The Congressional Budget Office (CBO) is required to prepare economic projections for the Congress twice a year, and the housing market has been—and will continue to be—an important factor in that outlook. Over the past two years, starts of new homes have fallen sharply, and the resulting decline in real residential construction over that period subtracted an average of 1.0 percentage point from the growth rate of real gross domestic product. Looking forward, several alternative paths for residential construction are possible, ranging from a fairly quick turnaround to a severe slump that lasts several years.

This background paper examines the various factors that have determined the number of housing starts in the United States in the past and will continue to determine it in the future. Those factors include the underlying demand for new housing units, especially the role of demographics; cyclical and financial conditions, such as unemployment rates and lending standards; and the number of excess vacant units. CBO expects that housing starts will fall far enough below underlying demand for a long enough period to eliminate the current glut of vacant units and any temporary shortfall of demand due to adverse cyclical and financial conditions; this paper presents three alternative scenarios that could achieve that outcome. In keeping with CBO’s mandate to provide objective, nonpartisan analysis, this paper makes no policy recommendations.