New Data for Examining How a Local Minimum Wage Policy Affects Diverse Young Adults

Background and purpose: While city-level minimum wages now cover one in 12 American workers, data limitations impede research on the impact of these local measures. Conventional sources of economic data do not have sufficient small area sample sizes to identify the impacts of local policies, much less create evidence about how local wage mandates might affect subgroups of people who are particularly at risk of exclusion from labor market opportunities.

The purpose of this paper is to report on the development of a uniquely detailed population-level administrative data that will permit analysis about how a prominent city-level minimum wage policy affected young workers, workers of color, and young workers of color.

Methods: Our data are built from administrative records from six different state agencies and sub-agencies, including records for drivers’ licenses, voter registration, human services program participation, and employment covered by the Unemployment Insurance (UI) system. We describe the creation of a common identifier across data sources and explain how we captured and imputed ethnoracial information and other demographic identifiers. We then benchmark the resulting merged data to Census population estimates.

Findings: We find that we are able to replicate Census counts of workers to within 4%-7% depending on the sub-population. Relative to Census survey data, this merged administrative data captures more workers with annual earnings below $10,000 and fewer workers with earnings above $10,000. Our methods of augmenting UI records with demographic indicators generates slightly larger numbers of young workers (age 16-34) and Black workers than does the Census, while the Census estimates more non-Hispanic Asian and Hispanic workers.

Conclusions and implications: Combining administrative data across multiple sources is a promising approach to providing highly accurate and disaggregated data within a state. Such data allow researchers to answer policymakers’ and advocates’ calls for more group and intersectional analyses of labor market and economic outcomes. For instance, we will be able to use this data to examine employment among young Asian American and Native American/Alaska Native persons, groups often functionally excluded from probability-based samples. While using administrative data has limitations, we believe that it is a potentially fruitful method to conduct rich research on workers and firms and build evidence on workers’ labor market experiences and the effects of local labor policies.