Notes

^1.

We are aware of only one previous study that uses repeated cross-sectional survey data to estimate the age profile of homeownership in Canada (Crossley and Ostrovsky 2003).

^2.

The Census of Population does not distinguish between household maintainers and homeowners. The household maintainers may not be the owners of the dwelling in which the household resides. In homes where adult children are living with their parents, it is possible that the children are the maintainers but not the owners. Looking at households according to the age of the maintainers may thus result in some bias.

^3.

The validity of synthetic cohort analysis rests on the assumption that the population composition of birth cohorts does not change over time (i.e., at different points of observation). In reality, the composition of cohorts may change when individuals with different characteristics join or leave a birth cohort at any of the different life stages. This may occur through immigration, emigration, or mortality. In future studies, we will address this issue through multivariate modeling (for example, see Crossley and Ostrovsky 2003). We did check whether immigration affects our estimates of homeownership age profiles across cohorts by computing rates for the Canadian-born only. The results are reported in Text table 1, Text table 2, and see first Chart in Appendix. There is little difference in the age and cohort profiles between the Canadian-born and the total population. This is because the Canadian- born account for over 80% of the total population and the age-adjusted difference in homeownership rates is small between immigrants and the Canadian-born. Immigrants tend to have a lower homeownership rate in the prime-working-age groups but a higher rate over age 55 than the Canadian-born. Over the period from 1971 to 2006, homeownership increased faster among the Canadian-born than among immigrants, primarily in the prime-working-age groups. As a result, the relative advantage of immigrants in homeownership diminished (see also Haan 2005).

^4.

Chiuri and Jappelli (2009) used the Statistics Canada Survey of Consumer Finance (1976 to 1998) and the Statistics Canada 2000 Survey of Labour and Income Dynamics. In order for us to follow birth cohorts up to age 80 using these data sets, individuals had to have been born before 1920.

^5.

The share of young adults aged 25 to 34 living with their parents decreased from 9.3% in 1971 to 8.3% in 1981.

^6.

Adult-equivalent-adjusted family income assumes that certain economies of scale accrue to people who live together in families and is calculated by dividing total family income by the square root of family size. Income quintiles based on adult-equivalent-adjusted family income are derived from all people (regardless whether they are household primary maintainers) within each broad age range.