Methodological Rift Over California's $20 Fast-Food Wage: Job Losses vs. Null Employment Effects

Methodological Rift Over California's $20 Fast-Food Wage: Job Losses vs. Null Employment Effects

The academic and policy debate over the employment impacts of California’s $20 fast-food minimum wage (AB 1228, effective April 1, 2024) is anchored in a deep methodological rift. While early studies suggested minimal to no disemployment effects, rigorous academic research from late 2024 through 2026 has split into two competing empirical camps utilizing different control groups, population adjustments, and data sources.

The debate is best understood by looking at the four primary studies analyzing AB 1228's labor market impacts:

1. Clemens, Edwards, and Meer (NBER Working Paper 34033, 2025): Utilizing Quarterly Census of Employment and Wages (QCEW) data, this study estimated that the policy led to an 8% wage increase in California's fast-food sector relative to the rest of the country, but also triggered a statistically significant employment decline of over 3% (representing a median estimate of approximately 18,000 lost jobs). Their calculated own-wage elasticity of employment was -0.40, placing it in "medium negative" territory where workers received only about 60% of the potential earnings increase due to job losses.
2. Sosinskiy and Reich (UC Berkeley IRLE, 2025): Also utilizing QCEW data, this study estimated a similar wage increase of about 7% but found an employment decline of just under 1% that was statistically indistinguishable from zero (own-wage elasticity of -0.12).
3. Hamdi and Sovich (2025): Utilizing Equifax payroll data to compare fast-food establishments within large firms across different states, this study found a 12% wage increase and a statistically insignificant 2% employment increase (own-wage elasticity of +0.19).
4. Schneider, Harknett, and Bruey (2024): Utilizing worker survey data from The Shift Project, this study found that the policy raised wages in the first quarter after implementation with no effect on the usual number of hours worked.

The Methodological Rift: Why Do Similar Datasets Yield Different Results?

The divergence between the NBER study (Clemens et al.) and the UC Berkeley study (Sosinskiy & Reich) despite using similar QCEW data comes down to two critical methodological differences:

• Population Adjustments: Net immigration and population growth in California fell rapidly around the time of the policy's implementation. The UC Berkeley study controlled for these population changes, whereas the NBER study did not. Failing to adjust for a shrinking customer base and labor supply in California can cause an analysis to overstate employment declines and incorrectly attribute them to the wage floor.
• Comparison/Control Groups: Clemens et al. compared California's fast-food sector to other states and high-wage industries in California, but did not directly control for California's full-service restaurant sector. Interestingly, full-service restaurant employment in California (which was not covered by the $20 wage) also declined by 1.55% to 2.75% over the same period. By failing to account for this broader service-sector slowdown in California, the NBER study attributed the entire decline in fast-food employment to AB 1228.

Ultimately, when averaging across these rigorous studies, the own-wage elasticity of employment is approximately -0.11, which aligns with the historical median of restaurant minimum wage literature (such as Dube & Zipperer 2025) and suggests that the policy succeeded in raising aggregate earnings for the low-wage workforce with modest, manageable employment adjustments.