This study seeks to explain the observed differences in the earnings of individual Canadians by sex. The study uses data from the micro data file of the 1986 Survey of Consumer Finances of individuals age 15 and over, with and without income. To a large extent, the study follows the examples presented in other Canadian studies conducted by Holmes (1974), Robb (1978), Gunderson (1980), Goyder (1981) and Ornstein (1983). Employment earnings account for an overwhelming proportion of the total income received by individuals. Thus, the examination of the earnings differential attempts to address the root causes of many of the problems faced by nontraditional families. Canadian society is no longer largely composed of the traditional family with a working father and the homemaking mother. The growing number of dual-earner couples, single and childless adults, and households headed by women presents a difficult challenge for social policy. The male-female earnings disparity is a key component in exacerbating problems that include the availability of credit for women, the feminization of poverty, access to affordable and adequate housing, and adequate incomes for retirement. To effectively address the problems that have resulted from the interaction of greater female participation in the labour force and the formation of alternate household types, planners and policy makers need to address the root problem of sexual inequality in the labour force, and not solely the symptoms. In the context of changing family structure and the economic position of women, the focus of this study is to identify the size of the male-female earnings gap, and to determine the extent to which the earnings gap can be explained by personal, work and productivity-related characteristics. The impact of these factors are analyzed from two points of view. First, the impact of individual factors on the level of earnings are analyzed through a simple comparison of mean earnings of men and women across a variety of characteristics. Second, the influence of these factors on earnings, and the degree of inequality between the earnings of men and women, is analyzed using multiple linear regression analysis. Regression analysis is used to estimate separate earnings equations for men and women. From the separate earnings equations, the wage gap can be partitioned into three parts, due to differences in (1) constant terms, (2) mean levels of the independent variables, and (3) the returns of the independent variables. Further, to assess the impact of occupational and industrial segregation on the earnings gap, a second set of earnings equations are calculated that do not include measures of occupational and industrial segregation. The calculations of separate earnings equations for men and women, for the selected sample, produced an unadjusted earnings ratio of 0.66. After adjustments were made for the ten productivity and productivity-related factors considered in the analysis, including occupational and industrial distributions, the ratio increased to 0.79. This left an earnings gap of $5,985 (1985 dollars) that could not be assigned to any of the measured variables. While part of the unexplained residual may be explained by variables not included in the analysis, or by more careful measurement of existing variables, it seems likely that at least 20 percentage points of the earnings gap is attributable to "an amalgam of different forms of discrimination which, taken together, disadvantage women relative to men", (Denton and Hunter, 1982). Discrimination is defined as different returns in earnings for equal productivity characteristics, as given by the regression coefficients. Of the total earnings gap of 34 percent, approximately 60% of this is attributable to wage discrimination, and approximately 40% is due to differences in productivity-related characteristics Occupational and industrial segregation account for a large proportion of the earnings gap. The adjusted earnings ratio, when occupational and industrial segregation are not considered endowments, is 0.69. Thus, the difference between the full-regression equation and the partial regression equation indicates that occupational and industrial segregation accounts for approximately 30% of the earnings gap. / Applied Science, Faculty of / Community and Regional Planning (SCARP), School of / Graduate

No description available.

No description available.