimations, the overall rate of return to education was between 14 percent and 16 percent.

By disaggregating the sample on the demographic basis, they found that rate of returns to education for women was higher than for male because education helped women in finding the employment outside the low paid professions. According to pseudo panel estimates the rate of return for education was higher for those workers living in urban areas than for rural workers because workers in urban areas had more opportunities to acquire skills and make use of them. Moreover, the rate of return to education was higher for unmarried workers than for married workers because unmarried workers had greater geographical and job mobility which allowed them to earn more by getting the higher education. The results of their study concluded that investment in the education of people led to increasing in per capita income of Thailand workers.

Wei (2010) estimated the economic returns to post-school education in Australia by using data from the year 1981 to 2006. He particularly focused on the returns of investment in bachelor degree. By applying Mincerian earning function and internal rate of return method he estimated that due to investment in bachelor degree the private rates of return increased over time for male from 13.1 percent in 1981 to 19.6 percent in 2001 and then decreased to 15.3 percent in 2006 and over the same period the range for female was 18.0 percent to 17.3 percent. He concluded that post school education plays a vital role in human capital stock.

Afzal (2011) empirically estimated that academic qualification or some other factors are key determinants of earning. He evaluated private financial returns to education in Pakistan, especially in Lahore district. For this purpose, he collected data from 3358 teaching and non- teaching employees of the school, college and universities in Lahore. He found that respondent’s age, gender, experience, spouse education, family background, occupation, working hours and family status contributed significantly and positively to earning for all education levels. He estimated that private financial returns to education at the college level were higher as compared to school and universities. Men related to teaching and non-teaching employees earned more than females and this difference was higher at primary level due to male gender biased society and better working environment for male staff. Teaching staff at all educational levels earned more than non- teaching staff, this difference was higher at university level due to their active role in financial decision making. Computer literacy and SSC sector of respondent institute contributed only to the respondents belonging to the university. The study found that for an additional year of schooling, an earning of the individual was increased by 5.1 percent.

Ashraf (2011) evaluated some recent returns to investment in education by using data from the Pakistan Integrated Household Survey (PIHS) for the year 2001-02. He used the modified form of Oaxaca’s model to estimate returns. He examined the effect of characteristics of working individuals, occupational categories and industrial classifications on salaries of workers. The sample selected was biased towards male and consisted of individuals between age 16 and 65. Most of the results were consistent with the previous findings. Overall the returns to education increased with the level of education. At middle level of education, the rate of return to education were higher for the female that was 13 percent as compared to 5 percent for male, therefore, more resources should be devoted to female education. Moreover, he examined that experienced male workers could earn more as compared to experienced female worker and workers with education from English medium institutions could earn more than those institutions using national languages. It is suggested that by promoting education in Pakistan poverty situation can be controlled.

Murillo, et al. (2011) examined the impacts of educational mismatch on wage level in the Spain’s labor market from the mid-1990s to 2006. They calculated the returns to education by matching with job requirement and worker’s schooling. The results indicated that the returns to

92 education were reduced by time. It also found that return of each more year of schooling was positive but less than that of an additional year of required education. They concluded that investment on education was positively alleviating the unemployment level in Spain.

Mohapatra and Luckert (2012) estimated the distribution of educational returns for male and female in India. It was found that wages are subject to substantial uncertainty for each level of education. Multivariate analysis indicated that returns of education changed widely across individuals. The results further indicated that uncertainty in returns to education was enhanced with the higher level of education. They estimated that the level of uncertainty for returns to education for males increases from 19 percent in secondary level to 25 percent and 30 percent in higher secondary and college level respectively. In another side, the uncertainty of returns to education for females increases from 41 percent in secondary level to 49 percent and 42 percent at the higher secondary and college level respectively. Thus it was concluded that uncertainty in returns to education was higher for women than for men at each level.

Qureshi (2012) linked the gender differences in parental resource allocation regarding demand for education in case of gender differences in returns to education at all levels for Pakistan. He suggested that there was prevail under investment in Pakistan in education for the female at all levels even though returns to education for females were much higher than males. Therefore he focused on the need for parental education so that female participation may increase in labor force.

Maazouz (2013) examined the relation which existed between education and the policy of labor market. He adopted the hypothesis of labor market theory of human capital and segmentation in order to estimate the profitability of the investment in the human capital. In segmented and competitor labor market, the rise of the level of education generated a rise in the rates of unemployment, and the manufacturing units proceed to use the level of education like a quantitative index of employment. He applied different econometric methods using the available resources for different educational levels to evaluate the rate of cost on investment in human capital. He implied that teaching at primary education is an essential level where the rate of return of investment in human capital was higher.

Saruparia and Lodha (2013) evaluated the returns to investment in technical and professional education offered by JNV University, Jodhpur (India). They presented three criteria for making investment decisions that were cost benefit ratio, net present value and internal rate of return. They explored that private rate of return was higher than that of the social rate of return. Moreover, subsidies do not go waste and investment in professional and technical education was beneficial for the state also.

Schundeln and Playforth (2014) investigated that whether differences between private and social returns to education can contribute to explain the micro-macro paradox for government sector employees. The data for analysis was taken from the Indian census for the years 1961, 1971, 1981, 1991, and 2001. They hypothesized that educated people found of privately rewarding jobs especially in those sectors where social returns are low like jobs in government sector. That could help to explain high returns of education in micro level and small or negative coefficients on education growth in macro level. It was indicated that effect of the size of government sector on the effectiveness of education in promoting growth was negative, statistically significant and economically large. The micro level evidence on public and private wage differentials shows that large positive wage differentials between the private and public sectors persisted through the 1990s, with the public sector wage premium being up to 140 percent.

Sohan (2013) explored some neglected issues about returns to education in Indonesia by using more recent data from Indonesia family life survey of 2007. By using Mincer (1974) specifications he explored that rate of return to an additional year of schooling was consistent with other studies that is 10.7 percent. The returns to education have increased with level of education. The inequalities between education’s groups were not severe in Indonesia as in other developed countries. The monetary returns to self-employed workers were low as compared to paid workers, both due to person and sector specific reasons. He also considered the non- monetary returns to education in the form of happiness and explored that education was positively linked to happiness. The non-monetary returns were again lower for self-employed workers than for paid employees.

Fulford (2014) evaluated returns to education in India by using data from Indian national sample surveys. He explored that individuals with more years of education lived in households with more consumption per capita and approximately on the additional year of education leads to 4

94 percent more consumption for men with no additional consumption for the female cohort. Thus overall the consumption returns were more than wage returns because consumption was shared among the households. Average returns to education were low for both men and women but most of the highly educated Indians got more returns because of India’s growth. The low returns were due to the low quality of education and poor mathematics and reading scores. He explored that low returns for women were due to ineffective use of female workers in production and due to lack of opportunities. Therefore, a large number of female workers must be brought in the work force of India.

3.4 Returns of investment to Health

The literature reveals substantial empirical studies on education but there are few studies on the benefits of investment in health. Very few studies focused on the estimation of returns to investment in health not only in Pakistan but in foreign countries also.

According to Schultz and Tansel (1997), most studies which estimated the morbidity was belonged to high-income group countries and focused on the disabilities among elderly. Other studies focused on the effects of productivity related to nutrition in under developed countries (Behrman and Deolalikar, 1988; Deolalikar, 1988; Sahn and Alderman, 1988 and Behrman, 1993) and some other studies discussed malnutrition effects for children and child morbidity (Rosenzweig and Schultz, 1982a, 1983; Rosenzweig and Wolpin, 1988). Schultz (1984) has found the relationship between public program interventions and child mortality. Health status of adults measures height, early childhood conditions, and cumulative health and nutrition investments by parents (Martorell and Habicht, 1986; Thomas and Strauss, 1997).

Strauss and Thomas (1997) used survey data of Brazilian economy that even after controlling for education, various aspects of health such as BMI, height, protein intake and calorie intake affect the wages of women and men positively. It was found that without controlling for health, the returns to education with health controls were 45 percent smaller for the male with secondary level of education or more. Schultz (1996) found that estimated returns of wage to schooling are reduced between 10 percent and 20 percent with the combination of three more inputs related to human capital in the regression, BMI, migration and height. They found that the effects of nutrition on adult productivity and height are subject to the decreasing returns to scale.

Mushkin (1962) described the importance of investment in health and returns associated with such type of investment and various other issues linked to it. He also described the similarities and differences in investment in health and education. Investment in education and health both were investments in the same person and increase their productivity and quality of life. Both types of investments depended on each other as basic health programs depended on education in personal hygiene and similarly a person’s formal schooling was impossible without a good health. He measured the human capital formation through investment in health care and explored that labor productivity and a number of labors in the work force that increases due to better health services and expenditure. Value of this investment was observed by the increase in productive time or by the increase in the value of money. He explored that average earning of labor multiplied by the years of age added as a result of cure of the disease that will lead to enhancing economic growth.

Kiker (1968) explained briefly with the help of capital analytical framework the benefits of investment in health care. He focused on the question of which management of disease or illness should resources be devoted? With the help of cost of production method and capitalized earning approach he commented on investment decision in health. The cost of production approach was helpful to estimate the total cost of the human capital formation while capitalized earning approach find the financial value of the stream of future earnings. These approaches assumed that health care affected the individuals but these individuals can not affect the economic activity. In another comment, he explained human capital analytical framework to make health investment decisions. This approach presented a comparison of loss and benefit of health investment. One should invest in health care service if it yields higher returns in the future otherwise investment should be made in other alternatives.

Cropper (1977) described two models of investment in health where the nature of death and illness was considered as random. In the first model he considered that investment in health was made to reduce illness thus he focused only on preventive expenditures. These preventive expenditures helped to get monetary returns and avoid the disutility associated with being ill, because a healthy person can get more utility and can earn more. In the second model, he presented a tradeoff between investment in health and choice of occupation. Because these types of occupations were dangerous for health, therefore, there was a tradeoff between job safety and

96 higher wages. Thus, if the worker wants to get higher wages he sacrifices health and the probability of illness or death increases. In such a situation a worker must not shift to the healthier working environment because the effects of these pollutants are long lasting. Moreover according to human capital theory preventive investment in health was higher in beginning of one’s life and may decrease as the individual ages and recuperative expenses increases.

Zaidi (1986) explored some salient problems relating to the health sector of Pakistan. He pointed out that in Pakistan doctors are taught western books. An important outcome of this type of training and education is the process of westernization for the doctors. It is a need that doctors in Pakistan should be trained about the western diseases. Our medical education system is the main reason for the brain drain of trained doctors from Pakistan. Unemployment of doctors is another main problem of Pakistan. Due to the absence of medical infrastructure to absorb the output of new doctors, many trained doctors have ended up their profession without jobs. In Pakistan, 85 percent production of pharmaceutical companies in Pakistan is controlled by 15 multinational companies. The doctors in Pakistan are given different incentives to promote the medicines of certain companies. He recommended that it cannot be expected any reasonable improvement in the health sector of Pakistan without substantial changes in the existing structure of powers.

Duraisamy and Duraisamy (1995) examined the determinants of investment in health of children which was measured by preventive health care and child survival of rural households in Tamil Nadu. The decision of family about the health of child was made according to the framework of neoclassical common preference. The Nash-bargaining model was developed as an alternative approach to studying the behavior of family and its implications. The empirical results showed the presence of differential effects of non labor income (measured through the value of assets) of mothers and fathers. They rejected the asset pooling implication of neoclassical model.

Foster (1995) defined the household savings and human investment behavior in terms of investment in health and nutrition. The results of the analysis showed that allocation of calories should be made according to the sector in which a person is working. It is shown by Foster and Rosenzweig (1994) by using the calorie data for the same person in different time periods that a person who was working as full-time job at piece rates was allocated per day 23 percent more amount of calories than when undertaking the calories on the basis of non-labor activity. He also observed that different aspects of health and calorie intake were incompletely rewarded in the

97 agricultural labor markets. Urbanization was considered as an important determinant for the availability of infrastructure and health personnel. Development expenditures on health were only affected by the changes in CG/ CP. Changes in political regimes had also positively and significantly affected the development expenditures. However, other socioeconomic factors were not found statistically significant.

Siddiqui et al. (1995) reported that favorable socioeconomic conditions, education and particularly changes to GDP effect the allocation of resources for improving the status of the health in Pakistan. It was also found that the assumption that payment of debt effect the allocation to the social sector was also true as the relationship from DSP to allocation of health is significant.

Cremieux, et al. (1999) investigated an important relationship between health care spending and health outcomes by using homogenous province specific data from Canada in the period 1978- 1992. As a proxy variable for health outcome, they used life expectancy and infant mortality rate which were also used in most of the previous studies. They examined the effect of various economic, social, lifestyle and nutrition variables on the health outcomes in Canada. They explored that greater alcohol and tobacco consumption resulted in a negative impact on life expectancy and infant mortality rate because such type of consumption affects lungs and cause heart attack. In social factors, they examined that high density led to more stress and dangerous lives and have reduced women’s life expectancy. Families with more poverty led to greater infant mortality although life expectancy remains unaffected. Moreover, they concluded that greater fats in diet led to lower health outcomes. In case of economic factors, they explored that higher income leads to higher health outcome because 10 percent reduction in health care spending will lead to 0.5 percent increase in infant mortality rate in males and 0.4 percent in females and life expectancy lower by six months for men and three months for women.

Filmer and Pritchett (1999) explored the impact of public spending on health and non-health factors on the child (under five) and infant mortality by using a cross national data. The data about child mortality was collected from UNICEF (1992) and country level health data obtained from different country sources as ministry of health. Findings of the study explored that among many socioeconomic factors some strongly affect health outcomes. It was observed that an additional year of female education led to 10 percent lower mortality and 1 percent increase in

98 income inequality measured by Gini coefficient led to 0.5 percent increase in mortality. The coefficient of being predominantly Muslim was found very strong for child mortality, showed a 45 percent rate of child mortality. The probability of two citizens speaking two different native languages led to higher mortality. Apart from these socio economic factors, public spending on health was associated with lower under-five mortality but this association was small and significant only at 10 percent level.

Arora (2001) investigated the impact of health on the growth paths in industrialized economies over last 100 to 125 years. He observed that pace of growth was increased by 30 to 40 percent due to changes in health and due to this the slope of growth path was altered. These results were robust across five measures of health in long run and it remains unchanged when controlled for investment in physical capital. Variables related to health were positively correlated with year of schooling. However, variables related to schooling did not replicate the findings which were obtained from the measures related to health. Improvements in health thus did not merely follow the economic growth.

Audibert and Etard (2003) explored the additional economic benefits after access to health facilities or treatment of diseases by using Quasi-experimental design model and by employing data from Office du Niger, an irrigated rice growing site in central Mali. Rice, Sorghum and onion were the main crops of that area. Schistosomiasis had been a persistent disease among the people of that area which was caused by a parasite. This study compared the labor productivity and production of the families who have benefitted from a disease control health programs related with those families who have not and concluded that returns on health by improving labor productivity of family provides direct more utility to the household and hence increase the time for leisure or in some other activities as increase in the cultivation of profitable crops.

Toor and Butt (2004) examined that how government health expenditure influenced household health behavior in Pakistan. They explored that public health expenditure were linked with both preventive and curative health services for children. Thus children of high expenditure households benefited more from government health spending than the children of low expenditure households.

Issa and Ouattara (2005) observed a negative association between health expenditure and infant mortality rate by taking data from 160 countries. They also discovered a negative relationship between infant mortality rate and per capita income and female education. Moreover, they concluded that at the low level of development public expenditure on health has the strong effect on mortality rate as compared with private expenditure at high development levels.

Okunade (2005) estimated an econometric model of the demographic and socio-economic determinants of post-SAP health care spending for a cross section of the 26 African countries by using 1995 cross sectional data and by employing flexible box-cox model regression methods. The study concluded that about 74 percent of the variation in health expenditure was due to economic and other determinants. The main determinants were official development assistance, Gini income inequality index, per capita GDP, population dependency ratio, internal conflicts and percentage of births attended by trained medical staff. Income inequality reduces health expenditure while ODA increased health expenditure. The GDP is found to be the most significant determinant as it yields on the elasticity of 0.65 percent in relation to health expenditure.

Datar et al. (2006) tested that whether parents compensate or reinforce for endowments of children. Birth weight was used as a proxy for endowments and found that how birth weight among siblings effect parental investment including duration of breast feeding duration and initiation, immunizations, well-baby visits, entry age in kindergarten and preschool attendance. It was also examined whether siblings’ endowment had effect the parental investment behavior. Results of the study indicated that the children with heavy birth weights received more parental investment than their sibling having lower birth weights. They suggested that parental investment in early childhood and infancy strengthened the differences in endowments.

Luce, et al. (2006) estimated the value of investment in health care in the USA by using three different approaches. These approaches were; returns to investment on the overall health, returns to investment in Medicare health care treatment service and returns to investment for selected major treatment innovations. In the first approach they observed two health outcome variables as decrease in death or increase in life expectancy and concluded that for every dollar spent on health care resulted in decline in death rate from 1039 to 872 per 100,000 persons and life expectancy from birth increased by 3.2 years and average life expectancy for all ages increased

100 by 2.18 years between 1998 to 2000. According to the second approach, they concluded that for every additional dollar spent by Medicare the return on health associated with heart attack treatment was $1.10, for stroke it was $ 1.49, $ 51.55 for type-2 diabetes and $ 4.80 gained for breast cancer treatment. According to the third approach, for every additional dollar spent on health innovations and treatments, the returns ranged between $ 1.12 and $ 33.00. They claimed that there were some other social and environmental factors improving or worsening the health. Therefore 33 percent of gains in health were due to these factors, thus, more spending on health in the USA leads to more improved health situations.

Nixon and Ulmann (2006) explored the causal relationship between health care expenditures and health outcomes by using a panel dataset from 15 member countries of the European Union for the period of 1980-1995. They argued that there were two main approaches used by previous researchers, these were Grossman’s human capital theory and health as a production function. According to the researchers, there were several quantitative and qualitative factors which can affect health outcomes as life expectancy and under-five mortality as dependant variables. Various social and environmental factors were considered as inputs, and fixed effect economic technique was employed on panel data and found that increase in health care expenditure were significantly associated with large improvements in infant mortality, but the only marginal relation with life expectancy in the EU countries. It had added only 2.6 years to the life expectancy of males and 2.8 years to that of female and these results were consistent with most of the previous studies.

Anyanwu and Erhijakpor (2007) examined the effect of per capita total, government health expenditure and per capita income on healthy outcomes as under- five mortality rate and infant mortality rate. This study was conducted by taking data from 47 African countries for the period of 1999-2004. The main sources of data collection were world development indicators, WHO and African development bank’s database. Along with above-mentioned factors they also explored the effect of ethno linguistic fractionalization, female literacy, urbanization and number of physicians on health outcomes and find that 10 percent increase in per capita total health expenditure reduce under-five mortality rate by 21 percent and infant mortality by 22 percent. Similarly, 10 percent increase in per capita public health expenditure reduced under-five mortality by 25 percent and infant mortality by 21 percent. Ethnological fractionalization had a

101 significant positive effect on health outcomes. Female literacy mattered for health outcomes and was negatively related to them. Urbanization had significant but weak negative association with health outcomes. Moreover, they explored that HIV prevalence positively affected health outcomes.

Akram et al. (2008) determined the connection among health and economic development by taking time series data from 1972 to 2006.Infant mortality rate and life expectancy were utilized as health indicators. Cointegration combined with Error Correction methods had been utilized as a part of the investigation. The outcomes showed that openness, age dependency, population per bed, life expectancy, secondary school enrolment and mortality rate were influencing per capita GDP. However, health consumptions had no association with per capita GDP. The outcome affirmed that health variable assumes an exceptionally huge part in deciding the long run economic development. As all the indicators related to health significantly affected the long run economic development. However, findings from the Error Correction model revealed that health indicators had no significant impact on economic growth in the short run.

Lhila and Simon (2008) estimated the behaviors of parents to allocate the resources in the prenatal health of their daughters and sons in the case when they know the sex of their child in advance. Prenatal health behaviors can be considered as decisions about investment among the immigrants from China and India in the United States. They found that preferences of parents about the gender of the child may affect their decisions about abortion first and then their decisions about investment. Health investments by parents can be measured by maternal choices during the time of pregnancy, like use of alcohol and tobacco during pregnancy, number of prenatal care visits and adequacy of weight gain. It was also tested whether gender preference persists among immigrants of first-generation who were born in China and India. Evidences were found consistent with sex-selective abortion among mothers from China and India. However, knowing about the sex of child in advance did not appear to affect the prenatal health investment neither among immigrant nor all U.S. mothers.

Mirvis et al. (2008) reviewed the literature about the relationship between health and economic conditions in the United States. The evidences from national, historical and transnational studies have indicated that improved health sector has increased economic growth through various micro and the macroeconomic factors. In U.S.A, peoples with better socioeconomic conditions and

102 higher income have better health outcomes than those persons who were less affluent. It is evidence that rich countries spend more money on health services than poor countries (CMH, 2001). It was observed that a one percent increase in per capita income would further enhance one percent spending on health related services. Poor health reduced the personal savings which provided capital for investment. Citizens with poor health spend more from their income for current health related expenditures and become concerned less with needs of future.

Sendi (2008) estimated the returns of health and non-health investment with the objective function of maximum returns on health investment from available funds. It had important implications for decision-making and may suggest that incremental cost-effectiveness ratios may overestimate the cost-effectiveness of interventions when non-health investments are considered from a broader perspective. Extended for of net benefit framework was presented under the conditions of certainty. He showed that the willingness of decision maker to pay per unit of impact for the treatment program needs to be more than its incremental cost-effective ratio to be considered cost effective. It is an important implication for the decision making and may suggest that when non-health investments are considered from a broader aspect, then incremental cost- effective ratio may overestimate the cost effectiveness of interventions.

Abbas and Hiemenz (2011) explored the determinants of health care spending in Pakistan by using time series data on economic, demographic, social and political variables. They concluded that urbanization and unemployment have negative long run impact on health care expenditure and income is a strong predictor of health expenditure.

Elgazzar (2011) estimated the overall value for money for the health system mainly as indicated by an analysis of public expenditure trends from 1997 until 2008 and by the degree to which health care benefits are conferred equitably to the population. Total health expenditure in 2007 accounted for 5.2 percent of GDP, or only PPP$ 41 per capita, with nearly 70 percent sourced by direct household payments (67%). Despite recent improvements in health status, Yemen continued to lag behind countries of similar or lower income and health expenditure levels. Levels of health outcomes in Vietnam, Indonesia and the Kyrgyz Republic were 2 to 6 times better than levels found in Yemen regarding the proportion of infants with low birth weight, the prevalence of malnutrition amongst children and the rate of births delivered by skilled attendants.

103

Although health facilities are relatively evenly distributed across the population, the operational status and quality of these facilities were highly variable.

Hadley and Reschovsky (2012) explained the impact of Medicare spending on reduction of deaths and avoiding hospitalization. For this purpose, they used instrumental variable approach and take a sample of 388,690 Medicare beneficiaries for this purpose. These beneficiaries were the high cost cases to evaluate the effect of health spending. They concluded that an increase in health spending leads to many health improvements which were quantitatively meaningful and significant. Moreover, they explored that 10 percent increase in health spending led to 3.8 percent reduction in hospitalization and 8.4 percent decrease in mortality rates.

Keyeke, Sackey and Azinim (2013) examined the relationship between public spending and health status in Ghana. They used under five mortality rate as an indicator of health status. They interpreted that health sector was interlinked with socioeconomic development; therefore health sector cannot be treated in isolation. They argued that availability of physicians and health insurance are the most important determinant of health status in Ghana.

Kim and Lane (2013) analyzed the relationship between public health expenditure and health outcomes in 17 OECD countries by taking infant mortality rate and life expectancy as the proxy for health outcome by using mixed effect model. They concluded that better health services increased life expectancy at birth and reduced infant mortality rate.

Reeves (2013) found the relationship between health investment and economic growth. He estimated the fiscal multiplier with the effects of government spending. By using models of cross national fixed effects of 25 EU economies, they estimated fiscal multiplier both after and before the recession of 2008. It was found that the government spending multiplier was 1.61, but there was heterogeneity across the types of spending. The value of fiscal multiplier ranged from −9.8 for defense expenditures to 4.3 for health expenditures. This difference can be explained by different degrees of absorption of spending by the government into the domestic economy. It was found that defense expenditures were associated to significantly larger trade deficits while education and health had no effect on the trade deficit.

Corso (2014) examined different types of revenues and expenditures, health services and health outcomes that can be tracked at local and state levels of public health. They gave

104 recommendations for strengthening the ability of local and state governments to associate the expenditures with public health outcomes within and across the jurisdiction. The revenue data source for most local jurisdictions was the accounting systems that were used for the auditing and budgeting activities. Policy change at the state level and implementation strategies by local health departments are required to access the health outcomes and costs of public health activities.

3.5 Relationship between Education, Health and Economic Growth

The papers which are based on the endogenous growth theory highlighted the significance of the human capital in the growth and development process. For example, in endogenous growth model, Lucas (1988) and Romer (1990) took education stock with human capital former; with research and technology later. In these studies, human capital is treated as a positive externality on the productivity of capital and its accumulation and also effect the welfare and economic growth of the community. In short, human capital is a stock of knowledge, health, training, including creativity and other investments. On the other hand, formation of human capital refers to the process of increasing and acquiring those people who have good health, skills, experience and education that are essential for economic development.

Although human capital is multifaceted, many theories explicitly connect investment in human capital development to education and occult the other aspects, mainly stock and investment in health. However, health also plays an important role in human capital accumulation and is closely connected to education. For instance, a healthy population is easy to educate and the efficiency of people to produce human capital is also high. Inversely, an increase in education involves the enhancement of health conditions as qualified people have a more responsible behavior.

A few economists have discovered that training has critical impact on economic development. For example, Mankiw et al. (1992), Barro and Sala-I-Martin (1995) located a positive relationship between initial levels of endowments of education and subsequent rates of growth. Utilizing cointegration analysis, Maksymenko and Rabani (2008) additionally found that training had a noteworthy constructive effect on development in both South Korea and India. Hanushek and Kimbo (2000) utilized the indexes of the quality of education for 38 nations in light of

105 academic performance in arithmetic and sciences over the period of 1965-1999 and proposed a solid connection between the quality of education and increase in per capita GDP.

By taking specification bias that is generated by ignoring the international differences in the schooling quality, Dessus (1999) analyzed 83 countries for the period of 1960 to 1990. He concluded that quality of education is measured by the pupil-teacher ratio, access to education facilities and government expenditures are considered significantly associated with the economic growth.

As for health is concerned, since 1990s, there are many studies which define the effect of health on the economic growth. For instance, Fogel (1994) showed that during last two centuries, about one-third of the economic groth in Britain is due to the improvement in health and nutrition. Barro (1996) found a positive relationship between economic growth and life expectancy. Similarly, Jamison et al. (1998) identified that those areas which were more prone to malaria had grown less because prevalence of malaria in these areas had decreased the growth rate of income.

Ainsworth and Over (1996) argued that HIV/AIDS epidemic has decreased the rate of growth by reducing human capital and per capita savings by 0.33 percent in the countries of Africa. Gyimah-Brempong (1998) showed a positive association between the share of government budget allocation and economic growth to health care in the countries of Africa. After measuring the health through the probability of adult survival by age group and sex and using Granger Causality test, Mayer (2001) concluded that in Latin America, health caused economic growth. Moreover, improvements in the status of health were related to 0.8-1.5 percent increase in the level of annual income which indicated that health had a significant role in economic growth.

Schultz (1993) evaluated the returns of investment in education and health of men and women and explored that private rate of returns to investment in women’s human capital has increased as compared to the private rate of return on men’s human capital or other alternative investments. In most of the countries, there was a large increase in women’s health measured by longevity and years of schooling as compared to men’s. There was some selection bias also like the inclusion of wage earns only, but these type of bias can not affect the higher private returns to schooling for women. These higher social returns to schooling for women favored great investment in women’s education.

106

Ross and Wu (1995) explained the association between health and education in different categories by using the different database for different age groups as the first database that came from the probability survey of US household ages 18-90 in 1990 and second database from US household survey ages 20 to 64 in 1979 and then again in 1980. These datasets covered different amount of samples and concluded a positive association between health and education because highly educated persons were less likely to be unemployed, they were more productive and got more rewarding jobs, and higher income as a result health was improved. They had a great sense of control over their lives, and they were less likely to smoke and more likely to do exercise, therefore, they get a healthy lifestyle.

Schultz (1999) evaluated the situation of health and education in Africa and found that people of Africa are experiencing the low level of both factors as compared to other regions of the world. The expectancy of life at birth was the most common indicator of health but it was wrongly measured in Africa, overall the life expectancy rate in Africa was at the bottom. Moreover, HIV was also threatening life expectancy. On the average, the level of education in Africa was also very low. Women’s enrollment was greater than the man that will be helpful in future to reduce child mortality and controlling fertility. In Africa education human capital can be accumulated by increasing public expenditure on education, education of mothers and private non-earned income per adult while the increase in relative price of education reduced enrollment, similar factors were found in case of health. It was observed that on the basis of efficiency Africa overemphasizes higher education. Due to higher public spending on the higher level of education, the social returns were lower as compared to private returns. Moreover, they explored that more investment was needed in the health sector as low level of health outcomes were proved to be a big hurdle in the path of development.

Asghar et al. (2009) estimated the impact of education and socioeconomic factors on health in Pakistan. They used an ordinal logistic regression model to regress the impact of different variables on self-reported health (SRH). Including the objective health status measures for all diseases, it was not hampered the significance for the employment. Different kinds of employment status except the disability were affecting the self-reported health measures significantly. Among different socioeconomic variables, gender, schooling, economic status, occupation and provinces were considered as significant determinants of the self-reported health.

107

Moreover, the relationship between both education and health was not very sensitive to either excluding or including the variables. The health level of an individual increased with higher level of education.

Schultz (2003) examined the pattern of investment by people in human capital as adult health, child nutrition and schooling and how these investments were related to changes in the earning opportunities. This study also discussed the issues related to the measurement of returns and biased introduced in this process. They concluded that health and education were the main factors which contribute to higher labor productivity in most countries.

Groot and Brink (2007) analyzed the relationship between health and education by taking data from a large cross sectional survey for the Netherlands. The 1999 supplementary provision survey of Dutch social and cultural planning bureau. They found that health returns to education were 1.3 to 5.8 percent. They also tested causality link between health and education and found that the heterogeneity and scale of reference bias had no strong effect on the health and education relationship.

Eide and Showalter (2011) explored the causal link between health and education irrespective of many difficulties regarding it and highlighted many recent researches on this issue. They found three main causality links as school causally affected health, health causally affected schooling and time preference affected both schooling and health. With the help of literature, they found that compulsory schooling laws were helpful to reduce mortality as a health outcome. They observed that in the USA, one additional year of education lowered the probability of dying in next ten years by 3 to 6 percent. It was also helpful in reducing hospitalization. Moreover, an increase in medical information also improved health outcomes. In case of the effect of health on education, they found that low birth weight affected child’s education attainment and future earnings similar was the result of shocks in the early childhood. Thus they concluded that both health and education had strong causal links which were emerging in the recent and upcoming literature.

Eggoh, et al. (2015) explored relationship between health, education and economic growth by examining 49 African countries for the period of 1996 to 2010. They used real GDP per capita as an indicator of economic growth and public investment in education and health as explanatory

108 variables and explored that due to corruption and inefficiency of these African countries, public investment in education and health has negative impact on economic growth. Moreover they explored that these two investments are interrelated therefore they should be increased jointly and efficiently to expect positive effects.

3.6 Conclusion

As the literature showed that although great work have been done related to education and health but only a few studies focused on the rate of return to investment in these sectors in Pakistan, by using primary data. Only monetary returns are focused and more attention is given to private rate of returns to education and health while non-monetary returns and social returns are ignored. Moreover, the quality factor is also ignored in most of the studies. A very limited number of studies are available on returns to investment in health not only in Pakistan but in other countries also because of data limitation. Therefore this study is planned to fill the gaps in literature by exploring various dimensions of returns to investment in education and health. Returns are calculated by using primary and secondary data sources. Quality of investment in education and health is also estimated by using various proxy variables and in case of returns to investment in education returns are disaggregated by gender, region and marital status by using education as a discrete and continuous variable with the help of mincer earning function. Returns to health are also estimated by using different measures of health which is a very novel contribution in literature.

109

Chapter 4 THEORATICAL & CONCEPTUAL FRAMEWORK

4.1 Introduction

If we review the recorded history then it is found that life expectancy was short, level of income was low, and there was less or no economic development. Now the people have healthier, longer, richer and hopefully happier lives. The shift in regime involved increment in knowledge and its diffusion, education and the great level of training, more migration, improved health, demographic transition and change in fertility. In short, the whole process involved more advancement in human capital.

The pivotal role of mankind in modern communities elaborates the necessity and importance of in depth studies related to human value. Plato had spoken about social heterogeneity and highlighted one of the main facts that all individuals are different regarding ethnic groups and races. Meanwhile, Aristotle presented the theory of natural slavery which is also called aristocratic theory which claims that humans are neither intellectually nor physically equal. Who follows and who leads is determined by nature. According to Darwin “the differences among humans are due to the process of natural selection which is also called survival of fittest. Theories and analysis of Marx about the elimination of social classes are well recognized. Such authors claimed that the diversities among human beings depend upon some external factors like moral and philosophical beliefs, climate and religion, and such diversities contribute to the development of anthropological inheritance.

The concept of human value is one of the supernatural concepts of the economy. In economics, the concept of individuals as a focal point of every social and economic system has made the monetary value of man thoroughly necessary. It is a complex task to give a monetary value to the individuals due to the range of determining factors and variability of human beings. Relegating a monetary value to individuals is a mind boggling issue given the immense arrangement of fluctuation of individuals and the scope of deciding elements they are subjected to. The primary trouble experienced when attempting to give a monetary value to people lies in the development of a target premise by which to evaluate such a sensitive question. The different researchers who

110 have taken up the issue of relegating an incentive to people have found that every single monetary value being equivalent, there is an idle, subjective appraisal concerning the level of individual fulfillment and prosperity which every individual appends to their own lives.

4.2 Concept of Capital

Expounding on the human capital must not avoid the problem of the semantics of the word" capital". Numerous researchers surround the significance of the capital rather barely, limiting it to the products produced by producers. So, the idea is not reached out to intangible goods and not stretched out to durable goods of consumers. The emphasis is on physical goods that were created and which are not given by nature (like natural resources, mineral deposits and lands).

Other researchers, be that as it may, have thought that it was helpful to broaden the importance of capital in different aspects. One expansion identifies with the users of capital: the utilization is never again confined to producers producing other goods yet is expanded towards consumers getting from it a long continuing stream of intangible services. To grasp this, one simply needs to consider private lodging. Most national-wage analysts incorporate the development of private homes in the process of capital formation. So, in this way, capital is not physical intermediate goods which can be utilized in the production of other physical goods, yet a strong structure that gives a stream of benefits for consumers. Economic scholars are different from economic statisticians in that they incorporate into the capital stock a wide range of durable consumer products, for example, home appliances, cars musical instruments, and so on. Still, despite the fact that the future advantages might be elusive, the assets including in the capital stock are physical and tangible.

When we talk about human resources then it is found that the concept of capital is no longer confined to tangible assets giving intangible or tangible services, however, expands to intangible assets which further yield substantial services. Still another refinement is to be made concerning HR, to be specific, regardless of whether the capital idea ought to be limited to interests in enhancements of the human asset or whether the human asset ought to be viewed as capital regardless of whether any uncommon cost has been made to expand its efficiency. Economists have sometimes condemning of the unusual false notion of with respect to the introduction of a sheep as an expansion in wealth, however the introduction of a human child as no such

111 increment at the same time, rather, as weight, as a reduction of income per capita. If the new born child is not capable of addition in capital or wealth then all the investment in caring and feeding for the future workers should be considered as capital formation.

4.2.1 Four Categories of Capital

John W. Kendrick divided the annual capital formation into four categories: (a) human tangibles, (b) non-human tangibles, (c) human intangibles, (d) non-human intangibles. To distinguish the four categories there are some specific items: (a) Expenditures for build up children till the working age are considered investments in tangible human capital; (b) Expenditures for machinery, construction and inventory are investments in tangible non-human capital; (c) Expenditures for training, education, mobility of people, safety and health are investments in intangible human capital: and (d) Expenditures for development and research are investments in intangible non-human capital.

One can't resist being inspired by the perfect uniformity of this arrangement, however, one may take exception to category b, and kids from age 0 to 14 or 15 are represented by tangible capital. Without a doubt, parents may get much joy from their kids and, consequently, a psychic profit for their interest in encouraging them and administering to them; and in many economies parents may rely on their kids as a wellspring of income in subsequent years, may be due to their support in old age. From the economies’ point of view, over the excessive increment in the population may decrease the efficiency of labor.

4.3 Concept of Human Capital

Human capital is known by the aggregate investment in such activities like health, migration, on- the-job training and education that increase the productivity of an individual in the labor market. In these days, non-market activities are also included in this concept. Human capital is the combination of innate abilities, knowledge, health and skills that a person develops and acquires throughout his life. The former is obtained free of cost by the individuals as well as from society, while latter can be acquired by the efforts of the individual involving a cost. The productivity of individuals is enhanced by both above-mentioned components in the production of ideas, goods and services in the market and non-market environment.

112

The human capital is a multi-faced and complex concept. Eight specific aspects can be considered for the broad definition of human capital.

First, it is confirmed that human capital consists of innate and acquired element. Innate abilities of humans include all intellectual, physical and psychological abilities that a person posses at the time of birth. These abilities are received as a gift without any choice or action and are differ from person to person due to parental decisions, heredity and random factors. On the other hand, innate abilities are represented by potential of an individual for the accumulation of human capital which includes skills and knowledge to help in actualization of this potential. These skills are obtained through inter-generational transfers in whole lifetime through personal contacts, work experience, knowledge, good health, on-the-job training, socialization and education.

Second, human capital can also be considered as non-tradable good whether acquired or innate; knowledge and skills are incorporated into people. As long as people remain non-tradable goods, there is no market in which interchange of human capital assets would be allowed.

Third, people do not always control pace and channels through which they achieve human capital. They might not make rational decisions about the needs of human capital in young age, nor can they realize the power of their innate capabilities. So, during the 1st year of life, the decisions about human capital are normally made by government, parents, and teachers and by society through its social and educational institutions. The individuals internalize the process of decision about human capital investments when they become able to make decisions independently. Since the ability of the owner to invest in human capital is based on the social environment, the behavior of past investments and the influence of their peers continually shape their acquirement of human capital, both in amount and type.

Fourth, human capital can be received either by informally or formally. Informal acquisition can be done through personal contacts, social organizations, self-teaching and through work experience. Formal acquisition can be obtained from institutions and programs were both skills and knowledge are transferred in educational environments.

Fifth, human capital has quantitative as well as qualitative aspects. An individual’s daily caloric intake or total years of schooling can be easily quantified. We cannot assume that investment in human capital is qualitatively homogeneous in nature. For example, those persons who are

113 graduated from Harvard University may obtain a better formation than those persons who have graduated from less renowned universities.

Sixth, human capital can be either specific or general. Human capital is considered general if it is used only for limited activities and if the relationship between firms and workers shows significant loss of value which can be recovered through costly investments. Human capital can be called general when abilities and knowledge can be used in different tasks, and they are easily shifted from one employer to another employer without any considerable loss of value.

Seventh, the stock of abilities and knowledge in every individual may not always be fully utilized. This mismatch may arise due to the incompatibility between skills which are demanded by market and skills which are acquired due to economic fluctuations and distortions of the labor market.

Finally, some external effects are also present in the definition of human capital. These spillovers show that effects which individuals have on the physical capital and productivity considering the situation that workers will be more productive in an environment, for any given level of skills containing the high level of human capital. So, there are different types of qualitative human capital due to the stock of the human capital present in the country, innate abilities of individuals, skills and education, teachers, means by which abilities and knowledge are transmitted and availability of financial resources to the society and individuals.

4.3.1 Uses of Human Capital

According to the standard definition of labour economics, ‘the human capital is a set of characteristics/skills that increase the productivity of the worker. This is considered as a useful starting point and quite sufficient for practical implications. This may be beneficial for differentiating between some alternative/complementary methods of explaining human capital. There are following categories:

1. The Becker View: human capital can be used directly in the process of production. It enhances the productivity of worker in all tasks through the possible differential in different tasks, situation and organizations. Although the human capital’s role in the process of production is complex but it can be represented by unidimensional objects,

114

such as skills and stock of knowledge and this is considered the part of the production function. 2. The Gardener View: according to this view, the human capital is not considered as unidimensional because the skills and knowledge have many dimensions and types. The simple version of this approach insists on physical vs. mental abilities in the form of different kinds of skills. 3. The Schultz/Nelson-Phelps View: human capital is mostly considered as the capacity of adaptation. According to this view, human capital is useful for dealing the situations of disequilibrium and for those situations where the environment is changing and workers have to change this environment. 4. The Bowles-Gintis View: human capital is represented by the capacity of work in organizations, obey the orders and to adopt the life in a capitalist/hierarchical society. According to this approach, the main responsibility of educational institutes is to give their students the correct approach and ideology towards life. 5. The Spence View: observable human capital measures are signals of ability that are independently useful for the production process. The first three views are quite similar despite their differences because human capital is valued in market which increases the profit of firm. In fact, the view of labor economists about the human capital is a combination of these three approaches. According to the view of Bowles-Gintis, the firms give higher wages to educated workers because they will be more useful for the firm, will be more trustable members of the firm’s hierarchy and can also obey orders better. Different implications of these views are discussed below.

4.3.2 Sources of Human Capital Differences

It is useful to discuss the possible human capital sources before explaining the incentives to the investment in human capital.

1. Innate Ability: due to innate differences, the workers have different amounts of human capital/skills. According to a research of social biology, there is some component of IQ level which is genetic to the origin. The authentication of this observation is twofold in labor economics: (i) if individuals have same economic constraints and same access to investment opportunities then human capital is likely to be heterogeneous; (ii) we have to

115

deal with this source of differences in empirical applications, especially when it is correlated with some other variables of interest. 2. Schooling: it is the most observable element in the investment of human capital. The analysis of school is very instructive if we assume that the forces which effect investment of schooling are also effected the non-schooling investment. So we can obtain from the behavior of investment of schooling that what would be happened to non-schooling investment which is more complex to be observed. 3. School Quality and Non-Schooling Investments: if the same year of schooling is completed by two identical twins then they may have different amounts of human capital. The reason is that they may have made different investments in other elements of human capital like studying some specific subjects, working hard or due to variety of circumstances/choices, or one may have better at communication and more assertive, etc. 4. Health: this is also a very important component of human capital difference. Suppose two persons having the same level of education and other skills have different earnings, this is because of differences in their health conditions like the disability to do an activity properly, Debility or missing days due to illness. Childhood health investments are also a great source of earning differences. 5. Training: this is that segment of human capital which workers obtain after completing their schooling, regularly connected with some arrangements of aptitudes helpful for a specific industry. At some level, training is fundamentally the same as schooling in that the workers have decided that how much to invest. It is troublesome for a worker to make preparing speculations without anyone else's input. The firm additionally needs to put resources into the preparation of the laborers, and frequently winds up bearing a vast part of the expenses of these preparation ventures. The part of the firm is considerably more prominent once we consider that training has a critical "coordinating" segment as in it is most helpful for the specialist to put resources into an arrangement of particular technologies that the firm will be utilized in a future.

4.4 Comparison between Physical and Human Capital

In economics, human capital is considered in the same way as physical capital. This section will highlight the variances between physical and human capital that applies to the public policy in

116 the knowledge based economy. Physical and human capital differs from each other concerning marketability and property rights, returns, accumulations and financing.

A. Property Rights and Marketability

Physical capital is tangible, and it can be easily touched or seen like factories, plants, machinery, patented processes, means of transportations, raw materials and communications and inventories by traders or producers. Moreover, physical capital can be transferred or sold from one person to another easily. On another hand, human capital is inseparable from individuals and its ownership cannot be transferable from one person to other.

B. Accumulation

The difference between the formation of new capital and the depreciation in the existing stock of capital is called the accumulation of capital. However, the process of accumulation of non- human and human capital is different concerning depreciation rates, technology and decision making which is used to produce these two forms of capital.

1. Decision Process: The process of decision in the accumulation and production of non- human and human capital involves decisions under the uncertainty by the firms and individuals. On the other hand, the decisions about the physical capital are made by managers or investors while the decisions about the production of human capital involve decisions by parents, peers and educators. Investment in human capital relies upon the existing stock of capital. If the abilities of individuals were not developed at an early stage then they would confront fewer opportunities of the accumulation of human capital during adulthood. 2. Accumulation of Capital: The accumulation of physical and human capital has some similarities. Both involve foregoing current consumption to increase the consumption and production in future. The human capital accumulation has a social aspect which is very less in physical capital. Infect, human capital is accumulated and developed with the interaction of ideas and individuals, hence making it a social activity (Lucas, 1992). This inherent characteristic shows that the production process of human capital is more labor- intensive than physical capital. Hence, human capital is produced by the interaction of

117

human beings so it is subject to the externalities and spillovers which have the power to change the accumulation and learning processes. 3. Depreciation: Both physical human capitals are depreciated with the passage of time. The physical capital depreciates when it is either used up or consumed. Technology, knowledge and abilities are the part of both kinds of capital, and when improved and new technology available then human and physical capital becomes obsolete. Inactivity impairs the skills so when human capital deteriorates then it becomes idle. A component related to the depreciation of human capital directly related to the ageing, involuntary unemployment and external shocks. Another component of human capital results from the conscious decision of individuals about the use of skills and knowledge in the production process. According to the investors’ point of view, at the time of retirement, human capital depreciates continuously and approaches to zero at the time of death. But according to the perspective of society, the individual who has invested in human capital does not imply the total loss of capital after his death, and the part of his knowledge would have been shifted in other generations through the production of goods and services, personal contacts and ideas prior to the death. This characteristic does not apply to non-human capital, and physical capital assets can be transferred from one person to another person.

C. Financing

Lenders easily lend physical than human capital accumulation because physical capital can be easily sold, jointly owned, seized and transferred by inheritance or by sale while human capital is dissociable and intangible from its owner. So it is harder to obtain private financing for the human capital acquisition. Markets fail in the financing of human capital due to liquidity constraints and income inequalities, so low investment in human capital would result in the suboptimal level of human capital. To overcome these potential inefficiencies which occur due to market failures, the government established programs to subsidize the financing of human capital. For example, financing the education through low tuition fees, scholarships and study loans.

118

D. Returns

The positive rate of returns can be achieved through investment in human and physical capital. Investment in physical capital produces only pecuniary returns while investment in human capital also produces non-pecuniary returns. It has been believed that both the flow and stock of human capital can be the argument in production and utility functions. So, it is difficult to differentiate the production motive from the consumption motive of human capital formation.

Returns to physical and human capital tend to behave differently. When people invest in physical capital then they are considered return takers and so they accept the returns which are dedicated by the market and cannot influence them. On the other hand, investors of human capital are return makers because there are no markets for human capital stock. The maintenance and quality of the human capital stock will decide the amount of offer by the market for their service. So, the returns to investment in human capital are more endogenous in nature.

Along with human and economic capital, social and psychological capital is also very important to enhance efficiency of a person and thus enhances monetary and non-monetary returns. Psychological capital enhances positive attitude towards future challenges, confidence and perseverance. Human capital shows the skills of a person that can be enhanced with the help of better social interactions and challenging working environment and by positive attitude towards work. Psychological capital improves the confidence of worker and he becomes optimistic thus performs much better and plays a significant role in economic growth.

4.5 Investment in Education and Health

Health and education are two basic objectives of development, and they are considered important ends in themselves. Health is necessary for the well-being of society, and education is essential for rewarding and satisfying life. Both are explaining human capabilities that present at the heart of the meaning of development. Education plays a fundamental role for the developing country to develop the capacity of self-sustaining development and growth and to absorb modern technology. Hence, health is necessary to increase the productivity of country and successful education system relies on the adequate health system. So, education and health can be seen as significant components of development and growth and also used as inputs in the aggregate

119 production function. The dual role of health and education as inputs and outputs gives them their main importance in economic development.

4.6 Producing Human Capital: Education and Training

A principal difference amongst people and different species is the broad transmission and safeguarding of information among people. This transmission and protection is the thing that had prompted present day economic development. In any case, the transmission couldn't have been wide based and couldn't have come to the "masses" of individuals notwithstanding foundations called schools.

4.7 Investment in Education

Education is considered as a major form of investment in human capital. Now, the question is that whether one should consider all expenditures of education as human capital formation regardless of considering whether a specific learning and teaching would enhance the labor productivity. Education can improve the quality of life by elevating the learner’s intellect, but it can also improve the efficiency and skills of individuals through producing useful things. A few economists see an essential distinction between the commitment that training may make to the stream of impalpable fulfillments to educated themselves (and their companions and peers) and the commitments which the skilled and effective laborers will make by producing material goods in more prominent amount or of better quality.

Durable consumption goods are treated as wealth but not as capital by many economists. Sources of intangible flows of consumer satisfaction are treated as capital by the national income accountants. On the other hand, some economists do not consider this as the main difference and they do not care whether the derived services are used for production purpose or only for consumption.

The oldest reference about the value of education is defined by a Chinese proverb:

“Give a man a piece of fish and you will feed him for one day. But if you teach him how to catch a fish, he will be fed whole the life."

120

The formation of human capital means to sacrifice the resources today for the benefits in the future. This tradeoff is explained in Figure 4.1. The profile of age-earnings for a less educated or untrained worker (thin line) is compared with the age-earnings profile of more educated worker (thick line).

Earnings More Educated

Less Educated

Age

Figure: 4.1 Basic Tradeoff in Human Capital theory

The profile of age-earnings is flat for the less educated worker while the same profile is steeper for the more educated worker. The earnings of the more educated worker are sacrificed by his/her career, so these earnings are less than less educated workers. However, when training period is over, as shown by the intersection of two profiles in the figure, the earning of the trained worker is higher than untrained workers. The reason is that the trained worker is more productive than untrained workers. For the validation of theory, it is necessary that the two profiles should cross each other. It is also necessary that they have to cross early in professional career so that time is allowed for the recovery of costs. Education is considered as a special case of training.

4.7.1 Defining Rates of Return to Education

The benefits of investments in education are broad, and many times difficult to quantify. These benefits might involve, apart from the social, non-monetary ones, benefits such as wages offered, wages received, and employment (Arrazola & de Hevia, 2008). However, there seems to be slight, but important, divide in the literature between defining and estimating returns to education. When defining returns to investments in education, most analysts observe earnings as

121 a function of the costs incurred to obtain such earnings. When estimating returns to investments in education, rates of return are usually seen as a measurement of the future net economic payoff of increasing the amount of education taken (Carnoy, 1995).

Measuring returns to investments in education means calculating the benefits of increased level of education at both the individual and the national level. The benefits of the increased level of education at the individual level are known as private rates of return, and the benefits of the increased education at the national level are known as social rates of return. More specifically, private returns to education refer to the benefits of individuals from investing in education, whereas social returns refer to the large scale benefits of such investments. Social returns also take into consideration the direct costs of schooling incurred by institutions or governments.

When calculating social and private rates of return based on educational level and income, attention is generally given to individual income tax payment. Because income tax is a cost to the individual but not to the whole society and income of a person should be calculated after the deduction of income taxes when the private rate of returns are calculated and before payment of income taxes when the economic (social) rate of returns are calculated (Perkins, et al., 2001). In short, Perkins indicates that one of the main differences between private rate of return to education and the social rate of return to education, apart from the obvious difference made by considering the cost of schooling incurred by institutions and governments when estimating social returns, is the after-tax, and before-tax calculation of earnings; private being after tax and social before.

Psacharopoulos differs with Perkins and Hicks’ views and argues that “despite the popular belief, the pre-tax versus post-tax treatment of income does not make a big difference in the calculation of the rate of return.” (1981). In other words, according to Psacharopoulos, it is the direct costs of schooling incurred by institutions or governments, which makes a significant difference when calculating private and social returns to education; not the pre- and post-tax measurement of earnings.

122

4.7.2 The Chicago School Approach

The contribution of Mincer is influenced by the theories which were formulated by Chicago School (Mincer, Schultz and Becker). The Chicago School had revived the concept of human capital during the second half of 20th century. Taking working experience and years of schooling as a baseline the authors analyzed the concept of human capital in detail and provided significant advancement in the mechanisms of accumulation and formation. The main idea presented by Chicago School is that the on-the-job-training and years of schooling are the main determinants of the evolution of labor-generated income for the workers over a life time. The main concentration of Schultz, Becker and Mincer was on human capital and they evaluated the costs, effects and advantages of investment in human capital rather than physical capital. According to Schultz the increase in national income of a country is derived by the growth of the human capital stock.

According to the view of Schultz, the human capital consists of the complex of training and education (professional experience, years of schooling and working years, etc.) but principally it contains stock of human capital which is measured by its national accounts. There are two main components of this stock of human capital: no earnings while students have gone to school with a commitment to activities which can generate income (called foregone earnings) and the current costs and direct costs due to education (school fees, buildings and the salaries of teaching staff).

4.7.3 The Mincer Equation

Mincer (1958) was the first person who used the term in his work to build earning function by using mathematical tools of the neoclassical theory of capital. He developed a very parsimonious model by using a year of schooling and found 60 percent of the variation in the income of adult white men of USA. His earning function was used in more than 100 countries, and same resounding success was achieved with USA data. The Mincer equation estimates the rate of returns in education via multiple regression analysis (Hough, 1993).

According to Bjorklund and Kjellstrom (2002), the Mincerian equation is considered one of the most commonly estimated relationships in the labor economics, and it related the year of experience and squared of experience and the logarithm of earnings to schooling. One of the reasons for the popularity of Mincerian equation is that it uses results from human capital theory

123 to obtain schooling coefficient and earnings equations which are closely related to the marginal internal rate of returns of the education. Despite the limitations of the Mincerian equation, if simplicity is considered for the estimating of the impact of work experience and schooling on wages, the Mincerian equation is hard to beat (Björklund & Kjellström, 2002).

4.7.4 The Elaborate Method

The elaborate method works with age earnings profiles by the education level and finds the rate of discount that equates the stream of educational cost to the stream of educational benefits at a given point in time. The annual stream of benefits is calculated by the advantage of earnings of graduates for that educational level in which rate of return is estimated. The stream of cost in the shape of foregone earnings is calculated by the calculation of private return augmented by the true cost of resources of schooling in the calculation of social rate of returns. Private rate of return is used to measure the behavior of people in seeking different levels and types of education, and as the distributive measures of the use of public resources. Social rates of returns can also be used for the educational investments in future. The limitation for the elaborate method is that it needs detailed data for the profile of age-earnings by the level of education and this data is difficult to come across in many countries.

4.7.5 Two Methods for Estimating Returns of Investments to Education Based on the Mincer Equation

The most important methodology for estimating the education returns are based on the Mincerian equation is short-cut methods and earnings function. The method of earning function calculates the private returns of education with the help of regression of logged earnings on years of experience, square of year of experience and year of schooling. The education returns are calculated by the regression coefficient on schooling while short-cut method calculates the private returns to education as the proportion of years of schooling and earning. The equation presented by Mincer has been widely accepted and referenced since the time of its inception in 1970 and has been used as a standardized method of calculation of returns to investment in education (Heckman, Todd and Lochner, 2003; Asaoka, 2006; Psacharopoulos, 1981; Patrinos & Psacharopoulos, 2010).

124

I. The Earnings Function Method

To estimate the educational returns, the earning function method is used which is a regression of the following basic form:

2 ln...Yabiiii=+++ Sc EXdEX

Here, Y shows the wages of individuals, S shows years of schooling, and EX denotes the experience of the labor market (Mincer, 1974; Psacharopoulos, 1995). Experience as a squared form (EX2) shows the nonlinear relationship between experience and earnings: In early years after the entering in the market the earning increases but later on it decreases through time (Harmon et al., 2003; Gunderson & Oreopoulos, 2010). The coefficient b in above equation can be denoted as the private rate of returns to one more year of schooling (r) which is shown as:

ln1Y YYYYzXzX−− br====  SYSYSXX.

Here ∆Y shows the relative change in earnings. Yz and Yx show the earnings with z and x year of schooling while z is highest level of education and x is immediate lower level. R can be considered as the limitations and assumptions of the model and proportional effect on wages of an increment to S (Psacharopoulos, 1981; Harmon, et al., 2003; Gunderson & Oreopoulos, 2010).

One of the important limitations of this approach is the assumption that rate of returns is same for all study levels by making difficult to solve the problem of resource allocation for different levels of education. When we estimate the return of one more year of schooling, then this approach assumes the same returns regardless of the level of education. Nevertheless, a more year of higher education have a potential to give more returns than a more year of education in a schooling of lower level. Also, the educational returns of the year previous to the year in which education is completed have been observed lower returns than the returns of the year in which education is completed: the sheepskin effect (Heckman, Lochner and Todd, 2006; Labor Department of USA, 2000).

125

Given these limitations, it is possible to incorporate the dimension of the level of education to the concept of rate of return. It is incorporated that levels of education can be attained by different series of dummy variables (i.e., PRIM, SEC and TERT) to the function of earnings model. The resulting regression function is as follows:

ln.....Yab=+++++ PRIMc SECd TERTe EXfEX 2

From which returns by educational level can be estimated. The rates of return to the different levels of education, relative to their immediate lower level, are derived from the estimated coefficients of b, c and d in the function, and are:

b r( primaryVs, Illiterate) = S p cb− r(sec.ondaryVs primary) = SSsp− dc− r(tertiaryVsondary.sec ) = SSts−

Here, S means the year of schooling for the subscripted level of education ( p = primary, t = tertiary, s = secondary) ( Psacharopoulos, 1981).

II. The Short-Cut Method

The short-cut method is considered the simplified version for the method of earning function and it calculate the private returns of education with the assumption that earning is proportional to the number of years of education (Mincer, 1974). Social returns to education are calculated by incorporating the education cost to the denominator of the equation.

Private returns of education are calculated according to the short-cut method by mathematical approximations in the earnings function.

YYk − (ks− ) privater k = SYk . (ks− )

126

rk Here, private shows the private rate of returns to the investment in the education level of k. Sk shows the number of years in the subscripted level of education. Social returns of education can be estimated by the following level of education:

YYk − (ks− ) Socialkr = SYC. + kk( (ks− ) )

Here, Ck indicates the public expenditures per student in k level of education.

4.8 Investment in Health

Investment in health as well as investment in education affects the productivity of an individual. Resource allocation of the household and the decisions about the consumption decide the health and nutritional status of the adults and children within the household. These decisions have impact on the anthropometric measurements of the adult such as body mass index (BMI), height, chronic and acute morbidity, and the patterns of disability and illness. Besides schooling, human capital also has many other forms like the capacity to avoid unwanted fertility, migration, and the health outcomes.

The framework which was set by Mincer (1974) has allowed for additional forms of human capital besides education. Schultz (1997) analyzed that how the investments by family and state influence the structure of reproducible human capital level and how these affect growth and earnings. The important questions he observed are: what are the wage returns and what determines the demand for the human capital in the labor market? Schultz (1997) observed that height of the adult is an important factor for the measurement of adult productivity and it is inversely related to the chronic problems related to the health among the middle aged and elderly people. It was also observed that height of a person was inversely related to mortality and directly related to the length of productive life.

4.8.1 Health, Human Capital and Income

In 1650, Thomas Hobbes had described in his writings that life was brutish, nasty and short. At that times life was filled with disease, infections and pestilential maladies. But now people have become taller and healthier. They live the long life and have lived with less suffering and pain.

127

People invest more in health and human capital due to increased resources and so they have become more productive. Due to better health and productivity, the income of the people has also increased.

Improvements to health for most history are due to increased and better resources. Due to more resources, people consume more protein and calories and to eat more nutritious foods. Improvements in health have served to increase the income. Improvements in the health of young enable the children to learn more and to attend the school for more days. Improvements in the health of the people allow them to work more years over their life.

4.8.2 Measures of Health Human Capital

Mortality is considered one of the main indicators for the status of health which exists for many places and across long periods. Historically, a large number of indicators of health exist like weights and heights for children and adults, body mass index (BMI) and the rate of infectious and chronic disease.

I. Demand for Health Care

According to McGuire et el. (1993), the analysis of health care is notorious due to the lacking of theoretical basis. Grossman (1972) provided a theoretical model for health, but due to imperfect health care markets it still needs to be adopted for better specifications for the models of health care. McGuire et.al.(1993) also explained that without any theoretical basis an additive functional form could be estimated by virtual studies (Newhouse, 1977; Hitiris and Posnett, 1992; and Wolfe, 1986) while it may be linear or nonlinear, which is yet to be known. For example; some studies have found that income after a certain level has a declining affect on infant mortality (Rodgers, 1979 and Younger, 2001). Linear models show that inputs which are used in models are independent and giving constant marginal products when an additional unit of input is used. On the other hand, log form models indicate the decreasing trend of marginal outputs (McGuire et.al, 1993).

The law of diminishing returns as applied to population health suggests that with an additional input the marginal increment to health improvements reduces. It is rationale because societies make more cost effective decisions keeping other things constant (Bishai, Opuni and Poon,

128

2007). It also proves that with the increase in income its effect on reducing mortality declines provided the distribution of income is egalitarian. Demand function approach is used in many studies related to the determinants of health expenditures to specify the models, specifically, real health care expenditures (HCE) is hypothesized to be a function of real GDP and a selection of economic and non-income variables. It is explained in many empirical studies on health and health care demand that the determination of health expenditures includes political and economic factors.

Changes in income of person affect the demand for health care and so the coefficient of income with respect to any health care measure (e.g., health expenditures) is termed as elasticity and based on its magnitude and health care can also be described as inferior, normal and superior good ( McGuire, et.al, 1993). While, there is also concerns about the status of this income elasticity which is in most of the cases above unity for Developed countries while it is less than unity in case of developing countries but the issue of whether health care is luxury or necessity is less settled yet (see for details for example; Grossman 1972; Hansen and King, 1996; Karatsaz, 2000; Okunade and Karakus, 2001 and Parkin, McGuire and Yule, 1987).

II. Production Function Approach

Grossman (1972) stated that the demand for health care is derived which is produced through a process defined by a production function. His model is widely used in empirical studies of health and health care. Grossman (1972) developed a theoretical model of health care which is commonly applied for analyzing the factors affecting health status and its relationship with economic and non-economic factors. The model can be specified as;

H= f (A)

Where H is any measure of health status like Life expectancy, infant mortality and (A) is a vector of other economic (income per capita), social (education), environmental (urbanization), demographic (population below or above certain age group) and health service variables (like population doctor ratio, population hospital ratio etc) variables affecting health status. Although, Grossman (1972) presented a model at the micro level but number of studies tried to employ his specification at the macro level also. Grossman (1972) production function specification talk about initial health endowments (at the individual level) as decreasing over the life cycle

129 provided the due investment has been made. While to make the analysis more closely to what Grossman specified it is better to represent the variables in their per capita form. This is done because; first, it avoids any sort of inequality that can distort the results of the analysis and create bias results. Second, it can be helpful to see the results in a more homogenous way. A country is a big unit of analysis and region may differ widely, but per capita estimation can be a possible source to avoid these regional biases to some extent. We can rewrite above equation in per capita extended form as follows;

h= f (e, d, p, s, n) e is economic factors in per capita terms affecting health status h and d is demographic factors, p political factors, s is social factors and n environmental factors. Classifying input and output for the production function is necessary, Feldstein (1967) suggested some output definitions one of them is improvements in health (i.e., increase in life expectancy or reduce mortality).

III. Estimation of Productivity of Health Investments by Extended Mincer Earning Equation

To estimate the return of investment in health, a Mincerian earning function is calculated which depends on the human capital. Mincerian log earning equations are estimated for this purpose which considers the indicators of health as hourly earnings determinants. The following earning function is estimated:

log(Yabijjik) =+ kihhiiXc + cd + Hf +

Here, Yi represents hourly earnings or productivity measure, Xji represents exogenous endowments which are not modified by the family or individual, Cki is the reproducible forms of human capital and Hhi denotes indicator of health status. In above equation, the indicators of health status are not correlated with the error term ‘f’ and exogenous to the function of hourly earnings.

Three variables are used in separate regressions in indicator of health status (Hhi ):

1. A dummy variable which shows the value of one in the case when a person could not go to school for at least one day during the previous month due to disability or illness.

130

2. Duration of disability (in days) during previous month; and 3. Height of a person

The number of days of disability was used as a threshold for the inability of work and gives the information that how long a person is incapacitated. On the other hand, the height of a person shows the exposure to disease, nutritional status and variation in other environment factors (Schultz, 1997).

The earning function is estimated separately for women and men because some control variables and health status differ by sex and height. The separate earning function is estimated for urban and rural areas although they are linked with each other by the choice of migration. The inclusion of health variables in the regression does not change the educational returns. To check the nonlinearities in the analysis quadratic terms in a number of days and height are included.

4.9 Role of Human Capital in Theories of Growth of Developing Countries

Education and information are main components to determine human capital. Education is considered a compulsory requirement for the people to attain an adequate quality and quantity of working skills. In education, vast activities are included which are connected with know-how, knowledge, referred to investment in human capital. Recently, the development of endogenous growth theories has stimulated the vast literature of empirical studies about human capital. The real contribution of human capital to economic growth was checked due to the available statistical data.

The capacity of physical capital tends to get smaller while human capital, with accumulation, become the inexhaustible motor for growth. Investment in human capital decided by the economic agents can generate continuous growth over time. Mankiw, Romer and Weil (1992) included human capital in Solow model in a rigorous manner. They measured the registration rate to secondary education and managed to explain a reasonable proportion of growth rates in different national economies. Lucas (1988) explained that positive externalities were produced due to human capital. Externalities occur when one economic agent make choices, and this creates benefit for the second agent without the first agent being rewarded for it, i.e. when investment in a person increases the productivity of another person.

131

The developing country cannot ignore the training of human capital; otherwise, comparative advantage on a globalized scale can be achieved only regarding the output of mature goods. Japan is an important example of growth through the replication and called ‘Asian Tiger’. In the process of replication the role of human capital is decisive (Nelson and Phelps, 1966). Technically skilled local labor is required for the popularity of imported technology as technology is not ready to be used anytime in anyplace. Moreover, technology also requires technical skills which are better than those which are generally adopted by the labor force in developing economies.

Backward countries might exploit the gap of technology which differentiates them from advance countries by adopting and replicating the technology developed by the latter country. The availability of critical mass of skilled and educated labor force is compulsory requirement to maintain the process of development in both backward and advanced economies (Nelson and Phelps, 1966). Some scholars (Romer, 1990; Azariadis and Drazen, 1991) have found threshold values in the stock of human capital of under developed countries. Some other studies have explained how the educated labor force was a necessary factor to explain the miracle of growth in some selected Asian countries (Amsden, 1989, Lucas, 1992).

4.10 Education, Health and Economic Growth

During 1960s theories related to economic growth comprised fundamentally of the neo- traditional models, created by Solow and Swan (1956). These economic growth models fixated the considerations of macroeconomists on the tangible capital arrangements by principle constraint of economic development. As per Solow models, capital and work couldn't clarify the economic development alone. His objective was to characterize the commitment of the elements of creation and innovation. Notwithstanding, these development models demonstrated that because of reducing the minor returns in substituting the physical capital for work. The aggregation of capital would not inconclusively bolster a steady rate of growth in labor productivity.

According to neo-classical models, the output of an economy develops as a contribution of capital and work. Non-economic variable, i.e. human capital has no part in these models, and furthermore the theory of diminishing returns to scale operates in the economy. With these

132 presumptions, these models demonstrated that as the capital stock expands, the development of economy backs off, and so to keep a similar pace, it must enhance the innovation. This sort of instrument in the neo-classical growth model is neither inalienable not does it endeavor to clarify much. It implies that the innovative advance is exogenous to the framework. During the 1950s and 1960s, Solow hypothetical structure bolstered arrangements that concentrated on the development of businesses, capital stock and the rate of sparing as the channel to empower economic development and higher wage per capita.

Endogenous Growth Models supported that, family units can spare and put resources into health and education which further increase the market value of labor. These savings may help the individual or family unit straightforwardly, or it might embrace a more respectable shape in which parents put resources into the education of their kids. Romer (1986) expands the idea of capital by including human capital. Another perspective of investigation moves consideration far from considering human capital as the immediate contribution to the generation of merchandise, rather its emphases on different activities by skilled labor, all the more imperatively advancement. Mechanical change revolting from innovative work venture that creates a superior scope of products is the fundamental type of development perceived by the endogenous growth literature.

These models indicated that the accumulation rate of both physical capital and human capital could be influenced by the policies of the government. Human capital contributes development by expanding profitability, helps in the innovation of new innovation and empowering their selection. Preparatory level of education will impact the development rate of economy and economies merge quicker when their underlying load of human capital is bigger, and this stock influences development rates by expanding productivity. Human capital is the fundamental factor in clarifying economic development (Mankiw 1992), as it builds the yield through a few sensible and unmanageable ways. Human capital enables a laborer to create more yields. The increment in human capital is an unquestionable requirement for greatest use of physical capital. Change in human capital will draw the interest of the physical capital which leads to increase the level of production (Abbas, 2010).

It is, for the most part, considered that the primary goal of instruction is to create the human capital, however, this sort of capital is not created by training alone. Some social activities and

133 different activities likewise create human capital and furthermore add to the gathering procedure. Verifiably, human capital has been characterized as instruction achievement just while disregarding the health. As of late, health has been recognized as an essential part of human capital. Till the second half of 1990s the role of human capital comprised of education, although some economists accepted the significance of other factors i.e. education and health. Mankiw, Romer and Weil (1992), Fogel (1994), Barro and Sala (1995) incorporated health with training in as a fringe idea of human capital. They analyzed the connection between economic development and health. As indicated by Fogel (1994), a piece of instruction, better health, and physical quality are the source of addition in the accumulation of human capital.

Health and economic growth interact with each other in various ways. Enhanced health expands the efficiency of a laborer by expanding physical capacities, i.e. fortifies, perseverance and furthermore their mental limits, intellectual working and thinking aptitudes. It has been demonstrated that health majorly affects expanding work profitability. Great wellbeing in kids directly affects school participation and understudy execution. It builds the subjective and thinking capacity. Youngsters with weakness have brought down instructive achievement (UNESCAP, 2011). As indicated by Case, Fertige and Paxson (2005), kids with bad health get less years of education.

4.11 Conclusion

This chapter provided the conceptual base for this study. It is clear from this discussion that education and health are two strong factors affecting human capital. A higher level of education and better health conditions leads to increased productivity in the form of higher earnings and to prove this Mincer earning function and extended Mincer earning function would be used for data analysis in case of Pakistan and especially in case of Multan district. We will also observe based on this discussion that how investment in education and health at macro level is related to economic growth in Pakistan. Proceeding chapter will describe data collection techniques and methodology adopted for data analysis.

134

Chapter 5 DATA AND METHODOLOGY

5.1 Introduction

This chapter deals with sources of data followed by methodology and analysis. Analytical framework and econometric models have also been presented and discussed in detail. For macro data analysis, Johansen Cointegration technique is used for cointegration between variables used in investment models and for relationship between health, education and economic growth. Error Correction Mechanism (ECM) is used to examine the relationship in short run, while Granger causality is used to know the causal relationships among the variables. For primary data analysis ordinary least square technique is used to estimate returns of investment to education and health in Multan District. E-views Software version 9 and Minitab Software version 16 is used for analysis.

5.2 Description of Variables

In this section the major concepts and definitions which are used to describe the data, are presented. In first section we present the variables used in micro data analysis while in second section we will describe all variables used in macro data analysis.

5.2.1 Variables for primary data analysis

Table: 5.1 Description of Primary variables

Variable Description of variable

Dependent variable

Y Monthly earnings

Independent variables

TED A continuous variable for completed years of education of respondent

PRI =1 if education level of respondent is up to Primary

135

=0 otherwise

=1 if education level of respondent is up to middle MDL =0 otherwise

=1 if education level of respondent is up to metric SEC =0 otherwise

=1 if education level of respondent is up to intermediate HSE =0 otherwise

=1 if education level of respondent is up to Graduation GRD =0 otherwise

=1 if education level of respondent is up to M.phil or PhD UNI =0 otherwise

=1 if respondent’s medium of instruction is English MDU =0 otherwise

A continuous variable defined as total age of respondent minus total years of EXR education minus five

EXQ Experience square

FQL Total years of respondent’s father qualification

MQL Total years of respondent’s mother qualification

=1 if respondents matriculation subject is science SBJ =0 otherwise

WHR Working hours of Respondent per day

136

HET A continuous variable measures height in meter

A measure of weight given height, calculated by dividing weight in kg by height in BMI meter square.

A continuous variable that show respondent’s expenditure on nutritious food per EXN month

=1 if pure drinking water is available to the respondent HYW =0 otherwise

=1 if medical facilities are available to respondent MDF =0 otherwise

=1 if respondent spent money for precautionary measures PCU =0 otherwise

=1 if respondent spent some time and money on daily exercise EXE =0 otherwise

=1 if respondent is living in a clean environment ENV =0 otherwise

=1 if respondent self-reported health status is excellent or good HLS =0 otherwise

=1 if respondent is having any physical disability DSB =0 otherwise

In this study for micro data analysis Mincer earning function method is used where log of monthly earnings is used as a return of investment in education and health. Education is treated both as a continuous and as a discrete variable for this purpose five categories of education are constructed from primary till university education. We also used parent’s qualification,

137 matriculation subjects and working hours as determinant of person’s earning. Similarly to evaluate the returns to quality of education a variable medium is used that show the medium of instruction in high school.

To measure the returns of investment in health a long run anthropometric variable height is used that is measured in meters while short run measure of health investment is body mass index. It is constructed by dividing respondent’s weight in kg by their height in meter square. Similarly expenditure on healthy food like dairy products, fruits, meat and pulses etc. availability of pure drinking water, availability of medical facilities, clean environment are also determinants of good health. Similarly investment of time and money on daily exercise and on precautionary measures are also reported. Self-reported health variable and disability variable is also used in this study. Self-reported health variable takes the value of zero if health status is poor and one if health status is good or excellent while in case of disability we are talking about disability in bending, walking or climbing upstairs.

5.2.2 Variables for Secondary Data Analysis

In this section we are describing some variables in detail which are used in secondary data analysis. Data on these variables was collected from economic survey of Pakistan and world development indicators.

Infant Mortality Rate (IMR)

The infant mortality rate represents the number of deaths under one year of age per one thousand live births in some specific geographical region during a year. In this study, this variable is used as an outcome of health investment.

Health Expenditure as a Percentage of GDP (HEX)

This is total general government expenditure that is spent on health sector as a percentage of GDP. This variable shows public investment in health sector of Pakistan. More investment is expected to give more returns with better utilization.

138

GDP Growth Rate (GDP)

GDP growth rate is the rate at which a nation's Gross Domestic product (GDP) changes/grows from one year to another while GDP is the market value of all the goods and services produced in a country in a particular time period. The growth rate is the percentage increase or decrease of GDP from the previous measurement cycle. In this study, it is used as a proxy for economic growth because if the GDP growth rate increases then it means that economy is growing.

Population per Doctor (PPD)

This variable is a proxy for availability of health services to the population. It is measured as number of doctors available per thousand populations. As the number of patients per doctor increases it means better services are not available for population thus mortality rate increases.

Literacy Rate (LTR)

It is the quality or condition of being literate, especially the capability to write and read above 10 years of age. So, the literacy rate is the percentage of people who have an ability of write and read. In this study, literacy rate is used as a proxy for awareness about better health condition and for higher demand of education.

Total Fertility Rate (FTR)

The number of children who would be born per woman (or per 1,000 women) in her childbearing age is called total fertility rate. It is expected that increase in fertility rate leads to increase in mortality rate.

Education Enrollment Index (EEI)

Earlier empirical studies used school enrollment rate and literacy rate as a measure of education attainment but this study used weighted education enrollment index as a measure of education outcome. The method of Wang and Yao is followed in this study with some amendments to construct education enrollment index. Wang and Yao (2003) built a human capital series by using perpetual inventory method. Rather than using enrolments at five year intervals, they used number of graduates completing different levels of schooling and constructed a weighted index

139 for the attainment of education derived from five different levels of schooling: primary level, junior secondary level, senior secondary level, special secondary and tertiary level.

A slightly different method from Wang and Yao (2003) is used in this study for the formulation of human capital index due to various data limitations. Instead of completed education, we have taken the enrolment ratio at different years of schooling due to some constraints in data availability. In this study, education enrollment index is constructed by dividing the sum of enrollment at primary, secondary and tertiary level of education to the total population of that age from primary till tertiary level of education.

Expenditure on Primary, Secondary and Tertiary Education (EXP, EXS, EXT)

These three variables show the percentage of public expenditures on primary, secondary and tertiary level of education out of total public expenditure on education. These variables are used to measure the impact of investment on different levels of education individually on enrollment rate in Pakistan.

Inflation Rate (INF)

Inflation rate is the rate at which prices increase over time and purchasing value of money falls. Consumer price index is used as a proxy for measuring inflation rate. The Consumer Price Index (CPI) is usually represented by a basket of goods or products. It measures the average change in the price of this basket of goods over a defined period of time. Here, we have used CPI as average year change in prices.

Unemployment Rate (UEM)

The unemployment rate is defined as the percentage of unemployed workers in the total labor force. Workers are considered unemployed if they currently do not work, despite the fact that they are able and willing to do so. The total labor force consists of all employed and unemployed people within an economy.

Total Labor Force (TLF)

The total labor force is the number of people who are employed plus the unemployed who are looking for work. The labor force, or currently active population, comprises all persons who

140 fulfill the requirements for inclusion among the employed (civilian employment plus the armed forces) or the unemployed. In this study it is used as a proxy for labor in growth model.

Gross Fixed Capital Formation (GCF)

The net increase in physical assets (investment minus disposals) in some specific period is called Gross Fixed Capital Formation. It is used as a proxy for capital and shows its role in economic growth.

5.3 Model Specification

In this section we are presenting models constructed for micro and macro data analysis separately.

5.3.1 Model Specification for Primary Data Analysis

The micro data analysis is based on the human capital model followed by Mincer (1974) that is known as "Mincerian" method or "basic earnings function." It involves using years of schooling (S), the fitting of a function of log-earnings (Ln Y), experience in labor market (in years) and the square of experience in labor market as independent variables. The coefficient for years of schooling (b) shows private returns of one more year of schooling in this semi-log specification. Thus, the year of schooling is:

2 Ln Y = α0 + b.S + c.t + d. t + U

 ln W b = = r S

In fact, the coefficient (b) in the above semi-log basic earnings function shows the rate of return to investment in education. This study followed the methodology adopted by Psacharopoulos (1990), Siphambe (2000), Awan and Hussain (2007), Mcnown (2010) and many other studies based on Mincer earning function. The operational model used in this study is as follows:

Ln Yi = β0 + β1 TED + β2 EXR + β3 EXQ + µi ------5.1

In order to investigate the impact of different levels of education on the earnings of wage earners, extended earning function is used where education is considered as a discrete variable.

141

In this case following extended earning function is used where education is treated as a discrete variable and expressed as different education categories.

LnYi = β0 + β1MDL+ β2SEC + β3HSE + β4GRD + β5UNI + β6EXR + β7EXQ + µi -5.2

In order to investigate the impact of different determinants of education on the earnings of individual following model is constructed that shows the impact of working hours per day and subjects in matriculation on the current earnings of the individual. Similarly role of some family background factors like parents qualification is also considered in this model to evaluate their impact on earnings of respondent.

LnYi = β0 + β1TED + β2WHR + β3SBJ + β4FQL + β5MQL + β6EXR + β7EXQ +µi-5.3

To evaluate the role of medium of instruction in determination of earnings, medium variable is added in above mentioned basic earning function. This variable shows the return of investment in quality of education of respondent.

Ln Yi = β0 + β1TED + β2MDU + β3EXR + β4EXQ + µi ------5.4

According to Mincer (1974), productive benefits of health and nutrition can be calculated by including characteristics of health of the worker in the function of wage determination, just like schooling and experience. According to Schultz (2003), a standard semi log linear approximation of the hourly wage function can be expanded by including the n form of human capital as input and vector of Y variables that affect logarithmic of wages exogenously.

n w r H  dY  where i=1,2,3.------,m i j ij i U i j1

Where, r shows a proportional increase in wages affiliated with a unit increase in human capital. Thus Mincer earning function (1974) is further expanded to include health related variables like anthropometric measures of health; self-reported health and some determinants of health along with education variables. In the first model of health investment that is stated in equation 5.5 height as a measure of long run health investment and body mass index as a measure of short run health investment are added in earning function.

LnYi = β0 + β1TED + β2HET + β3BMI + β4EXR + β5EXQ + µi ------5.5

142

In the second step expenditure on nutritional food and availability of pure drinking water are added in earning function to explore the impact of investment in nutritional factors on earnings.

LnYi = β0 + β1TED + β2EXN + β3HYW + β4EXR + β5EXQ + µi ------5.6

In model 5.7 we added self-reported measure of health status and self-reported disability measure as a proxy for health investment and explored their impact on earnings of individuals.

LnYi = β0 + β1TED + β2HLS + β3DSB + β4EXR + β5EXQ + µi ------5.7

In the last model of health investment that is model 5.8 we mentioned some health determinants like availability of medical services, and clean environment; expenditure on precautionary measures and daily exercise as determinant of earnings of respondent.

LnYi = β0 + β1TED + β2MDF + β3PCU + β4EXE + β5ENV + β6EXR + β7EXQ+ µi-5.8

5.3.2 Model Specification for Secondary Data Analysis

This study at macro level relies on neo-classical growth theory where economic growth depends on the amount of capital and labor, and these two factors are the main powerful forces of economic growth. New growth theories argue that the one-off increase in human capital will be associated with a permanent increase in the growth rate. The social benefits of education will clearly tend to be much greater in this cas. (Sianesi, 2003). Neo-classical production function is stated as;

Yt = f (Kt, Lt)

Here, y is output during period t, K is capital stock while L is labor. We have used extended neoclassical growth model by including health, education expenditure and some other variables related to growth.

In the first model that is mentioned in equation 5.9 we are measuring the impact of public investment in health on health outcome that is infant mortality rate. Population per doctor is taken as proxy for better health services while literacy rate is the proxy for demand for good health.

IMR = α0 + α1HEX + α2PPD + α3LFTR + α5LTR + Ut ------5.9

143

In model 5.10 we are measuring the impact of public investment in education on education enrollment index. In education enrollment index, we are taking school enrollment at primary, secondary and tertiary level simultaneously. Similarly we are taking public investment in education in the form of expenditure on three main levels of education in Pakistan.

EEI = γ0 + γ1 EXP + γ2 EXS + γ3 EXT + γ4 INF + γ5 LTR + εt------5.10

Finally in model 5.11 we are exploring the impact of investment in education and health on economic growth of Pakistan. GDP growth rate is considered as proxy of economic growth and total labor force and gross fixed capital formation are proxies for labor and capital respectively.

GDP = δ0 + δ1LTR + δ2LIMR + δ3TLF + δ4GCF + δ5UEM + µt------5.11

5.4 Study Area and Sampling Technique

In Pakistan, Punjab province is divided into three regions namely, central Punjab, upper Punjab and lower Punjab. Lower Punjab is also known as Southern Punjab which includes the division of Multan, Dera Ghazi Khan and Bahawalpur. In this study, Multan district is chosen purposely as a study area from Multan division because this area has been neglected from research point of view especially in economic related issues. Multan district consists of four Tehsils which are Multan Saddar, Multan City, Shujabad and Jalalpur Pirwala. In 2005, Multan district was recognized as City District. City District Multan consists of six towns; Bosan Town, Shah Rukne Alam Town, Sher Shah Town, Musa Pak Shaheed Town, Shujabad Town and Jalalpur Pirwala Town. These towns consist of 129 union councils in total out of which 61 are urban while 68 are rural union councils.

A total of 42 union councils were selected from these towns by simple random sampling in which 27 union councils were urban and 15 were rural. The sample size of 850 wage earners was selected from these union councils by purposive sampling. This sample of 850 wage earners in the age group 15-65 comprises of rural and urban, male and female respondents. The purpose of selecting only wage earners is to account for the variation in the wages of respondents due to changes in the investment in education and health.

144

5.4.1 Questionnaire Design

This study is mainly a survey research so the main dependency is on primary data, therefore, a questionnaire has been formulated to obtain the required data. The questionnaire contains information about various important factors, like, education, age, family members, marital status, occupation, qualification of parents, household income, family type, total number of children, and questions related to the status of health, illness, medical and food expenses, disability, etc. A structured questionnaire has been used and some dichotomous questions have also been inquired to collect the information from the sample. All the questions have predetermined choice of answers. Dichotomous questions have only two choices of response, “yes” or “no”. Multiple- choice questions were also given with different choices and respondents were requested to select one or more.

The questionnaire has four different segments. The first segment includes questions about personal information of the respondents. Second segment contains information about qualification, experience and skills acquired etc. In the third segment, questions were asked regarding respondents family. Last segment collected information on the health related issues of the respondent like any disability, major illness, expenditure on illness, health care, food and various environmental factors. As some of the respondents were unacquainted with English language, so to get appropriate response, the questionnaire for these persons was translated into Urdu.

5.4.2 Pre-Testing

A pretest survey, based on thirty respondents was conducted randomly to test the credibility of the questionnaire. Some shortfalls which have been founded in the questionnaire after pretest were detached and errors were omitted before finalizing the draft of the questionnaire.

5.4.3 Sample Size Determination

The important part of primary data analysis is the sample size determination. In this study sample size is selected with the help of process proposed by Levin et.al. (2005) as below;

145

2 pp1  nDz  2

Here, Z represents the value which is generated from expected confidence level. P is the proportion of wage earners in Multan district. D is the design factor and it lies between 1 and 10 while ε is the level of precision. By following this method 850 is the selected number of sample for this study and Multan district is sampling area and data is collected from all Tehsils of Multan district.

5.5 Secondary Data Sources

Along with primary source of data, this study is also based on secondary source of data. Time series data analysis is used to analyze the returns of investment to education and health in Pakistan. Annual data from the period 1972 to 2016 is used in this study. This period was selected specifically because data is easily available in this period for all the selected variables. There are various sources for secondary data collection like the World development indicators, various issues of Economic Survey of Pakistan, Hand Book of statistics of Pakistan economy (2005) and 50 years Statistics of Pakistan’s Economy (SBP). These all are authentic sources on publishing country wide macroeconomic data periodically and /or on yearly basis but in this study we get data for all variables from first two sources mentioned above.

5.6 Methodology

It is necessary to check for the stationarity of the time series data before empirical estimation of the model. So, we will check whether the series is stationary or not. Unit root of the residuals or checking stationarity is considered significant test for the validity of equilibrium in long run. Augmented Dickey Fuller (ADF) is used to find out the stationary. For macro data analysis Johansen cointegration technique is used while for micro data analysis ordinary least square is used. Following is the detail of all these techniques and tests applied in this study.

5.6.1 Augmented Dickey Fuller Test

We have used Augmented dickey fuller (ADF) test to check the stationarity of the variables. Following equations are used at level while application of ADF test:

146

 (1) yytt1 ut

With no intercept and trend.

   (2) yytt1 ut

With intercept,

 (3) yyttt 1 ut

With intercept and trend, H

Here, variable of interest is yt and α represents intercept and βt shows trend and Ut indicates error term. We check the following hypothesis for above mentioned equations. According to null hypothesis the series is unit root while according to alternative hypothesis the series is stationary.

:1(4)         H 0

:1(5)        H A

If the variable is stationary at level, then it is integrated of order zero or I (0) or it has no long run relationship. If the issue is not solved at level, then stationarity is checked by taking first difference of variables in the same way without intercept and trend, with intercept, and with trend and intercept respectively. If the variables are found to be stationary at first difference then they are integrated of order one or I (1). The same process is repeated for second difference if the problem is not solved by taking first difference.

q        (6) yyyttt 11 i i1

q    (7) yyyttt 11 i i1

q         (8) ytt y t11 i y t i1

147

5.6.2 Ordinary Least Square

To determine the returns of investment to education and health in Multan district, we use the method of least squares which is also known as ordinary least squares (OLS) method. In this method, we minimize the sum of squared residuals. This method is widely used in regression analysis because it is mathematically simple and naturally attractive. This method is invented by Carl Friedrich Gause in 1974. The least square process has some striking statistical properties under certain assumptions which has made it one of the mainly admirable and influential methods of regression analysis. First, we explain the least square principal to know this method. Population regression function with two variables is:

Yi = β0 + β2Xi + ei

This function is not exactly observable. So it is estimated by using sample values in regression function which is as follows:

Yi = β0 + β2xi + µi

= yi + µi

Here, yi shows the estimated value of Yi. Luckily, the process of least squares gives us such a shortcut. The technique or principle of the least squares has chosen the value of β1 and β2 in such a way, for a given sample, that the method or set of data, ∑ µi is as minute as possible.

The OLS method of estimation can simply be expanded to those models which involve two or more explanatory variables. Technique of multiple regressions which is used by Pearson’s (1998) is helpful to predict the variance or dummy variables. Multiple regressions can also create a set of dummy or independent variables which define a variance in the dependent variable. Powers are also used to define curvilinear effects. The multiple regressions can be written as follows:

Y = b1X1 + b2X2 + ------+ bnXn + c

148

In above equation, b’s indicate coefficients and c represents constant. R2 defines the percentage of variance in dependent variable. R2 also shows the proportion of reduction while estimating the dependent variable knowing about independent variable.

5.6.3 Johansen Cointegration

To estimate the long run relationship between variables, we use cointegration technique. Johansen (1988) approach and Engle and Granger (1987) approach is used for this purpose but in this study we used Johansen approach. This technique deals with investigation of cointegration by using VAR model and then we test the number of cointegration equations explicitly. In this way we can overcome the problem of Engle- Granger (1987) two-step procedure.

Johansen (1988) and Johansen and Juselius (1990) have defined few stages for reliable results which are given below:

i. For the application of Johansen Co-integration approach, it is necessary that all the variables of time series are integrated of order one [I(1)].

ii. In second stage, lag length should be selected using the VAR model on the basis of minimum values of Final Predication Error (FPE),Hannan and Quinn information criterion (HQ) and Akaike Information Criterion (AIC),

iii. In third stage, suitable model should be selected for the deterministic components in the multivariate system.

iv. Johansen (1988) and Johansen and Juselius (1990) have examined two methods for the determination of the number of co-integrating relations, and these methods include matrix Π estimation. Trace statistic and maximal eigenvalue statistics are applied in 4th stage to determine the co-integration relationships, and for the values of standard errors and coefficients in model.

5.6.4 Vector Error Correction

A vector error correction model is a restricted vector autoregressive (VAR) which is made for the series which is non-stationary, and is known to be co-integrated. This series may be tested for co-integration by using an estimated VAR object. The VECM has cointegration relations to show

149 the restrictions in the long run behavior of endogenous variables which later on converge towards their cointegrating relationships and allows dynamics of short run adjustment. The cointegration term is also known as speed of adjustment term or error correction term because any deviation from the equilibrium of long run is adjusted gradually through data series of short run partial adjustments.

The Short run equations for three macro level models presented above are as follows:

KKKK IMRIMRHEXPPDFTR01234 JJJJ1000 t jt jt jt j

K 5111LTRECM tt (5.12) J 0 tj

WWWW EEIEEIEXPEXSEXT01234t jt jt jt j JJJJ1000 WW 56111INFLTRECM t  jt jtt    (5.13) JJ00

llll GDPGDPIMRLTRTLF01234 t jt jt jt j jjjj1000 ll 56111LGCFUEMECM   t jt  jtt  (5.14) jj00

5.6.5 Granger Causality

Granger causality test is helpful in finding the direction of causal relationship among the variables. It is preliminary condition for the granger causality that long run co-integrating relationships among variables must exist. Then we can determine the direction of causality among variables on the basis of probability values of F- distribution. It is necessary that probability value must be less than 0.10 for significant results of granger causality. To find the causality relationship following equations are used:

kk xt a i x t j  i y t j  t jj11

kk yt i x t j   i y t j   t jj11

150

Here, εt and θt are called white noise terms and k denotes the maximum number of lags because causality is considered highly sensitive about the number of lags. We check for no granger causality, unidirectional and bidirectional causality among the variables in this process.

5.7 Conclusion

This chapter gives a base for the empirical estimation of the models because it provides theoretical models which will be estimated to verify the relationships among the variables at both macro and micro level. This chapter also gives some issues related to methodology that a researcher has to face when he empirically analyze the models. This chapter also describes the description of different variables and data sources that are included in this study both at macro and micro level. Sampling technique and questionnaire design is also discussed in this chapter. It also describes the models which will be estimated in the next chapter.

151

Chapter 6 RESULTS AND DISCUSSIONS: AN ELEMENTARY DATA ANALYSIS

6.1 Introduction

This chapter presents a snapshot of sample data collected from six towns of city district Multan. 850 respondents were selected randomly for this purpose and data was collected by using questionnaire technique. This sample data covers detailed information about respondents their family, qualification, skills and health and uncover many important aspects related to this study.

6.2 Personal Information of Respondents

This section deals with personal information of respondent. In this survey, 1.29 percent data was collected from non-Muslims while 98.71 percent data was collected from Muslims as they are in the majority in selected region. In case of regional distribution we can see that randomly 64.47 percent respondents were selected from urban area of all towns and 35.53 percent respondents were selected from rural areas of all towns in district Multan. Almost 75 percent respondents were male and 25 percent were female because data was collected from wage earners therefore male ratio is high. In case of marital status almost 60 percent respondents were married while 40 percent were unmarried. With the help of this information, we can do our basic analysis that is how returns of education and health differ with respect to region, gender and marital status.

Table: 6.1 Religion, Region, Gender and marital status distribution

Religion Region Gender Marital Status Non- Muslim Urban Rural Male Female Married Unmarried Muslim 839 11 548 302 641 209 502 348 (98.71%) (1.29%) (64.47%) (35.5%) (75.41%) (24.59%) (59.05%) (40.94%)

Source: Author’s calculation using sample data

152

Table 6.2 represents town wise distribution of the sample. As we can see that how many respondents were selected from each district and it is clear from elementary analysis that almost 15 percent respondents were selected from each town that covers both rural and urban areas.

Table: 6.2 Town wise sample distribution

Town Sample Selected Percentage Bosan Town 133 15.65 % Shah Rukn e Alam Town 173 20.35 % Sher Shah Town 137 15.41 % Mumtazabad Town 125 14.71 % Shujabad Town 167 19.65 % Jalal purpirwala Town 121 14.24 %

Source: Author’s calculation using sample data

It is clear from this analysis that researcher tried to cover almost all dimensions of society and all fields and levels of work. 46.47 percent respondents were selected from the education sector, 10.24 percent from health sector, 8.35 percent were selected from industry workers, 0.94 percent from armed forces, 12.17 percent respondents were related to the public and social services sector, 5.18 percent respondents were related to financial and legal services and 16.12 percent were related to other departments.

Table: 6.3 Profession wise sample distribution

Profession No of respondents Percentage Education 395 46. 47 % Health 87 10.24 % Industry 71 8.35 % Armed Forces 8 0.94 % Public & Social Services 108 12.71 % Financial & Legal services 44 5.18 % Other 137 16.12 %

Source: Author’s calculation using sample data

153

Among 850 respondents 1.65 percent respondents were on executive posts, 13.76 percent were on management level posts, 15.41 percent respondents were on admin and clerical post, 16.35 percent respondents were related to menial staff. While 52.82 percent respondents were related to social services staff like persons working in health, education, police, WAPDA, railway on different posts except for management and admin level posts. Detail is given below in table 6.4.

Table: 6.4 Designation wise sample distribution

Designation No of respondents Percentage Executive Staff 14 1.65 % Management Staff 117 13.76 % Admin & Clerical staff 131 15.41 % Social services Staff 449 52.82 % Menial Staff 139 16.35 %

Source: Author’s calculation using sample data

In table 6.5 sector of employment wise distribution of sample is presented. 11.06 percent respondents belong to federal sector, 42.71 percent respondents belong to provincial sector employment, 3.76 percent respondents were from semi-government departments while 41.76 percent respondents were selected from the private sector and 0.71 from other sectors of employment.

Table: 6.5 Sector of employment wise sample distribution

Sector Of Employment No of respondents Percentage Federal 94 11.06 % Provincial 363 42.71 % Semi government 32 3.76 % Private 355 41.76 % Other 6 0.71 %

Source: Author’s calculation using sample data

154

Sample distribution according to nature of the job is presented in table 6.6 in detail. From 850 wage earners randomly selected for this study, 63.18 percent respondent were doing a permanent job, 18 percent were working on temporary basis, 1.65 percent respondents were doing a part time job, 3.65 percent respondents were doing a full time job, 11.65 percent respondents were on contract and 1.53 percent were daily wage earners.

Table: 6.6 Nature of job wise sample distribution

Nature of job No of respondents Percentage 1Permanent 537 63.18 % Temporary 153 18 % Part Time 14 1.65 % Full Time 31 3.65 % Contract 99 11.65 % Daily Wages 13 1.53 % Other 3 0.35 %

Source: Author’s calculation using sample data

From table 6.7 we can see that 10.24 percent respondents work less than 6 hours daily, 50.71 percent respondents work between 6 to 8 hours, 27.29 percent respondents work between 8 to 10 hours daily, 8.35 percent works between 10 to 12 hours while 3.41 percent respondents work more than 12 hours daily. It is an important determinant of earning.

Table: 6.7 working hours wise sample distribution

Working hours No of respondents Percentage Less than 6 87 10.24 % 6 – 8 431 50.71 % 8 – 10 232 27.29 % 10 – 12 71 8.35 % More than 12 29 3.41 % Source: Author’s calculation using sample data

1 Part time worker works fewer hours and has temporary job nature with no extra benefits but full time worker has to work more hours with some additional benefits but not full benefits like pension etc. On the other hand a permanent worker gets full benefit like pension and health benefits etc as per company policy.

155

6.3 Qualification, Experience, Training/Skills of Respondents

In table 6.8 detail of respondent’s terminal degree and terminal degree faculty or subject is presented in detail. Nature of work is consistent with the study of Afzal (2011) but area of study is different with some more modifications. Among 850 respondents only 8 were primary literate, 18 were middle, 57 were secondary, 67 higher secondary, 203 graduation, 315 masters, 60 M.phil and only 2 respondents were PhD qualified while 120 respondents have other terminal degrees like law, MBBS, engineering and some other diploma holders.

Table: 6.8 Terminal degree and faculty-wise sample distribution

Terminal Degree No of respondents faculty No of respondents 8 84 Primary Arts & Humanities (0.94 %) (9.88 %) 18 160 Middle Social Science (2.11 %) (18.82 %) 57 85 Secondary Business education (6.70 %) (10 %) 67 19 Higher Secondary Engineering & technology (7.88 %) (2.24 %) 203 80 Graduation Physical science (23. 88 %) (9.41 %) 315 60 Masters Biological & Medical science (37.0 %) (7.06 %) 60 17 Mphil Agriculture & veterinary sciences (7.05 %) (2 %) 2 116 Ph.D Science (0.23 %) (13.65 %) 120 229 Other General category (14.12 %) (26.94 %)

Source: Author’s calculation using sample data

156

In case of faculty or subject we can see that 84 respondents have terminal degree related to arts and humanities, 160 from social science, 85 from business education, 19 from engineering and technology, 80 from physical science while 60 respondent from biological and medical sciences. Science and general category is for respondents having matric terminal degree that whether they have matric in science or arts.

In table 6.9 information regarding terminal degree institution, country and examination system is presented. 9.18 percent respondents receive terminal degree from private institution while 90.82 percent received from public institutions. Only 2 respondents receive terminal degree from foreign country while remaining 848 respondents receive their terminal degree from Pakistan. 565 respondents appear in annual examination for terminal degree while 285 respondents appear in semester system examination.

Table: 6.9 Terminal degree Institution, country and system of examination wise sample distribution

Institution Country Examination system Private Sector Public Sector Foreign Country Pakistan Annual Semester 78 772 2 848 565 285 (9.18 %) (90.82 %) (0.24 %) (99.76 %) (66.47%) (33.53%)

Source: Author’s calculation using sample data

Table: 6.10 High School region, subjects and medium of instruction wise sample distribution

Region Subjects Medium Rural Urban Arts Science Urdu English 288 562 244 606 712 138 (33.88 %) (66.12 %) (28.17 %) (71.29 %) (83.76 %) (16.24 %)

Source: Author’s calculation using sample data

In table 6.10 information regarding respondent’s high school region, subjects and medium of instruction is presented. 288 respondents did matriculation from rural high school while 562 from urban high school. 244 respondents did matriculation with arts while 606 respondents with science subjects. Similarly, for 712 respondents, medium of instruction was Urdu while only for

157

138 respondents medium of instruction was English. This dimension is also used as a proxy for quality of education in primary data analysis.

305 respondents have diploma while 545 have not any kind of diploma similarly 350 respondents have on the job training while 500 respondents do not have any training. These two are again two important determinants of increase in earnings of an individual. Detail is presented in table 6.11.

Table: 6.11 Diploma and Training distribution

Diploma Training Yes No Yes No 305 545 350 500 (35.88 %) (64.12 %) (41.18 %) (58.82 %)

Source: Author’s calculation using sample data

6.4 Respondent’s Household Information

In this section household information of respondent is presented in detail. In table 6.12 detail of linguistic distribution is presented. 28.12 percent respondents speak urdu as mother language, 43.53 percent Saraiki as it is native language of Multan while 27.88 percent are Punjabi and 0.47 percent are other tribal languages.

Table: 6.12 Linguistic sample distribution

Mother Tongue Urdu Saraiki Punjabi Other 239 370 237 4 (28.12 %) (43.53 %) (27.88 %) (0.47 %)

Source: Author’s calculation using sample data

In table 6.13 information regarding parent’s qualification of respondent is presented. It shows that 319 respondents have illiterate mother while 143 have illiterate father. 235 respondents have primary qualified mother and 160 with primary qualified father. 4 respondents with middle, 158 with secondary, 57 with higher secondary, 56 with graduation, 18 with masters, 1 with M.phil and 1 with PhD qualified mother. Similarly 8 respondents have middle, 197 with secondary, 116 with higher secondary, 125 with graduation, 85 with masters, 7 with M.phil and 3 with PhD

158 qualified father and 6 have some other terminal degree. Parent’s qualification is a very strong determinant of child schooling and in long term their earning.

Table: 6.13 Parent’s qualification wise sample distribution

Qualification Mother Father Illiterate 319(37.53%) 143(16.82%) Primary 235(27.65%) 160(18.82%) Middle 4(0.47%) 8(0.94%) Secondary 158(18.59%) 197(23.18%) Higher secondary 57(6.71%) 116(13.65%) Graduation 56(6.59%) 125(14.71%) Masters 18(2.12%) 85(10%) Mphil 1(0.12%) 7(0.82%) Ph. D 1(0.12%) 3(0.35%) Other 1(0.12%) 6(0.71%)

Source: Author’s calculation using sample data

Table: 6.14 Residence wise sample distribution

Residence Own House Rented Govt. Accommodation Other 706 131 10 3 (83.06 %) (15.41 %) (1.18 %) (0.35 %)

Source: Author’s calculation using sample data

Table: 6.15 Asset and Happiness wise sample distribution

Assets Happiness Yes No Happy Unhappy 144 706 814 36 (16.94 %) (83.06 %) (95.7 %) (4.23 %)

Source: Author’s calculation using sample data

159

According to figures presented in table 6.14 and 6.15 83.06 percent respondents have their own house, 1.18 percent has government accommodation and 15.41 percent live in a rented house while 0.35 percent have some other arrangement like in someone else house. Only 16.94 percent respondents have assets like car, property or livestock while 83.06 percent respondents do not have any assets. It may be due to non reporting error that respondents hesitate to provide a clear description of their assets. 95.7 percent respondents report that they are happy with their life while only 4.23 percent were not happy. This non-monetary aspect of study is linked with the work of Sohan (2013).

6.5 Respondent’s Health Information

In this section health status and some related information of respondent is presented. Although in literature self-reported health status is considered a biased estimator of health but according to estimates of this study 92.23 percent respondents reported a good health status while only 7.76 have poor health status while health indicators on macro level show a different picture. Similarly, only 83 percent respondents report a major illness and 90.24 percent respondents do not have any major illness. 101 persons have some sort of physical disability while 749 persons in the sample do not have any physical disability that affects their earnings.

Table: 6.16 Health Status, Major illness and physical disability wise sample distribution

Health Status Major illness Disability Excellent Poor Yes No Yes No 784 66 83 767 101 749 (92.23%) (7.76%) (9.76%) (90.24%) (11.88%) (88.12%)

Source: Author’s calculation using sample data

If we talk about nature of the disability and major illness then in table 6.17 we can see that 1.52 percent respondent have heart disease, 0.12 percent with cancer, 1.29 with arthritis, 2.23 with diabetes, 2.23 with high blood pressure and 2.35 respondents with some other major illness. In case of disability we can see that 0.35 percent respondents feel difficulty in bending, 1.88 percent in walking, 0.35 percent in climbing stairs while 9.29 percent respondents reported some other disabilities that hinder their daily routine.

160

Table: 6.17 Nature of Major illness and disability wise sample distribution

Nature of No of Respondents Nature of No of Respondents Major illness Disability Heart Disease 13 (1.52%) Bending 3 (0.35 %) Cancer 1 (0.12%) Walking 16 (1.88 %) Arthritis 11 (1.29 %) Climbing Stairs 3 (0.35 %) Diabetes 19 (2.23 %) Other 79 (9.29 %) High Blood pressure 19 (2.23 %) ------Other 20 (2.35 %) ------

Source: Author’s calculation using sample data

Table: 6.18 Medical facilities, Source of medication and precautionary measures wise sample distribution

Availability of Source of Precautionary Medical facilities Medication Measures Yes No Private Public Yes No 672 178 587 263 113 737 (79.06%) (20.94 %) (69.06 %) (30.94 %) (13.29 %) (86.71 %)

Source: Author’s calculation using sample data

Table: 6.19 Healthy diet, hygienic water and clean environment wise sample distribution

Availability of Hygienic condition Availability of Healthy diet of water Clean environment Yes No Good Bad Yes No 701 149 744 106 587 263 (82.47 %) (17.53 %) (87.52 %) (12.47 %) (69.06 %) (30.94 %)

Source: Author’s calculation using sample data

From table 6.18 and 6.19, it is clear that among the randomly selected sample, 79.06 percent respondents have access to medical facilities while 20.94 percent do not have proper access. 69.06 percent respondents use private source of medication while only 30.94 percent approaches to public health institutions due to long distance or due to poor quality of services in public

161 institutions. Only 13.29 percent respondents use precautionary measures against fatal diseases while 86.71 percent do not spend on precautionary measures that prove to be a big source of incidence of harmful diseases. 82.47 percent respondents have access to a healthy diet while 17.53 percent do not have a healthy diet, 87.52 percent respondents have access to safe drinking water while 12.47 do not have any access and 69.06 percent respondents live in a clean environment while 30.94 percent respondents do not have this opportunity.

6.6 Conclusion

In this chapter, a detailed elementary analysis of sample data is presented that includes respondent’s personal information like region, religion, gender, marital status, the sector of employment, designation and working hours. In second section, qualification and skills information is presented like terminal degree, faculty, institution, thecountry and examination system. Similarly, information regarding high school region, subjects and medium of instruction is also presented in second section. Next section deals with household information like mother tongue, parent’s qualification, residence, assets and happiness level. In the last section, a clear picture of health status and many health determinants are discussed in detail. Information presented in this chapter is very important for the primary data analysis presented in next chapter in detail.

162

Chapter 7

PRIMARY DATA ESTIMATION OF RETURNS OF INVESTMENT TO EDUCATION AND HEALTH: EMPIRICAL EVIDENCE FROM DISTRICT MULTAN

7.1 Introduction

In this chapter, returns of investment to education and health are discussed in detail by using Mincer earning function (1974). In the first section returns of investment to education are discussed and disaggregated by gender, region and marital status. Similarly, returns to education investment are also discussed for different sectors of employment, professions and faculty. Returns to different levels of education are also discussed by using extended earning function, and some important determinants of education are also empirically tested. Returns to investment in the quality of education are also explored in first section. The aspects discussed in these results are consistent with the objectives of various five year education plans.

In second section, we presented returns of investment to health in district Multan. Anthropometric measures of health, self-stated measures of health and investment in some other factors are also discussed in the second section by using a sample of 850 wage earners from Multan district. All results estimated by using Minitab-16 software are presented in appendix 21 in actual form as estimated for ready reference.

7.2 Returns of Investment to Education in District Multan

In this section, a comprehensive analysis of the Mincer earning function is presented. The explanatory variables of the earning function comprise of total years of education, experience, and experience square and then complete function is disaggregated for gender, marital status and geographical region.. We also analyzed the role of some earning determinants like working hours, matriculation subjects and a proxy for the family background that is father’s qualification. Along with the quantitative aspects of education we also evaluated qualitative aspect of education on earning in the form of medium of instruction in school.

163

The regression results for full sample are presented in table 7.1, where basic earning function is estimated. All categories had expected sign according to priori expectations and all are highly significant. The coefficient of total years of education showed that one extra year of schooling results in 12.9 percent increase in earning. The coefficient of experience is positive according to expectations showed that earnings increases with education. The coefficient of experience square is negative as expected because it shows the concavity of earning function that implies earning increases with experience but at a diminishing rate.

Table: 7.1 Mincer earning function (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 7.7542 0.1207 64.25 0.000 TED 0.1214 0.007146 16.99 0.000* EXR 0.062964 0.006063 10.39 0.000* EXQ -0.0007901 0.0001545 -5.11 0.000* R- Square 0.37 R- Square Adjusted 0.37 F-Statistic 165.18 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively These results are consistent with the findings of psacharopoulos (2006) that is 12.8 percent increase in earning on the average year of schooling for Asian countries and 12.5 percent for intermediate countries (Demetriades & Psacharopoulos, 1997) 12.6 percent rate of return was estimated for for Phillipines during 1998 (Schady, 2000) and 12.2 percent for China during 1993 ( Hossain, 1997). In case of Pakistan, our estimated returns are slightly higher than other studies on returns of investment in education in Pakistan, as 10.5 percent returns were estimated for Pakistan by Awan and Hussain (2007), this is due to the regional differences and difference in employment opportunities.

In table 7.2 we analyzed the Mincer earning function of gender where we estimated rate of return to investment in education for both genders separately. In case of men the coefficient on years of schooling in 0.1152 that is positive and significant according to expectations. Similar is the case

164 with experience and experience square, both variables are significant and signs are according to expectations that earnings increases with experience but at diminishing rate.

Table: 7.2 Mincer earning function of gender

Dependent variable (log of monthly earnings) Explanatory variables Male Sub-Sample Female Sub-Sample Coefficient t- ratio Coefficient t- ratio 8.0129 6.6719 C 54.40 15.61 (0.1473) (0.4274) 0.1152* 0.17514* TED 14.99 6.82 (0.00769) (0.0256) 0.04850* 0.05462* EXR 6.00 2.52 (0.0080) (0.0216) -0.0005245* -0.0001982 EXQ -2.90 -0.26 (0.000181) (0.000750) R-Square 0.36 0.28 R-Square Adjusted 0.35 0.27 F-Statistic 92.72 27.64 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively In case of female, the coefficient on years of schooling is remarkably higher than the male that is 0.17514 and it is highly significant. The coefficient of experience showed that female earning also increases with experience. In this case the coefficient of the square of experience is not significant although we expect experience to be an important variable in determining earning but the results here show otherwise. This could be a reflection of wages for female being less tied to experience because women had less cumulative work experience than man as a result of break in their work histories owing to the demand of motherhood and housework traditionally assigned to them.

Thus, in our findings, overall returns of investment in education for the male is 12.2 percent while 19.1 percent returns were estimated for female, so the returns of investment in education

165 are higher for female. All of these returns are calculated as (eβ-1)*100. Almost similar results were observed in the estimations of various researchers like Awan and Hussain (2007) estimated 9.3 percent returns for the male and 18.1 percent return for the female in Pakistan.

Table: 7.2.1 Rate of returns of investment to education by gender

Gender Private Returns Male 12.2 Female 19.1

Source: Authors calculation by using computer software Minitab-16 Our results are consistent with international literature also like Siphambe (2000) estimated 12 percent returns for male and 18 percent for female in Botswana. Chiswick (1976) estimated 9.1 percent returns for male and 13 percent for female for less developed countries. Asadullah (2006) evaluated returns for Bangladesh and found 6.2 percent for male and 13.2 percent for female. Although female earnings are low as compared to male irrespective of same education, our results suggest that additional investment made in female education have higher marginal returns than male. Doughterty (2005) argued that returns for women are higher than men because education helps women to find employment outside the traditional low paying female occupation.

In table 7.3 we estimated returns of investment returns of investment in education for individuals located in urban and rural areas of Multan district separately. The coefficient of years of schooling for urban respondents is positive and highly significant, that showed one year increase in education for urban residents leads to 13.4 percent increase in earnings. Experience and experience square both are significant and had priori expected signs.

In case of rural respondents, the returns of education are slightly lower that is 11.9 percent. This is because of the expectations that people living in urban areas have more opportunities to exploit skills acquired through higher education than do those living in rural areas. These results are consistent with the findings of Warunsiri & Mcnown (2010), Asadullah (2006) and Khan & Toor (2003). This issue was also addressed in 4th five year plan (1970-75).

166

Table: 7.3 Mincer earning function of region

Dependent variable (log of monthly earnings) Explanatory variables Urban Sub-Sample Rural Sub-Sample Coefficient t- ratio Coefficient t- ratio 7.6409 7.9837 C 48.71 42.36 (0.1569) (0.1885) 0.1258* 0.1128* TED 13.20 10.64 (0.00953) (0.0106) 0.0718* 0.04554* EXR 9.26 4.71 (0.00776) (0.00966) -0.001009* -0.000362 EXQ -5.07 -1.49 (0.000199) (0.0002429) R-Square 0.37 0.36 R-Square Adjusted 0.37 0.36 F-Statistic 109.30 57.54 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively Table: 7.3.1 Rate of returns of investment to education by region

Region Private Returns Urban 13.4 Rural 11.9

Source: Authors calculation by using computer software Minitab-16 Table 7.4 represents the returns of investment in education disaggregated by marital status. All findings are significant and had expected signs. The coefficient of total years of education in case of married respondents is 0.115 while in case of unmarried respondents it is 0.130 that showed returns to investment in education for married respondents is 12.2 percent while for unmarried respondents it is 13.9 percent. The coefficient of experience in both cases showed that earnings increases with experience for both married and unmarried respondents.

167

These findings are consistent with the estimation of Warunsiri and Mcnown (2010) where returns for married persons were 11.2 percent while for unmarried persons returns are 12.6 percent.

Table: 7.4 Mincer earning function of marital status

Dependent variable (log of monthly earnings) Explanatory variables Married Sub-Sample Unmarried Sub-Sample Coefficient t- ratio Coefficient t- ratio 8.0129* 7.4844* C 54.40 28.19 (0.1473) (0.2655) 0.1152* 0.1303* TED 14.99 8.39 (0.007691) (0.0155) 0.04850* 0.09199* EXR 6.00 4.78 (0.0080) (0.01925) -0.0005245* -0.002034* EXQ -2.90 -2.31 (-0.00018) (0.00088) R-Square 0.36 0.20 R-Square Adjusted 0.35 0.20 F-Statistic 92.72 29.78 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively Table: 7.4.1 Rate of returns of investment to education by marital status

Marital status Private Returns Married 12.2 Unmarried 13.9

Source: Authors calculation by using computer software Minitab-16 Returns of investment in education are higher for unmarried workers because they have greater geographical and job mobility which allows them to take advantage of greater potential earnings afforded by higher levels of education.

168

Table 7.5 provides a summary of the findings of Mincer basic earning function for different sectors of employment. Here we have evaluated that individuals employed in private sector are getting higher returns to investment in education that is 14.6 percent, federal government employees are earning 13.3 percent returns on additional year of schooling, semi government sector provides 11.6 percent returns while returns to investment in education for provincial sector are lowest that is 9.9 percent.

Table: 7.5 Rate of return to education by sectors of employment

Sector of employment Private Returns Federal 13.3 Provincial 9.9 Semi government 11.6 Private 14.6

Source: Authors calculation by using computer software Minitab-16 These findings are consistent with the most famous findings of Psacharopoulos (1994) where returns to investment in private sector were 11.2 percent while for public sector returns were 9.0 percent only.

Table: 7.6 Rate of return to education by profession

Profession Private returns Education sector 15.0 Health sector 17.3 Industrial sector 15.3 Social service sector 9.3 Finance & legal sector 10.2 Other professions 18.7

Source: Authors calculation by using computer software Minitab-16 In table 7.6, returns to investment in education for employees in different professions are estimated. In our data set, we have estimated that returns to individuals having other professions like working in the private companies have higher returns that are 18.7 as compared to other professions. Individuals working in health sector on different posts have shown 17.5 percent

169 return on investment in education. Industrial sector workers showed a 15.3 percent return, 15.0 percent return for employees in education sector, 10.2 percent for the financial and legal sector like bankers and lawyers etc. similarly 9.3 percent return was estimated for employees in social service sector like individuals working in police department, railway, WAPDA etc. Analysis of profession was also done for Czech Republic by Klazar, sedmihradsky and vancurova (2001).

In table 7.7 returns to investment in different faculties was estimated. Individuals were categorized into different faculties according to their terminal degree subject. We have observed that physical science had highest returns that is 20.5 percent then engineering and technology had 16.7 percent returns, 12.9 percent returns were estimated for business science, 12.8 percent for arts and humanities, 10.1 percent for biology and medical science, 8.5 percent for arts or general group, 7.9 percent for general science and 6.6 percent return of investment in education in social science was estimated for Multan district.

Table: 7.7 Rate of return to education by faculty

Subject Private Returns Arts & Humanities 12.8 Social Science 6.6 Business science 12.9 Engineering & technology 16.7 Physical science 20.5 Biology & medical science 10.1 Agriculture & veterinary science 2.02 Science 7.9 Arts & general group 8.5

Source: Authors calculation by using computer software Minitab-16 Psacharopoulos (1994) estimated returns to investment in education by faculty and found engineering with highest 19.0 percent returns while in our study these are 16.7 percent. He estimated returns for various other faculties but categories are different in both studies. Returns to different faculties vary due to the job requirement in different countries and regions or due to the availability of employment opportunities in particular area.

170

7.2.1 Mincer Earning Function with Working Hours, Family Background and Matriculation Subjects

In this section, we again estimated basic Mincer earning function (1979) with some other explanatory variables which are found to be very significant in determining returns in various other studies discussed in literature review section. In order to incorporate the role of family background in determining returns to education, we use qualification of family head as proxy, therefore, we used father’s qualification in our analysis. Moreover, we also introduced two other explanatory variables which are working hours and matriculation subject as determinants of earning.

Table: 7.8 Mincer earning function with working hours, family background and matriculation subjects (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 7.0557 0.1341 52.63 0.000* TED 0.124 0.007191 17.29 0.000* WHR 0.16172 0.01972 8.20 0.000* FQL 0.03688 0.01047 3.52 0.000* MQL -0.00381 0.01212 -0.31 0.753 SBJ 0.14172 0.03402 4.17 0.000* EXR 0.062384 0.005825 10.71 0.000* EXQ -0.0006827 0.0001475 -4.63 0.000* R- Square 0.44 R- Square Adjusted 0.44 0.000 F- Statistic 96.25

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively In table 7.8 we can see that all variables are significant and signs are according to expectations. The coefficient working hours is positive and significant showed that earning increased by 0.161 units for every increase in working hours. Similarly, individuals with the better family background in the form of higher father’s qualification have 0.036 units higher earnings than

171 individuals with less qualified fathers. Matriculation with science subjects plays a more vital role in determining returns to investment in education as to arts subjects. In education policy 1998 government encouraged high quality research in arts.

These findings are also consistent with international literature like Castellanous and Psacharoloulos (1990) used log of hours worked and found it highly significant in determining income. Siphambe (2000) used education of head and hours worked in Mincer earning function. Gillani, Khan and Faridi (2013) used weakly working hours in Mincer earning function and find it significant in increasing returns. Johnson and Stafford (1973) also used father’s education as a determinant of earning.

Importance of father’s education is due to the fact that more educated parents are more likely to get better information about employment and therefore obtain better-paying jobs for their children. Similarly increase in working hours leads to increase in earning in our dataset according to the nature of jobs or sectors from where data is collected. Sometimes jobs that require long working hours are low paying jobs like security guards, but this is not true in our case because most of the information is collected from persons with higher qualification and doing some skill acquired jobs.

In table 7.9 we repeated same analysis for two categories of gender that is male and female and observed that in both cases these additional explanatory variables are playing very significant role in determining returns. In both cases returns to education in the form of earnings increases with higher working hours and better family background. Similarly matriculation with science subjects is also a key determinant of earning in case of gender.

We also explored the role of family background, working hours and matriculation subjects for rural and urban workers in table 7.10 and for married and unmarried workers in table 7.11 and obtained almost similar results that individuals with more working hours get higher returns as compared to low working hours. Similarly, better family background helps in the better schooling of children and very helpful in getting better jobs and matriculation that is supposed to be the base of higher education leads to higher returns for all categories if completed with science subjects.

172

Table: 7.9 Mincer earning function with working hours, family background and matriculation subjects (Gender wise sub sample)

Dependent variable (log of monthly earnings) Explanatory variables Male sub-sample Female sub-sample Coefficient t-ratio Coefficient t-ratio 7.4693* 6.0670* C 44.46 14.75 (0.1680) (0.4113) 0.1152* 0.1495* TED 14.45 6.11 (0.0079) (0.0244) 0.1062* 0.2943* WHR 4.55 4.48 (0.0233) (0.0656) 0.0412* 0.0407** FQL 4.31 1.79 (0.0122) (0.0227) 0.0028 -0.0495* MQL 2.59 -1.83 (0.0152) (0.0270) 0.0910* 0.2814* SBJ 6.73 2.74 (0.032) (0.1029) 0.0522* 0.0592* EXR 2.59 3.00 (0.0077) (0.0197) -0.0005312* -0.0002922 EXQ -3.06 -0.42 (0.0001734) (0.000689) R- Square 0.42 0.42 R- Square Adjusted 0.41 0.40 F- Statistic 60.24 24.38 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively

173

Table: 7.10 Mincer earning function with working hours, family background and matriculation subjects (Region wise sub sample)

Dependent variable (log of monthly earnings) Explanatory variables Urban sub sample Rural sub sample Coefficient t-ratio Coefficient t-ratio 6.9630* 7.2559* C 40.47 33.94 (0.1721) (0.2138) 0.1249* 0.1191* TED 12.68 11.17 (0.0098) (0.0106) 0.1749* 0.1393* WHR 6.74 4.68 (0.0259) (0.0298) 0.0353* 0.0296** FQL 2.93 1.92 (0.0353) (0.0154) -0.0122 0.0304 MQL -0.81 1.39 (0.0151) (0,0219) 0.1409* 0.1513* SBJ 3.41 2.44 (0.0413) (0.0621) 0.0681* 0.0508* EXR 9.18 5.40 (0.0074) (0.0094) -0.0008450* -0.000364 EXQ -4.44 -1.57 (0.000190) (0.000232) R- Square 0.45 0.44 R- Square Adjusted 0.44 0.43 F- Statistic 73.41 39.46 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively

174

Table: 7.11 Mincer earning function with working hours, family background and matriculation subjects (Marital status wise sub sample)

Dependent variable (log of monthly earnings) Explanatory variables Married sub sample Unmarried sub sample Coefficient t-ratio Coefficient t-ratio 7.4693* 6.7846* C 44.46 25.67 (0.1680) (0.2643) 0.1152* 0.1228* TED 14.45 8.17 (0.0079) (0.0150) 0.1062* 0.2247* WHR 4.55 6.54 (0.0233) (0.0343) 0.0425* 0.0279** FQL 4.31 1.79 (0.00989) (0.0156) 0.0028 -0.0060 MQL 0.19 -0.31 (0.0152) (0.0194) 0.09076* 0.2616* SBJ 2.59 3.36 (0.0351) (0.0778) 0.0522* 0.0791* EXR 6.73 4.43 (0.00775) (0.0178) -0.0005312* -0.001687** EXQ -3.06 -2.06 (0.000173) (0.000818) R- Square 0.42 0.33 R- Square Adjusted 0.41 0.32 F- Statistic 60.24 28.11 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively 7.2.2 Extended Mincer Earning Function

In table 7.12 returns to different levels of education are presented where the primary level of education is considered as a reference category to avoid dummy variable trap. According to these

175 results returns to primary level of education are positive and significant and returns to middle level are positive and insignificant also. Similarly, returns to secondary and higher secondary level of education is positive but insignificant. Returns to investment in graduation and university level are positive and highly significant. So we can say that as the level of education increases returns also increases. Importance of investment in primary secondary levels is highlighted in almost all educational policies.

Table 7.12 Extended Mincer Earning Function (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 8.9243 0.1646 54.22 0.000* MDL 0.297 0.2037 1.46 0.144 SEC 0.062 0.1721 0.36 0.717 HSE 0.226 0.1699 1.34 0.182 GRD 0.439 0.1620 2.71 0.007* UNI 0.863 0.1595 5.41 0.000* EXR 0.066 0.0062 10.60 0.000*

EXQ -0.000855 0.00016 -5.34 0.000* R- Square 0.35 R- Square Adjusted 0.34 F-Statistic 64.11 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively By following the methodology adopted by Mincer (1958) and many other researchers we calculated returns for different levels of education by taking difference of higher level education coefficient and lower level coefficient and dividing by the total number of years of education and expressing the result as percentage. We estimated that returns to secondary education are highest as 17.95 percent due to favorable market situation and high absorbing power of market for secondary qualified workers while returns to higher secondary education are 8.2 percent only. On the other hand returns to graduation and university education are more or less same as 10.6 percent. Thus government should increase investment at secondary level and all other levels also.

176

7.2.3 Mincer Earning Function with Quality of Education

In this study we estimated returns to investment in education by focusing our attention on quantitative aspect of total years of education acquired, we did a detailed quantitative analysis that is presented in previous tables. We can also estimate returns to investment in the quality of education.

For this purpose, we used basic earning function (1974) with the quality of education variable. Different studies used different variables to capture the effect of quality of education like card and Krueger (1992) examined the effect of school quality on returns to education by using student-teacher ratio, average term length and relative pay of teachers as a proxy for quality of education. Awan and Hussain (2007) used private schooling as a proxy for quality of education and estimated that private schooling showed higher returns as compared to public schools. Khan and Toor (2003) and Jamal, Toor and Khan (2003) used private schooling and English as a medium of instruction as proxy for quality of education and examined a significant role of private schooling and English medium of instruction in increasing earnings.

Table: 7.13 Mincer Earning Function with medium of instruction (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values Constant 7.7356 0.1210 63.92 0.000 TED 0.12122 0.007138 16.98 0.000* MDU 0.06634 0.03852 1.72 0.085** EXR 0.06422 0.006100 10.53 0.000* EXQ -0.0008159 0.0001551 -5.26 0.000* R- Square 0.37 R- Square Adjusted 0.37 F-Statistic 124.92 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively In our study, we used English as a medium of instruction in school as a proxy for quality of education. In basic earning function we incorporated another explanatory variable that is medium

177 and found that it is highly significant and had a positive sign which shows that English medium plays a very important role in determining returns to investment in education and earning increases when we get the education from an English medium school.

In table 7.14 we estimated basic earning function separately for English medium schooling and Urdu medium schooling and found that all variables are significant and signs are according to priori expectations. In both cases, returns increase the level of education and experience, thus overall returns to investment in English medium schooling are 16.9 percent while for Urdu medium schooling returns are 12.6 percent. Thus, English language is a significant indicator of earning differential in Pakistan’s labor market.

Table: 7.14 Mincer Earning Function (medium of instruction wise sub sample)

Dependent variable (log of monthly earnings) Explanatory variables English medium Sub-Sample Urdu medium Sub-Sample Coefficient t- ratio Coefficient t- ratio Constant 7.0600 7.7912 11.84 66.17 (0.5965) (0.1177)

0.15617* 0.119030* TED 4.50 17.10 (0.03468) (0.006963)

EXR 0.10492* 0.06034* 4.19 9.83 (0.02501) (0.006136) -0.0017935* -0.0007278* EXQ -2.52 -4.74 (0.0007127) (0.0001535) R-Square 0.26 0.41 R-Square Adjusted 0.25 0.41 F-Statistic 16.12 162.77 Probability 0.000 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively

178

7.3 Returns of Investment to Health in District Multan

In this section returns of investment to health in Multan district are presented. There are several measures of investment in health but anthropometric measures like height and weight for height are considered as much better among all others. According to Schultz (2002) height is recognized as a long run outcome of investment in the health of an individual. It is initially determined by the genetic capacities or genotype of the individual but with the passage of time it is also affected by nutritional intake, exposure and treatment of diseases and environmental burden on individual can positively or negatively affect their genetic capacity. Hausman (1978) also discussed the positive and significant role of height as health investment in a wage function. Although this health investment variable is difficult to quantify accurately, therefore, less empirical literature is available about returns of health investment especially in Pakistan.

Along with height, weight is also potentially related to productivity therefore it is also preferable to use weight given height like body mass index in health investment model. BMI is related to maximum physical capacity independent of energy intake. Thus energy can be stored in body and expanded when needed. Height is considered as a long run measure of nutritional intake while BMI is considered as a short run measure of health status, nutritional status and work capacity (Komlos, 1994; and Steckel, 1995).

In Table 7.15 these anthropometric measures are used in the Mincer earning function to see their impact on earnings. Height is measured in meters while body mass index is calculated by dividing weight in kg with height in meter square. These results explore that returns to investment in education are almost same at 12 percent as stated in the section of returns to investment in education and experience also plays a positive and significant role in the determination of earnings. Similarly, it shows the concavity of earning function as the experience square term is negative and significant.

In case of height, we can see that the affect of height on earnings is positive and significant that 1 meter increase in height leads to 27.2 percent increase in earnings of full sample including male and female. In case of body mass index results are also positive and significant and showed that 1 unit increase in BMI leads to 1.3 percent increase in earnings of full sample.

179

Table: 7.15 Mincer Earning Function with Anthropometric Measures of Health (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 6.20 0.343 18.07 0.000* TED 0.119 0.007 16.84 0.000* EXR 0.054 0.006 8.86 0.000* EXQ -0.00063 0.0001 -4.10 0.000* HET 0.241 0.052 4.63 0.000* BMI 0.013 0.004 3.11 0.002* R- Square 0.38 R- Square Adjusted 0.38 F-Statistic 104.69 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively These returns of investment in health in the form of height as long run investment measure and weight as a short run measure of investment are disaggregated by gender and results are presented in table 7.16. From these results, we can observe that returns to investment in education are same as described in above section that female has higher returns to education as compared to male while experience is not related to the earnings of the female. These findings are discussed above in the same way.

In case of anthropometric measures, we can see that short run measure of health investment that is BMI is positively linked with earnings in both cases but plays a significant role for men because it is a measure of strength and power and male activities require more power or strength as compared to female especially for less educated men. While in case of height we can see that again it is positively linked with earnings in both cases and plays a significant role in the income determination of both male and female. Thus according to our findings, a one unit increase in height leads to 5.1 percent increase in height for male and 3.04 percent for female. Similarly, a one unit increase in BMI leads to 0.92 percent increase in earnings for male and 0.2 in case of female.

180

Table: 7.16 Mincer Earning Function with Anthropometric Measures of Health (gender wise sub sample)

Dependent variable (log of monthly earnings) Explanatory variables Male Sub-Sample Female Sub-Sample Coefficient t- ratio Coefficient t- ratio 7.57 5.240 C 21.9 5.18 (0.345) (1.012) 0.117* 0.1710* TED 18.45 6.45 (0.0063) (0.0265) 0.0415* 0.051* EXR 6.95 2.36 (0.0059) (0.021) -0.00043* -0.000093 EXQ -2.98 -0.12 (0.000144) (0.00075) 0.0518* 0.0376* HET 10.36 2.29 (0.005) (0.0164) 0.0092* 0.0024 BMI 2.23 0.22 (0.004) (0.011) R-Square 0.42 0.30 R-Square Adjusted 0.41 0.28 F-Statistic 92.54 17.16 Probability 0.000 0.000 Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively Thomas and Strauss (1996) estimated the effect of height and education on wages and explored that one percent increase in height is associated with a three percent increase in wages for male and two percent for female while in our case it is five percent and one percent respectively. According to Schultz and Tansel (1997) and Savedoff and Schultz (2000) an increment of 1cm in height leads to 6.8 percent increase in wages in Ghana but not in Cote d’Ivoire while a unit gain in BMI is associated with a 9 percent increase in wages for men in both countries but for women it is 15 percent in Cote d’Ivoire and 7 percent in Ghana. Our findings are also consistent with Thomas and Strauss (1997) that taller men and women earn more even after controlling for

181 education and other dimensions of health. BMI has a significant effect on wages of men but not of women.

In table 7.17, we discussed about the investment of individual in nutritional factors like better and healthy food and pure drinking water and explored that these two factors are highly significant and positively related to earnings when they are treated as health investment. Every 1 unit increase in expenditure on food like wheat, pulses, milk etc leads to 0.000037 units increase in earnings while availability of pure drinking water increase earnings by 0.120 units. According to Mushkin (1962) investment in pure drinking water and sanitation facilities controls the incidence of serious diseases that reduce death rate and improves health condition of a person so that he/she can actively participate and earnings increases. These factors are also discussed in SAP 1993-94.

Table: 7.17 Mincer Earning Function for Health Investment (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 7.683 0.129 59.45 0.000 TED 0.116 0.007 16.19 0.000* EXR 0.061 0.006 10.03 0.000* EXQ -0.0007 0.00015 -4.96 0.000* EXN 0.000037 0.0000097 3.81 0.000*

HYW 0.120 0.055 2.18 0.029* R- Square 0.38 R- Square Adjusted 0.37 F-Statistic 103.01 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively Although mortality is a good indicator of returns of investment in health but it is difficult to introduce in individual wage function because dead are not in the sample of wage earners. Therefore in a survey research according to Schultz (2003) self-reported morbidity is a good measure of health status, but it may suffer from reporting bias for example self-reported hypertension by a respondent might differ from medical reports.

182

Another measure of health status is the limitation on working ability of the respondent in the form of disability to do any physical activity like walking, bending and climbing upstairs etc. it may reduce the respondent utility of engaging in physical capital (Strauss et al. 1995). In this study, both of these self-reported variables like disability and self-reported health status are included in health investment model. According to Schultz and Tansel (1997) and Sawdoff and Schultz (2000) in Cote d’Ivoire and Ghana more disability per month is expected to reduce worker’s wage by 10 percent. According to the findings of this study, we can explore that as the health status of a person improves his earnings increase as he becomes able to do work more efficiently thus productivity increases but in case of disability we can see that as a person becomes more disable to do an activity his or her earnings fall as he cannot perform all duties efficiently but this is statistically insignificant in this study.

Table: 7.18 Mincer Earning Function With Self-Reported Health Indicators (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 7.871 0.126 62.14 0.000 TED 0.121 0.007 16.86 0.000* EXR 0.062 0.006 10.19 0.000* EXQ -0.00075 0.00015 -4.85 0.000* HLS 0.045 0.0198 2.29 0.022* DSB -0.041 0.056 -0.72 0.471 R- Square 0.37 R- Square Adjusted 0.36 F-Statistic 98.93 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively According to Grossman (1972), health investment is produced by household production function that includes exercise, recreation, housing, diet, medical care and environmental factors as inputs. Health is demanded by the consumer for two reasons as a consumption good where no of disability days creates disutility and as an investment good that shows more time available for economic activities that result in higher monetary returns. Hadley and Reschovsky (2012)

183 analyzed the role of medical spending in health improvement and found a positive and significant role.

Table: 7.19 Mincer Earning Function for some determinants of health (Full sample)

Dependent variable (log of monthly earnings) Explanatory variables Coefficient Standard deviation t- ratio P- Values C 7.617 0.125 60.73 0.000 TED 0.115 0.0071 16.20 0.000* EXR 0.061 0.0059 10.28 0.000* EXQ -0.00076 0.00015 -5.03 0.000* MDF 0.093 0.044 2.09 0.037* PCU 0.072 0.053 1.35 0.179 EXE 0.0013 0.00048 2.82 0.005* ENV 0.156 0.039 3.99 0.000* R- Square 0.39 R- Square Adjusted 0.38 F-Statistic 77.18 0.000

Source: Authors calculation by using computer software Minitab-16

*, ** shows level of significance at 5 percent and 10 percent respectively In table 7.19 some health inputs are used in earning function, and we explored a positive and significant role of the availability of medical facilities in the locality as availability of medical facilities increases earnings by 0.093 units. Similarly, time spent on daily exercise also has a positive and significant impact on earnings that is 0.0013 and 0.156 respectively. Similarly, expenditure on precautionary measures also has a positive effect on earnings although it is not statistically significant. Health aspects which are empirically tested in this section are also discussed in Alma Ata declaration 1978 and various health policies presented in various successive years and policies related to these aspects are mentioned earlier.

7.4 Conclusion

This chapter is based on primary data analysis by using a sample of 850 wage earners from District Multan. According to the findings, it is clear that increase in education leads to increase

184 in earnings for wage earners for both cases when education is treated as a continuous and as a discrete variable. Moreover, returns are higher for female as compared to male, and higher for urban workers as compared to rural workers. Similarly, returns to education are higher for unmarried workers as compared to married workers. Similarly, returns for individuals with English as a medium of instruction are higher as compared to Urdu medium of instruction. Some other factors like working hours, matriculation subjects and parents qualification are also helpful in the evaluation of returns to education.

In the second section, returns to investment in health are presented. From these results, it is clear that one unit increase in health investment in childhood or young age leads to increase in earnings. Similarly, some other factors like pure water, clean environment, and precautionary measures. Medical facilities and investment in nutritious food are also helpful in improving returns of investment.

185

Chapter 8 RETURNS OF INVESTMENT TO EDUCATION AND HEALTH IN PAKISTAN: SECONDARY DATA ANALYSIS

8.1 Introduction

This chapter is based on secondary data analysis of public investment in education and health services in Pakistan and impact of this investment on economic growth is also empirically explored in the ending section of this chapter. This chapter summarizes results of these economic and econometric models designed to attain these objectives by using data from Pakistan economy for the period of 1972-2016. Initially, preliminary data analysis is conducted where firstly we discussed some important descriptive statistic related to these variables and applied Augmented Dickey Fuller (ADF) test for checking stationarity of the variables. Johansen cointegration is applied after getting ADF results for the long run relationship among these variables, and error correction model is used to explore the existence of short run disequilibrium in the economy. Finally, we have used granger causality to determine the causal relationship among the variables in all three macroeconomic models. Actual E-views results of these three models are also presented in appendix 22 for ready reference.

8.2 Descriptive Analysis

This part of the study is based on time series data analysis of selected variables from Pakistan for the period of 1972 to 2016, and whole data set is based on 43 observations. In table 8.1 descriptive statistic of selected variables are presented. The average infant mortality rate is 96.89 in our data set with 22.80 variations while average health expenditure is 0.72 percent with 0.195 variations. Similarly mean value of expenditure on primary, secondary and tertiary education are 37.56, 29.50 and 20.05 respectively with standard deviation 7.13, 4.57 and 4.65 respectively. On the other hand average population per doctor, total labor force and gross fixed capital formation is 3736.9, 36.97 and 581705.2 respectively while the values of standard deviation are 3797.7, 13.33 and 654131.8 respectively. The mean value of literacy rate, fertility rate, unemployment rate, inflation rate and GDP growth rate are 41.82, 5.33, 4.93, 58.84 and 4.81 respectively and

186 the values with standard deviation 13.52, 1.43, 1.69, 56.53 and 2.11 respectively. On average the education enrollment index is 0.138 with 0.029 standard deviation from mean.

Table: 8.1 Descriptive Statistics

Variables Mean Std.Deviation Skewness Kurtosis LTR 41.82 13.52 -0.04 1.38 EXP 37.56 7.13 0.59 2.05 EXS 29.50 4.57 0.27 2.23 EXT 20.05 4.65 0.52 2.66 EEI 0.138 0.029 -0.09 1.49 IMR 96.89 22.80 0.12 1.60 FTR 5.33 1.43 -0.21 1.40 HEX 0.72 0.19 0.11 3.94 PPD 3736.9 3797.7 1.42 3.50 GCF 581705.2 654131.8 1.0 2.71 GDP 4.81 2.11 0.22 2.66 TLF 36.97 13.33 0.61 2.31 UEM 4.936 1.697 -0.012 2.23 INF 58.84 56.53 1.306 3.58 Source: Author’s own calculations based on E-Views 9

Skewness means lack of symmetry of a distribution, and it may be positive or negative. When a distribution is not symmetrical, then it is asymmetrical or skewed based on its numerical estimates. When numerical value of Skewness is greater than zero then it is positively skewed, otherwise negatively skewed for less than zero numerical estimates. Based on the calculations presented in table 8.1 we can say that almost all variables are slightly skewed while only literacy rate, fertility rate, education enrollment index and unemployment rate are skewed negatively and all other variables are skewed positively. Kurtosis is used to indicate the peakness of the distribution. It may be mesokurtic that is usual normal distribution, leptokurtic with a sharp peak having more values around its peak or platykurtic that is comparatively flatter. If the numerical estimate of kurtosis is equal to three, then distribution is mesokurtic, leptokurtic for more than three and platykurtic for less than three. According to this rule of thumb, we can say that health

187 expenditure as a percentage of GDP, population per doctor and inflation rate are leptokurtic while all other variables are platykurtic and no variable is mesokurtic.

8.3 Unit Root Test Table: 8.2 ADF Test Results

Level 1st Difference Variable Conclusion Intercept Trend and Intercept Intercept Trend and Intercept -0.6325 -0.5337 -4.831 -4.788 LTR I(1) (0.852) (0.978) (0.0003) (0.002) -1.881 -1.775 -7.592 -7.618 EXP I(1) (0.337) (0.699) (0.000) (0.000) -1.973 -1.876 -4.442 -9.469 EXS I(1) (0.297) (0.649) (0.001) (0.000) -1.412 -1.742 -10.516 -10.388 EXT I(1) (0.567) (0.715) (0.000) (0.000) -0.658 -2.689 -6.202 -6.133 EEI I(1) (0.846) (0.245) (0.000) (0.000) -0.864 -3.051 -7.179 -7.122 IMR I(1) (0.790) (0.131) (0.000) (0.000) 0.666 -2.103 -4.977 -5.032 FTR I(1) (0.990) (0.529) (0.0002) (0.001) -2.524 -2.976 -5.565 -5.455 HEX I(1) (0.116) (0.150) (0.000) (0.000) -0.683 -2.622 -4.709 -4.711 GCF I(1) (0.840) (0.2727) (0.000) (0.002) 2.545 -0.032 -5.373 -5.900 GDP I(1) (1.000) (0.994) (0.000) (0.000) -1.605 -0.897 -9.201 -9.440 PPD I(1) (0.471) (0.947) (0.000) (0.000) 2.875 0.272 -6.744 -7.806 TLF I(1) (1.000) (0.997) (0.000) (0.000) -2.124 -2.532 -7.891 -7.946 UEM I(1) (0.236) (0.312) (0.000) (0.000) 3.258 -1.868 -1.988 -3.349 INF I(1) (1.000) (0.653) (0.290) (0.07) Source: Author’s own calculations based on E-Views 9

Augmented Dickey Fuller (ADF) test is used to confirm whether data series is stationary or not. If probability distribution remains unchanged with the passage of time, then the process is called

188 stationary which means data generation process does not change. For the application of ADF test, the Null hypothesis of unit root presence is used against an alternative hypothesis of stationarity. Results of Augmented Dickey Fuller test are presented in table 8.2.

We checked all variables at level once with intercept and then with trend and intercept and found that at level all variables are not stationary and null hypothesis of unit root cannot be rejected. All variables are again checked at first difference once with intercept and then with trend and intercept and found that null hypothesis of unit root is rejected at first difference and all variables are I (1). These results are also verifies by using unit root test by Phillips and Perron and found similar decision about stationarity of variables.

These calculated statistics of the test show that all variables are integrated of the same order one, I (1) therefore, for the estimation of these investment models; it is most suitable to apply Johansen Cointegration technique and vector error correction model if cointegration exist among the variables.

8.4 Returns of Public Investment to Health in Pakistan

In this section, we discuss outcome or return associated with macro level investment in the health sector by Pakistan. Four steps are involved in testing the Cointegration. The order of stationarity is determined in first step for all the variables used in the model. As we can see from table 8.2 that all variables used in health investment model are stationary at first difference, for example infant mortality rate, fertility rate, literacy rate, number of persons per doctor and health expenditures as percentage of GDP all are I (1). Therefore, it is possible to determine Cointegration among these variables by using Johansen Cointegration technique.

8.4.1 Lag Length Selection

Table: 8.3 Lag Order Selection Criterions

Lag LogL LR FPE AIC SC HQ 0 -631.95 NA 2579750 28.952 29.155 29.027 1 -344.62 496.30* 7.24523* 17.028* 18.244* 17.479*

*Indicates lag order selection by the criterion like LR, FPE,AIC, SC and HQ

189

After carefully observing stationarity results of all variables next step is the appropriate lag length selection. For this purpose, we use various lag order selection criterions and results are presented in table 8.3.There are several information criterion which is helpful for the selection of appropriate lag length, for example, Schwarz information criteria, Akaike information criteria, and Hannan - Quinn information criteria. With the help of these criteria, we determine lag one suitable for this model.

8.4.2 Unrestricted Cointegration Rank Test

After the selection of lag length next step is the determination of the number of Cointegration vectors. Both Eigen value statistic and trace statistics are used in this study and results are formulated in table 8.4 and 8.5.

Table: 8.4 Results of Trace Statistic

Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob. ** None * 0.534 94.089 69.818 0.000 At most 1 * 0.466 61.167 47.856 0.001 At most 2 * 0.307 34.121 29.797 0.014 At most 3 * 0.235 18.301 15.494 0.018 At most 4 * 0.144 6.7334 3.8414 0.009

Trace test shows five cointegratingeqn(s) at the 0.05 level * shows that hypothesis is rejected at the 0.05 level ** p-value of MacKinnon-Haug-Michelis (1999)

Table: 8.5 Results of Maximum Eigenvalue

Hypothesized No. of Max- Eigen 0.05 Critical Eigenvalue Prob.** CE(s) Statistic Value None* 0.534 32.92 33.876 0.064 At most 1* 0.466 27.04 27.584 0.058 At most 2 0.307 15.82 21.131 0.235 At most 3 0.235 11.56 14.264 0.128 At most 4* 0.144 6.733 3.8414 0.009

Max-eigenvalue test shows no cointegration at the 0.10 level * indicates that hypothes is rejected at the 0.10 level ** p-value of MacKinnon-Haug-Michelis (1999)

190

According to trace statistic, there are five Cointegration vectors and according to Eigen value statistic, there are three Cointegration equations. It means that with the help of these two statistics, null hypothesis of no Cointegration can be rejected and we accept the alternative hypothesis of the high association between dependent and independent variables used in this study.

8.4.3 Johansen Co-integration (Long run Estimates)

In table 8.6 the long run estimates of Johansen Cointegration technique are presented where the first column represents explanatory variables. Values of coefficients, standard error and t-statistic are represented in column 2nd, 3rd and 4th respectively. Infant mortality rate is used as a dependent variable it generally reflects the level of mortality and the effectiveness of preventive care and the attention paid to maternal and child health while health expenditure as a percentage of GDP, population per doctor, fertility rate and literacy rate are used as dependent variable in this health investment model.

Table: 8.6 Normalized Cointegration Coefficients

Variable Coefficient Std.Error t-Statistic C 158.35 -- -- FTR 1.4369 3.455 0.415 HEX -11.839 4.009 -2.953 LTR -1.5444 0.435 -3.546 PPD 0.0010 0.000 2.983

Source: Author’s own calculations based on E-Views 9

Public investment in health sector can be represented as health expenditures as a percentage of GDP, while the reduction in infant mortality rate is the return of this investment. According to normalized cointegration coefficients, health expenditures are negatively and significantly affects infant mortality rate. So we can say that one percent raise in health expenditures decreases infant mortality rate by 11.8 points. Barenberget et.al. (2015) also found negative association between infant mortality rate and the health expenditure. The results are also consistent with the findings of Farahani, et.al, (2010) and Cremieux (1999) explored that greater alcohol and tobacco

191 consumption resulted in a negative impact on life expectancy and infant mortality rate because such type of consumption affects lungs and cause heart attack.

Population per doctor is used as a proxy for quality of provision of health services. As we know that Pakistan is an under developed country where the population is very high and provision of health services are inadequate and moreover unequally distributed therefore health status of individuals is not so good. This variable shows significant and positive relation with infant mortality rate because increase in population per doctor shows poor health services and leads to an increase in infant mortality rate. Frankenberg (1995) also explored that increases in the availability of health care decrease mortality. Within a village an increase of one maternity clinic decreases the odds of death of an infant with access to that clinic by about 15 per cent, relative to the infant born before the clinic existed. An additional doctor decreases an infant’s odds of death by around 1.7 per cent while increase in number of patients per doctor is a great threat for human life. Gulliford (2002) also highlighted the availability of general physicians as an indicator of better health services.

Total fertility rate is positively affecting infant mortality rate because the increase in fertility rate lowers the health of mothers and low space births harm the survival of new born babies therefore increase in fertility rate is threatening for infants survival. It is generally considered by Yamada (1985) that increase in fertility ratio leads to increase in mortality rate. These results are also consistent with the findings of Abbas (2010).

The literacy rate is used as a proxy for general attitude towards education or awareness. As people become more aware and educated, then they can adopt proper health safety measures and controls various infections. This study also shows that literacy rate is negatively and significantly affecting infant mortality rate as one unit increase in literacy rate leads to 1.54 units decrease in infant mortality rate. Wheatley (2015) explored negative impact of literacy rate on infant mortality rate. The results are also consistent with the findings of Gordon (2009), Shandraet.al.(2011), Oloo (2005), Boehmar and Williamson (1996) and Hanmeret.al. (2003). These studies explained that as the literacy rate of parents especially mother’s increases the chances of her being knowledgeable about birth control options and reproductive health. Moreover, increase in the literacy rate also increases the chances of providing the best care to the children so the survival rate of children increases.

192

8.4.4 Error Correction Model (Short run Estimates)

Error correction process also takes place if long run relationship exists between the variables. This process indicates the adjustment speed towards the equilibrium in long run after short run shocks. In table 8.7 the findings of error correction model are presented.

Table: 8.7 Error Correction Model

Variable Coefficient Std.Error t-Statistic CointEq1 -0.523 0.177 -2.950 D(IMR(-1)) 0.064 0.196 0.327 D(FTR(-1)) -1.523 4.116 -0.370 D(HEX(-1)) -7.085 4.680 -1.513 D(LTR(-1)) 0.340 0.801 0.424 D(PPD(-1)) 0.003 0.001 1.796 C -0.996 0.949 -1.049

Source: Author’s own calculations based on E-Views 9

The most important thing in these results is the speed of adjustment term that shows an economy’s speed of convergence towards equilibrium in long run. In speed of adjustment term, a negative sign indicates that economy will converge towards equilibrium in the long run after obtaining 52 percent annually adjustment rate in the short run and the coefficient has statistically significant value.

Other estimated coefficients showed that in short run infant mortality rate of last year is found to be positively related with infant mortality rate of the current year 2016. Similarly, in the short run population per doctor and literacy rate are positively related with infant mortality rate while health expenditure as a percentage of GDP and fertility rate has negative association. These results reveal that all explanatory variables are insignificant in the short run because health investment is a long run phenomenon, therefore, these health related variables are not significantly affecting infant mortality rate in the short run. After observing long run and short run adjustments various diagnostic tests were performed to check the stability of the model. This model is free from serial correlation and heteroskedasticity and graph of AR roots verified that

193 parameters are stable. Eviews results of these diagnostic tests are presented in appendix 23 in detail.

8.4.5 Granger Causality

In table 8.8 the estimated results of granger causality are presented and interpreted. These findings are quite interesting and used for policy relevance. These results found the bidirectional relationship between fertility rate and expenditures on health as a percentage of GDP; literacy rate and fertility rate.

Table: 8.8 Results of Granger Causality

Null Hypothesis Obs F-Statistic Prob. FTR does not Granger Cause IMR 0.855 0.433 43 IMR does not Granger Cause FTR 6.533 0.003 HEX does not Granger Cause IMR 1.831 0.174 43 IMR does not Granger Cause HEX 2.153 0.130 LTR does not Granger Cause IMR 2.688 0.080 43 IMR does not Granger Cause LTR 0.991 0.380 PPD does not Granger Cause IMR 0.631 0.537 43 IMR does not Granger Cause PPD 0.017 0.982 HEX does not Granger Cause FTR 6.135 0.004 43 FTR does not Granger Cause HEX 2.713 0.079 LTR does not Granger Cause FTR 6.174 0.004 43 FTR does not Granger Cause LTR 7.527 0.001 PPD does not Granger Cause FTR 6.119 0.005 43 FTR does not Granger Cause PPD 0.426 0.655 LTR does not Granger Cause HEX 2.284 0.115 43 HEX does not Granger Cause LTR 5.553 0.007 PPD does not Granger Cause HEX 0.705 0.500 43 HEX does not Granger Cause PPD 0.995 0.379 PPD does not Granger Cause LTR 8.633 0.000 43 LTR does not Granger Cause PPD 0.461 0.634 Source: Authors calculations based on computer software, E-Views 9

On the other hand, unidirectional relationship is found between infant mortality rate and fertility rate; literacy rate and infant mortality rate; population per doctor and fertility rate; health expenditure and literacy rate; population per doctor and literacy rate. There is no causality between fertility rate and infant mortality rate; health expenditure and mortality rate; literacy rate

194 and mortality rate; population per doctor and mortality rate; population per doctor and health expenditure; literacy rate and population per doctor.

8.5 Returns of Public Investment to Education in Pakistan

In this section we will discuss about the outcomes or returns associated with investment in education in Pakistan. Where public expenditure on primary, secondary and tertiary education is taken as investment and increase in education enrollment index is treated as return of this investment. According to the steps involved in the Cointegration process as stated earlier first step is to determine the level of stationarity and described in table 8.2 it is clear that all variables used in education investment model like education enrollment index, literacy rate, primary, secondary and tertiary education expenditure and inflation rate all are stationary at first difference, therefore, we can apply Johansen Cointegration Technique.

8.5.1 Lag Length Selection

Table: 8.9 Lag Order Selection Criterions

Lag LogL LR FPE AIC SC HQ 0 -557.55 NA 18189.13 26.835 27.084 26.926 1 -278.75 464.66 0.17648 15.273 17.011* 15.910 2 -240.32 53.064 0.17500 15.158 18.385 16.341 3 -179.45 66.673* 0.07209* 13.973* 18.690 15.702*

*Indicates lag order selection by the criterion like LR, FPE,AIC, SC and HQ

After we came to know about the ADF test results next step is to determine lag length by using suitable information criterion presented in table 8.9 we determine lag three as an appropriate lag length for this model.

8.5.2Unrestricted Cointegration Rank Test

After careful selection of lag length, we proceed with the determination of number of cointegration vectors.

195

Table: 8.10 Results of Trace Statistic

Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob. ** None * 0.9333 235.49 95.753 0.000 At most 1 * 0.6672 124.47 69.818 0.000 At most 2 * 0.6398 79.356 47.856 0.000 At most 3 * 0.4432 37.485 29.797 0.005 At most 4 0.2689 13.475 15.494 0.098 At most 5 0.0153 0.6339 3.8414 0.425

Trace test shows four cointegratingeqn(s) at the 0.05 level * shows that hypothesis is rejected at the 0.05 level ** p-value of MacKinnon-Haug-Michelis (1999) Table: 8.11 Results of Maximum Eigenvalue

Hypothesized No. of Max- Eigen 0.05 Critical Prob. Eigenvalue CE(s) Statistic Value ** None * 0.9333 111.03 40.077 0.000 At most 1 * 0.6672 45.113 33.876 0.001 At most 2* 0.6398 41.872 27.584 0.000 At most 3* 0.4432 24.009 21.131 0.019 At most 4 0.2689 12.841 14.264 0.082 At most 5 0.0153 0.6339 3.842 0.425

Max-eigenvalue test shows four cointegratingeqn(s) at the 0.05 level * indicates that hypothes is rejected at the 0.05 level ** p-value of MacKinnon-Haug-Michelis (1999) With the help of trace statistic and Eigenvalue statistic, we came to know about the number of cointegration equations. According to trace statistic and Eigenvalue statistic there are four cointegration equations. Thus, with the help of these two statistics, the null hypothesis of no cointegration is rejected, and the alternative hypothesis of cointegration is accepted. It is concluded that there is the strong relationship among these variables in long run.

8.5.3Johansen Co-integration (Long run Estimates)

In table 8.12 the long run estimates of Johansen cointegration are presented where education enrollment index is used as dependent variable while public expenditure on primary, secondary

196 and tertiary education, literacy and inflation rate are used as explanatory variables. Chaudhry et al. (2010), Ali et al. (2012) and Shahzad (2015) used weighted education enrollment index as an explanatory variable and explored its impact on economic growth, but in our study, we used this index as dependent variable and treated it as return to investment in education.

These results explain that expenditure on primary education is positively and significantly affecting the education enrollment index. Similarly, public expenditure on secondary and tertiary education has positive and significant impact on education enrollment index. We can see that one percent increase in secondary education expenditure increases enrollment by 0.002 points while primary and tertiary expenditures increases enrollment by 0.002 and 0.004 points respectively. Returns to investment in tertiary education are quite higher than others; this is mainly due to the availability of jobs, nature of jobs and labor market conditions which are more suitable for individuals with tertiary level of education.

Asif, Shah and Shabir (2014) explored positive impact of education expenditure on enrollment rate. Mcmahon (1999) also explored that increase in primary education expenditure leads to increase in primary gross enrollment rate. Winter, Ebmer and Wirz (2002) also studied impact of expenditure on primary and secondary education on enrollment rate. This issue is also discussed by Bergh and Fink (2006).Anyanwu and Erhijakpor (2007) also explored positive impact of expenditure on primary and secondary education on enrollment rate while Iyer (2009) found that primary education expenditure has a positive but negligible impact on enrollment rate.

Table: 8.12 Normalized Cointegration Coefficients

Variable Coefficient Std.Error t-Statistic C 0.06 -- -- EXP 0.0015 0.0004 3.663 EXS 0.0018 0.0003 5.551 EXT 0.0037 0.0006 5.649 INF 0.0005 5.9E-05 9.576 LTR 0.0008 0.0003 2.961

Source: Author’s own calculations based on E-Views 9

197

The literacy rate is used as a proxy for the demand of education or awareness, as more people become literate they invest more in education and on the other hand due to literacy their income level increases so they invest more in education thus enrollment increases; therefore, this variable has the positive and significant impact on EEI. Iyer (2009) also used literacy rate in education enrollment model and use it as a proxy for awareness and higher demand for education thus increase in literacy rate leads to increase in enrollment.

Inflation rate is used as a proxy for economic instability, but in this study, it is positively and significantly affecting education enrollment index. Stange (2013) discussed about the role of inflation in education enrollment and advocated that inflation can be helpful to increase enrollment. If institutions reallocate their budget for improvement in the educational institutions and their services, thus it will attract students towards education and as a result enrollment increases. On the other hand, as in some cases, inflation leads to increase in employment thus future demand for education increases and enrollment increase.

8.5.4 Error Correction Model (Short run Estimates)

To get the short run estimates next step in this process is to apply error correction model that shows the speed of adjustment for the economy towards long run equilibrium after the short run shocks. The negative sign of the adjustment term shows that the previous year deviations from long run equilibrium is corrected at a speed of 4 percent and economy converges towards long run equilibrium, moreover this term is statistically significant.

Other short run results reveal that previous year values of education enrollment index has insignificant impact and previous year values of inflation rate, primary, secondary and tertiary education has negative and insignificant impact on education enrollment index on average ceteris paribus in short run. Previous year values of literacy rate have a positive impact on current year value of education enrollment index but this effect is statistically insignificant. All variables has on average insignificant ceteris paribus effect on education enrollment index in the short run thus we can conclude from these results that investment in education is a long run process. After observing long run and short run adjustments various diagnostic tests were performed to check the stability of the model. This model is free from serial correlation and heteroskedasticity and

198 graph of AR roots verified that parameters are stable. Eviews results of these diagnostic tests are presented in appendix 23 in detail.

Table: 8.13 Error Correction Model

Variable Coefficient Std.Error t-Statistic CointEq1 -0.4268 0.127 -3.349 D(EEI(-1)) 0.0179 0.167 0.107 D(EEI(-2)) -0.0965 0.167 -0.577 D(EXP(-1)) -7.53E-05 0.0003 -0.208 D(EXP(-2)) -0.0001 0.0003 -0.436 D(EXS(-1)) -0.0006 0.0004 -1.429 D(EXS(-2)) -0.0002 0.0005 -0.517 D(EXT(-1)) -0.0014 0.0005 -2.388 D(EXT(-2)) -0.0005 0.0005 -1.114 D(LTR(-1)) 0.0051 0.0023 2.191 D(LTR(-2)) 0.0074 0.0024 2.987 D(INF(-1)) -0.0002 0.0005 -0.421 D(INF(-2)) -0.0001 0.0005 -0.307 C -0.0066 0.0032 -2.023

Source: Author’s own calculations based on E-Views 9

8.5.5 Granger Causality

In table 8.14 results of granger causality are presented for education investment model, and these results are evaluated with the help of significance of probability values of F-Statistic at 10 percent level of significance. Expenditure on primary, secondary and tertiary education shows independent causality with education enrollment index while literacy rate has uni-directional causality with enrollment index, expenditure on primary, secondary and tertiary education. Inflation rate has no or independent causality with enrollment index and primary education expenditure but unidirectional causality with secondary and tertiary education expenditure. There is no causal relationship of tertiary education expenditure with primary and secondary education

199 expenditure while there is uni-directional causality between primary and secondary education expenditure. Moreover, inflation and literacy rate are independent or have no causal relationship.

Table: 8.14 Results of Granger Causality

Null Hypothesis Obs F-Statistic Prob. EXP does not Granger Cause EEI 0.002 0.997 43 EEI does not Granger Cause EXP 1.081 0.349 EXS does not Granger Cause EEI 0.819 0.448 43 EEI does not Granger Cause EXS 0.217 0.805 EXT does not Granger Cause EEI 0.637 0.534 43 EEI does not Granger Cause EXT 0.489 0.616 LTR does not Granger Cause EEI 4.526 0.017 43 EEI does not Granger Cause LTR 1.043 0.362 INF does not Granger Cause EEI 1.083 0.348 43 EEI does not Granger Cause INF 1.058 0.356 EXS does not Granger Cause EXP 1.233 0.302 43 EXP does not Granger Cause EXS 2.677 0.082 EXT does not Granger Cause EXP 2.167 0.128 43 EXP does not Granger Cause EXT 0.739 0.484 LTR does not Granger Cause EXP 0.167 0.846 43 EXP does not Granger Cause LTR 5.741 0.006 INF does not Granger Cause EXP 0.449 0.641 43 EXP does not Granger Cause INF 0.045 0.955 EXT does not Granger Cause EXS 0.751 0.478 43 EXS does not Granger Cause EXT 2.197 0.125 LTR does not Granger Cause EXS 0.451 0.641 43 EXS does not Granger Cause LTR 2.666 0.082 INF does not Granger Cause EXS 1.493 0.237 43 EXS does not Granger Cause INF 7.122 0.002 LTR does not Granger Cause EXT 43 1.864 0.168 EXT does not Granger Cause LTR 7.216 0.002 INF does not Granger Cause EXT 0.406 0.669 43 EXT does not Granger Cause INF 3.079 0.057 INF does not Granger Cause LTR 1.474 0.241 43 LTR does not Granger Cause INF 1.901 0.163 Source: Author’s own calculations based on E-Views 9

8.6 Health, Education and Economic Growth

In this section, the impact of public investment in education and health on economic growth is discussed by using Johansen Cointegration technique. For this purpose, we use GDP growth rate as the proxy for economic growth, total labor force and gross fixed capital formation as proxy for

200 labor and capital in the growth model. The main human capital variables are literacy rate and infant mortality rate while unemployment rate is used as a proxy for economic stability.

8.6.1 Lag Length Selection

According to ADF test results presented in table 8.2, all variables stated above for growth model are stationary at first difference; therefore, we can apply Johansen Cointegration technique for data analysis. The first step of this technique is the appropriate lag selection. We have selected lag one for this model by using the appropriate information criterion.

Table: 8.15 Lag Order Selection Criterions

Lag LogL LR FPE AIC SC HQ 0 -1165.627 NA 5.42e+15 53.25579 53.499 53.346 1 -924.9720 404.7385* 5.02e+11* 43.95327* 45.656* 44.584*

*Indicates lag order selection by the criterion like LR, FPE,AIC, SC and HQ

8.6.2 Unrestricted Cointegration Rank Test

After selection of lag length, we proceed with this lag and determine number of cointegration vectors. By using trace statistic we can say that there are three cointegration vectors and by using Eigenvalue statistic there is one cointegration equation. It means that we can reject null hypothesis of no cointegration and concluded that there is cointegration among the variables in this model.

Table: 8.16 Results of Trace Statistic

Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob. ** None * 0.626 115.753 95.753 0.001 At most 1 * 0.436 73.4395 69.818 0.025 At most 2* 0.398 48.8034 47.856 0.040 At most 3 0.324 26.9309 29.797 0.103 At most 4 0.192 10.0427 15.494 0.277 At most 5 0.020 0.87347 3.8414 0.350

Trace test shows threecointegratingeqn(s) at the 0.05 level * shows hypothesis is rejected at the 0.05 level ** p-value of MacKinnon-Hsaug-Michelis (1999)

201

Table: 8.17 Results of Maximum Eigenvalue

Hypothesized No. of Eigenvalue Max- Eigen Statistic 0.05 Critical Value Prob. ** CE(s) None* 0.626 42.313 40.077 0.027 At most 1 0.436 24.636 33.876 0.410 At most 2 0.398 21.872 27.584 0.227 At most 3 0.324 16.888 21.131 0.177 At most 4 0.192 9.1693 14.264 0.272 At most 5 0.020 0.8734 3.8414 0.350

Max-eigenvalue test shows one cointegratingeqn(s) at the 0.05 level * shows hypothesis is rejected at the 0.05 level ** p-value of MacKinnon-Haug-Michelis (1999) 8.6.3 Johansen Co-integration (Long run Estimates)

In table 8.18 the long run estimates of this model are presented where we have explored that literacy rate has the positive and highly significant impact on GDP growth rate that one unit increase in literacy rate leads to 0.81 units increase in GDP growth rate. Romer (1990) used literacy rates as human capital stock measure and reported its impact on economic growth of Nigeria. Thus he found investment in education as a worthwhile venture and despite low budgetary allocation to education, its impact on economic growth was still felt during that period of study. Isola et al (2012) also finds that the average literacy score in a given population is a better indicator of growth. They reported that a country that focuses on promoting strong literacy skills widely throughout its population will be more successful in fostering growth and wellbeing than one in which the gap between high-skill and low-skill groups is large.

Literacy rate increases as a result of public investment in education and Saraswati (2012) also interpreted positive role of education expenditure on economic growth in Indonesia and stated that due to increase in government expenditure on education increases the efficiency and creativity of the people and they plays a significant role in improving public revenue thus accelerates economic growth. Mekdadet al. (2014) also explored positive relationship between economic growth and education expenditure. Our findings are also consistent with Khattak and Khan (2012), Baldwin and Borrelli (2008) and Bashir et al. (2012).

202

Similarly, infant mortality rate has the negative and significant impact on GDP growth rate that one unit decrease in infant mortality rate leads to 0.23 units increase in GDP growth rate. Due to increase in health expenditure and better provision of health facilities leads to reduction in infant mortality rate. According to Bloom et al. (2004) rising longevity in developing countries has opened a new incentive for the current generation to save—an incentive that can have dramatic effects on national saving rates.

High mortality rates may reduce investment. Saving rates are thus likely to fall, as the prospect of retirement becomes less likely. Therefore a reduction in infant mortality rate leads to increase in economic growth. Abbas (2010), Issa and ouattara (2005) and Bashiret al. (2012) also explored significant association between health expenditure and economic growth and argued that if resources are properly utilized and health facilities are equally provided to all regions and persons then we can control infant mortality rate and various other serious health related problems and it has fruitful effect on the economy.

Table: 8.18 Normalized Cointegration Coefficients

Variable Coefficient Std.Error t-Statistic C 50.56203 -- -- GCF 3.27E-06 8.5E-07 3.862 LTR 0.814560 0.12147 6.706 IMR -0.236687 0.06089 -3.887 TLF 0.065449 0.06099 1.073 UEM -1.417723 0.29435 -4.816

Source: Author’s own calculations based on E-Views 9

Total labor force has the positive but statistically insignificant impact on GDP growth rate as one unit increase in total labor force increases GDP growth rate by 0.06 units. Positive relationship between the labor force and economic growth is also found by the Galor (1997). He contributed in the literature and considered the educated labor force very productive as employed by sectors like scientist, professionals and technicians in Israel. Hanushek et.al (2000) also found a stable, strong and consistent relationship between labor force and economic growth and suggested that increasing size of the global workforce is giving an opportunity to gain economic expansion and

203 accelerate gross domestic product. Our results are also consistent with the findings of Zhu (2011) and kakar (2011).

Capital formation is one of the major determinants of economic growth. There is a conventional perception that the most pertinent obstacle to economic growth is shortage of capital. Therefore government should encourage savings, create conducive investment climate and improve the infrastructural base of the economy to boost capital formation and hence promote sustainable growth. In this model gross fixed capital formation has positive and significant impact on GDP growth rate. These results are consistant with the findings of Gibescu (2010), Lach (2010), Ongo and Vukenkeng (2014), Ibe and Osuagwu (2016).

Unemployment is an important issue in developing economies and used as a measure of economic stability in this model. High unemployment means that labor resources are not being used efficiently. Reduction in unemployment means more and more people are getting employed and production is increasing moreover the standard of living are also increasing. According to the findings of this study one unit fall in unemployment rate leads to 1.41 units increase in GDP growth rate. Thus a negative association is observed that proves the existence of Okun’s law in Pakistan. Noor et al. also explored negative association between unemployment and GDP and proved the existence of okun’s law in Malaysia. Our results are also consistent with the findings of Farsio et al (2003), Lal et al. (2010), Meidani et al. (2011), Fatai and Bankole (2013).

8.6.4 Error Correction Model (Short run Estimates)

In this section, results of error correction model are presented that shows the speed of adjustment and explains that how much economy converges towards long run equilibrium every year. This is clear from the coefficient of adjustment term that annually after 57 percent adjustment economy converges towards long run equilibrium after short run shocks.

If we talk about previous year values of other variables then we can see that previous year values of literacy and infant mortality rate has positive impact on GDP growth rate of current year while gross fixed capital formation, total labor force and unemployment rate has negative impact on per capita income. We can see that all variables are statistically insignificant in the short run because it is a fact that growth of the economy is not a short run process and a long period is required for beneficial changes in the economy. After observing long run and short run

204 adjustments various diagnostic tests were performed to check the stability of the model. This model is free from serial correlation and heteroskedasticity and graph of AR roots verified that parameters are stable. Eviews results of these diagnostic tests are presented in appendix 23 in detail.

Table: 8.19 Error Correction Model

Variable Coefficient Std.Error t-Statistic CointEq1 -0.575574 0.22686 -2.53717 D(GDP(-1)) -0.189537 0.15567 -1.21754 D(GCF(-1)) -8.68E-07 1.9E-06 -0.46664 D(LTR(-1)) 0.428182 0.42341 1.01127 D(IMR(-1)) 0.045426 0.10757 0.42230 D(TLF(-1)) -0.011596 0.23471 -0.04941 D(UEM(-1)) -0.205212 0.49619 -0.41357 C -0.278092 0.55479 -0.50126

Source: Author’s own calculations based on E-Views 9

8.6.5 Granger Causality

Granger causality is used to check causal relationship among the variables used in growth model and these results are represented in table 8.20. From these results, we can see that there if no causality between GDP growth rate and gross fixed capital formation similarly literacy rate has no causal relationship with GDP growth rate but unidirectional causality with gross fixed capital formation. Infant mortality rate has no causal relationship with GDP growth rate but a uni directional causality with gross fixed capital formation and literacy rate. Total labor force has independent causality with GDP growth rate, literacy and infant mortality rate but unidirectional causality with gross fixed capital formation. Unemployment rate has no causality with total labor force but unidirectional causality is observed with infant mortality rate, literacy rate, unemployment rate and GDP growth rate at 10 percent level of significance.

205

Table: 8.20 Results of Granger Causality

Null Hypothesis: Obs F-Statistic Prob. GCF does not Granger Cause GDP 1.680 0.199 43 GDP does not Granger Cause GCF 0.944 0.397 LTR does not Granger Cause GDP 1.276 0.290 43 GDP does not Granger Cause LTR 0.116 0.890 IMR does not Granger Cause GDP 1.263 0.294 43 GDP does not Granger Cause IMR 1.091 0.346 TLF does not Granger Cause GDP 0.747 0.480 43 GDP does not Granger Cause TLF 0.091 0.912 UEM does not Granger Cause GDP 0.401 0.671 43 GDP does not Granger Cause UEM 4.833 0.013 LTR does not Granger Cause GCF 3.927 0.028 43 GCF does not Granger Cause LTR 2.420 0.102 IMR does not Granger Cause GCF 3.751 0.033 43 GCF does not Granger Cause IMR 0.047 0.953 TLF does not Granger Cause GCF 3.825 0.031 43 GCF does not Granger Cause TLF 1.141 0.330 UEM does not Granger Cause GCF 2.582 0.088 43 GCF does not Granger Cause UEM 0.124 0.883 IMR does not Granger Cause LTR 0.992 0.380 43 LTR does not Granger Cause IMR 2.688 0.081 TLF does not Granger Cause LTR 0.726 0.490 43 LTR does not Granger Cause TLF 1.849 0.171 UEM does not Granger Cause LTR 3.314 0.047 43 LTR does not Granger Cause UEM 1.086 0.347 TLF does not Granger Cause IMR 1.908 0.162 43 IMR does not Granger Cause TLF 0.160 0.852 UEM does not Granger Cause IMR 2.992 0.062 43 IMR does not Granger Cause UEM 1.429 0.252 UEM does not Granger Cause TLF 0.195 0.823 43 TLF does not Granger Cause UEM 0.325 0.723 Source: Author’s own calculations based on E-Views 9

8.7 Conclusion

This chapter comprehensively describes results of all tests and techniques for the analysis of macroeconomic models and to discuss the required objectives. Initially, we have discussed some measures of central tendency and dispersion and then discussed the shape of the distribution.

206

According to the results of ADF test, we came to know that all the fourteen variables used in these three models are non-stationary at level but are stationary at first difference. Therefore it is suitable to apply Johansen Cointegration technique and vector error correction model for data analysis in all the models.

In the first health investment model by using lag one, we obtained five Cointegration vectors according to trace statistic while three vectors according to Eigenvalue statistic and estimated long run association between the variables. We came to know that effect of all the variables is according to priori expectations and highly significant in long run. We explored positive impact of fertility rate and population per doctor on infant mortality rate while all other variables have negative and significant impact on infant mortality rate. By using ECM, we came to know that 52 percent disequilibrium adjusts annually and then we discussed causal relationship among the variables by using Granger causality and found various uni and bi directional causal results. Thus overall this model suggest that with an increase in public investment for health sector we can get many profitable results and a check on fertility rate and awareness among parents regarding health issues is also playing a significant role in the longevity of children along with investment.

In the education investment model by using lag three as appropriate lag, there are four Cointegration equations according to trace statistic and Eigenvalue criteria. Long run results of this model explored that investment in education of all levels leads to significant increase in enrollment rate. Inflation and literacy rate also plays a significant role in this regard while results of ECM explored that annually 42 percent convergence occurs towards long run equilibrium. In education return model it is clear that almost all levels of education has equal contribution to increase enrollment but expenditure on tertiary level of education has a slightly greater impact to increase the enrollment ratio in the country.

In the last model of economic growth, we have used GDP growth rate as a proxy for economic growth and explored significant positive impact of investment in health and education on economic growth by using infant mortality rate and literacy rate as a proxy for these investments. Labor and capital have significant contributions to enhance economic growth. These models of health and education investments are free from any econometric problem and these are stable models as verified by various diagnostic tests.

207

Thus with the help of these findings we came to know that education and health sector are playing a highly significant role in economic growth especially to enhance the skills of human capital and to make them more productive. As we know that currently, the government is spending very little budget on these sectors therefore we are unable to meet international standards but if investment in these sectors increases then returns in the form of outcome variables stated above also increases and enhances economic growth.

208

Chapter 9 SUMMARY, CONCLUSION AND RECOMMENDATIONS 9.1 Introduction In this chapter, the summary of the findings derived from the study is presented. Conclusion based on the findings is also given, followed by appropriate suggestions and policy recommendations. 9.2 Summary of the Study The basic purpose of this study was to analyze the returns of investment to education and health in Pakistan. We conducted this study both at micro and macro level and tried to evaluate returns of investment in both components of human capital. In this section, a bird’s eye view on this study is presented. In the first chapter of this dissertation, we discussed about the background of this study and gave a detailed problem statement and discussed the significance of investment in these two components of human capital that is education and health. In this chapter, we discussed the millennium development goals and the indicators of education and health that are included as targets in MDG that encouraged researcher to investigate these problems in Pakistan that are addressed at international level also. Research questions and objectives of the study are clearly stated in this part of the study. At the end, a brief organization of the whole study is presented. In second chapter retrospective study of education and health sector in Pakistan are presented in detail. In this chapter, we discussed the historical performance of these sectors in Pakistan. All the plans and policies developed after the establishment of Pakistan till now are discussed in detail. It includes all five-year plans and some other reports regarding health and education. Budget allocated for these sectors, targets and the results obtained by using these policies are also presented here. Thus this part of dissertation provides a deep insight into this issue. District Multan is selected as sampling area in this study, therefore, in order to present situation of education and health services in Multan a detailed snapshot of these sectors in Multan district is also provided in chapter two. In this section, population in all towns and linguistic distribution in this district is presented. A detail of education services in Multan is also presented that includes the number of educational institutions of all types separately, enrollment situation, literacy rate and budget allocation for education sector in Multan. In the last section of this

209 chapter, an overview of health sector is presented that includes no of public and private hospitals and health units available in all towns and availability of other health services in Multan district. In chapter three detailed literature review is presented both at the national and international level about returns of investment to education and health and the relationship between education, health and economic growth. In the initial section of this chapter, a detailed history of human capital is also provided. In chapter four theoretical and conceptual framework is presented in detail. Moreover a detailed insight into the importance of human capital is presented where we came to know its importance and find that how it is different from investment in physical capital. Then we discussed the role of education and health in improving human capital. Several methods are presented to estimate the returns of investment in education and health and the models used in literature for this purpose are also discussed. Among all of them contributions of Mincer (1958, 1972, 1974, 1979), Psacharopoulos (1994, 1995, 2004), Grossman (1972), Becker (1974) and Schultz (1961, 1993, 2002) are of great importance in this field and for this study also. In the end, relationship between education, health and economic growth are also established. In chapter five, we discussed the data and methodology adopted in this study. Initially, we discussed about the variables used in micro and macro analysis and their sources. Then we discussed about the sampling technique, sample size and survey technique. Detailed presentations of econometric models which are empirically tested in this study are also included in this chapter. Finally, we discuss the econometric techniques that are used for data analysis like ADF test, Johansen cointegration, granger causality and ordinary least square. An elementary analysis of data from Multan district is presented in chapter six, where we can understand present situation of health and education in Multan at a glance. In chapter seven, a detailed micro analysis is presented by using Mincer earning function (1974) for both education and health returns. Earnings are treated as dependent variable and proxy for returns in both cases. In first section, returns to total years of education, different levels of education, gender, region, marital status, sector of employment, profession, terminal degree etc, are presented. Moreover, returns to the quality of education are also discussed in detail. In the next section of this chapter, we discussed the returns of investment in health by using anthropometric measures of health investment. We also discussed returns to self-reported health

210 status, investment in food, clean water, environment, exercise, precautionary measures and medical facilities are also discussed. In chapter eight macro data analysis is presented. In first section returns of investment to health is presented and infant mortality rate is used as measure of health outcome while in second section returns of investment to education are discussed where education outcome presented with education enrollment index. In last section GDP growth rate is used as a proxy for economic growth and impact of public expenditure on education and health sector in the form of lower infant mortality rate and better literacy rate is explored on it while this chapter nine is based on conclusion and policy recommendations. 9.3 Summary of the Findings The results of the Mincer earning function (1974) are estimated by using ordinary least square for investment in education and health separately. These results explored that private returns to total years of education are 12.9 percent in Multan district as a whole, but when disaggregated for gender then we found that these returns are higher for female as compared to the male that is 19.1 percent for female and 12.2 percent for the male. This means that although female earnings are low as compared to male but addition in female education yields higher marginal returns. Returns to an additional year of education in urban areas are 13.4 percent while in rural areas it is 11.9 percent only. This difference is due to the availability of highly paid jobs in urban areas. Similarly, returns to education for married workers are 12.2 while for unmarried workers it is 13.9 percent. This difference is due to greater geographical and job mobility for unmarried workers. Similarly, in case of employment sectors returns for individuals in federal sector are 13.3 percent while for workers in provincial, semi government and private sector it is 9.9, 11.6 and 14.6 respectively. Similarly, in case of professions, irrespective of 18 percent returns of other sectors workers in health sector earn highest returns that are 17.3 percent. Then 15.3 percent in the industrial sector, 15 percent in the education sector, 10.2 percent in finance and legal sector and finally 9.3 percent in the social service sector. Returns are also calculated for different faculties where physical sciences showed highest returns of 20.5 percent while agriculture and veterinary sciences showed lowest returns of 2.02 percent only. In case of returns of investment in the quality of education, we used the medium of instruction in high school as a proxy for quality of education and found it highly significant in Mincer earning

211 function. We explored by running the regression that returns to individuals with English medium of instruction are 16.9 percent while for Urdu medium it is 12.6 percent. We also discussed the role of total working hours, matriculation subjects and parent’s qualification as a determinant of earnings of individuals and found them significant. These results were again disaggregated by gender, region and marital status. Similarly, returns to different levels of education are discussed in detail where primary education is used as reference category to avoid dummy variable trap. Secondary level of education showed highest returns as compared to other levels of education while graduation and university level of education has almost same returns. In case of investment in health, we used height as a measure of long run health investment while body mass index is considered as a short run measure of health investment and nutritional status. In these results, we found that 1 unit increase in height leads to 0.241 units increase in earnings while in case of BMI this increase is 1.3 percent for the whole sample. These returns were then disaggregated by gender and found that for male increase in height leads to 5.1 percent increase in income for the male and 2.7 percent for the female. Similarly increase in BMI leads to 0.92 percent increase in income for the male while 0.24 percent for the female. BMI is significant especially for less educated workers and till a healthy range only as BMI approaches to obesity the situation may reverse. In case of investment in nutritious food and clean water, we came to know that each 1 unit increase in food investment leads to 0.0037 percent increase in earnings while in case of pure drinking water it is 12 percent because most of the diseases are spread due to impure drinking water. In case of self-reported health status, we found that better health status leads to 4.5 percent increase in income but due to increase in disability to perform physical activities earnings falls by 4.1 percent. In case of other determinants of health, we came to know that due to the availability of medical facilities earnings increases by 9.3 percent while due to investment in precautionary measures it is 7.2 percent. Daily time spent on exercise increase earnings by 0.01 percent while due to clean environment increase is 15.6 percent only. The study also examined public investment in education and health and their impact on economic growth by using proxies for outcomes of health and education with the help of recent developments in time series data, such as cointegration tests and stationarity test. Consequently, we were able to reveal the impact of health and education on economic growth in long run for the period of 1972 to 2016 by using annual time series data.

212

Returns of investment to education and health at macro level were difficult to estimate by using mincer earning function because of non-availability of data. Therefore multiple regression analysis was used to evaluate returns of public investment in education and health in Pakistan. In health investment model infant mortality rate was used as a proxy for health returns while health expenditure as percentage of GDP and population per doctor were used as a proxy for health investment and quality of provision of health services along with some other explanatory variables. This model showed very strong role of health related variables to increase the longevity of infants. In education investment model an education enrollment index was constructed and impact of expenditures on various levels of education was recorded on that index along with some other variables. This relationship highlighted the importance of public investment in education. Finally this study also analyzed the role of public investment in health and investment in enhancing economic growth by using some proxy variables and found it fruitful for the economy. 9.4 Conclusion The analysis provided in this study substantially improves understanding about the outcomes of public investment in education and health with special reference to district Multan and from Pakistan for the period of 1972 to 2016 by using suitable econometric techniques and software. The literature which studies the returns of investment to health and education is growing and is an important one. The basic aim of this study was contribution in existing literature by providing new dimensions for research in the fields of education and health investment because returns to investment in health are still debatable as fewer studies are available on this issue on the world level and in Asian countries also. Human capital is created initially by providing children with primary and then secondary schooling and basic health services. If poor countries wish to achieve high levels of national income, they need to provide public funding for the health and universal education of the poor, at least at the basic level. Health status and educational attainment are multidimensional concepts that cannot be directly measured by a single set of indicators. Social outcome should be seen as the result of a complex production process that involves interrelationships among many variables, including institutional factors and individual behavior (World Bank, 2000). Therefore this study uses a mix of proxy variables for investment returns at primary and secondary level.

213

In case of returns to education at micro level, by using Mincer earning function we found that increase in education increases earnings for all persons whether they are male or female, urban or rural, married or unmarried and from different sectors and faculties. Regional disparities, family background and medium of instruction are strongly affecting earnings of people. Thus equitable investment in education is always fruitful. Similar is the case with returns to investment in health that investment in health during childhood increases earnings of that person, similarly investment on other factors like food, pure water, clean environment, precautionary measures and availability of medical facilities all are helpful to increase earnings of an individual.

Health and education play a major and important role in determining the long run economic growth of Pakistan. The study confirmed that if the government increases the budget for education and health, more people will be educated which will result in more educated workers and resultantly more production. Similarly, it will also have a good impact on the health of the general public. The study suggests that the government of Pakistan should consider education and health sectors while formulating policies and must allocate sufficient budget for them and efficient allocation of budget is also necessary for fruitful results. 9.5 Policy Recommendations Investment in human beings is always underemphasized and neglected in Pakistan. But truly, this investment should be the growth of human capital. A major amount of budget should be allocated to health and education sectors for their efficient performance. Based on the findings of this research, following are some recommendations for the improvement of education and health sector in Pakistan both at micro and macro level; i. The standard of health and education should be improved in Pakistan including its availability. Good health not only raises the returns to investment in education but also improves the living standard of the people. ii. It is clear from the experience of developed countries that efficiency and technical change are two main factors which are important for economic growth. Technology and training should be given proper attention and training should design in such a way that it fulfills our local needs. Government should take steps in creating opportunities for new jobs. iii. The private sector should be encouraged to improve its share in hospitals and schools.

214

iv. Government should create environment for macroeconomic stability to encourage private investment in health and education sectors. v. Federal and provincial government should sustain and improve the programs of free education and healthcare. vi. Overhauling of public education system, method of teaching and curriculum should be done. vii. The problem of high dropout and low participation in the education system should be corrected by reshuffling the priorities of current education policy. Due attention should be given to the development and promotion of basic education. viii. Human capital embodied labor effects significantly and positively to economic growth, so government should increase investment in training programs for labor and technical education. ix. A careful evaluation and inspection of educational programs by neutral bodies may start to avoid implementation gaps and delays. x. Communities should be effectively involved in the promotion of basic education and health. xi. Market oriented approach in education should be emphasized. xii. More educational institutions should be established in sampling area and all other deprived regions and quality of education should be properly monitored. xiii. Use of English should be promoted in public institutions also. xiv. Training sessions should be organized for teachers to improve the quality of education. xv. Better educational and health facilities should be made available in far flung areas so that regional disparities should be removed. xvi. Government should take special measures to reduce unemployment among the educated people. xvii. Priority should be given in developing health infrastructure. xviii. Health Management and Information System should be strengthened. xix. A high level awareness in people is necessary about the health. xx. Preventive measures should be adopted to get rid of fatal diseases and public should cooperate with health teams in taking preventive measures against harmful diseases.

215

xxi. Government must accord its priorities to basic health care, primary and secondary education. Social spending must be targeted to ensure that the poor people of the community also have proper access to basic education and health care. xxii. In rural areas many fatal diseases are spreading just due to unhygienic drinking water. Therefore water purification plants should be established in rural areas at a large level. xxiii. The analytical basis for policy formation should be formulated by a core group of policy analysts and researchers. The group should consist of analysts from health and social sciences including economists’ educationists, health economists and public health experts. In spite of great effort, many issues are still uncovered and invite attention of other researchers to bridge up the gaps in this study especially in the field of returns of investment to health as almost very little attention has been focused on this problem in past.

216

REFERENCE

Abramovitz, M. (1956). Resource and output trends in the United States since 1870. In Resource and output trends in the United States since 1870. NBER. 1-23.

Abbas, F. (2010). Public health sector expenditures, health status and their role in development of Pakistan, Universitäts-und Landesbibliothek Bonn.

Abbas, F., & Hiemenz, U. (2011). Determinants of public health expenditures in Pakistan, Discussion Papers on Development policy No. 158, Center for Development Research, Bonn.

Afzal, M. (2011). Microeconometric analysis of private returns to education and determinants of earnings. Pakistan Economic and Social Review, 39-68.

Afzal, M., Farooq, M. S., Ahmad, H. K., Begum, I., & Quddus, M. A. (2010). Relationship between school education and economic growth in Pakistan: ARDL bounds testing approach to cointegration. Pakistan Economic and Social Review, 39-60.

Ainsworth, M., & Over, M. (1994). AIDS and African development. The World Bank Research Observer, 9(2), 203-240.

Akram, N., Ihtsham ul Haq Padda, & Khan, M. (2008). The long term impact of health on economic growth in Pakistan. The Pakistan Development Review, 487-500.

Akram, M. (2008). Health Care Services and Government Spending in Pakistan. PIDE Working Paper.

Ali, S., Sharif Chaudhry, I., & Farooq, F. (2012). Human Capital Formation and Economic Growth in Pakistan. Pakistan Journal of Social Sciences (PJSS), 32(1).

Alderman, H., & Lavy, V. (1996). Household responses to public health services: cost and quality tradeoffs. The World Bank Research Observer, 11(1), 3-22.

Alderman, H., Behrman, J. R., & Hoddnott, J. (2003). Nutrition, Malnutrition and Economic Growth, Health Economics Group, Pan American Health Organization (PAHO), Regional Office of the World Health Organization (WHO).

217

Amsden, A. H. (1992). Asia's next giant: South Korea and late industrialization. Oxford University Press on Demand.

Anyanwu, J., & Erhijakpor, A. E. (2007). Working Paper 92-Education Expenditures and School Enrolment in Africa: Illustrations from Nigeria and Other SANE Countries (No. 227).

Anyanwu, J. C., & Erhijakpor, A. E. (2009). Health expenditures and health outcomes in Africa. African Development Review, 21(2), 400-433.

Arias, O., & McMahon, W. W. (2001). Dynamic rates of return to education in the US. Economics of Education Review, 20(2), 121-138.

Arora, S. (2001). Health, human productivity, and long-term economic growth. The Journal of Economic History, 61(3), 699-749.

Arrazola, M., & de Hevia, J. (2008). Three measures of returns to education: An illustration for the case of Spain. Economics of Education Review, 27(3), 266-275.

Asaoka, H. (2006). Investing in education in india: inferences from an analysis of the rates of return to education across indian states. ProQuest.

Asghar, Z., Attique, N., Urooj, A., & Mahmood, N. (2009). Measuring impact of education and socio-economic factors on Health for Pakistan. The Pakistan Development Review, 48(4), 653- 674.

Asif, M., Shah, S., & Shabir, M. (2014). School expenditure, school enrollment and quality education: A case study of Gilgit Baltistan, Pakistan, Journal of Multidisciplinary Scientific Research, 2(5), 09-12.

Asadullah, M. N. (2006). Returns to education in Bangladesh. Education Economics, 14(4), 453- 468.

Ashraf, J. (2011). New evidence on rates of return to education in Pakistan, Global Journal of Business Research, 5(3), 113-120.

Audibert, M., & Etard, J. F. (2003). Productive benefits after investment in health in Mali. Economic Development and Cultural Change, 51(3), 769-782.

218

Awan, M. S., & Hussain, Z. (2007). Returns to education and gender differentials in wages in Pakistan, The Lahore Journal of Economics, 12(2), 49-68.

Azariadis, C., & Drazen, A. (1990). Threshold externalities in economic development. The Quarterly Journal of Economics, 105(2), 501-526.

Aziz, S. (2005). Investment in Education and Skill Development. Lahore Journal of Economics, 10 (Special Edition), 63-71.

Bagehot, W. (1953). Economic Studies. Stanford, Calif.: Academic Reprints.

Baldwin, N., & Borrelli, S. A. (2008). Education and economic growth in the United States: cross-national applications for an intra-national path analysis. Policy Sciences, 41(3), 183-200.

Bloom, D. E., Canning, D., & Jamison, D. T. (2004). Health, wealth, and welfare. Finance and development, 41, 10-15.

Barenberg, A., Basu, D., & Soylu, C. (2015). The Effect of Public Health Expenditure on Infant Mortality: Evidence from a Panel of Indian States, 1983-84 to 2011-12.

Barro, R. J., & Sala-i-Martin, X. (1995). Economic growth, Advanced series in economics. New York, London and Montreal: McGraw-Hill.

Barro, R. J. (1996). Health, human capital and economic growth. In Pan American Health Organization.

Bashir, F., Nawaz, S., Yasin, K., Khursheed, U., Khan, J., & Qureshi, M. J. (2011). Determinants of inflation in Pakistan: An econometric analysis using Johansen co-integration approach. Australian Journal of Business and Management Research, 1(5), 71-92.

Becker, G.S. (1964). Human Capital, (2nd ed.), Columbia University Press, New York.

Becker, G.S. (1974). Human Capital. New York : National Bureau of Economic Research.

Becker, G.S. (1975). Human Capital. Columbia University Press, 2nd ed., New York.

Behrman, J. R., & Deolalikar, A. B. (1988). Health and nutrition. Handbook of development economics, 1, 631-711.

219

Behrman, J. R. (1993). The economic rationale for investing in nutrition in developing countries. World Development, 21(11), 1749-1771.

Behrman, J. R., & Schneider, R. (1993). An international perspective on Pakistani human capital investments in the last quarter century. The Pakistan Development Review, 1-68.

Bélanger, C. H., & Lavallee, L. (1980). Economic returns to schooling decisions. Research in Higher Education, 12(1), 23-35.

Benhabib, J., & Spiegel, M. M. (1994). The role of human capital in economic development evidence from aggregate cross-country data. Journal of Monetary economics, 34(2), 143-173.

Bergh, A., & Fink, G. (2006). Higher education: Does public expenditure increase enrollment. Ratio Working Papers, (84).

Besley, T., & Burgess, R. (2003). Halving global poverty. The Journal of Economic Perspectives, 17(3), 3-22.

Bhargava, A., Jamison, D. T., Lau, L. J., & Murray, C. J. (2001). Modeling the effects of health on economic growth. Journal of health economics, 20(3), 423-440.

Bishai, D., Opuni, M., & Poon, A. (2007). Does the level of infant mortality affect the rate of decline?: Time series data from 21 countries. Economics & Human Biology, 5(1), 74-81.

Björklund, A., & Kjellström, C. (2002). Estimating the return to investments in education: how useful is the standard Mincer equation?. Economics of Education Review, 21(3), 195-210.

Bloom, D. E., Canning, D., & Sevilla, J. (2001). The effect of health on economic growth: theory and evidence (No. w8587). National Bureau of Economic Research.

Boehmer, U., & Williamson, J. B. (1996). The impact of women's status on infant mortality rate: A cross-national analysis. Social Indicators Research, 37(3), 333-360.

Bourguignon, F., & Morrisson, C. (1998). Inequality and development: the role of dualism. Journal of development economics, 57(2), 233-257.

Canton, E. (2007). Social returns to education: Macro-evidence. De Economist, 155(4), 449-468.

220

Card, D., & Krueger, A. B. (1992). Does school quality matter? Returns to education and the characteristics of public schools in the United States. Journal of political Economy, 100(1), 1-40.

Carnoy, M. (1995). Rates of return to education. The international encyclopedia of education.

Case, A., Fertig, A., & Paxson, C. (2005). The lasting impact of childhood health and circumstance. Journal of health economics, 24(2), 365-389.

Chaudhry, I. S., Malik, A., & Faridi, M. Z. (2010). Exploring the causality relationship between trade liberalization, human capital and economic growth: Empirical evidence from Pakistan. Journal of Economics and International Finance, 2(9), 175-193.

Compah-Keyeke, G., Gyimah Sackey, F. G., & Azinim, M. A. (2013). Public Expenditure and Health Status in Ghana. Journal of Economics and Sustainable Development, 4(11), 88-99.

Corso, P. S., Ingels, J. B., Taylor, N., & Desai, S. (2014). Linking Costs to Health Outcomes for Allocating Scarce Public Health Resources. eGEMs, 2(4), 234-256.

Crémieux, P. Y., Ouellette, P., & Pilon, C. (1999). Health care spending as determinants of health outcomes. Health Economics, 8(7), 627-639.

Cropper, M. L. (1977). Health, investment in health, and occupational choice. Journal of Political Economy, 85(6), 1273-1294.

Crum, F. S. (1919). Public Accidents and Their Cost. In Proc. Nat. Safety Council (8th Annual Safety Congress, 1919).

Daoud, Y. (2005). Gender gap in returns to schooling in Palestine. Economics of Education Review, 24(6), 633-649.

Datar, A., Kilburn, R., & Loughran, D. (2006). Health endowments and parental investments in infancy and early childhood, RAND labor and population working paper series, WR- 367.

Deininger, K., & Squire, L. (1998). New ways of looking at old issues: inequality and growth. Journal of development economics, 57(2), 259-287.

221

Denison, E. F. (1967). Why growth rates differ (pp. 140-141). Washington, DC: Brookings Institution.

Denison, E. (1979). Accounting for Slower Economic Growth: The United States in the 1970s, The Brookings Institution, Washington D.C.

Deolalikar, A. B. (1988). Nutrition and Labor Producvivity in Agriculture: Estimates for Rural South India. The review of Economics and Statistics, 406-413.

Dessus, S. (1999). Human capital and growth: the recovered role of education systems.

Dublin, L. I., & Alfred, J. (1930). Lotka. The Money Value of a Man.

Duflo, E. (2001). Schooling and labor market consequences of school construction in Indonesia: Evidence from an unusual policy experiment. American Economic Reveiw, 91(4), 795-813.

Duraisamy, P., & Duraisamy, M. (1995). Determinants of investment in health of boys and girls: Evidence from rural households of Tamil Nadu, India. Indian Economic Review, 51-68.

Eggoh, J., Houeninvo, H., & Sossou, G. A. (2015). Education, health and economic growth in African countries. Journal of Economic Development, 40(1), 93-113.

Eide, E. R., & Showalter, M. H. (2011). Estimating the relation between health and education: What do we know and what do we need to know?. Economics of Education Review, 30(5), 778- 791.

Elgazzar, H. A. (2011). Raising returns: the distribution of health financing and outcomes in Yemen, Health, Nutrition and population (HNP) discussion paper.

Engle, R. F., & Granger, C. W. (1987). Co-integration and error correction: representation, estimation, and testing. Econometrica: journal of the Econometric Society, 251-276.

Erdoğan, E., Ener, M., & Arıca, F. (2013). The strategic role of infant mortality in the process of economic growth: An application for high income OECD countries. Procedia-Social and Behavioral Sciences, 99, 19-25.

222

Fabricant, S. (1954). Economic progress and economic change. In Economic Progress and Economic Change. NBER.

Farahani, M., Subramanian, S. V., & Canning, D. (2010). Effects of state‐level public spending on health on the mortality probability in India. Health economics, 19(11), 1361-1376.

Farooq, M. (2011). The Returns to Education for Male and Female Workers in Pakistan: A New Look at the Evidence. Dialogue (Pakistan), 6(2), 231-255.

Farr, W. (1853). Equitable Taxation of Property, J. Royal Statis. Soc., 1-45.

Farsio, F., & Quade, S. (2003). An empirical analysis of the relationship between GDP and unemployment. Humanomics, 19(3), 1-6.

Fatai, B. O., & Bankole, A. (2013). Empirical test of Okun’s Law in Nigeria. International Journal of Economic Practices and Theories, 3(3), 227-231.

Filmer, D., & Pritchett, L. (1999). The impact of public spending on health: does money matter?. Social science & medicine, 49(10), 1309-1323.

Fisher, I. (1908). Cost of Tuberculosis in the United States and Its Reduction. Read before the International Congress on Tuberculosis, Washington.

Fisk, E. L., & Sharp, C. T. (1921). Health of Industrial Workers. Waste in Indastry. Washigton.

Fogel, R. W. (1994). Economic growth, population theory, and physiology: the bearing of long- term processes on the making of economic policy (No. w4638). National Bureau of Economic Research.

Forsyth, C. H. (1915). Vital and monetary losses in the United States due to preventable deaths. Quarterly publications of the American Statistical Association, 14(112), 758-789.

Foster, A. D. (1995). Nutrition and health investment. The American Economic Review, 85(2), 148-152.

Frank, J. W., & Mustard, J. F. (1994). The determinants of health from a historical perspective. Daedalus, 1-19.

223

Frankenberg, E. (1995). The effects of access to health care on infant mortality in Indonesia. Health Transition Review (5), 143 – 163.

Friedrich, C. C. H. (1938). U.S. Patent No. 2,131,024. Washington, DC: U.S. Patent and Trademark Office.

Fulford, S. (2014). Returns to Education in India. World Development, 59,434-450.

Gallup, J., & Sachs, J. (2000). The Economic Burden of Malaria. Center for International Development. Harvard University, Working Paper 52.

Galor, O., & Tsiddon, D. (1997). The distribution of human capital and economic growth. Journal of Economic Growth, 2(1), 93-124. Gibescu, O. (2010). Does the gross fixed capital formation represent a factor for supporting the economic growth? MPRA Paper No. 50135, Online at http://mpra.ub.uni-muenchen.de/50135/. Giffen, R. (1880). Essays in Finance. 1st ser. London: G. Bell & Sons.

Gillani, D. Q., Khan, R. E. A., & Faridi, M. Z. (2013). Earning Determinants and Urban Informal Sector: Evidence from District Multan. Pakistan Vision, 14(2), 132-150.

Gordon, R. M. (2009). Socio-economic determinants of infant and child mortality in Haiti. Journal of Eastern Caribbean Studies, 34(1), 1-21.

Groot, W., & Van Den Brink, H. M. (2007). The health effects of education. Economics of Education Review, 26(2), 186-200.

Grossman, M. (1972). On the concept of health capital and the demand for health. Journal of Political economy, 80(2), 223-255.

Guisinger, S.E., Handerson, J.W and Scully, G.W. (1984). Earning, Rates of Return to Education and the Earning Distribution in Pakistan. Economics of Education Review, 3(4), 257-267.

Guifu, C. H. E. N., & Hamori, S. (2009). Economic returns to schooling in urban China: OLS and the instrumental variables approach. China Economic Review, 20(2), 143-152.

Gulliford, M. C. (2002). Availability of primary care doctors and population health in England: is there an association?. Journal of Public Health, 24(4), 252-254.

224

Gunderson, M. & Oreopoulos, P. (2010). Returns to education in developed countries. International Encyclopedia of Education, 2, 298-304.

Gyimah-Brempong, K. (1998). The political economy of budgeting in Africa, 1971- 1991. Journal of Public Budgeting, Accounting & Financial Management, 9(4), 590-602.

Hadley, J., Reschovsky, J., Corey, C., & Zuckerman, S. (2009). Medicare fees and the volume of physicians' services. Inquiry, 372-390.

Hadley, J., & Reschovsky, J. D. (2012). Medicare spending, mortality rates, and quality of care. International journal of health care finance and economics, 12(1), 87-105.

Hamdani, K. A. (1977). Education and the income differential: an estimation for Rawalpindi city. The Pakistan Development Review, 16(2), 144-164.

Hanmer, L., Lensink, R., & White, H. (2003). Infant and child mortality in developing countries: analysing the data for robust determinants. The Journal of Development Studies, 40(1), 101-118.

Hansen, W. L. (1963). Total and private rates of return to investment in schooling. Journal of Political Economy, 71(2), 128-140.

Hansen, P., & King, A. (1996). The determinants of health care expenditure: a cointegration approach. Journal of health economics, 15(1), 127-137.

Hanushek, E. A., & Kimko, D. D. (2000). Schooling, labor-force quality, and the growth of nations. American economic review, 1184-1208.

Haque, N. U. (1992). The Rates of Return to Education in Rawalpindi, Pakistan. Sustainable Development Policy Institute.

Harmon, C., Oosterbeek, H., & Walker, I. (2003). The returns to education: Microeconomics. Journal of economic surveys, 17(2), 115-156.

Hausman, J. A. (1978). Specification tests in econometrics. Econometrica: Journal of the Econometric Society, 1251-1271.

225

Hitiris, T., & Posnett, J. (1992). The determinants and effects of health expenditure in developed countries. Journal of health economics, 11(2), 173-181.

Heckman, J. J., Lochner, L. J., & Todd, P. E. (2003). Fifty years of Mincer earnings regressions (No. w9732). National Bureau of Economic Research.

Hough, J. R. (1993). Educational Cost Benefit Analysis: Education Research Paper No. 02. Loughborough University, London.

Huebner, S. S. (1914). The Human Value in Business Compared with the Property Value, Proc. Thirty-fifth Ann. Convention Nat. Assoc. Life Underwriters, 17-41.

Hull, C. R. (1899). The Economic Writings of Sir William Petty. 2 vols, Cambridge Univ. Press.

Ibe, S., & Osuagwu, N. (2016). Impact of capital formation on the Economic Development of Nigeria. In Fifth International Conference on Global Business, Economics, Finance and Social Sciences (GB16 Chennai Conference).

Idrus, A., & Cameron, S. (2000). Returns to education between the self-employed and employed sectors: evidence from Malaysia. The Pakistan Development Review, 263-268.

Imad, U. (2014). The Impact Of Education And Health On Economic Growth An Econometric Study Of Pakistan (1970-2010) (Doctoral dissertation, University Of Peshawar, Peshawar).

Isola, W. A., & Alani, R. A. (2012). Human capital development and economic growth: Empirical evidence from Nigeria. Asian Economic and Financial Review, 2(7), 813-820.

Issa, H., & Ouattara, B. (2005). The effect of private and public health expenditure on infant mortality rates: does the level of development matters. University of Wales Swansea.

Iyer, T. (2009). Is public expenditure on primary education effective? Evidence from districts across India.

Jaffry, S., Ghulam, Y., Shah, V., & Iqbal, N. (2007). Returns to Education in Pakistan [with Comments]. The Pakistan Development Review, 833-852.

226

Jamal, H., Toor, I. A., & Khan, F. S. (2003). Private Returns to Education: Evidence for Pakistan. Social Policy and Development Centre.

Jamison, D.T., Lau, L.J., and Wang, J. (1998). Health’s Contribution to Economic Growth, 1965-1990, in Health, Health Policy and Health Outcomes: Final Report, Health and Development Satellite WHO Director-General Transition Team (Geneva: World Health Organization): 61-80. Johnson, G. E., & Stafford, F. P. (1973). Social returns to quantity and quality of schooling. Journal of Human Resources, 139-155.

Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of economic dynamics and control, 12(2-3), 231-254.

Johansen, S., & Juselius, K. (1990). Maximum likelihood estimation and inference on cointegration—with applications to the demand for money. Oxford Bulletin of Economics and statistics, 52(2), 169-210.

Junankar, P. N. (2003). Estimating the social rate of return to education for Indigenous Australians. Education economics, 11(2), 169-192.

Kakar, Z. K., Khilji, B. A., & Jawad, M. (2011). Relationship between Education and Economic Growth in Pakistan: A time series analysis. Journal of International Academic Research, 11(1), 123-132. Karatzas, G. (2000). On the determination of the US aggregate health care expenditure. Applied Economics, 32(9), 1085-1099. Kendrick, J. W. (1976). The formation and stocks of total capital. NBER Books.

Khan, S. R., Irfan, M., & Cohen, S. I. (1985). Rates of Returns to Education and the Determinants of Earnings in Pakistan [with Comments]. The Pakistan Development Review, 24(3/4), 671-683.

Khan, F. S., & Toor, I. A. (2003). Changes in Returns to Education in Pakistan: 1990-2002. The Lahore Journal of Economics, 8(2), 83-98.

227

Khattak, N. U. R., & Khan, J. (2012). The contribution of education to economic growth: evidence from Pakistan. International Journal of Business and Social Science, 3(4), 92-105.

Kim, T. K., & Lane, S. R. (2013). Government health expenditure and public health outcomes: a comparative study among 17 countries and implications for US health care reform. American International Journal of Comtemporary Research, 3(9), 8-13.

Kiker, B.F. (1968). A Criterion for Evaluating Investment in Health Care. Review of Social Economy, 26(2), 176-182.

Kimenyi, m. S., Mwabu, G. & Manda, D. K. (2006). Human capital Externalities and Private Returns to Education in Kenya. Eastern Economic Journal, 32(3), 493-513.

Klasen, S. (2002). Low schooling for girls, slower growth for all? Cross‐country evidence on the effect of gender inequality in education on economic development. The World Bank Economic Review, 16(3), 345-373.

Klazar, S., Sedmihradský, M., & Vančurová, A. (2001). Returns of education in the Czech Republic. International Tax and Public Finance, 8(4), 609-620.

Komlos, J. (Ed.). (1994). Stature, living standards, and economic development: Essays in anthropometric history. University of Chicago Press.

Lach, Ł. (2010). Fixed capital and long run economic growth: evidence from Poland. Systems Science, 36(4), 32-50.

Lal, I., Muhammad, S. D., Jalil, M. A., & Hussain, A. (2010). Test of Okun’s law in some Asian countries co-integration approach. European journal of scientific research, 40(1), 73-80.

Lazear, E. (1977). Education: consumption or production?. Journal of Political Economy, 85(3), 569-597.

Lenti, L. (1967). Gli investimenti in capitale umano. L’industria, 2, 145-148.

Levin et al. (2005)

228

Lhila, A., & Simon, K. I. (2008). Prenatal health investment decisions: Does the child’s sex matter?. Demography, 45(4), 885-905.

List, F. (1928). The National System of Political Economy, Translated by Sampson, S. L. New York: Longmans, Green & Co.

Lucas, R. (1988). On the mechanisms of development planning. Journal of Monetary Economics, 22(1), 3-42.

Lucas, R. E. (1992). Making a. Miracle’, Econometrica, 61(2), 251-72.

Luce, B. R., Mauskopf, J., Sloan, F. A., Ostermann, J., & Paramore, L. C. (2006). The return on investment in health care: from 1980 to 2000. Value in Health, 9(3), 146-156.

Maazouz, M. (2013). Return to investment in human capital and policy of labour market: Empirical analysis of developing countries. Procedia Economics and Finance, 5, 524-531.

Macinko, J., Guanais, F. C., & De Souza, M. D. F. M. (2006). Evaluation of the impact of the Family Health Program on infant mortality in Brazil, 1990–2002. Journal of Epidemiology & Community Health, 60(1), 13-19.

Maksymenko, S. V., & Rabbani, M. (2008). Economic reforms, human capital, and economic growth in India and South Korea: a cointegration analysis.

Mankiw, N. G., Romer, D., & Weil, D. N. (1992). A contribution to the empirics of economic growth. The quarterly journal of economics, 107(2), 407-437.

Martorell, R., & Habicht, J. P. (1986). Growth in early childhood in developing countries.

Mayer, D. (2001). The long-term impact of health on economic growth in Latin America. World development, 29(6), 1025-1033.

McCulloch, J. R. (1870). The Principles of Political Economy, Etc. A. Murray & Son.

McGuire, A., Parkin, D., Hughes, D., & Gerard, K. (1993). Econometric analyses of national health expenditures: can positive economics help to answer normative questions?. Health economics, 2(2), 113-126.

229

McMahon, W. W. (2000). Education and development: Measuring the social benefits. Clarendon Press.

Meidani, A. A. N., & Zabihi, M. (2011). The dynamic effect of unemployment rate on per capita real gdp in iran. International journal of Economics and Finance, 3(5), 170.

Mekdad, Y., Dahmani, A., & Louaj, M. (2014). Public spending on education and economic growth in Algeria: Causality test. International Journal of Business and Management, 2(3), 55.

Mill, J. S. (1909). Principles of Political Economy, New York: Longmans, Green & Co.

Mincer, J. (1958). Investment in human capital and personal income distribution. Journal of political economy, 66(4), 281-302.

Mincer, J. (1974). Schooling, Experience, and Earnings. Human Behavior & Social Institutions No. 2.

Mirvis, D. M., Chang, C. F., & Cosby, A. (2008). Health as an economic engine: evidence for the importance of health in economic development. Journal of Health and Human Services Administration, 30-57.

Mohapatra, S., & Luckert, M. K. (2012). Uncertain educational returns in a developing economy. International Journal of Educational Development, 32(4), 590-599.

Murillo, I. P., Rahona-López, M., & del Mar Salinas-Jiménez, M. (2012). Effects of educational mismatch on private returns to education: An analysis of the Spanish case (1995–2006). Journal of Policy Modeling, 34(5), 646-659.

Mushkin, S. J. (1962). Health as an Investment. Journal of political economy, 70(5, Part 2), 129- 157.

Muysken, J. (2003). Health as a Principal Determinant of Economic Growth(No. 024). Maastricht University, Maastricht Economic Research Institute on Innovation and Technology (MERIT).

Nasir, Z. M., & Ali, S. Z. (1999). Do Private Schools Produce More Productive Workers?[with Comments]. The Pakistan Development Review, 937-954.

230

Nasir, Z. M., & Nazli, H. (2000). Education and earnings in Pakistan, Pakistan Institute of Development Economics, 3(177), 95-110.

Nelson, R. R., & Phelps, E. S. (1966). Investment in humans, technological diffusion, and economic growth. The American economic review, 56(1/2), 69-75.

Newhouse, J. (1977). Medical care expenditure: A cross-national survey, Journal of Human Resources, 4(4), 115-125. Nicholson, J. S. (1891). The living capital of the United Kingdom. The Economic Journal, 1(1), 95-107.

Nixon, J., & Ulmann, P. (2006). The relationship between health care expenditure and health outcomes. The European Journal of Health Economics, 7(1), 7-18.

Noor, Z. M., Nor, N. M., & Ghani, J. (2007). The relationship between output and unemployment in Malaysia: Does Okun’s law exist. International Journal of Economics and Management, 1(3), 337-344.

Okunade, A. A., & Karakus, M. C. (2001). Unit root and cointegration tests: timeseries versus panel estimates for international health expenditure models. Applied Economics, 33(9), 1131- 1137.

Okunade, A. A. (2005). Analysis and implications of the determinants of healthcare expenditure in African countries. Health Care Management Science, 8(4), 267-276.

Oloo, J. A. (2005). Child mortality in developing countries: Challenges and policy options. Eastern Africa Social Science Research Review, 21(2), 1-17.

Ongo, E. N., & Vukenkeng, A. W. (2014). Does gross capital formation matter for economic growth in the CEMAC sub-region?. EuroEconomica, 33(2), 54-60.

Pakistan National Human Development Report (2002)

Parkin, D., McGuire, A., & Yule, B. (1987). Aggregate health care expenditures and national income: is health care a luxury good?. Journal of health economics, 6(2), 109-127.

231

Pastore, F., & Verashchagina, A. (2006). Private returns to human capital over transition: A case study of Belarus. Economics of Education Review, 25(1), 91-107.

Patrinos, H. A., & Psacharopoulos, G. (2010). Returns to education in developing countries.

Perkins, D., Radelet, S., & Snodgrass, D. R. M. Gillis, M. Roemer, (2001) Economics of Development.

Petty W. Political arithmetick [1676; published 1690]. Psacharopoulos, G. (1981). Returns to education: an updated international comparison. Comparative education, 17(3), 321-341.

Psacharopoulos, G. (1994). Returns to investment in education: A global update. World development, 22(9), 1325-1343.

Psacharopoulos, G. (1995). The Profitability of Investment in Education: Concepts and Methods Mimeo, Human Capital Development and operational Policy Working Paper, December, World Bank, Washington.

Psacharopoulos, G., & Patrinos, H. A. (2004). Returns to investment in education: a further update. Education economics, 12(2), 111-134.

Psacharopoulos, G. (2006). The value of investment in education: Theory, evidence, and policy. Journal of Education Finance, 113-136.

Qiu, T., & Hudson, J. (2010). Private returns to education in urban China. Economic change and restructuring, 43(2), 131-150.

Qureshi, M. G. (2012). The gender differences in school enrolment and returns to education in Pakistan. The Pakistan Development Review, 219-256.

Rebelo, S. (1991). Long-run policy analysis and long-run growth. Journal of political Economy, 99(3), 500-521.

Rehman, S. U., & Ahmad, H. K. (2016). The impact of foreign capital inflows on economic growth: pooled mean group analysis for developing countries. Pakistan Economic and Social Review, 54(2), 191-201.

232

Reeves, A., Basu, S., McKee, M., Meissner, C., & Stuckler, D. (2013). Does investment in the health sector promote or inhibit economic growth?. Globalization and health, 9(1), 43-51.

Rodgers, G. B. (1979). Income and inequality as determinants of mortality: an international cross-section analysis. Population studies, 33(2), 343-351.

Romer, P. M. (1990). Endogenous technological change. Journal of political Economy, 98(5, Part 2), S71-S102.

Romer, P. M. (1990) 'Human capital and growth: theory and evidence', Carnegie- Rochester Conference Series on Public Policy,32, 251-86. Rosenzweig, M. R., & Schultz, T. P. (1982). Child mortality and fertility in Colombia: individual and community effects. Health policy and education, 2(3-4), 305-348.

Rosenzweig, M. R., & Schultz, T. P. (1983). Estimating a household production function: Heterogeneity, the demand for health inputs, and their effects on birth weight. Journal of political economy, 91(5), 723-746.

Rosenzweig, M. R., & Wolpin, K. I. (1988). Heterogeneity, intrafamily distribution, and child health. Journal of Human Resources, 437-461.

Ross, C. E., Wu, C. L. (1995). The links between Education and Health. American Sociological Review, 60(5), 719-745.

Roscher, W. G. F. (1878). Principles of Political Economy, Translated by John J. Lalor. Chi- cago: Callaghan & Co.

Sahn, D. E., & Alderman, H. (1988). The effects of human capital on wages, and the determinants of labor supply in a developing country. Journal of Development Economics, 29(2), 157-183.

Saraswati, E. (2012). Public Spending Education and Inequality: A Case Study in Indonesia. International Journal of Social Science and Humanity, 2(5), 427-435.

Saruparia, C., & Lodha, S. L. (2013). Returns from Investment in Technical and Professional Education with Reference to JNV University, Jodhpur (Rajasthan) India.

233

Savedoff, W. D., Schultz, T. P. (2000). Wealth from health. Washington, DC: InterAmerican Development Bank.

Say, J. B. (1821). A Treatise on Political Economy, Translated by C. R. Prinsep. Boston: Wells & Lilly. 1.

Schmookler, J. (1966). Invention and economic growth.

Schady, N. R. (2000). The political economy of expenditures by the Peruvian Social Fund (FONCODES), 1991–95. American Political Science Review, 94(2), 289-304.

Schündeln, M., & Playforth, J. (2014). Private versus social returns to human capital: Education and economic growth in India. European Economic Review, 66, 266-283.

Schultz, T. W. (1959). Investment in man: An economist's view. Social Service Review, 33(2), 109-117.

Schultz, T. W. (1960). Capital formation by education. Journal of political economy, 68(6), 571- 583.

Schultz, T. W. (1961). Investment in human capital. The American economic review, 51(1), 1-17.

Schultz, T. P. (1984). Studying the impact of household economic and community variables on child mortality. population and Development Review, 10, 215-235.

Schultz, T. P. (1993). Investments in the schooling and health of women and men: quantities and returns. Journal of human resources, 694-734.

Schultz, T. P. (1997). Demand for children in low income countries. Handbook of population and family economics, 1, 349-430.

Schultz, T. P., & Tansel, A. (1997). Wage and labor supply effects of illness in Cote d'Ivoire and Ghana: instrumental variable estimates for days disabled. Journal of development economics, 53(2), 251-286.

Schultz, T. P. (1999). Health and Schooling Investment in Africa. The Journal of Economic Perspectives, 13(3) , 67-88.

234

Schultz, T. P. (2002). Why governments should invest more to educate girls. World Development, 30(2), 207-225.

Schultz, T. P. (2003). Human capital, schooling and health. Economics & Human Biology, 1(2), 207-221.

Sendi, P. (2008). Bridging the gap between health and non-health investments: moving from cost-effectiveness analysis to a return on investment approach across sectors of economy. International journal of health care finance and economics, 8(2), 113-121.

Sen, A. (1998). Mortality as an indicator of economic success and failure. The economic journal, 108 (446), 1-25.

Sen, A. (1999). Development as Freedom, Oxford University Press, UK.

Senior, N. W. (1939). An Outline of the Science of Political Economy, New York: Farrar & Rinehart.

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.

Shandra, C. L., Shandra, J. M., & London, B. (2011). World bank structural adjustment, water, and sanitation: A cross-national analysis of child mortality in Sub-Saharan Africa. Organization & Environment, 24(2), 107-129.

Sherman, A. (1994). Wasting America's future: The Children's Defense Fund report on the costs of child poverty. Beacon Press.

Siddiqui, R., Afridi, U., Haq, R., & Tirmazi, S. H. (1995). Determinants of Expenditure on Health in Pakistan [with Comments]. The Pakistan development review, 34(4), 959-970.

Sidgwick, H. (1901). The Principles of Political Economy. London: Macmillan & Co.

Siphambe, H. K. (2000). Rates of return to education in Botswana. Economics of Education Review, 19(3), 291-300.

235

Smith, A. (1937). An Inquiry into the Nature and Causes of the Wealth of Nations. The Modern Library, New York, New York, 15.

Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations, London: George Routledge & Sons. 1903. 78-79.

Sohn, K. (2013). Monetary and nonmonetary returns to education in Indonesia. The Developing Economies, 51(1), 34-59.

Solow, R. M. (1956). A contribution to the theory of economic growth. The quarterly journal of economics, 70(1), 65-94.

Solow, R. M., Arrow, K. J., Chenery, B. S., & Minhas, B.S. (1961). Capital labor substitution and economic efficiency, The review of Economics and Statistics, 43, 56-98.

Stange, K. (2015). Differential pricing in undergraduate education: Effects on degree production by field. Journal of Policy Analysis and Management, 34(1), 107-135.

Stark, W. (1952). Jeremy Bentham's Economic Writing. London: George Allen & Unwin.

Strauss, J., Gertler, P. J., Rahman, O., & Fox, K. (1993). Gender and life-cycle differentials in the patterns and determinants of adult health. Journal of Human Resources, 791-837.

Steckel, R. H. (1995). Stature and the Standard of Living. Journal of economic literature, 33(4), 1903-1940.

Thomas, D., & Strauss, J. (1997). Health and wages: Evidence on men and women in urban Brazil. Journal of econometrics, 77(1), 159-185.

Todaro, M. P. (1981). Economic development in the third world (No. 338.9 T6 1981).

Toor, I. A., & Butt, M. S. (2004). Health Demand and Outcomes in Pakistan. The Lahore Journal of economics, 9(1), 7-31.

United Nations Development Programme, (2017).

Upadhyay, A. K., Srivastava, S. (2015). International Institute for Population Sciences Mumbai, India.

236

Vella, F., & Gregory, R. G. (1996). Selection bias and human capital investment: Estimating the rates of return to education for young males. Labour Economics, 3(2), 197-219.

Vila, L. E. (2000). The non-monetary benefits of education. European journal of education, 35(1), 21-32.

Warunsiri, S., & McNown, R. (2010). The returns to education in Thailand: A pseudo-panel approach. World Development, 38(11), 1616-1625.

Webber, D. J. (2002). Policies to stimulate growth: should we invest in health or education?. Applied Economics, 34(13), 1633-1643.

Wei, H. (2010). Measuring economic returns to post-school education in Australia. Australian Bureau of Statistics Research Paper.

Weisbrod, B. A. (1962). Education and investment in human capital, Journal of Political economy, 70(5), 106-123.

Wheatley, L. (2015). Factors affecting child mortality.

WHO. (2001). Macroeconomics and health, Investing in Health for Economic Development. Geneva Switzerland: WHO.

Winter-Ebmer, R., & Wirz, A. (2002). Public funding and enrolment into higher education in Europe (No. 503). IZA Discussion Paper Series.

Wittstein, T. (1867). Mathematische Statistik und deren Anwendung auf National-Ökonomie und Versicherungs-Wissenschaft. Hahn'sche Hofbuchhandlung.

Wobst, P., & Seebens, H. (2005). The impact of increased school enrollment on economic growth in Tanzania. African Development Review, 17(2), 274-301.

Wolfe, B. L. (1986). Health status and medical expenditures: is there a link?. Social science & medicine, 22(10), 993-999.

Woods, E. A., & Metzger, C. B. (1927). America's Human Wealth: The Money Value of Human Life. FS Crofts & Company.

237

World Health Organization. Scientific Group on Epidemiology, & Prevention of Periodontal Diseases. (1978). Epidemiology, Etiology, and Prevention of Periodontal Diseases: Report of a WHO Scientific Group (No. 621). World Health Organization.

World Bank (2007). World development report.

World Bank Report. (2016).

Yamada, T. (1985). Causal relationships between infant mortality and fertility in developed and less developed countries. Southern Economic Journal, 364-370.

Yan, W., & Yudong, Y. (2003). Sources of China's economic growth 1952–1999: incorporating human capital accumulation. China Economic Review, 14(1), 32-52.

Younger, S. D. (2001). Cross-country determinants of declines in infant mortality: A growth regression approach.

Zaidi, S. A., & Sahibzada, S. A. (1986). Issues in Pakistan's Health Sector [with Comments]. The Pakistan Development Review, 25(4), 671-682.

Zhu, S., Wu, Q. J., & Wang, Y. (2011). Impact of labor force on China's economic growth based on grey metabolic GM (1, 1) model. In Proceedings of 2011 IEEE International Conference on Grey Systems and Intelligent Services, 262-265.

Ziderman, A. (1973). Rates of return on investment in education: recent results for Britain. Journal of Human Resources, 85-97.

238

Appendix: 1

QUESTIONNAIRE

This survey is conducted for academic purpose, to evaluate the returns of investment in education and health in Pakistan. You are requested to participate in this survey, all the information given by you will always remain confidential and will only be used for research purpose. Please encircle the relevant code e.g. 1, 2,3,4,5 etc throughout the questionnaire.

SECTION-I (PERSONAL INFORMATION) P1 Date of interview (dd/mm/yyyy) ------/------/2016 P2 Age ______Years ______Months P3 Religion 01. Muslim 02. Non-Muslim P4 Do you belong to? 01. Urban area 02. Rural area P5 Town/ Location where you live Town: ______Union Council: ______P6 Your gender? 01. Male 02. Female P7 Your current profession and designation? profession:______Designation:______Name of institution/ organization where P8 employed Town/ location in which institution/ P9 organization falls 01.Federal government 02. Provincial government P10 Type of institution where employed 03.Semi government 04. Private 05.Other (please specify) ______01. Permanent 05. Contract 02. Temporary 06. Daily Wages P11 Nature of Job 03. Part time 07. Other (Please Specify) 04. Full time ______How much amount you pay in taxes P12 ______Rs. annually? Net Salary in Rs. ______(Per Month) Income from other sources (If any) in Rs. ______P13 Per month Income Please specify other sources: 01. ______02. ______01. Less than 6 hrs 02. 06-08 hrs 03. 08-10 hrs P14 Daily working hours? 04.10-12 hrs 05.More than 12 hrs SECTION-II (QUALIFICATION, EXPERIENCE, TRAINING/ SKILLS ACQUIRED) Total number of years of Q1 ______Years education acquired Age at the time of admission in Q2 ______Years school Degree/ Certificate: ______Last/ terminal degree/ certificate Subject: ______Institution: ______Q3 obtained Country: ______Percentage/ Grade: ______Examination system: 01. Annual system 02. Semester system When did you qualify/ Pass Year: ______Percentage: ______Q4 matriculation/ O-level Subjects:______examination Your High School/ O-level 01. Urban Area Q5 school was situated in: 02. Rural Area

239

Q6 Which was your medium of 01. English 02. Urdu instruction in High school? 03. Any other language (Please specify) ______Do you have any kind of Q7 01. Yes 02. No diploma? Total working experience? Years: ______Q8 (from first job to the present Months: ______one) Did you have any local/ foreign 01. Yes Q9 training which is helpful for the 02. No job? SECTION-III (RESPONDENT’S HOUSEHOLD INFORMATION) 01. Yes 02. No (if answer is ‘Yes’ then skip to HH1 Are you head of household? HH4) HH2 Gender of head of household 02. Male 02. Female HH3 Income of head of household Rs. / month. ______Total number of persons in HH4 house 01. English 02. Urdu 03. Saraiki 04. Punjabi HH5 Your mother tongue 05.Any other regional language (Please Specify) ______Mother Father 01. Illiterate 01. Illiterate 02. Primary 02. Primary 03. Matriculation 03. Matriculation HH6 Your mother, father education 04. Intermediate 04. Intermediate 05. Graduate 05. Graduate 06. Masters 06. Masters 07. Mphil 07. Mphil 08. Ph.D 08. Ph.D 09. Any other (specify):_____ 09. Any other (specify):_____ HH7 Your Parents occupation Mother: ______Father: ______01. Married 03. Widowed HH8 Your marital status? 02. Unmarried 04. Divorced (if answer is ‘unmarried’ then skip to HH16) Male Children: ______Female children: ______HH9 Number of children None: ______(If ‘None’ then skip to HH12)

01. Under 05 years:______02. 05-09: ______Please specify number of HH10 03.10-12: ______04. 13-15: ______children in each age group 05. 16-18: ______06. Above 19 years:______If any child of you is earning hand, then please mention his/ her profile. Sr. No. Age Years of education Occupation Sector Income (Rs./month)

HH11

Spouse total number of years of HH12 ______years education acquired. HH13 Is spouse employed? 01. Yes 02. No HH14 Per month income of spouse HH15 Occupation of Parents in law Mother in law: ______Father in law: ______

240

01. Own house 02. Rented 03. Govt. accommodation HH16 Your residence 04.Other (please specify): ______HH17 What is area of your house? ______Marla Nature of asset Quantity Value of assets If you have some assets then Land (Acres) HH18 mention their quantity and value Livestock Other (specify)______

Taken all things together, how 1. Happy 2. unhappy HH19 would you rank happiness in

your life? SECTION-IV HEALTH (ILLNESS, EXPENDITURE AND ENVORONMENT) HE1 What you say about your health in general? 1.Excellent 2. Very good 3. Good 4. Fair 5. Poor HE2 Have you had any major illness (Like Heart disease, Cancer, Arthritis, 1. Yes Diabetes, High blood pressure etc ) during this Year? 2. No HE3 What is the nature of this illness and how many days did you remain Nature: ______absent from job due to this illness during whole year? Days missed: ______HE4 How much amount did you spent on treatment of this major disease in ______Rs. whole year? HE5 Have you suffered from disability of doing any physical activity (Like 1. Yes bending, walking or climbing upstairs etc) due to illness or injury in this 2. No year? HE6 What is the nature of this physical activity and how many days during a Nature: ______year did you remain absent from job due to this problem? Days missed: ______HE7 How much amount did you spent on treatment of this illness in whole ______Rs. year? HE8 Your height? ______feet ______inch HE9 Your Weight? ______pound or ______kg HE10 Are public health units available in your area? 1. Yes 2. No HE11 What is your source of medication? 1. Government 2. Private HE12 How much time you visit a health unit per month? HE13 How far is the nearest health unit from your house? ______km. HE14 How much amount you spent on health practitioner’s fees every month? ______Rs. HE15 How much amount you spent on purchasing prescribed and non- ______Rs. prescribed medicines in this year? HE16 Have you spent some money for precautionary measures against viral 1. Yes 2. No and other infections? If yes, specify the amount spent. ______Rs. HE17 Do you have complete availability of healthy diet? 1. Yes 2. No HE18 How much you have food expenditures (on average per day)? ______Rs. HE19 During a week, what is your expenditure on following food items per person on average: Sr. No Food Item Expenditure (Rs./week) 1 Dairy products & eggs 2 Fruits (fresh + dry) and vegetables 3 Cereal grains and products (Wheat, Rice, Barley, Millet etc) 4 Alcoholic drinks & Non alcoholic soft drinks 5 Meat and pulses 6 Tobacco 7 Food purchased from stores

HE20 Amount spends on miscellaneous health care expenditure. ______Rs.

241

HE21 Amount you spend on recreation in last four weeks ______Rs. HE22 How much time you spend for exercise daily? ______Hours HE23 What is the hygienic condition of drinking water? 1. good 2. bad HE24 What is the source of drinking water? 1. Hand pump 2. Electric motor 3. Public water system outside house 4. Open well HE25 What is the condition of your street? 1. Clean 2. Dirty HE26 Have you latrine in your house since last five years? 1. Yes 2. No Appendix: 2 MDGs Targets for Pakistan

MDG Targets up to Indicators 1990 2015 Infant Mortality Rate 120 40 (/1000 live births) Under five mortality Rate (/1000 live 140 47 births) Fully immunized children 25 >90 (12-23 months) (%) Births attended by skilled health staff N/A 90 (%) Maternal Mortality Ratio (/100,000 live 550 140 births) Prevalence of 12 553 Contraceptive (%) Source: Progress on Agenda for Health Sector Reform, Ministry of Health. Appendix: 3 Targets for the Vision 2010 of Health Policy 1997

Index 1998 2003 2010 Rate of Maternal Mortality 350 200 90 (per ten thousand) Infant Mortality Rate (per one thousand) 86 40 20

Children below one year 65 90 100 fully immunized (%) Life Expectancy at birth 62 65 69

Immunization of Mothers against 60 80 100 tetanus (%) Polio Eradication By Year 2000 Low birth weight babies 25 10 05 (%) Trained staff attending pregnancy at birth 20 70 100 (%)

Goiter prevalence Rate 15 10 10 (%) Iron Deficiency Anemia

242

(%) 40 20 5 i) Woman 30 20 5 ii)Children Oral Rehydration Therapy use (%) 70 90 100

Dental Surgeons 3000 6000 15000

Doctors 75000 133000 142000

Paramedics 115000 170000 215000 Nurses 24800 35000 50000

Community Health Workers (Female) 45000 60000 100000

Traditional Birth 50000 75000 65000 Attendant Source: Health Policy 1997

Appendix: 4 Detail of the Allocation for Sub-Sectors of Health under 4th Plan

Programs West Pakistan East Pakistan Center Total Eradication of Malaria 140 120 2 262 Health Programs in Rural Areas 300 400 --- 700 Tuberculosis 25 20 6 51 Communicable diseases 65 100 3 168 Beds in Hospital 225 440 49 714 Medical education 110 275 52 437 Research in Health ------8 8 Loan to private sector 35 45 --- 80 Centrally administered ------10 10 Regions Azad Kashmir Northern ------15 15 Regions Grand Total 900 1400 145 2445 Source: 4th Plan

Appendix: 5 Physical Achievements of Health Sector during 4th Five Year Plan

Estimated Benchmark S.No Sub-Sector achievements 1969-70 during 1970-78 1 Doctors 13400 9362 2 Lady Health Visitors 1881 1369 3 Training Centers for Nurses 21 7 4 Beds in Hospitals 32063 14029 5 Nurses 5400 4311 6 Sub-Centers of RHCs 250 491 7 Rural Health centers 86 203 8 Training Schools for Auxiliaries --- 4 9 Medical Colleges 6 9 Source: 4th Five Year Plan

243

Appendix: 6 Physical Targets of the Fifth Five Year Plan

S.No Item Benchmark Target End Position Physical Facilities 1 Rural Health Centers 289 625 914 2 Beds in Hospitals 46092 25820 71912 3 Basic Health units 5850 4596 10446 Health Manpower 1 Dentists 1047 595 1642 2 Doctors 12924 12917 25841 3 Nurses 4300 4780 9080 4 Dispensers 9000 --- 9000 Community Health 5 1621 50371 51992 Worker 6 Paramedical 15428 24886 40314 Source: 5th Five Year Plan

Appendix: 7 Infrastructures and Manpower Development during 6th Plan

Cumulative Facility Targets Total June, 1988 A Infrastructure i) Beds in Hospital 11770 63170 ii) Basic Health Units 2600 4315 iii) Rural Health Centers 355 729 iv) Dispensaries 2600 2600 B Manpower Development i) Dentists 600 1700 ii) Nurses 5000 10000 iii) Doctors 21000 36000 iv) Community health workers 30000 45000 v) Paramedics 38000 75000 Source: 6th Five Year Plan

Expenditures on Health during 6th Five Year Plan

S.No Sub-Sector Million Rs. % 1 Beds in Hospitals 3295 25.35 2 Rural Health Program 5660 43.54 3 Preventive Programs 1490 11.46 4 Medical education 975 7.5 5 Medical Research 85 0.65 6 Dental care etc. 250 1.92 7 Traditional Medicine 375 2.89 8 Nutrition programs 250 1.92 9 Disabilities 500 3.85 10 Others 120 0.62 Total 13000 100 Source: 6th Five Year Plan

244

Appendix: 8 Year Wise Expenditures on Health Sector during 7th Five Year Plan

Total Program 1988-89 19989-90 1990-91 1991-92 1992-93 (Rs. Million)

Hospital Beds 502 574 626 697 718 3117

Preventive 257 210 194 183 180 1024 Programs

Rural Health 1367 1330 1046 960 952 5655

Urban Health 42 45 189 201 215 692 Centers

Nutrition 10 35 49 58 60 212 Programs

Health Manpower 476 394 421 455 412 2158 Development

Traditional 8 37 47 45 51 188 Medicine

Medical 2 22 32 42 62 160 Rehabilitation 23 24 22 35 40 144 Others 2687 2671 2626 2676 2690 13350 Total Source: 7th Five Year Plan

Appendix: 9 Physical Targets of 8th Five Year Plan

S. No Sub-sector Targets

1 Training of Traditional Birth Attendant 57744

2 Immunization of Children ( in Millions) 22.08

3 ORS (Million Packets) 92.5

4 Health Workers 37811

5 New Rural Health Centers 45

6 New Basic health Units 252

7 Improvement of Basic Health Units 3874

8 Mohall Health Centers 616

9 Improvement of Rural Health Centers and Civil 492 Hospitals 10 616 Mobile Health Centers

245

11 40 Mobile Dispensaries 12 13 Doctors 17300 Doctors

13 Hospital beds 21500 14 925 Dentist 15 48500 Paramedics 16 18000 Nurses Source: 8th Five Year Plan

Appendix: 10 Expenditure Estimation for the 8th Plan 1993-98

Cost S.No Description (Million Rs.)

Federal

1 Population Welfare Services Islamabad District 59.7

2 Target Group Institutions 8.8

3 Administrative Organization 284.1

4 Non-Governmental Organizations 300

MCH population welfare services in Northern 5 12.4 Areas

6 MCH population welfare services in Azad Jammu Kashmir 22.1

7 Non-Clinical Training through RTIs 100.3

8 Clinical Training through RTIs 200

9 NRIRP 1.3

10 NRIFC 49.8

11 NIPS 108

12 Monitoring and Research Centers 8

13 Social Marketing of Contraceptive 14

14 Population Study Center 0.75

15 Contraceptive Requirement & Distribution 1418

16 Construction (population houses etc) 149.64

17 Consultancy 5

18 Communication Strategy 380

246

19 Unallocated 90.55

Sub-Total 3212.64

Provincial

1 Family Welfare Centers 1570.573

2 Reproductive Health Services 989.541

Tehsil, District, Divisional and Provincial 3 1385.88 Setup

4 436.593 Mobile Service Units 5 Contribution of TBAs in population welfare 93.723 Program Contribution of Provincial 6 71.486 Line Department and Health Outlets in population welfare programs 7 Family Planning Workers 975.218 8 Participation of Registered Hakeems and Medical 41.939 Practitioners in population welfare programs 9 20.004 Innovatives 10 222.403 Communication Strategy 11 80 Construction 5887.360 Sub-Total 9100 Grand Total Source: 8th Five Year Plan

Appendix: 11 Formal Educational Institutions in Pakistan

Gilgit No. Punjab Balochistan Sindh FATA AJ&K KPK ICT Total Baltistan Primary 52414 11079 46759 4836 4852 11079 24991 364 146185 Schools Middle 26831 1406 5928 616 1848 427 4921 170 42147 Schools High Schools 17958 917 5189 439 1081 268 3774 248 29874 Colleges 1241 68 471 62 199 35 202 40 2318 Universities 43 6 40 --- 6 1 29 16 141 Source: NEMIS 2013 Appendix: 12 Enrolments in Formal Schools in Pakistan

Gilgit No. Punjab Balochistan Sindh FATA AJ&K KP ICT Total Baltistan Primary 9123952 670143 3821191 374994 376501 107990 2980910 119168 17574849 Schools Male 4868313 402433 2206311 253494 194055 61403 1785509 60785 9832303

247

Female 4255639 267710 1614880 121500 182446 46587 1195401 58383 7742546 Middle 3474610 1634911 1081979 75713 162356 45662 1050771 64615 6119197 Schools Male 1883408 105889 600420 64325 86615 26324 691501 32294 3490776 Female 1591202 57602 481559 11388 75741 19338 359270 32321 2628421 High 1599465 71790 581326 29295 70205 20970 424661 37614 2835326 Schools Male 881297 47319 334637 24904 39231 11815 288592 19213 1647008

Female 718168 24471 246689 4391 30974 9155 136069 18401 1188318

Source: NEMIS 2013 Appendix: 13 4th Five Year Plan (Public Sector Allocation for Education)

West East Millian % of S.No Sub-Sector Center Pakistan Pakistan Rupees total 1 Primary Level Education 115 405 27 547 14.6

Secondary Level Education 253 400 22 675 18.4 2 (a) Middle 75 200 12 287 7.8 (b) Higher 178 200 10 388 10.6

3 Education for Teachers 45 100 --- 145 4 4 Vocational Education 290 630 --- 920 25.2 5 Colleges 135 220 10 365 9.9 6 Universities 90 170 80 340 9.4 7 Scholarships 120 125 21 266 7.3 8 Madrasas --- 25 --- 25 0.7 Cultural and Social Activities 9 including Library Related 22 50 55 127 3.5 Services 10 Tribal Areas ------10 10 0.3 11 Special Areas 25 --- 15 40 1.1 12 Adult Education 35 50 --- 85 2.3 13 Miscellaneous Projects 5 5 --- 10 0.3 14 Research Development 15 20 20 55 1.5 15 Education on TV ------15 15 0.4 16 Statistics and Planning --- 30 --- 30 0.8 17 Publicity Scheme ------10 10 0.3 Total 1150 2230 285 3665 100 Source: 4th Five Year Plan Appendix: 14 Requirement of the Recurring Expenditure during 4th Plan (1970-75)

S.No Items East Pakistan West Pakistan Center Total 1969-70 Expenditure 1 extended over the 5-years of 1060 1850 110 3020 4th plan Expenses required for the 2 annual increments of the 90 350 5 445 educational staff in institutions The expenses required for additional staff and expenses on 3 1170 850 35 2055 introduction of New Pay Scales Total 2320 3050 150 5520 Source: 4th Five Year Plan

248

Appendix: 15 Financial Requirement Estimates for Education during 5th Plan

Non- Total Development S.No Name of the Sub-sector development (Rs in Expenditure Expenditure Million) 1 Primary Education 2508 7361.5 9869.5 2 Secondary Education 3475.7 4422.3 7898 3 Teachers Education 466 310 776 4 Vocational Education 779 630.1 1409.1 5 College level Education 867.5 1971.7 2839.2 6 University level Education 590 720.2 1310.2 7 Non Formal Education 89.7 6 95.6 8 Scholarship 576.5 49.6 627.1 Supply and Production of 9 42.3 --- 42.3 Books 10 Curriculum Development 22.5 2 24.5 i) Examination 6.9 1 7.9 Reforms 11 ii) Guidance & 6.9 6.5 71.9 Counseling 12 Libraries 34.1 5 39.1 13 Physical Education 425.5 5 430.5 14 Arts & Culture 43 2 45 15 Museums and Archeology 73.7 2 75.7 16 Special Education 34 10 44 17 Miscellaneous 93.6 1400 1494.6 Sub Total 10134.9 16963.8 27098.7 Other Divisions Establishment Division 35 1 36 Planning Division 35 --- 35 Technological and Scientific Research 25 1 26 Division Religious Affairs 25 1 26 Division Grand Total 10254.9 16966.8 27221.7 Source: 5th Five Year Plan

249

Appendix: 16 Allocations for Educations and Manpower in 6th Five Year Plan Total SUB-Sector Federal Punjab Sindh NWFP Baluchistan (Rs in Millions) Primary 3980 1320 770 740 190 7000 Education Secondary 945 1615 750 570 245 4125 Education Teacher 60 80 80 60 25 305 Education Technical 1155 700 180 180 120 2335 Education College 370 430 240 190 70 1300 Education University 2100 491 431 310 126 2100 Education Mass Literacy 750 ------750 Scholarship 395 170 35 60 50 660 Development of Public Library System 300 80 25 30 20 455 Miscellaneous 145 330 40 30 20 455 Expenses for Establishment 250 ------250 Division Total 10450 4725 2120 1860 695 19850 Source: 6th Five Year Plan

Appendix: 17 Expenditures for Public Sector Development Allocated for Education in 7th Five Year Plan (1988-93) S.No. Sub-sector Million Rupees Percentage 1 Primary level Education 10129 43.8 2 Secondary level Education 6539 28.3 3 Education for Teachers 287 1.3 4 College Education 2000 2.7 5 Vocational Education 615 8.6 Literacy & Mass 6 300 1.3 Education 7 Scholarships 760 3.3 8 Library System 181 8 9 University level Education 2000 8.6 10 Other Divisions 150 0.65 11 Other Programs 150 0.65 Total 23111 100 Source: 7th Five Year Plan

250

Appendix: 18 Expenditure Allocations for Education and Training during 8th Plan (In Million Rs.) Sub-sector Allocation SAP Component Secondary Schools 16521 ---- Primary Schools 32669 32669 Education for Teachers 3360 234 Vocational Education 2447 ---- Colleges 2507 ---- Universities 4100 ---- Scholarships 1400 ---- Literacy services and 200 ---- Museum Literacy And Mass 1750 1750 Education Establishment Division 200 ---- Education Foundation 3200 2560 Miscellaneous 677 ---- Total (Rupees in Million) 69031 39319 Source: 8th Five Year Plan

Appendix: 19 Targets for the Education Sector Reforms S.No Sub-Sector Bench Mark 2001 Target 2005 1 Literacy 49% 60% 2 Gross Enrollment at Primary Level 83% 100% 3 Net Enrollment at Primary Level 66% 76% 4 Enrollment in Middle Schools 47.5% 55% 5 Enrolment in Secondary Schools 29.5% 40% 6 Polytechnics/Mono-Technics 77 160 7 Technical Streams Schools 100 1100 8 Madaris Mainstreaming 148 8000 9 Higher Education Enrollment 2.6 5% 10 Public-Private Partnership 200 2600 Source: Education Sector Reforms 2001-05

Appendix: 20 Financial Requirements for ESR during 2001-05 S.No Programs 2001-02 2002-03 2003-04 2004-05 Total % Elementary level 1 4 9 10 11 34 34 Education Mainstreaming 2 0 5 5 4 14 14 Madaris 3 Literacy Campaign 0.8 2 2.5 3 8.3 8.3 Secondary level 4 1 3 3 3 10 10 Education College/higher 5 1 3 3 3 10 10 Education Vocational 6 0 3 5 7 15 15 Education Public-Private 7 0.1 0.2 0.2 0.2 0.7 0.7 Partnership 8 Quality Assurance 1 2 2 3 8 8 Total 7.9 27.2 30.7 34.2 100 100 Source: Education Sector Reforms 2001-05

251

Appendix 21: Minitab Results of Returns of investment to education and health: Primary data evidence from district Multan

1. Returns of investment to education Full sample Regression Analysis:

The regression equation is ln (Y) = 7.75 + 0.121 TED + 0.0630 EXR - 0.000790 EXQ

Predictor Coef SE Coef T P Constant 7.7542 0.1207 64.25 0.000 TED 0.121405 0.007146 16.99 0.000 EXR 0.062964 0.006063 10.39 0.000 EXQ -0.0007901 0.0001545 -5.11 0.000

S = 0.531486 R-Sq = 36.9% R-Sq(adj) = 36.7%

Analysis of Variance

Source DF SS MS F P Regression 3 139.982 46.661 165.18 0.000 Residual Error 846 238.976 0.282 Total 849 378.957 Regression Analysis: The regression equation is ln (Y) = 7.06 + 0.124 TED + 0.0624 EXR - 0.000683 EXQ + 0.162 WHR + 0.142 SBJCT + 0.0369 FQL - 0.0038 MQL

Predictor Coef SE Coef T P Constant 7.0557 0.1341 52.63 0.000 TED 0.124310 0.007191 17.29 0.000 EXR 0.062384 0.005825 10.71 0.000 EXQ -0.0006827 0.0001475 -4.63 0.000 WHR 0.16172 0.01972 8.20 0.000 SBJ 0.14172 0.03402 4.17 0.000 FQL 0.03688 0.01047 3.52 0.000 MQL -0.00381 0.01212 -0.31 0.753

S = 0.500017 R-Sq = 44.4% R-Sq(adj) = 44.0%

Analysis of Variance

Source DF SS MS F P Regression 7 168.443 24.063 96.25 0.000 Residual Error 842 210.514 0.250 Total 849 378.957 Regression Analysis:

The regression equation is ln (Y) = 7.74 + 0.121 TED + 0.0642 EXR - 0.000816 EXQ + 0.0663 MDU

Predictor Coef SE Coef T P Constant 7.7356 0.1210 63.92 0.000 TED 0.121229 0.007138 16.98 0.000 EXR 0.064227 0.006100 10.53 0.000 EXQ -0.0008159 0.0001551 -5.26 0.000 MDU 0.06634 0.03852 1.72 0.085

252

S = 0.530869 R-Sq = 37.2% R-Sq(adj) = 36.9%

Analysis of Variance

Source DF SS MS F P Regression 4 140.818 35.204 124.92 0.000 Residual Error 845 238.140 0.282 Total 849 378.957 Regression Analysis: The regression equation is ln (Y) = 8.92 + 0.298 MDL + 0.062 SEC + 0.227 HSE + 0.439 GRD + 0.863 UNI + 0.0661 EXR - 0.000855 EXQ

Predictor Coef SE Coef T P Constant 8.9243 0.1646 54.22 0.000 MDL 0.2976 0.2037 1.46 0.144 SEC 0.0623 0.1721 0.36 0.717 HSE 0.2268 0.1699 1.34 0.182 GRD 0.4390 0.1620 2.71 0.007 UNI 0.8630 0.1595 5.41 0.000 EXR 0.066097 0.006236 10.60 0.000 EXQ -0.0008554 0.0001603 -5.34 0.000

S = 0.541836 R-Sq = 34.8% R-Sq(adj) = 34.2%

Analysis of Variance

Source DF SS MS F P Regression 7 131.758 18.823 64.11 0.000 Residual Error 842 247.199 0.294 Total 849 378.957

Male sub sample

Regression Analysis: The regression equation is ln (Y) = 8.01 + 0.115 TED + 0.0485 EXR - 0.000524 EXQ Predictor Coef SE Coef T P Constant 8.0129 0.1473 54.40 0.000 TED 0.115265 0.007691 14.99 0.000 EXR 0.048506 0.008080 6.00 0.000 EXQ -0.0005245 0.0001811 -2.90 0.004

S = 0.471947 R-Sq = 36.0% R-Sq(adj) = 35.6% Analysis of Variance

Source DF SS MS F P Regression 3 61.955 20.652 92.72 0.000 Residual Error 495 110.254 0.223 Total 498 172.209 Regression Analysis: The regression equation is ln (Y) = 7.47 + 0.106 WHR + 0.115 TED + 0.0412 FQL + 0.0523 EXR - 0.000531 EXQ + 0.0028 MQL + 0.0910 SBJ Predictor Coef SE Coef T P Constant 7.4675 0.1685 44.33 0.000 WHR 0.10617 0.02340 4.54 0.000 TED 0.115298 0.007988 14.43 0.000

253

FQL 0.04124 0.01228 3.36 0.001 EXR 0.052250 0.007765 6.73 0.000 EXQ -0.0005312 0.0001736 -3.06 0.002 MQL 0.00283 0.01521 0.19 0.852 SBJ 0.09102 0.03516 2.59 0.010 S = 0.449646 R-Sq = 42.4% R-Sq(adj) = 41.5% Analysis of Variance

Source DF SS MS F P Regression 7 72.938 10.420 51.54 0.000 Residual Error 491 99.271 0.202 Total 498 172.209

Female sub sample

Regression Analysis: The regression equation is ln (Y) = 6.67 + 0.175 TED + 0.0546 EXR - 0.000198 EXQ Predictor Coef SE Coef T P Constant 6.6719 0.4274 15.61 0.000 TED 0.17514 0.02568 6.82 0.000 EXR 0.05462 0.02166 2.52 0.012 EXQ -0.0001982 0.0007505 -0.26 0.792 S = 0.673945 R-Sq = 28.8% R-Sq(adj) = 27.8% Analysis of Variance

Source DF SS MS F P Regression 3 37.667 12.556 27.64 0.000 Residual Error 205 93.111 0.454 Total 208 130.778 Regression Analysis: The regression equation is ln (Y) = 6.07 + 0.294 WRH + 0.150 TED + 0.0408 FQL + 0.0593 EXR- 0.000292 EXQ + 0.281 SBJ Predictor Coef SE Coef T P Constant 6.0670 0.4113 14.75 0.000 WRH 0.29434 0.06563 4.48 0.000 TED 0.14956 0.02448 6.11 0.000 FQL 0.04076 0.02278 1.79 0.075 EXR 0.05927 0.01975 3.00 0.003 EXQ -0.0002922 0.0006899 -0.42 0.672 SBJ 0.2814 0.1029 2.74 0.007 S = 0.612761 R-Sq = 42.0% R-Sq(adj) = 40.3% Analysis of Variance

Source DF SS MS F P Regression 6 54.9319 9.1553 24.38 0.000 Residual Error 202 75.8462 0.3755 Total 208 130.7781

Urban sub sample

Regression Analysis: The regression equation is ln (Y) = 7.64 + 0.126 TED + 0.0719 EXR - 0.00101 EXQ Predictor Coef SE Coef T P Constant 7.6409 0.1569 48.71 0.000 TED 0.125868 0.009534 13.20 0.000 EXR 0.071877 0.007762 9.26 0.000 EXQ -0.0010099 0.0001993 -5.07 0.000 S = 0.551269 R-Sq = 37.6% R-Sq(adj) = 37.3%

Analysis of Variance

Source DF SS MS F P Regression 3 99.646 33.215 109.30 0.000 Residual Error 544 165.320 0.304 Total 547 264.966

254

Regression Analysis:

The regression equation is ln (Y) = 6.97 + 0.176 WHR + 0.125 TED + 0.0418 FQL + 0.0676 EXR - 0.000834 EXQ - 0.0122 MQL + 0.138 SBJ Predictor Coef SE Coef T P Constant 6.9727 0.1725 40.41 0.000 WHR 0.17616 0.02601 6.77 0.000 TED 0.125139 0.009858 12.69 0.000 FQL 0.04176 0.01444 2.89 0.004 EXR 0.067601 0.007469 9.05 0.000 EXQ -0.0008340 0.0001910 -4.37 0.000 MQL -0.01225 0.01512 -0.81 0.418 SBJ 0.13842 0.04146 3.34 0.001 S = 0.519746 R-Sq = 44.9% R-Sq(adj) = 44.2% Analysis of Variance

Source DF SS MS F P Regression 7 119.093 17.013 62.98 0.000 Residual Error 540 145.873 0.270 Total 547 264.966

Rural sub sample

Regression Analysis: The regression equation is ln (Y) = 7.98 + 0.113 TED + 0.0455 EXR - 0.000363 EXQ Predictor Coef SE Coef T P Constant 7.9837 0.1885 42.36 0.000 TED 0.11281 0.01060 10.64 0.000 EXR 0.045542 0.009664 4.71 0.000 EXQ -0.0003626 0.0002429 -1.49 0.137 S = 0.491816 R-Sq = 36.7% R-Sq(adj) = 36.0% Analysis of Variance

Source DF SS MS F P Regression 3 41.753 13.918 57.54 0.000 Residual Error 298 72.081 0.242 Total 301 113.834 Regression Analysis:

The regression equation is ln (Y) = 7.24 + 0.137 whr + 0.119 TED + 0.0297 FQL + 0.0509 EXR - 0.000365 EXQ + 0.0305 MQL + 0.154 SBJ Predictor Coef SE Coef T P Constant 7.2406 0.2137 33.88 0.000 whr 0.13743 0.02978 4.61 0.000 TED 0.11864 0.01065 11.14 0.000 FQL 0.02969 0.01545 1.92 0.056 EXR 0.050947 0.009396 5.42 0.000 EXQ -0.0003653 0.0002323 -1.57 0.117 MQL 0.03047 0.02191 1.39 0.165 SBJ 0.15429 0.06205 2.49 0.013 S = 0.461959 R-Sq = 44.9% R-Sq(adj) = 43.6% Analysis of Variance

Source DF SS MS F P Regression 7 51.0930 7.2990 34.20 0.000 Residual Error 294 62.7414 0.2134 Total 301 113.8344

Married sub sample

Regression Analysis:

255

The regression equation is ln (Y) = 8.01 + 0.115 TED + 0.0485 EXR - 0.000524 EXQ

Predictor Coef SE Coef T P Constant 8.0129 0.1473 54.40 0.000 TED 0.115265 0.007691 14.99 0.000 EXR 0.048506 0.008080 6.00 0.000 EXQ -0.0005245 0.0001811 -2.90 0.004

S = 0.471947 R-Sq = 36.0% R-Sq(adj) = 35.6%

Analysis of Variance

Source DF SS MS F P Regression 3 61.955 20.652 92.72 0.000 Residual Error 495 110.254 0.223 Total 498 172.209 Regression Analysis: The regression equation is ln (Y) = 7.47 + 0.106 whr + 0.115 TED + 0.0412 FQL + 0.0523 EXR - 0.000531 EXQ + 0.0028 MQL + 0.0910 SBJ Predictor Coef SE Coef T P Constant 7.4675 0.1685 44.33 0.000 whr 0.10617 0.02340 4.54 0.000 TED 0.115298 0.007988 14.43 0.000 FQL 0.04124 0.01228 3.36 0.001 EXR 0.052250 0.007765 6.73 0.000 EXQ -0.0005312 0.0001736 -3.06 0.002 MQL 0.00283 0.01521 0.19 0.852 SBJ 0.09102 0.03516 2.59 0.010 S = 0.449646 R-Sq = 42.4% R-Sq(adj) = 41.5% Analysis of Variance

Source DF SS MS F P Regression 7 72.938 10.420 51.54 0.000 Residual Error 491 99.271 0.202 Total 498 172.209

Unmarried sub sample

Regression Analysis:

The regression equation is ln (Y) = 7.48 + 0.130 TED + 0.0920 EXR - 0.00203 EXQ

Predictor Coef SE Coef T P Constant 7.4844 0.2655 28.19 0.000 TED 0.13034 0.01553 8.39 0.000 EXR 0.09199 0.01925 4.78 0.000 EXQ -0.0020349 0.0008806 -2.31 0.021

S = 0.597773 R-Sq = 20.7% R-Sq(adj) = 20.0%

Analysis of Variance

Source DF SS MS F P Regression 3 31.926 10.642 29.78 0.000 Residual Error 343 122.565 0.357 Total 346 154.491

Regression Analysis: The regression equation is

256 ln (Y) = 6.78 + 0.226 whr + 0.123 TED + 0.0311 FQL + 0.0788 EXR - 0.00167 EXQ - 0.0060 MQL+ 0.260 SBJ

Predictor Coef SE Coef T P Constant 6.7832 0.2647 25.62 0.000 whr 0.22589 0.03460 6.53 0.000 TED 0.12323 0.01511 8.15 0.000 FQL 0.03109 0.01860 1.67 0.096 EXR 0.07880 0.01792 4.40 0.000 EXQ -0.0016745 0.0008204 -2.04 0.042 MQL -0.00600 0.01948 -0.31 0.758 SBJ 0.25969 0.07822 3.32 0.001

S = 0.551834 R-Sq = 33.2% R-Sq(adj) = 31.8%

Analysis of Variance

Source DF SS MS F P Regression 7 51.2587 7.3227 24.05 0.000 Residual Error 339 103.2325 0.3045 Total 346 154.4912 English Medium

Regression Analysis:

The regression equation is ln (Y) = 7.06 + 0.156 TED+ 0.105 EXR - 0.00179 EXQ

Predictor Coef SE Coef T P Constant 7.0600 0.5965 11.84 0.000 TED 0.15617 0.03468 4.50 0.000 EXR 0.10492 0.02501 4.19 0.000 EXQ -0.0017935 0.0007127 -2.52 0.013

S = 0.672527 R-Sq = 26.4% R-Sq(adj) = 24.7%

Analysis of Variance

Source DF SS MS F P Regression 3 21.8662 7.2887 16.12 0.000 Residual Error 135 61.0595 0.4523 Total 138 82.9257

Urdu medium

Regression Analysis: The regression equation is ln (Y) = 7.79 + 0.119 TED + 0.0603 EXR - 0.000728 EXQ

Predictor Coef SE Coef T P Constant 7.7912 0.1177 66.17 0.000 TED 0.119030 0.006963 17.10 0.000 EXR 0.060340 0.006136 9.83 0.000 EXQ -0.0007278 0.0001535 -4.74 0.000

S = 0.497656 R-Sq = 40.9% R-Sq(adj) = 40.6%

257

Analysis of Variance

Source DF SS MS F P Regression 3 120.933 40.311 162.77 0.000 Residual Error 707 175.097 0.248 Total 710 296.030 2. Returns of Investment to health

Anthropometric measures

Regression Analysis:

The regression equation is ln (Y) = 6.20 + 0.120 TED+ 0.0550 EXR - 0.000639 EXQ + 0.242 HET + 0.0131 BMI

Predictor Coef SE Coef T P Constant 6.2031 0.3433 18.07 0.000 TED 0.119575 0.007099 16.84 0.000 EXR 0.054970 0.006202 8.86 0.000 EXQ -0.0006389 0.0001558 -4.10 0.000 HET 0.24180 0.05278 4.58 0.000 BMI 0.013055 0.004197 3.11 0.002

S = 0.526432 R-Sq = 38.3% R-Sq(adj) = 37.9%

Analysis of Variance

Source DF SS MS F P Regression 5 145.059 29.012 104.69 0.000 Residual Error 844 233.898 0.277 Total 849 378.957

Male sub sample

Regression Analysis:

The regression equation is ln (Y) = 7.57 + 0.118 TED+ 0.0415 EXR - 0.000431 EXQ + 0.0519 HET + 0.00928 BMI

Predictor Coef SE Coef T P Constant 7.5740 0.3459 21.90 0.000 TED 0.117948 0.006395 18.45 0.000 EXR 0.041516 0.005977 6.95 0.000 EXQ -0.0004312 0.0001446 -2.98 0.003 HET 0.05189 0.005263 10.36 0.000 BMI 0.009280 0.004167 2.23 0.026

S = 0.441263 R-Sq = 42.2% R-Sq(adj) = 41.7%

Analysis of Variance

258

Source DF SS MS F P Regression 5 90.098 18.020 92.54 0.000 Residual Error 635 123.643 0.195 Total 640 213.741

Female sub sample

Regression Analysis:

The regression equation is ln (Y) = 5.24 + 0.171 TED + 0.0512 EXR - 0.000094 EXQ + 0.0376 HET + 0.0024 BMI

Predictor Coef SE Coef T P Constant 5.240 1.012 5.18 0.000 TED 0.17103 0.02653 6.45 0.000 EXR 0.05122 0.02173 2.36 0.019 EXQ -0.0000935 0.0007512 -0.12 0.901 HET 0.03761 0.0164 2.29 0.016 BMI 0.00244 0.01099 0.22 0.825

S = 0.672939 R-Sq = 29.7% R-Sq(adj) = 28.0%

Analysis of Variance

Source DF SS MS F P Regression 5 38.8502 7.7700 17.16 0.000 Residual Error 203 91.9278 0.4528 Total 208 130.7781

Nutritional factors Regression Analysis: The regression equation is ln (Y) = 7.68 + 0.116 TED+ 0.0606 EXR - 0.000762 EXQ + 0.000037 EXN + 0.120 HYW

Predictor Coef SE Coef T P Constant 7.6833 0.1292 59.45 0.000 TED 0.116379 0.007187 16.19 0.000 EXR 0.060571 0.006039 10.03 0.000 EXQ -0.0007624 0.0001537 -4.96 0.000 EXN 0.00003721 0.00000976 3.81 0.000 HYW 0.12016 0.05507 2.18 0.029

S = 0.528054 R-Sq = 37.9% R-Sq(adj) = 37.5%

Analysis of Variance

Source DF SS MS F P Regression 5 143.616 28.723 103.01 0.000 Residual Error 844 235.342 0.279 Total 849 378.957

259

Self reported health

Regression Analysis: The regression equation is ln (Y) = 7.87 + 0.121 TED + 0.0620 EXR - 0.000753 EXQ - 0.0457 HLS - 0.0409 DSB

Predictor Coef SE Coef T P Constant 7.8706 0.1267 62.14 0.000 TED 0.120998 0.007175 16.86 0.000 EXR 0.061975 0.006082 10.19 0.000 EXQ -0.0007530 0.0001553 -4.85 0.000 HLS -0.04565 0.01989 -2.29 0.022 DSB -0.04087 0.05672 -0.72 0.471

S = 0.532062 R-Sq = 37.0% R-Sq(adj) = 36.6%

Analysis of Variance

Source DF SS MS F P Regression 5 140.030 28.006 98.93 0.000 Residual Error 844 238.928 0.283 Total 849 378.957 Precautionary measures

Regression Analysis:

The regression equation is ln (Y) = 7.62 + 0.116 TED+ 0.0616 EXR - 0.000768 EXQ + 0.0932 MDF + 0.0726 PCU+ 0.00136 EXE + 0.157 ENV

Predictor Coef SE Coef T P Constant 7.6175 0.1254 60.73 0.000 TED 0.115595 0.007136 16.20 0.000 EXR 0.061617 0.005997 10.28 0.000 EXQ -0.0007679 0.0001527 -5.03 0.000 MDF 0.09320 0.04451 2.09 0.037 PCU 0.07256 0.05392 1.35 0.179 EXE 0.0013566 0.0004814 2.82 0.005 ENV 0.15691 0.03935 3.99 0.000

S = 0.523600 R-Sq = 39.1% R-Sq(adj) = 38.6%

Analysis of Variance

Source DF SS MS F P Regression 7 148.117 21.160 77.18 0.000 Residual Error 842 230.840 0.274 Total 849 378.957

260

Appendix: 22 E-views results of returns of investment to education and health in Pakistan

Model 1: Returns of Public Investment to Health in Pakistan

VAR Lag Order Selection Criteria Endogenous variables: IMRTR FRTR HXGDP LITR PPDR Exogenous variables: C Date: 09/20/17 Time: 17:42 Sample: 1972 2016 Included observations: 44

Lag LogL LR FPE AIC SC HQ

0 -631.9561 NA 2579750. 28.95255 29.15530 29.02774 1 -344.6242 496.3005* 17.24523* 17.02837* 18.24487* 17.47951*

* indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion

Date: 09/20/17 Time: 17:43 Sample (adjusted): 1974 2016 Included observations: 43 after adjustments Trend assumption: Linear deterministic trend Series: IMRTR FRTR HXGDP LITR PPDR Lags interval (in first differences): 1 to 1

Unrestricted Cointegration Rank Test (Trace)

Hypothesized Trace 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.534957 94.08932 69.81889 0.0002 At most 1 * 0.466857 61.16742 47.85613 0.0018 At most 2 * 0.307839 34.12192 29.79707 0.0149 At most 3 * 0.235860 18.30065 15.49471 0.0184 At most 4 * 0.144947 6.733466 3.841466 0.0095

Trace test indicates 5 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

Hypothesized Max-Eigen 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None 0.534957 32.92190 33.87687 0.0647 At most 1 0.466857 27.04551 27.58434 0.0585 At most 2 0.307839 15.82126 21.13162 0.2355 At most 3 0.235860 11.56719 14.26460 0.1280

261

At most 4 * 0.144947 6.733466 3.841466 0.0095

Max-eigenvalue test indicates no cointegration at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

Vector Error Correction Estimates Date: 09/20/17 Time: 17:43 Sample (adjusted): 1974 2016 Included observations: 43 after adjustments Standard errors in ( ) & t-statistics in [ ]

Cointegrating Eq: CointEq1

IMRTR( -1) 1.000000

FRTR(-1) -1.436939 (3.45561) [-0.41583]

HXGDP(-1) 11.83944 (4.00921) [ 2.95306]

LITR(-1) 1.544429 (0.43543) [ 3.54692]

PPDR(-1) -0.001096 (0.00037) [-2.98341]

C -158.3566

Error Correction: D(IMRTR) D(FRTR) D(HXGDP) D(LITR) D(PPDR)

CointEq1 -0.523091 0.011970 -0.002715 0.067265 31.14760 (0.17732) (0.00811) (0.00660) (0.04622) (9.96137) [-2.95004] [ 1.47553] [-0.41138] [ 1.45547] [ 3.12684]

D(IMRTR(-1)) 0.064302 0.000607 -0.001408 -0.042324 -12.91141 (0.19635) (0.00898) (0.00731) (0.05118) (11.0305) [ 0.32749] [ 0.06759] [-0.19270] [-0.82704] [-1.17052]

D(FRTR(-1)) -1.523772 -0.078785 0.222530 -1.500607 -367.7878 (4.11605) (0.18832) (0.15322) (1.07279) (231.233) [-0.37020] [-0.41836] [ 1.45238] [-1.39879] [-1.59055]

D(HXGDP(-1)) -7.085186 0.266739 0.141346 -0.391446 368.8237 (4.68068) (0.21415) (0.17424) (1.21996) (262.954) [-1.51371] [ 1.24556] [ 0.81123] [-0.32087] [ 1.40262]

D(LITR(-1)) 0.340639 -0.037706 0.006101 0.019532 -12.68342 (0.80183) (0.03669) (0.02985) (0.20899) (45.0459) [ 0.42482] [-1.02780] [ 0.20441] [ 0.09346] [-0.28157]

262

D(PPDR(-1)) 0.003275 -0.000229 1.73E-05 -0.000346 0.637066 (0.00182) (8.3E-05) (6.8E-05) (0.00048) (0.10243) [ 1.79614] [-2.74476] [ 0.25550] [-0.72749] [ 6.21956]

C -0.996697 -0.126985 0.020937 0.540923 -137.9101 (0.94977) (0.04345) (0.03535) (0.24754) (53.3565) [-1.04941] [-2.92230] [ 0.59221] [ 2.18516] [-2.58469]

R-squared 0.298332 0.302789 0.154064 0.170154 0.846772 Adj. R-squared 0.181387 0.186587 0.013075 0.031846 0.821234 Sum sq. resids 288.4379 0.603775 0.399679 19.59402 910316.4 S.E. equation 2.830577 0.129505 0.105367 0.737752 159.0175 F-statistic 2.551048 2.605714 1.092736 1.230258 33.15731 Log likelihood -101.9349 30.69934 39.56895 -44.11588 -275.1618 Akaike AIC 5.066738 -1.102295 -1.514835 2.377483 13.12381 Schwarz SC 5.353445 -0.815588 -1.228128 2.664190 13.41051 Mean dependent -1.632558 -0.088372 0.005349 0.851163 -274.0930 S.D. dependent 3.128496 0.143592 0.106063 0.749788 376.0991

Determinant resid covariance (dof adj.) 8.826344 Determinant resid covariance 3.630372 Log likelihood -332.7925 Akaike information criterion 17.33919 Schwarz criterion 18.97751

Pairwise Granger Causality Tests Date: 09/20/17 Time: 17:44 Sample: 1972 2016 Lags: 2

Null Hypothesis: Obs F-Statistic Prob.

FRTR does not Granger Cause IMRTR 43 0.85560 0.4331 IMRTR does not Granger Cause FRTR 6.53332 0.0036

HXGDP does not Granger Cause IMRTR 43 1.83141 0.1740 IMRTR does not Granger Cause HXGDP 2.15345 0.1300

LITR does not Granger Cause IMRTR 43 2.68827 0.0809 IMRTR does not Granger Cause LITR 0.99177 0.3803

PPDR does not Granger Cause IMRTR 43 0.63134 0.5374 IMRTR does not Granger Cause PPDR 0.01772 0.9824

HXGDP does not Granger Cause FRTR 43 6.13580 0.0049 FRTR does not Granger Cause HXGDP 2.71353 0.0791

LITR does not Granger Cause FRTR 43 6.17458 0.0048 FRTR does not Granger Cause LITR 7.52720 0.0018

PPDR does not Granger Cause FRTR 43 6.11964 0.0050 FRTR does not Granger Cause PPDR 0.42661 0.6558

LITR does not Granger Cause HXGDP 43 2.28434 0.1157 HXGDP does not Granger Cause LITR 5.55389 0.0077

263

PPDR does not Granger Cause HXGDP 43 0.70566 0.5001 HXGDP does not Granger Cause PPDR 0.99527 0.3791

PPDR does not Granger Cause LITR 43 8.63300 0.0008 LITR does not Granger Cause PPDR 0.46109 0.6341

Model 2: Returns of Public Investment to Education in Pakistan

VAR Lag Order Selection Criteria Endogenous variables: EEI EXPE EXSE EXTE INF LITR Exogenous variables: C Date: 09/20/17 Time: 17:45 Sample: 1972 2016 Included observations: 42

Lag LogL LR FPE AIC SC HQ

0 -557.5515 NA 18189.13 26.83579 27.08403 26.92678 1 -278.7515 464.6667 0.176480 15.27388 17.01155* 15.91081 2 -240.3255 53.06443 0.175001 15.15836 18.38546 16.34122 3 -179.4498 66.67338* 0.072096* 13.97380* 18.69033 15.70260*

Date: 09/20/17 Time: 17:46 Sample (adjusted): 1976 2016 Included observations: 41 after adjustments Trend assumption: Linear deterministic trend Series: EEI EXPE EXSE EXTE INF LITR Lags interval (in first differences): 1 to 3

Unrestricted Cointegration Rank Test (Trace)

Hypothesized Trace 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.933332 235.4998 95.75366 0.0000 At most 1 * 0.667241 124.4707 69.81889 0.0000 At most 2 * 0.639861 79.35691 47.85613 0.0000 At most 3 * 0.443227 37.48502 29.79707 0.0054 At most 4 0.268902 13.47549 15.49471 0.0985 At most 5 0.015343 0.633945 3.841466 0.4259

Trace test indicates 4 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

264

Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

Hypothesized Max-Eigen 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.933332 111.0291 40.07757 0.0000 At most 1 * 0.667241 45.11375 33.87687 0.0015 At most 2 * 0.639861 41.87190 27.58434 0.0004 At most 3 * 0.443227 24.00953 21.13162 0.0191 At most 4 0.268902 12.84154 14.26460 0.0829 At most 5 0.015343 0.633945 3.841466 0.4259

Max-eigenvalue test indicates 4 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

Vector Error Correction Estimates Date: 09/20/17 Time: 17:47 Sample (adjusted): 1975 2016 Included observations: 42 after adjustments Standard errors in ( ) & t-statistics in [ ]

Cointegrating Eq: CointEq1

EEI( -1) 1.000000

EXPE(-1) -0.001589 (0.00043) [-3.66317]

EXSE(-1) -0.001888 (0.00034) [-5.55195]

EXTE(-1) -0.003733 (0.00066) [-5.64912]

INF(-1) -0.000566 (5.9E-05) [-9.57661]

LITR(-1) -0.000874 (0.00030) [-2.96038]

C -0.063952

Error Correction: D(EEI) D(EXPE) D(EXSE) D(EXTE) D(INF) D(LITR)

CointEq1 -0.426883 20.26430 -142.6078 43.00373 -7.096033 17.67233 (0.12743) (73.2770) (43.8633) (49.4057) (45.7142) (11.4702) [-3.34989] [ 0.27654] [-3.25119] [ 0.87042] [-0.15523] [ 1.54071]

D(EEI(-1)) 0.017954 72.35243 -29.42810 -24.90229 -43.28108 6.450333 (0.16732) (96.2141) (57.5933) (64.8705) (60.0235) (15.0606)

265

[ 0.10730] [ 0.75199] [-0.51096] [-0.38388] [-0.72107] [ 0.42829]

D(EEI(-2)) -0.096590 -45.53467 47.47789 -29.81713 5.944087 -6.332569 (0.16724) (96.1652) (57.5641) (64.8376) (59.9931) (15.0530) [-0.57757] [-0.47350] [ 0.82478] [-0.45987] [ 0.09908] [-0.42069]

D(EXPE(-1)) -7.53E-05 -0.112184 -0.152563 -0.016874 0.004740 0.032300 (0.00036) (0.20751) (0.12422) (0.13991) (0.12946) (0.03248) [-0.20870] [-0.54062] [-1.22822] [-0.12060] [ 0.03661] [ 0.99439]

D(EXPE(-2)) -0.000147 -0.223478 -0.116596 0.041630 0.050412 0.045117 (0.00034) (0.19397) (0.11611) (0.13078) (0.12101) (0.03036) [-0.43672] [-1.15213] [-1.00420] [ 0.31832] [ 0.41660] [ 1.48595]

D(EXSE(-1)) -0.000667 -0.328005 -0.445578 -0.097485 -0.426941 -0.064158 (0.00047) (0.26813) (0.16050) (0.18078) (0.16727) (0.04197) [-1.42981] [-1.22330] [-2.77616] [-0.53924] [-2.55234] [-1.52863]

D(EXSE(-2)) -0.000294 -0.252814 -0.125338 -0.319170 -0.156553 -0.035123 (0.00057) (0.32716) (0.19583) (0.22058) (0.20410) (0.05121) [-0.51744] [-0.77276] [-0.64002] [-1.44697] [-0.76705] [-0.68587]

D(EXTE(-1)) -0.001407 -0.063120 -0.395345 -0.560334 0.116228 0.090088 (0.00059) (0.33878) (0.20279) (0.22842) (0.21135) (0.05303) [-2.38858] [-0.18632] [-1.94951] [-2.45314] [ 0.54993] [ 1.69882]

D(EXTE(-2)) -0.000583 -0.154191 -0.364566 -0.052955 0.071769 0.012898 (0.00052) (0.30103) (0.18019) (0.20296) (0.18780) (0.04712) [-1.11436] [-0.51222] [-2.02319] [-0.26091] [ 0.38216] [ 0.27373]

D(INF(-1)) -0.000230 -0.008952 -0.034248 -0.226734 0.565459 0.023188 (0.00055) (0.31378) (0.18782) (0.21156) (0.19575) (0.04912) [-0.42085] [-0.02853] [-0.18234] [-1.07174] [ 2.88867] [ 0.47212]

D(INF(-2)) -0.000164 0.009891 0.005730 0.296521 0.360693 -0.008607 (0.00053) (0.30711) (0.18383) (0.20706) (0.19159) (0.04807) [-0.30751] [ 0.03221] [ 0.03117] [ 1.43204] [ 1.88262] [-0.17905]

D(LITR(-1)) 0.005092 -1.539157 1.152899 -0.971252 0.001226 -0.005822 (0.00232) (1.33603) (0.79974) (0.90080) (0.83349) (0.20913) [ 2.19157] [-1.15203] [ 1.44158] [-1.07821] [ 0.00147] [-0.02784]

D(LITR(-2)) 0.007405 0.926774 1.743565 0.107829 -0.838527 -0.066763 (0.00248) (1.42541) (0.85325) (0.96106) (0.88925) (0.22312) [ 2.98736] [ 0.65018] [ 2.04345] [ 0.11220] [-0.94296] [-0.29922]

C -0.006626 0.651522 -1.994281 0.740638 1.467553 0.872982 (0.00327) (1.88296) (1.12713) (1.26955) (1.17469) (0.29474) [-2.02338] [ 0.34601] [-1.76934] [ 0.58339] [ 1.24931] [ 2.96184]

R-squared 0.363626 0.239106 0.476604 0.391421 0.803375 0.428665 Adj. R-squared 0.068166 -0.114167 0.233599 0.108867 0.712084 0.163402 Sum sq. resids 0.001656 547.6229 196.2223 248.9428 213.1314 13.41803 S.E. equation 0.007691 4.422438 2.647251 2.981747 2.758956 0.692254 F-statistic 1.230712 0.676831 1.961291 1.385295 8.800207 1.616001 Log likelihood 153.3640 -113.5217 -91.96857 -96.96604 -93.70444 -35.63294 Akaike AIC -6.636382 6.072461 5.046122 5.284097 5.128783 2.363473

266

Schwarz SC -6.057159 6.651684 5.625346 5.863320 5.708006 2.942697 Mean dependent 0.002062 0.015037 0.218217 -0.012053 4.627214 0.859524 S.D. dependent 0.007967 4.189736 3.023898 3.158635 5.141761 0.756845

Determinant resid covariance (dof adj.) 0.022436 Determinant resid covariance 0.001970 Log likelihood -226.7451 Akaike information criterion 15.08310 Schwarz criterion 18.80668

Pairwise Granger Causality Tests Date: 09/20/17 Time: 17:51 Sample: 1972 2016 Lags: 2

Null Hypothesis: Obs F-Statistic Prob.

EXPE does not Granger Cause EEI 43 0.00282 0.9972 EEI does not Granger Cause EXPE 1.08071 0.3496

EXSE does not Granger Cause EEI 43 0.81919 0.4484 EEI does not Granger Cause EXSE 0.21701 0.8059

EXTE does not Granger Cause EEI 43 0.63755 0.5341 EEI does not Granger Cause EXTE 0.48959 0.6167

INF does not Granger Cause EEI 43 1.08310 0.3488 EEI does not Granger Cause INF 1.05897 0.3568

LITR does not Granger Cause EEI 43 4.52624 0.0173 EEI does not Granger Cause LITR 1.04348 0.3621

EXSE does not Granger Cause EXPE 43 1.23340 0.3027 EXPE does not Granger Cause EXSE 2.67754 0.0817

EXTE does not Granger Cause EXPE 43 2.16707 0.1285 EXPE does not Granger Cause EXTE 0.73972 0.4840

INF does not Granger Cause EXPE 43 0.44910 0.6415 EXPE does not Granger Cause INF 0.04591 0.9552

LITR does not Granger Cause EXPE 43 0.16701 0.8468 EXPE does not Granger Cause LITR 5.74163 0.0066

EXTE does not Granger Cause EXSE 43 0.75145 0.4786 EXSE does not Granger Cause EXTE 2.19745 0.1250

INF does not Granger Cause EXSE 43 1.49358 0.2375 EXSE does not Granger Cause INF 7.12282 0.0024

LITR does not Granger Cause EXSE 43 0.45068 0.6406 EXSE does not Granger Cause LITR 2.66601 0.0825

INF does not Granger Cause EXTE 43 0.40601 0.6692 EXTE does not Granger Cause INF 3.07977 0.0576

267

LITR does not Granger Cause EXTE 43 1.86423 0.1689 EXTE does not Granger Cause LITR 7.21641 0.0022

LITR does not Granger Cause INF 43 1.90120 0.1633 INF does not Granger Cause LITR 1.47437 0.2417

Model 3: Health, Education and Economic Growth

VAR Lag Order Selection Criteria Endogenous variables: GDPG GFCF LITR IMRTR TLF UNEMP Exogenous variables: C Date: 09/20/17 Time: 17:58 Sample: 1972 2016 Included observations: 44

Lag LogL LR FPE AIC SC HQ

0 -1165.627 NA 5.42e+15 53.25579 53.49909 53.34601 1 -924.9720 404.7385* 5.02e+11* 43.95327* 45.65636* 44.58486*

Date: 09/20/17 Time: 17:59 Sample (adjusted): 1974 2016 Included observations: 43 after adjustments Trend assumption: Linear deterministic trend Series: GDPG GFCF LITR IMRTR TLF UNEMP Lags interval (in first differences): 1 to 1

Unrestricted Cointegration Rank Test (Trace)

Hypothesized Trace 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.626202 115.7533 95.75366 0.0011 At most 1 * 0.436131 73.43957 69.81889 0.0250 At most 2 * 0.398701 48.80343 47.85613 0.0406 At most 3 0.324801 26.93091 29.79707 0.1033 At most 4 0.192038 10.04279 15.49471 0.2774 At most 5 0.020108 0.873475 3.841466 0.3500

Trace test indicates 3 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

268

Hypothesized Max-Eigen 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.626202 42.31368 40.07757 0.0275 At most 1 0.436131 24.63614 33.87687 0.4100 At most 2 0.398701 21.87252 27.58434 0.2270 At most 3 0.324801 16.88813 21.13162 0.1774 At most 4 0.192038 9.169313 14.26460 0.2723 At most 5 0.020108 0.873475 3.841466 0.3500

Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

Vector Error Correction Estimates Date: 09/20/17 Time: 18:00 Sample (adjusted): 1974 2016 Included observations: 43 after adjustments Standard errors in ( ) & t-statistics in [ ]

Cointegrating Eq: CointEq1

GDPG( -1) 1.000000

GFCF(-1) -3.27E-06 (8.5E-07) [-3.86225]

LITR(-1) -0.814560 (0.12147) [-6.70601]

IMRTR(-1) 0.236687 (0.06089) [ 3.88714]

TLF(-1) -0.065449 (0.06099) [-1.07304]

UNEMP(-1) 1.417723 (0.29435) [4.81639]

C -50.56203

Error Correction: D(GDPG) D(GFCF) D(LITR) D(IMRTR) D(TLF) D(UNEMP)

CointEq1 -0.575574 74715.54 -0.029685 -0.492226 0.180156 -0.089134 (0.22686) (16028.7) (0.08850) (0.34231) (0.18786) (0.09940) [-2.53717] [ 4.66135] [-0.33540] [-1.43795] [ 0.95899] [-0.89668]

D(GDPG(-1)) -0.189537 -35458.23 0.012243 -0.055533 -0.144757 -0.008973 (0.15567) (10999.2) (0.06073) (0.23490) (0.12891) (0.06821) [-1.21754] [-3.22372] [ 0.20158] [-0.23641] [-1.12292] [-0.13154]

269

D(GFCF(-1)) -8.68E-07 0.182855 7.07E-07 -2.12E-07 -1.47E-06 -3.73E-07 (1.9E-06) (0.13148) (7.3E-07) (2.8E-06) (1.5E-06) (8.2E-07) [-0.46664] [ 1.39075] [ 0.97342] [-0.07534] [-0.95375] [-0.45691]

D(LITR(-1)) 0.428182 19818.85 0.281244 0.213016 -0.595351 -0.071664 (0.42341) (29916.4) (0.16519) (0.63890) (0.35063) (0.18553) [ 1.01127] [ 0.66247] [ 1.70257] [ 0.33341] [-1.69797] [-0.38626]

D(IMRTR(-1)) 0.045426 -18261.00 -0.002633 0.073719 0.001947 0.049961 (0.10757) (7600.33) (0.04197) (0.16231) (0.08908) (0.04713) [ 0.42230] [-2.40266] [-0.06273] [ 0.45418] [ 0.02186] [ 1.05996]

D(TLF(-1)) -0.011596 11771.36 0.045003 -0.681030 -0.293948 0.006098 (0.23471) (16583.6) (0.09157) (0.35416) (0.19436) (0.10285) [-0.04941] [ 0.70982] [ 0.49146] [-1.92293] [-1.51237] [ 0.05929]

D(UNEMP(-1)) -0.205212 90068.82 0.295478 -2.304222 0.151582 -0.407690 (0.49619) (35058.9) (0.19358) (0.74872) (0.41090) (0.21742) [-0.41357] [ 2.56907] [ 1.52637] [-3.07754] [ 0.36891] [-1.87510]

C -0.278092 -31619.19 0.499712 -0.801015 1.993507 0.290906 (0.55479) (39198.9) (0.21644) (0.83714) (0.45942) (0.24310) [-0.50126] [-0.80663] [ 2.30876] [-0.95685] [ 4.33921] [ 1.19666]

R-squared 0.438987 0.475254 0.187917 0.302220 0.162906 0.120739 Adj. R-squared 0.326784 0.370304 0.025500 0.162664 -0.004513 -0.055113 Sum sq. resids 125.9794 6.29E+11 19.17462 286.8393 86.38921 24.18853 S.E. equation 1.897211 134048.7 0.740166 2.862763 1.571071 0.831325 F-statistic 3.912446 4.528415 1.157005 2.165586 0.973042 0.686595 Log likelihood -84.12509 -564.2447 -43.65068 -101.8154 -76.01410 -48.64495 Akaike AIC 4.284888 26.61603 2.402357 5.107692 3.907633 2.634649 Schwarz SC 4.612553 26.94370 2.730022 5.435357 4.235298 2.962314 Mean dependent -0.054285 39064.93 0.851163 -1.632558 1.125349 0.099302 S.D. dependent 2.312270 168926.3 0.749788 3.128496 1.567537 0.809322

Determinant resid covariance (dof adj.) 2.58E+11 Determinant resid covariance 7.50E+10 Log likelihood -904.4759 Akaike information criterion 44.58027 Schwarz criterion 46.79201

Pairwise Granger Causality Tests Date: 09/20/17 Time: 18:00 Sample: 1972 2016 Lags: 2

Null Hypothesis: Obs F-Statistic Prob.

GFCF does not Granger Cause GDPG 43 1.68051 0.1998 GDPG does not Granger Cause GFCF 0.94454 0.3978

LITR does not Granger Cause GDPG 43 1.27664 0.2907 GDPG does not Granger Cause LITR 0.11654 0.8903

IMRTR does not Granger Cause GDPG 43 1.26361 0.2942

270

GDPG does not Granger Cause IMRTR 1.09118 0.3461

TLF does not Granger Cause GDPG 43 0.74764 0.4803 GDPG does not Granger Cause TLF 0.09176 0.9125

UNEMP does not Granger Cause GDPG 43 0.40176 0.6719 GDPG does not Granger Cause UNEMP 4.83381 0.0135

LITR does not Granger Cause GFCF 43 3.92760 0.0282 GFCF does not Granger Cause LITR 2.42066 0.1024

IMRTR does not Granger Cause GFCF 43 3.75027 0.0326 GFCF does not Granger Cause IMRTR 0.04782 0.9534

TLF does not Granger Cause GFCF 43 3.82584 0.0306 GFCF does not Granger Cause TLF 1.14103 0.3302

UNEMP does not Granger Cause GFCF 43 2.58251 0.0888 GFCF does not Granger Cause UNEMP 0.12436 0.8834

IMRTR does not Granger Cause LITR 43 0.99177 0.3803 LITR does not Granger Cause IMRTR 2.68827 0.0809

TLF does not Granger Cause LITR 43 0.72610 0.4904 LITR does not Granger Cause TLF 1.84987 0.1711

UNEMP does not Granger Cause LITR 43 3.31480 0.0471 LITR does not Granger Cause UNEMP 1.08626 0.3477

TLF does not Granger Cause IMRTR 43 1.90811 0.1623 IMRTR does not Granger Cause TLF 0.16003 0.8527

UNEMP does not Granger Cause IMRTR 43 2.99243 0.0621 IMRTR does not Granger Cause UNEMP 1.42979 0.2519

UNEMP does not Granger Cause TLF 43 0.19529 0.8234 TLF does not Granger Cause UNEMP 0.32649 0.7235

271

Appendix: 23 E-views results of diagnostic tests of returns of investment to education and health in Pakistan by using secondary data.

Model 1: Returns of Public Investment to Health in Pakistan

Inverse Roots of AR Characteristic Polynomial

1.5

1.0

0.5

0.0

-0.5

-1.0

-1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

VEC Residual Serial Correlation LM Tests Null Hypothesis: no serial correlation at lag order h Date: 06/22/19 Time: 22:44 Sample: 1972 2016 Included observations: 43

Lags LM-Stat Prob

1 28.64668 0.2790 2 32.85618 0.1347

Probs from chi-square with 25 df.

VEC Residual Heteroskedasticity Tests: No Cross Terms (only levels and squares) Date: 06/22/19 Time: 22:45 Sample: 1972 2016 Included observations: 43

Joint test:

Chi-sq Df Prob.

193.3196 180 0.2357

272

Model 2: Returns of Public Investment to Education in Pakistan

Inverse Roots of AR Characteristic Polynomial

1.5

1.0

0.5

0.0

-0.5

-1.0

-1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

VEC Residual Serial Correlation LM Tests Null Hypothesis: no serial correlation at lag order h Date: 06/22/19 Time: 22:48 Sample: 1972 2016 Included observations: 42

Lags LM-Stat Prob

1 27.00794 0.8606 2 19.50795 0.9886

Probs from chi-square with 36 df.

VEC Residual Heteroskedasticity Tests: No Cross Terms (only levels and squares) Date: 06/22/19 Time: 22:49 Sample: 1972 2016 Included observations: 42

Joint test:

Chi-sq df Prob.

518.6755 546 0.7941

273

Model 3: Health, Education and Economic Growth

Inverse Roots of AR Characteristic Polynomial

1.5

1.0

0.5

0.0

-0.5

-1.0

-1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

VEC Residual Serial Correlation LM Tests Null Hypothesis: no serial correlation at lag order h Date: 06/22/19 Time: 22:56 Sample: 1972 2016 Included observations: 43

Lags LM-Stat Prob

1 33.73609 0.5767 2 35.52024 0.4912

Probs from chi-square with 36 df.

VEC Residual Heteroskedasticity Tests: No Cross Terms (only levels and squares) Date: 06/22/19 Time: 22:57 Sample: 1972 2016 Included observations: 43

Joint test:

Chi-sq df Prob.

299.4221 294 0.4014

274