A calibrated model is used to determine the welfare impacts of various regulatory instruments for improving health. The results of a lab experiment are integrated in a partial equilibrium model representing demands for two kinds of fish, one with higher nutritional benefits (canned sardines) and one with higher contamination risks (canned tuna) in France. In the laboratory, information about health effects leads to a statistically significant decrease (increase) in the willingness to pay for tuna (sardines). Simulations with the laboratory results show that, for most cases, a per-unit tax on tuna and a per-unit subsidy on sardines without any information revealed to consumers lead to the highest welfare, because both the tax and subsidy directly internalize health characteristics. The information policy combined with a per-unit tax on tuna and a per-unit subsidy on sardines is socially profitable only if a large proportion of consumers (greater than 95%) receives health information.
Marette, Stéphan; Roosen, Jutta; and Blanchemanche, Sandrine, "Tax, Subsidy, and/or Information for Health: An Example from Fish Consumption" (2007). CARD Working Papers. 464.