People make poor decisions when buying health insurance. So demonstrates health economist Austin Frakt in a review of studies probing Americans' health insurance decisions and knowledge base. Frakt cites studies finding that:
The upshot: When choosing among Medicare plans, employer-sponsored plans, or private plans on offer in the Affordable Care Act marketplace, we all need help.
But what kind of help? That depends, in part, on the menu of choices. And the ACA marketplace is unique -- different in at least one vital way from the markets for Medicare or employer-sponsored plans.
For low-income ACA marketplace buyers -- that is, most buyers -- the choices are fundamentally skewed by the fact that Cost Sharing Reduction (CSR) subsidies are available only with silver plans.
For most ACA marketplace customers with incomes under 200% of the Federal Poverty Level (FPL) -- more than half of current enrollees - silver plans are the only route to adequate insurance. Cheaper bronze plans usually carry single-person deductibles of $6,000 and up, rendering them close to useless for low income buyers (though all include the ACA's mandated free preventive care, and some offer limited benefits before the deductible kicks in). In a country where nearly half the population reports that they could not handle a $400 emergency without borrowing money or selling something, a $6,000 deductible may as well be $6 million for many buyers.
On the other end of the ACA menu, gold and platinum plans are generally too expensive for low income buyers -- and in any case, CSR renders silver plans superior to both (for buyers under 150% FPL) and to gold (for buyers between 150 and 200% FPL).
In short, for marketplace shoppers with incomes up to 200% FPL, one financial decision dwarfs all others: bronze or silver? For a single 40 year-old with an annual income of $23,000 (a bit under 200% FPL), it will typically play out like this: Buy a silver plan with a $500 deductible for about $120 per month, or a bronze plan with a $6,500 deductible for $55 per month (that spread can vary widely, however).
Are we better decision-makers when the stakes are high?
The studies Frakt highlights suggest that most people make less than ideal choices when faced with more fine-grained decisions, such as between $30 per month difference in premium versus $500 per year in deductible. In the ACA marketplace, as in all other markets, people doubtless make lots of mistakes in decisions on that scale -- say, between six silver plans offering minor variations in deductible/copay formulas. We could all use help with such subtle choices -- even Frakt.
But on the far more consequential decision facing most ACA shoppers -- bronze or silver? -- the human decision-making apparatus functions better. About three quarters of marketplace enrollees who qualify for CSR are enrolled in silver plans. That includes buyers with incomes in the 200-250% FPL range, where the benefit is much weaker, weak enough for some buyers to rationally forgo the freebie. Among buyers up to 200% FPL, for whom CSR makes silver plan coverage more comprehensive than that offered by most employer-sponsored plans, a bit over 80% choose silver.
This despite the fact that the CSR subsidy structure, like the deliberately deceptive study Frakt cites that switches the terms of gold and bronze plans, scrambles the metal levels, by rendering silver plans superior to gold and in some cases platinum. Despite, too, the daunting price of silver plans for those at the higher ranges of CSR eligibility. $121 per month, the current subsidized premium for the benchmark (second-cheapest) silver plan for a person earning $23,000 per year, is a very big bite (6.3% of income, to be exact).
Why do most CSR-eligibles forgo bronze? Perhaps a $6,000-plus deductible gets people's attention, even if they've never heard the word "deductible" before. The deductible does figure as prominently as premium in the plan summaries posted on healthcare.gov and most state exchanges.
The ACA marketplace has "special needs" for decision support
Still, more than 15% of low-income buyers choose bronze plans, and that's too many. Bronze may make sense for a healthy low-income young person whose parents can cover several thousand dollars in unexpected medical expenses, or for a healthy 50-something who's lost a job or just started a business and has at least tens of thousands of dollars in savings. But for most low-income buyers, it's a ticket to underinsurance, leading to financial strain and forgone needed medical treatment.
Here's where I think that current research and thinking about insurance decision support may mislead. To boost CSR takeup, blunt tools work best. From day 1 of ACA implementation, the Connecticut ACA exchange steered CSR-eligibles toward silver plans simply by showing them silver plans first in the menu of available plans (also providing a clear, if rather long explanation of CSR). In 2015, 89% of Connecticut buyers with incomes up to 200% FPL bought silver plans; just 6% bought bronze. Maryland cloned Connecticut's technology in 2015 and achieved 86% silver plan selection among all CSR-eligible buyers. In New York, which also "defaults to silver" for CSR-eligibles, 89% of buyers under 200% FPL in 2014 and 86% in 2015 chose silver.
Many other state exchanges have followed suit. Healthcare.gov does not "default to silver" for CSR eligibles, though it does provide a pop-up warning to those who move to buy a metal level other than silver. CSR takeup varies quite a bit in the 37 states that used the platform in 2015, averaging about 80% for enrollees up to 200% FPL.
TMI tools?
Given the "coverage cliff" yawning between silver and bronze for CSR-eligible buyers, subtler decision support tools may actually prove counterproductive. Take the federal exchange's new "total yearly cost estimator," which asks users whether their "expected care" is likely to be low, medium or high and uses the response to estimate the total yearly cost of each plan, including premiums and out-of-pocket costs.
When users anticipate low usage, the estimator usually shows bronze plans as the most cost-effective, even when the buyer is eligible for CSR, which is forfeited by buying bronze. For example, a 40 year-old man earning $17,000 per year is told that a bronze plan with a premium of just $4 per month and a $6,850 deductible is likely to cost him just $163 a year, including $119 for medical care. A silver plan with a premium of $47 per month, no deductible, and a $500 yearly out-of-pocket maximum is estimated to cost $643 for the year, including $75 for medical care.
These estimates leave risk out of the equation. Insurance is about risk management. The cheapest auto insurance policy on the market will cost me less than one that offers adequate liability coverage -- but may bankrupt me if I hit someone and get sued. In the case of the health plans quoted above, one trip to the emergency room or one MRI would flip the script. The bronze plan is only cheap if you don't use it. To someone earning under $1500 a month, a posted saving almost $500 in "likely" costs is "likely" to be tempting, and to obscure the risk of being on the hook for 42 times the estimate.
A cost estimator of this sort could be contextualized and hedged with a warning: your costs will be much higher if you require significant medical care. In my view it should not be part of the initial "shop and compare" plan preview, but rather an option for later drill-down. Results should not determine the order in which available plans are shown (and the healthcare.gov tool does not do so).
Once a marketplace shopper has selected a metal level, the cost estimator could help clarify choices within that band. Improved physician listings and drug formularies, promised soon on healthcare.gov, will also be a boon to users. But in the ACA marketplace, getting CSR-eligible buyers into silver plans -- at least those whose incomes are under 200% FPL -- should be the first priority. Analyses of tools to help marketplace shoppers optimize their choices should reflect that.
- most people choose wrong when faced with relatively small tradeoffs between premium and deductible;
- Americans have a poor grasp of core insurance terms like "coinsurance;
- low income people will choose a plan labeled "gold" over one labeled "bronze" even if the "gold" plan is manifestly inferior (researchers swapped the labels to test comprehension); and
- when given an estimate of our yearly medical costs, a typical ACA marketplace shopper can't determine which plan would cost them least.
The upshot: When choosing among Medicare plans, employer-sponsored plans, or private plans on offer in the Affordable Care Act marketplace, we all need help.
But what kind of help? That depends, in part, on the menu of choices. And the ACA marketplace is unique -- different in at least one vital way from the markets for Medicare or employer-sponsored plans.
For low-income ACA marketplace buyers -- that is, most buyers -- the choices are fundamentally skewed by the fact that Cost Sharing Reduction (CSR) subsidies are available only with silver plans.
For most ACA marketplace customers with incomes under 200% of the Federal Poverty Level (FPL) -- more than half of current enrollees - silver plans are the only route to adequate insurance. Cheaper bronze plans usually carry single-person deductibles of $6,000 and up, rendering them close to useless for low income buyers (though all include the ACA's mandated free preventive care, and some offer limited benefits before the deductible kicks in). In a country where nearly half the population reports that they could not handle a $400 emergency without borrowing money or selling something, a $6,000 deductible may as well be $6 million for many buyers.
On the other end of the ACA menu, gold and platinum plans are generally too expensive for low income buyers -- and in any case, CSR renders silver plans superior to both (for buyers under 150% FPL) and to gold (for buyers between 150 and 200% FPL).
In short, for marketplace shoppers with incomes up to 200% FPL, one financial decision dwarfs all others: bronze or silver? For a single 40 year-old with an annual income of $23,000 (a bit under 200% FPL), it will typically play out like this: Buy a silver plan with a $500 deductible for about $120 per month, or a bronze plan with a $6,500 deductible for $55 per month (that spread can vary widely, however).
Are we better decision-makers when the stakes are high?
The studies Frakt highlights suggest that most people make less than ideal choices when faced with more fine-grained decisions, such as between $30 per month difference in premium versus $500 per year in deductible. In the ACA marketplace, as in all other markets, people doubtless make lots of mistakes in decisions on that scale -- say, between six silver plans offering minor variations in deductible/copay formulas. We could all use help with such subtle choices -- even Frakt.
But on the far more consequential decision facing most ACA shoppers -- bronze or silver? -- the human decision-making apparatus functions better. About three quarters of marketplace enrollees who qualify for CSR are enrolled in silver plans. That includes buyers with incomes in the 200-250% FPL range, where the benefit is much weaker, weak enough for some buyers to rationally forgo the freebie. Among buyers up to 200% FPL, for whom CSR makes silver plan coverage more comprehensive than that offered by most employer-sponsored plans, a bit over 80% choose silver.
This despite the fact that the CSR subsidy structure, like the deliberately deceptive study Frakt cites that switches the terms of gold and bronze plans, scrambles the metal levels, by rendering silver plans superior to gold and in some cases platinum. Despite, too, the daunting price of silver plans for those at the higher ranges of CSR eligibility. $121 per month, the current subsidized premium for the benchmark (second-cheapest) silver plan for a person earning $23,000 per year, is a very big bite (6.3% of income, to be exact).
Why do most CSR-eligibles forgo bronze? Perhaps a $6,000-plus deductible gets people's attention, even if they've never heard the word "deductible" before. The deductible does figure as prominently as premium in the plan summaries posted on healthcare.gov and most state exchanges.
The ACA marketplace has "special needs" for decision support
Still, more than 15% of low-income buyers choose bronze plans, and that's too many. Bronze may make sense for a healthy low-income young person whose parents can cover several thousand dollars in unexpected medical expenses, or for a healthy 50-something who's lost a job or just started a business and has at least tens of thousands of dollars in savings. But for most low-income buyers, it's a ticket to underinsurance, leading to financial strain and forgone needed medical treatment.
Here's where I think that current research and thinking about insurance decision support may mislead. To boost CSR takeup, blunt tools work best. From day 1 of ACA implementation, the Connecticut ACA exchange steered CSR-eligibles toward silver plans simply by showing them silver plans first in the menu of available plans (also providing a clear, if rather long explanation of CSR). In 2015, 89% of Connecticut buyers with incomes up to 200% FPL bought silver plans; just 6% bought bronze. Maryland cloned Connecticut's technology in 2015 and achieved 86% silver plan selection among all CSR-eligible buyers. In New York, which also "defaults to silver" for CSR-eligibles, 89% of buyers under 200% FPL in 2014 and 86% in 2015 chose silver.
Many other state exchanges have followed suit. Healthcare.gov does not "default to silver" for CSR eligibles, though it does provide a pop-up warning to those who move to buy a metal level other than silver. CSR takeup varies quite a bit in the 37 states that used the platform in 2015, averaging about 80% for enrollees up to 200% FPL.
TMI tools?
Given the "coverage cliff" yawning between silver and bronze for CSR-eligible buyers, subtler decision support tools may actually prove counterproductive. Take the federal exchange's new "total yearly cost estimator," which asks users whether their "expected care" is likely to be low, medium or high and uses the response to estimate the total yearly cost of each plan, including premiums and out-of-pocket costs.
When users anticipate low usage, the estimator usually shows bronze plans as the most cost-effective, even when the buyer is eligible for CSR, which is forfeited by buying bronze. For example, a 40 year-old man earning $17,000 per year is told that a bronze plan with a premium of just $4 per month and a $6,850 deductible is likely to cost him just $163 a year, including $119 for medical care. A silver plan with a premium of $47 per month, no deductible, and a $500 yearly out-of-pocket maximum is estimated to cost $643 for the year, including $75 for medical care.
These estimates leave risk out of the equation. Insurance is about risk management. The cheapest auto insurance policy on the market will cost me less than one that offers adequate liability coverage -- but may bankrupt me if I hit someone and get sued. In the case of the health plans quoted above, one trip to the emergency room or one MRI would flip the script. The bronze plan is only cheap if you don't use it. To someone earning under $1500 a month, a posted saving almost $500 in "likely" costs is "likely" to be tempting, and to obscure the risk of being on the hook for 42 times the estimate.
A cost estimator of this sort could be contextualized and hedged with a warning: your costs will be much higher if you require significant medical care. In my view it should not be part of the initial "shop and compare" plan preview, but rather an option for later drill-down. Results should not determine the order in which available plans are shown (and the healthcare.gov tool does not do so).
Once a marketplace shopper has selected a metal level, the cost estimator could help clarify choices within that band. Improved physician listings and drug formularies, promised soon on healthcare.gov, will also be a boon to users. But in the ACA marketplace, getting CSR-eligible buyers into silver plans -- at least those whose incomes are under 200% FPL -- should be the first priority. Analyses of tools to help marketplace shoppers optimize their choices should reflect that.
So as far as I understand it, Bronze plans, with their high deductibles relative to income, are simply catastrophic coverage, right? Yes, that's under insurance given what the person could get if they were willing to pay more in upfront premiums, but the actual baseline we're working with is probably someone who wouldn't otherwise have any coverage at all. If "no coverage" was how this segment of the population previously dealt with health care needs, then catastrophic coverage seems like a solid improvement. (And it seems like the predictable outcome of political bargaining between centrist and liberal Democrats who passed the ACA--liberals were willing to pay for more, but centrists weren't, thus you end up with access to universal catastrophic coverage, not universal well-insured coverage.)
ReplyDelete