A CAUSALITY BETWEEN CAPITAL FLIGHT AND ECONOMIC GROWTH: A CASE STUDY INDONESIA

A CAUSALITY BETWEEN CAPITAL FLIGHT AND ECONOMIC GROWTH: A CASE STUDY INDONESIA Setyo Tri Wahyudi Department of Economics-Brawijaya University INDONESIA setyo.tw@ub.ac.id; setyo_triwahyudi@yahoo.com Ghozali Maski Department of Economics-Brawijaya University INDONESIA ghozalimaski@ymail.com Abstract Foreign capital is a source of financing for investment and consumption, as well as to strengthen the country's foreign exchange reserves. The phenomenon indicates that in Indonesia the flow of foreign capital in the form of Foreign Direct Investment (FDI) was dominated the composition of foreign capital since 1990, which was largely dominated by the flow of portfolio capital in the form of bonds, stocks, equity, and other short term instruments compared with the flow of capital in form of FDI itself. These conditions have indicated the presence of capital flight in Indonesia. The purpose of the study is to describe and identify the causal relationship between capital flight and Indonesia's economic growth over the period 2000-2009. Although the results of study was indicated that economic growth in Indonesia moves towards positive economic growth rate is quite high, but on the other hand, Indonesia has problems enough attention, which is still a high level of capital flight out of the country. Furthermore, the causality test results show that the capital flight have impact on economic growth and not vice versa. Keywords: Capital flight, Economic growth. JEL classification: F32, F43 1

BACKGROUND The economic crisis that was hit southeast Asia in 1997-1998 has brought a significant influence especially on the economic structure in the region. For Indonesia, the impact of the economic crisis has resulted in fundamental changes to the economy. The impact can be seen from several aspects such as the fall of the Rupiah up to a level of Rp.17.000 per US dollar, high inflation, the number of banks that had to be closed down, many workers is "laid off", and various other problems. This suggests that the economic structure of Indonesia is still very vulnerable to protect changes in external factors. Recently, although the current macroeconomic conditions of Indonesia had steadily improved, but the shadow of economic crisis that had devastated the economic structure still continue to haunt. Moreover, the facts have shown that the process of economic recovery in post-crisis in Indonesia as very slow compared to the neighboring country which was also experiencing a similar crisis, such as Malaysia and Thailand. Therefore, it is not surprising if confidence of the investor to invest in Indonesia is still very low, even be said that Indonesia is no longer a country that is safe as an investment destination. Further, the survey regarding investment climate of Indonesia that has been done by World Bank and International Financial Corporation was shown that the position of Indonesia is the lowest when compared with other countries in Southeast Asia. In other hand, although the survey from Doing Business 2010 which show that Indonesia has raised the position of the order of 129 to 122, but the position is still a long way when compared with neighboring countries like Singapore which are able to maintain a great image, perched in the first rank. Meanwhile, Thailand is ranked at position 12, Malaysia in position 23, Vietnam ranked 93, and Brunei Darussalam ordered at 96, while, Indonesia's position just ahead of the Philippines which are in the order of 144 and Laos in position 167. The results of this ranking indicated that, the ease of investing in Indonesia is very low; further, this also means that Indonesia is not an attractive investment destination in Southeast Asia. The implementation of free foreign exchange system that began in 1967 was support the Indonesia's financial system to be integrated with world financial system. As a result is more increasingly open the flow of foreign capital in term of free exit and entry. If the capital outflows very high, the phenomenon indicates that there has been a capital flight. Generally, the phenomenon of foreign capital flight is usually indicated by the type of short-term investments such as investment in portfolio. Investments in portfolio could be affecting the domestic financial market with transaction forms such as equity and cash securities. For developing countries like Indonesia, the flow of foreign capital is important as the source of financing for investment and also consumption, further to strengthen the country's foreign exchange reserves. The purpose of this study was to (1) describe the capital flight in Indonesia during the period 2000-2009, and (2) identify the causality between capital flight and economic growth. 2

Concept of Capital Flight The main problem when describing the capital flight is that there is no clear consensus on the definition of the phenomenon. Although, it was agreed that the phenomenon is a response to both economic and political uncertainty, but based on the literatures, no further consensus regarding this concepts. Several studies have tried to define and declare that the capital flight with capital outflows, while others argue that it is only a part of all outflows. Therefore there are two categories of definitions of capital flight: 1) those who distinguish between capital flight and 2) those who are no distinguished between capital flight. Schneider (2001) gives two main characteristics of the concept of capital flight are: Capital flight - A response to the handling of domestic capital Schneider (2001) argues that capital movements can occur in response to a perceived change and uncertainty are not always captured by portfolio theory, as summarized as follows: Capital flight is part of the deployment of international asset or portfolio adjustments in response to an unusual decline in the perceived risk/reward profile associated with the assets located in certain countries, faced with the conflict between asset holders and the government. Two-way capital flows due to different effects experienced by domestic and foreign investors, arising due to several factors such as information asymmetry, risk, return, and the impact of political risk. Capital Flight - An illegal transaction Capital flight is often defined as an illegal transaction that occurs when traders get foreign capital by way of falsified trade documents. Capital flight can be done in a way that intentionally makes transactions without invoice exports and imports. Ways thus is easily detected by comparing the trading partner country statistics. Capital flight is defined as occur only when the foreign exchange traders illegally transfer funds out of the country in hopes to avoid the domestic market. A serious drawback of the definition is that for the calculation of the transfer mechanism may include income that is stored outside the country to avoid quotas and tariffs, as well as income from criminal activity and smuggling that does not have to be included in the concept. However, the concept of capital flight which assumes that this is an illegal transaction is a good indicator for the effort to prevent such activities. The question that arises is whether we should limit the occurrence of capital flight is only for illegal transactions? Does that not mean that the use of the concept, we will underestimate the actual capital flight? In fact, there is evidence that the illegal flight of capital as a transaction should be considered as part of the total capital flight and should be included in the calculation of capital flight despite using different definitions. 3

Causality between Capital Flight and Economic Growth A wide range literature regarding the causality between capital flight and economic growth in around the world. This section will review the relevant empirical studies linking capital flight and economic growth. Li and Liu (2005), on the other hand, uses the panel data of 84 countries to investigate the influence of FDI on growth. The study found a significant relationship between FDI and economic growth. Additionally, a stronger relationship was extracted when FDI interacted with human capital. This is because stronger human capital poses better absorptive capacities due to the complementary nature of the FDI and the human capital, most importantly for the developing countries. In contrast, Akinlo (2004) investigated the impact of FDI on economic growth in Nigeria using the ECM showed an insignificant negative influence of FDI on growth. The author further argued that extractive FDI might not extract significant impact on growth compared to the FDI in manufacturing sector. Additionally, FDI may influence growth negatively once there is an evidence of the foreign investors transferring profits, or other investment gains to their home country. Kadochnikov (2005) analyzes the determinants and effects of capital flight on the Russian economy by using an institutional approach that is rarely raised in the study. The New Institutional Economics approach as basis for analyzing the impact of capital flight. To support his analysis, he used a modification of non-granger Causality test to determine whether capital flight dynamics have a causal effect on interest rates, and vice versa. The study concluded that the handling of capital flight in Russia do not require a strict policy of capital, which in fact it will worsen the condition, because of some policy tightening will not increase investment opportunities, reduce the quality of project financing, the accumulation of bad debt, and may be cause crisis. Thus, the policy of restrictions on capital flight in the case is not part of the pro-growth policies. Instead, it takes only an increase in institutional functions where it can improve the investment process and encourage investment activity, the impact of capital flight would likely decline. Ayadi (2008) investigates the linear determinants of capital flight in Nigeria utilizing the ordinary least squares (OLS) and the error correction method (ECM). The study found that the validity of the portfolio theory which postulates how risk-averse investors can build portfolios in order to optimize or maximize expected returns given a level of market risk. Further, the study confirmed that capital flight is caused by the interest rates deferential both in the short and in the long run. In addition, Ayadi found that exchange rate depreciation significantly increases capital flight in Nigeria. Output growth which measures the domestic opportunity cost of flight in Nigeria is negative and significant in the short-run indicating that non performance of domestic resources can trigger capital flight. Recent study by Ogundipe and Aworinde (2011) explored the causality between Foreign Direct Investment and economic growth in Nigeria using Granger causality. 4

The study, using annual data covering the period between 1970-1985, 1986-2007 and 1970-2007, showed causality relationship from economic growth (GDP) to FDI in the prederegulation era, which implies that there is causality relationship from economic growth to FDI. In the post-deregulation era there is no casual relationship between GDP and FDI. However, in the whole period 1970-2007 economic growth (GDP) is the cause of FDI in the pre-deregulation era, which implies that there is causality relationship from economic growth to FDI. In other words, there is a one-way relationship between FDI and economic growth. Other noteworthy studies examining the influences of FDI employs the Granger causality test (Knoldy, 1995; Nair Reichert and Weinhold, 2001) but the results vary according to country, method used and time frame under study. METHODOLOGY Data and Sample The data used in the study is secondary quarterly time series from the first quarter 2000 to third quarter 2009. The data collected from International Financial Statistics, World Development Indicators, financial statistical data issued by Bank Indonesia and the Center of Statistical Office (BPS). Model Specifications In this study Granger causality test will be used in order to test the hypotheses regarding the presence and the direction of causality between Capital Flight and Economic Growth. The models suggested for this test are as follows:...(1)...(2) The methods and procedures testing is Causality Test Model. The test procedure is as follows: (1) Stationarity Test Stationarity test used to see whether the observed data are stationary or not. Although it was just a natural test of Granger causality test, but if the results show that the observed data is stationary, this will improve the accuracy of the analysis of Granger causality. As a consequence of the use of time series data, the stationarity test will give 5

a profit, this is because the data analysis has been to eliminate the variables are nonstationary in the model. This means that the outcome would avoid biased estimates of standard error. If the estimate was biased, it could lead to the conventional criteria used to justify the causality between two variables becomes invalid. This means that estimation using a variable that has the data non-stationary (unit root) can result in incorrect conclusions because of the regression coefficient estimator is inefficient. In order to apply Granger causality test, the series that belong to variables should be stationary. Therefore; it is necessary to make test for unit roots to examine whether the series for these two variables are stationary or not. Macroeconomic time series are usually not stationary. Such series are made stationary by calculating logarithms or taking first or second differences. There are many tests used to determine stationary. In this study, the stationary of the variables will be tested by using Augmented Dickey- Fuller unit root test. Technically, procedures in the ADF test is based on MacKinnon critical values instead of t-test, the t-ratio is compared with critical value of t-statistics in ADF table in order to determine the presence or absence of unit roots. If the hypothesis is accepted, the variable was not stationary, and is necessary to test the degree of integration. Test the degree of integration is intended to look at the degree or order difference to how the observed data be stationary. (2) Granger Causality Test The direction of causality determines the direction of the relationship among variables and Granger causality test has three different directions for these purposes: a) One way causality: In a single equation model, Y is the dependent variable and X independent. Here, there is a causality relationship from X towards Y Independent variable is the cause and causes a one-way effect on dependent variable, which shows the presence of one-way causality and the relationship is determined as Y on X b) Two-way causality: There can be a reciprocal effect between variables. c) Lack of Causality: There is no relationship among variables, therefore no causality. To find out the possible existence of various forms of causality as mentioned in equations (1) and (2), the F-test performed for each regression model. Null hypothesis is: Test F-test using the formula: 6

where: SSRr = Sum of squared residuals for the restricted equation; SSRu = Sum of squared residuals for the unrestricted equation; w = number of regressors in the equation n = Number of observations; k = number of regressors in the equation. Based on the Granger causality test model, the hypotheses to be tested are: H0: Economic growth does not affect the capital flight Hi: Economic growth affects capital flight H0: Capital flight does not affect economic growth Hi: Capital flight affecting economic growth Here, H0 hypothesis are tested by comparing the value obtained in this test with the values calculated by Dickey-Fuller. Null Hypothesis shows that series is not stationary and has a unit root (Ho: γ=0), and alternative hypothesis shows that series is stationary. If the absolute value of calculated statistics is higher than the absolute value of critical values, we cannot reject the hypothesis which shows that series is stationary. However, if this value is lower than critical value, time series is not stationary (Gujarati, 2004). RESULTS AND ANALYSIS Overview of Indonesia s Economic Growth An indicator of a country s economic growth is represented by the process of the production capacity of an economy that embodied in the form of increased national income. The development of economic growth in Indonesia during the period 2007-2009 at current prices (ADHB) and at constant prices 2000 (ADHK) are shown in Table 1. In 2009, Indonesia's economy was grown by 4.5 percent compared to 2008. The value of GDP at constant prices (ADHK) in 2009 reached Rp2,177 billion, while in 2008 and 2007 is Rp2,082 trillion and Rp.1,964 trillion, respectively. When viewed by current prices (ADHB), GDP in 2009 rose by Rp662 billion, from Rp4,951.4 trillion in 2008 amounted to Rp5,613.4 trillion in 2009. During 2009, all economic sectors experiencing growth. The highest growth occurred in transport and communications sector which was reached 15.5 percent, followed by sector electricity, gas and clean water (13.8 percent), construction sector (7.1 percent), services sector (6.4 percent), the financial, real estate, and company services (5.0 percent), mining and quarrying (4.4 percent), agricultural (4.1 percent, and trade, hotel and restaurant (1.1 percent). GDP growth in oil and gas in 2009 reached 4.9 percent. 7

No Sectors Table 1: Indonesia s GDP, 2007-2009 Current Prices (Billion Rupiah) Constant Prices 2000 (Billion Rupiah) 2007 2008 2009 2007 2008 2009 1 Agriculture 531.9 716.1 858.3 271.5 284.6 296.4 2 Mining 440.6 540.6 591.5 171.3 172.4 180 3 Manufacture 1,068.7 1,380.7 1,480.9 538.1 557.8 569.5 4 Electrical, Gas, and Clean Water 34.7 40.9 46.8 13.5 15 17.1 5 Construction 305 419.6 555 121.8 131 140.2 6 Trade, Hotel, and Restaurant 592.3 691.5 750.6 340.4 363.8 367.9 7 Communications and 264.3 312.2 352.4 142.3 165.9 191.7 Transportations 8 Financial, Real Estate, and 305.2 368.1 404.1 183.7 198.8 208.8 Company services 9 Services 398.2 481.7 573.8 181.7 193 205.4 GDP 3940.9 4951.4 5613.4 1964.3 2082.3 2177 GDP Non-Migas 3,534.4 4,427.2 5,146.5 1,821.8 1,939.5 2.035.1 Source: BPS, 2010 Indonesia's GDP by nine sectors of economic activities during 2007-2009 period shows that the largest contributing sector is manufacturing, followed by agriculture and services sectors. While the sector with smallest contribution is electricity, gas, and water. The contribution of agriculture sector continued to decline, it demonstrates the ongoing structural transformation in Indonesia (Table 2). Compared with 2007 and 2008, in 2009 there was an increase in some sectors except: Industry Sector, Trade Sector, Hotel and Restaurant, Mining and Quarrying, and the Financial Sector, Real Estate and Business Services. The Role of Agriculture Sector increased from 14.5 percent to 15.3 percent, services sector from 9.7 percent to 10.2 percent, construction sector from 8.5 percent to 9.9 percent, while the Transport and Communications Sector and the Sector Electricity, Gas and Water respectively provide the same role from 2008 that is equal to 6.3 percent and 0.8 percent. While the Manufacturing sector fell from 27.9 percent to 26.4 percent, Trade Sector, Hotel and Restaurant down from 14.0 percent to 13.4 percent, Mining and Quarrying sector decreased from 10.9 percent to 10.5 percent, and the Financial Sector, Real Estate and Business Services dropped from 7.4 percent to 7.2 percent. Furthermore, if viewed in total, the role of oil and gas GDP rose from 89.4 percent in 2008 to 91.7 percent in 2009. 8

Table 2: The Structure of Indonesia s GDP (%) No. Sectors 2007 2008 2009 1 Agriculture 13.7 14.5 15.3 2 Mining 11.2 10.9 10.5 3 Manufacture 27 27.9 26.4 4 Electrical, Gas, and Clean Water 0.9 0.8 0.8 5 Construction 7.7 8.5 9.9 6 Trade, Hotel, and Restaurant 15 14 13.4 7 Communications and Transportations 6.7 6.3 6.3 8 Financial, Real Estate, and Company services 7.7 7.4 7.2 9 Services 10.1 9.7 10.2 GDP 100 100 100 GDP Non-Migas 89.5 89.4 91.7 Source: BPS, 2010 Causality Test Results This section discusses the results of causality testing between capital flight and economic growth. Causality testing stages are starting with a unit root test that aims to determine the stationarity of variables in the model. If the stationarity test results concluded that the variables in the study were stationary at the same degree, and then can be continue to Granger Causality test. (1) The results of stationarity tests Testing stationarity aims to test the stationarity of data due to the use of time series data in the research. This must be done in order to avoid the problem of model bias estimation or spurious model. In this study, the unit root test conducted by using Augmented Dickey-Fuller test (ADF). The criteria is when the ADF test statistic is smaller than the Mackinnon critical values, it is said that the variable are stationary. Meanwhile, if the test results concluded that the data are not stationary, then the differentiation procedure done, which is further analyzed to obtain data that are stationary. The results of the testing unit roots using the ADF test are shown below: Table 3: Stationarity Test Results Variable ADF statistics Degree CF 1.034664 Level GDP -1.502480 Level Note: Mackinnon critical values are -4.420595 (1% level); -3.259808 (5% level); -2.771129 (10% level). 9

As the presented in Table 3, it is known that the ADF statistic for both variables (CF and GDP) is smaller than the Mackinnon critical values then it can be concluded that the variables stationary on the same degree of integration of 0 or the I(0). That is, all variables used in this study at degree level stationary are significant. (2) The results of Granger Causality Test Granger causality test is used to look at the relationship between two variables statistically, the capital flight and economic growth in Indonesia. Through this test can be seen whether the two variables are unidirectional relationship, two-way (mutual influence), or have absolutely no linkage (not affect each other). Granger causality test results can be seen in Table 4 below: Table 4: Granger Causality Test Results Pairwise Granger Causality Tests Sample: 2000 2009 Lags: 2 Null Hypothesis: Obs F-Statistic Prob. GDP does not Granger Cause CF 8 0.78936 0.5304 CF does not Granger Cause GDP 17.2931 0.0225 To find out the relations between the two variables, hypothesis testing is done as follows: a) H0: Economic growth (GDP) does not affect the capital flight (CF) Hi: Economic growth (GDP) affects the capital flight (CF) b) H0: Capital flight (CF) does not affect economic growth (GDP) Hi: Capital flight (CF) affects economic growth (GDP) If the probability of the hypothesis is smaller than the errors then both decided to reject H0, so that interpretation is economic growth and capital flight interplay (Causality). Conversely, if only one hypothesis H0 is rejected, then the relationship between economic growth and capital flight is only a one-way causal relationship. Based on the test results for both hypotheses, obtained the following results: a) H0: Economic growth (GDP) does not affect the capital flight (CF) Hi: Economic growth (GDP) affects the capital flight (CF) 10

Granger test indicates the probability of F-statistic is 0.5304. Probability value is greater than the tolerable error (α = 5%). Means to accept the null hypothesis (Ho), namely economic growth (GDP) does not affect the flight of capital (CF). b) H0: Capital flight (CF) does not affect economic growth (GDP) Hi: Capital flight (CF) affects economic growth (GDP) Granger test indicates the probability of F-statistic is 0.0225. Probability value is less than the tolerable error (α = 5%). It means to reject the null hypothesis (Ho), the capital flight (CF) effects on economic growth (GDP). Thus, it can be concluded that the causal relationship between capital flight and economic growth in Indonesia did not show a two-way relationship, but only one-way relationship, which influence the direction of capital flight (CF) effects on economic growth (GDP) and not vice versa. CONCLUSION The study has shown the causality between capital flight and economic growth in Indonesia. During 2000 to 2009, Indonesia's macroeconomic show that Indonesia s economic growth moves towards positive rate which is quite high. On the other hand, Indonesia is still high dependence on foreign capital loans as one of the engine to promote economic growth. The causality test concluded that capital flight have impact on economic growth and not vice versa. It indicates that Indonesia is also experiencing problems enough attention, which is still a high level of capital flight out of the country. Further, Indonesia is still very vulnerable to external shocks, particularly from short-term foreign loans. As a result, Indonesia's economic growth is interrupted. Several suggestions and recommendations regarding the study are as follows: 1) the need for appropriate policies that were taken and run the government of Indonesia to prevent capital flight. Improvement of the investment and licensing procedures in Indonesia should be reexamined, so that capital flight can be prevented so as not to interfere with the process of economic growth. 2) Necessary efforts to promote economic growth in Indonesia. Although Indonesia's economic growth performance has been pretty good, but needs to be improved and maintained so that growth occurs truly reflect the level of welfare of society as a whole. Several attempts to do to improve economic growth is to maintain the stability of inflation and exchange rates, increased domestic production and encourage export activities. And 3) reduce dependence on foreign aid or financing. The higher foreign debt, the high economic growth will never be enjoyed by the public, otherwise used for debt repayments and interest from time to time due to the increasingly burdensome in the long run, fluctuations in inflation rates, as well as the exchange rate is difficult to control. 11

References Akinlo, A.E., (2004). Foreign Direct Investment and Growth in Nigeria: An Empirical Investigation. Journal of Policy Modelling, Vol 26, pp 627-639. Ayadi, F.S. (2008). Econometrics analysis of capital flight in developing countries: a study of Nigeria. Paper presented at 8th Global Conference on Business & Economics. October, 18-19th, Italy. Bank Indonesia. (2011). Indonesia s Foreign Debts Statistics. Jakarta. Gujarati, D. (2004). Basic Econometrics. McGraw-Hill Company: New York. Kadochnikov, D.V. (2005). Economic Impact of Capital Flight from Russia and its Institutional Context: Why Capital Controls cannot be a Part of a Pro- Growth Policy. Working Paper, June. Knoldy, S. (1995). Causality Between Foreign Direct Investment and Spillover Efficiency. Applied Economics, 27(8), pp 745-749. Li, X. and Liu, X., (2005). Foreign Direct Investment and Economic Growth: An Increasingly Endogenous Relationship. World Development, vol 33(3), pp 393-407. Nair-Reichert, U., and Weinhold, D. (2001). Causality Tests for Cross-country Panels: A New Look at FDI and Economic Growth in Developing Countries. Oxford Bulletin of Economic and Statistics, 363(2), pp 153-171. Ogundipe, M.A., and Aworinde, O.B. (2011). An analysis of causality between Economic Growth and Foreign Direct Investment in pre- and postderegulated Nigerian economy (1970-2007). European Journal of Scientific Research, 53 (3), pp.317-325. Schneider, B. (2001). Measuring Capital Flight: Estimates and Interpretations. London: Overseas Developement Institute. (www.odi.org.uk/publications/working_papers/wp194_maintext.pdf) 12