Consequences of the ACA Medicaid Expansions
Sarah Miller
Business Economics
Research Overview
Health Policy
What happens when we use public policy to expand health insurance coverage to low-income kids, pregnant women, and adults?
- Short term: Improvements in financial well-being (Mazumder and Miller, AEJ: Economic Policy 2016), access to care and use of services (Miller and Wherry, New England Journal of Medicine 2017), use of appropriate care (Miller Journal of Public Economics 2012), provider response (Buchmueller, Miller, and Vujicic, AEJ: Economic Policy 2016).
- Long term: lower health care costs in adulthood, potentially offsetting the initial outlays (Wherry, Miller, Kaestner and Meyer, Review of Economics and Statistics forthcoming), improved educational and labor market outcomes (Miller and Wherry revise and resubmit, Journal of Human Resources).
- Intergenerational: Can health interventions be the key to breaking the cycle of poverty? (Miller and Wherry, in progress)
Credit and Borrowing Behavior of Subprime or Low-Income Borrowers
What determines credit market choices?
- How do lenders use information to screen in subprime populations (Journal of Financial Intermediation 2015), how does your neighborhood affect your credit market choices such as payday borrowing (Miller and Soo, in progress), what do payday loan borrowers do with a fresh start (Miller and Soo, in progress).
- How does a basic income affect characteristics such as discount rate, present bias, risk aversion? (Universal Basic Income Experiment with Y Combinator, in progress)
The ACA Medicaid Expansions and Credit Market Outcomes
Affordable Care Act was one of the largest expansions of public health insurance coverage since the 1960s. One motivation for expanding coverage is to provide financial protection.
- “No longer will illness crush and destroy the savings that[older Americans] have so carefully put away over a lifetime” -President Johnson (Signing of Medicare)
- “That’s what health insurance reform is all about–the peace of mind that if misfortune strikes, you don’t have to lose everything.” - President Obama 2014 State of the Union.
The Impact of the ACA Medicaid Expansions
As the result of a 2012 Supreme Court decision, the program was not expanded in all states as originally intended.
As of today: 32 states and DC have adopted the expansions; 18 states have not. Michigan adopted the expansion in April 2014.
What are the consequences of these expansions on financial well-being?
The MEdicaid Expansion and Financial Outcomes
In Michigan, the ACA Medicaid plan is called the "Healthy Michigan Plan" and has enrolled over 600,000 individuals since it was launched in April of 2014.
To conduct this analysis, we partnered with the Michigan Department of Health and Human Services to match HMP enrollees to their credit reports.
- MDHHS provided information on use of health care services, enrollment, income, and demographic characteristics.
- These were matched using a "double blind" matching method with TransUnion credit reports.
HMP and Financial Outcomes
Data:
HMP side: # of hospitalizations and ED visits in first 12 months of enrollment, presence of a chronic illness diagnosis code on an encounter during first 12 months of enrollment, some demographic information.
HMP AND FINANCIAL OUTCOMES
Exclusions:
- Those with third party liability
- Those enrolled on or after 7/2014
- Those not matched to a TU credit report
HealthY Michigan Enrollees
|
|
Pre-ACA | Post-ACA |
|---|---|---|
| Amount in Collections | $2083.35 | $1759.30 |
| Medical Collections | $1065.02 | $1002.85 |
| Amount past due | $846.62 | $762.16 |
|
|
|
|---|---|
| Matched to TU credit report | 90.1% |
| Age | 39.2 |
| Gender=Female | 52% |
| Income as % FPL | 44.3 |
| Hospitalizations | 0.13 |
| ED Visits | 1.05 |
| Chronic Illness | 72% |
| # of Individuals: | 477,803 |
FInancial Outcomes for Low Income Zip Residents
There may be trends in credit market outcomes due to macroeconomic conditions. To account for this, we also draw a random sample of about 800,000 individuals living in low-income zip codes in states that did not expand Medicaid.
|
|
Pre-ACA | Post-ACA |
|---|---|---|
| Amount in Collections | $1014.18 | $996.65 |
| Medical Collections | $401.74 | $472.09 |
| Amount past due | $636 | $668.48 |
HMP and FInancial Outcomes
Who doesn't have a credit report?
|
|
Not Matched | Matched |
|---|---|---|
| Age | 30.6 | 41.24 |
| Gender=Female | 44% | 51% |
| Hospitalizations | 0.11 | 0.15 |
| ED Visits | 1.00 | 1.19 |
| Chronic Illness | 63.9% | 77.8% |
| Income as % FPL | 30.4% | 41.1% |
| N | 47,760 | 429,240 |
Results
|
|
All Collections | Medical Collections | Past Due |
|---|---|---|---|
| Effect of HMP: | -301.98 (20.13)*** | -129.10 (14.41)*** | -115.25 (39.47)*** |
Results
|
|
Chronic Condition | No Chronic Condition | Hospitalization or ED visit | No Hospitalization or ED visit |
|---|---|---|---|---|
| All Collections | -314.23 (20.12)*** | -244.99 (20.36)*** | -243.16 (20.14)*** | -351.44 (20.28)*** |
| Medical Collections | -125.06 (14.43)*** | -129.59 (14.31)*** | -105.24 (14.45)*** | -149.59 (14.31)*** |
| Past Due | -128.92 (30.47)*** | -67.90 (30.57)*** | -90.61 (30.50)*** | -136.09 (30.50)*** |
Results
|
|
0 Income | Income 1-50% FPL | Income > 50% FPL & <= 100% FPL | Income >= 100% FPL |
|---|---|---|---|---|
| All Collections | -312.04 (20.16)*** | -301.12 (20.29)*** | -284.33 (20.27)*** | -311.61 (20.25)*** |
| Medical Collections | -137.43 (14.42)*** | -120.07 (14.32)*** | -119.57 (14.31)*** | -136.76 (14.31)*** |
| Past Due | -109.29 (30.50)*** | -111.36 (30.55)*** | -129.42 (30.52)*** | -117.49 (30.51)*** |
Results
Main takeaway:
- Large reductions in delinquencies
- Effects are (surprisingly) similar across many subgroups
- Beneficiaries of HMP are likely also hospitals, other creditors who serve this population
Other outcomes
Improvements in financial outcomes--but what about other outcomes we might care about, like health, use of care, and access?
No linked data available but we can learn about the ACA Medicaid expansions in general using a large survey, the National Health Interview Survey.
National Health Interview Survey
Not linked to Michigan data-- use low-income respondents in all states that expanded Medicaid and compare to those that did not expand.
Outcomes:
Health insurance coverage
Use of health care
Ease of access
Financial strain (self reported)
National Health Interview Survey
National Health Interview SUrvey
National Health Interview Survey
National Health Interview Survey
Conclusion
Thank you for the opportunity to discuss some current work surrounding the ACA Medicaid expansions.
Hosmer Hall
By umich
