Abstract
Domestic products have a disproportionately high market share on many goods markets. We examine the contribution of preferences to such “home bias”, using detailed data on wine sales in New Hampshire (weekly sales by brand by store for one year). In counterfactual simulations, where we use the same set of products as currently available, the U.S. market share falls from 58 percent to 38 percent if all country-of-origin effects are set equal. Home bias on this market is not explained by higher marginal costs for imports or by lesser store coverage of imported brands. The evidence rather points to higher foreign fixed costs of entry, coupled with a preference for U.S. wines, as the main sources for the high domestic market share.