Decentralizing crime control: the political economy perspective.

• Author: Teichman, Doron
• Position: Response to articles in this isssue, p. 1713, 1725, 1733

INTRODUCTION

In an article recently published on the pages of this Law Review, The Market for Criminal Justice: Federalism, Crime Control, and Jurisdictional Competition ("The Market"), (1) I put forward a theory of crime control in a decentralized government. Specifically, I made three distinct claims. First, criminal justice policies affect the geographic decision of criminals as to where to commit their crimes. (2) Other things being equal, criminal activity will tend to shift to areas in which the expected sanction is lower. Second, local jurisdictions attempting to lower their crime rates will react to policies adopted by neighboring jurisdictions and try to keep up with their neighbors' sanctioning levels. (3) In other words, the optimal expected sanction for a certain jurisdiction cannot be derived from the characteristics of that jurisdiction alone; it must incorporate the expected sanctions of neighboring jurisdictions. (4) Third, competition among local jurisdictions in the area of criminal justice could be both efficient (a race to the top) and inefficient (a race to the bottom) depending on the specific context over which jurisdictions are competing. (5)

In three insightful comments, Professors Rachel Barkow, Sam Gross, and Wayne Logan deepen and broaden the discussion I attempted to start in The Market. (6) They flesh out in great detail some of the theoretical complexities and practical difficulties regarding the competitive forces driving criminal justice policies that I did not fully treat in The Market. Further, they demonstrate that we should use great caution before adopting any policy recommendations based on the insights of The Market. The comments clearly help a great deal to understand the issues at hand; one should not read The Market without reading this correspondence.

My goals in this short Reply are twofold. On one hand, I will answer some of the criticisms leveled against the arguments made in The Market. Obviously, due to the constraints of this format, I cannot answer every point made by the commentators, so I will focus my remarks on the main areas of disagreement. On the other hand, I will try to build on the comments to extend my initial analysis. The Reply is divided into two parts. In the first, I shall deal with the existence of competition in the area of criminal justice and defend my claim that competition affects the design of criminal justice polices in a decentralized government. In the second, I will turn to the policy implications of my claims and argue that The Market offers constructive policy recommendations for any decentralized criminal justice system, including the United States.

1. THE POTENTIAL PROBLEM

   In The Market I argued that competition in the area of criminal justice could potentially lead to a race to the top or a race to the bottom. The focus of all three comments published here is on the potential problem--namely, the race to the bottom. Yet at the outset I would like to point out that in The Market I spoke of a potential race to the bottom and emphasized the benefits of jurisdictional competition in the area of criminal justice. (7) Competition might drive local governments to innovate and adopt efficient crime control measures. In addition, decentralization could allow criminal laws to be tailored to the sensitive preferences of local communities. Furthermore, there might be organizational inefficiencies associated with creating large national agencies that can be avoided by creating small local agencies. Finally, there might be informational advantages to relying on local rather than national systems. Thus, while this Reply will focus on the potential race to the bottom, the reader should note that that race is only one possible factor to be weighed in determining crime policy.

   The commentators express skepticism as to the existence of competition with respect to criminal justice. Gross, for example, argues that there might not even be a question of a race to the top or a race to the bottom for the simple reason that there is no race to begin with. (8) Gross's skepticism arises from the lack of empirical data supporting the competition hypothesis. This concern is legitimate, and that is why the claims made in The Market were tentative, pending further empirical research. But Gross does more: he presents data that supposedly suggest that competition does not play an important role in the area of criminal sanctions. (9) Yet a close evaluation demonstrates that the data he presents cannot rebut the competition hypothesis. For instance, Gross points out that between the years 1980 and 2001, while the drug related state prison population rose by a factor of thirteen, the federal drug related prison population rose by a factor of sixteen. (10) According to Gross, since the federal government is not part of the jurisdictional race, the fact that its sanctions rose faster than state sanctions falsifies the competition hypothesis. (11) This claim is unpersuasive for two reasons. First, I am not sure that the federal government is not functioning in a competitive setting in the area of drug crimes. Drug crimes in the United States are part of an international market for drugs, and from that perspective it might be the case that the United States is attempting to place itself as a country in which the cost of committing crimes is high. If, for example, the United States manages to confiscate twice as many drugs as Canada, a drug exporter in South America might decide to focus his activity on the Canadian market rather than on the American market. Secondly, and perhaps more importantly, as Barkow clearly described in her reply, the politics of criminal law at the state and federal level are not identical. (12) Thus, simply presenting the raw numbers (as is done in Gross's response) does not explain the causation behind the numbers. I do not claim to present a rigorous statistical analysis here, but it might well be the case that displacement was a driving force at the state level, while other political forces were active on the federal level. If, for example, the decision rule of federal legislatures in the area of criminal law is "copy all state legislation and make it a little tougher"--and Barkow's analysis seems to indicate that that is a distinct possibility--one would expect to see the exact statistical picture Gross presents. Everything else isn't equal and thus we simply cannot draw the conclusions that Gross would like us to draw from the evidence he presents. (13)

   Gross's skepticism as to the competition hypothesis also rests on a rejection of the underlying assumptions as to the behavior of criminals. As he puts it, "[i]t's hard to imagine that black crack users from Cleveland have responded to the policy of the Cuyahoga County prosecutor by driving over the border [and] smoking crack in the nearly all-white suburbs...." (14) Even without conducting extensive empirical research one has to concede that this observation must be true in a large number of cases. Yet as the commentators acknowledge, some displacement clearly cannot be ruled out; (15) likewise, the possibility that the perception of displacement is driving policies cannot be ruled out. (16) That is an empirical question, and aside from citing anecdotes supporting our positions, not much can be done at this point in time. (17) What I would like to emphasize here is that at least one competitive force I described in The Market requires no assumptions as to the mobility of criminals: the force of competition for capital investments. (18) If criminals are completely immobile, but capital is mobile and tends to shift to jurisdictions in which the crime rate is lower, (19) then the competition for capital investments will drive the harshening of the criminal justice system. (20) To draw on Gross's hypothetical, a criminal in Cleveland might not move to Seattle because of a lower expected sanction, but a multinational corporation might decide to invest in Seattle rather than in Cleveland because of lower crime rates created by higher expected sanctions.

   Finally, Logan also raises doubts as to the validity of the competition hypothesis due to the large scope of interstate reliance in the area of criminal justice. Logan points out that a sizeable number of jurisdictions (sixteen) adopted the external approach with respect to the registration requirement of their Sex Offender Registration and Notification Law (SORNL). (21) Logan notes many negative consequences that arise from the use of the external approach. (22) I am not sure why this evidence is contrary to my assertion. The fact that so many jurisdictions adopted the external approach supports the competition hypothesis. The very content of those laws demonstrates that legislatures are concerned about sex offender migration when enacting SORNLs. In other words, a state cannot design its optimal SORNL without taking into account the equivalent legislation of other states. That is precisely the positive claim I made in The Market.

   While I disagree with the comments' denial of the very existence of competition...