The Journal of Things We Like (Lots)
Select Page

W. Jonathan Cardi, Randy Penfield & Albert Yoon, Does Tort Law Deter?, Wake Forest Univ. Legal Studies Paper No. 1851383 (2011), available at SSRN.

Why have tort law? After all, other laws and institutions cover a lot of the same ground. If we want to punish bad actors, there is criminal law. If we want to ensure safety, there is regulatory law. If we want to aid injury victims, there is public and private insurance. What’s left for tort?

A standard answer is that tort, with its threat of liability, deters people from engaging in conduct that unduly risks harm. Tort law is needed for this purpose, the answer continues, because a good deal of risky conduct falls outside the ambit of criminal and regulatory law. For example, injuring someone by careless driving usually isn’t a crime, so the threat of negligence liability fills the void.  Of course this answer assumes that the prospect of liability will actually induce people to act more carefully.  Does Tort Law Deter?, by Professors Cardi, Penfield, and Yoon, is an innovative attempt to harness social science methods to investigate that assumption.

The authors anonymously surveyed 700 first-year students at multiple law schools. (Most respondents completed the survey before or during their first Torts class.) Each was invited to state the likelihood that s/he would undertake certain specified activities that carry risks of injury to others.  For example, each was asked to rate the likelihood that s/he would make a watercraft available to house guests without providing them with life jackets. At the same time, participants were randomly assigned a description of one of four legal regimes governing the activities: (1) no liability (i.e., no legal consequences for engaging in the activities); (2) a nonquantified possibility of a significant criminal fine, but no tort liability; (3) an absence of “laws or rules against” the activities, but a nonquantified possibility that the subject “might be sued and might have to pay for any injury caused by [the activities]”; or (4) no description of the legal regime.  The main goal was to determine whether regime (3) – which features tort but not criminal liability – reduced participants’ stated inclination to undertake any of the activities as compared to regime (1)’s no liability scheme.

The results suggest that the prospect of tort liability had little or no deterrent effect: participants reported being just as likely to undertake the risky activities in regimes (1) and (3).  By contrast, participants reported that they would be less likely to undertake the activities in regime (2).  If valid, these findings support the notion that the threat of tort liability has no significant deterrent effect, at least with respect to the kinds of activities mentioned in the survey. However, the authors are for appropriately cautious about reading too much into their study, acknowledging both limitations of survey-based research and their own particular survey.

This is an imaginative and provocative paper, well worth a read. The authors deserve praise for use of a novel research design to try to make headway in assessing tort law’s effects.  Still, I have concerns about the design, particularly the description of the tort liability regime that is built into the survey.

Recall that the survey asked participants to gauge their willingness to participate in a set of risky activities under different legal regimes. The tort liability regime – regime (3) – was defined as one in which “[t]here are no laws or rules against” the risky activities, and instead only a risk of liability. I gather the authors meant by this language to convey to participants that, in this regime, risky activities generated possible tort liability, but no risk of criminal punishment.

My worry is that this language could just as easily be interpreted to specify a regime in which liability is completely divorced from wrongdoing. And if respondents understood regime (3) in this way, then it is perhaps not surprising that the threat of liability had no deterrent effect.  Imagine a legal system in which the courts make clear that it is perfectly acceptable, as far as the law is concerned, to engage in certain risky activities, so long as one is prepared to pay for any resulting harm.  By deeming such activities acceptable, the law might not only fail to deter them, it might actually encourage them, in part by undermining social norms against them. Perhaps, then, the survey (inadvertently) assessed the deterrent effect of a liability regime that, instead of enjoining certain risky activities, condoned them (or deemed them permissible). If so, it probably won’t tell us much about the deterrent effect of the tort law we actually have.

Indeed, the study’s findings instead may help to explain why tort law overwhelming ‘speaks’ in a different voice from the one ascribed to it in the survey. Tort is a law of wrongs, not a law of permissions and fees. When liability attaches to risky conduct that causes injury, it attaches precisely because there are “laws and rules against” such conduct. Even when not a crime, careless driving that injures another is against the legal rules – the rules of tort law.  Specifically, the tort of negligence contains within it a rule that imposes on each of us a legal obligation to refrain from driving carelessly.

One might defend the survey’s framing of tort liability as disconnected from a notion of rule-breaking or law-breaking on the ground that legal economists – among the leading proponents of deterrence theory – tend to adopt the view that tort liabilities are mere fees. But then the question posed by the study is not whether tort law deters; it is whether tort law understood on a liability-rule model deters. That is a different question.

Download PDF
Cite as: John C.P. Goldberg, Against the Law?, JOTWELL (January 18, 2013) (reviewing W. Jonathan Cardi, Randy Penfield & Albert Yoon, Does Tort Law Deter?, Wake Forest Univ. Legal Studies Paper No. 1851383 (2011), available at SSRN),