Saturday, May 10, 2014

McKinsey's really important finding: the still-uninsured don't know about ACA subsidies

McKinsey & Company is out with a new survey -- conducted online and in English only -- of participants in the 2014 individual market for health insurance. One headline stat seized on by ACA skeptic Avik Roy (see note below) is that of those who purchased ACA-compliant plans, just 26% reported themselves uninsured in 2013.  That number is apparently at wild variance with HHS' report that of the 5.18 million enrollees who applied for financial assistance (95% of the total), 87% reported being uninsured at the time of application.The huge difference is due in large part to these factors:

1) HHS asked those applying for financial assistance whether they were insured at the time of application. McKinsey asked whether they were covered through most of 2013.

2) McKinsey's numbers include people who bought off-exchange -- and people not eligible for subsidies* (as Charles Gaba has noted).

3) McKinsey's respondents were self-selected via an online survey, and the survey was conducted in English only.

The truth probably lies somewhere in between. Charles Gaba estimates that about half of ACA exchange signups were previously uninsured. The question of who was previously uninsured is in any case complex, given the high volatility of the individual market for insurance (and of the job market). According to an estimate by healthcare scholars Rick Curtis and John Graves, less than half the people who will be eligible for ACA enrollment at the end of 2014 will have been eligible a year prior. According to Graves and Jonathan Gruber, 7.6 million people lost health insurance in 2012; approximately the same amount must have gained it.  That kind of churn is perpetual.

While the questions, "how many signups were uninsured?" and "how many paid?" suck up all the political air, the real import of the McKinsey survey (to the extent that it is representative) lies elsewhere, in my view. The survey indicates pervasive ignorance among the subsidy-eligible of what the ACA had to offer them.

Forty-nine percent of the 2,874 survey participants did not enroll in any insurance plan, on- or off-exchange*  Of those 49%, 44% did not shop for insurance; 56% shopped but did not buy. Most cited cost as the key factor keeping them out of the market. For most of those unfortunates, the really determinative factor was ignorance (my emphasis):
In our April survey, 72 percent of the respondents who reported that they shopped but did not buy were both previously uninsured and subsidy-eligible. Perceived affordability remained the most common reason for exiting the purchase process; it was cited by 59 percent of all April respondents who reported shopping but not enrolling and by 64 percent of the subsidy-eligible respondents who reported that behavior. Eighty-eight  percent of all those citing perceived affordability challenges were subsidy-eligible. Consistent with our February findings, most of the subsidy-eligible respondents (66 percent) who cited perceived affordability as the reason they stopped shopping were aware of neither their eligibility nor the amount for which they were eligible.

Nevertheless, both factors appear to be strong contributors to enrollment. For example, previously uninsured, subsidy-eligible shoppers who indicated that they knew their subsidy Amounts were almost three times as likely to report having enrolled as those who did not (Exhibit 6). Consistent with this finding, 71 percent of the respondents who reported shopping but not enrolling indicated that they would be more likely to enroll if they had more information about the cost of different plans (including subsidy eligibility, net -of-subsidy premium amounts, out-of-pocket maximums, and expected out-of-pocket total cost).

Perceived affordability was also the reason offered most often for not enrolling among those reporting that they had not even shopped for coverage; it was cited by 71 percent of all respondents in this group and by 75 percent of the subsidy-eligible individuals in this group. Ninety percent of those citing perceived affordability challenges who did not shop were subsidy-eligible. Seventy-nine percent of the subsidy-eligible respondents who cited perceived affordability as the reason they did not shop were unaware of their eligibility or the amount of subsidy for which they were eligible.
Those who deliberately spread disinformation about the ACA and actively encouraged the uninsured to remain in that blessed state of freedom can be really proud of themselves.

That said, educating the uninsured about ACA offerings was always understood to be a slow undertaking. All CBO projections, starting in 2009, lay out a four-year path to full or nearly full takeup; the most recent estimate envisions 13 million exchange enrollments in 2015, 24 million in 2016, and 25 million in 2017. Red state governments' refusal to actively promote ACA offerings -- and in some states, measures to actively impede outreach -- are doubtless taking a toll. At the same time, it must be admitted that ignorance is not confined to states with ACA-hostile governments. In Kentucky, under the leadership of Democratic Governor Steve Beshear, outreach has been vigorous and effective: the state signed up almost all of its Medicaid-eligible and 33% of those estimated to be eligible to buy private plans on the state exchange. Yet only 54% of Kentuckyans surveyed in April by the New York Times and the Kaiser Family Foundation knew that "the law provides financial help to low and moderate income people to purchase coverage."

Continued education and outreach is a vital priority in the long slog to get affordable coverage to the bulk of uninsured Americans. It's too bad that a GOP Congress will impede funding to power that outreach, and that GOP state governments won't lift a finger to help uninsured citizens access the coverage that the law makes available -- not to mention blocking the Medicaid expansion that would make coverage available to their neediest citizens, as 24 states are still doing.

*     *     *

* Roy neglects to mention how McKinsey defines "unininsured" and erroneously asserts that McKinsey is surveying the subsidy-eligible only. He therefore downplays the fact that McKinsey stats for those who bought "ACA plans" include those who bought off-exchange.UPDATE: Roy has corrected his assertion that McKinsey surveyed the subsidy-eligible only, while neglecting to note that I alerted him to the error at 8:40 ET on Saturday morning, with Charles Gaba then picking up the thread. And he hasn't changed his misleading headline: "New McKinsey Survey: Only 74% of Obamacare Signups Were Uninsured." That includes off-exchange buyers as "Obamacare signups," as well as oversimplifying generally, per the caveats above. I've previously noted similar sleight-of-hand from Roy re the ACA and healthcare reform:

What Avik Roy won't tell you about healthcare in Switzerland and Singapore
No, Avik, 93 million Americans will not lose their health plans under Obamacare 

**Of those McKinsey respondents who did buy insurance, 53% either renewed a prior policy or bought a pre-ACA policy with an effective date prior to Jan. 1, 2014.


  1. This is all nonsense.

    McKinsey did not estimate that only 26% of individual market QHP purchasers were “previously uninsured”. Nor did they offer any evidence for such a conclusion. To do so, one would have to draw a random sample from the population of QHP buyers, and then count the number of previously uninsured in that sample. McKinsey did not do that. Rather they drew several samples from different populations and then aggregated the results. You can’t extrapolate to any population if you do this – you can only look for trends, comparing results to previous, identically constructed studies. Which is what McKinsey did. This why they called their survey "directional only".

  2. By the way - I should have mentioned that your headline is dead on. That is the real take-away from the McKinsey survey. I hope people are paying attention.