Do Entry Costs Provide an Empirical Basis for Poverty Traps? Evidence from Mexican Microenterprises

Recent theoretical literature in development economics has shown that nonconvex production technologies can result in low-growth poverty traps. This article uses detailed microenterprise surveys in Mexico to examine the empirical evidence for these nonconvexities at low levels of capital stock. While theory emphasizes nondivisible start-up costs that exceed the wealth of many potential entrepreneurs, this article shows start-up costs to be very low in some industries. Semiparametric methods are then used to flexibly estimate returns to capital in microenterprises. Much higher returns are found at low levels of capital stock than at higher levels, and this remains true after controlling for firm characteristics and measures of entrepreneurial ability. Overall, little evidence of production nonconvexities is found at low levels of capital. The absence of nonconvexities is a significant finding because it suggests that access to start-up capital does not determine the ultimate size of the enterprise.