Thursday, June 22, 2017

Trading Medicaid coverage for high deductible private market coverage

The Senate version of the AHCA, the Orwellianly named Better Care Reconciliation Act, does even more fundamental damage to the U.S. healthcare system than the House bill. While it phases out the ACA Medicaid expansion more slowly, stepping down the enhanced federal contribution over several years, it imposes even tighter per capita caps on Medicaid, limiting annual growth to the straight CPI after 2024. The damage to Medicaid will be continuous in perpetuity, barring further legislation.

The BCRA does toss a bone to the dis-insured poor by offering private-market subsidies to those who are shut out of Medicaid. Under the ACA, in the 31 states plus D.C. that accepted the law's Medicaid expansion (rendered optional to states by the Supreme Court), anyone whose household income is below  139% of the Federal Poverty Level (FPL) qualifies for Medicaid, and so not for subsidies in the private plan marketplace (with one class of exceptions*).  In states that refused the expansion -- a possibility not envisioned by the law's drafters -- eligibility for Marketplace subsidies begins at 100% FPL, and those below that level are left out in the cold -- because their state's governors and legislatures wanted it that way.

The BCRA allows people with incomes in 0-100% FPL range to buy a "benchmark" plan for 2% of income, and those in 100-133% FPL range** to buy one for no more than 2.5% of income:

But there's a huge catch. The ACA based income-adjusted subsidies are based on a benchmark "silver" plan with an actuarial value (AV) of 70%, meaning the plan is designed to cover 70% of the average enrollee's annual medical costs. For low income enrollees - the majority of marketplace enrollees -- silver plans are enhanced by Cost Sharing Reduction (CSR) subsidies that raise AV to 94% for those with incomes up to 150% FPL, 87% for those in the 150-200% FPL range, and 73% (a near-negligible boost) for those in the 200-250% FPL range.  That usually means deductibles in the $0-250 range for people with incomes up to 150% FPL and $500-1000 for the 150-200% FPL cohort. Medicaid enrollees in expansion states generally have no deductible and little to no out-of-pocket expense.

The Senate bills drops the AV of a benchmark plan to 58% -- below that of the ACA Marketplace's bottom-level bronze plans, which have an AV of 60%. Bronze plans generally have single-person deductibles over $6,000 -- a fact Republicans have cited endlessly in asserting that the ACA makes coverage unaffordable. As I noted late last year, for example, here is Johnny Isakson, responding when asked what might become of the 20 million newly insured by the ACA if it's repealed:
Most of those 20 million got bronze policies with a great big deductible and not much insurance, so I don’t know that there’s going to be a big backlash,..There are some minefields out there but we can deal with them.
And here's Bill Cassidy, expressing purported concern for low income ACA enrollees:
Right now, that person might have a $6,000 deductible, which for someone who makes 150 percent of the federal poverty line might as well be $6 million. 
Isakson's assertion is a flat-out lie, and Cassidy's is misleading at best. Over 85% of ACA marketplace enrollees with incomes up to 200% FPL -- who constitute almost two thirds of all enrollees -- select silver plans and so access Cost Sharing Reduction subsidies. It's true, as Cassidy avers, that an enrollee with an income of 150% FPL might have a $6,000 deductible, but most don't.  In fact the weighted average AV for marketplace enrollees with incomes up to 200% FPL is about 86% -- better than the average for those who get insurance from their employer. In any case, "most of those 20 million" who newly gained coverage did so through the Medicaid expansion and have zero deductible.

The BCRA ends CSR in 2020, when it introduces its new subsidy schedule. At that point, plans skimpier than the bronze plans Republicans have been excoriating for four years will be the principle offering for people who now have access to Medicaid -- or would, if their state of residence had not rejected the expansion. In expansion states, under the BCRA, many Medicaid expansion enrollees will have to transition from premium-free coverage at AV 98-100% to coverage costing 2%*** of income and providing AV 58%.  They'll be moving from no deductible to a $7,000 deductible -- unless their states opt out of ACA coverage rules and so enable insurers to offer coverage with lower out-of-pocket costs  for those more limited services that they do cover (possibly excluding services such as maternity care, mental health care or pharmaceuticals). [Update: David Anderson notes that states could use BCRA "stability funds" to reduce out-of-pocket costs to some degree for low income enrollees, but not nearly to the extent CSR -- or Medicaid -- do.]

In short, about 16 20 million people currently covered either through the ACA Medicaid expansion (11 approaching 16 million) or Marketplace coverage with strong CSR (5 million) (or rather, their successors in similar situations) will soon have access only to private coverage that renders actual healthcare prohibitively expensive. [Update: Charles Gaba points out to me that my Medicaid expansion numbers are out of date, partly because some who were covered by waiver or on a state's dime pre-ACA are now receiving the enhanced federal match for the expansion population. Longish story; see Charles's post.]

I don't mean to sugar-coat the weaknesses of the ACA marketplace. Out-of-pocket expenses are too high for most buyers with incomes over 200% FPL, and premiums are painfully high for many of the modestly affluent, especially older buyers in the 400-600% range. The subsidies were too skimpy from the start. But this legislation will render conditions far worse. Not only will out-of-pocket costs in private plans be prohibitively high, but with the individual mandate repealed, premiums are likely to skyrocket.

And the ill effects of all this mis-engineering and short-dollaring pale in comparison to the bill's evisceration of Medicaid. I'm almost ashamed to have burned a post on the individual market, as the bill's long-term crime is to strangle federal Medicaid funding in perpetuity. But then, part of the crime here is to push Medicaid enrollees into this misbegotten commercial market.

UPDATE, 6/24: I am reading repeatedly (e.g. here), as a basis for comparison with BCRA benchmark plans at AV  58%, that benchmark silver plans under the ACA have an average deductible of ~$3,500. While that is true with respect to the way insurers need to structure a silver plan independent of CSR, which is credited and reimbursed separately, it distorts the contrast between what people actually get in the ACA marketplace vs. the BCRA.  A July 2016 CMS data brief shows that as of April 2016, fully 52% of all enrollees were enrolled in silver plans enhanced by CSR to AV 94% or 87%, which, as noted above, translates to deductibles in the $0-250 range for the former and $500-1000 for the latter. Based on the CMS chart, the average weighted AV for silver plan enrollees in April 2016 was 85%, which suggests a deductible under $1,000 (more on average AV for different income groups here).  The average weighted AV for all enrollees at this point was 81.4%, about comparable to employer sponsored insurance. Here is the CMS breakout:

A caveat: this breakout reflects enrollment on only, covering 38 states, and as of April 2016, when the post-open-enrollment attrition had already been recorded. After the attrition, the percentage of enrollees in silver plans ticked up; the overall percentage of enrollees with "strong CSR" -- AV 94% or 87% -- was about 49% immediately after enrollment.
Related: The real cost of ACA benchmark silver for low income BCRA enrollees


* Legally present noncitizens who are time-barred from Medicaid enrollment are eligible for marketplace subsidies even if their income is below 100% FPL.

** In practice, the 133% FPL cutoff is 138% FPL, since all applicants get a 5% initial deduction.

*** The benchmark plan will cost 2% of income; a plan cheaper than the benchmark could cost less.

1 comment:

  1. Many thanks. I am a long-time reader and benefit from your clear, but also detailed, work on this issue. I can't get as good information anywhere else.