The United States is unique in that most states select their prosecutors using popular election. An interesting feature of prosecutor elections is that the pool of potential challengers to an incumbent comes primarily from subordinates within the office. We develop a signaling model that analyzes whether entry into the political race by a subordinate provides voters with useful information. We challenge the commonly held presumption that the election mechanism is ineffective. Exploring the decision of an insider to challenge the boss, we demonstrate that a challenger can strategically convey information about the incumbent's quality by entering the race. Anticipating this, poor incumbents exit. Thus, the model matches empirical observations of rare contested elections and rather modest vacancy rates.Compared to the situation where outsiders challenge the incumbent, the election mechanism is effective.

We revisit the relationship between collective bargaining by teachers unions and school performance. The empirical literature in this area has found mixed results at both the state and district levels. We contribute to this literature in two ways. First, rather than simply dummy union status, we proxy for the restrictiveness of collective bargaining agreements with the number of pages per agreement. Second, we employ Bayesian spatial methods to deal with spatial dependence in school district activities. Our reduced-form results indicate that collective bargaining directly lowers scores on high school math scores, but that the total effect is zero.

No abstract

No abstract