Mind the Income Gap


So, I just recently finished my graduate studies thesis that required econometric analysis on an international development topic, and thought I would make a post about it and share it here. For those that are learning regression analysis through school, like me, might find this post fairly insightful. Especially, those who might be preparing to venture on a serious research study will find empirical economic research can be very daunting. Being given a four-month time frame to begin and fully complete a research study, I absolutely do understand the hair-pulling stress that comes with it! Though I must say, I am really proud of the end product.

Anyhow, I found it appropriate to write my Master’s Thesis on “Economic Openness and Its Effect on Country-Level Income Inequality” given the current political and economic environment. Income inequality has been a hot topic for many economists, namely noble-prize recipient Joseph Stieglitz, for quite some time. But a hot topic not just for economists, but it has also caught the public eye in the recent year or two. Unless you live in a cave devoid of connection to the outside world, the well-known nationwide Occupy Wall Street demonstrations (which I believe are technically still going on, just not as prominent and are not receiving the attention compared to a year ago) are built around this very issue. By no means is this a domestic (US) problem, but a more global issue, and is isolated to not only developed countries, but also the developing ones.

A popular general example is to look at China and their very notable and recent economic developments. With China’s extremely high GDP growth rates, hitting double digits in past years, a majority of that growth is taking place in urban areas. And what we see is the divergence of standard of living between urban and rural areas, which reflects labor wages growing disproportionally faster in the industry sector comparatively to the agricultural sector. Ultimately, widening the income gap.

Economists have argued that developing nations that open themselves up to the force of globalization increase their risk of widening household income gaps. Although the exact definition is debatable, it is generally agreed that the main catalysts driving the forces of globalization are economic interests among nations through international trade and financial integration into the global economy. With that said, in my study I define economic openness as the joint effect of a country’s trade openness and net foreign direct investment (FDI) flows.

My study implements the following fixed-effects model, along with instrumental variable techniques, using a monthly panel of country data for the period 1970 through 2006.

Screen Shot 2013-02-07 at 4.56.42 PM

I sampled 147 countries, where each country’s income distribution is measured by a Gini coefficient (dependent variable). The key independent variables measuring economic openness are: (i) a country’s trade openness, which is measured by the sum of a country’s imports and exports of goods and services as a share of its GDP emphasizing the share of a country’s economy that is devoted to international economic interaction, and (ii) a country’s net FDI inflows, where an attraction of FDI inflows strengthens a country’s connection to world trade networks. Additionally, a number of control variables are used since other significant factors may influence income inequality. These control variables include: GDP per capita, education, and a country’s sectorial share, which are incorporated into the model. To appropriately account for the Kuznets Hypothesis and its non-linear relationship on income inequality, the square of GDP per capita is also incorporated into the model. Empirical analysis also accounts for two separate developmental groups, developed and developing countries, as it is suggested and nearly unanimously determined by other studies, that there exists differentiating inequality consequences for different development levels.

However, the methodological specification has potential endogeneity problems of income influencing trade openness, where some studies have found that countries with high incomes are relatively high for motivations other than openness of trade, and cause them to be more open to trade. Frankel and Romer’s (1999) discussion of the gravity model demonstrates that a country’s geographic characteristics are powerful determinants of trade. Moreover, a country’s geographic characteristics are also not affected by a country’s income, government policies, or even other elements that influence income. Thus, I employ the use of a country’s geographic characteristics, and utilize them as suitable instrumental variables to address trade openness’ issue of endogeneity.

Screen Shot 2013-02-07 at 4.57.02 PM

All regressions employ robust standard errors to account for heteroscedasticity. Regressions 1 through 4 utilize standard OLS regression, while taking into account for potential regional differences. Regressions 5 through 8 utilize fixed-effects regression and instrumental variables to account for unobservable characteristics and potential endogeneity.

Beginning with Regression 1, general indications display a negative and significant relationship of trade openness on income inequality as well as a statistically significant positive relationship of a net gain in FDI inflow on income inequality across all countries. However, after accounting for time-invariant unobserved characteristics, Regression 5 reveals coefficient estimates lose relative significance, which encourages further investigation. Mentioned prior the model accounts for additional control variables. Comparing regression 2 and 6 suggest that endogeneity and unobservable time-invariant characteristics were placing an upward bias on our coefficient estimate of trade openness. After implementing fixed-effects regression and instrumental variables, in Regression 6, results show to a robust negative relationship of income inequality on trade openness, while also sustaining a significant and positive relationship of a net gain in FDI inflow and the adverse effect on a net loss in FDI inflow. In regression 7 and 8, suitable fixed-effects regressions are estimated separately for both developed and developing countries, shedding light to the finalized results. The negative sign for trade openness in both regressions are statistically significant. Furthermore, a net gain in FDI inflows do have a significant impact on income inequality at the 1 percent level of significance for both developed and developing countries, and a net loss in FDI inflows also have a significant adverse effect on income inequality.

Also, comparing the marginal effects of net FDI inflows between Regression 7 and 8, FDI has a much larger impact on inequality in developing countries than developed countries. This result may be because developing countries tend to have less regulated labor markets and a weaker presence of unionization, where Calderon and Chong (2009) find that labor market regulations have an adverse effect on income inequality. Thus, causing income inequality for developing countries to have a stronger positive relationship with a gain in net FDI inflows.

Now from a policy standpoint…

As seen from recent events this issue of income inequality has significant social impacts, which may lead to social unrest. Results show trade reduces inequality and typically trade is associated with transfer or welfare programs that compensate the losers of increased trade. From a policy standpoint, and as countries continue to experience growth, it is important that governments in developing countries have well established welfare programs, as it is a proven and common practice in developed countries dealing with the consequences of increased trade. In general however, theory shows that the benefits of trade tend to diffuse through a country’s economy, having a relative benefit on all households with an increase in trade. Results also show that FDI raises income inequality and intuition might suggest that recipients of FDI, countries experiencing a net gain in FDI inflows, should also similarly experience a negative relationship of income inequality. However, FDI tends to be concentrated in industries and the richest part of a country where wages are higher. So, this explanation weakens the relationship of FDI and low wages, which consequently increases income inequality. Thus, policymakers need to take measures to reduce the negative effects of FDI, perhaps by offering tax breaks or subsidies to industries or subsets of these industries without FDI to bridge the wage gap between those within-industries that are with and without FDI.


~ by Roger Sutton on February 7, 2013.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: