THE COURTS AND RISK ASSESSMENT1

In the past decade the federal courts have come to play an important role in reviewing agency decision-making on prospective risks. Questioning the conventional wisdom that judges are poorly equipped for the task, the authors outline the range of choices facing courts in such cases and contend that they cannot avoid making ultimate decisions on risk policy. However, recent Supreme Court cases on nuclear hazards and occupational benzene indicate narrowing of the scope for judicial review.     

REGULATION AND THE THEORY OF AGENCY1

The major arguments of this paper are that there is a need for an integrating framework for the study of regulation, including the design of regulatory institutions, and that the theory of agency may provide such a framework. The paper provides a brief overview of this approach. The theory of agency is a general theory of social relationships of “acting for” that is now under development in several disciplines, particularly economics and accounting. Regulation is seen as a generic relation observed widely in social behavior, and as a particular type of agency relationship. The problems of agency relations–e.g., the problems of principals in controlling agents and of agents in acting according to the principals' desires–have their counterparts in regulation.     

THE RELATIONSHIP BETWEEN CRIME AND INCARCERATION AMONG THE AMERICAN STATES1

Data from the 50 states for 1975 are analyzed to try to separate the presumed reciprocal effects on each other of crime rates and incarceration rates. Two-stage least-squares regression shows no significant effect of crime on incarceration nor of incarceration on crime. Incarceration rates are affected mainly by the percentage of blacks in the state population. Crime rates are significantly influenced by urbanization, region, mean income, and unemployment.     

BUREAUCRATIC DECENTRALIZATION AND THE REDISTRIBUTION OF URBAN SERVICES: THE CASE OF TEAM POLICING1

Two extreme positions dominate the decentralization debate in urban politics: one school consistently opposes decentralization, while the other maintains that only complete radical community control can solve the problems of politics and administration in the city. Positions and policies between these two extremes tend to get lost in rhetoric and argument, or receive only passing recognition. This paper is an attempt to carve out, theoretically and empirically, some space for an intermediate position. Following a brief exploration of the decentralization debate, we discuss and present some analysis from a team policing experiment, a type of ecentralization that falls in between the two extremes. The data suggest that this type of decentralization can be successful if the program is designed as a non-zero-sum game, and the paper concludes that intermediate programs represent attractive third choices to discredited or unworkable alternatives.