Abstract
This article proposes a theoretical framework and more accurate methods for projecting the household growth component of estimates of housing needs. These estimates combine empirical evidence with normative assumptions about the quantity of housing expected with population growth. Recent California experience illustrates the theoretical and practical issues involved. Alternative empirical methods are used to model changes in per capita household formation and homeownership rates over time.
The results show great instability between 1960 and 2000 in the linkage between population and housing needs, casting doubt on which linkage to use for future projections. Past changes in housing growth are attributed to changing population composition and occupancy patterns for subgroups. Estimates based on a cohort method are lower than those using constant rates of housing consumption and conform much more closely to recent experience, but it may not be desirable to lock in the deficiencies of the past when projecting needs.