Directorates
·	Director General's Office
·	Data Development
·	Strategic Evaluations
·	Partnership and Continuous Evaluations
Publications
·	Evaluation Reports
·	Evaluation Briefs
·	Labour Market Development Agreement Series
·	Lessons Learned Series
·	Evaluation Tool Kit Series
·	Evaluation Forum Newsletter
·	From Theory To Practice

The main focus of this report is on relative income and financial dependence in couples, considered as socio-economic factors that may affect the standards of living, social interactions and self-esteem of wives. In addition, as was noted in Section 2.3, it is possible that low relative incomes and financial dependence among wives might affect the level of their spending on other family members, especially children. Since the well-being of children is an important policy issue, some observations on this point are needed.

The 1997 Canadian Out of Employment Panel (COEP) Survey included questions on household expenditures, which were asked of four cohorts. Unfortunately, none of the information collected was specific to children. It is, therefore, not possible to say with any precision how much money is spent on children in particular types of families. However, it is possible to describe expenditures on the basic necessities of food, clothing and shelter for the household as a whole.

Table A-1 and Table A-2 report the mean amounts spent on food and clothing per capita, as well as the mean housing affordability index, for different families. ³⁰ The housing affordability index is a measure of the degree of difficulty that a household has in paying for shelter. It also indicates the extent to which the magnitude of shelter costs limits the amount of money available to be spent on other household needs. Housing affordability is calculated here as the proportion of household income that is spent on rent, mortgage, property taxes and condominium fees. The mean score on the housing affordability index for the 1997 COEP couples subsample is 19.3 percent. A higher score is therefore indicative of above-average housing affordability problems. Any household that spends more than 30 percent of total income on shelter costs is conventionally assumed to be in great financial difficulty.

Families with children have greater housing affordability problems than families without children, and they also spend less per capita on food and clothing. In general, household spending patterns are more closely related to the presence or absence of children than they are to the distribution of incomes between spouses.

Housing affordability seems to be unrelated to the relative incomes of wives, but it does have a small relationship with the relative incomes of husbands (r = -0.15). ³¹ Households of husbands who earn less than half the household income after an employment separation have a little more difficulty paying their shelter costs than do comparable households (see Table A-1).

A somewhat higher correlation exists between personal expenses coverage (for which negative values imply financial dependence) and the housing affordability index (r = -0.30). The relationship is stronger among husbands (r = -0.40) than it is among wives (r =-0.21). The households of financially dependent husbands with children spent an average of 43.6 percent of household income on shelter costs (see Table A-2). However, despite their obvious economic stress, ³² food expenditures per capita in these households were not noticeably depressed.

Food expenditures per capita, and clothing expenditures per capita, do not appear to be affected much by degrees of relative income or by extent of personal expenses coverage, among wives or husbands. This is demonstrated by very low zero-order correlation coefficients. ³³ Nevertheless, it should be pointed out that the households of dependent wives with children spend less per capita on food and clothing than any other household type (see Table A-2).

Finally, the amount of a wife’s personal income is found to be unrelated to the amounts spent on food and clothing per capita in households with children. Although the zero-order correlation coefficients suggest a small positive relationship (+0.13 for food and +0.18 for clothing), multivariate analysis shows that there is no relationship when other factors are controlled (see Table A-3). Amount of household income is a factor influencing food and clothing expenditures in the families of wives with children, but the amount of the wife’s personal income is not. The evidence from this study suggests that income transfer programs that target payments on low-income couples with children will have a modest positive impact upon children’s well-being, regardless of whether the transfer payment is received by the wife or not.

Footnotes

30	COEP respondents’ estimates of food and clothing expenditures by their households tended to be rounded to the nearest $50.00 or $100.00, producing clusters of cases at particular points in the frequency distributions for these variables. The mean, rather than the median, is therefore used as the measure of central tendency here, since the median can be unduly sensitive to such clustering when the number of cases is small. The main difficulty with using the mean as the measure of central tendency is that it can be distorted by a few extreme values, or outliers. Therefore, upper and lower limits were set to the ranges of values, by top-coding and bottom-coding the data at two standard deviations above or below the mean for the 1997 COEP couples subsample. The same data adjustment procedure was used when calculating all correlation and regression statistics.
31	In Appendix A, relative income refers to the respondent’s personal portion of household income at the time of interview, that is, after his or her employment break
32	In declining order of severity: 20.3 percent of financially dependent husbands said that their lives were very stressful: 12.5 percent of wives who were not financially dependent said they had very stressful lives; 8.7 percent of financially dependent wives reported having very stressful lives; and 8.1 percent of husbands who were not financially dependent said their lives were very stressful.
33	The low correlations referred to here seem to be consistent with Martin Browning’s (1998) observation that the links between personal income, household income and household expenditure are relatively weak.