How stable and predictable are welfare estimates using recreation demand models?

Economic analyses of environmental policy projects typically use pre‐existing estimates of welfare measures that are then transferred over time to the policy relevant periods. Understanding how stable and predictable these welfare estimates are over time is important for applying them in policy. Yet, revealed preference models of recreation demand have received few temporal stability assessments compared to other nonmarket valuation methods. We use a large administrative panel dataset on campground reservations covering 10 years to study temporal stability and predictability of environmental quality welfare estimates. Welfare estimates are statistically different across years in 62% of the comparisons, and this ranges from 47%–71% depending on modeling assumptions. Using an event study design, we find evidence that week‐specific welfare estimates are stable after an initial adjustment week in response to a change in environmental quality. Our findings further reveal that using 2 years of data in the modeling compared to a single year improves the prediction of future welfare measure estimates substantially, but further prediction improvements are modest when including more than 2 years of data. Predictions of welfare estimates are more consistent when using data closer in time to the prediction year. We discuss the implications of our results for using revealed preference studies in policy analysis.