A revised model of unilateral accidents

Jacob Nussim*, Avraham D. Tabbach

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


This paper suggests and justifies a revised formulation of the unilateral accident model based on relaxing two assumptions of the standard model: the precaution function and the harm function. The revised model is, therefore, more general and corresponds better to various situations. A resulting trait of the generalized model is its account for the interaction between the injurer's care and activity levels, which was implicitly assumed away so far. The revised model is examined using a few standard issues in tort and the analysis brings new results and insights for the unilateral accident case. For example, the view that, under a negligence regime, due care can be defined regardless of the optimal level of activity holds under very restrictive assumptions. In general, due care must be defined simultaneously with the optimal activity level. In addition, the common view suggests that underestimation of the level of actual damages under strict liability would induce injurers to take insufficient care and to engage excessively in a risky activity (and vice versa, for overestimation). This paper shows that underestimation of actual damages may counter-intuitively lead to insufficient activity or excessive care levels. Similarly, the results of underestimating harm under a negligence regime prove to be different than commonly thought. In addition, the revised model questions the intuitive similarity between the underestimation of harm and the judgment-proof problem, and provides some new results for the latter problem.

Original languageEnglish
Pages (from-to)169-177
Number of pages9
JournalInternational Review of Law and Economics
Issue number2
StatePublished - Jun 2009


  • Damages
  • Judgment proof
  • Negligence
  • Strict liability
  • Tort
  • Unilateral accident model


Dive into the research topics of 'A revised model of unilateral accidents'. Together they form a unique fingerprint.

Cite this