The increase in part-time employment post-ADA found in Chapter 4 is worth exploring at the state level, as well. It is of interest since flexibility in hours may serve as an important mechanism through which employers can accommodate many types of disabilities. If this is the case, then requirements to accommodate workers' disabilities at the state level should result in similar adjustments as seen post-ADA. Figure 6.3 plots the average proportion across states of disabled and nondisabled workers that are employed part-time. The reference vertical line corresponds to the time when legislation was in place in each state. There appears to be an increase in the proportion of disabled workers that are employed part-time, as well as a modest divergence in the two series.

The pooled, cross-sectional analysis of Chapter 4 is repeated here in order to determine whether there is any significant growth in parttime employment among disabled workers, post-legislation, relative to

nondisabled workers. The model estimated is the bivariate probit with selection:

(6.3) emp is = a1 + yXu + p1 disable, + posts + 01 disable, X posts + e1is pt,s = a2 + iX + p2 disable,- + posts emp,, = 1 if person i in state s is employed, 0 otherwise, and pt,, = 1 if person , in state s is employed part-time and is not observed unless empb = 1. disable,- is equal to 1 if person i is disabled, 0 otherwise; posts is equal to 1 if the person is observed after passage of the state legislation; Xu and X2i include individual demographic characteristics; e1is and e2is are distributed as a bivariate normal with means equal to 0, variances equal to 1, and correlation equal to p.

Again, the coefficient in the part-time equation on the disable X post regressor is what tells us whether there is any change in the probability of part-time employment among disabled workers, post-legislation, relative to nondisabled workers. Table 6.4 details the regression results. Using the parameter estimates, the difference in the impact of having a work-limiting disability on part-time employment across the two time periods can be calculated by evaluating the probabilities of interest for each person, varying the disable and post dummy variables, then taking the difference between these probabilities and averaging the differences across the sample. This calculation translates the estimated coefficients into a 2-percentage-point greater probability of disabled workers being employed part-time than nondisabled workers, post-legislation relative to pre-legislation. This result is only significantly different from zero at the 85 percent confidence level, but it does provide some support for the notion that disability legislation, whether by the states or national, influences the hours of work of disabled workers.4 Also, as with the preceding wage analysis, these findings suggest that the impact of the ADA was dampened somewhat by the adjustments in hours that had already taken place as a result of the state-level legislation.

Was this article helpful?

## Post a comment