Are all legal probabilities created equal?

Yuval Feldman*, Doron Teichman

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

33 Scopus citations

Abstract

At the core of the economic analysis of law lies the concept of expected sanctions, which are calculated by multiplying the severity of the sanction that is applied to wrongdoers by the probability that it will be applied. This probability is the product of several sequential probabilities involving the different actors responsible for sanctioning wrongdoers (e.g., police, prosecutors, judges, jurors, etc.). Generally, legal economists treat different legal probabilities as fungible, simply multiplying them much like any other sequential probabilistic situation. This Article challenges this assumption, demonstrating that people perceive and are affected by different types of legal probabilities in distinct ways. More specifically, it shows that uncertainty associated with the substance of the law and uncertainty associated with imperfect enforcement should not be treated equivalently. To demonstrate this point, this Article presents a series of between-subjects experimental surveys that measure and compare participants' attitudes toward compliance in conditions of uncertainty. Study participants-several hundred students from Israel and the United States-answered questions in the context of one of several variations on the same hypothetical scenario. While the expected sanction was the same in each variation, the source of uncertainty differed. These studies confirmed that people are less likely to comply when uncertainty stems from the imprecision of law's substance than when uncertainty stems from the imperfect enforcement of clear law.

Original languageEnglish
Pages (from-to)980-1022
Number of pages43
JournalNew York University Law Review
Volume84
Issue number4
StatePublished - Oct 2009

Fingerprint

Dive into the research topics of 'Are all legal probabilities created equal?'. Together they form a unique fingerprint.

Cite this