We document the numerical challenges we experienced estimating random-coefficient demand models as in Berry, Levinsohn, and Pakes (1995) using two well-known data sets and a thorough optimization design. The optimization algorithms often converge at points where the first- and second-order optimality conditions fail. There are also cases of convergence at local optima. On convergence, the variation in the values of the parameter estimates translates into variation in the models' economic predictions. Price elasticities and changes in consumer and producer welfare following hypothetical merger exercises vary at least by a factor of 2 and up to a factor of 5.

Climate policy has favored costly measures that implicitly or explicitly subsidize lowcarbon fuels.We simulate four transportation sector policies: cap and trade (CAT), ethanol subsidies, a renewable fuel standard (RFS), and a lowcarbon fuel standard. Our simulations confirm that alternatives to CAT are 2.5 to 4 times more costly but are amenable to adoption due to right-skewed distributions of gains. We analyze voting on the Waxman-Markey (WM) CAT bill. Conditional on a district’s CAT gains, a district’s RFS gains are negatively correlated with the likelihood of voting for WM. Our analysis supports campaign contributions as a partial mechanism.