OFFERS, NOT TAKE UP, DETERMINE HEALTH INSURANCE GAP BETWEEN IMMIGRANT AND NATIVE WORKERS

Native-born workers are 18 percentage points more likely to have health insurance coverage than immigrant workers. This difference largely reflects differences in the extent to which natives and immigrants have employer-sponsored insurance. In a study funded by the Economic Research Initiative on the Uninsured, Thomas Buchmueller, Anthony LoSasso, Ithai Lurie and Sarah Senesky investigate the underlying sources of this coverage gap, examining differences between natives and immigrants in the probability of working for an employer that offers health insurance, eligibility for offered coverage, and take-up conditional on eligibility. In all analyses they account for differences between citizen (naturalized) and non-citizen immigrants as well as differences by gender.

FINDINGS
Descriptive statistics

• Among adults residing in the U.S., 13.3% are immigrants. About 40% of those immigrants are naturalized citizens. Employment is similar across natives and non-natives (75.3% of native adults employed; 71.7% of non-native adults), with larger differences between native and non-native women (69.1% of native women and 57.9% of non-native women employed.)
• The native/immigrant insurance coverage gap is largely accounted for by differences in holding coverage from employers (84.5% of natives v. 66.1% for the foreign born hold coverage through employment.)

Effect of immigrant and citizenship status on employer-sponsored insurance (ESI)

• Among immigrants, there are large differences in coverage between naturalized citizens and non-citizens. The percentage of naturalized citizens with ESI is 6 points lower than the rate for natives, whereas the native/non-citizen gap is 26 percentage points.

• Between 40 and 50 percent of the gap in ESI coverage between natives and foreign-born adults is explained by observed demographic and human capital characteristics.
• The most important individual characteristic is educational attainment, which accounts for roughly 60% of the explained gap between natives and non-citizens.

Effect of immigrant status on offer, eligibility, and take up

• Differences in overall coverage can be decomposed into differences in employer offers, employee eligibility and employee take-up.
• Natives and naturalized citizens have similar probabilities of working for an employer that offers health insurance.
• In contrast, non-citizens are much less likely to work for a firm that offers ESI. The unadjusted gap between natives and non-citizens is 22.9 percentage points. As with the gap in coverage, adjusting for differences in individual and job characteristics reduces the native/non-citizen gap considerably (to 6.6 percentage points).
• Conditional on working for an employer that offers health insurance, nativity-related differences in eligibility are small and become statistically insignificant when personal and job characteristics are included in the model.
• Similarly, conditional on being eligible for coverage, differences in take-up are small and are not statistically significant after controlling for individual and job characteristics.

POLICY IMPLICATIONS
The lower rates of health insurance coverage among immigrants are not reflective of immigrant status per se. Rather, they reflect the demographic traits of the immigrants and low rates of health insurance coverage among jobs for less-skilled workers. Immigrants will benefit from general policies addressed to low income families and less-skilled workers.

METHODOLOGY
Probit regression models.

CAVEATS
The study is a descriptive study of associations between natives, and naturalized and non-naturalized citizens. The paper does not consider differences in selection into the labor force across natives, foreign-born citizens, and non-citizens, or, once in the labor market, for nativity and citizenship to effect job choice. While the evidence is consistent with the idea that non-citizens becoming naturalized citizens would increase their health insurance coverage, the paper does not show a causal connection between citizenship and health insurance.

Because the analysis is based on a single cross-section,, the paper cannot provide any insights on how duration of time in the US or country of origin affect coverage.

DATA SOURCE
Survey of Income and Program Participation, 2001 panel, data for 2002. Immigrant status is from wave 2 (July to October 2001) and employer offers are from wave 5 (July to October 2002.) Adults ages 18 - 64. 38,041 adults interviewed in both 2001 and 2002; 22,975 in the "employed" sample.

CITATION
Immigrants and Employer-Provided Health Insurance
Thomas Buchmueller, University of California, Irvine, Anthony T. Lo Sasso, University of Illinois at Chicago, Ithai Lurie, Northwestern University, and Sarah Senesky, Mathematica Policy Research

ERIU Working Paper #38

The final version of the paper is forthcoming in Health Services Research.

Back to top