Citation
Myers, C. (2007). A cure for discrimination? Affirmative action and the case of California’s Proposition 209. Industrial and Labor Relations Review, 60(3), 379-396.
Highlights
- The study’s objective was to gauge the impact of California’s Proposition 209, a 1996 repeal of affirmative action, on employment, unemployment, labor force participation, and wages among women and racial minorities.
- The author used a nonexperimental design and estimated a difference-in-difference-in-differences regression model on monthly Current Population Survey data from 1994 to 2001.
- The study found that the employment rate of all minorities in California relative to white men dropped 2.8 percentage points from 1995 to 1999, a statistically significant decrease.
- The quality of causal evidence presented in this report is low because the author did not control for pre-intervention minority employment trends in California relative to the rest of the United States or preemptive hiring decisions that might have anticipated the law’s passage. This means we are not confident that the estimated effects are attributable to California’s affirmative action repeal through Proposition 209. Other factors are likely to have contributed.
Intervention Examined
California's Proposition 209
Features of the Intervention
In 1996, California repealed state affirmative action policies through Proposition 209, leaving federal affirmative action statutes, particularly those applicable to federal contractors, intact. The legislation went into effect in November 1997 after a prolonged legal challenge. Before that time, private and public agency employers in California had been subject to state fair employment practice laws and affirmative action requirements as well as federal legislation. Title VII of the Civil Rights Act of 1964 (the Civil Rights Act) forbids employers from discriminating against their employees on the basis of race, sex, color, religion, or national origin. Affirmative action Executive Order 11246, signed by President Lyndon B. Johnson in 1965, reinforced prohibitions in the Civil Rights Act and earlier executive orders, requiring the federal government and federal contractors with 50 or more employees or $50,000 in contracts to develop written affirmative action plans and ensure equal opportunities for all workers.
Features of the Study
The author estimated difference-in-difference-in-differences regression models using individual-level monthly Current Population Survey outgoing rotation group data from 1994 to 2001. Data from Washington State were not included in the analysis because the state enacted a comparable affirmative action ban in 1998. The analysis estimated impacts of the affirmative action repeal through Proposition 209 on white, African American, and Hispanic women and on African American and Hispanic men separately. The outcomes of interest were employment rates, unemployment rates, labor force nonparticipation rates, and wages in three post-intervention years (1999, 2000, and 2001). Impacts were identified as the difference between changes in outcomes for groups covered by affirmative action legislation relative to other groups in California and changes in outcomes for covered groups relative to other groups in all states except Washington. The regression models controlled for age, marital status, educational attainment, month of interview, central city location status, nativity, and citizenship status.
Findings
- The study found that the employment rate of all minorities in California relative to white men dropped 2.8 percentage points from 1995 to 1999, a statistically significant decrease.
- The employment rate of white women in California relative to white men fell significantly by 3.4 percentage points over this period, while the corresponding relative employment rate for Hispanic women in the state fell significantly by 4.8 percentage points.
- Labor force nonparticipation rates relative to white men in this period increased significantly by 2.9 percentage points for white women, 4.6 percentage points for African American women, and 5.2 percentage points for Hispanic women.
- Findings were similar for outcomes measured in 2000 and 2001.
Considerations for Interpreting the Findings
Although the difference-in-difference-in-differences regression model controlled for many sources of variation related to the outcomes of interest, several potential confounding factors remained. First, the author did not address the possibility that the employment effects attributed to Proposition 209 might also reflect trends already in motion among the demographic groups in California and elsewhere before the legislation was enacted. Second, the author noted that the governor of California issued an executive order related to affirmative action in 1995, the year from which pre-intervention data were drawn. Because this executive order could have affected employers’ hiring decisions, differences in minority employment from 1995 to 1999 might capture the consequences of both this executive order and Proposition 209, rather than Proposition 209 alone.
Finally, the author estimated many related impacts on outcomes related to employment, earnings, and labor force participation. Performing multiple statistical tests on related outcomes makes it more likely that some impacts will be found statistically significant purely by chance and not because they reflect program effectiveness. The author did not perform statistical adjustments to account for the multiple tests, so the number of statistically significant findings in these domains is likely to be overstated.
Causal Evidence Rating
The quality of causal evidence presented in this report is low because the author did not control for pre-intervention minority employment trends in California relative to the rest of the United States or preemptive hiring decisions that might have anticipated the law’s passage. This means we are not confident that the estimated effects are attributable to California’s affirmative action repeal through Proposition 209. Other factors are likely to have contributed.