Did California’s Medi-Cal Expansion Increase Health Insurance Coverage Among Young Adult Noncitizens?

Leonard and Lipton (2026) examine whether California’s 2020 expansion of full-scope Medi-Cal increased health insurance coverage among noncitizens ages 19 to 25, regardless of immigration status. They analyze 2016–2022 American Community Survey data and compare young California noncitizens with older noncitizens in California and similar age groups in six other states. Using a triple-difference design, they find that the expansion increased Medicaid coverage by 4.2 percentage points and any insurance coverage by 3.5 percentage points. Among newly eligible young adults, they estimate gains of 24.4 percentage points in Medi-Cal coverage and 20.3 percentage points in overall coverage.

Why This Article Was Selected for The Policy Scientist

This article addresses a broadly important policy question: whether extending public insurance to populations excluded by immigration rules materially changes access to coverage. The issue is timely because several states have recently expanded, paused, or reconsidered such programs amid rising enrollment and fiscal pressure. Published in JAMA Network Open, a prominent peer-reviewed medical journal, the study adds credible evidence to a limited literature on coverage expansions for undocumented adults. Lipton has previously published related research on immigrant insurance expansions. The American Community Survey provides a large, representative data set, although self-reported coverage and the use of noncitizenship as a proxy for undocumented status introduce measurement limits. The triple-difference design is a strong causal-inference approach.

Full Citation and Link to Article

Leonard, R., & Lipton, B. J. (2026). The California 2020 Medi-Cal expansion to young adults and coverage among noncitizens. JAMA Network Open, 9(5). https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2848982

I

Central Research Question

The study examines whether California’s 2020 expansion of full-scope Medi-Cal increased health insurance coverage among noncitizens ages 19 to 25. Before the expansion, undocumented young adults in this age range could generally receive only restricted-scope Medi-Cal, which covered emergency services but not the broader set of benefits available through full-scope coverage. Beginning January 1, 2020, California extended full-scope Medi-Cal eligibility to income-qualified young adults regardless of immigration status.

The authors ask whether this policy increased Medicaid enrollment and overall insurance coverage among the affected population. They also examine whether the expansion changed private insurance coverage, which helps distinguish between reductions in uninsurance and substitution from private plans to public coverage. In addition, they test whether the estimated effects differed by race and ethnicity, sex, and age. The broader objective is to determine whether extending public insurance eligibility to undocumented young adults produced measurable gains in coverage during the first several years of implementation.

Previous Literature

The study builds on a substantial literature documenting that undocumented immigrants have much higher uninsured rates than United States citizens. Prior research attributes much of this disparity to legal restrictions on Medicaid, Medicare, and subsidized Marketplace coverage, as well as lower rates of employer-sponsored insurance. Because most undocumented immigrants are Hispanic or Asian, citizenship-based coverage restrictions also contribute to racial and ethnic differences in health insurance access.

Several earlier studies examined insurance expansions for pregnant immigrants and children. Wherry, Fabi, Schickedanz, and Saloner found that state and federal coverage for pregnant immigrants increased prenatal care, although they did not detect corresponding changes in infant health. Miller, Wherry, and Aldana analyzed a large prenatal coverage intervention for undocumented immigrants and contributed additional evidence on the effects of extending public insurance during pregnancy.

Research on children has generally found that expanding eligibility improves coverage and access to care. Lipton, Nguyen, and Schiaffino studied California’s Health4All Kids expansion and found increased insurance coverage among low-income noncitizen children. Cousineau, Stevens, and Pickering examined preventable hospitalizations following children’s insurance initiatives in California counties. Howell, Trenholm, Dubay, Hughes, and Hill evaluated new coverage for undocumented and other low-income children in three California counties. Rosenberg, Shabanova, McCollum, and Sharifi also studied insurance and health care outcomes in regions where undocumented children were eligible for Medicaid.

The literature on nonpregnant undocumented adults is much smaller. This limitation is important because adult expansions may operate differently from programs directed toward children or pregnant patients. Adults face different employment conditions, insurance alternatives, health needs, and administrative barriers. The present study therefore extends earlier research by evaluating an age-specific expansion for young adults and by using the timing and eligibility structure of the policy to estimate its association with coverage outcomes.

The authors also draw on research concerning the Affordable Care Act’s dependent-coverage provision. Chatterji, Liu, and Yörük examined effects on geographic mobility and living arrangements. Hahn and Yang studied employment decisions, while Blascak and Mikhed analyzed financial distress. These studies demonstrate that insurance eligibility can affect behavior beyond enrollment itself, although the current article focuses primarily on coverage.

Data

The principal data source is the 2016–2022 American Community Survey. The survey is conducted continuously and represents approximately 1 percent of the United States population each year. It contains demographic, economic, social, and health insurance information and supports comparisons across states, age groups, and years.

The analysis includes noncitizens ages 19 to 32 in California and six comparison states: Arizona, Florida, Illinois, Nevada, New York, and Texas. The comparison states were selected because each contained at least one county with an undocumented population of 100,000 or more and had not adopted a similar coverage expansion by the end of 2022.

The principal treatment group consists of California noncitizens ages 19 to 25. California noncitizens ages 26 to 32 serve as one comparison group because they were not yet eligible for the age-based expansion during the study period. Noncitizens in both age groups from the six other states provide additional comparisons.

The final sample contains 19,773 California noncitizens ages 19 to 25 and 32,515 California noncitizens ages 26 to 32. It also includes 28,535 noncitizens ages 19 to 25 and 43,213 ages 26 to 32 in the comparison states. Survey weights were used to produce population-representative estimates.

The outcomes are any health insurance coverage, Medicaid coverage, and private insurance coverage. The study excludes 2020 from the main analysis because the pandemic disrupted American Community Survey data collection and created significant nonresponse concerns. The prepolicy period therefore covers 2016 through 2019, while the postpolicy period covers 2021 and 2022.

The authors supplement the survey evidence with administrative enrollment data from the California Department of Health Care Services. Those records report monthly enrollment under the young-adult expansion and provide an external reference point for the self-reported survey measures.

Methods

The primary method is a triple-difference design. This approach compares changes in coverage among California noncitizens ages 19 to 25 with changes among California noncitizens ages 26 to 32, while also comparing those age-based differences with corresponding changes in the six comparison states.

The coefficient of interest is the interaction among three indicators: residence in California, observation after the expansion, and membership in the 19-to-25 age group. This structure is intended to isolate changes specific to the population newly eligible for full-scope Medi-Cal.

The regression models include single-year age indicators, year fixed effects, and public use microdata area fixed effects. The geographic fixed effects account for stable local differences, including health care infrastructure and immigrant-serving resources. Year fixed effects account for national or broadly shared changes over time. The models also include all two-way interactions among state, age group, and postpolicy status.

Additional controls include sex, race and ethnicity, and marital status. Regressions are estimated using ordinary least squares with American Community Survey weights. Standard errors are clustered at the public use microdata area level.

The authors conduct event-study analyses to evaluate whether the treatment and comparison groups followed similar trends before the expansion. They also perform sensitivity tests that include or exclude different pandemic years, impose alternative insurance hierarchies, add education controls, remove demographic controls, restrict the sample by income or local disadvantage, and estimate a California-only difference-in-differences model.

Because the survey does not directly identify undocumented status, the analysis uses noncitizenship as a proxy. The authors then conduct a post hoc calculation to approximate the effects among undocumented, income-eligible young adults specifically.

Findings/Size Effects

The expansion was associated with a 4.2-percentage-point increase in Medicaid coverage among California noncitizens ages 19 to 25. The 95 percent confidence interval ranged from 1.3 to 7.1 percentage points. Relative to the group’s baseline Medicaid coverage rate of 27 percent, this represents an increase of approximately 15.5 percent.

The policy was also associated with a 3.5-percentage-point increase in any insurance coverage, with a 95 percent confidence interval from 0.2 to 6.8 percentage points. Relative to the baseline overall coverage rate of 73.7 percent, this amounts to an increase of approximately 4.7 percent.

The estimate for private insurance was close to zero and statistically insignificant. This pattern suggests that the Medi-Cal gains primarily reflected reductions in uninsurance rather than large-scale replacement of private coverage.

Subgroup results indicate a 6.7-percentage-point increase in Medicaid coverage among Hispanic young adults. Medicaid coverage increased by 3.6 percentage points among males and 5.0 points among females. The estimated increases were 4.4 points for those ages 19 to 22 and 4.0 points for those ages 23 to 25. Differences between the male and female estimates and between the two age groups were not statistically significant.

In post hoc calculations focused on the newly eligible population, the authors estimate a 24.4-percentage-point increase in Medi-Cal coverage and a 20.3-percentage-point increase in any coverage. They translate these estimates into approximately 30,665 additional young adults enrolled in Medi-Cal and 25,554 additional young adults with some form of insurance.

Administrative data show that enrollment under the expansion reached 122,365 young adults by the end of 2022. The discrepancy between this figure and the survey-based estimate may result partly from the inability of the survey to distinguish restricted-scope from full-scope Medi-Cal.

The event-study results generally do not show differential prepolicy trends, supporting the design’s central identifying assumption. Most sensitivity analyses produce qualitatively similar findings, although some estimates become less precise under narrower sample restrictions.

Conclusion

California’s 2020 expansion of full-scope Medi-Cal was associated with significant increases in both Medicaid enrollment and overall insurance coverage among noncitizens ages 19 to 25. The absence of a significant decline in private insurance indicates that the policy primarily expanded coverage among people who were previously uninsured or who had only restricted emergency coverage.

The findings provide evidence that immigration-inclusive public insurance expansions can produce measurable coverage gains among young adults. At the same time, several limitations qualify the conclusions. The American Community Survey relies on self-reported insurance information, does not directly identify undocumented immigrants, and cannot distinguish restricted-scope from full-scope Medi-Cal. The observational design also cannot eliminate every possible source of confounding.

Generalization to other states should be cautious because California has an unusually large immigrant population, extensive safety-net institutions, and substantial experience administering immigrant-focused programs. Nevertheless, the study contributes evidence from a large population survey, a clearly defined policy change, multiple comparison groups, administrative enrollment records, and extensive sensitivity analyses. Further research is needed to determine whether the coverage gains translated into improved access to preventive care, reduced emergency department use, better health outcomes, or changes in public expenditures over time.