I model global imbalances as arising from changes in preferences for housing relative to tradable goods. The key ingredients in the model are labor reallocation across sectors and consumption smoothing between housing and tradable goods. Countries import goods during periods when more domestic labor is devoted to housing construction. Housing booms are larger in countries that can run trade de…cits. This occurs despite the absence of wealth e¤ects, and even if trade is not primarily concentrated in capital goods. I provide several types of evidence to support the theory. First, over the last decade housing variables have decoupled from the business cycle while durable and total consumption expenditures have not. Second, for the same period there has been a strong cross-country correlation between housing variables and current account dynamics. Third, in a parameterized version of the model, housing demand shocks that match the cross-country dynamics of housing quantities generate current account dynamics matching recent global imbalances. Fourth, I use sign restrictions implied by the model to estimate a vector autoregression and identify the e¤ects of housing shocks on the U.S. trade de…cit. The results suggest that housing shocks are an important driving force of current account dynamics.