Suchergebnisse

In the past decade, political science has witnessed a substantial amount of research using formal models to explicate the rationale for and effects of myriad aspects of bureaucratic institutions. Whereas previous waves of formal modeling on bureaucratic structure emphasized bureaucracy as a device for making policy commitments last, more recent formal research has grappled with information asymmetries and more explicitly considered the principal-agent relationship between bureaucracies and political authorities. We review several major recent themes in this literature, particularly the effects and development of bureaucratic hierarchies, the agency dilemmas inherent when policy-making authority is delegated to bureaucrats, and the effects of institutional structure on the development and sharing of expertise and capacity in bureaucracies. Adapted from the source document.

We investigate how aspects of "civil service" systems of personnel management interact with bureaucratic discretion to create expert bureaucracies populated by policy‐motivated agents. We construct a dynamic model in which bureaucrats may invest in (relationship‐specific) policy expertise and may or may not be interested in policy choices per se. The legislature makes sequentially rational grants of discretion, which serve as incentives for expertise investment and continued service only for policy‐motivated bureaucrats. Bureaucratic policy preferences and the legislature's agency problem vis‐à‐vis bureaucracies develop endogenously in the model. Bureaucratic expertise can be supported in equilibrium only at a cost of its politicization; "neutral competence" is inconsistent with strategic incentives of bureaucrats. We identify several conditions that support the development of an expert bureaucracy in equilibrium, including security of job tenure and control over policy issues for policy‐motivated bureaucrats.

This paper examines inference and attribution in a simple and ubiquitous strategic situation: a voter is faced with discerning whether a leader worked on his or her behalf after observing an informative, but noisy signal about the leader's performance. We characterize perfect Bayesian equilibria, quantal response equilibria (QRE), and provide a simple model of a heuristic-based approach, referred to as strategic naivete, within a wide class of such environments. We also discuss experiments conducted to examine human behavior within such an environment. While it is clear that the observed behavior is inconsistent with perfect Bayesian equilibrium, distinguishing between QRE and strategic naivete will require further work. We conclude with a discussion of the broader implications of probabilistic and/or heuristic-based attribution processes for electoral politics and political economy. Adapted from the source document.

In this article, we extend the well-known 'agreeing-to-disagree' and 'no-trade' results from economics and game theory to international relations. We show that two rational countries should never agree to go to war when war is inefficient and when rationality is common knowledge. We argue that this result might provide one possible explanation for the empirical finding, often referred to as the 'democratic peace', that modern democracies rarely go to war with one another. We propose that the informational properties of pluralistic institutions (as opposed to oligarchies or dictatorships) lead to better decision-making by democracies and that democracies are therefore more likely to be the rational actors necessary for the 'no-war' result. We discuss empirical evidence in support of this proposition.

Few, if any, elected representatives are capable of unilaterally implementing their platforms. Rather, they choose between options generated by other actors and/or external events. We present a theory of voters' preferences over representatives who will cast votes on their behalf, and show that in this setting voters' preferences over candidates' platforms will not look like voters' preferences over policies. We demonstrate that these induced preferences for representation tend to favor more extreme representatives, and we present two models of electoral competition in which induced preferences over representatives lead to elite polarization.

Kenneth J. Arrow was one of the most important intellectuals of the twentieth century, and his "impossibility theorem" is arguably the starting point of modern, axiomatic social choice theory. In this review, we begin with a brief discussion of Arrow's theorem and subsequent work that extended the result. We then discuss its implications for voting and constitutional systems, including a number of seminal results—both positive and negative—that characterize what such systems can accomplish and why. We then depart from this narrow interpretation of the result to consider more varied institutional design questions such as apportionment and geographical districting. Following this, we address the theorem's implications for measurement of concepts of fundamental interest to political science such as justice and inequality. Finally, we address current work applying social choice concepts and the axiomatic method to data analysis more generally.