Expanding Insurance Coverage to Undocumented Immigrants in Connecticut

Background

There are roughly 113,000 undocumented immigrants living in the state of Connecticut, and almost 60 percent of this population lacks access to health insurance (Migration Policy Institute, undated). Although the Affordable Care Act (ACA) greatly expanded access to health insurance coverage in the United States, the undocumented population—excluded from these reforms—represents a disproportionate share of the uninsured population nationwide (Kaiser Family Foundation, 2021). In particular, undocumented immigrants are not eligible for Medicaid and cannot enroll in marketplace coverage or receive marketplace subsidies. Legally present immigrants are also at high risk for being uninsured (Kaiser Family Foundation, 2021), and legally present recent immigrants are ineligible for Medicaid until they have been legally present in the country for five years.

Policymakers in Connecticut are considering removing immigration status requirements from Medicaid and Children's Health Insurance Program (CHIP) eligibility, enabling both undocumented and legally present recent immigrants to enroll in these programs. The state would fully fund this expansion, with no federal matching. In addition, the state is considering a look-alike individual market plan, to be offered off-marketplace, that would allow undocumented immigrants to purchase coverage and receive state-funded advance premium tax credits (APTCs) and cost-sharing reductions (CSRs). State-funded APTCs on the look-alike plan would be available to people with incomes between 138 and 400 percent of the federal poverty level (FPL) with no other affordable source of insurance. State-funded CSRs on the look-alike plan would be available to people with incomes between 138 and 250 percent of FPL, but—unlike on the marketplace—the cost of these CSRs would not be loaded onto the silver plan.¹ As with CSRs available to legally present Connecticut residents, CSRs on the look-alike plan would be enhanced via Covered Connecticut, a 2020 state law that used state funding to enhance CSR amounts for residents with incomes below 175 percent of FPL. These reforms would build on 2021 state legislation that expanded Medicaid to undocumented immigrants who are pregnant or children through the age of eight years (Connecticut General Assembly, 2021) and prior legislation that expanded Medicaid eligibility to legally present recent immigrants who are pregnant or through the age of 20 (Division of Medicaid and Children's Health Operations, Boston Regional Office, 2014).

In this study, we use a microsimulation modeling approach to analyze the effects of such polices on health insurance enrollment and state spending in Connecticut. We consider three specific policy scenarios, driven by policy changes under consideration in the state. The specifics of each policy scenario are shown in Table 1. In Scenario 1, Medicaid eligibility is further expanded to the full population of otherwise-eligible undocumented and legally present recent immigrants. Scenarios 2 and 3 additionally expand marketplace subsidy eligibility to undocumented immigrants, with two different levels of income eligibility.

We also consider any unintended effects for U.S. citizens and legally present immigrants who have resided in the United States for more than five years.

Methods

We developed a Connecticut-specific version of the COMPARE microsimulation model (Cordova et al., 2013) to estimate how undocumented immigrants and legally present recent immigrants would respond to new health insurance enrollment options. Using COMPARE, we created a representation of the Connecticut population by reweighting national-level data to resemble Connecticut in terms of key demographic characteristics (e.g., age, income, race, and ethnicity). Modeled individuals then choose among various health insurance enrollment options by comparing their costs and benefits.

Because it is not reported in most data sources, we imputed immigration documentation status using logic that was informed by algorithms used in the literature (Van Hook et al., 2021; Passel and Cohn, 2018) that prioritized noncitizens and ruled out individuals who reported having jobs that are typically unavailable to undocumented immigrants (e.g., police officer) or if they reported benefits that are unavailable to the undocumented population, such as Social Security income.

A key challenge for this analysis was determining what share of undocumented immigrants would be likely to take up insurance coverage if it were available to them. Literature suggests that take-up would be lower among the undocumented population than among the legally present recent immigrant population because of such factors as information barriers and hesitancy to present to a government agency (Whitener, 2020). However, because few states have expanded coverage to their undocumented populations and because the denominator is uncertain, estimates of take-up rates are highly uncertain. There is similar uncertainty in estimating how much health care undocumented populations will use once they become insured. To address these uncertainties, we conducted sensitivity analyses that varied both the take-up and utilization rates.

Results

Figure 1 shows the estimated insurance enrollment of undocumented and legally present recent immigrants in the coverage expansion scenarios relative to current law. We estimate that enabling undocumented and legally present recent immigrant populations to enroll in Medicaid will increase total insurance enrollment among this population by approximately 21,400 individuals, or 43 percent (Scenario 1 relative to current law). In sensitivity analyses, we found that the change in enrollment ranged from 17,500 to 25,100 people (in Scenario 1), depending on assumptions about the take-up rate. The number of individuals enrolling in Medicaid (28,900) exceeds the total increase in insurance enrollment because some people move to Medicaid from other sources of coverage. In particular, individuals move from employer-sponsored insurance (ESI) and unsubsidized individual market coverage into Medicaid as undocumented and legally present recent immigrants gain Medicaid eligibility. Allowing undocumented populations to access subsidies in the individual health insurance market leads to a small increase over the Medicaid-only scenario, adding 2,200 to 3,000 additional enrollees (Scenarios 2 and 3, respectively).

Scenario	Employer-sponsored insurance	Medicaid	Subsidized individual market	Unsubsidized individual market	Total
Current Law	43,200	800	200	5,200	49,400
Scenario 1	39,100	28,900	0	2,800	70,800
Scenario 2	39,000	28,900	2,600	2,500	73,000
Scenario 3	39,000	28,900	4,400	1,500	73,800

Figure 2 shows the estimated cost to Connecticut for 2023 from enacting these policies. We estimate that spending will increase by nearly $83 million in Scenario 1, which expands only Medicaid. In sensitivity analyses, we found that this estimate could range from $70 to $93 million, depending on assumptions about take-up rate and health care spending, although even these estimates carry a high degree of uncertainty. Scenarios 2 and 3, which expand marketplace subsidies to undocumented immigrants, have a relatively small effect on coverage while having a larger effect on costs (Scenarios 2 and 3), which increase by 32 to 45 percent over the Medicaid-only scenario.

Current Law

$1.1 million

Scenario 1

$83.7 million

Scenario 2

$110.4 million

Scenario 3

$121.7 million

NOTE: Costs include new state spending on Medicaid and marketplace subsidies (APTCs and CSRs). We do not account for the costs of reduced spending on emergency Medicaid services.

Our microsimulation analysis focuses only on the costs of providing Medicaid and marketplace subsidies and does not directly consider potential savings to hospitals from reduced uncompensated care spending or potential savings to the state from reduced spending on emergency Medicaid services or recouped federal APTC spending. However, we provide a high-level estimate of potential savings to hospitals from reduced spending on uncompensated care, which could amount to roughly $63 million in Scenario 1 to $72 million in Scenario 3. In addition, high-level data provided by the state suggest that the state spent approximately $15 million on emergency Medicaid services in 2021.

We estimate that these policies would have minimal effects on enrollment and premiums for U.S. citizens and legally present residents who have lived in the United States for five or more years.

Discussion

We estimate that expanding Medicaid and individual market subsidy eligibility to otherwise-eligible undocumented and legally present recent immigrants would lead to an increase in coverage among this population by 21,000 to 24,000 individuals and some movement from ESI and unsubsidized individual market coverage into the new health insurance options. These changes did not substantially impact total insurance coverage among U.S. citizens and other immigrants residing in Connecticut. Allowing undocumented immigrants to receive individual market subsidies (in addition to expanding Medicaid eligibility) had only a minor impact on total insurance enrollment among this population, although it would substantially improve affordability of coverage. In turn, changes in individual market premiums were negligible. We estimate costs to the state to extend Medicaid coverage to undocumented and legally present recent immigrants would be approximately $83 million. On top of the costs to extend Medicaid coverage to individuals who are currently ineligible because of their immigration status, the cost to Connecticut to expand individual market subsidies to undocumented immigrants with incomes less than 200 percent of FPL would be $27 million, while expanding subsidies to undocumented immigrants with incomes less than 400 percent of FPL would cost the state $38 million. Because the federal government would not fund these programs, we assumed that the costs of these insurance-eligibility expansions would be borne entirely by the state.

In each scenario, almost one-quarter of undocumented and legally present recent immigrants newly insured by Medicaid or receiving subsidies on the individual market were previously insured via ESI or unsubsidized individual market coverage. As a result, the state will absorb costs for some individuals who would have been insured without the policy change. However, these policies would also substantially increase the affordability of health care coverage for this population, particularly for individuals who qualify for Medicaid.

Another consideration for policymakers is the potential cost savings from such policy options. These could stem from (1) cost savings to hospitals on uncompensated care spending, and (2) direct savings to the state that we did not estimate in our analysis. For example, Connecticut currently has an emergency Medicaid program that covers emergency care to individuals who qualify based on income, regardless of immigration status. The costs for this program (approximately $15 million in 2021) would presumably be substantially reduced by the decrease in uninsurance among undocumented and legally present recent immigrants. Furthermore, by offering Medicaid to legally present recent immigrants, the state would forego federal APTC funding for the portion of this population that would have otherwise enrolled in marketplace coverage and been eligible for subsidies. Therefore, it is possible that Connecticut could recoup some of the forgone APTC funding via a Section 1332 waiver.

There are several limitations associated with this analysis that are important to note. In particular, there is uncertainty around the size and demographics of the undocumented population, the Medicaid take-up rate, and expected health care spending among undocumented immigrants. However, we conducted sensitivity analyses around the take-up rate and health care spending assumptions that we used in the model, and they had relatively small impacts on the overall findings, in the expected directions.

This analysis suggests that removing immigration status requirements for Medicaid and individual market subsidy eligibility would decrease uninsurance among the undocumented and legally present recent immigrant populations by 32 to 37 percent. Costs to the state would range from $83 to $121 million (or approximately $3,900 to $4,900 per newly insured individual), without factoring in savings to hospitals on uncompensated care or any costs that the state could recoup from forgone federal APTC payments. Overall, expanding Medicaid and individual market subsidy eligibility to individuals who would qualify were it not for their immigration status has promise to improve insurance coverage and affordability in Connecticut for undocumented and legally present recent immigrant populations, while not substantially impacting legal residents.