. International Journal of Economics and Financial Issues ISSN: 2146-4138 available at http: www.econjournals.com International Journal of Economics and Financial Issues, 2016, 6(4), 1299-1303. International Journal of Economics and Financial Issues | Vol 6 • Issue 4 • 2016 1299 The Role of Domestic Private Investment in Promoting Per Capita Income Convergence amongst Association of Southeast Asian Nations-5 Economies Sallahuddin Hassan1*, Musa Murtala2 1School of Economics, Finance and Banking, Universiti Utara Malaysia, Malaysia, 2Department of Economics, Usmanu Danfodiyo University, Sokoto, Nigeria. *Email: din636@uum.edu.my ABSTRACT Using data spanning 1980-2014, this study investigates the phenomena of per capita income convergence across a panel of Association of Southeast Asian Nations-5 (ASEAN-5) economies. The study also delves on the role of domestic private investment in facilitating per capita income convergence amongst the economies. Employing seemingly unrelated regression augmented Dickey-Fuller test to investigate per capita income convergence of the countries to the group average, it was observed that, as opposed to Malaysia and Singapore, economies of Indonesia, Philippines and Thailand tend convergence to the group average. On the impact of gross domestic investment (GDI) on per capita income convergence, results from system generalized method of moments indicate that gross domestic product gap ratio between a pair of economies tend to diminish with a reduction in the GDI gap ratio between the economies. Keywords: Convergence, Domestic Capital Formation, Seemingly Unrelated Regression Augmented Dickey-Fuller, Generalized Method of Moments, ASEAN-5 JEL Classifications: E31, E52, E58 1. INTRODUCTION One of the major implications of neoclassical growth theory is that relatively poor economies, with low initial per capita, tend to grow faster than relatively richer ones to a point where convergence in per capita income across world economies would be achieved. On the whole, the economic convergence proposition rests upon two assumptions: Diminishing marginal return on physical capital and technological gap (Khan and Kumar, 1993). One of the assumptions of neoclassical growth model of Solow (1956) is that relatively poor economies have low capital to labour ratios compared to the richer ones. As a result, capital investment in such economies with low capital-labour ratios is expected generate higher returns on capital than in the relatively richer economies. Furthermore, in search for higher returns on investments, international capital flows from economies with relative capital abundance to economies where relative capital scarcity exists. Consequently, poor economies would grow faster than the richer ones. Investment is very central the idea convergence. This is the case because the rate of return on investment is an important determinant of catch-up among economies. Therefore, the high the volume of investment in a low-performing economy, the higher the chances that the economy would do better in terms of catch-up. The second basis for income convergence has to do with technology transmission. Richer economies have high stock of technology relative to the poor ones and as a result technology is transferred from richer economies to the poor ones. However, the rate of technology absorption of a country is a positive function of technology gap between the economy and world technology stock. Hence, relatively poor economies would growth faster than the richer ones. Recent decades have witnessed proliferation of studies aimed at testing the validity of convergence hypothesis as well as exploring on various determinants of convergence. Islam (2003) provides a comprehensive survey of convergence literature. By and large, Hassan and Murtala: The Role of Domestic Private Investment in Promoting Per Capita Income Convergence amongst Association of Southeast Asian Nations-5 Economies International Journal of Economics and Financial Issues | Vol 6 • Issue 4 • 20161300 it can be said that results emanating from cross-section and panel data studies are supportive of the convergence hypothesis. Prominent among cross-section based studies include: Kormendi and Meguire (1985), Baumol (1986), Grier and Tullock (1989), Barro and Sala-i-Martin (1992) and Rodríguez-Pose et al. (2012). On the other hand, Knight et al. (1993), Islam (1995), McCoskey (2002), Weeks and Yao (2003) and Lei and Tam (2010) are among the famous studies that used panel data. In contrast to cross-section and panel data related studies, time-series based studies, such as Carlino and Mills (1993), Aubyn (1999), Cellini and Scorcu (2000) and Habibullah et al. (2012), reported more conflicting findings. However, despite such efforts, there appears to be an apparent neglect on investigating the role domestic investment in determining per capita income convergence in general and particularly Association of Southeast Asian Nations-5 (ASEAN-5) economies. Therefore, this study seeks to examine the role of domestic investment in achieving per capita income convergence amongst the ASEAN-5 countries. The paper further delves on the performance of individual economies in terms of achieving convergence the group average. In addition to conventional unit root testing procedures, the paper utilized an alternative approach based on seemingly unrelated regression augmented Dickey-Fuller (SURADF). ASEAN-5 comprises of: Indonesia, Malaysia, Philippines, Singapore and Thailand. The economies are pioneer member of ASEAN. Present membership composition of the association is made up of 10 economies, the additional ones being Brunei, Cambodia, Lao DPR, Myanmar and Vietnam. ASEAN-5 economies are considered suitable sample for the fact that the economies for a couple of reasons. The countries are the pioneer members of the association, and as a result forming a better sample to explore the extent to which the association achieve its first objective of accelerating “the economic growth, social progress and cultural development in the region through joint endeavours ….” Given the fact that one of the cardinal objectives of establishing the group has to with achieving economic growth, and by extension economic convergence, investment is required. To provide some insight on economic performance amongst ASEAN-5 countries, the rankings of 10 year average per capita gross domestic product (GDP) of the countries over the period 1960-2014 are presented in Table 1. For the entire period under review, Singapore maintained the lead in terms of average annual per capita GDP. The economy has not been overtaken throughout the period. As for the Malaysian economy, it was also observed that the economy maintained consistent performance, ranking second for the whole sample period. Philippines on the other hand ranked third, next to Malaysia, for the decades 1960-1969 and 1970-1979 before being overtaken to rank forth for the subsequent decade – 1980-1989 – by the Thailand economy. Prior to overtaking Philippines in 1980-1989 to rank third for the rest of period under review, the Thailand economy ranked forth for the decades 1960-1969 and 1970-1979. Finally, for the first three decades reviewed (1960-1969, 1970-1979 and 1980-1989) Indonesia ranked fifth and forth for the last three decades. The ranking provided above may not be unrelated to the capital stock available in each economy and rate of return on investment. Furthermore, Figure 1 shows plots for the standard deviation, as a measure of convergence, for both per capita income and gross domestic fixed capital formation over the period under review for the ASEAN-5. As can be observed, both measures depicts similar pattern, except for the period 1998-2002, when the convergence in gross domestic capital formation increased as shown by fall in the standard deviation. The rest of the paper is structured into three sections. Section two hosts methodology. Section three contains empirical results. Finally, section four provides policy implications and conclusion. 2. METHODOLOGY 2.1. Data Data covering the period 1980-2014 on the variables of interest were obtained from the World Bank. Our choice of timeframe for the study was mainly influenced by data availability. In order to remove the inherent cyclical effect present in most macroeconomic time series data, we follow the common practice of averaging data points over 5 year non-overlapping periods. Thus, arriving at seven observations per variable for each panel member. 2.2. Empirical Model In the standard literature of economic convergence, GDP deviations are calculated using three alternatives: Deviation of individual economies from their group average; difference between the GDP of a particular country and that of a chosen reference economy; and pair-wise output differences of all possible pairs for a given sample of economies. Figure 1: Convergence in per capita gross domestic product and gross fixed capital formation, 1980-2014 Table 1: Real GDP per capita ranking ASEAN-5 economies Country 1960-69 1970-79 1980-89 1990-99 2000-09 2010-14 Indonesia 5 5 5 4 4 4 Malaysia 2 2 2 2 2 2 Philippines 3 3 4 5 5 5 Singapore 1 1 1 1 1 1 Thailand 4 4 3 3 3 3 GDP: Gross domestic product, ASEAN-5: Association of Southeast Asian Nations-5 Hassan and Murtala: The Role of Domestic Private Investment in Promoting Per Capita Income Convergence amongst Association of Southeast Asian Nations-5 Economies International Journal of Economics and Financial Issues | Vol 6 • Issue 4 • 2016 1301 Following Ben-David (1993) and Ben-David and Bohara (1997), for a sample of n economies with annual per capita GDP given by yit and its average as y̅t, convergence or divergence of incomes can be described by an estimation of Equation 1. yi,t−y̅t=Φ(yi,t-1−y̅t-1)+ϵi,t, (1) If Φ<1, it shows evidence of per capita income convergence. Letting zit=yit−y̅it and zi,t−1=yi,t-1−y̅t-1, Equation 1 can be written as: zit=Φzi,t−1 (2) Furthermore, Equation 2, according to Nelson and Plosser (1982), can be presented in an ADF framework as: Z Z Z eit i i t j p i j i t j i t= + ∑ ∆ +− = −φ δ, , , ,1 1 (3) Therefore, testing the hypothesis that Φ<1 narrows down to a unit root test using ADF approach. In this study, SURADF procedure is employed to test the convergence hypothesis using Equation 3. SURADF was proposed in Breuer et al. (2001; 2002). The approach entails performing ADF unit root test within the framework of SUR. The method offers a number of advantages over single equation ADF and other traditional panel data unit root testing procedure. The main disadvantage associated with SURADF method is the lack of standard distribution. Simulations using the underlying dataset have to be carried out in the course of using the method. Critical values for the SURADF test were generated by simulation using 10,000 replications using the underlying dataset. Two alternatives are commonly employed in testing the hypothesis that Φ<1 in economic convergence literature: One, employing the traditional time-series based single equation ADF test. Two, pooling data to perform a panel data based unit root test, as ones due to Levin et al. (2002) and Im et al. (2003). However, both single equation ADF test and traditional panel data unit root testing procedures suffer some drawbacks. At least within the context of testing the convergence hypothesis across a sample of economies located within a given region, conclusions based single equation ADF may be erroneous. This is the case because such economies, as the case of ASEAN-5, are very much unlikely to be independent. As noted in Ben-David and Bohara (1997), in such instances, the error covariance structure of ϵit is not a diagonal one. This by implication implies that shocks transmit across sampled countries and thus results based on single equation ADF test can be inaccurate. As for the traditional panel data based unit root test procedures, rejecting the null hypothesis that a given series does not contain a unit root does not mean that the series are independently stationary for each of the panel members. In other words, traditional panel unit root testing methods cannot be used to detect the mix of I(0) and I(1) in a panel setting (Breuer et al., 2001). Taking into account these issues, we resort to SURADF method to test the convergence hypothesis for the ASEAN-5 economies. Given the assumption that marginal return on investment, and by extension economic growth, is a negative function of capital-labour ratio in a given economy, it is logical to believe that as the capital- labour ratio of two economies converge; the growth rate gap between the economies diminishes. It is against this background that this study hypothesised that for any pair of countries, GDP gap ratio is a positive function of domestic capital formation gap ratio as depicted by Equation 4. PGDPRijt=PGDIRijt+GFCERijt+TOPNRijt+ϵijt (4) Where, PGDPijt=|PGDPit−PGDPjt|/(PGDPit+PGDPjt), GDIRijt=|PGDIit−PGDIjt|/(PGDIit+PGDIjt), GFCEijt=GFCEit+GFCEjt/2, TOPNijt=TOPNit+TOPNjt/2, εit is the error term and subscripts i, j and t account for data points for countries i and j in year t. The variables used in this study to estimate Equation 4 are GDP per capita (PGDP), gross domestic investment (GDI), government final gross consumption expenditure (GFCE) and trade openness (TOPN). With the exception of TOPN, which is measured as a total trade share of GDP, all other variables are measured in constant USD. Following Choi (2004; 2009), this study pooled series gap pairs for 10 possible pairs of countries in estimating the hypothesized relationship.1 Equation 4 was estimated using Model 1, Model 2 and Model 3. Model 1 estimates the relationship between PGDPR and GFCF alone. Model 2 estimates the relationship between the variables of interest, controlling for TOPN. Finally, in Model 3 we control for both GFCE and TOPN. Generalized method of moments (GMM) proposed in Blundell and Bond (1998) was used in estimating the relationship. Parameter estimates using system-GMM are more consistent when handling data with shorter time span. 3. EMPIRICAL RESULTS 3.1. Unit Root Test Table 2 shows SURADF test results. The test was performed on the deviation series (yit−y̅t) for each economy. For comparison purposes, we perform both ADF and SURADF tests on the series. As the table shows, the economies of Indonesia, Philippines and Thailand show evidence of convergence to the group average over the period 1960-2014. However in Malaysia and Singapore 1 The pairs are: Indonesia-Malaysia, Indonesia-Philippines, Indonesia- Singapore, Indonesia-Thailand, Malaysia-Philippines, Malaysia-Singapore, Malaysia-Thailand, Philippines-Singapore, Philippines-Thailand and Singapore-Thailand. Hassan and Murtala: The Role of Domestic Private Investment in Promoting Per Capita Income Convergence amongst Association of Southeast Asian Nations-5 Economies International Journal of Economics and Financial Issues | Vol 6 • Issue 4 • 20161302 the null hypothesis of no convergence cannot be rejected over the sample period. 3.2. System GMM Estimation Table 3 contains estimation results of Model 1, Model 2 and Model 3. Results for Model 1 show a positive relationship between the gap ratios of real PGDP and that of real GDI per capita. The relationship is statistically significant at 10% level of significance. According to the results, on the average, 1% increases in the level of real GDI gap ratio between two economies leads to a 0.17% increases in the GDP gap ratio amongst the ASEAN-5 economies. By controlling for TOPN in Model 2, the estimated parameter for GDI increases to 0.34. This suggests that the convergence is stronger after controlling for the average TOPN for each pair of economies. The relationship is also statistically significant at 5% level of significance. Model 3 that controls for the government final consumption expenditure and TOPN also indicate that lower GDP gap ratios are associated with low GDP gap ratio amongst the ASEAN-5 countries. On the average, a 1% decreases in the GDI gap ratio tends to lower real GDP ratio by 0.36%. The relationship is also statistically significant at 5%. 4. CONCLUSION AND POLICY IMPLICATIONS Findings from the study show a positive link between GDI gap ratio and real GDP gap ratio (convergence). This implies that GDI plays a significant role in facilitating per capita income convergence across ASEAN-5. In other words, as two economies become similar in terms of GDI, the GDP gap ratio between them shrinks. In line with this finding, it is recommended that governments in ASEAN-5 should strengthen the existing investment promotion policies and create new incentives (both tax and non-tax) policies capable of enhancing domestic private investment in order to facilitate convergence across the economies. REFERENCES Aubyn, M.S. (1999), Convergence across industrialised countries (1890–1989): New results using time series methods. Empirical Economics, 24(1), 23-44. Barro, R.J., Sala-i-Martin, X.X. (1992), Convergence. Journal of Political Economy, 100(2), 223-251. Baumol, W.J. (1986), Productivity growth, convergence, and welfare: What the long-run data show. The American Economic Review, 76(5), 1072-1085. Ben-David, D. (1993), Equalizing exchange: Trade liberalization and income convergence. The Quarterly Journal of Economics, 108(3), 653-679. Ben-David, D., Bohara, A.K. (1997), Evidence on the contribution of trade reform towards international income equalization. Review of International Economics, 5(2), 246-255. Blundell, R., Bond, S. (1998), Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87(1), 115-143. Breuer, J.B., McNown, R., Wallace, M. (2002), Series-specific unit root tests with panel data. Oxford Bulletin of Economics and Statistics, 64(5), 527-546. Breuer, J.B., McNown, R., Wallace, M.S. (2001), Misleading inferences from panel unit-root tests with an illustration from purchasing power parity. Review of International Economics, 9(3), 482-493. Carlino, G.A., Mills, L.O. (1993), Are US regional incomes converging? A time series analysis. Journal of Monetary Economics, 32(2), 335-346. Cellini, R., Scorcu, A.E. (2000), Segmented stochastic convergence across the G-7 countries. Empirical Economics, 25(3), 463-474. Choi, C. (2004), Foreign direct investment and income convergence. Applied Economics, 36(10), 1045-1049. Choi, C. (2009), Does bilateral trade lead to income convergence? Panel evidence. Journal of Economic Development, 34(1), 71-79. Grier, K.B., Tullock, G. (1989), An empirical analysis of cross-national economic growth, 1951–1980. Journal of Monetary Economics, 24(2), 259-276. Habibullah, M.S., Dayang-Affizzah, A.M., Puah, C.H. (2012), Regional income disparities in Malaysia: A stochastic convergence analysis. Malaysian Journal of Society and Space, 8(5), 100-111. Im, K.S., Pesaran, M.H., Shin, Y. (2003), Testing for unit roots in heterogeneous panels. Journal of Econometrics, 115(1), 53-74. Islam, N. (1995), Growth empirics: A panel data approach. The Quarterly Journal of Economics, 110(4), 1127-1170. Islam, N. (2003), What have we learnt from the convergence debate? Journal of Economic Surveys, 17(3), 309-362. Khan, M.S., Kumar, M.S. (1993), Public and Private Investment and the Convergence of Per Capita Incomes in Developing Countries, IMF Working Paper No, 93/51. Knight, M., Loayza, N., Villanueva, D. (1993), Testing the neoclassical theory of economic growth: A panel data approach. Staff Papers- International Monetary Fund, 40(3), 512-541. Kormendi, R.C., Meguire, P.G. (1985), Macroeconomic determinants of growth: Cross-country evidence. Journal of Monetary Economics, 16(2), 141-163. Lei, C.K., Tam, P.S. (2010), A panel data approach to the income convergence among Mainland China, Hong Kong and Macao. Table 2: Unite root test results Country t-statistic SURADF critical values ADF SURADF 0.01 0.05 0.10 Indonesia −4.356 [1]* −3.391[1]* −2.706 −1.932 −1.498 Malaysia −2.232 [1] −1.075[1] −2.935 −2.154 −1.724 Philippines −0.358 [1] −3.477[1]* −2.796 −2.027 −1.605 Singapore −0.890 [1] −1.560[1] −2.903 −2.118 −1.691 Thailand −2.316 [1] −1.831[1]** −2.912 −2.197 −1.786 * and ** indicate rejection of the null hypothesis that real per capita GDP of a particular economy is not converging to the group average at 5% and 10% levels of significance, respectively. Numbers in brackets represent the lags included to ensure that serial correlation is removed. The critical values for ADF test statistic are: −4.143, for 1% level of significance; −3.497, for 5% level of significance; and−3.178, for 10% level of significance. ADF test was carried out including both intercept and trend. Critical values for SURADF are generated by Monte Carlo simulation using 10,000 replications based on the underlying dataset. GDP: Gross domestic product, ADF: Augmented Dickey-Fuller, SURADF: Seemingly Unrelated Regression Augmented Dickey-Fuller Table 3: System GMM estimates on the relationship between per capita income convergence and gross domestic investment Regressors Model 1 Model 2 Model 3 LGDIRijt 0.174 [0.090]** 0.341 [0.099]* 0.360 [0.103]* LOPNijt 0.031 [0.005]* 0.033 [0.010]* LGFCEijt 0.685 [0.420] CONSTANT −10.792 [2.240]* −16.2397 [4.452]* −28.606 [8.636]* * and ** indicates that a parameter estimate is statically significant at 5% and 10% levels of significance, respectively. Numbers in brackets are standard errors. GMM: Generalized method of moments Hassan and Murtala: The Role of Domestic Private Investment in Promoting Per Capita Income Convergence amongst Association of Southeast Asian Nations-5 Economies International Journal of Economics and Financial Issues | Vol 6 • Issue 4 • 2016 1303 Journal of the Asia Pacific Economy, 15(4), 420-435. Levin, A., Lin, C.F., Chu, C.S.J. (2002), Unit root tests in panel data: Asymptotic and finite-sample properties. Journal of Econometrics, 108(1), 1-24. McCoskey, S.K. (2002), Convergence in Sub-Saharan Africa: A nonstationary panel data approach. Applied Economics, 34(7), 819-829. Nelson, C.R., Plosser, C.R. (1982), Trends and random walks in macroeconomic time series: Some evidence and implications. Journal of Monetary Economics, 10(2), 139-162. Rodríguez-Pose, A., Psycharis, Y., Tselios, V. (2012), Public investment and regional growth and convergence: Evidence from Greece. Papers in Regional Science, 91(3), 543-568. Solow, R.M. (1956), A contribution to the theory of economic growth. The Quarterly Journal of Economics, 70(1), 65-94. Weeks, M., Yao, J.Y. (2003), Provincial conditional income convergence in China, 1953–1997: A panel data approach. Econometric Reviews, 22(1), 59-77.