Following popular discourse, we abuse economic terminology by defining the “housing shortage” in the United States as the difference between the number of homes that would be built in the absence of supply constraints and the actual number of homes. The magnitude of the housing shortage is important to policymakers, who use it to measure the scope of the housing supply problem and the extent to which proposed policies would solve it. However, previous studies understate the housing shortage because they estimate how many more homes would have been built if historical building or household formation trends prevailed today, even though historical trends were also affected by supply constraints. We are the first to use a supply and demand framework to estimate the full housing shortage in the United States. Using county-level data on land shares of home prices, we estimate that the U.S. housing shortage was 20.1 million homes in 2021, 14.1 percent of the national housing stock. Our housing shortage estimate is 4 to 5 times as large as previous estimates, and 13 times as high as the shortage cited by the White House to contextualize the effects of policies intended to close the gap. Consistent with predictions of economic theory, our estimated housing shortage is uniformly low in areas with low regulation but varies in areas with high regulation, since a housing shortage requires both stringent regulations and strong housing demand.