Tuesday, October 20, 2015

Who leaves employer-sponsored insurance on the table?

The Kaiser Family Foundation's latest estimate of the still-uninsured population in the U.S. includes about 4.8 million whose employers offer health insurance that they decline to buy.

Some are locked out by the ACA's so-called "family glitch," which denies marketplace subsidies to those for whom an employer's insurance offering would cost less than 9.5% of income for individual coverage, even if the family coverage offered by the employer costs far more than that. But many low income workers find even individual employer-sponsored insurance (ESI) unaffordable. Probably the bulk of Kaiser's estimated 4.8 million who forgo ESI could do well in the ACA marketplace if the ESI offer didn't disqualify them from subsidies.

The New York Times' Stacy Cowley has a good story that spotlights the plight of those who can't afford insurance offered by their employers:

Mr. Sewell’s insurance meets the [ACA's affordability] test, but $65 per biweekly paycheck is more than most of his workers are willing — or able — to pay for insurance that still carries steep out-of-pocket costs, including a $2,500 deductible.

Clarissa Morris, 47, has been a server at the Golden Corral here for five years, earning $2.13 an hour plus tips. On a typical day, she leaves the restaurant with about $70 in tips. Her husband makes $9 an hour at Walmart but has been offered only a part-time schedule there, without benefits. Their combined paychecks barely cover their rent and daily essentials.

“It’s either buy insurance or put food in the house,” she said. On the rare occasions that she gets sick, she visits a local clinic with sliding-scale fees. It costs her $25 for a visit, and $4 to fill prescriptions at Walmart.
Cowley cites a 2013 study by payroll processor ADP finding that just 37% of full-time workers with incomes in the $15-20,000 range will accept an ESI offer. Takeup rises to 58% in the $20-25,000 range and tops out at 82% for those earning $40-45.000. ADP points out that the upper figure, coincidentally or not, is close to the 400% FPL ceiling for subsidized coverage in the ACA marketplace (for an individual).

Who leaves employer-sponsored insurance on the table?

ADP, which surveyed 300 employers with more than 1000 employees, has another very useful finding (and prescient, considering that marketplace plans had not yet been created or priced at the time the study was conducted:
...the ADP Research Institute compared employee premium contributions (as a percent of income) with the potential premium contributions required to purchase coverage in a Public Healthcare Exchange. The data shows that, up to 200% of FPL, single employees might pay less for coverage through a Public Exchange for self-only coverage — in part because of the subsidy provided. However, employees at higher income levels would benefit far more by retaining their current employee group health plan....
Based on the available data, for single employees earning less than $22,340 per annum...a Public Exchange is less expensive than group health benefits for self-only coverage.
200% FPL is the threshold below which silver plans (and only silver plans) offered in the ACA marketplace carry strong Cost Sharing Reduction (CSR) subsidies. CSR raises the actuarial value of a silver plan to 94% for a buyer whose household income is below 150% FPL, and to 87% for those in the 150-200% FPL range -- levels superior to those of the average ESI plan. A single worker earning $17,500 in 2015 (just under 150% FPL) would pay $54 for the benchmark cheapest silver plan in her region; a worker earning $23,300 (just under 200% FPL) would pay $118. The deductible would likely be zero in the first category and $500 in the second.

ADP was dead-on in estimating that marketplace coverage would likely be less attractive than ESI for workers earning more than 200% FPL. At 201% FPL, CSR raises the actuarial value of the benchmark silver plan to just 73% -- inferior to the ESI average of AV 82% cited by ADP. At 251% FPL, CSR fades out entirely, and the subsidized benchmark silver premium rises to just shy of $200 per month, with deductibles usually over $2,000. That's worse coverage at a higher price than that offered by the two smallish businesses featured in the Times article.

Many who forgo ESI would do better in the ACA marketplace, if eligible

The nearly 5 million Americans whom Kaiser estimates are forgoing ESI are likely concentrated below 200% FPL.  While ADP estimates that just 8.6% of employees in its survey group pay more than 9.5% of base pay for their insurance, that's because those for whom insurance would take the biggest bite are the ones who forgo it:
Within the ADP sample population, 30% of benefits-eligible unmarried employees have incomes within the $15,000 – $30,000 range. If their participation rate increased to the norms of higher-income individuals, we would expect a 40% increase in enrollment for this lower-income segment
While the ACA's tax penalty for remaining uninsured might push some lower-income employees into accepting ESI, many if not most in this income range would be better off on the exchanges if they were eligible for subsidized coverage there (we don't know how many, as ADP is recording subjects' individual earned income, not their household income  e.g., if they're single parents or cohabiting).

Not surprisingly, those who forgo ESI are at income levels where the nation's uninsured are concentrated. According to Census data*, in 2014, 54% of the uninsured were under 200% FPL, and 72% were under 300% FPL:

Reduction in Uninsured Population by Income Group, 2013-2014

Pop 2013
Pop 2014
Uninsured 2013
Uninsured 2014
Drop uninsured
< 100% FPL
 46.27 mil
 46.66 mil
10,87 mil
 9.00 mil
1.87 mil
 58.77 mil
 58.69 mil
11.99 mil
 8.86 mil
3.13 mil
 49.58 mil
 51.45 mil
 7.83 mil
 6.02 mil
1.81 mil
 40.82 mil
 40.82 mil
 4.20 mil
 3.43 mil
  .77 mil
Over 400%
117.67 m
118,19 m
 6.59 mil
 5.67 mil
  .92 mil
313.44 m
316.17 m
41.47 mil
32.98 mil 
8.50 mil

Under 200%
105.04 m
105.34 m
22.86 mil
17.86 mil
5.00 mil
Under 300%
154.61 m
156.29 m
30.68 mil
23.88 mil
6.81 mil

As with ADP's numbers, the Census's income categories don't quite match up with those used by the ACA marketplace. The Census surveys query income for everyone who lives under one roof, while the Marketplace calculates subsidies on the basis of a tax filing household. The marketplace may therefore place enrollees in a lower income band than the Census surveys would -- in which case, an even higher percentage of the uninsured may be concentrated in the ACA's sweet spot -- under 201% FPL as defined by the marketplace.

A good number of those low-income uninsureds may be forgoing ESI.  It's dispiriting that the substantial number of employees who find ESI unaffordable and who could do better in the marketplace are ineligible for subsidies there (though a portion of the low-income ESI refuseniks may be eligible for Medicaid, for which an ESI offer is not disqualifying). The family glitch exacerbates that affordability gap, but it would exist even if the glitch were fixed.

If the ACA's employer mandate were scrapped -- a change that many progressives support -- then employers might indeed be well advised to cut off ESI for lower-income workers. That would probably be good for many workers with incomes under 200% FPL.  It would also cost billions. But with marketplace enrollment projected by HHS to be just half of what CBO projected for 2016, perhaps we could afford it.

UPDATE: An astute set of tweets on the ACA's skewed "affordability" scale for low income workers from Louise Norris, a broker and my fellow contributor at healthinsurance.org:

For ppl w/ ESI, the threshold is 9.66% of income in 2016, even if they only earn 150% of FPL.
Whereas if they could get exchange coverage, they'd only pay 4.07% of income for benchmark plan at 150% of FPL
Affordability rules are skewed; ESI isn't as affordable for lower-income workers as HIX coverage would be.
It doesn't make sense that there's no sliding scale to determine ESI affordability.

I would think, though, that it would be impracticable to require employers to offer insurance that matches marketplace affordability scales for buyers under 200% FPL. What would seem to make sense would be granting marketplace eligibility to those whose (compliant) ESI offer did not match the affordability of a benchmark silver plan -- perhaps for buyers under 200% FPL only.
* The Census' Health Insurance Coverage in the United States: 2014 is based on two large-scale annual surveys of households' income and insurance status, the Current Population Survey and the American Community Survey.


  1. Don't forget workers who don't want to have their health habits surveilled by their employers in ill-executed wellness programs:

    One big wellness program we looked at led previously happy employees in a stable job environment to become anxious about losing their jobs. It seemed to make them think they needed to be more attractive to their employer, and if they did something like smoking a cigarette, they felt it affected their employability.

  2. The low wage worker was a terrible demographic problem for health insurance before the ACA, and the Times has shown in several articles that it is not yet solved by any means.
    The worker in this article appears to clear about $10 an hour and her husband makes $9 an hour part time. So let's put their income at $32,000 a year.

    That income is too high for Minnesota Care where I live, which would charge them $50 a month for very decent insurance.

    The family might get tax credits if they could squeeze by the family glitch in Health Care.gov, but they would still have to pay at $150 a month for an ACA plan, and they do not have $150 a month for health insurance as already proven.

    A previous article in the Times described a woman who did get an ACA policy for a net cost of just $58 a month in her fifties, but she dropped the policy due to the pressure of other bills and then racked up many thousands in medical debt.

    Seems like some form of free care is the only answer for this demographic.