Foreign Capital Inflows and Economic Growth in GCC Countries

This paper provides a literature review and an empirical study of the particular independent growth impact of various capital inflows in the GCC countries during the period 2005-2018. It augments the standard growth determinants with inflows of foreign direct investment (FDI), foreign portfolio investment (FPI) and other investments (OI) and estimates this equation with the LSDV estimator. Main results are that capital inflows as a whole don’t exert any effect on economic growth. However, when it is decomposed in different inflows kinds, other investment and FDI don’t affect growth while foreign portfolio investment has a significant negative effect.


INTRODUCTION
Recent turmoil in oil prices sheds the light on the importance of financial resource diversification for GCC countries and the need to reduce their dependence on oil and gas revenues by seeking alternative financing sources. A strategic priority was the attraction of foreign direct investment, portfolio investments and other investments. In fact, capital inflows reached, 18% of GDP in Oman, 21% in Kuwait in 2007 and9% of GDP in KSA andalmost 13% in Qatar in 2009. Meanwhile, in several emerging markets, massive capital inflows resulted in tremendous crises which posed questions on the ability of these flows to stimulate growth and about their real profitability. The controversial question was whether GCC countries did well or not by opening their capital accounts.
Proponents of financial liberalism vaunt the positive externalities of foreign capital inflows in term of risk diversification and consumption smoothing, Obstfeld (1994), in term of managerial and technological knowledge transfer, Grossman and Helpman (1991) and in term of financial system deepening through greater competition in the banking system and higher liquidity in the equity market, Levine (2001). The presence of foreign capital inflows provides additional capital to local saving and promotes capital accumulation thus increasing growth through knowledge spillover and market efficiency effect, Borensztein et al. (1998). These arguments were tested and verified empirically with Bailliu (2000), Quinn and Toyoda (2008), Klein and Olivei (2008) and Garcia and Santana (2004) who found a positive effect of net private capital, equity market liberalization and capital account openness on growth.
However, other economists such as (Krugman, 1979;McKinnon and Pill, 1997;Rodrik, 1998 andStiglitz, 2000) were skeptic and highlighted the negative consequences of capital inflows especially the financial crisis driven by the increase in systemic risk and sudden reversals when moral hazard problem is present. In fact, episodes of crisis in Latin America and Asian countries resulted in an exchange rate collapse, an asset prices fall and an overall financial instability. Through an empirical work Reinhart and Reinhart (2008) showed that the likelihood of financial and economic crisis is higher in emerging markets in the presence of foreign capital inflows. Moreover, Bussiere and Fratszcher (2008) concluded that the financial openness promotes growth in the short run but not in the medium and the long run.
The cleavage between these economists both theoretically and empirically raised the interest for new studies to focus on the specific effect of every kind of capital instead of financial integration as a whole and to test the hypothesis that these inflows aren't equal towards growth, some could be beneficial to country growth while others could have deleterious effects. This question will be examined for the GCC countries during the period 2005-2018 through the estimation of a standard growth equation augmented with various capital inflows using LSDV estimator. In fact, every kind of flow in percentage of GDP is introduced in the equation in order to evaluate the volume effect.
The paper is organized as follows: the next section will be devoted to a theoretical and empirical studies review on the effects of every kind of capital on growth. The third section presents the empirical study and the last one will conclude.

THEORETICAL ARGUMENTS AND LITERATURE REVIEW
Capital inflows are decomposed in the financial account as foreign direct investments 1 , foreign portfolio investments 2 and other investments 3 . In the first subsection theoretical arguments concerning their contribution to economic growth will be exposed while in the second section empirical studies involving these inflows will be detailed in a table.

Theoretical Arguments
According to academics and policy makers FDI is supposed to have positive effects on a host country's growth as it is a provider of capital, a creator of jobs and a supplier of foreign currency. It is also considered to increase the rate of technical progress through the contagion effect since it facilitates the diffusion of managerial skills and technical know-how (Findlay, 1978;Borensztein et al., 1998). Besides, it results in efficient productivity gains by increasing the competition in product and factor markets (De Mello, 1999;Markusen and Venables, 1999). In addition, FDI inflows enrich the government with taxes on FDI income, Feldstein (2000) and allow the country to maintain a steady level of consumption even in crisis period thanks to their resilience Sarno and Taylor (1999). However, institutional, human and financial threshold should be reached to expect to benefit from FDI.
1 Foreign Direct investment is an investment of 10% or more of a business undertaken by an entity resident in one economy in an enterprise resident in another economy. 2 Portfolio investment consists of equity (shares, stocks, being <10% ownership in an entity) and debt securities (bonds and notes, debentures, treasury bills, commercial and financial paper). 3 Other investment item is a residual item that includes all financial transactions not covered under direct investment, portfolio investment, financial derivatives or reserve assets. It mostly consists of cross border bank lending and we find also trade credits.
The Portfolio investment flows, by spurring the stock market efficiency through specialization, acquisition and information dissemination, improve firm control, reduce monitoring costs and result in investment increase (Jensen and Meckling, 1976). Specifically, equity portfolio investments through risk sharing help to finance innovation (King and Levine, 1993) and increase capital productivity by insuring against liquidity risk (Greenwood and Jovanovic, 1990). However, detractors of these flows advance that portfolio investment inflows are often short term and speculative driven by the attempts of foreign investors to diversify their risks and to have instant liquidity so they can be counterproductive as they hinder economic growth through externalities emanated both during the surges and sudden reversals.
Foreign bank lending can supplement low levels of domestic credit and enhance investment given the higher activity level and the greater capitalization of foreign banks, De Haas and Van Lelyveld (2004). Despite the possible positive effects, in the presence of moral hazard problems, the buildup of short term debt can cause crisis, Rodrik and Velasco (1999) since the ratio of short term debts to reserves is a robust predictor of financial crisis.
In addition, trade credit is an alternative way of finance in countries with less developed financial markets, Fisman and Love (2007). It allows firms unable to access to credit market to be financed through their suppliers which permits the production process continuity, Petersen and Rajan (1996).

Empirical Literature Review
Empirical studies cited above have explored the impact of capital flows in aggregate or have examined the effect of a specific kind of foreign capital inflows only, few studies have dealt with all types of capital inflows included in the same study we will report them in this table.

EMPIRICAL STUDY ON GCC COUNTRIES
The study specificity relative to other researches is that, it disaggregates capital inflows relative to their kind to show the effect of every one on growth in the GCC countries. Moreover, the study will use gross capital inflows that are more suitable while focusing on foreign investments. Indeed, after the global financial crises gross inflows are used instead of net flows 4 since the latter conceal the economy vulnerabilities. Gross inflows represent net sales of domestic financial instruments and assets to foreign residents; they increase when the economy incurs external liabilities and decrease when foreign investors are retrenching.
In the first subsection, some stylized facts about Gulf countries will be presented and the second subsection will deal with the econometric framework.

Statistical Facts
While looking to GCC 5 countries as a whole we can remark that capital inflows were increasing till 2007 where they reached a pick 4 The net capital flows are the difference between gross inflows and outflows 5 GCC countries here refers to all countries except UAE because of the lack of data Authors Sample and methodology Results Reisen and Soto (2001) GMM estimator 44 countries 1986-1997 Both FDI and portfolio equity investment exert significant impact on growth.
Private capital flows don't help but don't hurt economic growth in developing countries Durham (2003) 2SLS 88 countries 1977-2000 Foreign portfolio investments have no statistically significant effect on growth.
OFI has negative effect that is mitigated only when corruption rating is favorable and equity markets are large. Durham (2004) 2SLS Extreme bound analysis 80 countries 1979-1998 Lagged FDI and EFPI don t have direct unmitigated positive effects on growth and some data prove that their effects are contingent on the absorptive capacity of host countries.

Baharumshah and
Thanoon (2006 FDI has a positive effect on growth but portfolio investment and foreign debt have a negative one. The interaction between private capital inflows and stock market indicators has a positive effect on growth if domestic stock market reaches a certain threshold. Aizenman et al. (2013) OLS estimator 100 countries 1990-2010 Lagged FDI inflows are associated with higher growth while the association between lagged equity flows and growth is smaller and not as stable.
The association of growth and lagged short-term debt is nil before the crisis, and negative and large during the crisis. Baharumshah et al.
Threshold regression technique 80 developed, emerging and developing countries  Only countries achieving better financial market development beyond a certain threshold level can facilitate the positive growth effects of FDI and portfolio equity while debt portfolio inflows has no positive effects on growth. Okafor et al. (2015) OLS and Granger causality test Only FDI has positive effect on growth. Foreign portfolio investment and other investments are not significantly affecting growth.
of more than 15% of the region GDP. Then they witnessed a drastic drop during the crisis period where only FDI proved to be more resilient, portfolio investment collapsed and cross border-lending turned negative at about 5% of the region GDP.
Since 2013, flows started to climb steadily to attain 8% before contracting again in 2017 after the Qatar crisis and in 2018 these flows soared again.
The most noticeable facts are first, that these swings in capital inflows are mainly led by cross border lending and that FDI inflows are losing their importance in favor of portfolio investments. In fact, despite policy efforts to diminish administrative barriers, FDI inflows have stalled because of restrictions on foreign ownership and are still concentrated on green field investment in real estate, petroleum and chemical fields. However, the increase in portfolio investment is motivated by the recent inclusion of GCC countries in the global bond and equity indices.
Taking capital inflows to GCC countries masks great disparities relative to the flows attracted by every country. Figures 1 and 2 will detail inflows relative to GDP and in millions dollars for every country of our sample.  In order to evaluate the consequence of these inflows and their changing composition on the economic growth of the Gulf cooperation council countries an econometric study is deployed in the next section.

Empirical Study
The dynamic panel data model that arises in the convergence literature is written in this form: The panel estimation is a better method for capturing the relation between the dependent variable and the explanatory variables within a country; it allows capturing specific regional factors that affect the dependent variable, which are not captured by the explanatory variables, so it permits to reduce problems of the omitted variable bias.
However, since the dataset used is too small in both dimensions no other approach than OLS can be implemented, as it remains the only estimator with known small sample properties. The choice is either a pooled OLS approach, or a standard fixed-effects estimator which may introduce some bias but this may be preferable to omitting the fixed effects. That's why the LSDV estimator will be used to estimate this equation.
This choice is justified by Islam (2001) findings. In fact, while investigating the small sample properties of dynamic panel estimators for a growth convergence equation and comparing a set of ten different estimators, he surprisingly found that the LSDV estimator proves to be a relatively superior estimator that outperforms more sophisticated ones 6 .
In order to test the hypothesis that capital inflows have an effect on economic growth and that this effect differs with the nature of the capital inflow considered, the study proposes to decompose these flows and to add them to a set of controlling variables that proved to be relevant in the growth literature. The lagged per capita GDP is anticipated to have a negative sign because of the catching up effect.
The domestic saving is used here instead of the investment rate because investment is financed by domestic saving and foreign saving. In this paper, investment rate is replaced by its means of finance: on the one hand the saving rate and on the other hand foreign capital inflows. So we expect that the coefficient of the saving rate should have a positive sign.
Secondary school enrollment rate is used to catch the level of human capital and it is thought that its level is positively correlated with growth.
Inflation rate could have a positive or a negative sign as mentioned in the literature.
For different inflows relative to GDP, the positive effect of FDI inflows seems to make a consensus whereas portfolio investment and other investment could have either positive or negative sign. 6 Islam (2001) explained that the implementation of sophisticated estimators requires the use of estimated weighting matrices and their estimation pick up the sample variability and other noise and this results in their worse performance compared to their simpler counterparts Section 4 presents the results and try to give some explanations and recommendations. Table 1 presents the results of five estimations of the effect of capital inflows on economic growth taking into account traditional variables influencing growth.

RESULTS AND DISCUSSION
It can be noticed, as mentioned in the convergence literature, that the lagged real per capital GDP has a significant negative effect on real per capital growth rate through all equations confirming the catching-up effect.
Moreover, the ratio of domestic saving to GDP is significantly positive in the five equations traducing that an increase of 1 percent point in the share of saving in the GDP results in an increase of about 0.2% of the real per capital growth rate.
Similarly, the education indicator which is the secondary school enrollment rate proved to be positive and significant highlighting the importance of human capital building in the economic growth.
However, population growth rate is significant with the expected negative sign only in one equation.
Concerning inflation, it is positive at the level of 10% in two equations, this can be explained in the case of gulf countries that are not running at capacity and where inflation can help to increase production.
Regarding the interest variables, equation (1) reports that total capital inflows as a share of GDP has no effect on real per capita GDP growth as found by Soto (2003) who justifies that in developing countries these inflows don't help and don't hurt growth.
For GCC, this result isn't surprising since these countries don't receive capital inflows in a stable manner and inflows are not important relative to region GDP.
Equation (2) proves that FDI inflows have no significant effect on economic growth. This result is comparable to those of Durham (2004) and Baharumshah et al. (2015) and can be due to the fact that FDI inflows to GCC countries have been decreasing since 2008 and most of them are turned to resource seeking activities. This observation highlights the fact that in order to profit from FDI spillovers on growth, countries need to attain a certain threshold in institutional quality and financial market development.
Equation (3) confirms the significant negative effect of foreign portfolio investment as an increase of 1% in their share to GDP reduces real per capital GDP growth by 0.33%. This result is coherent with those of Choong et al. (2010) and De Vita and Kyaw (2009) and traduces that the short term nature of these inflows, their instability in the region and their ability to reverse with any turmoil make them deleterious for growth. Equation (4) investigates the effect of cross-border lending that proved to be insignificant in GCC countries. Usually these inflows when they are of long term nature contribute to growth in countries with well capitalized banking system but could be a source of trouble if they are of short term nature as it was the case in Latin America because of the maturity mismatch problem that they introduce.
Equation (5) introduces these three types of inflows and finds the same results as everyone introduced alone. So, all results are confirmed again: FDI and other investments are insignificant and portfolio investments have negative effect on economic growth.
Concerning dummy variables, we can notice that they are insignificant for Bahrain and Kuwait which means that there is no difference between these countries and Saudi Arabia in the constant term. However, Oman and Qatar dummies are significant meaning that Oman has a small constant term and Qatar a higher one relative to Saudi Arabia.

CONCLUSION
Studies that used capital account liberalization as a whole in order to examine the effect of international capital on growth neglect the difference that exists between different kinds of capital in influencing economic growth. Recently, some studies turned to decompose capital inflows and to study the effect of every kind on economic growth. This paper is part of this line of research. It uses data on GCC countries to estimate a standard growth equation augmented by capital inflows type through an LSDV estimator that proved to be appropriate in small sample.
Main results show that FDI have no significant effect on economic growth since most of them are resource seeking investments and they have been losing momentum in the region during the last years. In addition, cross-border lending proved to be insignificant also because of their high volatility and consequently their inability to sustain the domestic banking system. Besides, foreign portfolio investments have a significant negative effect because of their short term nature since they are only driven by diversification and speculation purposes which increase their ability to reverse with any turmoil.
Further investigations can be conducted to refine these results by introducing proxies of country absorption capacity in institutional and financial terms in order to detect thresholds that permit to the region to profit from these inflows or to reduce their negative effects.