Earlier this week, I had a new manuscript, co-authored with Dr. Petra Rasmussen (RAND) published in the Milbank Quarterly.
We examined what happened to plan choice on Covered California after a huge relative price shock for silver plans in 2018.
Silver plans became comparatively way more expensive relative to gold plans in 2018 compared to 2014-2017. This was because the Trump Administration stopped paying Cost Sharing Reduction subsidies directly to insurers, so many states, including California, had insurers place the CSR costs directly into Silver plan premiums. Silver plan premium levels are what determines the value of the subsidy, and thus the price that subsidized people pay for every plan. California, luckily for us as researchers, also institutional details of only allowing standardized benefit designs to be offered so we could readily compare apples to apples. Those benefit designs produced clear breaks between metal levels and CSR variants.
In 2018, every California insurer Silverloaded. Most insurers had their silver plans still a smidge lower premium than the gold plans that they offered on the same plan type and network. However two insurers priced their gold plan a smidgeon cheaper than the corresponding silver plan. The high value CSR designs from these two insurers still offered less cost-sharing than the corresponding gold plan. This means that there still existed a premium and cost-sharing trade-off for people who earned under 200% FPL between CSR silver and gold plans. Enrollees could tell a facially plausible rational story as to why either choice could make sense. However, these two insurers’ gold plans had both less cost-sharing and lower premiums compared to silver for anyone earning over 200% FPL. Silver was transparently dominated for buyers of these two insurers if they made more then 200% FPL.
So what happened?
8/N from a policy POV, ideally no one earning over 200% FPL in these triads ever buys Silver.
That sooooo did not happen pic.twitter.com/TP2PU08v1s
— David Anderson (@bjdickmayhew) July 7, 2021
About 20% of the eligible population in 2018 bought a dominated plan. These folks ended up paying about $450 more in annual premiums and were exposed to several thousand dollars more in annual deductible.
There was significant variance. Individuals who had to make an active choice did better with about 8 to 9% of that subpopulation choosing a dominated plan. Automatic re-enrollment and choice inertia led to people who had a silver plan in 2017 being way more likely to either choose or just end up in a dominated plan. We saw a steep gradient in dominated choice by income — lower income households were more likely to have a dominated plan. We were surprised to find that assistance did not matter all that much on the likelihood that these folks ended up in dominated plans.
Choosing insurance in the best of choice environments is tough.
Very interesting work! Hard to expect people to make rational choices when choice set itself appears irrational – you get more for less?! Then again, plenty of precedent to show we choose insurance plans poorly. (Aside: I assume you control for insurer and plan type.)
— Leemore Dafny (@LeemoreDafny) July 7, 2021
The presence of dominated plans on the choice menu makes an already tough challenge even tougher as the quick rules of thumbs and heuristics that people use to sort through their options fail.
Brad F
Dave
Could you demonstrate (simply) what an individual would see with >200%FPL in a side by side with gold vs silver on the web page (assuming that function was available)? How much of an absolute dollar difference would stare them in the face? That’s not to say your findings are solid as a rock. Its more to show what levels of cost would force action (“good Lord, that’s too much for me”).
Brad
Adam Lang
Side note: “The high value CSR designs…”
You shouldn’t assume that everyone knows what this means. Professional s in the space, as well as health care policy groupies* like me, do, but even other frequent readers may not.
* Possibly a class of one?
Martin
I’ll add an anecdotal point here. I’m planning to retire, and so I’m looking on the CA marketplace to see what the best option for us is. My decision is a bit more complex – if I start taking from my pension (something I don’t want to do until I’m 60-65) I can simply continue my health insurance by paying my employers premiums – which are pretty good. Alternatively, I can buy off the marketplace. I started pricing silver, looked at the plans, then looked at gold. I settled toward gold for two reasons:
@Brad F:
Coverage for me, my wife, and my daughter at $75K income, silver is $501/mo, gold is $824/mo, and platinum is $992/mo. Differences between gold and platinum seem pretty small, IMO – mostly copays – but silver has a $8K yearly deductible vs $0 for gold. This is just for Kaiser HMO, so it’s not representative if I was cost shopping, but probably gives a ballpark. I’m ignoring the coinsurance options.
At $50K silver is $228/mo, gold $551/mo, platinum $719/mo.
silver-gold are pretty similar for copays, the deductible is doing most of the work there. Platinum is where the copays really start to vanish, and the copays don’t bother me as they’re already pretty reasonable.
So for me, I’m looking at a $323/mo delta between silver and gold. But, that avoids an $8000 ($667/mo) potential deductible which depends on services. But I’ve got two pre-existing conditions in this household and we’d guaranteed use at least $323/mo of that deductible if not the entire thing. It gets a lot harder for me to see scenarios where platinum pays for the additional costs. We’d have to rack up a doctor visit a week and a bunch of other stuff, and we just don’t have those kinds of expenses – it’s mostly a bunch of drugs, occasional check-ins, blood tests, etc. and a surgery or ER visit every other year or so.
I priced for two different incomes because I’d be coasting until 65 on my investment income which isn’t the easiest thing to control for in it’s current configuration, plus, it factors in a decision to pay off the house or not which would eliminate the need to take as much income. We’re near the end of the mortgage, so there’s not much interest left to deduct. So $50K is probably pay off the house, $75K is probably not paying off the house.
My work policy would get me something between gold and platinum coverage for about $450/mo, so cheaper than even the $50K gold plan and I wouldn’t have to watch my income so closely, but I’d have to start drawing off my pension which I’d rather not do as the benefit steps up quite a bit at age 60.
Now, if I was legit working and making $50K, that gold option would almost certainly be out of reach. Even at $75K it’d be tough. The gap is 7% of income at $50K and 5% at $75K. That’s a lot, especially if you thought you could avoid much of the deductible. But we’re in our 50s and shit is starting to break a lot more regularly. We’re no longer on the ‘get less out than you pay in’ side of insurance, we’re on the other side now, or at least break-even.
But there’s a decent number of folks that are buying the plans as bridges from employer insurance to Medicare and we have a fairly different cost profile.
J R in WV
Dominated Health care plans…
. . do they require submissive customers?
. . . . ;~)