State governments have recently adopted many policies creating competition in service provision. These policies typically are expected to reduce costs and improve quality through organizational innovation. I present an analysis of one such initiative—the creation of charter schools. Using evidence from a survey-based study comparing charter schools and traditional public schools, I explore the relationship between competition and organizational innovation. I find that competition does appear to promote innovative local practices. But the evidence also suggests that competition can limit communication within the practitioner community, which, in turn, can stifle the diffusion of innovative ideas. Thus, policy design for local innovation may be more effective when competition is augmented with state-level strategies that promote inter-organizational cooperation and learning.

This study uses a three-state survey to examine how state legislators and legislative candidates view the ballot initiative process. We hypothesize that state legislators see the initiative as a rival to the legislature, and that the initiative is evaluated in terms of expectations about policy outcomes. We test whether opinions about the initiative are related to legislative incumbency, partisan interests, or both. We find that incumbents, regardless of party, are more interested in constraining the initiative than non-incumbents. We also find an independent effect of partisanship, with Republicans having a more positive view of the initiative than Democrats.

We examine the effects of state economic conditions on gubernatorial popularity using time-series analyses of gubernatorial approval ratings for 10 states. We find that although unemployment rates affect some governors' approval ratings, the dynamics of gubernatorial approval are highly idiosyncratic. Our results suggest that a single model may be inadequate to describe patterns of public approval, even for candidates elected to similar offices. The unique personal characteristics of the officeholder, as well as differing constituent expectations, may be at least as important as the general economic variables typically included in models of public approval.

Term limit laws for state legislators now exist in 19 states. While these laws were passed as early as 1990, they have only recently actually forced legislators from office. Prior to the 2000 elections, only two states (California and Maine) had experienced virtually complete legislative turnover in at least one chamber as a result of term limits. Nine other states have had at least one cohort of legislators forced out of office. Now that a significant portion of these legislative bodies has been replaced by term limits, we can begin to examine the consequences of this reform. We report findings from a survey of lobbyists in five term-limits states. We find strong consensus among these lobbyists that term limits have caused the state political influence structure to shift away from the legislature and toward the governor, administrative agencies, and interest groups. With regard to the impact on internal operations and procedures of the legislature, we discover considerable variation across states, variation that, in part, is related to legislative professionalism.

How do people hold elected officials accountable for expected economic conditions in a system of government in which jurisdiction and responsibility are divided such as they are in the United States? Due to the president's visibility, people may hold all elected officials of the president's party responsible for economic conditions. Or, people may be able to differentiate between state and national economic conditions and jurisdictional responsibilities and hold governors of either party accountable for state economic conditions. I use ordered probit to analyze a pooled model of 13 individual-level mid-1990s Michigan opinion surveys and find no evidence that individuals consistently rely on a presidency-centered referenda model of cognition when evaluating the state's governor. Instead, gubernatorial performance ratings reflect an incumbent-centered accountability model. The evidence also suggests jurisdictional accountability, where gubernatorial evaluations are influenced by state economic expectations, but not by national economic expectations.

Term limits provide a rare opportunity to employ pre- and post-intervention research designs to investigate the effects of turnover in legislatures. This article describes a study of term limits in Michigan that takes advantage of this opportunity. With eight states implementing term limits in 2002 or soon thereafter, there are opportunities for other scholars to replicate all or parts of the study described here. The payoffs for such projects in terms of generating systematic answers to the impacts of legislative turnover and term limits are considerable.