Wednesday, December 28, 2005


Prepare for a bit of brain strain below as Arnold Kling applies economic thinking to health insurance and concludes that few people want real health insurance. What they want is a freebie that someone else pays for. I think he is a bit too cynical but agree that the only realistic way to prevent free-riders is to mandate that everyone contribute to some health insurance scheme or other -- as is already done in Australia

Policy pundits are unhappy with the state of health insurance. What is the problem? After considering some alternative theories, I believe that the best explanation is simply that most people do not want health insurance. The cost of health insurance has been rising, leading to well-publicized problems in the employer-provided health insurance system and increasing numbers of uninsured. But blaming insurance companies for that is like saying that the calories in a double-fudge chocolate cake are all in the icing. The cake of health care expenses consists of health care services -- doctor visits, surgeries, and all the rest. The icing consists of health insurance -- administrative costs, profits and all that. In dollar terms, the icing represents less than ten percent of the iced cake.

Many proposals to reform health care finance mistake the icing for the whole cake. They act as if the cost problem is concentrated in health insurance per se, rather than the medical system as a whole. They make proposals to change the system of icing in various ways, with the most dramatic proposal being single-payer health care, with the government providing people with health insurance. The reality is that re-doing the icing will not have much effect on the cake, as the icing is not the reason that the cake has so many calories.

On the Left, the arguments runs: Europeans have lower per capita spending on health care than Americans. Europeans have much more of their health care paid for by government than Americans. Therefore, we could lower our spending on health care by switching to single-payer health insurance. However that is like saying that because a fruit cup topped with powdered sugar has lower calories than double-fudge chocolate cake topped with icing, that we could have a low-caloried dessert by replacing the icing on the cake with powdered sugar. Consider this list of factors that affect the cake of health care costs.

* health insurance industry greed
* health insurance industry paperwork, inefficiency, and overhead
* asymmetrical information in the health care market (see below)
* premium medicine -- expensive specialists and medical equipment

All of these factors are present, and they all contribute to the calories in the cake. But when we talk about the breakdown of the health insurance market, we need to ask: why now? Have insurance company executives become greedier in recent years? Have they become less and less efficient? Have information asymmetries increased? Has there been an increase in specialization and use of high-tech medical equipment?

The first three possibilities are dubious, at best. But the fourth explanation holds up pretty well. For example, since 1975, we have more than quadrupled the number of gastroenterologists, pulmonologists, and diagnostic radiologists. Since 1980, the annual number of CT scans has grown from less than 4 million to more than 50 million and the annual number of MRI's has grown from 0 to close to 25 million.

Or consider the "natural experiment" of Medicare and Medicaid, which is a much thinner layer of icing according to New York Times columnist Paul Krugman and other advocates of single-payer health care. If the icing really is thinner, and if the icing is a big factor in the total calories of the cake, then total health care spending under Medicare and Medicaid should be noticeably lower than spending under private insurance, after controlling for population characteristics. Instead, if one uses other OECD countries as a control group, our spending on the elderly is as excessive relative to other countries as is our spending on those without private insurance. In fact, Medicaid and Medicare, which together cover less than half the U.S. population, absorb a higher proportion of our GDP than many other countries' single-payer systems that cover their entire population.

Massachusetts Governor Mitt Romney has a proposal to reform health insurance in his state. One interesting fact about Massachusetts is that per capita health care spending there is more than 20 percent above the national average of over $5000 per person. Do you suppose that is because insurance companies are greedier or more inefficient in Massachusetts than in other states? Or could it be because Massachusetts is home to top-flight medical schools and world-class hospitals, giving it an unusually high number of specialists per capita as well as plenty of high-tech equipment? I suspect the latter, in which case Governor Romney's proposal to tinker with the icing will probably not work.

Krugman has pointed out, correctly, that 5 percent of the population accounts for 50 percent of health care spending. This suggests that in a health insurance pool, the sickest 5 percent could have their expenses paid by everyone else. In any given year, 95 percent of people would not be paid claims, but the sickest 5 percent would receive payments from health insurance.

The catch is that the cutoff for being in the sickest 5 percent is $10,000 of health expenditures. Real health insurance would have a deductible of $10,000 per person, so that the other 95 percent of people would not receive any money. Do you think that politicians will propose single-payer health insurance with a $10,000 deductible? Don't hold your breath. We do not observe insurance policies with $10,000 deductibles in the market. There are two possible explanations for this:

* asymmetrical information
* people do not really want insurance

The asymmetrical information story is that people know too much, relative to insurance companies, about their risk of requiring expensive care. As Tim Harford puts it in The Undercover Economist:

"the insurance company only sells insurance to people who are confident they will use it. As a result, the insurer loses clients who are unlikely to make claims and acquires the unwanted clients who are likely to make costly claims, and then the insurer has to cut back on benefits and raise premiums...More and more people cancel their policies... The curious conclusion, which is obvious in retrospect, is that an insurance policy depends on mutual ignorance."

This is a clever story, due originally to Nobel Laureate Joseph Stiglitz and much beloved in the economics profession, for why health insurance markets might break down. The very term "asymmetrical information" is esoterically cool. For example, it is the title of one of the best economics blogs around. In reality, however, when it comes to forecasting our health care needs, consumers and insurance companies operate in an environment that more closely resembles mutual ignorance than asymmetrical information. Mark Pauly and Bradley Herring, who, unlike the many armchair theoreticians, have examined actual health insurance markets, find that insurance companies can pool risks reasonably well.

Recently, Alex Tabarrok commented on a number of markets where economists like to trumpet information asymmetries. His conclusion is that real people and real markets have found solutions to prevent the "adverse-selection death spiral."

What we are left with, then, is that people do not want real health insurance. I would gladly take a health insurance policy with a $10,000 deductible per individual, and I suspect that many of my wise, risk-averse TCS readers would, too. But we are in a tiny minority! Most people do not want to be responsible for the first $10,000 in medical expenses, and most people believe that an insurance policy that is expected to pay no claims 95 percent of the time is a bad deal.

I am willing to claim that no insurance market in history ever arose because of spontaneous demand on the part of consumers. Maritime insurance, which was one of the first forms of insurance, was demanded by creditors as a condition for lending money to shippers. Life insurance also initially arose to meet the needs of creditors who were lending money to pensioners. Homeowners' insurance is standard because it protects mortgage lenders. Collision insurance for autos is optional if you own yours free and clear, but not if you still owe money to the finance company. William Tucker is right. For the most part, people buy insurance because it is mandated by others. Insurance does not have a large natural market.

What we call health insurance also arose to meet the needs of creditors. In this case, the creditors were doctors and hospitals, who wanted assurance that they would be paid for service. Comprehensive, first-dollar health coverage, which is not really health insurance, protects suppliers, not consumers.

During World War II, employers entered the picture. According to Milton Friedman, they offered health care benefits instead of wage increases, because the latter were capped by wartime controls. My strong suspicion is that people like health insurance as it exists today because they mistakenly believe that they are getting something for free.

Tucker argues that government should mandate a low-premium, high-deductible health care policy. (In the Romney plan for Massachusetts, the only way to avoid such a mandate is to post a $10,000 "bond" that guarantees that you will pay your medical bills.) Ironically, this is a relatively libertarian proposal. It is relatively libertarian because the only realistic alternative is for government to continue to provide and/or subsidize the comprehensive "insurance" that is prevalent today.

In fact, I think that Americans are too mentally ill to accept a proposal as sensible as mandatory catastrophic health insurance. We will continue to act as if health care is something that should be paid for only by someone else, never by oneself. We will forever be demanding the double-fudge chocolate cake, and expecting someone to come up with a way to get rid of the calories by messing with the icing. In that sense, mandatory catastrophic health insurance would be just one more attempt to mess with the icing. However, it might get people to pay attention to the calories that are in the cake.

Ultimately, I believe that America could use a commission to provide more information about the calories in the cake of modern medicine. In other words, consumers need data on the costs and benefits of various medical protocols. But as long as consumers are insulated from cost by the so-called health insurance that exists today, there is no incentive for them to pay attention to such data.



For greatest efficiency, lowest cost and maximum choice, ALL hospitals and health insurance schemes should be privately owned and run -- with government-paid vouchers for the very poor and minimal regulation. Both Australia and Sweden have large private sector health systems with government reimbursement for privately-provided services so can a purely private system with some level of government reimbursement or insurance for the poor be so hard to do?

Comments? Email me here. If there are no recent posts here, the mirror site may be more up to date. My Home Page is here or here.


No comments: