Much has been written about the general issues associated with higher premiums for plans purchased on the various Exchanges under the Affordable Care Act. The fear is that higher premiums will price individuals out of the market — at least lower risk individuals — and result in a spiral of premium increases. Supporters of the ACA counter, however, that increases in gross premiums — the amount insurers say they charge — does not matter much because the amount most individuals pay under the Affordable Care Act does not depend on gross premiums so much as it depends on one’s income. Recent studies, after all, suggest that about 87% of purchasers receive tax credits to purchase plans that depend solely on their income.
The truth, however, about the effect of premium increases under Obamacare is more complex. Basically, increases in gross premiums for basic silver plans can have a non-linear and frightening effect on increases in the net premiums Americans pay to persist in plans that are more generous either because they afford greater choice in selecting one’s doctors or because they require less cost sharing by insureds. The remainder of this blog post explains why.
Let’s start with a simple example drawn from real life to illustrate the general idea. Data taken directly from healthcare.gov shows that in my home city of Houston (Harris County, Texas), the gross premium for the second lowest silver plan (an HMO plan) is about $250 per month for a 40 year old individual. If that person’s net income is $25,000, under 26 U.S.C. § 36B their net premium can be computed as $143 per month for that second lowest silver plan; the remainder is supposed to be paid via a $107 tax credit from the federal government. (Mathematica notebook containing the analysis available on request) Suppose, however, the individual does not want an HMO plan but wants a PPO plan so they can have greater choice of doctors. The cheapest one in Harris County has a gross premium of $338 per month or about 35% more. But, because the tax credit stays constant at $107, the net premium is $231. This means that the right to select one’s own doctor costs our hypothetical individual 57% more in the amount they actually pay.
The situation is similar if one is willing to accept an HMO plan but one wants the plan to pay a higher percentage of expected medical expenses. The second least expensive gold plan in Harris County is $297 per month or 18% more than the second lowest silver plan. Again, however, because the tax credit remains the same, there’s a problem: the actuarial value of the gold plan is actually just 14% higher than its silver counterpart, but because of the way subsidies are calculated, the cost is 33% higher. Only two types of people would likely be willing to incur this “double payment” to get a better plan: people who are unhealthy for whom a small improvement in a plan might mean a great deal or people who are just extremely risk averse.
The situation persists for the few platinum purchasers. If one goes from the second lowest silver plan all the way to the second least expensive platinum plan, it turns out that the net premium increases a whopping 138%. And this is true even though the while the gross premium goes up by 79% to $448 per month and the actuarial value of the policy goes up only about 29% and the gross premium goes up “only” by 79%. One can imagine that the only people of this income level willing to incur this high a premium increase in exchange for only somewhat better coverage would be those who expected to use the coverage extensively — exactly the people that force insurers to increase future prices.
Some of the same mathematical logic that drives the disproportionate increases in net premiums in a single year applies in a similar way to premium changes over time. Let us consider our same individual and project forward to 2016. Imagine that they purchased that PPO silver plan in 2015 and wish to continue in it. To do the computations, we’ll have to make a few assumptions: (1) that the federal government’s expected contribution from income increases for 2016 at the same rate it increased for 2015 and that the federal poverty levels for 2016 likewise increase at the same rate that they did in 2015. Suppose further that the gross premium for the second lowest silver plan and the gross premium for the increases by 4% but the gross premium for the gold HMO plan or the silver PPO plan increase by 8%. If our individual’s income again increases by 10%, the net price of the more generous policy jumps by 13.5%. The effect is non-linear.
And what if we start talking not about individuals, but about families? Consider, now, the family of four with two adults each age 40. We’ll give the family an income of $50,000 per year. Again, for concreteness, I’ll place them in Harris County, Texas. The gross premium for the second lowest silver plan (an HMO) is $748 per month. But with a tax credit of $456, the net premium for that policy is $292 per month. If the family wants to purchase a silver PPO, the cheapest one will feature a gross premium of $1013. Since the tax credit stays the same at $456 per month, this means the net premium is $557 per month, an increase of 91% just to get more choice in picking doctors. Or, if the family wants to purchase the second cheapest Gold HMO, that will cost $890 gross and $434 net. This is an increase of 49% just to get a plan with 14% richer expected benefits. Now, whom do you suppose might pay such a disproportionately higher amount?
The diversity of metal levels and of plan types has always been touted as a benefit of Obamacare. It is supposed to distinguish the ACA from administratively simpler (“one size fits all”) regimes such as single payor plans. But the existence of premium subsidies pegged to the price of the second lowest silver plan means that the present diversity of plans may be a short run phenomenon or that the diversity may exist only on paper. There may technically still be gold plans or PPOs, but very few may be purchasing them. Assuming Obamacare lasts a few more years, we may effectively see the demise in the marketplace of Gold and Platinum plans and, even more likely, the demise of Gold and Platinum PPO plans. Choosing one’s doctor may, as a practical matter, become a sensible option only for the few wealthy purchasers that do not depend on subsidies.
Might this all be just sort of bug that would be easy to fix if only Congress were more cooperative? Actually, not. The high marginal costs of more generous policies is a fundamental feature of Obamacare’s architecture. It is one that is simply becoming more apparent as the program matures. Once you tell people that government will essentially buy you an HMO silver plan if you contribute some amount based on your income – but will pay no more – the net costs of buying anything more generous than that inevitably look very high.
I’ve been doing a lot of research on the state of policies sold on the health insurance exchanges. It’s not easy because the Obama administration, as even its friends acknowledge, has not been forthcoming with information. It has, however, placed some useful information in the public domain: two large databases of the plans being sold on the “Federally Facilitated Marketplace.” That’s the health insurance Exchange for states, like Texas and many others, that declined to establish their own exchanges. With the help of the R computer language, I’ve been sorting through this database and have reached the following conclusions.
The change in premiums between 2014 and 2015 depends significantly on the metal level of the plan and whether it is a PPO or HMO.
Gross premiums for platinum plans are up significantly in price, 21%, whereas bronze plans and catastrophic plans are down over 11%. The high increase in gross premiums for platinum plans creates a serious potential for an adverse selection death spiral in that segment of the market.
Net premiums will show larger percentage increases and decreases than gross premiums for many individuals. This is so because substantial parts of the premiums are paid via subsidies from the federal government.
PPO plans are up substantially in price, 8%, whereas HMO plans are down substantially, -18.%.
The combination of increases in the more generous platinum and PPO plans and decreases in the less generous bronze and HMO plans may start to divert Americans into healthcare plans that offer lower benefits and somewhat less choice, albeit at a lower price than was paid this past year.
Among plans that persisted between 2014 and 2015, the premium variations are less extreme: persistent bronze plans increased in price by 9.5% whereas persistent platinum plans increased in price by 14%. The larger variation in gross premiums overall is thus likely due to the exit of carriers who priced at extremes and low pricing by new entrants for bronze plans but very high prices for the more generous plans.
Cost sharing for the plans has increased somewhat, but many cost sharing arrangements have remained largely the same.
Competition, as measured by the number of unique issuers offering plans in each county, has increased substantially since 2014, but a market in which three or more insurers are actively competing is still a rarity, particularly for the plans that give consumers a greater amount of choice in selecting their doctor.
You can read a copy of the full report, which contains 29 tables of information, here.
I’ve been doing some research into the effects of market concentration on health insurance premium pricing on the health insurance Exchanges run by the federal government. During the course of that research, I discovered what I first thought had to be a programming error on my part or a database error on the part of healthcare.gov: Silver and Gold plans that were costing individuals age 50 upwards of $2,000 per month. Yes, per month!
It turns out, however, that these exorbitant prices are not errors. They represent a clever attempt by several insurers in Virginia — Optima Health, Coventry Health Care of Virginia, Inc., Innovation Health Insurance Company, and Aetna — to get the federal government to pick up a substantial part of the tab for bariatric surgery. Here’s how it works. The insurer offers the consumer a premium that is often $2,000 per month ($24,000 per year) more than it charges for other essentially identical plans. The bonus is that the insurer offers the consumer, in addition to the usual benefits, bariatric surgery, which is otherwise subject to coverage restrictions in Virginia. Now, the only person who would rationally purchase such a policy is one who is pretty certain to undergo such surgery. And, as it happens, bariatric surgery (such as a gastric bypass) appears to cost between $20,000-$25,000. In effect, then, the insured prepays for the surgery via augmented premiums and perhaps piggybacks on the insurer’s bargaining power with surgeons to get a cheaper price.
So far, however, this does not seem like a compelling business model for insurance; at best it converts insurance into an elaborate financing scheme. But wait: if the insured has a relatively low income (and obesity correlates with poverty in modern America), under the cost sharing reductions provisions of the ACA (42 U.S.C. § 18071) the federal government now picks up much of the deductible and coinsurance that would otherwise be owed. Instead of there being, say, an $3,500 deductible and a $6,350 coinsurance limit, as there is under the Aetna Classic 3500 PD:MO policy offered in Virginia, if the person is poor enough (100-150% of federal poverty level), the deductible under the Aetna Classic 3500 PD: CSR 94% MO is now $300 and the out-of-pocket limit is now $1,250. The federal government is thus likely to pay for $3,200 to $5,100 of the bariatric surgery that would otherwise come out of the patient’s pocket.
Is this legal under the ACA? I believe it may well be. I don’t see a violation of the “metal tiering” provisions of the ACA. Under section 1302 of the ACA (42 U.S.C. § 18022), whether something qualifies as a Silver or Gold plan depends on the cost to the insurer of providing essential health benefits to a standard population, not on the cost to the insurer of providing its actual health benefits to the population it anticipates attracting. That may not be a very good system, but is the one in the law; it is probably simpler than some alternatives. Moreover, section 1302(b)(5) of the ACA makes clear that a health plan may provide “benefits in excess of the essential health benefits described in [the ACA].” And, since some states apparent include bariatric surgery in their list of essential health benefits, it’s hard to say that Congress implicitly rejected paying for this procedure.
Footnote: I suppose there could be an issue as to whether this plan conforms to Virginia insurance regulations. I’m not an expert on that, but my working assumption is that the Virginia regulatory apparatus has approved these plans.
Is what these insurers are doing appropriate? That’s a tricky question. Basically what they are doing is the result of a decision by the Department of Health and Human Services relating to implementation of sections 1201 and 1302 of the ACA. HHS, instead of creating some uniform concept of Essential Health Benefits for those states that elected not to make their own decision, instead decided to try and mimic features of the “largest plan by enrollment in the largest product by enrollment in the State’s small group market.” 45 C.F.R. 156.100) That essentially made it a bit a matter of luck as to the circumstances under which bariatric surgery or other weight loss programs would be covered by plans permitted to be sold after 2013 on the individual market. It meant that in some states the risk of needing (or badly wanting) bariatric surgery would be spread among all those purchasing non-grandfathered plans after 2014 whereas in other states either the risk would not be transferred at all or would be transferred, as in Virginia, only at a high price. The map below created by the “Obesity Care Continuum” shows how the states differ.
And should bariatric surgery itself be covered? It’s not an easy decision. On the one hand, bariatric surgery frequently results in part from poor health choices made by the individual. Yes, there may be contributing factors such as access to healthy foods, genetics, access to safe methods of exercise, but, still, most people have a choice not to become obese. And, if the condition is viewed as substantially the result of individual choice, the case for socializing and spreading the risk is weaker. On the other hand, there are plenty of risks that health insurance policies do pay for — both before and after the ACA — that likewise result substantially from personal choice. They cover orthopedic surgery for (mostly wealthy) people who choose to ski. They cover smoking related conditions — albeit for an additional premiums which, if actually collected, would still probably be less than the actuarial risk of tobacco use. They cover treatment in at least some forms for the variety of conditions created by substance abuse (drugs, alcohol). They sometimes cover non-surgical costs to which obesity contributes even when those problems are partly the result of individual choices. And they covers the costs of treating sexually transmitted diseases even when those diseases might, in some instances, have been prevented by safer sexual practices. Untangling fault out of medical need is often a tricky proposition indeed.
So, perhaps these Virginia insurers are doing the public a service by evading/working around restrictions in the Obamacare package of essential benefits provided in some states that were unduly narrow. Indeed, on this view, the problem is not that the federal government is subsidizing bariatric surgery, it is that individuals have to pay these enormous extra premiums for a risk that should be shared and that are shared in some states. It will be interesting to see what happens with these Virginia plans and whether what has started there extends to other states in which bariatric surgery is not presently considered an essential health benefit.
Those optimistic about the success of the Affordable Care Act have been noting over the past several months that the premiums offered by insurers have been lower than those earlier forecast. But if one looks carefully at the original rhetoric, the comparison tends to be between some of the lowest premiums offered within a jurisdiction and those originally forecast. And this metric, according to ACA proponents, is appropriate because they expect consumers to focus purchases on the lowest cost policies.
But what if the lowest premiums are lower than expected not because the mix of purchasers is thought to be fine or because of cost cutting measures enabled by the ACA, but simply because all this metric exposes is the work of the insurers who priced their policies below actual risk? The “winner’s curse” is the term economists and game theorists give to situations in which, in an atmosphere of uncertainty, people bid on an item in an auction environment. What will often happen is that the “winning” bidder will tend to be one that loses money.
It is quite possible that all we are seeing with “low” ACA pricing, as measured by ACA proponents, is “the winner’s curse” in action. We may well be looking at insurers who (a) got it wrong or (b) thought the government would most greatly subsidize their losses or (c) for strategic reasons, decided to sell a “loss leader” in the first year or so of the ACA in order to lock consumers into their networks and their doctors with the idea that they could substantially raise premiums in the future. If this hypothesis is correct, individual policies under the Exchange are a lot less stable than many ACA proponents are asserting.
To summarize the results of the computations shown below, if the mean premium charged by insurers selling a type of policy (Silver HMO Plans, for example ) in a given geographic region (Harris County, Texas, for example) reflects the true risk posed by ACA policy purchasers, about 20% of the low bidders — the ones that I suspect will get a disproportionate share of the business — stand to lose at least 20% on their policies before the Risk Corridor program bails them out.
The big story as the ACA unfolds may be that some insurers — the ones who ended up with the business — simply made an error of exuberance in a new market and priced their policies too low. While these insurers will, thanks to a federal subsidization program for losing insurers called Risk Corridors, not entirely lose their shirts in the first year of the program as a result, they do stand to lose a lot of money that they will likely want to make up in any subsequent years of the Affordable Care Act.
New data analysis finds significant dispersion in plan premiums
This post will contribute some new data analysis that suggests the likelihood of the winner’s curse materializing as well as the magnitude of such a curse. Basically, I have sucked into my computer official government data on the 78,000 plans sold on the federal marketplace and done a lot of number crunching. The data shows a significant dispersion of prices offered by insurers for plans in the same geographic area, of the same metal tier and offering the same form of coverage (PPO, POS, HMO, or EPO) . While this dispersion does not prove that the low prices are outliers reflecting either miscalculation by some insurers or only-temporary use of low prices, it does suggest a significant possibility that such is the case.
Let’s take an example. Here are the prices offered where I live, Harris County, Texas — mostly Houston — for an HMO Silver Policy to a couple with two kids. The couple has an average age of 50 years old. We’ll call this hypothetical family “The Chandlers,” as a matter of convenience. The graphic shows the dispersion of premiums normalized so that the lowest price for a given policy is given a value of 1.
As one can see, for the Harris County, Texas policies shown here, although there are three policies that have premiums fairly close to the minimum, there are, however, two policies that have premiums more than 30% more than the minimum. If the mean premium estimated by insurers is “correct,” the insurer selling a Silver HMO policy at the lowest price will lose about 17%. The implication, if the Harris County plan is representative and if the mean premium is closer to the true risk than the low premium, is that the insurers most likely to win business due to low prices are likely to lose a considerable amount of money.
There are several potential rejoinders to the suggested implication of the graphic. Let me address each of them in turn.
Might Harris County, Texas be unusual?
One response is that the example for Harris County, Texas Chandlers is unrepresentative. Houston, for example, has some very fancy hospitals and some not so fancy hospitals; so maybe premium dispersion for Harris County simply reflects whether one has access to the fancier hospitals (and the doctors who have admitting privileges to them). I have considered this possibility and find that, actually, the example I provide is pretty representative. Here, for example, are 20 randomly selected examples. For each plan, I show the amount the low bidder would lose if the average premium is “correct,” the dispersion of premiums, and the plan and purchaser randomly chosen. Of the ones in which there are any significant number of policies available, most of the premiums show a dispersion pattern qualitatively similar to that in Harris County for The Chandlers. Indeed, some of the random examples show dispersion considerably greater than that for the Harris County silver HMO policies. Except where there is little competition for plans and the low bidder is thus selling at the average price, the result presented above does not look like a fluke.
I can double check this result by computing for 5,000 random combinations of plans and purchasers the losses of the low bidder if the true risk was equal to the mean premium charged for policies and purchasers of that type. The graphic below shows the “survival function” (or “exceedance curve”) for the resulting distribution of these losses. The value on the y-axis is the probability that the losses will exceed the value on the x-axis. The results shown below confirm that the situation for Harris County Silver HMO plans sold to The Chandlers is not all that unusual. As one can see, losses of more than 10% take place more than 30% of the time and losses of more than 20% take place about 17% of the time. A rather scary picture.
In fact, however, the situation may be even worse than depicted in the graphic above. Sometimes the losses computed by this method are low because the low bidder is also the only bidder. If we consider situations in which there is more than one bidder, here is the resulting survival function (exceedance curve) of the distribution. As one can see in the graphic below, the distribution of risks is shifted slightly to the right. Now 40% of the low bidders stand to lose at least 10% and about 21% stand to lose at least 20%.
Maybe the higher priced policies are better?
Another potential explanation for price dispersion is that, even if the policies are priced differently, that does not mean that the cheapest policies are selling for too low a price. All Silver HMO policies sold in Harris County, Texas to The Chandlers may not be the same. Some may have different deductibles or different networks.
The first response to this rejoinder is that the actuarial value of the policy — the relationship between expected payments by the insurer and premiums — should be about the same for each metal tier of policies. Silver policies should all have actuarial values, for example, of 70%. So it should not be the case that one silver policy has cost sharing different than the cost sharing of another silver policy in a way that would affect the premium charged for the policy. Moreover, the calculations underlying this post keep HMOs, PPOs, POS plans and EPOs apart; so it should not be the case that observed premiums differ because, perhaps, the cheaper plans are HMOs whereas the more expensive ones are PPOs.
Of course, cost sharing is not the only way in which policies within a given location, of the same metal tier and sold to the same purchaser could vary. One policy might offer richer benefits than another. It could have a richer network with more doctors available or more prestigious and expensive hospitals inside the network. Could that be responsible for a substantial part of the premium dispersion we see? It’s impossible to tell for sure — the data published by HHS does not attempt to quantify the richness of the network being offered. I do find it difficult to believe, however, that such differences are responsible for the entirety of differences in excess of 20% between the low bidder and the mean bid, or, for that matter, differences in excess of 40% that sometimes occur between the low bidder and the higher bidders.
Maybe the average premium is meaningless; the low bidder got it right
Of all the potential rejoinders I have considered, the one now forthcoming is the one that is most troubling. There is nothing the data standing by itself can tell us whether most of the insurers have it right and the low insurers are about to lose their shirts or whether the low insurers have been more insightful or have managed to keep costs down such that they will break even (or even make money) selling their policies at low premiums. And, yet, I am doubtful. One can view the mean or median of the premiums as an “ensemble model” of the true cost of providing care under the Affordable Care Act. And there is research (examples here, here and here) suggesting that ensemble models predict better in many open-textured situations than individual models. So, while it’s possible, I suppose, that in every jurisdiction the low bidder is predicting more accurately than the group of insurance companies as a whole, such a result would be surprising. A far simpler explanation is that the low bidder — the one who is likely to win business from price sensitive insureds — is succumbing to “the winner’s curse.”
Maybe the disaggregation of plans is misleading
This is a very technical objection, but consider carefully what I have done. I have looked at all policies of a given metal tier and a given plan type in a given geographic location sold to a certain family type such as “all silver policies in Harris County, Texas, sold to The Chandlers.” But, really, plans are sold not to just to The Chandlers but to all family types. So, it could conceivably be that while the plans sold to the family type I am looking at are highly dispersed, the average premiums over all family types (weighted by prevalence of the family type) are far less dispersed. This strikes me as unlikely — why would an insurer be overcharging one family type relative to another — but you can not rule it out a priori. Maybe — just maybe — the dispersion we are observing is not real; it is just an artifact of my disaggregation of the data.
I would, of course, love to aggregate the data and see if the high degree of dispersion persists. The difficulty with this cure comes with the problem of weighting the data. We don’t know the distribution of policies sold among family types. We don’t know, for example, whether The Chandlers constitute 2% of policies sold or 5% of policies sold. So, I can’t perform a perfect aggregation of the data. One way to get a feel for the objection, however, is to simply take an unweighted average of the premiums for all the family types identified in the database and aggregate it that way. This is far from perfect, and we could spend a lot of time refining it, but it should provide a clue as to whether the disaggregation of plans is significantly responsible for the high degree of observed dispersion.
The graphic below shows the exceedance curve for losses of the low bidder assuming the mean premium is the true risk based on an unweighted average of family types purchasing the policies. One can see that 20% of the low bidders will lose at least 20% if it turns out that the mean premium charged for similar policies reflected the true risk. Upwards of 35% will lose more than 10%. A quick comparison of this curve with those above shows that it is essentially the same. There is nothing that I can see suggesting that the fundamental result shown in this blog entry — high dispersion of premiums among what should be similar policies and the potential for significant losses by low bidders — is an artifact of the methodology I have employed.
In the end, even the extensive data that the government is put out is insufficient to determine definitively whether the lower priced insurers in the individual Exchanges are about to lose money. There are more optimistic interpretations of the observed premium dispersions: maybe it is the low bidders who are “getting it right” or maybe the low bidders have just found ways to keep costs down through better negotiating or cheaper care networks. But if these optimistic explanations prove insufficient, what this post shows is that while some insurers will likely do just fine there are a substantial minority of insurers who are about to get bitten by the “winner’s curse” and get a large volume of purchasers for whom the premiums charged will be insufficient to defray the expenses incurred.
The data used here was taken directly from the United States Department of Health and Human Services. It was analyzed using Mathematica software, which was also used to produce the graphics shown here.
Exploring the likely implosion of the Affordable Care Act