University of South CarolinaArnold School of Public Health HSPM 714 J50

Insurance theory concepts


premium -- what you pay to get the insurance coverage. In return, the insurance company agrees to pay you or provide you with benefits if certain events, spelled out in your policy, occur.


indemnity benefits -- insured gets fixed amount of money if specific event occurs. Life insurance does this. (This is how Arrow uses the term "indemnity" and how I learned it.)

service benefits -- insured gets service, insurance pays the cost. Blue Cross-type health insurance provides a service benefit.

In recent years, "indemnity" has come to be used by some to mean non-managed care. An "indemnity" plan pays for whatever service you choose to get and your provider chooses to provide. In this usage, the opposite of an "indemnity" policy is an HMO that determines what service you need and pays for it.

deductible -- patient pays the first $X of the cost.

coinsurance -- patient pays X% of the cost. Generally, there is a ceiling on how much the patient will have to spend in total.

copayment -- deductible or coinsurance

limits -- insurance will pay only so many dollars per incident (a concept also familiar from automobile liability insurance), or only for so many days in the hospital. E.g. years ago Blue Cross would only pay for the first 30 days of hospitalization. Before 1986, SC Medicaid only paid for 12 days hospitalization per year. Today it has a limit of how many prescriptions can be bought per month. SC state employees' health insurance plan has a limit of $1 million lifetime payment per insured.

Conservative economists argue that insurance with copayment is better for both individual and society than comprehensive 100%-pay insurance.

How can worse insurance be better? It can be, if these crucial assumptions are true:

  1. Premiums are set by perfect competition, so that your premium = your expected claims cost + expected administrative cost (which includes a normal profit as a cost of capital).
  2. The insurer has no market power as a buyer of health care services.
  3. Transactions cost for collecting copayments are negligible.
If the insurance company has market power, and can get discounts that you can't get as an individual, and if it passes some of the savings on to you in lower premiums, then insurance may be a winning gamble even if it's comprehensive. In this case the insurance company acts as a buyers' cartel.

Why insurance is a losing gamble, and why comprehensive insurance loses more

To understand this argument, you must already be OK with the concepts of risk, expected value, and risk aversion. Please try the economics tutorials on risk and risk aversion if you have not already done so.
The conservative argument is important to understand, so let us go over it. Three reasons why insurance must be a losing gamble, even if there is perfect competition among insurance companies, are:
  1. administrative cost
  2. moral hazard
  3. self-selection
1. Administrative cost. If there were no percentage favoring the "house," no insurance company could survive.

The existence of administrative cost in the premium makes "rational" consumers avoid buying insurance against three kinds of events:

a. high-probability events

If it's sure to happen then you're going to be paying for it anyway. If you make the insurance company pay for it, they'll charge you what you'd pay plus their administrative cost. Pay for it directly yourself and you save the administrative cost. (Assumes insurance companies have no buying power to force discounts individuals can't get.)

b. low-cost events

If you can afford to assume the risk yourself, why pay extra to somebody else? (Assumes that the insurance is tailor-made for you. If the premium is based on some group's experience and your actual risk is higher than your group average, the insurance may be a winning deal for you.)

c. ultra-low probability events

It costs some minimum amount to write a policy, no matter what the policy covers. Suppose it costs an insurance company $100 just to set up a new policy. Would you buy insurance against being struck by lightning this week? No. The insurance company's fixed per-policy administrative cost makes the risk premium too high. This explains why you ordinarily buy one health insurance policy that covers a lot of possibilities, rather than buying a separate policy for each individual bad thing that might happen.

2. Moral hazard further increases the cost of health insurance, making it a poorer gamble for the buyer. Important jargon term.

Moral hazard exists if having insurance increases the probability that the insured-against event will take place. E.g., a person buys life insurance and then contemplates suicide.

The fact that health care demand is not completely inelastic with respect to out-of-pocket payment implies that having health insurance increases the quantity of health care demanded = moral hazard.

Moral hazard increases as health insurance becomes more comprehensive. When you buy comprehensive first-dollar insurance, you must pay for other people's extra intensive use of medical services.

3. Selection

People often make insurance decisions based on differences between their personal situation and the insurance group they are in. Much depends on how premiums are set. The two basic methods are:

  1. Community rating = everyone in the community pays the same premium.
  2. Experience rating = people are classed into subgroups. Premium determined by the "experience" of the subgroup. This attempts to customize the premium so that it reflects each individual's probability of having a claim. Subgroups can be determined by age, sex, habits, medical history, etc. Underwriting is the insurance term for calculating risks and adjusting premiums accordingly.

Under competition, experience rating drives out community rating. Here's why: Suppose you start with a market in which the insurance companies are charging everyone the same premium. That is community rating. Competing insurance companies will go into such a market, identify the lower-risk individuals, and offer them a cheaper premium. That will leave the community-rating insurance companies with only the higher risk people, forcing them to raise their premiums. This in turn will encourage more competing insurance plans offer lower prices to the less risky of those people still with the community-rating plans. This process is self-feeding. Eventually, the community rating companies will be left with only the highest risks. They will have to charge the highest premium. At that point, everyone will have an experience-rated policy, in effect.

Self-selection happens when consumers choose among insurance plans according to their particular probabilities of having a claim. Self-selection tends to make comprehensive insurance more pricey. A simplified example shows why: Suppose there are two insurance policies available in your community or workplace. One -- a "bronze plan" -- has a big deductible and copayments. The other -- a "gold plan" -- has a lower deductible and copayments. The premiums for each are initially set according to the expected average utilization for your whole group. Now, some individuals in your group will realize that they are more likely than average to have medical expenses. They'll flock to the gold plan, hoping to minimize their out-of-pocket spending. The low-co-pay plan will consequently have more claims to pay, so it will have to raise its premium. This means that when you choose your insurance plan, even if you are highly risk averse and want a comprehensive policy, you'll find that the cost of the comprehensive plan is increasingly out of line, unless you're a high utilizer yourself.

Summarizing the Conservative Critique of Comprehensive Insurance

Administrative costs, moral hazard, and selection combine to imply that the more comprehensive an insurance policy is, the worse a deal it is in terms of expected value. Thus "rational" consumers will trade off risk aversion for premium savings, and presumably stop short of buying first-dollar coverage.

That is the standard conservative economic analysis of insurance. As pointed out above, the analysis assumes away two important reasons why comprehensive insurance might actually be better:

  1. If insurers have market power, so that they act as a buyer's cartel, they may be able to get a better price than an indivual buyer could get. The price discount could offset the administrative and moral hazard cost. Medicare, for example, is like a buyer's cartel for the elderly in the U.S. That is why pharmaceutical companies are so afraid of having Medicare pay for prescription drugs.
  2. Billing and collecting copayments itself involves administrative expense. In Canada, hospitals operate on lump-sum annual payments from the provinces, financially backed by the Canadian federal government. Patients pay nothing for hospital service, nor does private insurance. Hospitals need billing departments only to charge visitors from the U.S. or elsewhere. The administrative cost savings are immense.


The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of South Carolina.
E-mail: