Lesson 17 — SIL 1

install.packages("remotes")
remotes::install_github("reder206/ma206data")

library(tidyverse)

df <- ma206data::ada92 |>
  filter(age_range == "20-39", white == FALSE)

with_disability <- sum(df$disability == "Disability")
total <- length(df$disability)
phat <- with_disability / total

xbar <- mean(df$weeks_worked)

Summary of Consequences of Employment Protection? The Case of the Americans with Disabilities Act

Daron Acemoglu and Joshua Angrist (2001) study the labor market effects of the Americans with Disabilities Act (ADA), which prohibited discrimination and required employers to provide reasonable accommodations for disabled workers beginning in 1992. The law aimed to increase employment opportunities and earnings for disabled individuals, but it also created potential costs for employers in terms of accommodations and legal risk.

Using data from the 1988–1997 Current Population Survey, the authors compare employment and wages between disabled and nondisabled workers aged 21–58. They find that employment among younger disabled men and women (ages 21–39) fell sharply after the ADA’s implementation. The decline was especially large in medium-sized firms and in states that saw high levels of ADA-related litigation. For older disabled workers (40–58), the effects were smaller and mixed. In contrast, nondisabled workers’ employment was largely unchanged.

Wages for disabled workers did not display consistent changes across the study period. The authors also examine whether the growth of disability benefit rolls could explain the decline in employment but conclude that this factor accounts for only a portion of the change. The results suggest that the ADA itself, by raising the expected costs of employing disabled workers, contributed to the decline in their employment.

The study highlights that legislation intended to improve outcomes for disabled workers may have unintended consequences. While the ADA did not negatively affect nondisabled workers, its impact on the disabled population raises questions about how broadly such results apply beyond the specific groups and time period studied. The findings also raise important considerations for how research on sensitive policies can be interpreted and applied.