Direct Income and Substitution Effects in Participation Decisions

Abstract

The work of Jacob Mincer (1962) and others during the 1960s brought to light the existence of a positive correlation between female real wage rates and their labor force participation. This stood in contrast to earlier findings of a negative correlation between the same two variables for adult males. Following traditional labor-leisure theory, these correlations may be interpreted as a triumph of the substitution effect over the income effect for females, while the relative strength of the two effects is reversed in the case of adult males. This interpretation now appears to have acquired the seal of approval of labor economics textbooks.1 This dichotomy of effects has no doubt been strengthened by Mincer's persuasive emphasis on the wider area of substitution possibilities for females than for males. The female has an alternative, of home work in place of market work, which for cultural and biological reasons has been denied to males. The purpose of this brief note is to warn against casual interpretations of Mincer's stress on the home-market work alternatives and, in particular, against acceptance of the view that the dominant effect generally reverses itself with a change of sex. On such a view it would be difficult to explain: (a) historically declining hours of work of the subgroup of females in the labor force; (b) positive earnings participation coefficients for some groups of males (Bowen and Finegan 1969); and (c) historically rising participation rates for males age twenty-five to sixty-four, at least in the United States (Mincer 1968, p. 474). These puzzles are easily resolved, however, when it is realized that the size of an income effect, for males or females, will vary, ceteris paribus, with the hours of work already undertaken (or contemplated in the case of lifetime decisions). The greater the proportion of time spent in the labor