Search

I describe and analyze a model of strategic communication and deliberation in decentralized decision‐making settings. I show that, in a cheap‐talk environment, inclusion and exclusion of agents can affect the credibility of messaging between agents and, accordingly, the quality of policy decisions and overall social welfare. Somewhat surprisingly, the inclusion of agents can aid information aggregation and social welfare even when the added agents do not themselves communicate truthfully. Analogously, the results suggest an informational, social welfare–based rationale for excluding agents not only from observing policy‐relevant deliberation but also from observing the product of the communication precisely because the excluded agents possess decision‐making authority. [ABSTRACT FROM AUTHOR]

Much of policy-making involves prioritization —deciding not only what to do but also when —and uncertainty —not knowing exactly how the choices made will affect actual policy outcomes. I present a theory of dynamic prioritization within a hierarchical organization. The model illustrates how notions such as an agency's performance, mission, and critical tasks are linked with details such as institutional structure and the preferences of both front-line bureaucrats and their overseers. The theory highlights some reasons why even sincere, representative policy-making decisions might appear irrational, inconsistent, or "captured" to outside observers. This is in contrast to classical "spatial models" of policy that abstract from the more quotidian details of how policy is actually made as opposed to simply being "chosen." The theory also generates traditional comparative static-style predictions about the features of the policy-making tasks, the preferences of bureaucrats and political overseers, and agency structure that affect the substance and quality of policy-making. Finally, the theory offers a general explanation for why real-world agencies employ widely varying processes to organize and implement policy-making: optimal policy-making is—in a precise sense—"sufficiently complicated" to render a succinct and robust summary of optimal management impossible. [ABSTRACT FROM AUTHOR]

"Banning the Box" refers to a policy campaign aimed at prohibiting employers from soliciting applicant information that could be used to statistically discriminate against categories of applicants (in particular, those with criminal records). In this article, we examine how the concealing or revealing of informative features about an applicant's identity affects hiring both directly and, in equilibrium, by possibly changing applicants' incentives to invest in human capital. We show that there exist situations in which an employer and an applicant are in agreement about whether to ban the box. Specifically, depending on the structure of the labor market, banning the box can be (1) Pareto dominant, (2) Pareto dominated, (3) benefit the applicant while harming the employer, or (4) benefit the employer while harming the applicant. Our results have policy implications spanning beyond employment decisions, including the use of credit checks by landlords and standardized tests in college admissions. [ABSTRACT FROM AUTHOR]

Algorithmic fairness is a new interdisciplinary field of study focused on how to measure whether a process, or algorithm, may unintentionally produce unfair outcomes, as well as whether or how the potential unfairness of such processes can be mitigated. Statistical discrimination describes a set of informational issues that can induce rational (i.e., Bayesian) decision‐making to lead to unfair outcomes even in the absence of discriminatory intent. In this article, we provide overviews of these two related literatures and draw connections between them. The comparison illustrates both the conflict between rationality and fairness and the importance of endogeneity (e.g., "rational expectations" and "self‐fulfilling prophecies") in defining and pursuing fairness. Taken in concert, we argue that the two traditions suggest a value for considering new fairness notions that explicitly account for how the individual characteristics an algorithm intends to measure may change in response to the algorithm. [ABSTRACT FROM AUTHOR]

To purchase or authenticate to the full-text of this article, please visit this link: http://dx.doi.org/10.1111/j.1540-5907.2007.00286.x Byline: Sean Gailmard (1), John W. Patty (2) Abstract: 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-a-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. Author Affiliation: (1)University of California, Berkeley (2)Harvard University Article note: Sean Gailmard is assistant professor of political science, University of California, Berkeley, 210 Barrows Hall, Berkeley, CA 94720 (gailmard@berkeley.edu). John W. Patty is assistant professor of government, Harvard University, 1737 Cambridge Street, Cambridge, MA 02138 (jpatty@gov.harvard.edu).