A bout 8 cents out of every dollar spent in the United States-including both spending by consumers on final goods and spending by firms on intermediate goods-is spent on imports, according to the World Input-Output Database (WIOD). What if, because of a wall or some other extreme policy intervention, these goods were to remain on the other side of the US border? How much would US consumers be willing to pay to prevent this hypothetical policy change from taking place? The answer to this question represents the welfare cost from autarky or, equivalently, the welfare gains from trade.There is little direct empirical evidence about the impact of autarky on prices and quantities. The Jeffersonian trade embargo at the beginning of the nineteenth century is one rare exception (Irwin 2005). It is not clear, however, what such historical evidence can tell us about the magnitude of US gains from trade today. In order to make progress on this important issue, we therefore propose an alternative strategy combining both theory and empirics.Our strategy is based on two observations. First, when countries exchange goods, it is as if they were indirectly exchanging the factor services embodied in the production of these goods: unskilled labor, skilled labor, physical capital, land, and