Friday, May 13, 2016

Do Democrats and Republicans get the same coverage in the ACA marketplace?

UPDATED 5/20: Liz Hamel, Kaiser's director of survey research, was kind enough to send me some cross-tabs addressing the core question here -- whether Democrats and Republicans obtained similar coverage in the individual market.  See below!

Kaiser Family Foundation president Drew Altman, previewing a Kaiser survey of enrollees in private plans sold in the ACA marketplace, highlights the startling fact that partisanship seems to be the dominant factor in whether enrollees report being harmed or helped by the law.

While no one would likely dispute Altman's main conclusion, it does, to my mind, raise a question. Here's the gist:
In the Kaiser survey, which will be published next week, 29% of Republicans in marketplace plans (i.e., Obamacare) say they have benefited from the ACA compared with 75% of Democrats, a 46-point difference. There is no reason to believe that there are demographic differences between these Republican or Democratic marketplace enrollees that would explain this large of a difference in their responses. They are all purchasing coverage in the ACA marketplaces, and most members of each group are receiving premium subsidies under the law. Overall, substantially more marketplace-plan enrollees say that they benefited from the ACA (54%) than say they were negatively affected (35%).
Demographic differences between Democrats and Republicans doubtless don't explain this large of a difference in perception. But it seems possible to me that demographic differences -- specifically income differences -- might explain part of the perception divide.

With some exceptions, the lower a marketplace enrollee's income, the more comprehensive and affordable the purchased coverage is likely to be. I suspect, controlling for party affiliation, that marketplace customers' satisfaction level is inversely proportionate to their reported income

That's for two reasons. First: the lower an enrollee's income, the lower the percentage of income she'll pay for a benchmark silver plan. At 138% FPL, she's pay 2%; at 200% FPL, over 6%; and at 300% FPL, almost 10%. (That's supposed to be proportionate to ability to pay, but I don't think it is.) Second: those below 200% FPL who buy silver plans, as about 80% in that income range do, qualify for strong Cost Sharing Reduction (CSR) subsidies.

While Altman notes that most Democrat and Republican enrollees alike receive premium subsidies, he doesn't delve into whether CSR recipients are evenly divided between the parties.

CSR is the ACA marketplace's principle defense against underinsurance. It raises the actuarial value of a silver plan -- the percentage of the average user's annual medical costs -- from a baseline of 70% to 94% for enrollees with incomes under 150% FPL and  to 87% for those in the 150-200% FPL range. CSR fades to near insignificance at 201% FPL (raising AV to 73%) and cuts off at 251% FPL.

About a third of Americans live in households with incomes under 200% FPL. So do about 60% of ACA marketplace enrollees.

Almost everyone who obtains a high-AV plan in the ACA marketplace does so via CSR, and so has an income under 200% FPL. Here is an estimated breakdown (mine) of the actuarial values obtained by enrollees who qualify for premium subsidies in the 38 states that used the federal marketplace,

Actuarial value obtained by subsidized enrollees on

Actuarial Value
and metal/CSR level
Percent of subsidized
94  (Silver - CSR1)
37.0 (under 150% FPL)
90  (Platinum)
87  (Silver - CSR2)
22.6 (150-200% FPL)
80  (Gold)
73  (Silver - CSR3)
70  (Silver - no CSR)
60  (Bronze)
weighted average

A question for Kaiser: are buyers with incomes up to 200% FPL likelier to be more satisfied? What about those up to 150% FPL?*

The answer is almost certainly yes. Altman reports that satisfaction among those in lower-deductible plans is 61%, with 30% dissatisfied. For those with higher deductible plans, the split is 46-48%. Over 90% of marketplace customers accessing low deductible plans do so via the strong CSR available to those with incomes under 200% FPL.

The next question is whether enrollees under either income threshold (150% or 200% FPL) are likelier to identify as Democrats. As Altman stresses, as a factor in itself, partisan affiliation weighs even more heavily than deductibles in respondents' perception of help or harm. But if the partisan gap somewhat matches an income gap (a big if -- I don't know whether it does), the partisan lenses may be ground by reality, and least in part.

Update: It's well known that lower income people tend to vote Democratic (and also vote less). But we don't know (or at least I don't) whether marketplace enrollees under 200% FPL lean more Democratic than marketplace enrollees over that threshold.

UPDATE, 5/20: here is the breakout of those who obtained high- vs. low-deductible plans, by party affiliation (among the 90-92% who offered one) and income:

While Democrats and Republicans make up near-identical percentages of those who obtained high-deductible plans, more than three times as many Democrats as Republicans (in an admittedly small sample) obtained low-deductible plans.  I think it's fair to infer that among the large majority of low-deductible enrollees with incomes below 250% FPL, Democrats are far more heavily concentrated than Republicans. It's worth keeping in mind, too, per the footnote below, that as more holdout states embrace the ACA Medicaid expansion, the large chunk of enrollees below 138% FPL (the cutoff for Medicaid eligibility in expansion states) will shrink. That will reduce the percentage of marketplace enrollees with low-deductible coverage, and so probably degrade enrollees' ratings of their plans. In fact, it may already already have.  In 2015, 64% of enrollees in low-deductible plans had incomes under 250% FPL. compared to 59% this year, while the overall percentage in lower deductible plans dropped from 46% to 35%.** In 2015, 74% of individual market enrollees rated their plans excellent or good, versus 64% this year.


* It might be particularly useful to break out the responses of buyers in the 100-138% FPL range. With a handful of exceptions, those buyers exist only in states that refused the ACA Medicaid expansion, which makes Medicaid available to all adults with incomes up to 138% FPL. In non-expansion states, eligibility for private plan subsidies begins at 100% FPL, rather than 138%. Buyers in that range pay just 2% of income for a silver plan that covers 94% of average costs. In nonexpansion states such as Florida, Texas, Mississippi, Alabama and the Carolinas, they're likely to be heavily minority.

There may be some dissatisfaction among such enrollees, because their expenses, while comparatively low,  can be burdensome at that income level, as Kaiser's recent focus group study of newly insured people indicated (Kaiser also reported that many in marketplace coverage were  hit with unexpected bills for services they thought were covered). But still, I suspect they are likelier to be more satisfied than higher income enrollees.

** Updated and corrected: originally, I read misread those percentages as the total percentage of enrollees under 250% FPL.


  1. When you are living at 100% of FPL, any health care expenses over $1 are burdensome. There are many accounts in health care literature where a $10 copay actually lowered utilization. The NY Times had an article not long ago about a woman who had a great ACA plan for $30 a month, but she let it lapse when a car repair took precedence.

    Of course there are other cases where buying cigarettes and I-phones takes away the money for health care. I feel there will always be some portion of our population that need free health care.