On the imputation of rental prices to owner-occupied housing

This paper challenges the usual objections to the possibility of applying the rental equivalent approach to determine the weight that nonrental housing services should have in the Consumer Price Index (CPI). Using data from two Spanish household budget surveys, it is shown that market rents can be well represented in terms of an index of housing quality, two geographical variables, and the year of occupancy. This parsimonious empirical model is used to impute a rental value to nomental housing units, taking into account the possible selection bias induced by systematic differences in housing characteristics between the market rental sector and the nonrental stock. On average, the estimated hedonic values are relatively close to the self-imputations provided in the household surveys by the occupants of such dwellings. Therefore, using either of the two alternatives to assess the importance of nonrental housing services in the CPI have small consequences for inflation. Instead, dropping these services from the CPI creates a downward bias in the measurement of inflation of 0.33 percentage points per year during 1985-1992, and an upward bias of 0.38 percentage points per year during 1993 to 2000.