Reading Response: “Inequality does cause underdevelopment: Insights from a new instrument”

Big Question: The effect of income inequality on economic development has long been a topic of debate within growth economics. Early growth literature, published by influential economists such as Lewis and Kaldor, argued that high levels of inequality encouraged economic development by funneling income to high-saving capitalists. More recent theoretical and empirical studies, however, reversed this prediction and suggested that inequality actually harms growth through its effect on political economy and restrictions on human capital accumulation. Such studies have brought about more research on the relationship between growth and inequality, which have produced conflicting results. In this paper, William Easterly builds on Engermann and Sokoloff empirically-based research to examine the extent to which agricultural endowment predicts inequality and inequality predicts growth.

Little Question: At the outset of the paper, Easterly makes a clear distinction between structural and market inequality. Structural inequality reflects non-market mechanisms and historical events such as slavery, colonization, and conquest that resulted in a ruling elite. Market inequality is the result of the uneven success of market sources across the global economy. Easterly claims that market inequality has ambiguous effects on growth- inequality obviously results in some negative externalities, but the elimination of market forces would negate any economic incentives for those within an economy. Structural inequality, however, is unambiguously bad for economic development.

Easterly’s paper follows an empirical strategy based on the work of economic historians Engermann and Sokoloff. Engermann and Sokoloff, over the course of several papers, suggest that agricultural factors endowments are the central determinants of structural inequality, which in turn leads to underdevelopment. Easterly uses this framework- using the suitability of land for wheat growth as opposed to sugar growth- to build an econometric model in which he examines the effects of inequality, instrumented through agricultural endowments, on institutions, human capital, and economic development.

Relation to Class Discussion: In class, we have discussed Galor and Zeira’s model, which examines market frictions, inequality, and the relationship between finance and economic growth. Through this model, we have gained insight into how inequality affects the accumulation of human capital, as well as the manner in which frictions within financial markets exacerbate economic inequality. Easterly’s model informs our understanding of the foundations of inequality (or structural inequality) and its expands our discussion to include the effects of inequality on not only human capital accumulation, but also institutions and economic development.

Methods: Easterly builds on the work of Engermann and Sokoloff and uses the suitability of land for wheat or sugarcane production as a predictor for inequality. The production of wheat, Easterly suggests, promotes small family farms and a strong middle class. Sugarcane production, however, promoted economies of scale and the use of slave labor, which typically resulted in high levels of inequality. Easterly uses a WIDER dataset from 2000, to construct a regression equation with several different measures of income and two measures- Gini coefficient and wealth distribution- of inequality. Easterly uses data from the Food and Agriculture Organization to measure crop endowments. Wheat-sugar ratio is the independent variable within the regression, which is strongly predictive of the historical number of family farms in a certain location. The proportion of family farms within a given area serves as a proxy for historical inequality; the higher proportions predicting less inequality.

Results: Using the methodology described above, Easterly finds that a one standard deviation increase in the Gini coefficient reduces income by 1.1 standard deviations, institutional quality by 1.0 standard deviations, and schooling by 1.3 standard deviations. The results also indicate that inequality hinders economic development and growth. Easterly also checks for three possible omitted variables: ethnic fractionalization, legal origins, and tropical location. All three variables are possible alternatives to the inequality hypothesis. Ethnic fractionalization and Socialist legal origins yield significant results when controlling for inequality. Easterly’s results confirm Engermann and Sokoloff’s claims that institutions and schooling are the mechanisms by which inequality hinders development, while also finding inequality to be the underlying cause acting through these mechanisms.