Cap'n Arbyte's


Other sites

What Everyone Needs to Know about Health Insurance

Why do people desire health insurance? If you answer "to pay for medical expenses", stop for a minute. Do you have food insurance to pay for food expenses, or textile insurance to pay for clothes? Of course not. We recognize that insurance makes no sense for food and clothing. Why do we treat medical expenses differently, expecting routine care to be covered by insurance? What is insurance really about?

Insurance is a vehicle for spreading risk. The participants prefer a small but certain cost instead of a large but unlikely cost. Let's proceed by analogy, using household fire insurance to explain some important features of insurance in general, and health insurance in particular.

The odds of your home burning down are low, but the cost if it happens is tremendous. For most people, the uninsured loss of their home would be a ruinous financial burden. By purchasing insurance, they are protected from the catastrophe, but must regularly pay an insurance premium to get this protection.

As economists never tire of saying, there are no free lunches. In total, insurance premiums must exceed insurance claims or the insurer will go out of business. The cost of anything covered by insurance is paid for through insurance premiums. For insurance companies to pay more or higher claims, premiums must correspondingly increase.

In the case of health care, the tax-advantaged status of insurance premiums has created a large distortion in the market. General medical spending is only tax-advantaged if it exceeds an certain (uncommon) percentage of your income, but employer-paid insurance premiums are always tax-advantaged.

The other mantra of economics (besides "no free lunches") is that incentives matter. Most people don't qualify for the tax benefit on out-of-pocket medical spending, but through their employer they do get the tax benefit for insurance premiums. Because insurance premiums fundamentally match insurance claims, in aggregate, an insurance premium can be considered as simply another way — a tax-advantaged way — of paying for health care.

The incentive is to pay for medical care through insurance premiums rather than out-of-pocket. This favors insurance plans that offer wider coverage and smaller co-pays or deductibles.

If you haven't ever thought of it this way, and therefore think you haven't been affected by the tax incentive, I have good news for you: You didn't have to think of it, because your employer and the insurance companies already figured it out. The process of economic competition has brought you the tax benefit even as you were unaware of it.

An interesting historical note in this area is that employer-provided health insurance grew out of price and wage controls during World War II. Prohibited from raising wages, employers competed for employees by raising non-wage benefits such as health insurance. If you stop to think for a moment, doesn't it seem strange that individuals themselves pay for all kinds of insurance except health insurance? What sense does it make for any kind of insurance to be tied to your job? It doesn't; it's a legacy of wartime economic intervention that tax incentives have made permanent. Whenever a market doesn't make sense, look for government intervention. You'll find it.

What are the effects of low co-pay/deductible, wide-coverage health insurance? Consumers don't pay much attention to price. (When was the last time you compared different doctors' prices for an office visit? If your co-pay is always the same, you don't care what the total cost is.) The aggregative nature of health insurance means that any individual's health care decisions have a negligible impact on premiums. The incentive is to consume more, because other people bear almost all the cost. When large numbers of people do this, premiums increase. This is a description of what has been happening in the United States over the past several years.

As aggregate systems of payment grow, the system more and more resembles a socialist one. The end state of this trend, nationalized health care, is within sight and is positively desired by many. Socialism only strengthens the perverse incentive to consume as much health care as possible. Socialist systems "solve" the problem of ballooning costs by rationing. In Canada it takes months to get an MRI scan; in England it takes years. Rationing has come to the United States also, in the form of HMOs denying treatments. Fortunately, in the United States, federalism has made it possible to experiment with and to end socialist systems before they reach their full destructive potential. (That program, incidentally, was very similar to Kerry's national health care plan. Its cancellation — by a Democratic governor — should give supporters of Kerry's plan serious pause and reflection.)

What would I like to see?

I would like to see a move back toward insurance as insurance — protection against catastrophe — instead of an aggregate payment system for health care. The way to achieve this is to end the tax advantage of employer-paid premiums over out-of-pocket payment. Health Savings Accounts do this by giving the same tax advantage to both, making high-deductible (catastrophic) coverage attractive again. This form of coverage restores the incentive of individuals to pay attention to price, rekindling economic competition and simultaneously reducing excessive consumption of health care.

The most serious objection to my proposal is that people who have chronic conditions will be worse off. As people who are low consumers of health care migrate to catastrophic coverage, the subsidy they had been implicitly providing to the chronically ill will be reduced. This is true, and in fact is a major personal motivation for me to switch to a catastrophic coverage plan with a Health Savings Account. I don't want to be involved in a quasi-socialist scheme paying for other peoples' medical bills.

This is a complicated matter to address and requires additional background.

Imagine a town filled with very similar homes, each with a small risk of fire. If everyone buys an insurance policy, the people whose houses burn down are covered. The others have lost a little money paying premiums. Ordinarily people are happy to do this; most people prefer a small but certain cost over a large but unlikely cost.

So far we have assumed the risk is evenly distributed. What if it isn't? What if the investigation of a fire showed that it was caused by particularly bad wiring, and that it was known exactly half the homes in town had been wired similarly and were therefore at increased risk of burning down?

This discovery changes the risk assessment. Fires in half of the homes are more likely than had been previously realized. Higher claims must mean higher premiums. (There are no free lunches.) At this point the insurer faces an alternative. Should everybody pay higher premiums to cover the overall higher rate of claims, or should just the high-risk group pay higher premiums?

The free-market outcome would be for only the high-risk group to pay higher premiums. If the insurer tried to raise everyone's premium, the low-risk group would be unhappy and would switch to a competing insurer who offered the original premium on the condition of knowing the house had low-risk wiring. Because the risk groups are identifiable, the natural outcome is a separation between them. Each person would be in an insurance pool with others of a similar risk profile, and their premiums would reflect their risk. (DWL: Please no nitpicking about situations without stable equilibria, I'm trying to keep this simple!)

The alternative of everyone paying a higher premium could be obtained through government intervention. However, it contains an implicit transfer of funds from the low-risk group to the high-risk group. That arrangement would be favored by the high-risk group but resented by the low-risk group. Because the situation is politically created, the friction between those two groups becomes a political matter, with all that that implies.

The example of discovering bad wiring is analogous to a person discovering they have significant risk factors for a serious disease. It increases their likelihood of making insurance claims.

Adjusting fire insurance premiums based on the quality of the house's wiring is analogous to adjusting health insurance premiums based on risk factors. Chronic conditions may be thought of as a risk with probability 1.

Just as bad wiring is in no way the fault of the homeowner, a chronic condition may be (or in the case of genetics, is) in no way the fault of the patient. However, insurance is based on risk, not on fault or need.

There is no free lunch in insurance or anywhere else. On an aggregate basis, every person should expect to pay more in premiums than they ever collect in claims. Insurance isn't a vehicle to save money. Insurance isn't a vehicle to pay for expenses. Insurance is a vehicle to dilute risk. If you have any other expectation, you will be disappointed.

In a free market, people with chronic conditions will not be subsidized through the insurance system by people without chronic conditions. And that is as it should be. No one has the right to force the cost of their higher risk to be shouldered by someone else. Not in fire insurance, and not in health insurance. The fair outcome is to be in an insurance pool with people of similar risk.

You may find it morally repulsive that the chronically sick must pay the most out-of-pocket. Yet it is undeniably true that they are the highest-risk group. I encourage you to ease your conscience with your wallet, not your vote — give to medical charities or directly to those in need. Do not hijack and politicize the insurance system. That leads to socialism and the destruction of effective health care for everyone.

Let insurance be insurance, and let charity be charity. Do not confuse the two.

UPDATE 2004-11-22 01:30:47 UTC: Further thoughts on financing care for the chronically ill.

Tiny Island