Skip navigation to content
eriu: Economic Research Initiative on the Uninsured Initiating and disemminating research to spark new policy discussion on health coverage issues.
Fast Facts  
Funded Research Home

      Sort by Author (A-Z)
      Sort by Topic


      Costs of Health Insurance
      Demand for Health Insurance   
      Employment-Based Coverage
      Labor Markets
      Near Elderly Population
      Vulnerable Populations
      Welfare, Medicaid, and SCHIP

Find a Document

      Research Highlights
      Research Findings
      Working Papers
      Q & A with the Author
      Conversations with       Economists

Home > Funded Research Home > All > Sort by Author (A-Z) > Haas & Swartz

ERIU Funded Research Projects

Sort by: Author (A-Z) | Topic

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

Author: Haas, Jennifer ; Swartz, Katherine
Working Paper: The Effects of Worker, Firm, and Market Characteristics on Access to Employer Sponsored Health Insurance (PDF) ; October 2004

This paper analyzes how the characteristics of a worker, the firm where the worker is employed, and the local labor market are related to racial/ethnic disparities in employer-sponsored health insurance (ESI). The analyses are based on data from the annual surveys of the Medical Expenditure Panel Survey Household Component (MEPS-HC) for 1996 through 2000, which were merged with 2000 Census data about the characteristics of each individual’s county of residence. Adults 18 to 64 years old who were employed at least part-time for some portion of the year, and were not full-time students, and described their race/ ethnicity as white, African American or Latino were included in this analysis (n = 26,813). The county of residence was assumed to represent the local labor market in which an individual searches for a job. Two models were estimated to determine the relative effects of an individual’s characteristics, the firm’s characteristics, and the labor market’s characteristics on disparities in ESI coverage. The first model is of workers sorting to jobs that do or do not offer ESI. The second reflects the full range of possible outcomes: (1) ESI is offered and accepted, (2) ESI is offered and declined but person is insured elsewhere, (3) ESI is offered and declined and person is uninsured, (4) ESI is not offered but person is insured elsewhere, or (5) ESI is not offered and person is uninsured.

Approximately two-thirds of the workers in our sample were offered ESI. African Americans and Latinos were more likely than whites to be uninsured, either because they were not offered ESI or because they declined offered insurance. However, race/ethnicity was not associated with being offered ESI when characteristics of the worker, firm, and labor market are controlled for. Firm size and industry were significantly associated with all possible insurance outcomes. The percentage of county residents who were foreign-born and the percentage living below poverty were the only market characteristics associated with not being offered ESI and being uninsured. Individual characteristics explained almost two-thirds of the risk of not being offered ESI, while firm characteristics explained almost another third of the risk, and local labor market characteristics explained less than 1%.