Tuesday, May 10, 2016

Clinton's ACA subsidy sweeteners: Something for everyone

Of the roughly 24 million people whose health insurance is currently subsidized as a result of the Affordable Care Act*, over 80% have incomes below 200% the Federal Poverty Level (FPL).

About 14 million more people are enrolled in Medicaid than would have been had there been no ACA. They live in households with incomes below 138% FPL.  Of approximately 9 million subsidized enrollees in private plans sold in the ACA marketplace**, about 60%, or 5.4 million, have incomes under 200% FPL.

If banks are where the money are, the poor and "near-poor," as The CDC's National Health Interview Survey (NHIS) defines people in the 100-200% FPL range, are where the uninsured are most concentrated. A third of the U.S. population lives in households with incomes under 200% FPL. In 2013, just prior to ACA implementation, 55% of the uninsured had incomes below that level, according the most recent update of the Census Bureau's Current Population Survey. As of the first three quarters of 2015, according to the NHIS, the uninsured rate among those under age 65 had been cut by 35% for the poor (under 100% FPL) and by 38% for the near-poor (100-200% FPL).

Those with incomes over 200% FPL were not that far behind: the uninsurance rate for those above that income level fell by 30% from 2013 to 2015, according to the NHIS (from 9.6% to 6.7%).  But that may be a result more of stick than carrot: the tax penalty for remaining uninsured is quite stiff for higher earners. Among those in the upper range of ACA subsidy eligibility, 200-400% FPL, takeup has been poor. Above 200% FPL, subsidies require enrollees to pay between 6.4% and 9.7% of their incomes for benchmark silver plans. Cost Sharing Reduction subsidies fade to insignificance above 200% FPL and phase out entirely at 251% FPL, leaving silver plan holders with high deductibles and out-of-pocket costs.

Here comes Aunt Hillary's cavalry 

The concentration of ACA benefits on the poor and near-poor probably underlies the law's low approval ratings.  Enter Hillary Clinton's wish list. Her proposed package of supplements and sweeteners to ACA subsidies has something for people at all points in the income distribution, very much including those at the top of it (my emphasis below):
Hillary believes that in order to expand coverage for families, we need to reduce the cost of purchasing health insurance on the Affordable Care Act exchanges. Her plan will provide enhanced relief for people on the exchanges, and provide a tax credit of up to $5,000 per family to offset a portion of excessive out-of-pocket and premium costs above 5% of their income. She will enhance the premium tax credits now available through the exchanges so that those now eligible will pay less of a percentage of their income than under current law and ensure that all families purchasing on the exchange will not spend more than 8.5 percent of their income for premiums. Finally, she will fix the “family glitch” so that families can access coverage when their employer’s family plan premium is too expensive.
The tax credit "enhancements" sketched out above seem to be based on a package of proposals published last summer by Urban Institute scholars Linda Blumberg and John Holahan. The cost is estimated by the Center for a Responsible Federal Budget  (CRFB) at about $170 billion over ten years -- actually quite moderate, especially in light of CBO's reductions in estimates of the long-term cost of the ACA's programs as is. In the short run, little to none of it has any chance of becoming law, unless the Trump wrecking ball wrecks mainly the GOP and the Democrats regain control of Congress. In that case, chunks of it might indeed come to pass.

The plan behind Clinton's

Now for the details, insofar as Hillary's plan relies on Blumberg and Holahan's. Noting that the ACA marketplace confronted many uninsured people with premiums they found daunting, with high out-of-pocket costs, or both, the Urban scholars proposed the following subsidy enhancements:

Blumberg and Holahan also proposed reducing out-of-pocket costs by changing the "benchmark" plan on which subsidies are based from silver-level to gold-level. Via Cost Sharing Reduction subsidies (CSR), they would raise the "actuarial value" of marketplace plans -- that is, the percentage of the average enrollee's yearly medical costs that the plan is designed to pay -- as follows:

            N.B. At 250-300% FPL, the silver plan currently has an AV of 70%, with no additional CSR subsidy.

For those above 300% FPL, benchmark gold would be AV 80%, compared to the current benchmark silver (AV 70%).

Raising the AV and lowering the premiums of marketplace plans as outlined here would make them dramatically more attractive to subsidy-eligible buyers, particularly those over 200% FPL.  What interests me here, though, is the prospect of capping premiums for all marketplace customers at 8.5% of income.

Aiding the affluent and near-affluent (and the near-elderly)

The "cap with no cap" is aimed at two problems. First, the current ACA subsidy structure leaves those who earn more than 260% of the Federal Poverty Level (FPL) -- about $30,600 for an individual this year -- paying more than 8.5% of income for a benchmark silver plan (currently the second-cheapest silver plan in each area). Secondly, and more importantly, it eradicates the so-called "subsidy cliff" -- that is, the cutoff of eligibility for premium subsidies at 400% FPL.

There's generally no cliff to speak of for a single young person. Those over 300% FPL ($35,640) have to pay almost 10% of income for a subsidized benchmark silver plan -- and for young adults, an unsubsidized silver plan rarely costs more than 10% of 400% FPL ($47,520).

But premiums for older adults are up to three times as high as for young ones, and the subsidy covers the difference. Thus, according to the Kaiser subsidy calculator, a pair of 55 year-olds earning 399% FPL will pay $513 for a benchmark silver plan covering both of them -- as will a pair of 27 year-olds. If the older couple earns 401% FPL, however, they'll pay $1043 on average (though unsubsidized premiums vary widely by region).  If they have to insure two kids, they'll pay an average of $1340 -- whereas if they earn 399% FPL ($96,757), they'll pay an average of just $770 for the four of them.

Blumberg and Holahan point out that the subsidy cliff leaves the moderately affluent with a very heavy burden of combined premium and out-of-pocket costs:

The 8.5% cap provides significant relief. Below, I've charted how the cap would affect premiums for current benchmark plans for a pair of 55 year-olds at different income levels. Once again, the premiums are the U.S. average, and on the ground they vary widely, both among states and within them. These premiums are silver plans with an actuarial value of 70%. In the Blumberg-Holahan plan -- though not necessarily in Hillary's -- the premium capped at 8.5% would buy a plan with AV 80%.

Cost of average benchmark marketplace plan for two 55 year-olds:
Current vs. with 8.5% income cap

300% FPL
350% FPL
399.9% FPL
400% FPL
500% FPL
600% FPL
Subsidy with
Avg US premium
Avg US premium

Premium with
8.5% cap

Blumberg and Holahan aim their premium cap primarily at those within shouting range of the current 400% FPL subsidy cutoff, claiming that the cap "would not affect many of the higher income individuals potentially eligible for it because premiums would not reach that level relative to income for most of them."

That may be true for younger and mostly single shoppers.  But with no limit to subsidy eligibility, an 8.5% cap on premiums would mean that our 55 year-old couple, paying the $1043 average premium for benchmark silver, would be subsidized up to an income of $147,247. That's 924% FPL. Two 64 year-olds subject to the average premium ($1700 per month) would be subsidized up to an income of  $240,000. That's 1500% FPL.

Stepping down a bit, our 55 year-old couple earning 800% FPL ($127,440) would pay $902 per month under an 8.5% income cap. That's a monthly subsidy of $141. At 700% FPL ($111,510) the premium would be $790, the subsidy, $253.

Why subsidize the affluent?

Subsidies at that income level may sound outrageous. But many affluent Americans who are insured through their employers actually get larger subsidies via the tax exclusion for employer-sponsored insurance, enjoyed by both the employer and the employee. And as the examples above illustrate, U.S. healthcare costs are so high that unsubsidized premiums take a disproportionate share of income even at many multiples of FPL.

Blumberg and Holahan are probably right that the numbers of truly affluent people -- say, over 600% FPL -- taking advantage of the 8.5% cap would be relatively small. For two 40 year-olds, the average unsubsidized monthly premium for benchmark silver this year is $598. That's 8.5% of $84,423, or 529% FPL. And most of the affluent have access to employer-sponsored insurance, barring them from premium subsidies under current law .***

According to the CPS, as of 2014, a bit more than a third of uninsured had incomes that would benefit from an 8.5% cap on premiums. That suggests about 11 or 12 million in that category at present. In 2014, 5.6 million Americans, or 17% of the uninsured, had incomes over 400% FPL Some people in these would probably be excluded from any premium subsidy by access to employer-sponsored insurance, and a few by immigration status.

Extending premium subsidies to people in the $100-200,000 may make political sense, especially given the ACA's current skew toward those in the lower third of the income distribution. In a way, the proposal recalls Democrats' promises not to raise taxes above individuals earning $200,000 (circa 2009) and families earning $250,000.  Upper middle class people vote, and Democrats are generally loathe to alienate them. They paid in political blood to pass the ACA -- but along that rocky road, thanks mainly to conservative Democrats in the Senate, they shorted the middle middle, not to mention the upper middle, classes.

The Senate was penny-wise and pound-foolish. The ACA subsidy schedule works reasonably well for those under 200% FPL, but it's too skimpy above the threshold (and arguably below it too). The exchanges would be healthier with a well-calibrated $20 billion per year boost.

How much of that boost, if any, should go to people with incomes above 400% FPL? Of the roughly $17 billion per year cost that CRFB estimates for Clinton's subsidy package, I don't know what share goes to the 8.5% "cap with no cap." Partly that depends on whether 8.5% of income buys a silver plan (AV 70%) or gold, as Blumberg and Holahan propose (AV 80%).  In any case, the cost is relatively small. And a cliff at any income level (600% FPL?) is still a cliff. Maybe it's more coherent to simply provide that no American household should have to spend more than 8.5% of income for health insurance designed to cover at least 80% of costs. That would be a long step toward making healthcare affordable for all Americans.

Update, 5/11: As commenter Bob Hertz points out below, the Blumberg-Holahan plan calls for a small increase in the Medicare hospital insurance tax on wages. That makes sense as a pay-for for the cap in particular. Like Medicare itself, the benefit would be paid for by all and available to all. While most of us are not in the individual market at any given time, and most indeed are not over 400% FPL, the cap can be thought of as a  kind of stop-loss insurance for all. Any one of us can end up in the individual market at any given time, and for many of us, income fluctuates pretty wildly over the course of a lifetime.


* It should be noted that this is not a measure of how many people have been newly insured by the ACA, as people in both categories (Medicaid and subsidized private plans) may have been previously insured, or might now be insured had there been no ACA. It's a measure of currently enrolled direct beneficiaries of the law.

** That estimate is based on assumed attrition since the close of open enrollment, when there were 10.5 million subsidized marketplace enrollees. Based on scattered state report, Charles Gaba has estimated that about 13% of initial enrollees have dropped off to date.

*** According to current ACA rules, those who have an offer of insurance from an employer are ineligible for subsidies. But Clinton's brief proposal suggests that anyone who buys insurance in the marketplace would have premiums capped at 8.5%. Not excluding those with an ESI offer is probably an oversight -- or perhaps was just judged not worth specifying in a brief paragraph outline.


  1. Thanks for the details. I read the Holohan piece in detail when it first came out and liked it a lot.

    The writers also had solid recommendations of how to pay for it. My memory is that they suggested an increase in the Medicare payroll tax, and/or a cap on the tax-free status of employer-paid health insurance premiums.

    I wonder if Ms Clinton will be honest about paying for her good proposals.

    1. Yup -- "Small increase in Medicare hospital insurance tax on wages." Especially appropriate for the 8.5% cap on all indiv market premiums -- like Medicare, it's available regardless of income.

  2. Phenomenally informative post. Thank you!