What if the Obama administration had ended direct CSR reimbursement in 2016?

Trump's bid to fund his wall with money not appropriated by Congress has set me thinking about the Obama administration's far better grounded, but still dubious, determination to fund the ACA's Cost Sharing Reduction (CSR) subsidies without an explicit appropriation from Congress.

That decision was challenged in court by the Republican Congress in 2014 and, with the suit still pending, countermanded by Trump in October 2017, with disruptive but far from catastrophic consequences.

I'm thinking not of tendentious comparisons between Obama's executive actions and Trump's, but rather about a counterfactual: What if the Obama administration had taken the likely Congressional refusal to appropriate funds at face value, and declined to directly fund CSR?

Two key factors in state-by-state ACA enrollment performance

Last week, I noted that in the 39 states using the HealthCare.gov platform, enrollment declines from 2018 to 2019 were generally steeper in states that had accepted the ACA Medicaid expansion than in states that have so far refused to expand.

Excluding Virginia and Maine, which are expanding Medicaid as of Jan. 1 2019, 13 of the 18 states with enrollment performance above the hc.gov median are nonexpansion states, while 15 of 18 below the median are expansion states (as is Kentucky, at the median).

My working assumption is that a measure of enrollment stability in non-expansion states is grounded in the one third-plus of enrollees in those states who have incomes that would qualify them for Medicaid in expansion states (100-138% of the Federal Poverty Level). This is the income level at which ACA marketplace offerings are most affordable and comprehensive.  At 100-138% FPL, the premium for a benchmark silver plan with an actuarial value of 94% (average deductible $355) costs just 2% of income.

Apparently cutting against this hypothesis is the fact that in HealthCare.gov states in 2018, enrollment dropped 7% at 100-150% FPL while remaining essentially flat at 200-400% FPL, as at the higher income levels, silver loading discounts (see note at bottom) largely offset the effects of cuts to advertising and enrollment outreach. But the drop at 100-150% FPL was much steeper in the expansion states on the platform, where subsidy eligibility begins at 138% FPL (so that the income band is effectively 138-150% FPL). In expansion states on HealthCare.gov, enrollment in this income band cratered 14% in 2018, compared to 6% in the nonexpansion states.* At 138-150% FPL, a benchmark silver plan -- still at 94% AV -- costs 3-4% of income, compared to 2% of income for those below 138% FPL.  And while silver loading reduced enrollment losses at 200-400% FPL in 2018, the resulting discounts in many cases evaporated in 2019 (though new discounts also emerged).**

The chart below sets 2019 enrollment performance beside the percentage of enrollees in a state who obtain silver plans with the highest level of CSR, which raises the actuarial value of a silver plan to 94%.  That percentage is generally much higher in nonexpansion states, where over three quarters of the enrollees in 94% AV silver would be in Medicaid if their state had accepted the expansion. The relationship is significant, I think,  but can be trumped by silver loading effects, as they are in some (not all) of the highlighted states below, as well as by other factors noted below. In states marked in red, the relationship between the percentage of enrollees at 94% AV silver and enrollment performance is most sharply out of whack; in those marked in tan, more moderately so. The apparent effects (or lack thereof) of premium changes from 2018 to 2019 in highlighted states are discussed below.

The median percentage of enrollees with 94% AV silver plans in these 37 states is 19%. The median enrollment change is Kentucky's -5.38%.  I have excluded Virginia and Maine, as they are in the midst of Medicaid expansion.

Sources: Charles Gaba, CMS state-level Public Use Files, 2018.

Actuarial Value in the ACA Marketplace: Down a bit

In 2018, CMS has for the first time ever published data making it possible to precisely calculate the average weighted actuarial value of health plans sold in the 39 states that use the federal marketplace, HealthCare.gov.

The actuarial value is the percentage of the average enrollee's costs a health plan is designed to pay, calculated according to a fixed formula mandated by CMS. Take that with a grain of salt, sprinkled on in a couple of caveats at bottom. Still, AV is a standardized measure that makes comparisons possible. Average AV for both employer plans and Medicare has been calculated to be a bit over 80%.

The ACA mandated specific AV for each of four metal levels (with some wiggle room, per below): 60% for bronze plans, 70% for silver, 80% for gold, 90% for platinum. But secondary Cost Sharing Reduction (CSR) subsidies, available to enrollees with incomes up to 250% of the Federal Poverty Level, complicate the formula. CSR, available only with silver plans, raises AV: to 94% for those with incomes up to 150% FPL; to 87% for those in the 151-200% FPL range; and to 73% at 201-250% FPL.

This year, for the first time, CMS reported exactly how many people were enrolled at each CSR level, albeit only in HealthCare.gov states, which account for just about three quarters of marketplace enrollees.* So who obtained what in the ACA marketplace this year? Here's the data.

Actuarial value obtained by enrollees on HealthCare.gov, 2018 

Metal level	Actuarial value	Total enrolled	% enrolled	Weighted AV
Silver - CSR1	94%	2,630,842	30%	28.20
Silver - CSR2	87%	1,437,131	16%	13.92
Gold	80%	528,087	6%	4.80
Silver - CSR3	73%	656,895	8%	5.84
Silver - no CSR	70%	963,242	11%	7.70
Cat/platinum*	64%	74,907	1%	.64
Bronze	60%	2,452,538	28%	16.80
Total		8,743,642	100%	77.9%

* There were just 16,731 platinum (90% AV) enrollees on hc.gov, and 58,176 catastrophic plan (57% AV) enrollments. Combined, that comes to a weighted AV of 64%, used above to avoid splitting percentages.

Covered California to middle-class enrollees: There (may be) gold in them thar hills

Update (!) Oct. 11, 2017: CoveredCA has gone live with  its 2018 plan preview tool and has implemented the CSR surcharge discussed below -- 12.4% added to silver on-exchange plans only. That makes gold cheaper than silver in some areas, and CoveredCA is indeed in some cases highlighting gold plan offerings ahead of silver.

Back on August 1 Covered California, the California ACA exchange, announced a proactive approach to enable health insurers to cope with the uncertainty over whether the Trump administration (or Congress) would guarantee them continued reimbursement for Cost Sharing Reduction subsidies.

Insurers would file rate requests under the assumption that CSR will continue to be reimbursed. If reimbursement was not guaranteed by a certain date, however, a surcharge (later pegged at 12.4%) would be added to silver plans sold on-exchange. CSR is available to enrollees with qualifying incomes only if they select silver plans, and only if they buy on-exchange. ACA-compliant silver plans sold off-exchange would not include the surcharge. That surcharge is likely to be triggered on or around October 11, CoveredCA tells me (postponed from an original target of Sept. 30).

The likely results have been widely noted (e.g., here). Off-exchange, plan pricing would remain proportionate to the actuarial value offered at each metal level. On-exchange, the only buyers of silver plans should be those with incomes up to 200% of the Federal Poverty Level, for whom CSR raises the value of silver plans roughly to that of platinum.

Mysteries of the Maryland ACA Marketplace, cont.

As I noted on Oct. 3, Maryland Health Connection, the state's ACA exchange, sent me enrollment data indicating that Maryland may have the highest "CSR takeup rate" in the country. That is, in 2015, Maryland private plan buyers whose incomes qualify them for Cost Sharing Reduction (CSR) subsidies, which are available only with silver plans, chose silver and access the benefit at market leading rates. In Maryland, 86% of CSR-eligible buyers chose silver, compared to a national average of about 76%. That's doubtless in large part because Maryland cloned the web interface of the Connecticut exchange, which does an excellent job steering CSR-eligibles toward silver.

There are some peculiarities in the Maryland data, though, and they're thrown into sharper relief by the raw numbers of silver plan enrollees at each income level, which I did not have when I posted last week. Chief among them is a relatively huge number of buyers, 22%, with incomes low enough to qualify them for Medicaid, unless they're lawfully present non-citizens not yet eligible for Medicaid because they've been in the country less than five years.

In Washington State, too many low-income bronze plan buyers

Washington HealthPlanFinder, the state's ACA exchange, has set the standard for enrollment data reporting, providing a more detailed and complete account of private plan buyers' demographics and behavior than any other state to date. Washington is a wealthy state, with a median household income (2013) of $60,106, compared to a national median of $51,939.   Its buyers of private plans on the exchange (known as Quality Health Plans, or QHPs) are accordingly a much wealthier group than the average among the 37 states that used the federal exchange, healthcare.gov.

Only 12.5% of Washington's QHP buyers have incomes under 150% of the Federal Poverty Level (FPL), compared to 24% in those healthcare.gov states that accepted the Medicaid expansion -- and 50% in healthcare.gov states that refused the expansion. (In non-expansion states, eligibility for QHP subsidies began at 100% FPL, versus 138% FPL in expansion states, and those between 100 and 138% FPL swelled the QHP enrollments, accounting for about a third of all enrollments in non-expansion states.) Low takeup in this low-income band perhaps explains in part why Washington has reached just 32% of its target QHP market as calculated by the Kaiser Family Foundation, versus 42% for the nation as a whole. Conversely, Washington has been very successful in expanding its Medicaid rolls. The enrollment report tallies 533,628 "Medicaid expansion adults," far exceeding a 2012 Urban Institute forecast of 330,000.

Too many poor buyers of bronze plans

While Washington's relatively small number of enrollees in the 100-150% FPL income band may be in large part a matter of demographics, there is one way in which the state exchange has seemingly failed lower-income buyers. Takeup of Cost Sharing Reduction subsidies, available only with silver plans and only to buyers with incomes below 250% FPL, is lower in Washington than on healthcare.gov, and much lower than in states like New York and Connecticut that take special measures to highlight CSR for those who are eligible for it.

HHS expected *more* low-income ACA buyers to select silver plans

In my multi-stage exploration of the data on ACA private plan buyers' metal level selections, I have been pleasantly surprised to determine with a fair degree of confidence that about 90% of shoppers on healthcare.gov with incomes under 200% of the Federal Poverty Level selected silver plans -- and so availed themselves of the Cost Sharing Reduction (CSR) subsidies available only with silver. For all ACA buyers on all exchanges, the figure is a few percentage points lower, since many state exchanges did a much poorer job than healthcare.gov in steering CSR-eligible buyers toward silver.

While CSR is available to buyers with incomes up to 250% FPL, the rate of silver plan selection (necessary to access CSR) falls off a cliff at 201% FPL. That's wholly rational, as CSR weakens as you move up the income scale. For those with incomes between 200 and 250% FPL, CSR bumps up the actuarial value of a silver plan -- that is, the percentage of the average plan holder's yearly medical costs paid by the insurer -- just three points, from 70% to 73%.  At lower income levels CSR is much more valuable, raising AV  to 87% (for incomes at 150-200% FPL) or 94% (for incomes under 150% FPL). Accordingly, in New York (one of the few states that breaks out metal level selection by income level), 89% of buyers with incomes under 200% FPL bought silver -- but just 59% of those between 200 and 250% FPL did.

With all this in mind, I was surprised to read* in the latest CBO updated ACA cost projections that the Department of Health and Human Services and the Joint Committee on Taxation have been surprised that more buyers eligible for CSR have not selected silver plans:

Outlays for cost-sharing subsidies over the 2015–2024 period are currently projected to be $39 billion less than previously estimated, primarily because CBO and JCT now expect that more people will forgo those subsidies by choosing to enroll in a bronze plan instead of a silver plan...

The ACA's invisible deductible discounts

The healthcare Twittersphere was riveted yesterday by Robert Pear's spotlight on Harvard professors up in arms about comparatively modest deductibles and copays added to their excellent health plan.

The main underlying story, as Adrianna McIntyre pointed out, is that cost-shifting to employees is accelerating in employer-sponsored insurance, driven in part by the ACA's Cadillac tax on the most generous plans.

There's a second story, however, in Pear's use of the ACA exchanges as a foil to highlight Harvard employees' relative privilege. Pear contrasted the Harvard plan's 91% actuarial value with that of a silver ACA plan, which he said "typically" covers just 70% of the average user's medical costs. As I pointed out yesterday, only about 20% of silver plan buyers get plans with an AV of 70%. For the rest, income-based Cost Sharing Reduction (CSR) raises the AV -- to 94%, 87% or 73%, depending on the buyer's household income. Available stats indicate that over 60% of silver plan buyers are at the AV 87% or 94% levels.

The broader point here is that the apparent prevalence of high deductibles on the ACA exchanges is somewhat misleading. As has been widely reported, bronze plans single-person deductibles average over $5,000, and silver plans (unenhanced by CSR) close to $3,000. Moreover,if you take a spin on the shop-arounds offered by Healthcare.gov and most state exchanges, the first price quotes you'll see are for bronze plans with deductibles in the $5,000-6,600 range, since available plans are most often displayed sorted by premium, lowest first.

Not quite the Harvard plan, but....

The NYT's Robert Pear reports that Harvard profs are up in arms because a small deductible and copays have been added to the university health plan, taking the actuarial value all the way down to... 91%.

Pear contrasts this to the 70% AV mandated for silver ACA plans, the "most popular" metal level. Which is fine, except...more than half of silver plan buyers qualify for Cost Sharing Reduction (CSR) at a level strong enough to take the AV up to 87% or 94%, depending on income level.

I have a post coming up on healthinsurance.org that teases out of enrollment stats published by New York State and HHS the likely percentages of ACA private plan buyers who access the higher levels of CSR, and the overall percentages of direct ACA beneficiaries who have obtained high-AV insurance.

The broader point: clunky or overly complex as the ACA's subsidy and means-tested benefit formulas may seem, they do work broadly to keep out-of-pocket costs relatively low for low-income beneficiaries.

Hope you'll tune in to the healthinsurance.org piece.

Update: The New York enrollment numbers also show that only 18% of silver plan buyers in the state had no CSR at all. I tend to discount the value of CSR to buyers with income between 200-250% FPL, for whom CSR boosts the silver plan AV to just 73%.  But it's really inaccurate to say that silver plans "typically" cover 70% of costs. That's probably true for less than a quarter of silver plan buyers.

New data: health exchange design matters

Why was there such huge variation among states in the proportion of ACA private plan buyers in the first open season who bought bronze plans -- the plans with the lowest premiums and highest deductibles and copays? In Hawaii, 41% of ACA shoppers bought bronze; in Mississippi, 8% did.

Part of the answer, as I've noted before, lies in a state's relative levels of wealth and health. Lower income buyers are eligible for generous Cost Sharing Reduction (CSR) subsidies that reduce deductibles, co-pays and yearly out-of-pocket (OOP) maximums -- but only if they buy silver plans.

Fortunately, most did. If you're sick and poor, a $6,000 deductible is likely to give you pause -- even if the plan is all but free and you don't come in knowing what a deductible is. On Healthcare.gov, only 15% of buyers eligible for any kind of subsidy bought bronze. The percentage is probably considerably lower among those eligible for strong CSR subsidies -- that is, buyers with incomes under 200% of the Federal Poverty Level (data from states that broke out metal level selection by income band, cited below, suggests as much).

Another factor plainly has a strong impact, though, and may account for some wealth/health anomalies in state performance. That's website design. In Connecticut, which had a 2013 median household income of $67,781, second highest in the nation, just 16% of all buyers in the first open season selected bronze. In Colorado, with a median income of $63.371, 40% bought bronze.

That's doubtless because the Connecticut site shows CSR-eligible applicants silver plans first; that is, the menu of plans available to a given user defaults to silver, both in the pre-application "shop-around" feature and in the actual application. The Colorado site, in contrast, does next to nothing to steer CSR-eligible buyers toward silver. The shop-around is cumbersome; the filter by metal level is hard to find (at the bottom of the screen, and you have to scroll back up to activate it). Perhaps more importantly, unlike on Healthcare.gov, applicants who qualify for CSR and make a move to buy a bronze plan receive no warning that they're leaving benefits on the table.

New data from Connecticut

The strongest evidence of the impact of site design comes from data about the choices of buyers with household incomes below 200% FPL (CSR is available but much weaker for those between 200-250% FPL). HHS did not provide this information for the 36 states that used Healthcare.gov in the first open season, and neither did most states. Colorado did, however, and so did New York (median 2013 household income $53,843).  And now, Access Health Connecticut has provided me with CT numbers for the current open season, which began on November 15.

Obama: Drop that "Plans for under 100!" pitch

Hey, Mr. President: you may have the right audience here. But you have partly the wrong message:

Obama appeared Dec. 8 on Comedy Central’s “The Colbert Report,” where he took the place of host Stephen Colbert for a regular segment called “The Word” -- retitled “The Decree” for his appearance -- to pitch enrollment to young viewers.

“Most young people can get covered for less than $100,” Obama said, using lines purportedly meant for Colbert. “How is the president going to get that message out to the kids? 

The administration really needs to take down this "plans for under $100!" banner. Here's why: if an adult under age 35 has an income low enough to get a plan with a premium under $100, she almost certainly qualifies for Cost Sharing Reduction (CSR) subsidies that reduce deductibles and copays -- but only if she buys a silver-level plan. Keeping the premium under $100 for a single person in many cases means buying a bronze plan -- and the average deductible on a bronze plan is over $5,000.