This paper aims to identify the obstacles to school progression by using field surveys that were conducted in twenty-five Pakistani villages. The full-information maximum likelihood (FIML) estimation of the sequential schooling decision model reveals important dynamics of the gender difference in educational attainment, intrahousehold resource-allocation patterns, and transitory income and wealth effects. In the descriptive statistics as well as the econometric analyses, we find a high educational retention rate and observe that school progression rates between male and female students after secondary school are comparable. In particular, we find gender-specific and schooling-stage-specific birth-order effects on education. Our overall findings are consistent with the implications of optimal schooling behavior under binding credit constraints and the self-selection of education-friendly households. Finally, we find serious supply-side constraints which might arise from a village-level lack of demand for primary schools for girls.
In the early twentieth century, a large number of households resettled from the European to the Asian part of the Russian Empire. We propose that this dramatic migration was rooted in institutional changes initiated by the 1906 Stolypin land titling reform. One might expect better property rights to decrease the propensity to migrate by improving economic conditions in the reform area. However, this titling reform increased land liquidity and actually promoted migration by easing financial constraints and decreasing opportunity costs. Treating the reform as a quasi-natural experiment, we employ difference-in-differences analysis on a panel of province-level data that describe migration and economic conditions. We find that the reform had a sizeable effect on migration. To verify the land liquidity effect, we exploit variation in the number of households participating in the reform. This direct measure of the reform mechanism estimates that land liquidity explains approximately 18% of migration during this period.
We examine spatial spillovers in institutional development. Dependent variables are institutional measures reflecting politics, law, and governmental administration. The explanatory variable of interest is the level of institutions in bordering countries—a spatial lag of the dependent variable. Our spatial model directly leads to the identification strategy for the endogenous spatial lag. We implement new results in spatial econometrics to counter missing-data problems usually rife in spatial empirics. Spatial institutional spillovers are statistically significant and economically important. A counter-factual exercise – the non-existence of the USSR – reveals large direct and indirect spillovers. Numerous robustness exercises bolster conclusions, including yearly cross-section regressions, fixed effects estimates, and adding many extra explanatory variables. Moreover, we provide a new theoretical result showing the robustness of estimates in the presence of omitted variables. We extend the core model, allowing different effects for better and worse neighbors, using inverse distance weights, estimating the spatial-Durbin model, and using Polity's institutional measure.