On his Substack, Arnold Kling, resident book reviewer at Econlib, wrote:
If people pay for their own health insurance, the market is subject to selection games. The individuals with the most incentive to buy health insurance are those that will cost the insurance company the most in claims. (Although it turns out that there is a selection effect that goes in the other direction. People who are high in conscientiousness are more likely both to obtain health insurance and to take better care of themselves.) Insurance companies, by the same token, have an incentive to try to avoid writing policies for people who most need health insurance. Nobel Laureate Joseph Stiglitz was known for pointing out that this selection game might have no viable solution: the health insurance market could collapse entirely. (italics added)
I think Arnold engaged here in what Ronald Coase called “blackboard economics.” The idea is to think through the incentives that the various players face and, on that basis, make conclusions about the way the world is. It’s appropriate that he cited Joe Stiglitz because Stiglitz has been one of the masters of blackboard economics.
The problem is two fold. First, under this approach, you can sometimes be tempted not to think through the incentives all the way. A clear incentive is for the insurance company not to deny insurance to a high-risk person in the individual market but to underwrite insurance. That means assessing risk and charging a premium that reflects that risk.
Second, the blackboard approach ignores the evidence. In a blog post in 2010, I discussed some interesting economics in the 2010 Economic Report of the President and the way the author of the health care chapter, whom I assume was the CEA’s health economist Mark Duggan, now head of SIEPR at Stanford, twisted himself into a pretzel to justify Obamacare. The whole post is worth reading. Here’s the particular part I want to emphasize here:
A House committee investigation found that three large insurers rescinded nearly 20,000 policies over a five-year period, saving these companies $300 million that would otherwise have been paid out as claims (Waxman and Barton 2009). (p. 188)
20,000 policies over 5 years is 4,000 policies per year. So the average number of policies rescinded by the 3 companies individually was about 1,333. Is that a large number? Given that the author says these were large companies, I don’t think so. As one commenter on my post noted at the time:
The five largest insurers nationwide each have between 10 and 40 million members, which is ~4-15M policies. The next half-dozen or dozen are all million+.
The bottom line is that it appears that very few people who wanted health insurance and were willing to pay a premium that reflected their risk went without.
READER COMMENTS
steve
Jul 11 2023 at 11:10am
It’s difficult to get access to private health insurance company records. I doubt very much their numbers were accurate. People who follow the industry estimated much higher numbers than they provided.
Many people were unable to obtain insurance as it was not offered. From personal experience my brother-in-law was a high risk pt. He was self employed and was unable to obtain even an offer. The wife and I had significant income so I called our corporation’s insurance broker whom I knew well. I explained to him that we would cover the costs. He could not find any plan willing to take him.
Talk with the actuaries. At the extremes it gets hard to predict numbers. With high risk people the costs can be very high and harder to predict when they are incurred. Note that the profit for insurance companies lies with healthy pts.
You forgot a couple of words. “were willing AND ABLE” to pay a premium.
The problem was common enough that many states offered state financed insurance to those unable to find insurance. It was generally expensive but for those who could afford it better than nothing. They usually had waiting lists.
Steve
David Henderson
Jul 11 2023 at 5:21pm
You write:
I didn’t forget; I used “willing” the way economists use the term. When we say that someone is willing to pay something, we mean that when presented with the option at that price, they will choose to pay. We don’t mean that they won’t pay now but if someone, say, doubled their wealth, they would pay.
Thomas Hutcheson
Jul 11 2023 at 12:22pm
I agree that the problem of adverse selection turned out to be less of a problem than anticipated.
For me the strong point of ACA is that is could be the kernel of a universal system in which purchase their own insurance with tax credits rather than having some people’s employers buy it for them. I think this would be better than Medicare for all because insurance companies would have more incentive to incentivize cost-effective treatments than DHS. Addionally as we create the new system, we should have the opportunity to finance it with a VAT rather than a tax on wages
steve
Jul 11 2023 at 4:27pm
“I think this would be better than Medicare for all because insurance companies would have more incentive to incentivize cost-effective treatments than DHS.”
Not much evidence this actually exists. Private insurers pay more than Medicare and Medicaid. If you look at the rate of increase of cost private insurance (per person) has increased faster than public insurance. There are periods where it is close and in a given decade private might be the same or a bit lower, but overall private insurance costs grow faster. From 2008-2021 about 54% for privates, 37% Medicare and 19% Medicaid. This unfortunately ignores the costs on the provider side. It costs much more to bill and collect from private insurers than public. I try to avoid putting in a second link but if you are interested Health Affairs has a number of papers on this.
My personal observation based upon being in charge of billing for a 200 person provider group for a long time is that insurers seem much more concerned about market share than holding down costs. They are very averse to negative publicity about not covering therapies even if marginal and would rather just pass on cost increases.
https://www.healthsystemtracker.org/chart-collection/u-s-spending-healthcare-changed-time/#Total%20national%20health%20expenditures,%20US%20$%20Trillions,%201987-2021
Steve
Thomas Hutcheson
Jul 12 2023 at 1:29pm
The USG has some monosomy power.
Matthias
Jul 11 2023 at 7:23pm
Why not just have a system were people by insurance on the open market? (And no silly games were employers can buy insurance with pre-tax money.)
If you want to help poor people and/or those with bad health, just give them money?
Alternatively, it’s almost always worthwhile to have a look at what Singapore does. As a society, the US spends about 20% of their GDP on healthcare (that’s both public and private expenditure.) The UK spends about 10%. And Singapore spends about 5% of GDP. At no worse health outcomes.
Thomas L Hutcheson
Jul 12 2023 at 1:32pm
That’s what I said. Individuals get tax credits and buy their own insurance.
steve
Jul 12 2023 at 2:35pm
At this point probably inertia and insurance company preference. Most people are insured through their place of work. If you had added an extra 100 million people to the insurance markets in the 70s or 80s it would have cost a lot to individually evaluate risk without the aid of computers. It was a lot safer/cheaper for insurance companies to insure larger groups where the risks average out. That’s not quite so much of an issue now as I believe you could use computers to do the processing quickly and cheaply. So now there just isn’t a model of this kind of insurance working anywhere in the US and in other countries it always associated with a lot more govt intervention than we have.
Steve
nobody.really
Jul 11 2023 at 2:32pm
The title of this post is “Pre Obamacare, Did Health Insurance Companies Refuse to Insure a Lot of People[?]”. David Henderson cites a quote addressing how many policies insurers rescinded. I don’t know that counting rescinded policies answers the question.
David Henderson states, “very few people who wanted health insurance and were willing to pay a premium that reflected their risk went without.” Steve appropriately remarks that this statement says nothing about a person’s ability to pay such terms.
Moreover, this discussion begs the question, what counts as an individual’s risk? Insurers provide a lot of insurance via contracts to employers. Given 1) economies of scale and 2) the fact that employers tend to hire people based on criteria that do not correlate positively with (and generally correlate negatively with) high health-care costs, insurers tend to offer these group policies at lower rates—and least, compared with rates offered to individuals. So when we talk about a premium that “reflected their risk,” we’re often talking about risks arising from not being employed by a large employer. Does it make sense to characterize this as risks arising from an individual—or from a system?
I suspect many of the comments David Henderson reads arise within a context in which people are seeking to defray risks arising primarily from the vagaries of chance. For example, Hayek wrote—
Friedrich A. Hayek, The Road to Serfdom, Chap. 9, “Security and Freedom” (1944).
People with this attitude seek to design policies that take the economics of group health policies and extend it to a larger share of the population. To be sure, these policies will lack some of the advantages of group health plans—specifically, the policies do NOT seek to charge higher premiums to individuals born with expensive genetic defects, because they regard the risk of having such defects as a SOCIAL risk (e.g., evaluated from a standpoint preceding birth, a la Rawls) rather than an individual risk. For people who cannot grasp the concept of a social risk, this argument may seem nonsensical.
It’s entirely possible that, in the absence of government meddling, the health care market would operate every bit as efficiently as the market for luxury cars. But if I were to make that argument to Hayek, would he have responded with, “By golly, I’d never thought of it that way—you’re right! Concern about the ‘common hazards of life’ is clearly misguided and muddle-headed!”? I guess we’re all left to draw our own conclusions.
Thomas L Hutcheson
Jul 12 2023 at 1:42pm
I would add unemployment to Hayek’s list of risks that the state might insure and remove most of the Acts of God, because even if God sent the flood, the individual did not have to build a house in the floodplain. Forward-looking rates of hazard insurance premia ought to be part of our mitigation of the effects of climate change.
National Jester
Jul 11 2023 at 4:36pm
What percentage of Americans “bought” their healthcare as an individual, and how many purchased it through a group (such as an employer) prior to Obamacare? My experience, as a government employee, was decidedly negative. The group purchase prevented rejection based on health, which was good. But non-smokers paid the same rates as smokers, which I believe was a mistake. My experience was that my rates increased significantly, my benefits and choice of doctors/hospitals decreased and I was less sure of what was actually covered. At what level can negative lifestyle choices be factored into rates before people drop coverage? I’m now a market consumer. My insurer pays practically nothing towards my care, but their negotiated rates substantially lower my out of pocket expenses. If I could obtain those rates as an individual, pay cash for service and place the money now going to the insurance company into a health savings account that would be awesome.
vince
Jul 11 2023 at 4:57pm
Yes, sir. Wouldn’t it be great if insurance companies would pool risks according to verifiable voluntary behavior? Engage in all those risky behaviors you want. Just be prepared to pay for it without subsidies from those who don’t want to take those risks.
Dylan
Jul 11 2023 at 5:55pm
It seems to me that insurance markets can only exist if we can assess risk relatively accurately at some group level, but not all that accurately at the individual level. The more fine tuned and predictive the risk-assessment, the less value there is. If you’re part of a pool that has a million people in it, and we know that all but 500 of those people aren’t going to cost them anything, but those 500 people are going to cost them an amount that would cost more than 10x the lifetime income of the average person, then insurance likely makes sense and you can have a premium that is a good bet for everyone. Let’s say you get really, really good with your models and you can split those pools into 999,000 people that have zero risk, and 1000 people that have those 500 risky people in it. Now, neither side has an incentive to buy insurance, the 999,000 don’t have any risk and the pool of 1000 is super risky, but insurance for that market has to be more than they could possibly afford.
Jon Murphy
Jul 11 2023 at 7:05pm
I don’t follow your logic. What you describe as “the only way the insurance market can work” is the adverse selection problem discussed above and you show that it results in the market collapsing.
Dylan
Jul 11 2023 at 9:44pm
Adverse selection is when one side of the market has more information than the other. In my scenarios both sides have the same info, but in the first insurance works because of the big uncertainty. All I was trying to say is that the more certainty there is, the less viable insurance as a profitable venture can be. In health insurance there are plenty of “pre-existing conditions” that remove a good chunk of uncertainty. Yes, an insurance company might write you a policy, but only at a rate that would incorporate that condition. At that point, if you can afford the rate, you could also afford to self insure for less money.
Jon Murphy
Jul 12 2023 at 7:31am
I know what adverse selection is and how it works in insurance. I teach that. My point is you’ve incorrectly identified it. The story you lay out is an adverse selection problem. The insurance companies are at an informational disadvantage, so they have to pool groups with different risk profiles. That does lead to the market breaking down (in theory). But if the company was able to accurately distinguish who in the group is risky and who in the group is not, then premiums will be different for each group.
Dylan
Jul 12 2023 at 9:24am
Hey Jon, I’m sorry I must not be explaining my hypothetical correctly. I’m trying to say explicitly that the insurers are not at an information disadvantage.
Yes! And that works great up to a point. But the more you stratify into different groups, the more risk (which is just another word for uncertainty) goes down, not up. And risk is the thing that is necessary for insurance markets to work. The EV of any insurance policy is negative, it has to be for the insurance company to stay in business. As long as there is a big variance in outcomes and people are risk averse, it makes sense to still buy insurance with a negative EV to protect against catastrophic loss. I think I’m not saying anything controversial up to this point, right?
But, to illustrate the point with an absurd example. Let’s assume that both the insurance company and the customer knows with 100% certainty that the customer is going to need exactly $x worth of service over the lifetime of a policy. The insurance company would have to charge the PV of $x + their margin. The customer wouldn’t take the deal, because they could just invest the PV of $x without having to pay the insurance company’s margin. Still in agreement?
What I’m saying is, up to a certain point, accurately distinguishing who is likely to cost more and who is likely to cost less probably makes insurance markets better. But, there is a tipping point, long before you get to 100% certainty, where insurance markets will stop working. Even if both sides of the market have symmetrical information.
steve
Jul 12 2023 at 2:45pm
Dont forget the actuarial side and marketing. The actuaries have more trouble predicting costs at the extremes as they dont have enough data. That makes very sick people a high risk. It’s safer to just avoid them. They could maybe adjust and quote super high rates to be safe but they know at that point people wont pay. Besides that from a marketing POV they really dont want word going around that they are charging $200k a year for insurance.
Steve
Knut P. Heen
Jul 12 2023 at 8:30am
The adverse-selection problem may not be that big when it comes to health insurance. The insurance company may use its own people to assess the health of the customer. The customer is no expert on the issue either. How many think they have a serious disease without having one?
Moral hazard seems like a more serious problem. That affect people who are insured, but do not take appropriate steps at maintaining their health (for example, by exercising regularly). Moral hazard is more of an issue for the state provided health care in many European countries because taxes cannot be tied to BMI. You can tie insurance fees to BMI. They try to increase taxes on sugar, but that does not get people to exercise more. People who exercise regularly may even need to consume more sugar than people sitting in the sofa.
Robert Hertz
Jul 13 2023 at 4:20pm
There does not seem to be a lot of hard data on how many applicants were denied health coverage before the ACA. I was an insurance agent myself, and I advised many persons not to even try for coverage because I knew they would be denied.
The Kaiser Foundation (which is pretty reputable) did a study in 2016 titled ‘Pre-existing conditions and Medical Underwriting in the Insurance Market before the ACA’ (12/12/2016)
They looked at the actual underwriting guidelines of several insurers in the individual market.
There were a lot of declinable conditions (and even some declinable occupations, like firefighting, logging, mining, etc.) The declinable conditions were what you would expect — rheumatoid arthritis, Hep C, Congestive Heart Failure, diabetes, MS, Crohn’s, fibromyalgia, heart bypass.
Not all of the sufferers went uninsured. Some got onto a spouse’s plan, others took any job that offered insurance.
Some of the comments to this post seem very naive. You cannot underwrite someone who has just been diagnosed with cancer, and who will incur $250,000 in costs for the next 12 months….or a boy with hemophelia, whose drugs might cost that much each and every month. These people exist, and they are vexing to all private insurers.
Vivian Darkbloom
Jul 15 2023 at 8:23am
I’m surprised that in this comment thread (and the blog post) no one has mentioned that, with a few exceptions, the problem of rescinding a health insurance contract or refusing to insure someone would not exist if having health insurance would be mandatory and people actually adhered to that requirement. Both these issues are those that “Obamacare” attempted to address (with some limited success, in my view).
That person who has just been diagnosed with cancer should not have a problem if he or she had insurance at the time of diagnosis. He or she has (justly) a big problem if that person had gone uninsured until the time when the problem had become evident.
The problem of contract rescission as referenced in the cited study normally arose when a person who developed problems was found to have been less than truthful in questionaires completed in an insurance application. To be sure, there were a relatively few cases (dutifully reported, of course) where the insurance company attempted to rescind a contract when an insured failed to disclose relatively minor and unrelated pre-existing conditions. But contract law should be a remedy in those few cases.
The exceptions that need to be addressed are cases in which a person has a congenital defect at birth or a condition before the age of majority. The other possible exception is where a person has relied on employer-provided insurance and, after leaving or losing that employment, is forced to seek insurance elsewhere. Despite Obamacare’s many shortcomings, I think it attempted to address these problems as well.
Critics of the existing regime need to propose solutions to these problems (lack of mandatory insurance, pre-existing conditions prior to the age of majority and portability of insurance) that existed prior to Obamacare rather than listing the latter’s failures.
Mark
Jul 23 2023 at 11:24pm
A broader application of “discrimination” is refusing coverage for preexisting conditions. That was a key problem pre-Obama.
The greatest problems were and continue to be the availability of care. Insurance cost is just built on cost of care, but we don’t need insurance. We need care.
Low cost.
Consistent positive results.
Provider nearby.
Services available in a timely manner.
Obamacare has not addressed any of these. #1 continues to rise, and #2,3,4 have not improved.
Comments are closed.