Exploring Variation in States’ Changing Uninsured Rates From 2014 to 2015

Click here to open the publication as a PDF.

By Molly Warren, Senior Policy Analyst, Best Practices Institute, Enroll America | January 2016

To guide outreach efforts leading up to each open enrollment period since 2013, when the health insurance marketplaces opened and coverage expansions created by the Affordable Care Act (ACA) took effect, Enroll America created and updated a data model to predict who in the United States has health insurance and who does not. With three years of data, we can now look at the changes and trends over time — and last year’s data tell a more nuanced story than before.[i]

The data continue to show that overall more uninsured Americans across the country have been able to get health insurance, but there is more state-by-state variation in how uninsured rates have changed between 2014 and 2015 than between 2013 and 2014. Despite outreach and enrollment efforts during the second open enrollment period that were arguably smarter and more coordinated, and engaged more partner organizations than during the prior year, some of the states that had seen some of the steepest drops in their uninsured rate between 2013 and 2014 (after the first enrollment period) actually saw slight increases in the uninsured rate between 2014 and 2015 (after the second open enrollment period). This paper explores some of the reasons why this might have happened, and some potential factors that do not seem to have played a role.

Key Findings

Although the national uninsured rate continued to decline in 2015, following the second open enrollment period, the changes at the state level are more complicated. Not surprisingly, most states (34) and the District of Columbia continued to see their uninsured rates drop between 2014 and 2015. However, our model estimates that in 14 states the uninsured rate actually increased slightly between 2014 and 2015: Alaska, Arizona, Arkansas, California, Colorado, Florida, Mississippi, Montana, Nevada, New Mexico, Oklahoma, Texas, West Virginia, and Wyoming. The two remaining states (Missouri and North Carolina) saw no changes in their uninsured rates between 2014 and 2015. See figure below for the estimated uninsured rate nationally, compared to the 14 states that experienced increases between 2014 and 2015.


Explanations for State-Level Variation

We spent some time trying to better understand what factors besides enrollment might be contributing to the trends we saw in those 14 states (since marketplace enrollment went up between 2014 and 2015 in each of those states, and was on par with national average enrollment growth). We considered things like state demographic shifts, affordability of coverage, unemployment rates, and prevalence of common barriers to coverage, like language barriers and verification challenges for immigrants.

Before delving into some trends in state-level variation and potential explanations behind some of the increases in states’ uninsured rate, it is important to keep a few things in mind:

  1. Uninsured rates were still lower in 2015 than in 2013 (before the ACA coverage expansions took effect) in every single state.
  1. There are far more factors that contribute to a state’s uninsured rate than marketplace and Medicaid enrollment alone.
  1. Our uninsured estimates include individuals who don’t have affordable coverage options right now, including individuals who fall into the family glitch[ii] or the coverage gap in states that have not expanded Medicaid, which encompasses 3 million Americans in Texas and Florida alone according to recent estimates[iii].

With this in mind, here are some of our takeaways about the increased uninsured rate in some states between 2014 and 2015:

  1. Some demographic groups experience greater barriers to getting and keeping coverage: Overall, states with relatively large Latino populations and immigrant populations as well as those with high poverty rates (according to 2014 American Community Survey data) generally saw large decreases in their uninsured rate between 2013 and 2014, but saw slower declines, and in some cases increases, in their uninsured rates between 2014 and 2015.[iv] Given that these three demographic factors tend to overlap and are markers for a number of acknowledged barriers, it is hard to determine what one factor is most responsible. A combination of all of them seems most probable. For example, barriers like citizenship and immigration data matching problems during the application process, affordability/eligibility issues, and language barriers all likely contribute to these states having higher uninsured rates in 2015 than in 2014. 
  1. Migration and effectuated enrollment patterns may also explain some differences: We found that states with some of the larger increases in uninsured rates between 2014 and 2015 (North Carolina, Florida, Nevada, and Texas) have some of the highest numbers of inbound residents according to the Annual Movers Study.[v] So, despite success at getting current residents enrolled, there is still a continual, disproportionate influx of new residents, who are more likely to be uninsured (since insurance is typically attached to geography; when you change addresses, your insurance options often change, either because you have changed jobs, or there are different plans or programs available to you depending on where you live). Also, some states (Arizona, Arkansas, Florida, Mississippi, and North Carolina) that saw increases in uninsured rates between 2014 and 2015 had lower 2014 enrollment effectuation and retention rates, which would certainly could influence uninsured rates.[vi]
  1. Some seemingly compelling factors do not seem to drive state variation: We also looked at a number of other factors that do not seem to drive variation among states. While age and gender are important when looking at trends among populations, the age and gender distributions do not seem to diverge enough from state to state to cause the variation we are observing in state uninsured rates. We also looked at changes in unemployment rates and premium costs but did not see any clear patterns. 


Looking ahead, we expect a continuing, moderate decline in the uninsured rate nationally as most Americans stay with their current coverage; more Americans learn about and take advantage of the expanded Medicaid and marketplace options they now have; Medicaid and marketplace enrollment platforms, processes, and assistance continue to improve; and the amount of the fine for not having coverage increases. As of December 26, 2015 — with more than a month to go yet in the open enrollment period — plan selections nationwide had nearly reached last year’s final open enrollment tally.[vii] While marketplace enrollment is only one factor influencing uninsured rates, it is a positive sign that more Americans are signing up this year, and that coverage gains will contribute to lower uninsured rates in 2016.



[i] For Enroll America’s 2015 estimates for uninsured rates, see: https://www.enrollamerica.org/research-maps/maps/changes-in-uninsured-rates-by-county/, and for more detailed background information and methodology behind Enroll America’s uninsured model, see: https://www.enrollamerica.org/research-maps/maps/changes-in-uninsured-rates-by-county/detailed-background/.

[ii] Financial assistance on the marketplaces is available to individuals and families based on income and family size (as measured by the Federal Poverty Guidelines) as well as whether an individual or family has an offer of affordable coverage through their employer. The test to determine whether employer coverage is affordable is linked to the cost for covering the individual employed and not the cost of the covering the whole family, which can be considerably higher. Therefore, in some instances, families may not be eligible for financial assistance on the marketplaces due to an employer coverage option for an individual while the family plan is truly unaffordable. This is often called the family glitch. For more information, see: http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=129.

[iii] Kaiser Family Foundation, Distribution of Eligibility for ACA Health Coverage Among those Remaining Uninsured as of 2015, October 2015. Available online at: http://kff.org/health-reform/state-indicator/distribution-of-eligibility-for-aca-coverage-among-the-remaining-uninsured/.

[iv] 2014 American Community Survey, U.S. Census Bureau, October 2015. Available online at: https://www.census.gov/programs-surveys/acs/news/data-releases/2014/release.html.

[v] 2014 National Movers Study, United Van Lines, January 2015. Available online at: http://www.unitedvanlines.com/about-united/news/movers-study-2014/index.html.

[vi] Plan Selections by County in the Health Insurance Marketplace, Office of the Assistance Secretary for Planning and Evaluation at the U.S. Department of Health and Human Services, July 2015. Available online at: https://aspe.hhs.gov/report/plan-selections-county-health-insurance-marketplace-july-2015-updated.

[vii] Nationwide Enrollment Reaches 11.3 Million, Enroll America, January 2016. Available online at: https://www.enrollamerica.org/blog/2016/01/nationwide-enrollment-reaches-11-3-million/.

Enroll America Menu
Are you looking for affordable health insurance? GetCoveredAmerica.org