This paper investigates the influence of wives' earnings on the distribution of family earnings. In the process, some differences in the manner in which family earnings are distributed within racial groups are highlighted. Earnings of wives equalize income distributions in white families but increase dispersion among blacks. Because they have conflicting effects, covariances between spouses in their wage rates and labor supply are isolated. Male and female wage functions are adjusted for sample censoring to fill out the true population variances and covariances in wages across all families. Due to the larger positive correlation in wages of black spouses, black family earnings would be distributed more unequally even if all individuals worked the same amount. Our labor supply analysis indicates that white families attempt to stabilize family earnings with some family members increasing their labor supply in response to a decline in participation of other family members. This compensatory function of wives' earnings is much less prevalent in black families.

In the most frequently used microdata sets, over a quarter of all respondents now refuse to answer some questions about their incomes. The Census Bureau has dealt with this problem, which has been increasing in severity over time, by imputing incomes of non-respondents. Their imputation procedure, called the "hot deck," essentially matches nonrespondents with demographically similar donors. In this paper we evaluate the census imputation methodology and raise some questions. First, the census procedure is tied to commonality of events in the population rather than the more appropriate informational content of regressors. Clearly, the census procedure severely understates income in certain occupations. Because it is based on the apparently invalid assumption that income does not affect reporting propensities, it most likely understates average incomes as well.