Executive summary
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
The financial security of widowed and divorced women during their retirement years has long been a concern. This paper places this issue within the context of research on replacement rates, that is, the extent to which family income during the working years (here, the mid-50s) is replaced as individuals move into their late 70s. Using a longitudinal database and fixed-effects econometric models, the paper assesses the effect of widowhood or widowerhood and divorce after age 55 on replacement rates during the retirement years. This paper is not an analysis of the effect of widowhood or widowerhood or divorce on low income or financial well-being in general, but rather an analysis of their effect on income replacement rates in retirement. To address this issue, the Longitudinal Administrative Database (LAD) maintained at Statistics Canada is used to track the income trajectories of individuals who were aged 54 to 56 in 1983 over the subsequent 23 years, until they were aged 78 to 80, in 2007. Since this study is concerned with economic welfare, family income rather than individual income is used to estimate replacement rates. The income replacement rates of women and men who became widowed or divorced or separated during this period are compared to those of individuals who remained married over the period.
The results indicate that, among women, separation or divorce has a larger negative effect on replacement rates than does widowhood. The effect of divorce or separation is greatest among women from higher-income families, where there is more reliance on private- pension and investment income. The heavy reliance on public-pension income tends to reduce the effect of divorce on replacement rates for women in lower-income families. Moreover, it is possible that, because of changes in the labour force participation of women, the effect of divorce or widowhood may be reduced among more recent cohorts. To test this possibility, the research also focused on the cohort of women aged 54 to 56 in 1993, a "1993 cohort," rather than the earlier "1983 cohort." No support was found for the notion that the effect of widowhood or divorce on replacement rates was reduced in the 1993 cohort as compared to the earlier cohort. However, this does not mean that these results are necessarily generalizable to future cohorts of women, for the reasons mentioned.
Among men, separation or divorce had little effect on replacement rates. Widowerhood tended to increase replacement rates among men from middle- and higher-income families. This is likely because, although family income may fall after the death of the wife or female partner, economic requirements fall more. Hence, the family income, which is adjusted for changes in family size, would rise.
Finally, the adult-equivalent-adjustment scale selected to account for differences in family size and the economies of scale associated with larger families can influence the results. In a sensitivity analysis, the results indicate that using alternative scales does influence the magnitude of the effects; however, the basic findings hold in most cases.
- Date modified: