Preferences for criminal justice error types: Theory and evidence

Yehonatan Givati*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


What shapes individuals’ preferences for criminal justice error types, that is, the preferences for convicting the innocent versus letting the guilty go free? The strong correlation between preferences for criminal justice error types and incarceration rates across countries highlights the importance of these preferences. I develop an instrumental theory and an intrinsic theory of the preferences for criminal justice error types. Using individual-level data from the United States, I find support for both theories. Consistent with the instrumental theory of preferences, gender, race, and concern about crime shape preferences. Consistent with the intrinsic theory of preferences, education and ideology also shape preferences. I confirm these findings using individual-level data from 22 countries and provide some suggestive evidence that culture shapes preferences too.

Original languageAmerican English
Pages (from-to)307-339
Number of pages33
JournalJournal of Legal Studies
Issue number2
StatePublished - 1 Jun 2019

Bibliographical note

Funding Information:
yehonatan givati is Professor of Law at Hebrew University Law School. I am grateful to Edward Morrison and to an anonymous referee for their very helpful comments, which significantly improved the paper. I am also grateful to Henrik Lando, Murat Mun-gan, Holger Spamann, Megan Stevenson, Gui Woolston, and Asaf Zusman and to participants in seminars at the University of Virginia Law School, the University of Chicago Law School, the University of Texas Law School, Pepperdine Law School, Stanford Law School, and the American Law and Economics annual meeting at Boston University for helpful comments and to Hebrew University’s Center for Empirical Legal Studies for financial support.

Publisher Copyright:
© 2019 by The University of Chicago. All rights reserved.


Dive into the research topics of 'Preferences for criminal justice error types: Theory and evidence'. Together they form a unique fingerprint.

Cite this