Macro-Economic Influences on the Distribution of Income: A Case Study of Canada

Abstract

<p>The purpose of this thesis is to estimate the effects of macro-economic variables, such as unemployment, inflation and labor force participation rates, on the size distribution of income in Canada. Most previous studies of this kind have been based on U.S. data. These include studies by T. Paul Schultz Lester Thurow, Charles Metcalf, Charles Beach, Alan Blinder and Howard Esaki, and others. Recently, however, a study in the Canadian context has been published by Adolf Buse, in which he analyzes the distribution of incomes of individuals, based on income tax returns. In our study, in addition to the income distribution for all individuals, we analyze a number of other distributions. These includes ones for all male individuals combined, particular age groups of male individuals, all families and unattached individuals combined, all families only, and families with age of head in a particular age group. There are two sources of distribution data: Taxation Statistics of Revenue Canada and the Survey of Consumer Finances of Statistics Canada. Data for family distributions are available only from the latter source. Our framework of analysis is highly disaggregative. Changes in income inequality are characterized by fluctuations in decile income levels relative to the mean of the distribution. The effects of macro-economic variables on these relative decile levels are estimated by econometric methods. The problem of multicollinearity is handled by using approaches based on principal components, including an approach suggested recently by Yair Mundlak. The responses of various points on the distribution to changes in the macro-economic variables are examined first. The decile model results are then analyzed in terms of single-valued measures of inequality such as the coefficient of variation, the Gini coefficient and the Atkinson index. The results indicate that the effects of unemployment are disequalizing, but this is not true for all the distributions. The effects of inflation on inequality are not very clear in most cases. On the other hand, the effects of the unanticipated component of inflation, as approximated by changes in the inflation rate appear to be disequalizing. There are some interesting results relating to the effects of labor force participation rates. Unemployment insurance benefits appear to improve the relative position of lower income groups. Evidence seems to suggest that lower income groups in the age range 65 and over gain more from retirement and old age benefits than do other groups in that age range.</p>