Causation and Standard of Proof From an Economic Perspective

Causation is a problematic notion, as explained by Ronald Coase regarding the “bilateral nature” of externalities. However, causation has played only a minor role in standard economic models of civil liability. An exception is the sub-literature on Uncertainty Over Causation and the Determination of Civil Liability, the benchmark paper written by Steven Shavell in 1985: “. . . the familiar notion that for parties to be led to reduce accident risks appropriately, they should generally face probability-discounted or ‘expected’ liability equal to the increase in expected losses that they create. This, of course, is naturally the case in the absence of uncertainty over causation, for parties then face liability if and only if they cause losses.” Thus, difficulties would seem to arise only when causation is uncertain.

This paper looks at situations where causation is unambiguous, i.e., there is no “uncertainty over causation”. However, we focus on situations where the tort system may provide sub-optimal incentives because of (i) limited liability problems (judgment-proofness), and (ii) other sources of uncertainty, for example, about the injurer’s actual level of care. We ask whether information about causation then plays a useful role in assigning liability.