Decomposing Canada's Growing Housing Affordability Problem: Do City Differences Matter?

Abstract

This article examines the role of eight factors that affect the prevalence and incidence of housing affordability problems: geography, demography, migration/immigration/ ethnicity, income recipients, income source, employment and education. It develops bivariate probit models that use the 1991 and 1996 Canadian census public use microdata to predict the joint probability that a household spends more than half of its income on housing and that its income is below the poverty line. The conclusions show that city and regional differences are negligible after the effects of the factors common to all the cities have been accounted for. Changing employment levels and sources of household income are the most important factors explaining the prevalence and growth of housing poverty. While single parents have the highest incidence, the growth of the problem is mostly in the young non-family households. Migration, immigration and ethnicity play a role that is independent of the other factors. Education has almost no effect. The changes are in the underlying structures and in the variable profiles that differ remarkably across the factors. There were minor adjustments in the demographic and occupational profiles that would tend to reduce the problem but these are unlikely to stem its growth in the foreseeable future.