How would a 20% drop in base ACA marketplace premiums affect enrollment?

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

My last post delved into the paradox of reinsurance in the ACA marketplace: lower base premiums tend to produce higher premiums for subsidized buyers who select plans that cost less than the benchmark (second cheapest silver) plan. Lower premiums also tend to reduce bronze plan discounts generated by silver loading* -- though potentially increasing gold plan discounts.

Just how much do premiums reductions reduce discounts in below-benchmark plans? The effects can be sampled with reference to the average premium nationwide for the lowest cost plan (LCP) at each metal level in 2020, courtesy of the Kaiser Family Foundation.

Below, I compare net-of-subsidy premiums for those averages to net-of-subsidy premiums when those base premiums are cut by 20%. That's the average reduction effected by the seven state reinsurance programs already in operation, according to Avalere. The 20% reduction is compared to what premiums would have been absent the reinsurance, not what they were in the previous year.  In New Jersey, for example, reinsurance cut premiums by 15% in 2019, and a state individual mandate by an additional 7% -- but premiums were 9% below 2018 levels. Still, let's stick with 20% to make the effect easily visible.

Kaiser's estimates are for a 40 year-old. I've plugged in net-of-subsidy premiums for a single individual at that age with an income of $30,000, or slightly below 250% of the Federal Poverty Level (FPL). At this income level, silver plans come with the weakest level of Cost Sharing Reduction (CSR) -- too weak to stop a strong exodus out of silver and into discounted bronze and gold in the silver loading era (2018 and 2019).

Here are the premium differences. LC = "lowest cost."

Average Premiums in ACA Marketplace (unsubsidized 40 year-old)

Lowest cost at each metal level
2020 vs. 2020 discounted by 20%

Year/future year	LC bronze	LC silver	Benchmark	LC gold
2020	$331	$442	$462	$501
2020 - 20%	$265	$354	$370	$401

Here's how the spread from the benchmark changes.

Spread from benchmark (40 year-old)

Year/future year	LC bronze	LC silver	LC gold
2020	$131	$20	$39
2020 - 20%	$105	$16	$31
Difference w/ discount	-$26	-$4	-$8

And here's how those spread differences affect subsidized premiums at $30k income/40 years old.

Premium net of subsidy (40 year-old, income $30,000)

Year/future year	LC bronze	LC silver	LC gold
2020	$68	$179	$238
2020 - 20%	$94	$183	$230
Difference w/ discount	+$26	+ $4	-$8

Here's how premiums change for a 55 year-old at the same income.

Premium net of subsidy (55 year-old, income $30,000)

Year/future year	LC bronze	LC silver	LC gold
2020	$0	$164	$267
2020 - 20%	$17	$171	$254
Difference w/ discount	+$17	+$7	-$13

Note that while bronze plans get more expensive as base premiums drop -- as does the one silver plan cheaper than the benchmark -- gold plans get cheaper. That's also true for silver plans that cost more than the benchmark, not to mention platinum plans. Premium reductions (if evenly distributed**) compress spreads, putting downward pressure generally on plans priced above the benchmark.

A bit more than half of on-exchange enrollees in ACA-compliant plans have incomes below 200% FPL. For them, the most important spread is between LC silver and the benchmark. While the average spread there is narrow, all spreads vary enormously by region, and narrower spreads on average mean significantly less affordable silver plans for many of those who qualify for strong CSR.  That matters, since 83% of enrollees at 100-200% FPL select silver, and enrollment at this income level has been dropping.

Discounted bronze and gold plans matter most at 201-400% FPL, where they have boosted enrollment by several hundred thousand since silver loading began in 2018.  There's been a broad shift out of silver in this income bracket since 2017, and a slight increase in enrollment that contrasts markedly with a 10% enrollment drop at 100-200% FPL. Here's how the shift evolved in the 39 HealthCare.gov states from 2017 to 2019.

Enrollment by metal level at 201-400% FPL in HealthCare.gov States

Year	Total bronze	% bronze	Total silver	% silver	Total gold	% gold	Total
2017	969,190	34%	1,706,780	60%	135,235	5%	2,851,601
2018	1,299,845	45%	1,251,385	43%	337,995	12%	2,891,851
2019	1,428,582	50%	986,957	35%	427,824	15%	2,863,824

 Source: CMS  state-level public use files.***

While enrollment was all but flat at these income levels -- again, in contrast to a 10% drop at 100-200% FPL -- bronze enrollment was up by about 459,000 and gold by about  254,000.  Cuts in base premiums, by reducing bronze plan discounts, could discourage some who took those discounts from enrolling. Aviva Aron-Dine has calculated that in the 300-400% FPL income range, where silver loading effects are most pronounced, "each additional percentage-point increase in the silver-bronze wedge [spread] results in a 0.75 percent increase in enrollment." In the example above, the 20% across-the-board reduced the silver-bronze spread by 25%. That would probably wipe out much of the rise in bronze enrollment since 2017 in that top subsidized income bracket.

Enrollment by metal level at 301-400% FPL in HealthCare.gov States

HealthCare.gov States

Year	Total bronze	% bronze	Total silver	% silver	Total gold	% gold	Total
2017	316,400	40%	409,600	52%	49,844	6%	786,678
2019	506,610	58%	210,295	24%	144,850	17%	870,637

 Source: CMS  state-level public use files.***

The loss would presumably be partially if not fully offset by the increase in gold discounts and the premium cut for the unsubsidized. Returning to the results of New Jersey's 9%  net premium reduction in 2019:  on-exchange enrollment fell 7% as of the end of Open Enrollment. We have since seen major offsets, however, as of the second quarter (NJ's Dept. of Banking and Insurance publishes quarterly enrollment updates, on- and off-exchange):

• Retention has improved. The on-exchange enrollment drop is just 1.1% year-over-year as of Q2.  Perhaps some of those lured in by discounts are also quickest to drop (although, conversely, the discounts may be a factor in improved retention, in NJ and nationally).


• Off-exchange (unsubsidized) enrollment was up 3% as of Q1 and almost 9% as of Q2 -- again, retention is better. That partly reversed a 14% slide from 2017 to 2018, when premiums spiked.

More detail on the NJ experience in 2019 is here.


---
* Silver loading is the byproduct of Trump's October 2017 cutoff of direct federal reimbursement to insurers for the Cost Sharing Reduction (CSR) subsidies they are required to provide to low income marketplace enrollees who select silver plans. Faced with the cutoff at the brink of open enrollment for 2018, most state insurance departments allowed or encouraged insurers to price CSR into silver premiums only. Since premium subsidies, designed so that the enrollee pays a fixed percentage of income, are set to a silver plan benchmark (the second cheapest silver plan), inflated silver premiums create discounts for subsidized buyers in bronze and gold plans.

**  In fact insurers tend to cut silver premiums more than premiums at other metal levels. This year, on average, benchmark silver premiums are going down a percentage point more than the LCP at each metal level, according to both CMS and Kaiser. That hurts spreads in both directions. But reinsurance should cut premiums proportionately at each metal level.

*** In the PUF tables, enrollment totals at each metal level are rounded as a percentage of all enrollment at that metal level; I have added together those rounded totals at income brackets ranging from 200-400% FPL. The rounding creates some slight distortion. In 2017, CMS broke out metal level by income for bronze and silver plans only. I have estimated the gold as follows: in all income brackets, gold enrollment was 77% of all enrollment other than bronze and silver. For the 2017 gold estimate, I used the formula (total - bronze/silver) x .77.