A recent paper in the American Journal of Managed Care led by Vicki Fung estimated the distribution of ARPA subsidies by income bracket for off-exchange buyers. Their top line finding was the benefits of subsidizing households over 400% FPL were wildly dispersed:
The median annual PTC in 2021 for eligible off-marketplace enrollees was $311 but varied greatly by age, family or individual plan, and household income (5%-95% range, $0-$14,836).
However there was a secondary finding that has been bugging me since I read the paper:
We know that choosing health insurance in the ACA’s marketplaces and other American individual choice markets (Medicare Advantage, Medicare Part D, Medicaid managed care etc) is challenging. People routinely make expensive choice errors that are fairly persistent. We know that the program knowledge elements are both costly and obscure. We also have a huge industry of assistors who are supposed to have the ability to acquire that costly detailed knowledge and then translate it, for a fee, to people who are rationally ignorant of the details. In the insurance industry, we call those people brokers and agents.
When I was looking at dominated plan choices with Petra Rasmussen, we found that people who got help versus those who did not barely had any difference (unadjusted) in picking a dominated plan. Picking a dominated plan is wicked expensive and objectively an inferior choice relative to picking a non-dominated plan. This is a clear choice error. It is an error that a broker or an agent should never advise. Yet, it happened fairly frequently. I question the effectiveness of help in this choice space.
The new finding by Fung implies that least ~40% (69%-(1-51%))/51%) of the people who received broker help were not aware of a critical programmatic fact — subsidies were only available to people who bought plans categorized as “on-Exchange” instead of “off-exchange”. This is an important fact as buying on-Exchange for people who estimate that they are likely but not certainly going to earn too much money to qualify for subsidies means they are giving up a nearly free option for subsidies that gets triggered if they modestly overestimated their income. Now this case does not apply to everyone who buys non-subsidized, off-exchange coverage. A consultant making 1900% FPL with a 500% FPL termination clause in their contract could remain rationally ignorant of this fact even if they are being advised by a broker.
This sub-finding is making me go “hmmmm” even more on the quality and value of assistance on plan choice.