Over the last few decades, scholars have increasingly turned to identifying the causal impact of changes in policies and laws on criminal justice outcomes. This literature, too enormous to overview here, has yielded important insights into the actual functioning of crime and judicial systems. Typically, however, legal actors are not modeled and identification strategies assume they act naively (or even randomly). Moreover, “policy implications” are made with far too little concern for unanticipated consequences or methodological limitations.

“Statistics first” is the zeitgeist in many areas of the social sciences.Footnote 1 As Hipp and Williams (2020) summarize, “the data and methods used by researchers in recent years are so novel that theory is at best used in a post-hoc manner to explain findings rather than to generate testable hypotheses”. While the balance of theory and empirics is an active research area (Todd & Wolpin, 2021; Imbens, 2021; Al-Ubaydli et al., 2021; Card, 2022),Footnote 2 Recent research now clearly documents the problems with a “statistics first” mentality (Vivalt, 2020; Akerlof, 2020; Brodeur et al., 2020; Huntington-Klein et al., 2021; Weill et al., 2021; Eliaz et al., 2021; Jardim et al., 2022; de Chaisemartin & D’Haultfoeuille, 2022; Young, 2022; Frankel & Kartik, 2022; Beiser-McGrath & Beiser-McGrath, 2023; Simonsohn, 2022). Criminal justice research is not immune. Our knowledge of crime and courts does not improve lexicographically—progressing along rungs of a statistical hierarchy with a single gold-standard of evidence on top.

While many empirical fields face tradeoffs between identifiability and practical relevance, criminal justice research faces unique measurement issues (Adamson & Rentschler, 2023). Most of our empirical records result from actions by criminal actors (who do not want any record of their activity) as well as civilians, police, coroners, and many others. Scholars often make assumptions that the behavior of some, or even all, of such actors are exogenous to satisfy statistical criteria. This ignores much economic literature and is a main way that a “statistics first” mentality creates false findings and obscures true ones,Footnote 3 Additional issues surround the research questions and cases that a “statistics first” mentality selects for. Hayek (1943) is again pertinent:

The blind transfer of the striving for quantitative measurements to a field where the specific conditions are not present which give it its basic importance in the natural sciences is the result of an entirely unfounded prejudice. It is probably responsible for the worst aberrations and absurdities produced by scientism in the social sciences. It not only leads frequently to the selection for study of the most irrelevant aspects of the phenomena because they happen to be measurable, but also to “measurements” and assignments of numerical values which are absolutely meaningless. What a distinguished philosopher recently wrote about psychology is at least equally true of the social sciences, namely that it is only too easy “to rush off to measure something without considering what it is we are measuring, or what measurement means.”

Once we acknowledge measurement and methodological issues, we must also acknowledge that many links to policy are tenuous at best. While “legal capacity” research is in some sense the branch of criminal justice that explicitly adopts the “best of motives” assumption, experimental and applied microeconomic studies also seem to have a well-intending policymaker in mind with their implied policies. (Experimental studies are more focused on justice biases and quasi-experimental studies are more focused on legal biases, but both evade the question of “who will nudge the nudgers?”) The pure theory of self-interested policymakers is not sufficient to determine policy, but it is necessary for deducing policy implications from criminal justice statistics.

A mentality of complements, we argue, is a promising path forward. The cost-benefit analysis of crime and courts (such as Becker (1968)) are crowning achievements of economics, and a “statistics first” mentality to criminal justice is in many ways a return to a pre-economic analysis of law. (Criminologists have a long history of field experiments, including the Cambridge-Somerville Youth Study of 1939 and the Kansas City preventive patrol experiment of 1972; see Oakley (2000).) Although there are gains from adopting the credible methods developed by criminologists, the gains are probably larger with specialization and exchange. While public choice scholars have contributed both empirics and theory, economists might have a comparative advantage in theory. In particular, public choice theory has much to contribute in answering two broad questions: what do empirical analyses actually show when criminal justice data are generated from self-interested legal actors and researchers? What legal policies are feasible and desirable when being determined by self-interested policy-makers?

We call for public choice scholars to analyze legal systems and actors using a variety of methodologies, and to exploit the complementarities between them. For this special issue, we solicited papers that took all legal actors to be rational, and which used experiments, observational data, or applied theory. We fostered complementarity across these papers by requesting at least one reviewer with a different methodological specialty. As might be expected, the papers that appear in this special issue are methodologically diverse.

One topic that features prominently is the importance of civilian views and beliefs with regard to legal actors. Hong and Zhang (2023) theoretically examine the role of bureaucratic beliefs in corruption. Candelo et al. (2023) report an insightful experiment exploring how trust for public officials varies with income. Baumann et al. (2023) provide a theoretical and experimental analysis of the effects of fine revenue going directly to the police or to the public purse. DeAngelo et al. (2023) provide an observational study examining how prosecutors “upcharge” to protect the police from bad publicity.

Another theme emerged on the effect of incentives and institutions surrounding trials on behavior prior to trials. Michaeli and Zohar (2023) provide a theoretical demonstration that the availability of plea bargaining will tend to dramatically reduce the prevalence of trials. Ralston et al. (2023) provide a theoretical and experimental analysis of the effects of incentives that change how prosecutors value plea bargains or trial sentences. Bienenstock and Kopp (2023) provide a theoretical model that helps explain why foreign companies have consistently chosen to settle cases brought against them by the American Foreign Corrupt Practices Act. Guerra et al. (2023) provide a theoretical and experimental analysis of behavior before trials under adversarial and inquisitorial systems.

The special issue also includes Ball et al. (2023), which experimentally evaluates whether how rights are framed can affect cooperative behavior. Finally, Di Liddo and Morone (2023) reports the results of an interesting experiment that explores the relationship between fiscal disparities and corruption.