Does the government size affect output growth in developing countries? Evidence from selected ASEAN countries.

Kaushik, Neetu ; Dhakal, Dharmendra ; Upadhyaya, Kamal 等

Abstract

This paper studies the effect of government size economic growth in selected AEAN countries. A standard growth model is developed in which the output growth is a function of the size of government, growth in labor supply and total stock of capital as percentage of GDP. The estimated results suggest that increase in the capital stock is the primary factor that helps to grow the economy. Government size did not have either positive or negative effect on output growth. Growth rate of labor does not affect the output growth in Thailand, Malaysia, and Indonesia. In case of the Philippines labor supply seems to have a negative effect on output growth probably because of excess labor supply in relation to other inputs.

JEL Category: E61, H50, O40

Key words: government size, economic growth, ASEAN countries, growth model

I. INTRODUCTION

The relationship between the size of government and economic growth has been a debatable issue for several decades. In early fifties and sixties there was some consensus among development economists as well the policy makers that an ideal system is a mixed economic system in which the government sector requires to play an important role in the economy. Particularly, in developing countries where there was not enough infrastructure development the government had to play an important role in developing modern modes of transportations as well as communications. In addition to developing the infrastructures some governments are also engaged in supplying the basic goods as well as services to the general public at some fair price to ensure that the people at the margin have access to the basic needs.

By early eighties it was realized that mixed economic system with big government sector essentially distorts the production as well as consumption in the economy which eventually slows down the pace of growth. That is why the last two decades of twentieth century saw a massive privatization in developing countries. The industrialized countries, particularly the USA and the United Kingdom started to reduce the size of government arguing that the smaller government with less economic regulation is much better for growth of an economy. The recession that started in 2008 brought the issue of role of government again in the forefront. Many social scientists now have started arguing that we need a bigger government in order to sustain economic growth (Madrick, 2008; Sachs 2009). In view of this controversy in this paper we estimate and analyze the effect of government size on economic growth in selected ASEAN nations (Association of South East Asian Nations) namely, Indonesia, Malaysia, the Philippines, and Thailand. We hope that the findings of this paper will shed some light on the relationship between government size and economic growth in Southeast Asian countries.

II. REVIEW OF LITERATURE

As indicated above the issue relating to the government size and economic growth has been an issue of contention for a long time. Tomas Hobbes in 1651 described life without government was "nasty, brutish, and short" and argued that the law and order provided by government was a necessary component of civilized life (Gwartney et al. 1998). Obviously, Hobbes was indicating that the role of government in maintaining the law and order, protecting the property rights smooth functioning of the judicial system was important in a society not only in maintaining a civilized society but also to enhance economic growth. In other words, secure property rights, contracts enforcements and a stable monetary regime provide the foundation for a smooth operation in a market economy (Gwartney et al. 1998).

In addition to ensuring the smooth operation of market economy the government also provides stimulus for economic growth by providing quality infrastructures and public goods which market often fails to supply. There is no contention on the provision of public goods by the government either in well-developed western nations or in developing countries of Asia or Africa. Although the infrastructures such as roads, railways, and public utilities can be supplied by the private sector but because of its high cost to build and the free rider problem the private sector shy away from such ventures forcing the governments to provide these services.

Social services such as providing social security to the marginalized group of the society, health care and education for the low income people of the society have become other responsibilities of the government. In most developed countries expenses for providing these services have been skyrocketing over last few decades which is bloating the government budgets leading to a continuous growth of the government size. Often it is argued that such expenses do help the economy to grow by creating additional demand for goods and services through multiplier effect. But the effectiveness of such government programs on economic growth is still an empirical question.

There are several empirical studies on the government expenditure (size of government) and its relationship with economic growth. The findings of these researches are inconsistent. In other words, some researchers find government size negatively affecting the economic growth and other find this not to be true. A careful look, however, reveal that the findings that these differences are due to the measure of government size and type of countries studies e.g. rich and poor (Bergh and Henrekson, 2011). When looked at the industrialized countries there is almost a unanimous consensus that the size of government and per capita real GDP growth are negatively associated (Di Pietro et al., 1993) and it is also found that during the economic downturn the government size grows (Bergh and Henrekson, 2011).

Using a systematic analysis on the relationship between increases in spending, lower rates of economic growth, high unemployment, increases in deficits and inflation among 19 nations Cameron (1982) concludes that high levels of spending and large increases in spending have not caused stagflation. In addition, his conclusion also suggests that a large and expanding welfare state is beneficial to the market economy. Ram (1986) in his seminal paper which uses a cross-sectional and time series data for 1960-70 and 1970-80 concludes that government size has a positive effect on economic performance and growth. He further concludes that government size has a positive externality effect on the rest of the economy and the factor productivity was higher in the government sector than in the rest of the economy during 1960s.

Landau (1993) in his cross-country analyses with 48 countries added control variables for education, energy consumption and some dummies for geographical variables and finds a negative relationship between government spending and economic growth. Likewise, Marlow (1986), based on his study of 19 countries for a period of data from 1960 to 1980 also argues that the public sector size retards overall economic growth.

There is relatively few research on this issue with developing countries data set. One such study is by Landau (1986) which essentially is an extension of his earlier work mentioned above. In this study he used 96 country cross-sectional data from 1960 to 1980 and finds a negative relationship between the government size and economic growth in LGCs. Guseh (1997) in his study differentiates the effects of government size on economic growth across political and economic systems in developing countries. The results show that growth in government size has negative effects on economic growth, but the negative effects are three times as great in nondemocratic socialist systems as in democratic market systems.

III. THEORETICAL BACKGROUND, METHODOLOGY AND DATA

In order to estimate the effect of government size on economic growth the following model is developed, which is based on the classical growth model:

Y = f(K, L) (1)

where, Y = total output (real GDP).

K = capital stock in the economy

L = total amount of labor in the economy

Assuming a constant returns to scale an increase in either the amount of capital or number of labor will increase the production of total output in the economy. Once we add the government spending in above equation we have:

Y - f(K, L, G) (2)

where, G = Government spending. The statistical form of equation (1) is as follows:

Y = [c.sub.0] + [c.sub.1] K + [c.sub.2] L + [c.sub.3] G + e (3)

In equation (3) as indicated above [c.sub.1] and [c.sub.2] are expected to be positive. The variable of interest is G. If it's coefficient [c.sub.3] is positive and statistically significant government size has a positive effect on economic growth, if it is negative and statistically significant it has a negative effect on economic growth. If it is not statistically significant the size of government has no effect on economic growth. In equation (3) e is random error term.

The first difference of log of Y gives the real GDP growth rate and log of L gives the labor growth rate in the economy. Once these variables are converted into growth form equation (3) can be written as:

y = [c.sub.0] + [c.sub.1] k + [c.sub.2] l + [c.sub.3] g + e (4)

where, y = real output growth rate

k = total capital stock as percentage of GDP

I = labor force growth rate

g = government spending as percentage of GDP

As indicated above this study is conducted for four ASEAN countries namely, Indonesia, Malaysia, the Philippines and Thailand. Annual time series data from 1976 to 2012 is used. All the data are derived from the WDI of the World Bank.

IV. EMPIRICAL FINDINGS AND ANALYSIS

Estimations of equation (4) are reported in Table 1. In the initial estimation an autocorrelation problem was detected for both the Philippines and Thailand. In order to eliminate the autocorrelation problem an AR(1) model is estimated for these two countries. All the estimations seem fit well in terms of the sign of the coefficients of the variables, coefficient of determination, Durbin Watson values and F statistics.

Few interesting findings are observed from the estimated results. First, in all the estimations the coefficient of capital growth is positive and statistically significantly different from zero. This finding is consistent with the conventional wisdom that in developing countries there is a scarcity of capital, therefore, the marginal product of capital is positive. Indeed our estimated results suggest that if the percentage of capital as GDP is increased by 10 percent the real GDP grows anywhere from 2.1 to 3.8 percent in these countries. Second, the estimated coefficient of labor growth shows that an increase in the growth of labor has no effect in output growth in Indonesia, Malaysia and Thailand. In these countries a significant percentage of labor force are employed in agriculture where the marginal product of labor is almost zero or even negative. Indeed in case of the Philippines the marginal product of labor is negative presumably because of excess labor use and disguised unemployment.

As indicated above, the coefficient of g i.e. the government size is the main focus of this study. The estimated results could not detect any statistically significant effect of the size of government on the output growth in any of the country under study except for Indonesia. Even in case of Indonesia the coefficient is marginally significant (only at 11% critical level). In order to see if there is any lagged effect we estimated the model with one and two years lag of the government size variable but the result did not change. Based on the coefficients of the government size in all the estimations it can be concluded that the size of government expenditure has neither positive nor negative effect in these four ASEAN countries. In order to see if there is any lagged effect we estimated the model with one and two years lag of the government size variable but the result did not change.

V. SUMMARY AND CONCLUSION

This paper studies the effect of government size on economic growth in selected AEAN countries namely, Indonesia, Malaysia, the Philippines, and Thailand. A standard growth model is developed in which the output growth is a function of the size of government, growth in labor supply and total stock of capital as percentage of GDP. The model is estimated using annual time series data from 1976 to 2012. Since the initial estimation suggested an autocorrelation problem the model is estimated with an AR(1) term for the Philippines and Thailand. The estimated results suggest that increase in the capital stock is the primary factor that helps to grow the economy. Government size did not have either positive or negative effect on output growth. Growth rate of labor does not affect the output growth in Thailand, Malaysia, and Indonesia. In case of the Philippines labor supply seems to have a negative effect on output growth presumably because of excess labor supply in relation to other inputs.

References

Bergh, A. and Henrekson, M. (2011), "Government Size and Economic Growth: A Survey and Interpretation of the Evidence" Journal of Economic Surveys, vol. 25, pp. 872-897.

Cameron, D. (1982), "On the Limits of Public Economy" Annals of Academy of Political and Social Science, vol. 459, pp. 46-62.

Di Pietro, W. R., Sawhney, B. and Jampani, R. (1993), "Determinants of Economic Growth: Does the Stage of Economic Development Matter?" Rivista Internazionale di Scienze Economiche e Commercially vol. XL, 1993, pp. 731-739.

Guseh, J. (1997), "Government Size and Economic Growth in Developing Countries: A Political Economy Framework" Journal of Macroeconomics, vol. 19. pp. 175-192.

Gwartney, J., Lowson, R. and Holocomb, R. (1998), "The Size and Functions of Government and Economic Growth" Joint Economic Committee, Washington, D.C.

Landau, D. (1993), "Government Expenditure and Economic growth: A Cross-Country Study" Southern Economic Journal, vol. 49, pp. 783-792.

--(1986), "Government and Economic Growth in Less Developed Countries: An Empirical Study for 1960-1980," Economic Development and Cultural Change, pp. 36-75.

Madrick, J. (2009), The Case for Big Government. Princeton, N.J.: Princeton University Press.

Marlow, M. (1986), "Private Sector Shrinkage and Growth of Industrialized Economies" Public Choice, vol. 49, pp. 143-154.

Sachs, J. (2009), "The Case for Bigger Government" Time, Thursday, January 8, pp. 1-3.

Ram, R. (1986), "Government Size and Economic Growth: A New Framework and Some Evidence from Cross-Section and Time-Series Data," American Economic Review, vol. 76(1), pp. 191203.

World Bank (1998), World Development Indicator, Washington, D.C.

NEETU KAUSHIK, Assistant Professor of Economics, Delmar College, Corpus Christi, Texas, E-mail: [email protected]

DHARMENDRA DHAKAL, Professor of Economics and Finance, Tennessee State University, Nashville, Tennessee, E-mail: [email protected]

KAMAL UPADHYAYA, Professor of Economics, University of New Haven, West Haven, CT 06516, E-mail: [email protected]

Table 1
Estimation of equation (4); dependent variable y

Country

Variable         Coefficient     Indonesia       Malaysia

CONST.            [C.sub.0]        10.16           0.69
                                 (2.54) **        -1.23
k                 [C.sub.1]         0.34           0.21
                                 (2.89) ***     (2.49) **
l                 [C.sub.2]         0.05           1.49
                                   (004)          (0.96)
g                 [C.sub.3]         0.74           0.21
                                  (1.60) #        (0.67)
AR(1)                                --             --

Adj [R.sup.2]                      0.288          0.143
F                                   5.82           2.99
D.W.                                1.67           1.72

COUNTRY

Variable         Philippines      Thailand

CONST.               8.11           4.06
                    -0.87          -0.32
k                    0.37           0.38
                   (1.91) *      (2.32) **
l                   -5.16           1.06
                  (2.05) **        (1.17)
g                    0.11          -0.33
                    (0.20)         (0.40)
AR(1)                0.39           0.39
                  (2.12) **      (2.26) **
Adj [R.sup.2]        0.31           0.37
F                    4.96           6.14
D.W.                 1.73           1.94

Note: Figures in the parentheses are t-values for the corresponding
coefficients.

***, **, *. # significant at 1%, 5%, 10%, and 15% critical level
respectively.