We use a probit model to estimate the following equation:
(1) yi = β0 + β1veteran + β2veteran*post +X' β3+ δt + ζs + ζst + ε
The various dependent variables, yi, include indicators for wives’ labor
supply outcomes including not working, self-employed, working part-time
conditional on working, hours worked last week, weekly hours worked
conditional on working any hours, weekly earnings, and ln(weekly earnings).11
The variable not working is 0 if the wife is employed and 1 otherwise. The part-
time variable reported is coded as 1 if the number of weekly hours worked is
between 0 and 35 hours, and 0 if the individual works more than 35 hours. Self-
employed is an indicator that is equal to 1 if the class of worker is self-employed
(either incorporated or not incorporated) and 0 otherwise.
Among the independent variables, veteran is a dummy equal to 1 if the
husband has been honorably discharged from active military duty, post is a
dummy equal to 1 in the post-policy period, X is a vector of the wife’s individual
characteristics including age, race, education, and indicators for employer-
11 Weekly earnings are earnings during a usual work week. This question is limited only to
respondents in their fourth and eighth months of the survey, greatly reducing sample size. For the
weekly earnings outcomes we code respondents in these months who did not have positive
earnings as having zero earnings. Hourly earnings are constructed from weekly earnings and are
available from the authors. We present weekly earnings because the results for hourly wages are
similar to those for weekly earnings but, as a created variable, introduce more measurement error,
and it is more problematic dealing with top-coding. Direct hourly earnings are available only for
the subset of the sample that earns an hourly wage.
13