Why the ACA remains unpopular, cont.

Over at the Huffington Post, Jonathan Cohn has teamed up with pollster Mark Blumenthal for a deep dive into why the ACA's approval ratings remain underwater* and why more people continue to say that the law has personally harmed than helped them (though the gap had narrowed. to 22-19 when Kaiser last polled this question in March).

There are two main takeaways: 1) polling results are overwhelmingly partisan, and Republicans are more passionate in their hatred of the law than Democrats are in support of it, and 2) Americans tend to attribute any changes in their health plans -- usually price hikes or coverage cutbacks -- to the ACA. That's especially true of people with employer-sponsored insurance, a third of whom said they'd been hurt by the law.**

While those conclusions are spot-on, and Cohn and Blumenthal provide a nuanced overview of the ACA's effects on various groups, I'd like to throw one sidelight and add a couple of caveats.

First, the sidelight. Noting that the largest category of those who say the law hurt them say it drove their costs up, Cohn and Blumenthal suggest that the perception is not accurate:

But insurance premiums go up every year. That was true before the ACA became law and it remains true after. And since the law's enactment they've actually risen more slowly than before. The historically low inflation is actually one of the most remarkable developments in health care today. And while economists debate over what role, if any, the health care law has played in this progress, there's no compelling evidence that the law has made employer insurance premiums -- the premiums most people see -- rise more quickly.

Similarly, out-of-pocket costs really are rising and they are rising more quickly than wages, which is a big reason why people feel their impact. The law's critics frequently complain about rising deductibles, as if the law were responsible for them -- citing, among other things, a recent Commonwealth Fund report on the increases in out-of-pocket costs for consumers. But as Sara Collins, a vice-president of the Fund and co-author of that study, told HuffPost, "the trend in higher deductibles began well before the Affordable Care Act... the trend is entirely separate from the Affordable Care Act." 

What I'd like to add is that people blame cost-shifts and price hikes in their employer-sponsored insurance on the ACA because their employers tell them to. That's reflected in a poll of employers conducted early this year by the Society for Human Resource Management (SHRM). Here's one key finding:

The biggest changes from 2013 are that more organizations have communicated the impact of the ACA to employees (from 32% o 67%) and to retirees (from 7% to 17%).

Reflecting on the SHRM results in April, I uttered a plaint similar to Cohn and Blumenthal's:

The ACA has had and is likely to continue to have a complex and mixed effect on employer-sponsored insurance. Thus far, Kaiser's findings indicate that the marked slowdown in recent years of overall healthcare spending growth -- which may be partly attributable to ACA cost controls and pilot programs --  has offset new expenses to employers such as the requirements to cover employees' children up to age 26 and the ACA's new taxes on insurers and drug and medical device makers.

At the same time, a majority of employers are now attributing changes in health insurance to the ACA.  Usually such changes include hikes in employee costs or reductions in benefits. While such cuts and price hikes have not accelerated in recent years, they've been continuous for so long that many employees are exposed to thousands of dollars in out-of-pocket costs. Many probably now blame what's been a relentless climb on the ACA.

At the same time, it should be acknowledged that the ACA has put some upward pressure on employee costs, and while the law's cost control measures may have so far roughly offset (or helped offset) those pressures, that may not be the case going forward.  In the SHRM survey, 33%, of responding employers said that they plan to take action to ward off the so-called "Cadillac" tax on plans offering particularly rich benefits, scheduled to take effect in 2018 -- though only 9% have taken such action to date.

Why the cost of unsubsidized coverage rose in the individual market

My other caveats and quibbles with Cohn and Blumenthal are really just matters of emphasis. Regarding the wrath of the rate-shocked -- those who bought insurance in the pre-ACA individual market and saw their premiums rise when the law was implemented --we're told that the ACA's market reforms "raised the underlying costs of these policies, since previously carriers could sell plans that left out key benefits (like prescription drugs or mental health) or simply refuse insurance to people who already had medical problems."

That's true, but it's important to note that "guaranteed issue" -- the ban on denying coverage or charging more on the basis of the applicant's medical history -- had a much bigger impact on pricing than did the ACA's mandated Essential Health Benefits (EHBs) and other coverage rules. Republicans love to beat up on the EHBs -- it's easy to note that a rate-shocked 55 year-old doesn't need maternity coverage -- but they usually make at least vaguely sympathetic noises about guaranteed issue, the ACA's most popular provision.  As I've noted before, analyses conducted both prior to and after ACA implementation indicate that guaranteed issue is the main driver of post-ACA increases in the price of unsubsidized individual market plans:

...benefits consultant Milliman estimated in March 2013 that guaranteed issue would drive the cost of insurance in California up 26.5%.  The ACA's benefit requirements, in contrast, would raise premiums only by an estimated 4.8%. More recently, with the data for the ACA's first open season available, a NBER study by a team of health economists led by the Wharton School's Mark Pauly identified guaranteed issue as the primary cause of cost increases averaging 14 to 28% in 24 states.

Footnote: more than half of individual market customers are unsubsidized

As for who was harmed (at least temporarily) by those increases: Cohn and Blumenthal note that most private plan buyers who obtain their insurance on ACA exchanges are satisfied with their coverage and its cost, according to surveys by Kaiser and Commonwealth. They also note that exchange customers account for somewhat more than half of all those buying insurance in the individual market.

I would add, though, that according to the May 2015 Kaiser survey of individual market customers, a bit more than half of those who buy their insurance in the individual market are not subsidized. Sixteen percent are in grandfathered or grandmothered noncompliant plans, 24% bought an ACA-compliant plan off-exchange, and about 13% of those who bought on-exchange earned too much to qualify for subsidies.

Liz Hamel, Kaiser's lead researcher in the May survey, gave me a breakout for the 24% of respondents who bought ACA-compliant plans off-exchange. In that subgroup, Hamel wrote, "30% said they’ve benefited from the ACA and 40% said they’ve been negatively impacted." Among the 16% of respondents in non-compliant plans (i.e., those who don't want to let go of plans they held pre-ACA), just 8% said the law had helped them, and 41% said it had harmed them.

While we don't have a breakout for those who bought on-exchange and did not qualify for a subsidy***, it's fair to say that the majority of the more-than-half of individual market customers who are unsubsidized are more likely to say that the law has harmed than helped them. The fact that the negative tilt is relatively modest for those who bought ACA-compliant plans off-exchange (30-40) may reflect the high percentage of Americans (estimated in various government surveys in a range from 19-50%) who have a pre-existing condition. Many of them could not get insurance at all in the pre-ACA market or had to pay more than they're paying now, even if they don't qualify for subsidies. Those who are still in noncompliant plans are almost by definition healthy or at least were not significantly penalized for any health issues.

In the individual market as a whole -- subsidized and unsubsidized, on-exchange and off-exchange, in ACA-compliant and ACA-noncompliant plans -- 40% said that they had benefited from the ACA, and 33% said that they had been negatively affected. [Per update below, in Kaiser's regression analysis published last week, the favorable/unfavorable split was 51-43 in favor].

Added to the plus side of the ACA "helped/harmed" ledger should be the 12 million-plus who have taken advantage of the ACA's Medicaid expansion. That underscores another ACA perception problem:: the overwhelming majority of the law's most direct beneficiaries are low income. The Medicaid beneficiaries have incomes under 138% of the Federal Poverty Level (FPL), and most private plan buyers -- 68% in the 37 states using healthcare.gov -- are under 200% FPL.  That is good policy but tough politics.

UPDATE: I neglected above to note that Cohn and Blumenthal link to a Kaiser regression analysis of its individual market polling that Kaiser put out last week, weighing the factors that might lead to respondents saying that the ACA has helped or harmed them. While partisanship was the biggest driver and ultimately the only reliable predictor, as Cohn and Blumenthal stress, income also had a large effect, as my crude parsing above indicates. Naturally enough, those who are ineligible for subsidies are far likelier to say that the law harmed them, and those with incomes low enough to qualify them for Cost Sharing Reduction subsidies are likeliest to say that they benefited. Here's the upshot:



In case the type's too small: The predicted likelihood that the “average” non-group enrollee will say the law has negatively impacted them is 40 percent for those with incomes above 400 percent of the federal poverty level (FPL) compared with 23 percent for those with incomes less than 250 percent FPL.

I would like to see a score for those with incomes below 200% FPL instead of (or alongside) those with incomes below 250% FPL. Cost Sharing Reduction subsidies (available only with silver plans) are really strong under 200% FPL and fade away to almost nothing at 201-250% FPL; CSR takeup accordingly falls off sharply at 201% FPL. As a rule, the lower your income, the likelier you are to be satisfied with the ACA's private plan offerings. The most satisfied customers are probably those with incomes from 100-138% FPL in states that refused the Medicaid expansion -- if their states had accepted the expansion, they would have been eligible for Medicaid. At that level, the cost of a benchmark silver plan premium is capped at 2% of income, and CSR raises the plan's actuarial value to 94%.

According to Kaiser, partisanship is the only reliable predictor of whether a person will have a favorable or unfavorable view of the law -- "other demographic factors..were not significant predictors." Given a large enough sample size, however, I imagine that that would be true only up to a point. Would a Republican silver plan buyer with an income of 101% FPL be likelier to say the law has harmed or her than a Democrat who earns too much to qualify for a subsidy (and buys a bronze plan with a $6,000 deductible)?

UPDATE 2, 6/24: Just reflecting on the overall 51-43 favorable/unfavorable ACA rating among all nongroup market participants Again, a slight majority of nongroup market customers are unsubsidized, and their ratings are understandably negative. Of course, those nongroup market participants who view the ACA positively are a lot poorer than those who view it negatively -- most of them are what the National Health Interview Survey (NHIS) deem the "near-poor," with incomes between 100-200% FPL.  These reactions illustrate the tough tradeoffs that Democrats embraced in crafting and passing the ACA. In brief, they raised prices on the then-majority of long-term nongroup market participants to protect those with preexisting conditions (which eventually could be any of us) and give access to those who were previously priced out. Republicans would like to unwind those tradeoffs but not make any of their own to expand access beyond where it was in 2010.

----
* The April Kaiser tracking poll actually had the ACA in positive territory, 43-42. But by the June update it was back in negative territory, 39-42, and polling averages still have it underwater (42.6-49.2 according to HuffPost's long-term average, going back to the beginning of 2013).

** That's according to Cohn and Blumenthal. Unless I missed it, Kaiser's published survey results don't show results specifically for those with ESI.

*** Those who bought on-exchange and did not qualify for a subsidy were naturally significantly less pleased than the subsidized. Among the subsidized, 57% said that they benefited and 25% said that they were negatively affected. Among all exchange buyers, the split was 53-27. Since only 13% of on-exchange buyers were unsubsidized, they moved the needle quite a bit given their numbers. I'e asked Kaiser for a breakout of the figures for the unsubsidized.