Findings from the ACA Special Initiative
RSF Accomplishments: Special Initiative on the Social, Economic, and Political Effects of the Affordable Care Act
These summaries of 13 research awards made in the first four rounds of our ACA special initiative highlight key research questions and findings for each project. The write-ups are divided into five sections devoted to different aspects of the ACA’s impact. The findings were presented by the PIs at a meeting at RSF on Friday, December 8, 2017.
The ACA, Politics, Public Opinion, and Public Policy
- 1. Policy Uptake as Political Behavior: Evidence from the Affordable Care Act; Amy E. Lerman and Meredith Sadin, University of California, Berkeley
- 2. Attitudinal Policy Feedback and the Affordable Care Act; Julianna Pacheco, University of Iowa
- 3. Policy Making Politics? The Mass Political Impact of Medicaid Expansions; Joshua D. Clinton, Vanderbilt University, and Michael W. Sances, University of Memphis
- 4. Explaining Stability and Change in Public Attitudes Toward the Affordable Care Act; Daniel J. Hopkins, University of Pennsylvania, and William R. Hobbs, Northeastern University
The ACA, Labor Market Outcomes, and Public Program Participation
- 5. Employer-Sponsored Insurance under the Affordable Care Act and Comparisons with Insurance in the Exchanges; Jean M. Abraham, University of Minnesota, and Anne B. Royalty, Purdue University
- 6. The Impact of the ACA Medicaid Expansion on Public Program Participation and Labor Market Outcomes of Low-Wage Workers; Lara Shore-Sheppard, Lucie Schmidt, and Tara Watson, Williams College
The ACA, Financial Security and Economic Wellbeing
- 7. Can the ACA’s Medicaid Expansions Protect the Creditworthiness of Low-Income Americans?; Heidi Allen and Tal Gross, Columbia University
- 8. Accounting for the Impact of the Affordable Care Act on Poverty; Sanders Korenman and Dahlia K. Remler, Baruch College, City University of New York
- 9. Health Care and the Financial Wellbeing of Low Income Families; Robert Kaestner, University of Illinois, Chicago; Bhashkar Mazumder, Federal Reserve Bank of Chicago; and Sarah Miller, University of Michigan
- 10. The Impact of Medicaid Expansion on Household Consumption; Sayeh Nikpay, Vanderbilt University; Thomas Buchmueller and Helen Levy, University of Michigan
- 11. Effects of the Affordable Care Act on the Use of Alternative Financial Services Providers; Katie Fitzpatrick, Seattle University, and Anne Fitzpatrick, University of Massachusetts, Boston
The ACA and Previously-Underserved Populations and New Beneficiaries
- 12. The Effects of Medicaid Enrollment on Recidivism Outcomes; Mia Bird, Public Policy Institute of California and the University of California, Berkeley; Shannon McConville and Viet Nguyen, Public Policy Institute of California
A Mixed-Methods Approach to the Study of Federal Health Policy Change
- 13. Tracing Effects in Non-Metropolitan Counties of the 2017 Federal Health Policy Changes; Katherine Swartz, Theda Skocpol, and Mary Waters, Harvard University
The ACA, Politics, Public Opinion, and Public Policy
1. Policy Uptake as Political Behavior: Evidence from the Affordable Care Act; Amy E. Lerman and Meredith Sadin, University of California, Berkeley; award amount: $133,285 (July 2015)
About 20 million Americans have gained health insurance coverage since the Affordable Care Act (ACA) became law. However, millions remain uninsured. We hypothesize that the incomplete uptake is in part a function of persistent political rancor. We find that self-identified Republicans have been less likely than Democrats to enroll in an insurance plan through a state or federal exchange since the ACA was passed.
Because a variety of partisan differences might predict insurance status, we also estimate the effects of partisanship on uptake. We find that, all else equal, being a Republican increases the likelihood that an individual will either be uninsured or purchase off-marketplace insurance, as opposed to signing up for a plan offered through the ACA marketplaces.
We then examine the implications of partisanship for policy implementation and political strategy. In a large-scale field experiment, we randomly assigned uninsured individuals to one of two websites through which they could enroll in the exchanges: either a private website called HealthSherpa.com, or the government’s website, Healthcare.gov. Republicans assigned to the private website were 20 percentage points more likely to enroll in a plan through the ACA marketplace than those assigned to the public site. In other words, framing enrollment in a way that emphasizes the “private” nature of the ACA and de-emphasizes the government’s role can substantially increase uptake among Republicans.
Our results make clear that partisanship interacts with policy framing to influence enrollment behavior. An abundance of research illustrates that partisanship is a primary driver of citizens’ attitudes toward public policy. However, we do not yet know the extent to which party loyalties play a role in shaping policy behavior, such as whether to take advantage of public benefits. Those who study the decision to enroll in a public program focus on many factors: the availability and accuracy of information and the structure of incentives and penalties. We show that ACA insurance uptake is about both information and incentives and politics.
2. Attitudinal Policy Feedback and the Affordable Care Act; Julianna Pacheco, University of Iowa; award amount: $34,564 (July 2016)
This project examines how state policy decisions on the ACA influence public opinion toward healthcare reform. By focusing on the states, we can better understand the nuances of the effects of the ACA on changing public opinion. We can also assess the extent to which the ACA influences particular subgroups within states. This is useful for determining whether personal gain, partisan attachment, or policy knowledge influences reactions to policy.
We relied on the Kaiser Family Foundation (KFF) polls to measure state attitudes, and, where appropriate, combined the KFF polls with others (e.g., CBS/NYT, Gallup, and Pew) that asked similar questions. We validated our estimates using state surveys, such as the Kentucky Health Issues Poll and the Ohio Health Issues Poll. We measured quarterly state ACA attitudes on five topics (including overall favorability and support for keeping or expanding the law) from 2009 to 2015. We also estimated partisanship, health status, health insurance, and policy knowledge across states from individual-level survey data using a multi-level regression with post-stratification technique. With this specialized data, we estimated quarterly ACA favorability for Democrats, Republicans, respondents in excellent/good health, and respondents in fair/poor health from 2009 to 2015 in each of the states.
We find that state opinion toward the ACA varies both across time and states, but the amount of variance differs across question types. While favorability toward the ACA is generally low at the national level, we find that in some states (e.g., CA), a majority of residents favor the ACA while in others (e.g., WV) support is much lower than the national average. The public generally has a more positive outlook about the ACA’s impact on the country in general versus for themselves or their families.
We find significant movement over time in the public’s understanding of the ACA, with little variance across states. All states experienced increasing understanding of the ACA until the late 2010s. As expected, we find significant differences in levels of ACA support based on the type of marketplaces across states. Finally, we also find evidence that state policy decisions spill over to influence opinion in neighboring states. In a paper published in The Journal of Politics, Policy, and Law (JHPPL), we find that gubernatorial ACA announcements and grant activity in one state increased the likelihood of ACA support in nearby states. This suggests that public opinion is important to state policies as well as for the spread of policies. DOI: https://doi.org/10.1215/03616878-3766737
3. Policy Making Politics? The Mass Political Impact of Medicaid Expansions; Joshua D. Clinton, Vanderbilt University, and Michael W. Sances, University of Memphis; award amount: $28,095 (March 2016)
Our project examines the extent to which the Medicaid expansions have had any political effects on voter participation and public opinion. To study voter participation, we compared changes in registration and turnout using similar counties in states that do and do not expand Medicaid over time. This allowed us to account for the many ways in which states may vary politically, economically, and socially, and enabled us to identify the impact of the expansion. We find that the expansion of Medicaid has, at least in the short term, created a politically engaged constituency of beneficiaries. Overall, however, the effects are modest.
In counties with a higher than average share of citizens making less than 138 percent of the poverty line, we see an increase in the percentage of registered voters of about 3 percent in counties located just over the border of a state that expands Medicaid relative to a neighboring county located in a state that did not expand Medicaid between 2010 and 2014; the percentage of registered voters increased by about 4 percent when comparing the difference between expansion and non-expansion counties from 2012 to 2016. However, in counties with a lower than average share of potential beneficiaries—individuals making less than 138 percent of the poverty line—we observe no differences between expansion and non-expansion counties. When we examine voter turnout, we find modest effects in 2014 that disappear by 2016.
We also analyzed the extent to which the expansion changed public opinion towards the ACA. Specifically, do those living in expansion states become more favorable toward the ACA after the expansions, relative to residents of non-expansion states? We constructed a state-by-quarter panel data set of ACA opinions using 230,000 respondents. We found no difference in ACA approval between expansion and non-expansion states prior to the expansion. But there is a marked increase in support for the ACA following the expansion in expansion relative to non-expansion states. The impacts are small – about 0.2 on a 0 to 100 scale. Supplementary analysis at the individual level suggests the limited effects are driven by partisanship: only political independents adjust their opinion of the ACA in response to the expansion. We also found that the impacts are dampened in states where the expansions were implemented under Republican governors, which may reflect those governors actively obscuring any connections to the ACA.
4. Explaining Stability and Change in Public Attitudes Toward the Affordable Care Act; Daniel J. Hopkins, University of Pennsylvania, and William R. Hobbs, Northeastern University; award amount: $33,000 (November 2016)
Understanding public opinion toward the ACA can help provide insights about the future of health policy. This project consists of three surveys of public opinion on the ACA. First, we conducted a fall 2016 wave of the Institute for the Study of Citizens and Politics (ISCAP) panel survey among a sample of nationally-representative adults. The 12-wave panel includes extensive questions about health, insurance status, ACA attitudes, partisanship, prejudice, and other political attitudes over a decade, offering a unique opportunity to see how Americans respond to Republican repeal efforts.
Since the ACA was implemented in 2014, public opinion held stable with a slight majority against the law. However, in the short period between our October 2016 and November/December 2016 waves, we saw a marked increase in support, one that was most pronounced among Republicans. What underpins this shift? Trump’s victory made the prospect of ACA repeal much more realistic, and so may have led many to perceive a threat to the law and the benefits and protections it confers.
Second, in September 2017 we conducted a series of survey experiments with 1,631 SSI respondents which helped us identify the causal factors at work in the ISCAP findings. We found that priming partisanship makes Republicans (and also Democrats, though to a lesser degree) more opposed to the ACA. There was a strikingly large effect of a race-related prime: Trump supporters who read that the 2010 health care law “has allowed approximately 18 million non-elderly adults to get health insurance, including more than 7 million black and Hispanic Americans” were markedly more supportive of repealing the ACA than respondents who did not read the clause about black and Hispanic Americans. Even years after the ACA’s implementation, symbolic attitudes such as those related to race and ethnicity remain an important element in explaining attitudes toward the ACA.
Third, in order to get around the limitations of close-ended survey questions, we used a text-analysis survey to analyze justifications cited in individual attitudes toward the ACA over the full course of the law. We analyzed open-ended responses to see if different types of justifications were more or less common before the law was signed, before the ACA was implemented, and before the 2016 presidential election. Among both supporters and opponents, we see a move from general to concrete justifications over time. Republican respondents spoke more about the role of government than cost considerations in 2009 compared to 2015, and Democrats did not mention specific benefits of the law until after it was signed. One impact of implementation may be to shift the grounds on which it is evaluated. This shift from abstract to concrete may have also changed the criteria on which Republican repeal efforts were evaluated.
The ACA, Labor Market Outcomes, and Public Program Participation
5. Employer Sponsored Insurance under the Affordable Care Act and Comparisons with Insurance in the Exchanges; Jean M. Abraham, University of Minnesota, and Anne B. Royalty, Purdue University; award amount: $110,303 (March 2016)
The employer-based health insurance system has traditionally meant a strong link between insurance and employment. The ACA dramatically changed options for obtaining non-job-related insurance. New options include Medicaid expansion, the young adult mandate, and individual market expansion through subsidized marketplaces. The ACA implementation suggests that expanding these options may affect employment outcomes. We ask how the Medicaid expansion has affected employer-sponsored insurance (ESI) offers, workers’ ESI out-of-pocket premiums, and workers’ ESI eligibility.
Using the 2010-2015 Medical Expenditure Panel Survey-Insurance Component and a difference-in-differences modeling strategy, we find no evidence that the expansion of Medicaid eligibility affects an establishment’s probability of offering ESI. We also find no evidence to support a causal impact of the expansion on out-of-pocket premiums for single coverage. In line with other studies, we find that employer provision of health insurance is largely unaffected by the Medicaid expansion. Employer-sponsored insurance continues alongside the new options.
Given the uncertainty still surrounding the law and the repeal of the individual mandate through the tax reform bill, we note some caveats and potential ways these outcomes could change in the future. First, uncertainty may have deterred employers from changing to ESI up until now. Uncertainty and possible political sabotage may continue to deter employers from making changes or could hasten changes if marketplace premiums rise faster than before or if insurers decide to pull out of the markets. As more flexible waivers create greater differences among states in their Medicaid expansion provisions, including potential work requirements, the labor market effects of the Medicaid policy expansion could also change.
6. The Impact of the ACA Medicaid Expansion on Public Program Participation and Labor Market Outcomes of Low-Wage Workers; Lara Shore-Sheppard, Lucie Schmidt, and Tara Watson, Williams College; Award amount: $96,931 (December 2016)
The ACA significantly increased the availability of public insurance. However, the safety net is a nexus of public programs addressing overlapping needs and serving an overlapping set of beneficiaries. As a result, the expansion of eligibility to groups previously ineligible for Medicaid may have spillover effects to other public assistance programs. Individuals may be induced to reduce their income below the Medicaid income limit, potentially increasing cash and food program participation. Or the increased income limit for some groups could allow more earnings while still qualifying for Medicaid, potentially reducing cash and food program participation. Finally, the availability of insurance separate from employment may reduce labor force participation, or could promote participation by improving health. These potential labor market effects may affect participation in safety net programs, particularly programs such as the Earned Income Tax Credit (EITC) that are linked to labor market outcomes.
This project investigates the impact of the Medicaid expansion on participation in the disability benefit programs Supplemental Security Income (SSI) and Social Security Disability Income (SSDI), the Earned Income Tax Credit (EITC) and the Supplemental Nutrition Assistance Program (SNAP) using county-level administrative data. We compare changes in program participation in contiguous counties across state lines where one county expanded Medicaid and the other did not.
We find robust evidence that Medicaid expansions result in differing levels of coverage between such counties. Results for disability program participation show small effects for SSI only. We find that counties with high rates of SSI participation among children prior to the expansion (a proxy for knowledge about the program not directly affected by the ACA) had a larger increase in SSI participation among adults. We find no statistically significant impacts on SSDI caseloads. Based on IRS tax returns containing EITC claims at the county level, we find that Medicaid expansion is associated with a roughly 2 percent increase in EITC claims in previously high uninsurance counties compared to neighboring counties across the state line. Finally, our preliminary estimates for SNAP participation indicate that the Medicaid expansion increased participation in counties with high levels of uninsurance prior to the expansion. These findings are not surprising, as Medicaid expansion is likely to have had the greatest importance for individuals in previously high uninsurance areas.
Overall, the increases in safety net program participation in counties that expanded Medicaid suggest that access to one safety net program may increase participation in others. Although we cannot yet determine the mechanisms behind these connections with the data we have used so far, in the second year of our research we are incorporating household-level data that will allow us to explore the nature of these connections in more depth.
Financial Security and Economic Wellbeing
7. Can the ACA’s Medicaid Expansions Protect the Creditworthiness of Low-Income Americans? Heidi Allen and Tal Gross, Columbia University; award amount: $142,800 (March 2016)
Our project will produce two separate papers that assess how the ACA has affected families’ financial wellbeing. The first paper, co-authored with Ashley Swanson and Jialan Wang, was published in the October 2017 issue of Health Affairs. It tested how California’s early expansion of Medicaid affected consumers’ demand for payday loans, a form of high-interest borrowing popular with low- and middle-income Americans. We used a unique dataset of payday loans made available to us by an industry trade group, the Community Financial Services Association of America (CFSA). It contained information on over 93 million loans originated by five national storefront payday lending chains. For each loan, we observed the dates that the loan was made and was due, the outcome of the loan (for example, a default or late payment), and the ZIP code of the lender’s storefront. The data set covered all loans from 2009 through early 2014.
We merged that dataset with information on when each county in California expanded Medicaid and constructed counts of payday loan volume for each county and month. We then ran difference-in-difference regressions that measured how payday borrowing changed in California counties after they expanded Medicaid, relative to counties in other states that did not expand Medicaid. The results suggest that consumers initiated 789 fewer loans after their county expanded Medicaid, which amounts to an 11 percent drop in borrowing. We found similar drops in other measures of total loan volume: total dollars loaned and the number of unique borrowers appearing in payday-lending storefronts each month. Our findings suggest that Medicaid coverage reduces the demand for high-cost loans. DOI: https://doi.org/10.1377/hlthaff.2017.0369
Our second paper will measure the effect of Medicaid expansion on eviction. We have purchased a large dataset consisting of a near-universe of evictions from fifteen states. We are currently merging that information into a dataset that will allow us to test how county-level and state-level expansions of Medicaid affected the frequency with which families were evicted.
8. Accounting for the Impact of the Affordable Care Act on Poverty; Sanders Korenman and Dahlia K. Remler, Baruch College, City University of New York; award amount: $149,967 (March 2016)
How much do ACA benefits, specifically Medicaid expansion and premium subsidies, reduce (health-inclusive) poverty rates and gaps? How does the impact of ACA benefits on poverty vary by family type and by race and age? How much did poverty rates fall in states that expanded Medicaid between 2014 and 2015? Since there is no official health-inclusive poverty measure, we have no ready answer to these and other related questions. Yet few would disagree that health care is a basic need and that health insurance helps meet that need. Our primary goal is to construct a Health-Inclusive Poverty Measure (HIPM) and use it to assess the impact of health insurance benefits, including the ACA premium subsidies and Medicaid expansions nationally.
Our first project paper, “Estimating the Effects of Health Insurance and Other Social Programs on Poverty Under the Affordable Care Act,” was published in Health Affairs (October 2017). Among our key findings: public health insurance benefits accounted for nearly one-third of the overall (health-inclusive) poverty reduction from all public benefits; Medicaid had a larger effect on child poverty than all non-health means-tested benefits combined, an impact nearly as large that of refundable tax credits; poor adults with neither children nor a disability experienced little poverty relief from public programs, and what relief they did receive came mostly from premium subsidies and other public health insurance benefits. Hispanics and non-Hispanic blacks also experienced major poverty reductions from employer and government insurance. Overall, health insurance benefits reduced poverty greatly among beneficiaries. Strikingly, Medicaid reduced poverty by 18 percentage points among Medicaid recipients and ACA premium subsidies reduced poverty by nearly 7 percentage points among those who purchase non-group insurance. DOI: https://doi.org/10.1377/hlthaff.2017.0331
Our second project paper, “The Impact of State Medicaid Expansions on Poverty,” estimates differences in health-inclusive poverty rates in 2015 between states that did and did not expand Medicaid. The HIPM poverty rate in 2015 was two percentage points lower in expansion states than in non-expansion states. Medicaid expansion was associated with a 20 percent reduction in HIPM deep poverty, controlling for socio-demographic characteristics and non-health income. These poverty impacts can only be detected with a health-inclusive poverty measure; the impact of Medicaid expansion on deep poverty according to the Census Bureau’s Supplemental Poverty Measure is less than 6 percent. Medicaid expansion is particularly important for reducing deep HIPM poverty among less-educated persons, blacks, Hispanics, and children.
9. Health Care and the Financial Well-Being of Low Income Families; Robert Kaestner, University of Illinois, Chicago; Bhashkar Mazumder, Federal Reserve Bank of Chicago; Sarah Miller, University of Michigan; award amount: $150,000 (Mar. 2016)
Our project examines the relationship between Medicaid enrollment under the expansion in Michigan (“Healthy Michigan”) and the financial wellbeing of households. Our analysis links individual-level administrative data on enrollment and health from the Healthy Michigan Plan (HMP) enrollees to their TransUnion credit reports. Our main findings are that enrollment in HMP was associated with significant reductions in medical debt sent to a collection agency, overall debt sent to a collection agency, and credit market debt that was 30 days past due or more.
The fundamental research challenge in studying the effects of Medicaid expansion is understanding what would have happened in the absence of the reform. To address this issue, we used a difference-in-differences design that compared changes in amount of credit market debt past due, debt sent to a collection agency, and medical debt sent to a collection agency among HMP enrollees prior to the expansion (July 2011- January 2014) and after the expansion (July 2014- January 2016) relative to non-elderly adults in low-income zip codes in states that did not adopt expand Medicaid. Our sample consisted of 477,803 HMP enrollees linked to TransUnion credit reports and a random sample of TransUnion credit reports of 830,181 non-elderly adults in non- expansion states.
Enrollment in HMP was associated with reductions in the amount of debt sent to a collection agency, the amount of medical debt sent to a collection agency, and the amount of credit market debt 30 days past due or more. The results remained robust to a variety of alternative specifications. Similar sized effects were estimated across subgroups defined by income level, medical service usage, and presence of a chronic disease. This is surprising, given the likely differences in risk of financial shock due to differences in health and income. Despite these differences, Medicaid enrollment appears to have salutary effects on financial outcomes even among apparently healthier and higher-income groups. Overall, we find that Healthy Michigan improved the financial well-being of enrollees across all groups and subgroups.
10. The Impact of Medicaid Expansion on Household Consumption; Sayeh Nikpay, Vanderbilt University, and Thomas Buchmueller and Helen Levy, University of Michigan; award amount: $126,871 (December 2016)
The Medicaid expansion represented a large transfer of resources to very low-income households. We aimed to measure household consumption to better understand how this transfer of resources affected low-income households. We estimated the impact of Medicaid expansion on consumption using an analytic sample including 24,927 observations on 12,502 households from 2010 through 2016 from the Consumer Expenditure (CE) Survey. Outcome measures were different elements of household consumption. The data includes information on household spending on health insurance premiums, out-of-pocket medical care, and prescription drugs. We combined these costs for a measure of total health spending by household.
We find that while Medicaid expansion significantly reduced average health spending by low-income households, these reductions were small relative to the size of total household budgets and did not lead to any significant increases in non-health spending. One way to understand our results is to note that low-income households spent very little out-of-pocket on medical care to begin with, and paid for very little of what they consumed. However, the costs of medical care by low-income households prior to Medicaid expansion seems to have resulted in some combination of debt that the households could not pay or that was written off by providers. Providing coverage through Medicaid improved the financial position of both these households and the hospitals that served them.
We found that the change in the fraction of households with an unpaid credit balance before and after 2014 was 1.5 percentage points lower in expansion states, suggesting that households may have used savings from Medicaid to pay down debt. While our findings demonstrate the beneficial impact of the expansion on economic wellbeing and financial security, the ACA’s premium tax credits may have an even greater effect on economic well-being.
11. Effects of the Affordable Care Act on the Use of Alternative Financial Services Providers; Katie Fitzpatrick, Seattle University, and Anne Fitzpatrick, University of Massachusetts, Boston; award amount: $23,740 (June 2017)
We examine the extent to which the Medicaid expansions improved the financial well-being of low and moderate-income (LMI) households, measured by their use of alternative financial services (AFS). AFS products tend to be high-cost, high-fee services, used by households with few other options from mainstream financial providers such as banks. AFS products include payday loans, pawn loans, auto-title loans, tax refund anticipation loans, rent-to-own contracts, check cashing services, and money orders. An estimated 24 percent of all households report using an AFS provider in the past 12 months. This industry processes at least $320 billion worth of transactions annually, including more than $200 billion credit transactions. Regulation of these products varies widely; some states tightly regulate or ban these products, while others have little or no regulation. Relatively little federal regulation exists.
We first document that Medicaid expansion expanded health insurance coverage among LMI households. We then examine whether increased health insurance coverage changed AFS use by linking four years (2009, 2011, 2013, 2015) of the FDIC-sponsored Unbanked and Underbanked Supplement to the Current Population Survey (CPS) with the March CPS Supplement on health insurance and income. We then test whether AFS use decreased in states that expanded Medicaid over time compared to states that did not.
We find that households in Medicaid expansion states are 6 percent less likely to report any AFS use in the past 12 months, driven by a 16.4 percent decrease in AFS credit usage. Households in expansion states are 1-2 percentage points less likely to report use of check cashing services, pawn shop loans, and payday loans; there is increased use of rent-to-own contracts. We find that these financial outcomes are due to reduced medical debt. Households in Medicaid expansion states report on average $365 to $470 lower medical expenditures. Our results suggest that Medicaid expansion improved financial security by reducing medical debt and expenses. As a result, households decreased their use of AFS providers, notably pawn shops, check cashing outlets, and payday loan outlets.
The ACA and Previously Underserved Populations and New Beneficiaries
12. The Effects of Medicaid Enrollment on Recidivism Outcomes; Mia Bird, Public Policy Institute of California and the University of California, Berkeley; Shannon McConville and Viet Nguyen, Public Policy Institute of California; award amount: $137,810 (July 2015)
The ACA created new opportunities to address a key criminal justice challenge: the justice-involved population has high health needs but historically limited access to health insurance coverage. Expansion of state Medicaid programs extended a public insurance option for the first time to many in the criminal justice population. To the extent that coverage confers access to needed medical services—particularly behavioral health services—these expansions could have important impacts on recidivism outcomes and public safety.
In California, policymakers enacted a law (AB 720) in October 2013 that removed barriers to enrolling jail inmates into public insurance coverage. The legislation authorizes counties to assist inmates in applying for health insurance coverage and changes state law to allow inmates to retain or initiate coverage while in custody. Policymakers had passed a law (AB 82) in 2012 that provides $25 billion in funding to support targeted outreach and enrollment efforts for the Medicaid program with a focus on high need groups, including those under the supervision of correctional agencies. These steps reinforced California’s efforts to expand its Medicaid program (Medi-Cal) to the justice-involved population under the ACA.
This study draws on data collected through the Multi-County Study (MCS), an effort to link local and state data on the criminal justice population in California to examine the effects of Medicaid enrollment on recidivism outcomes. We use a pre-post design, leveraging the 2014 Medicaid eligibility expansion as a policy experiment to identify the effects of enrolling into Medicaid on recidivism rates for individuals released from jail custody terms in two counties. Drawing on a rich set of individual-level characteristics, including demographics and criminal histories, we use a matching technique to construct an unenrolled, pre-ACA control group similar in characteristics to the enrolled, post-ACA treatment group. We then use regression models, adjusting for any remaining differences in observed characteristics, to estimate the effects of realignment on overall and felony re-arrest and reconviction rates for a one-year window. We find that Medicaid enrollment substantially reduced recidivism rates, with overall re-arrest rates over 5 percent lower for the post-ACA treatment group, and reconviction rates nearly 13.5 percent lower.
While this study includes only two counties and our findings are preliminary, the implications of this research are that facilitating health insurance enrollment for the justice-involved is a crucial intervention with gains beyond health outcomes, including improvements in public safety and reductions in justice system expenditures.
A Mixed-Methods Approach to the Study of Federal Health Policy Change
13. Tracing Effects in Non-Metropolitan Counties of the 2017 Federal Health Policy Changes; Katherine Swartz, Theda Skocpol, and Mary Waters, Harvard University; award amount: $50,000 (March 2017)
The objective of this project is to understand how communities in non-metropolitan counties (i.e., not urban and not very rural counties) are planning for and coping with reductions in federal health care funding. To do so, we are interviewing a variety of community leaders—health care providers, business executives, local government officials, county leaders of political parties and independent groups, newspaper publishers—in two counties in each of four states: North Carolina, Ohio, Pennsylvania, and Wisconsin.
We selected counties with similar demographic and income characteristics. Two states expanded Medicaid eligibility and two did not; all four states have state legislatures dominated by Republicans, but two have governors who are Democrats; all are mid-size population states with diversity in terms of urban-rural areas and sizeable manufacturing and agricultural sectors. All of the counties voted for Trump in 2016, and three of those counties voted for Obama in 2012. The counties’ populations are primarily white, with not insignificant numbers of African Americans, Latinos, and other immigrant populations, and the population with incomes below the poverty line is roughly similar across the eight counties.
Our research provides a broader context for the problems faced by the uninsured, the underinsured and Medicaid enrollees in obtaining care. The need for health care is high among these communities, particularly with regard to mental health and substance abuse disorders and maternal-child health. But health care is only one of a host of issues related to the communities’ attempts to maintain a thriving place to work and live. Leaders see the health needs (especially those related to mental health) as intertwined with their efforts to attract and retain good jobs so younger people will stay (which also relates to the schools, housing stock and other amenities).
There is a pervasive sense that the federal government cannot be counted on to provide funds needed for health-related services for lower-income individuals. If Medicaid funds are reduced or fewer people are able to enroll, many more people will be unable to obtain care. But the loss of Medicaid funds also has community-wide ripple effects. The burden of trying to make up for the loss of federal funding falls on public health and social services agencies and community organizations like United Way and local foundations.
The efforts over the past year to repeal and destabilize the ACA as well as curtail funding for Medicaid and other health programs have been a catalyst for increased citizen activism. Independent groups have formed to battle congressional and Trump administration efforts to dismantle the ACA and tax proposals. Many of the grass-roots groups that have grown up over the past year are led or co-led by women, many of whom have ties to churches and/or concerns about local schools. Tying health issues to people’s interests in local schools seems to be an avenue for engaging the larger community.