Is employer-sponsored insurance superior to ACA marketplace plans? For whom?

So far, the Affordable Care Act has not curtailed the availability of employer-sponsored insurance (though it may have curtailed its growth). On the front page of today's New York Times, Reed Abelson reports that "health care remains an important recruitment and retention tool as the labor market has tightened in recent years." One part of her explanation is "mostly true," as the fact-checkers say -- but hides an important fact about what kind of insurance is available to whom:

The law has resulted in more coverage for low-income people, as expected. But...the plans on the exchanges remain less generous than those offered by many employers, with significantly higher deductibles and a significantly narrower choice of hospitals and networks.

That's probably true for most workers whose employers offer them insurance. It's not true, however, with respect to out-of-pocket costs for workers whose household income is below 201% of the Federal Poverty Level (FPL). As I noted last Friday, for ACA enrollees with incomes below that threshold, the average weighted actuarial value (AV) of the plans they obtain -- that is, the percentage of the average user's costs that the plan will cover -- is 86%.  The average AV in employer-sponsored insurance was estimated by the Kaiser Family Foundation at 82% back in 2011 -- and it's probably dropped since then.  It's likely to be still lower for lower income workers.

The high actuarial value obtained by ACA enrollees with incomes under 200% FPL is a result of the Cost Sharing Reduction (CSR) subsidies available to them if they buy silver plans. CSR raises the AV of a silver plan from its baseline 70% to 94% for buyers with incomes under 150% FPL, and to 87% for buyers with incomes in the 150-200% FPL range.*

Over 60% of ACA marketplace enrollees have incomes under 201% FPL.**  Over 80% of enrollees below that threshold select silver plans and so access CSR at AV 94% or 87%. While the provider networks available to these enrollees are likely to be narrower than those offered by employer-sponsored insurance, their out-of-pocket costs (if they stay in network) are likely to be lower.

According to Census figures, a third of Americans live in households with incomes below 200% FPL. As of 2013, just prior to ACA implementation,  55% of the uninsured had incomes below that level. This is precisely the population that suffered severe loss of access to ESI once the Clinton-era boom faded. According to the latest update of the National Health Interview Survey (NHIS), the percentage of Americans with incomes in the 100-200% FPL range with any form of private insurance dropped from 56.2% in 1997 to 34.7% in 2010. Since full ACA implementation, it's surged back from 36.4% in 2013 to 44.3% in 2015. And thanks in large part to the Medicaid expansion, the overall uninsurance rate for people in this income group is probably lower than it's ever been. In 1997, 34.9% of the "near-poor" (100-200% FPL) were uninsured; in 2015, 24.1% were.

Reed's statement is not untrue, because most people with access to ESI have households incomes over 200% FPL. For those who don't, however, the contrast between what's available in the marketplace and what an employer might offer is likely to be sharper than between the overall ESI average and the CSR-enhanced marketplace offerings. The Kaiser Family Foundation's 2015 Employer Benefits Survey finds that employers with a high percentage of low income workers pay a lower share of their health insurance costs that employers with more high income workers. Among the findings (Section 6):

• For family coverage, covered workers in firms with many lower-wage workers (35% or more earn $23,000 or less annually) contribute a greater percentage of the premium than those in firms with fewer lower-wage workers (41% vs. 28%) (Exhibit 6.21). 
• Covered workers in firms with many higher-wage workers (35% or more earn $58,000 or more a year) contribute less on average than those in firms with a lower proportion of higher-wage workers (26% vs. 33%).
• Twenty-nine percent of covered workers at firms with many lower-wage workers pay more than 50% of the premium for family coverage, in contrast to 14% at firms with fewer lower-wage workers (Exhibit 6.19).
• Looking at dollar amounts, covered workers in firms with many lower-wage workers (35% or more earn $23,000 or less annually) on average contribute $6,382 for family coverage versus $4,829 for covered workers in firms with fewer lower-wage workers (Exhibit 6.13).
• Covered workers with family coverage in firms that have at least some union workers contribute a significantly lower percentage of the premium than those in firms without any unionized workers (23% vs. 32%) (Exhibit 6.21).

Once more: The majority of ACA marketplace enrollees have incomes under 201% FPL, and the majority of that group obtain plans with AV superior to what's generally available from employers. Conversely, though, most Americans with access to ESI would do worse in the marketplace than they do with their employers.
----
*Beyond that threshold CSR fades to almost nothing, raising the AV to 73% for buyers up to 250% FPL. It is not available to buyers above that level.

** In the 38 states that use the federal marketplace, HealthCare.gov, 66% of enrollees for whom income data is available have incomes below 201% FPL. That's 61% of all buyers on the platform, since over 7% don't report incomes.  In the twelve state-based marketplaces, which account for a quarter of all enrollees, the percentage of lower income buyers is somewhat lower. HHS tables for 2016 enrollees are available here and here.

P.S. I originally attached a male pronoun to Reed Abelson, now corrected. Sorry about that!