Determinants of Foreign Portfolio Investment (FPI): Empirical Evidence from Pakistan

ASIAN JOURNAL OF EDUCATIONAL RESEARCH & TECHNOLOGY Vol. 5 (2), July 2015: 161-169 ISSN (Print): 2249-7374 Website: http://www.tspmt.com ISSN (Online): 2347-4947 RESEARCH ARTICLE Determinants of Foreign Portfolio Investment (FPI): Empirical Evidence from Pakistan Bilal Aziz 1, Zeeshan Anwar 2 and Sundas Shawnawaz 2 1 Professional Development Center (PDC), IBM, University of Engineering and Technology (UET) Lahore, Pakistan 2 University of Lahore, Pakistan E-mail: biiilal@live.com Received: 22 nd April 2015, Revised: 24 th May 2015, Accepted: 28 th June 2015 ABSTRACT This study intends to examine the impact of determinants of foreign portfolio investment (FPI) in Pakistan during the period of 2005-2014. The study has used the time series data and Ordinary Least Squares (OLS) regression model has been applied to estimate the relationship of dependent and independent variables. The independent variables which have been used to explain the FPI inflows include foreign exchange rate, trade degree of openness, inflation rate, growth rate of real GDP and market capitalization. The results of the OLS regression model have shown that trade degree of openness, growth rate of real GDP and market capitalization have positive and significant relationship with foreign portfolio investment (FPI) in Pakistan, whereas, inflation rate have negative and significant relationship with foreign portfolio investment (FPI) in Pakistan. The variable of foreign exchange rate has positive but insignificant relationship with foreign portfolio investment (FPI) in Pakistan. Therefore, it is being suggested that the government of Pakistan should provide the risk free environment to the foreign investor and local investors to increase foreign investment. Key words: Exchange rate, trade degree of openness, Inflation rate INTRODUCTION Foreign portfolio investment (FPI) is an investment activity that involves the purchase of bonds, stocks or money market instruments for shorter time period in foreign country. Because of its short term nature, it provides opportunity to investors to take advantage of favorable interest rate and exchange rate for buying and selling the security. It is an investment in the foreign secondary market and its purpose is to obtain higher return. This task can be completed through passive holding of shares or through active trading of securities in the financial or capital markets. It improves liquidity position of host economy and also helps to increase the foreign reserves that result in stability of exchange rate. Firstly, the foreign portfolio investment (FPI) is affective for obtaining higher return and decreases risk through international diversification. Secondly, it plays a significant role in the economic growth of the host country. Thirdly, it encourages investment of new funds in the country due to which investment level would increase. Foreign portfolio investment (FPI) is one of the crucial capital flows and in current economic conditions, it is extremely important as the market risk is very high due to instable political conditions, and the foreign investors are avoiding investment in Pakistan, therefore, foreign investors can be attracted to invest in Pakistan through FPI. Although, it is a shorter time period investment but it may be helpful to attract foreign investors in Pakistan. But unfortunately, the present conditions of Pakistan s economy and stock markets are unstable. Economists suggest that stock exchange is the indicator of an economy, and the stock market of Pakistan is in deep crisis. Table 1 depicts that FPI has increased to $1,820 million in year 2006-07 from previous value of ($10) in year 2001-2002. Then it started decreasing and reached at ($46.9) in year 2011-12 and then again started increasing and reached at $622.8 in year 2013-14. Because of the importance of foreign portfolio investment (FPI) in boosting economic growth of Pakistan, the objective of this study is to examine the impact of exchange rate,

interest rate, inflation rate, market capitalization, money supply and real growth rate of GDP on Foreign portfolio Investment (FPI) for the period of 2005 to 2014 through applying Ordinary Least Squared (OLS) regression model. This study is very important because before there are very few studies which have investigated the determinants of FPI in Pakistan. The rest of the paper has been organized as follows: the literature review has been discussed in section two, the research methodology has been discussed in section 3, the theoretical model has been presented in section 4, results have been presented in section 5 and conclusion has been discussed at the last section. Table 1: Foreign Portfolio Investment inflows in Pakistan ($Million) Year Portfolio Investment 2001-02 -10.00 2002-03 22.00 2003-04 -28.00 2004-05 153.00 2005-06 351.00 2006-07 1,820.00 2007-08 19.30 2008-09 -510.30 2009-10 587.90 2010-11 344.5 2011-12 (46.9) 2012-13 119.5 2013-14 622.8 Source: State Bank of Pakistan LITERATURE REVIEW Researchers have investigated varies determinants of foreign portfolio investment (FPI). According to Grabel (1998) there had been dramatic increase in level of global foreign Portfolio investment (FPI) in the world. He argued that foreign portfolio investment (FPI) was considered a comparatively safe and efficient mean of moving capital to those countries where the demand of capital is greater, on the other hand this opinion has been challenged by the sequence of financial crises of Mexico in 1994 and the Southeast Asia in 1997 to 1998. Therefore, several economists debated that these crises were strange and reflect outstanding circumstances. However, a closer look depicts that the foreign portfolio investment (FPI) plays an important role in economic growth of a country. Agarwal et. al (2003) examined both Asian countries and firm-level authority and representation policies which effected investment distribution decisions. Based on the country-level survey, the authors argued that the U.S. investor invested more in open market which had stronger investor rights, better accounting standards and had improved legal frameworks. After controlling for the country characteristics, the U.S. funds were discovered to be invested more in firms which adopted policies which resulted in better development of economic growth. Onuorah et. al (2003) found that inflows of foreign direct investment (FDI) and foreign portfolio investment (FPI) had a positive effect on the economic performance of developing countries as many of the developing countries have established financial markets with the help of Foreign Direct Investment (FDI) and foreign portfolio investment (FPI). The authors observed the influence of international capital flows on the growth of Indian economy and financial markets. By utilizing the monthly time series data, the authors found that the Foreign Direct Investment (FDI) is positively affecting economic growth, whereas, the Foreign Portfolio Investment (FPI) is negatively affecting economic growth. ~ 162 ~

Agarwal (2006) investigated the determinants of foreign portfolio investment (FPI) in Asian developing countries.the regression analysis identified that the real exchange rate, inflation rate, split of host capital market and indicators of economic movement in the global stock market capitalization were statistically important determinants of foreign portfolio investment (FPI). The results showed that the variable of inflation rate had negative relationship, whereas the other three variables had positive relationship. The variable of foreign trade, foreign direct investment (FDI) and current account had statistically insignificant relationship. Rai and Bhanumurthy (2007) observed the factors which affect Foreign Portfolio Investment (FPI) in India, which crossed approximately 12 billion US$ at the end of year 2002. These investment flows had significant influence on performance of the other household markets. The authors found that FPI inflows were influenced by the return of stock market, ex-ante risk and inflation rate (both foreign and domestic). The impact of the ex-ante risk and stock market returns had turned out to be most important factors of FPI inflows. The author suggested that the stock market stability and reducing the ex-ante risk would attract further Foreign Portfolio Investments (FPI) inflows which might have positive impact on the national economy. Bordo and Miessner (2007) discovered the relationship between the international capital market and economic growth. They used standard growth regressions analysis and found the varied evidence of any relationship between the foreign capitals inflows and economic growth. If there was an influence, it came with a long gap and it is temporary which have no impact on growth rate. They takes a balance sheet viewpoint on Crises and discovered determinants of debt crises and money crises which included the money conformation of debt, the debt prejudice and the role of governmental associations The Countries which have reliable commitments and comprehensive fiscal and financial policies were able to avoid major financial crises and achieve greater per capita income. Ekeocha and Chukwuemeka (2008) examined that comparatively low yield in developing countries together with higher economic growth rate and higher rate of return encourage foreign investors to shift their resources and funds in developing countries. The authors suggested that increase in foreign portfolio investment causes increase in trade development, international economic linkages and increase of production resources in host country. Ekeocha and Chukwuemeka (2008) found that improvement of home capital markets and reductions in capital flow restrictions attracted foreign investors to make foreign portfolio investment (FPI) in host countries which resulted in economics growth. The authors also evaluated the costs and benefits of foreign portfolio investment (FPI) for the recipient countries and presented the empirical evidence about the relationship between market development and foreign portfolio investment (FPI), level of capital market and market volatility. Duasa and Salina (2009) determined the relationship of foreign portfolio investment (FPI) and economic growth in Malaysia during the period of 1991-2006 through applying ganger causality Test, Toda and Yamamoto's (1995) non-causality analysis. They used gross domestic product as dependent variable and foreign portfolio investment (FPI) was used as independent variable. The authors suggested that foreign portfolio investment (FPI) positively influenced economic growth in host country. Hasnan and Masood (2002) investigated the relationship of foreign capital inflows and domestic savings in Pakistan with sector analysis by using the time series data through co-integration technique for the period of 1985-2000. They found that domestic savings rate had insignificant relationship with foreign capital inflows in Pakistan. Zafar and Sattar (2005) determined the relationship of remittances and economic growth in Pakistan with sector analysis by using the time series data through multiple regression analysis. The variables which were analyzed include GDP at current price, public investment (PG), inflation rate (INF), worker remittances (WR), private investment (PI), per capita income, squared per capita income and changes in terms of trade. They found that there was positive relationship among workers remittances and economic growth. Asma and salma ~ 163 ~

(2006) determined the relationship of foreign investment in which the authors included foreign portfolio investment (FPI), foreign direct investment (FDI) and economic growth in Pakistan by using the time series secondary data through ganger causality technique for the period of 1980-2004. The author found that the foreign portfolio investment (FPI) and foreign direct investment (FDI) had positive and significant relationship with economics growth in Pakistan. Azam and Naeem-ur-Rehman (2010) discovered the relationship of social and political factors and foreign direct investment (FDI) in Pakistan by using the secondary data for the period of 1971-2005 by utilizing the least square regression technique. The results showed that the human capital had positive and significant relationship, whereas political instability had negative and insignificant relationship with foreign direct investment (FDI) in Pakistan. Farman Afzal at. al (2011) empirically investigated that the impact of foreign direct investment (FDI) on economics growth in Pakistan by using the secondary data for the period of the 1990-2008 by utilizing the ordinary least square technique (OLS). The variables which were included in this model include, GNP growth rate, BOP, FDI, trade of goods and services, GDP growth rate, employment, inflation, exchange rate, openness factor, GNP deflator and GDP deflator. They found that inflation, GDP growth rate and growth of economy had positive relationship. Arshad and Sujaat (2011) empirically investigated the impact of foreign direct investment (FDI) on economic growth in Pakistan by using time series data for the period of the 1981-2008, by utilizing granger causality test and panel co-integration test. The variable which were included in the model were foreign direct investment, economic growth, and residual time effect. They observed long run relationship between these variables and having a direct relationship with each other. The authors recommended that foreign direct investment (FDI) is a primary factor for economic growth in developing country. Ali Raza, et al. (2011) empirically investigated the influence of international capital flows on Pakistan s economic growth for the period of 1985 to 2010. The results explained that FPI, FDI and remittances positively and significantly affected economic growth, whereas, foreign support significantly but negatively correlated with economic growth. They further suggested that FPI, FDI and remittances had major influence on the economic growth. They also observed that Pakistan should improve the domestic resources to improve foreign aid. Talat and Zeshan (2013) investigated the association of political instability, cost of war against terrorism, electricity generation, inflation rate, market size, trade openness, incentives provided to investors and exchange rate stability with FDI inflows in Pakistan for the period of 1980 to 2010 by using ARMA regression model. The results have shown that political instability and war against terrorism had negative relationship while electricity generation had positive influence on FDI inflows in Pakistan. The results have also shown that exchange rate stability, market size, incentives provided to investors and trade openness had positive relationship with FDI while the inflation rate had a negative association with FDI inflows. Zeshan and Talat (2014) explored the relationship of governance indicators including political stability and absence of violence, voice and accountability, regulatory quality, government effectiveness, governance index and control of corruption with FDI inflows in Pakistan from 1996 to 2010 by utilizing OLS regression model. The results have depicted that political stability, voice and accountability, government effectiveness, absence of violence/ terror, control of corruption, regulatory quality and governance index positively and significantly affect inflows of FDI in Pakistan. It can be seen from the above mentioned literature review that much research has been conducted on determinants of FDI but there are very few studies which explored the determinants of FPI in Pakistan. Therefore, this study is intended to fill this research gap by identifying the important determinants of FPI in Pakistan for the period of 2005-2014 ~ 164 ~

RESEARCH METHODOLOGY In this study, time series secondary data has been uses for the period of year 2005-2014. The data for the dependent and independent variables have been gathered from World Bank, State Bank of Pakistan (SBP), World debt tables and world table, IMF and National Bureau Statistics. The Ordinary Least Squares (OLS) regression technique has been employed to determine the relationship of dependent variable i.e. foreign portfolio investment (FPI) in Pakistan and independent variables i.e. Real exchange rate, Trade degree of openness, Inflation rate, and Growth rate of real GDP and Market capitalization. The following regression model has been used in this study: LnFPI= + ER + TDO + INF + RGDP + Mcap + Where: FPI= foreign portfolio investment RER= Real exchange rate in percentage TDO= Trade degree of openness measured as average of imports and exports INF= Inflation rate in percentage GDP= Growth rate of real GDP in percentage Mcap= Market capitalization measured number of outstanding shares multiplied by current shares market price SPSS 16.0 has been used to measure the relationship of dependent and independent variables. Theoretical Model: Independent Variable Dependent variable Exchange Rate Trade degree of openness Inflation Growth rate of real GDP Foreign Portfolio Investment (FPI) Market capitalization ANALYSIS AND DISCUSSION OF RESULTS Table 2: Descriptive Statistics FPI ER TDO INF GDP Mcap N Valid 10 10 10 10 10 10 Missing 0 0 0 0 0 0 Mean 9.1160-1.1330.7004 9.6200 4.5678 23.9440 Std. Error of Mean 4.55069 1.93484.04823 1.67536.67247.25089 Median 19.7900-3.3750.6705 7.8000 4.2394 24.1300 Mode 7.60-5.63 a.48 a 7.70 1.60 a 22.32 a Std. Deviation 1.43905E1 6.11849.15252 5.29797 2.12653.79338 a. Multiple modes exist. The smallest value is shown ~ 165 ~

The results of descriptive statistics have been shown in table 2. The results of Descriptive Statistics show that the mean value for foreign portfolio investment (FPI) is 9.11 million; it shows that the average foreign portfolio investment (FPI) is 9.11 million for the period of 2005-2014. Median value is 19.79 million, it shows that the middle value of FPI is 19.79 million; mode value is 7.60 million, which means that the foreign portfolio investment (FPI) is 7.60 million repeated of more than one time. The value of standard deviation for FPI is 1.43 million, it means that the foreign portfolio investment (FPI) can vary from its mean within a range of 1.43 million, the value of standard error of mean is 4.55 million, it shows that if the mean value of FPI will be calculated repeatedly, it may have an error of 4.55 million. The mean value of exchange rate is -1.13 percent; it means the average exchange rate is -1.13 percent for the period of 2005-2014. Median value is -3.37 percent; it means that the middle value of exchange rate is -3.37 percent. Mode value is -5.63 percent; it means that the exchange rate is -5.63 percent repeated of more than one time. Standard deviation is 6.11 percent, it means that the exchange rate can vary from its meant within a range of 6.11 percent, the value of standard error of mean is 1.93 percent, it shows that if the mean value of exchange rate will be calculated repeatedly, it may have an error of 1.93 percent. The mean value of Trade degree of openness is 0.70 percent, it means the average trade degree of openness is 0.70 percent for the period of 2005-2014. Median value is 0.67 percent; it shows that the middle value of trade degree of openness is 0.67 percent, Standard deviation value is 0.15 percent, it means that the trade degree of openness can vary from its meant within a range of 0.15 percent. Mode value is 0.48 percent; it means that the trade degree of openness is 0.48 percent repeated of more than one tome. Standard error of mean is 0.04 percent; it shows that if the mean value of trade degree of openness will be calculated repeatedly, it may have an error of 0.04 percent. The mean value of inflation rate is 9.62 percent; it means that the average inflation rate is 9.62 percent for the period of 2005-2014. Median value is 7.80 percent; it shows that the middle value of inflation rate is 7.80 percent. Mode value is 7.70 percent; it means that the inflation rate is 7.70 percent repeated of more than one time. Standard deviation value is 5.29 percent; it means that the inflation rate can vary from its meant within a range of 5.29 percent, and the standard error of mean is1.67 percent, it shows that if the mean value of inflation rate will be calculated repeatedly, it may have an error of 1.67 percent. The mean value of GDP is 4.56 percent; it means that the average GDP is 4.56 percent for the period of 2005-2014. Median value is 4.23 percent. It shows that the middle value of GDP is 4.23 percent, mode value is 1.60 percent, it means that the GDP 1.60 percent repeated of more than one time, Standard deviation value is 2.12 percent, it means that the GDP can vary from its meant within a range of 2.12 percent, and the standard error of mean is 0.67 percent, it shows that if the mean value of GDP will be calculated repeatedly, it may have an error of 0.67 percent. The mean value of market capitalization is $23.94; it means that the average market capitalization is $23.94 for the period of 2005-2014. Median value is $24.13; it shows that the middle value of market capitalization is $24.13. Mode value is $22.32; it means that the market capitalization is $22.32 repeated of more than one time, Standard deviation value is $.79; it means that the market capitalization can vary from its meant within a range of $.79, and the value of standard error of mean is $0.25, it shows that if the mean value of market capitalization will be calculated repeatedly, it may have an error of $0.25. After measuring the descriptive statistics, OLS regression technique has been employed to determine the relationship of dependent and independent variables. The results of the regression model have been presented in table 3, 4 and 5. The results of table 3 depicted that the value of adjusted R Square is 0.94, which means that the independent variables which have been used in the model have explained about 94 percent of the variations occurring in foreign portfolio investment. The value of Durbin Watson is 2.21 which is within the acceptable range of 1.5-2.5, it means that there is no problem of multi co-linearity among the independent variables. ~ 166 ~

Table 3: Model summery Model R R Square Adjusted R Square Std. Error of the Estimate Durbin-Watson 1.987 a.973.940 3.52031 2.219 a. Predictors: (Constant), Mcap, INF, ER, GDP, TDO b. Dependent Variable: FPI Table 4: ANOVA smodel Sum of Squares Df Mean Square F Sig. 1 Regression 1814.216 5 362.843 29.279.003 a Residual 49.570 4 12.393 Total 1863.786 9 The results in table 4 describes that the overall model is significant because the p-value is 0.003 which is less than 0.05. It means that the overall model used in this study is significant and the independent variables which have been included in the model can be used to explain the dependent variable. Table 5: Coefficients Model Unstandardized Coefficients Standardized Coefficients B Std. Error Beta 1 (Constant) 26.900 10.814 1.372.242 ER.510.342.217 1.492.210 TDO 2.238.620.872 3.988.016 INF -.966.380 -.356-2.546.034 GDP 7.584 1.326 1.121 5.719.005 Mcap 7.651 6.075.422 1.259.026 a. Dependent Variable: FPI The table 5 shows the level of significance of individual independent variables. The regression results depicts that the variable of trade degree of openness (TDO) has positive and significant relationship with inflows of foreign portfolio investment in Pakistan. Its value of coefficient is 2.238, which means that if trade degree of openness (TDO) would be increase by 1 percent, it will cause an increase of 2.238 percent in inflows of foreign portfolio investment in Pakistan. The regression results show that the variable of inflation has negative and significant relationship with inflows of foreign portfolio investment in Pakistan. Its value of coefficient is -0.966 percent, which means that if inflation rate would be decreased by 1 percent, it will cause an increase of 0.966 percent in inflows of foreign portfolio investment in Pakistan. The regression results show that the variable of GDP has positive and significant relationship with inflows of foreign portfolio investment in Pakistan. Its value of coefficient is 7.584 percent, which means that if inflation rate would be decreased by 1 percent, it will cause an increase of 7.584 percent in inflows of foreign portfolio investment in Pakistan. The results further show that the variable of market capitalization has positive and significant relationship with inflows of foreign portfolio investment in Pakistan. Its value of coefficient is $7.651, which means that if market capitalization would be decreased by 1 percent, it will cause an increase of $7.651 percent in inflows of foreign portfolio investment in Pakistan. The regression results show that the variable of exchange rate has positive but insignificant t Sig. ~ 167 ~

relationship with inflows of foreign portfolio investment in Pakistan; it means that it does not significantly effect inflows of foreign portfolio investment in Pakistan. CONCLUSION AND RECOMMENDATION Foreign Portfolio Investment (FPI) plays a crucial role in economic growth of developing countries and it also has a crucial role in the economic development of Pakistan. A developing country like Pakistan is extremely dependent on the foreign capital inflows as it plays a vital role is economic growth of Pakistan. Therefore, this study investigated the variables which influence inflows of FPI in Pakistan. The relationship of Exchange rate, trade degree of openness, Inflation rate, Growth rate of real GDP and market capitalization with foreign portfolio investment (FPI) in Pakistan has been determined for the period of 2005-2014 with the help of OLS regression technique. The results of the regression model depicts that the trade degree of openness, growth rate of real GDP and market capitalization have positive and significant relationship with foreign portfolio investment (FPI) in Pakistan whereas, inflation rate have shown the negative and significant relationship with foreign portfolio investment (FPI) in Pakistan. The variable of exchange rate has shown positive but insignificant relationship with foreign portfolio investment (FPI) in Pakistan. It is recommended that the Government of Pakistan should reduce the inflation, encourage trade degree of openness which will cause an increase in economic growth. It is also recommended that the Government of Pakistan should provide the risk free environment to the foreign investor and local investors for getting more investment in order to attract the foreign investors which will cause an increase in FPI. LIMITATIONS Following are the limitations of the study: 1. Some other important variables like gross capital flow, institutional quality can also be included in the model. 2. The data for the extended time period can be included in the model. 3. The burning issues of Pakistan like terrorism, political instability energy crisis can also be included in the model. REFERENCES 1. Agarwal. (2006). Foreign portfolio investment in some development countries: A study of determenents and macroeconomic impact. 2. Aggarwal, & et, a. (2003). portfolio preferences of foreign institutional investor. 3. Ali Raza, et al. (2011) FPI, FDI and economic growth: the role of domestic financial sector. Islamabad: PIDE working paper. 4. Asma and salma, (2006) Empirical Analysis of Private Investments: The Case of Pakistan. 1 Fatima Jinnah Women University, Rawalpindi, Pakistan. 5. Durham, J. (2003). Foreign Portfolio Investment, Foreign Bank Lending and Economics Growth. International Finance Discussion Paper no 757. 6. EKEOCHA, & CHUKWUEMEKA, P. (2008). Modeling the long run determinants of foreign portfolio investment in an emerging market. International conference on Applied Economics-ICOAE. 7. Ekeocha, P. C., Stella, E. C., Victor, M., & Onyema, O. M. (2012). Modeling the Long Run Determinants Of Foreign Portfolio Investment in Nigeria. Journal of Economics And Sustainable Development. 8. Farman Afzal, at al, (2011) factors influsing foreign direct investment (FDI) in Pakistan. (The Case of Pakistan), Air University Islamabad, Pakistan. 9. Grabel. (1998). "Portfolio investment" Foreign Policy in focus. Inter Hemispheric Resource Center and Institute vol 3 no 13. 10. Grieg-Gran et al. (1998). Foreign Portfolio Investment and Sustainable Development A study of the Forest Product sectors in Emerging Markets. International Institution for Environment and Development. 11. Hasnan and Masood, (2002) foreign capital inflow and domestic saving in Pakistan with sector analysis. PDR (Pakistan development review), WP : 67. 12. Lee, & H, C. (2007). A survey of the literature on the Determinants of Foreign Portfolio Investment in the United State. Journal Review Of World Economics. 13. Muhammad Azam Naeem-ur-Rehman Khattak (2010) social and political factors effecting on foreign direct investment (FDI) in Pakistan. ~ 168 ~

14. Onuorah, Chi, A. c., Akujuobi, & Eze, L. (2013). Impact of Macroeconomics Indicators on the performance of Foreign Portfolio Investment in Nigeria. Europen Journal of Business and Management. 15. Onuorah, chi, A. c., Akujuobi, & Eze, L. (n.d.). Impact of Macroeconomics Indicators on the performance of foreign portfolio investment in Nigeria. European Journal of Busaness and Management. 16. Prasand, E.S, Rajan, R., & Subramanian, A. (2007). Foreign Capital and Economics Growth. National Bureau of Economics Research. 17. Rai, K., & Bhanumurthy, N. R. (2007). Determinants of foreign Instititional Investment in India: the role of return,risk and inflation". Institute of Economic Growth Delhi Universty Enclave. 18. Talat Afza and Zeshan Anwar (2013), Foreign Direct Investment (FDI) in Pakistan: Measuring Impact of Cost of War against Terrorism, Political Instability and Electricity Generation, Caspian Journal of Applied Sciences Research, pp: 117-127, Vol: 2(3). 19. Zafar and Sattar (2005). Economic growth and its determinants in Pakistan. PDR (Pakistan development review, 471-486. ~ 169 ~