earticle

영어

Rising income inequality witnessed in most countries especially in advanced economies since the early 1990s has drawn attention of many scholars who have presented several possible explanations for this phenomenon. While FDI has been the fastest growing economic activity around the world, most of the findings in the literature of FDI focus on the relationship of complementarity between FDI and economic growth and disregard how FDI flows affect income inequality. Especially it remains unclear whether widening income inequality in the advanced economies is a consequence of massive increase in cross border M&A FDI. This paper investigates how the effect of FDI and cross border M&A FDI on income inequality within the host country may depend on the level of human capital of that country. To formalize this mechanism this paper extends the theoretical framework of Aghion et al. (2002) where MNEs have advanced technologies and management know-hows, offering them productivity advantage over domestic firms in the host country. In their model MNEs can be regarded as vehicles for adopting and disseminating advanced technologies in the host economy. Based on this idea we try to build a model to make theoretical predictions concerning the effects of FDI, especially that of cross border M&A FDI on income inequality in advanced host countries. Domestic firms learn by imitating and adopting advanced technology, which leads to higher demand for skilled labor and changes in relative wage structure. This theoretical considerations lead us to proposes hypothesis: If FDI increase, income inequality of host country with higher level of human capital widens, whereas income inequality of host country with lower level of human capital narrows. And such a logic also holds for cross border M&A FDI inflows. This paper uses a large cross-section and panel of 119 developed and developing countries for a recent period 1990～2010 to analyse how the causal relationship between FDI and income inequality is affected by the level of human capital in the host country. The OLS estimation results show that the effect of FDI and cross border M&A FDI is statistically insignificant; these OLS results are biased because of an endogeneity problem associated with FDI. To tackle endogeneity problem, we conduct IV estimations using ‘remoteness’ and ‘internal distance’ as instrumental variables for FDI. The results of the F-tests and of Hansen-t test of over-identifying restrictions imply that these instruments are valid. The IV estimation results indicate that FDI and cross border M&A flows have a significantly negative impact on inequality in host country group with higher level of human capital even when controlling for various explanatory variables. To complement cross-country analysis this paper performs the dynamic panel data analysis by using the system GMM estimators. The system GMM estimation enables us to control for unobserved country-specific effects and to tackle the endogeneity problems by using the internal lagged variables as the instrumental variables. Unlike cross-country analysis, we do not divide the entire sample into countries with higher or lower level of human capital because performing system GMM needs as many countries as possible. Instead, we add interaction term between inflow of FDI and human capital to test our hypothesis. By using and estimating interaction term the partial effect of FDI on inequality, can be estimated. Results show that is negative and is positive. Although the coefficient of FDI, is insignificant, the null hypothesis of is rejected by the F tests at the conventional significance level. Because the coefficients of FDI are negative and significant and the interaction terms of human capital with FDI are positive and significant, FDI inflows widens inequality in a country with higher level of human capital, whereas it narrows inequality in a country with lower level of human capital. To confirm the validity of the set of instrumental variables, we perform the Hansen tests for overidentifying restrictions, which do not reject the orthogonality conditions at the conventional significance level. With appropriate and valid instrumental variables the estimation results are consistent with our hypothesis: the FDI inflow widens inequality in a country with higher level of human capital, whereas it narrows inequality in a country with lower level of human capital. The host country of higher level of human capital that desire to decrease inequality resulting from FDI or cross border M&A should consider implementing social policies to mitigate the widening inequality if they are oriented to sustain economic growth and job creation with equitable income distribution.

한국어

외국인 직접투자의 유입이 호스트 국가의 소득분배에 미치는 영향에 대한 대부분의 연구는 직접투자의 유형을 구분하지 않고 있다. 본 논문은 외국인 직접투자와 인수합병형 직접투자로 구분해서 이들이 호스트 국가의 소득분배에 미치는 영향을 분석하고, 소득분배에 미치는 영향이 호스트 국가의 인적자본 수준에 의해 좌우되는지의 여부를 검토한다. 인적자본 수준은 경제성장에 영향을 미치는 조건(Borensztein et al., 1995)이고, 동시에 숙련/미숙련 상대수요 및 상대임금수준에 영향을 줌으로서 소득불평등을 좌우하는 조건이기도 하다. 이러한 아이디어를 Aghion et al.(2002)의 모형을 수정하여 가설을 도출한다. 가설을 실증분석하기 위해 1990～2010년 기간 동안 119개 국가의 국가별 횡단면 자료를 이용한 OLS, IV추정, 그리고 패널 데이터를 이용한 system GMM 추정한다. 추정 결과를 요약하면 다음과 같다. 첫째, 인수합병 직접투자는 전체 외국인 직접투자와 동일한 방향으로 소득불평등에 영향을 미친다. 둘째, 외국인 직접투자와 인수합병형 직접투자는 인적자본 수준이 높은 국가그룹에서는 소득불평등을 악화시키고, 인적자본수준이 중하위 국가그룹에서는 소득불평등을 완화시켜준다. 셋째, 패널 추정 결과는 횡단면 분석결과를 지지해 주고 있다. 인적자본 수준이 높은 선진국의 경우 외국인 직접투자를 통해 높은 수준의 기술이 인적자본을 통해 소득불평등에 부정적 영향을 미칠 수 있으므로, 경제성장과 일자리 창출을 위한 외국인 직접투자 유치정책을 추진함과 동시에 소득불평등 악화 효과를 감소시키는 사회 정책도 동시에 모색되어야 할 것이다.