Jeffrey F. Timmons

How do fiscal institutions, partisanship, and governance affect federal transfers to municipalities? We address this question using a novel research design and data set for Mexico. We compare the state-level obligations for federal transfers to municipalities with the distribution of these funds as reported by municipalities. This strategy allows us to know whether state-level formulas are binding, whether there are partisan skews in the formula, and how and why governors reallocate funds. We find that state-level fiscal institutions are quite binding; even so, deviations from the formula total approximately US$300–500 million annually. Whereas Partido Revolucionario Institucional governors appear to reallocate to municipalities when they are governed by their co-partisans, Partido Acción Nacional and Partido de la Revolución Democrática governors appear to reallocate funds to municipalities for equity, stabilization, and disasters (with no detectable partisan bias).

How much growth do (economic and legal) institutions cause? To quantify this effect, we adapted the baseline regression in Acemoglu, Johnson and Robinson's (2002, Quarterly Journal of Economics, 117(4), pp. 1231–1294) seminal work on the causal relationship between the quality of institutions and differences in modern-day income levels was adapted. We found that improving institutional quality by one standard deviation increased a country's average annual growth rate by only 0.4% from 1820 to 1995.

The relationship between specific skills and the welfare state has been the subject of considerable debate. To help resolve the conflict, we present a general model of preferences over social insurance with endogenous wages and investment in specific skills and a variety of exogenous constraints. Our dynamic model underscores the link between wages, skills and unemployment risks. It shows that skill-specificity is irrelevant for preferences over social insurance when wages adjust for investment costs and unemployment risks. We validate the adjustment mechanism with U.S. data. We then extend the model to show how different conditions, including centralized wage bargaining, capital market imperfections, and taxation, affect skill formation and skill-based preferences for social insurance. Our model provides an analytical framework that can reconcile the disparate empirical findings and demonstrates how they, along with Iversen and Soskice’s seminal results, are special cases of the interaction between labor markets and politics.