Do Markets Worsen Economic Inequalities? Kuznets in the Bush

• Published in: Human ecology : an interdisciplinary journal 2004-06, Vol.32 (3), p.339-364
• Main Authors: Godoy, Ricardo A., Gurven, Michael, Byron, Elizabeth, Reyes-García, Victoria, James Keough, Vadez, Vincent, Wilkie, David, Leonard, William R., Apaza, Lilian, Huanca, Tomás, Pérez, Eddy
• Format: Article
• Language: English
• Subjects: Agricultural Production; Bolivia; Coefficient of variation; Commercial production; Crop production; Economic Development; Economic value; Economics; Gini coefficient; Globalization; Households; Hunting and Gathering Societies; Income; Income distribution; Income Inequality; Indigenous peoples; Indigenous Populations; Market economies; Market Economy; Measurement; Native peoples; Quality of life; Rainforests; Rice; Rural areas; Socioeconomics; Tropical forests; Villages; Wealth
• ISSN: 0300-7839; 1572-9915
• DOI: 10.1023/B:HUEC.0000028085.47707.08

Integration into a market economy or economic development can erode the quality of life of indigenous people by, for example, increasing income inequalities. The Kuznets hypothesis predicts that the link between income inequality and income (a proxy for economic development) resembles an inverted U. We test the hypothesis using a survey of 511 households from 59 villages of Tsimane' Amerindians, a horticultural-foraging society in the tropical rain forest of Bolivia. We measure village inequalities of three economic outcomes: income, imputed annual value of rice production, and wealth. We used three indices of inequality: the coefficient of variation, the standard deviation of the logarithm, and the Gini coefficient. Explanatory variables include either income and income squared, wealth and wealth squared, or imputed annual rice production and production squared. We used village-to-town distance as a control. We find little evidence that integration to the market increases inequalities of economic outcomes, with two exceptions: Wealth bore the predicted inverted U-shaped relation with wealth inequalities, and imputed rice production bore a U-shaped relation to inequality, but only when (a) using adult equivalents to express household size and (b) the Gini coefficient and the coefficient of variation to measure inequality; in no case were results robust to different econometric specifications. We advance several explanations for why economic development might not accentuate economic inequalities among relatively autarkic rural economies.