Human capital and natural resource curse in Islamic countries

Eslamloueyan, K. & Ebrahimi Torkamani, A.

Human capital and natural resource curse in Islamic countries

DOI:10.13140/RG.2.1.3817.9928

Human capital and natural resource curse in Islamic countries Karim Eslamloueyan1and Ali Ebrahimi Torkamani Department of Economics, Shiraz University WP 06072012 June 2012 DOI: 10.13140/RG.2.1.3817.9928

Abstract Using a panel data model, this paper studies the impact of natural resource abundance and human capital accumulation on economic growth in Islamic countries. More specifically, it examines whether the resource-rich Islamic countries tend to grow more slowly than resource-poor Islamic countries. In other words, we investigate the validity of the so called "resource curse hypothesis" for members of Organization of Islamic Countries (OIC). Unbalanced panel data GLS with period fixed effects method is used to estimate the model. The results show that there is a negative relationship between natural resource abundance and economic growth rate for OIC members. This negative relationship holds true even after controlling for other variables that might affect output growth. We also find that physical capital has positive impacts on economic growth rate in these countries. Moreover, our finding supports the crucial role of education in boosting economic growth and in reducing the adverse effect of resource abundance on output growth rate for OIC as a block. This finding might have important policy implication for OIC policymakers and international institutions. Keywords: Islamic countries; Resource curse; Growth; Human capital; Panel data. JEL: E23, O13, O47

1. Introduction Do easy riches lead to sloth? The possible negative effects of natural resources on economic growth rate are largely explored through the so-called Resource Curse Hypothesis (RCH). Moreover, recent studies on RCH have shown the importance of human capital for building a sustainable development path. Are natural resources a curse for Islamic countries? If yes, can a large level of human capital offset the negative effect of natural resource abundance on output growth rate in these countries? Since so many OIC2 members are energy-rich economies and have abundant natural resources, it is important to examine whether the 1 2

Corresponding author, Department of Economics, Shiraz University, Shiraz, [email protected] Organization of Islamic Countries

1



DOI:10.13140/RG.2.1.3817.9928

resource curse hypothesis is valid for these counties. It might have important policy implications for policymakers and international institutions and can help to better understand the role of human capital accumulation in economic growth and development and hence the failure in natural resource-led development for these countries. Using a panel data model, this paper will study the impact of natural resource abundance and human capital accumulation on economic growth in selected Islamic countries. Our explanatory variables include initial GDP per capita, index of natural resource, labor force, physical capital, human capital, and other control variables. More specifically, we will examine whether there is a negative relationship between natural resource abundance and economic growth rate in these countries. In other words, we will estimate a panel data model to investigate the impact of natural resource abundance on output growth rate of selected OIC members. To the best of our knowledge no study has examined "the resource curse hypothesis" for a group of Islamic countries. Moreover, given the high emphasize that Islam puts on knowledge and education, our model will allow us to examine whether the accumulation of human capital can offset a possible negative impact of resource abundance on economic growth rate in Islamic countries. The experience of nineteen century and first half of twenty century show that natural resources was considered to be an important source of economic growth for many countries. However, since the second half of twenty century one can hardly find a resource-rich country with a high long-run average per capita growth rate. For example, Gylfason (2001, p. 848) discusses that in the second half of twenty century the average long-run economic growth rates of Asian tigers with relatively low natural resource abundance such as South Korea, Singapore, and Hong Kong was higher than many energy-rich countries such as Iran, Iraq and Kuwait.

2



DOI:10.13140/RG.2.1.3817.9928

In the literature there are many researches that support the resource curse hypothesis. Sachs and Warner (1995) in their influential paper show that there is a negative relationship between the exports of primary goods as a percentage of GDP and per capita income growth rate for a sample of 97 developing countries. They also show that resource abundance might indirectly affect output growth rate through its impact on the degree of trade openness, law, investment and the price of capital goods. Following seminal work of Sachs and Warner (1995) many other researchers developed cross section and panel data models to study the hypothesis of resource curse for a group of developed and developing countries. Sachs and Warner (1997) approve the resource curse hypothesis for 87 developing counties. They show that natural resource-rich economies tend to grow slowly. The negative relationship between resource abundance and output growth rate holds true even after they control for variables that are found to be important for economic growth. Gylfason, Herbertsson and Zoega (1999) develop a two sector endogenous growth model to show the presence of an inverse relationship between the size of primary sector and economic growth for sample of 125 countries for the period 1960-1992. Sachs and Warner (2001) show that resource-abundant countries are high-price economies. They find out that natural resource abundance has negatively affected the economic growth of 99 countries for the period 1970-1980 through its positive impact on general price level in these countries. Moreover, they find little evidence that geographical or climate variables might explain the natural resource curse. Gylfason (2001) studies the relationship between human capital and natural resources. This researcher shows that natural resources has negative impact on human capital in resource

3



DOI:10.13140/RG.2.1.3817.9928

rich countries. He also shows that an increase in human capital can decrease the negative impact of natural resources on economic growth rate. Stijns (2001) shows that there is a positive relationship between resource abundance and human capital in a sample of 102 countries for the period of 1970-1999. Bravo-Ortega and Gregorio (2002) develop a two sector endogenous growth model to study the relationship between resource abundance, human capital and economic growth. Using a panel data model, they show that the presence of human capital might decrease the negative impact of resource abundance on output growth rate. Using a cross section data for 85 countries, Gylfason and Zoega (2002) show that there is a negative relationship between resource abundance and investment in physical capital, human capital, and financial development. Atkinson and Hamilton (2003) show that if natural resource abundance is used mostly to finance government consumption expenditure, it negatively affects the economic growth rate. They also show that initial per capita income has negative and years of schooling has positive impact of output growth. Contrary to the finding of Sachs and Warner (1995), Sala-i-Martin and Subramanian (2003) find that resource abundance has no direct negative impact on economic growth for a sample of 71 counties. However, they show that resource abundance can negatively affect the quality of institution and through this might have adverse impact on economic growth. Papyrakis and Gerlagh (2004) studies the direct and indirect effects of resource abundance on economic and show that natural resource adversely affect output growth rate. Bulte et al. (2005) investigate the relationship between resource abundance, institutions' quality, human development for a sample of 100 countries and show that resource-rich

4



DOI:10.13140/RG.2.1.3817.9928

countries suffer from low level of human development. Low level of human development in turn negatively affects the quality of institution and hence economic growth rate. Stijns (2006) also shows that there is negative relationship between human capital accumulation and different indices of natural resource abundance. Papyrakis and Gerlagh (2007) estimate cross section data model for 49 states of the US and approve the resource curse hypothesis. In other words, they show that the resource abundance has negatively affected the economic growth. They conclude that resource abundance has negatively affected output growth through decreasing the level of investment, education, trade openness and research and development and increasing the corruption level in in resource-rich states. Papyrakis and Gerlagh (2007) support resource curse hypothesis for US states. We find that natural resource abundance decreases investment, schooling, openness, and R&D expenditure and increases corruption in US states. They show that these effects can fu explain the negative effect of natural resource abundance on growth. Brunnschweiler (2008) focuses on the quality of institutions to study the relationship between resource abundance and economic growth for 79 countries during the period 19702000. This author challenges the resource curse hypothesis and show that natural capital has positive impact on output growth rate. To sum up many empirical studies approve "the resource curse hypothesis". It is also argued by some researchers that human capital accumulation can offset the negative impact of natural resource abundance on output growth rate. However, to the best of our knowledge no study has examined "the resource curse hypothesis" for Islamic countries as a block. Moreover, given the high emphasize that Islam puts on knowledge and education, it is also important to know whether the accumulation of human capital can adversely affect the possible negative impact of natural resources on output growth rate for the OIC members.

5



DOI:10.13140/RG.2.1.3817.9928

Hence, our panel data model includes a variable to capture the interaction between human capital and natural resource abundance for Islamic countries. The rest of the paper is organized as follows. We present the basic model in Section 2. Section 3 is devoted to data and methodology. Estimation results are also presented in this section. Section 4 is the concluding remarks.

2. The Basic Model

Resource curse hypothesis is inconsistent with the expectation of economists about the positive role of natural resources in economic growth. One might expect that natural resource abundance to boost economic growth and development through attracting foreign skills and capital, increasing exports and bringing foreign exchange, and providing raw materials for domestic production. However, the experience of many resource-rich countries in the past sixty years shows that it might not be true.

Many researchers attempt to explain this phenomenon. However, there is no universal accepted theory to explain the resource curse hypothesis.3

The abundance of natural

resources can adversely affect output growth rate through various mechanism such as lowering the quality of institutions, increasing mismanagement and corruption, raising fluctuation in revenue from exporting natural resources, and also through the so called Dutch disease.

In order to show how the presence of natural resource might negatively affect economic growth, we very briefly present the model provided by Romer (2006). His propsed

3

Sachs and Warner, 2001, P. 833.

6



DOI:10.13140/RG.2.1.3817.9928

model introduces natural resources to a standard neoclassical growth model. Consider the following Cobb Douglas production function: Y (t )  K (t ) R(t )  T (t ) A(t ) L(t )

1   

,  0,   0,   0,      1.

(1)

In which the output, Y is produced by using physical capital, K, natural resources, R, land, T, labour, L. A is level of technology or knowledge and  ,  , and  are parameters. It is assumed 





that K (t )  sY (t )  K (t ), R(t )  bR(t ); T (t )  0,



A(t )  gA(t ),



L(t )  nL(t ), and b  0. s,

 , n, g, and b are saving rate, rate of depreciation, growth rate of labour force, growth rate

of knowledge and growth rate of natural resources, respectively. The growth rate of land is zero. All of these rates are assumed to be exogenous. After solving the model, per capita output growth rate on the long-run equilibrium growth path is:

gYbgp  gYbgp  g Lbgp  L

(1       )(n  g )  b (1       ) g  b  (   )n n  1 1

(2)

Equation (2) shows that the per capita growth on the balanced growth path can be positive or negative. It means that the presence of natural resources in the model might have negative impact on per capita output growth rate. But the growth of technical progress might be large enough to compensate the adverse effect of natural resources on output growth rate.

Review of the empirical researches on natural resource abundance and economic growth shows cross-country evidence and some individual case studies that support the linkages between output growth and various determinants of economic growth including natural capital itself. We follow the literature on resource curse and economic growth and use Sachs and Warner (1995, 1997, and 2001), Gylfason, Herbertsson and Zoega (1999), Stijns

7



DOI:10.13140/RG.2.1.3817.9928

(2001 and 2006), Gylfason and Zoega (2002), Sala-i-Martin and Subramanian (2003), and Bravo-Ortega and Gregorio (2002) and estimate the following panel data econometric model:

git  0  1GDPi 0  2 Nit  3 Lit  4 Iit  5Git  6 Hit  7OPENit   it (3)

In which g it is growth rate of real GDP per capita of country i in period t.

GDPi 0 is real GDP per capita at the beginning of period for country i.

N it : Ratio of raw material exports to GDP

Lit : Labor force participation rate, (% of total population ages 15+)

I it : Ratio of capital formation to real GDP in country i in period t

Git : Ratio of government consumption spending to GDP in country i in period t

H it : Total high school (Secondary) attending (% of population aged 15 and over) Source: Barro R. & J.W. Lee 2010

OPEN it : Ratio of sum of total exports and imports to GDP for country i in period t.

The data are the average five-year periods are used for estimation purpose.

In order to study the impact of the interaction between the abundance of natural resources and human capital on economic growth, we introduce variable N it * H it to the above model. Hence, the following models will be estimated: git  0  1GDPi 0  2 Nit  3 Lit  4 Iit  5Git  6 Hit  7OPENit  8 Nit * Hit   it (4)

8



DOI:10.13140/RG.2.1.3817.9928

The variable N it * H it captures the interaction effects among natural resources and human capital. Hence, the estimation of the above equation allows us to see whether the effect of natural resources on economic growth can be influence by human capital. The differentiation of equation (4) with respect to N, gives the following equation:

dg   2  8 * H  0 dN

(5)

The above equation can be solved to find the threshold level of human capital. This means that if the level of human capital is above its threshold level, it might offset the negative impact of natural resource abundance on output growth rate.

3. Data and Methodology

The data are obtained from World Development Indicators (WDI) CD (2010) published by the World Bank. Human capital index are obtained from Barro R. & J.W. Lee (2010). Our sample includes 34 OIC member countries. The countries are selected based on availability of their data. The sample includes Albania, Algeria, Bahrain, Bangladesh, Benin, Cameroon, Egypt, Gabon, Gambia, Indonesia, Iran, Jordan, Kazakhstan, Kuwait, Kyrgyzstan, Malaysia, Maldives, Mali, Mauritania, Morocco, Mozambique, Niger, Pakistan, Qatar, Saudi Arabia, Senegal, Sudan, Syria, Togo, Tunisia, Turkey, Uganda, United Arab Emirates, and Yemen.

The data cover the average of five-year period from 1980 to 2008. Since five-year average period is used there is no need to conduct panel data unit root and cointegration tests. However, in other to choose appropriate estimation techniques we conduct Chow and Hausman tests. 9



DOI:10.13140/RG.2.1.3817.9928

We use the Chow test to choose between pooling and fixed effects models. In other words, this test allows to examine fixed effects versus poolability of data. The result of Likelihood Ratio (LR) test which is reported in Table1 shows that the null hypothesis of redundant fixed effects is rejected at about 1 percent significant level.

Table 1. Redundant Fixed Effects Tests Test period fixed effects Statistic 2.837298

Effects Test Period F

d.f. (5,159)

Prob. 0.0175

Next we conduct Hausman test to examine the correlated random effects. This test might help us to choose between random and fixed effects models.The result of this test is reported in Table 2. As the result show the correlation between explanatory variables and error term is not rejected at 8 percent significant level.

Table 2. Hausman Test Test cross-section random effects Test Summary Chi-Sq. Statistic Chi-Sq. d.f. Cross-section random 13.998754 8

Prob. 0.0818

Based on Chow and Hausman tests, we used fixed effects method to estimate our model. More specifically, we use panel Generalized Least Squaure method with period fixed effects to estimate equation 4. The estimation result is presented in Table 3.

Table 3. The panel data estimation result for period 1982-2008 Variable C GDP0 N I L G

Coefficient 2.078757 -0.000134 -8.142114 7.110932 -0.721976 -0.323273

Std. Error 1.598008 4.68E-05 3.280028 2.390667 2.041085 2.320440

10

t-Statistic 1.300842 -2.855611 -2.482331 2.974456 -0.353722 -0.139315

Prob. 0.1952 0.0049 0.0141 0.0034 0.7240 0.8894



DOI:10.13140/RG.2.1.3817.9928

H N*H OPEN R-squared 0.280610 Adjusted R-squared 0.221792

-0.011629 0.160685 0.039719

0.021958 -0.529612 0.5971 0.080840 1.987701 0.0486 0.738102 0.053812 0.9572 Durbin-Watson stat 1.763144 F-statistic 4.770814 Prob(F-statistic) 0.000001

The result shows that real GDP at the beginning on the period (1982) has expected negative impact on growth rate. This confirms the prediction of standard growth theories about conditional convergence. The coefficient of N is negative and statistically significant. Hence, it means that an increase in natural resources decreases the output growth rate. Hence, we cannot reject the resource curse hypothesis for OIC members. This negative relationship holds true even after introducing some control variables into the model. We also find that the ratio of physical capital to GDP, I has positive impacts on economic growth rate. However, the effects of government expenditure and trade openness on output growth rate are not statistically significant. Moreover, the coefficient of N*H is positive and statistically significant. It means that the interaction of human capital and natural capital can offset the negative impact of resource abundance on output growth rate for OIC as a block. However, the effects of government expenditure, G and trade openness, Open, on output growth rate are statistically insignificant.

Now we can use equation (5) to calculate the threshold level of human capital that is need to offset the negative effect of natural resource abundance on economic growth rate. After substituting for  2 and  8 in equation (5), we get:

dg  8.1421  0.1607 H  0  H  50.6665 dN

11

(6)



DOI:10.13140/RG.2.1.3817.9928

According to the above result, the threshold level for the block of OIC is about 50.7 percent. It means that the human capital needed to offset the negative impact of natural resource abundance on output growth rate, is at least 50.7. In other words, if more than 50 percent of total population in these countries have high school education, human capital accumulation can decrease the negative impact of natural resources on economic growth rate.

4. Concluding remarks The negative effects of natural resources on economic growth are largely explored through the so-called resource curse hypothesis. Recent studies show the importance of human capital for building a sustainable development path. Given the importance of education in Islamic teaching and also knowing that many energy and natural resource-rich countries are members of OIC, we developed a panel data model to examine whether the hypothesis of resource curse is valid for these countries as a block. Moreover, this paper studies the interaction between natural resource abundance and human capital for selected members of OIC. Our modeling allows to examine whether human capital accumulation can offset a possible negative impact of natural resource abundance on output growth rate in these countries. The explanatory variables include initial GDP per capita, index of natural resource abundance, labor force, physical capital, human capital, and other control variables. Unbalanced panel data Generalized Least Square (GLS) with fixed period effects method is used to estimate the model. The results show that there is a negative relationship between natural resource abundance and economic growth rate for OIC members. This negative relationship holds true even after controlling for other factors that might affect

12



DOI:10.13140/RG.2.1.3817.9928

output growth. We also find that physical capital and human capital have positive impacts on economic growth rate. However, the effects of government expenditure and trade openness on output growth rate are statistically insignificant. Moreover, we find out that human capital accumulation can offset the negative impact of resource abundance on output growth rate for OIC as a block. Hence, it emphasizes on the importance of investment in human capital for building a sustainable development path. This finding might have important policy implication for OIC policymakers and international institutions. More specifically, it supports the crucial role of education in boosting economic growth and in reducing the adverse effect of resource abundance on output growth rate.

13



DOI:10.13140/RG.2.1.3817.9928

References: Atkinson, G. and Hamilton, K. (2003). "Savings, growth and the resource curse hypothesis." World Development, Vol. 31, No. 11, pp. 1793-1807. Bravo-Ortega, C. and Gregorio, J. D. (2002). "The relative richness of the poor: natural resources, human capital and economic growth." Central Bank of Chile Working Papers, No. 139, pp. 1-36. Brunnschweiler, C. N. (2008). "Cursing the blessings? Natural resource abundance, institutions, and economic growth." World Development, Vol. 36, No. 3, pp. 399-419. Bulte, E. H., Damania, R., and Deacon, R. T. (2005). "Resource intensity, institutions, and development." World Development, Vol. 33, No. 7, pp. 1029-1044. Costantini, V. and Monni, S. (2008). "Environment, human development and economic growth." Ecological Economics, Vol. 64, pp. 867-880. Gylfason, T. (2001). "Natural resources, education, and economic development." European Economic Review, Vol. 45, pp. 847-859. Gylfason, T., Herbertsson T. and Zoega, G. (1999). "A mixed blessing: Natural resources and economic growth." Macroeconomic Dynamics, Vol. 3, pp. 204–225. Gylfason, T. and Zoega, G. (2002). "Natural resources and economic growth: The role of investment." Central Bank of Chile Working Papers, No. 142, pp. 1-47. Papyrakis, E. and Gerlagh, R. (2004). "The resource curse hypothesis and its transmission channels." Journal of Comparative Economics, Vol. 32, pp. 181-193. Papyrakis, E. and Gerlagh, R. (2007). "Resource abundance and economic growth in the United Stats." European Economic Review, Vol. 51, pp. 1011-1039. Romer, D. (2006). Advanced macroeconomics, Third Edition, McGrow-Hill.

14



DOI:10.13140/RG.2.1.3817.9928

Sachs, J. D. and Warner, A. M. (1995). "Natural resource abundance and economic growth.", National Bureau of Economic Research Working paper No. 5398, Cambridge, MA. Sachs, J. D. and Warner, A. M. (1997). "Natural resource abundance and economic growth." Center for International Development and Harvard Institute for International Development, November. Sachs, J. D. and Warner, A. M. (2001). "Natural resources and economic development: The curse of natural resources." European Economic Review, Vol. 45, pp. 827-838. Sala-i-Martin, X. and Subramanian, A. (2003). "Addressing the natural resource curse: An illustration from Nigeria." NBER Working Paper, No. 9804. Stijns, J. P. (2001). "Natural resource abundance and human capital accumulation." Stijns, J. P. (2006). "Natural resource abundance and human capital accumulation." World Development, Vol.34, No. 6, pp. 1060–1083.

15