Abstract Most models of the rental housing market assume a close link between the level of residential rents and the after-tax user cost of rental housing capital. However, little empirical evidence exists to establish the strength of this link or the speed with which rents adjust to changes in user cost or tax policy. This paper develops and estimates an econometric model of the rental housing market to shed light on both of these issues. U.S. annual data for 1964–1993 are used to generate two-stage least-squares estimates of a four-equation structural model. Although the results are generally consistent with expectations and reveal several interesting relationships among the system variables, the estimates fail to identify a strong relationship between rent and user cost. About half of any increase in user cost is ultimately passed along as higher rent. The adjustment process also takes a long time, with only about a third of the long-run effect realized within 10 years of a user cost shock. The fundamental reason for this result is that our estimate of the user cost series, based on widely accepted procedures, is much more volatile than the residential rent series.