Health Insurance Markets

These are the notes I used for my remarks on health care reform at the forum I attended last week:

Health Insurance Markets, by Mark Thoma: In the ten minutes that I have, I’d like to talk about insurance markets. In general, insurance gives us financial protection from unexpected events -- a tree falls on our house, we have a car accident, we become unemployed, we become sick and need health care, and so on. In some cases, the insurance is provided by the private sector, though it’s usually with government oversight, but in other cases the government itself provides the insurance. When the government provides the insurance, we call it social insurance. Social insurance provides individuals or households protection against certain events such as unemployment, poverty during old age, health expenses, and disability. It is distinguished from other insurance by the fact that it is “undertaken, facilitated, or enforced by government as a social policy.”

We might leave all of our insurance needs to the private sector, but unfortunately insurance markets do not always function in a way that provides adequate levels of insurance at the lowest possible price. Economists use the term market failure to describe such instances and there are many reasons markets might fail. For insurance markets, the main problems are called moral hazard and adverse selection. Moral hazard is the tendency for people to engage in riskier behavior when they know they are covered by insurance – when the downside is covered there's no reason to take precautionary behavior. This causes too much risk to be taken raising costs. Fortunately, this class of problems can be addressed, though not completely eliminated, through deductibles, co-payments, and other cost sharing arrangements that cause the insured to pay a penalty when there is a need to use the insurance. But so long as the insurance company pays some share of the costs, some degree of moral hazard is likely to be present.

While moral hazard can be reduced, adverse selection is not as easily overcome. The specific form of adverse selection at work in these markets is asymmetric information. Individuals have much better information about their health histories and likely health outcomes than insurance companies. Because premiums are set based upon the average member of group, some members of the group will need more health care than they pay in premiums and other members will need less. Those who expect to use less health services than it costs based upon knowledge of their likely health outcomes will not get insurance – it doesn’t pay to do so – while those who expect to have higher costs will stay in the market. Then, as the lower expected health care cost people drop out of the market and higher cost people make sure to enter, average costs rise, premiums increase, and this then motivates more people to drop out leading to market failure.

The solutions here are to require people to disclose known conditions at the time they purchase insurance and to fully disclose their health histories to insurance companies, or to mandate that everyone must have coverage thereby preventing anyone from dropping out. This leads to concerns, however. The first is privacy. How much should people be required to tell insurance companies about their personal health histories? How much testing should insurance companies be allowed to do before signing you up? For example, is genetic testing and subsequent exclusion form coverage okay? That brings up the second problem, whether we want to discriminate in terms of the cost of health insurance based upon conditions people are born with, their genetics, or is that a risk that is properly shared across the population since an individual can do nothing to affect the genes they are born with?

There is another problem in these markets that also causes people to drop out and forego coverage. When people believe they will end up in the same general condition whether or not they have insurance, there is no need to purchase protection. If the unemployed believe society will take care of them when they are unemployed if they don’t have insurance, there is no need to purchase private unemployment insurance. If a person believes their retirement will be the same whether or not they save - that the government will not let them live in squalor and will provide the same basic standard of living they would have otherwise, or nearly so, if health care can be obtained whether or not the individual is insured, then there is no incentive to purchase insurance and these markets will fail.

This is one reason social insurance or regulations mandating everyone have insurance can be useful. By forcing everyone to be pooled together and share risks that are out of their personal control and by preventing people from dropping out of the market, the adverse selection problem is reduced and risks are shared broadly, but note that this requires government intervention through regulation, tax incentives, and other policies, or the government provision of insurance, and a societal judgment that these risks ought to be broadly shared.

In recent history, one way we have addressed the adverse selection problem is by purchasing health insurance through our employers. The bigger the group, the lower the costs from reduced administrative costs and other savings. Employer provided health insurance has been encouraged and facilitated by tax-subsidies which have also helped to hold the market together and to convince people who might not buy insurance on the individual market to purchase it through their employer.

Employer-provided health insurance is not a perfect solution to adverse selection – and it is especially imperfect for small employers – but it has presumably helped. More generally, regulation and subsidies can help with the adverse selection problem but, in the real world, they will never completely eliminate the adverse selection problem and will not, therefore, result in universal coverage. The only way to do this is to force people to buy health insurance, which is the approach that Massachusetts is attempting right now.

All markets fail to one degree or another. But that doesn’t automatically imply that the government should step in and try to correct the problem. Government is inefficient. Thus, when government gets involved, it is costly in terms of wasted resources and other problems. But market failures are costly too. These costs need to be balanced – doing nothing brings about the cost of the market failure and stepping in and doing something has the cost of government inefficiency plus whatever residual part of the problem remains after the intervention. Therefore, the government should get involved only if the costs it imposes are smaller than the costs of allowing the market to operate without any intervention. Often this test is not met – markets are not perfect but government interference would be worse – but that is not always the case and there are certainly instances where government intervention can improve the economic outcome.

In general, we can think of two types of solutions to the market failure problem in insurance markets. The first is based in the private sector and it involves the government creating laws and regulations that get the incentives right in these markets, and then letting the markets themselves provide the insurance. In general, given that intervention can be justified due to the presence of a substantial market failure, economists believe in the ability of markets to efficiently allocate resources and when market based solutions are available, they are generally preferred to more heavy handed regulatory structures. But it is not always possible to solve the problems in the private markets through regulation and other policies to create the right incentives, and in these cases the second type of solution, the government provision of the good, is the most efficient way to provide the good or service.

Do health markets suffer from substantial market failure? My assessment of these markets says the answer is yes and, in particular, the most difficult problem to solve, adverse selection, is the most serious problem plaguing these markets. My preferred solution is a universal coverage single-payer system, but there is room for disagreement on this. In any case, solving these problems will require us to choose between two courses of action, solutions such as Health Savings Accounts which reside in the private sector but are guided by a regulatory structure that attempts to create incentives for both providers and consumers of health care to behave optimally, and solutions such as expanding Medicare nationally into a single-payer system. For me the choice is not ideological, it is not Democrats versus Republicans or free marketeers versus advocates of the welfare sate, but rather it is a matter of economics. It comes down to which type of solution is likely to produce the most desirable outcome from a social policy perspective.

To summarize, let me emphasize the difficult tradeoffs we face. In the context of health insurance, there is no perfect solution that will solve all of the problems in these markets. We will have to make choices and there will always be trade-offs. If we have a private sector based system with wide and generous insurance coverage with small out of pocket expenses, there will always be moral hazard and the higher insurance costs that come with it. If we have a single payer system, there will always be problems arising from the government setting prices different from their optimal values leading to misdirected resources and other inefficiencies. If you don't want the government involved at all, you will have to accept that some segments of the population will go uninsured. There is no way to have it all. Instead, the question is which type of system is likely to produce the outcome most consistent with our social policy objectives.

Comments

[reason]

Mark,
isn't there a more fundamental problem here.

Society doesn't really believe that questions of life and death should be a function of income or wealth. Economics in western societies normally deals with elective consumption not basic needs. There is a fundamental conflict between a needs based and a money demand based supply paradigm.

I think only the most extreme idealogical libertarians would agree that: -
"It is better that poor, sick people be allowed to die".

[anne]

Nice summary presentation, though a gentle word about looking to an Australia or Germany for comparative structure could have been helpful. Looking abroad comparatively though is resisted fiercely. Nice.

[james]

two points

1. "Government is inefficient" - when did this statement become a law of nature? - an undeniable natural phenomenon - like gravity - if it is a law of nature - what does it mean when you use it like a trump card when arguing a position?

2. "moral hazard" and "adverse selection" are cliches and like most cliches they emphasize a subjective reality as opposed to an objective reality - selling insurance for health costs is an extremely dificult area of human affairs in which to try to sell services and make a profit - it is inevitable that you will incur more liabilty than you have accumulated in fess so you better set thresholds which alleviate some of that liability (co-called moral hazard") and you better make sure you sell your services to people who are not likely to use them

[reason]

james ...
has a point, I have worked both for government and in private industry and cannot say I have noticed a great difference in efficiency. Wouldn't it be better to say...
Government like all monopolies can tend to inefficiency if badly managed.

[slink/js paine]

mark nice clear expo

obviously its balanced
you are trying to provide insight not conclusions

[Richard]

Mark, two comments I'd like to make: First, most plans covering more than 10 or 20 participants self-insure: they retain the insurer not as insurance but as a third party administrator. They may reinsure some of the risk (such as claims in excess of 120% of some measure of the previous year's claims).

Second, it does seem to me that a fair number of people that I know who would start their own business do not do so because of the health care tether: pre-existing condition for a spouse or a child couple with coverage within corporate america. I often wondered whether the way we provision coverage doesn't create and "entrepreneuial damper", since the health characteristics that lead one to choose corporate health care provisioning are in no way correlated with the abilities to start a business.

[Bernard Yomtov]

Following up on Richard's comment, I think the problem is broader than just the "entrepreneurial damper" he mentions. Employer-based isurance creates potential problems for people wanting to switch jobs, retire early, or do any one of a number of things that involve leaving a job that provides coverage.

This should be seen not only as one more economic inefficiency, but also as a constraint on many people's ability to live their lives as they wish.

[evagrius]

I'm curious about the question of "moral hazard". If moral hazard exists in the health insurance area, then why do countries with single-payer or universal coverage seem to have better health outcomes for their population as a whole?

[Bruce Wilder]

Good talk. Some general observations:

"Government is inefficient"? That's an assumption or a political conviction?

Moral hazard and adverse selection are just two aspects of a single problem, which is that all risk/imperfect information has incentive effects on decision-making. The incentive effects of insurance are not limited to the behavior of insurer and insured. The system of private insurance is breaking down, not because of moral hazard and adverse selection problems, but because of escalating costs of health care, a problem which is not well understood and doesn't have clever names.

Ultimately, we would like people to be autonomous agents, able to make responsible choices for themselves, and able to trust the choices made by others, including medical care providers, professionals and technicians. Both distributing and attenuating risk throughout the system of health care is a general problem of organization. Markets don't really allocate resources outside theoretical models -- administrative organizations do that in real life -- but they do help to objectify value and cost, something that administrative organizations are not able to do.

A universal system does not have to be, literally, single payer, nor does it have to be financed by a single tax. Separate agencies might administer payments for emergency care, medical education, public health measures, childhood vaccinations, hospice care, etc. Health care is over 15% of the total economy. Monolithic administration is unlikely to be effective or efficient, and "single payer" proposals should not be taken to imply monolithic administration.

One of the critical failures of private insurance in health care is the complexity of administration, which it has created. Health insurance administration is one of the fastest growing sectors; it is sucking up resources and those resources are being wasted.

The government, in administering a universal health care system, is free to finance health care from a variety of taxes and fees, which can manipulated to address the problems that might manifest as adverse selection or moral hazard in a system where insurance is optional. For example, a tax on cigarettes can finance related health care measures.

The rules governing private accident and liability insurance ought to be harmonized with universal health care. (Personal note: I don't carry accident insurance on myself, beyond the absolute minimum mandated by law, because my health insurance requires that I transfer any proceeds to them.)

[save_the_rustbelt]

I have trouble with the moral hazard argument here, I think people drink, smoke, eat too much and exercise too little because they enjoy it (neat term) or have a lack of discipline, rather than because they have insurance.

Having said that the system is a mess.

The "insurance tether" argumentis quite true, and is a drag on the economy, and I think the tether gets tighter as we get older and have more responsibilities.

Single payer, though, scares me for many reasons.

[Barry]

Posted by: Bernard Yomtov: "Following up on Richard's comment, I think the problem is broader than just the "entrepreneurial damper" he mentions. Employer-based isurance creates potential problems for people wanting to switch jobs, retire early, or do any one of a number of things that involve leaving a job that provides coverage."

(Warning - unsourced claim!): I remember hearing about a study comparing the US and Canada, which found that a higher percentage of the labor force was self-employed in Canada. The researchers had concluded that the major factor was that, in the USA, quitting a medium-to-large employer to start one's own business meant giving up health insurance. This was, of course, a particularly large obstacle for people with children. One significant health claim in the course of a few years could cause bankruptcy. In addition, many policies exclude coverage for 'pre-existing conditions', so, even if one were to purchase coverage one their own, they might have serious holes in that coverage.

In contrast, a Canadian who wanted to start his/her own business would not lose health coverage.

[Mark]

Portability of the health care coverage under HIPAA has solved the job lock dilemma for most in the United States. If you leave a large employer to start your own business you can usually obtain similar coverage at a similar price if you are in good health. In fact the cost can often be less! If you or a family member has an ongoing medical condition or recent adverse health history, the cost of coverage may prevent that person from leaving a large employer where insurance cost would be less. That is also a barrier to early retirement for many in the US.