Principles of Insurance

Insurance is intended to provide financial protection against unexpected or untimely events. In particular, life and health insurance are purchased not in anticipation of imminent death or illness—although it is understood that death is inevitable and serious illness is fairly common. Rather, life insurance is obtained to protect dependents or business associates from the financial disadvantages that can occur in the event of unexpected death, and health insurance is meant to provide protection in the event of a significant financial loss associated with an unanticipated illness.

How does private insurance work? Basically, policyholders pay a relatively small, affordable amount into a common "pool," and the benefits of the pool are distributed to the unfortunate few who die (life insurance), become disabled (disability insurance), or develop illness (health insurance). In this way, the financial loss attendant to

these events can be mitigated even through the events themselves cannot be prevented.

But not all people are alike. The likelihood of occurrence and magnitude of loss varies across those differences. Some people apply for large amounts of insurance and others for small amounts. Some are young and others elderly. Occupations and avocations modify the likelihood of unexpected death or illness, as do health-enhancing activities such as exercise, proper diet, and not smoking. And some applicants are already in poor health or at known significant risk of developing poor health in the future. These different factors are evaluated by the insurance company through a process known as "risk selection and classification." The more common term for this process is "underwriting." Through underwriting, the insurance company determines the appropriate contribution to the risk pool by an individual policyholder.

The fundamental goal of the underwriting process is equity: policyholders with the same or similar expected risk of loss are charged the same. The higher the risk, the higher the premium. The lower the risk, the lower the premium. Note the distinction between equity and equality. With equity, premiums vary by risk; with equality, everyone—young or old, healthy or ill, and with or without associated factors that significantly increase the likelihood of making an early claim—would pay the same price.

During the underwriting process, risk classifications are created that recognize the many differences that exist among individuals in order to place applicants into groups with comparable expectations of longevity and health. Although the risk presented by any single individual cannot be determined with absolute precision, if people are assigned to groups with reasonable accuracy and the total number of insured people is large, then the estimate of the risk of the entire group of insured people is likely to be accurate.

Traditionally, characteristics important for risk classification have included factors such as age, gender, health history, physical condition, occupation, the use of alcohol and tobacco, family history, and serum cholesterol. These factors serve to identify individuals that have a greater or lesser likelihood of premature death or illness in the future. Because of this process, costs are held down for the great majority of insurance applicants since premiums more closely match the risks taken on by the insurance company.

How are rates determined that reflect the principle of equity? Under state laws and in the opinion of most observers, rates are considered equitable when they allow the insurer to earn enough income to pay claims and expenses and generate a reasonable margin for the risks they accept. In other words,

rates should be adequate but not excessive and should discriminate fairly between insureds. They should be adequate in order to provide insurers with sufficient income. They should not be excessive, for excessive rates impose undue burden on insureds. And they should discriminate fairly so that each insured will pay in accordance with the quality of his life.^[1]

The previous statement reflects the rate-setting philosophy of a private insurance company—not equal but equitable treatment of all. It recognizes differences among classes of insured persons, with products priced at a level that will result in a payment by each insured of an amount that is fair. Such fairness is accomplished by equating the anticipated cost to the company and the amount of the premium.

The vocabulary of insurance can be confusing. In the context of private insurance, discrimination is not necessarily bad and equality good. For example, in accordance with the insurance philosophy just set out, it would not be equitable to collect the same annual premium for the same life coverage from a sixty-year-old man in poor health as collected from a twenty-year-old woman in good health. To charge an equal premium would be inequitable. An insurer may—and must—discriminate to achieve equity, insofar as the discrimination remains fair. In fact, the statutes, regulations, and case laws that regulate the insurance industry compel discrimination; what they forbid is unfair discrimination.

Adverse selection, also known as antiselection, is a consideration that is of great importance to insurers. Adverse selection is a well-known phenomenon. It occurs when people with a greater likelihood of loss than what they are charged for continue insurance coverage to a greater extent than other people. It occurs when applicants withhold significant information from the insurer and/or choose amounts and types of insurance that are most beneficial to themselves. For example, someone with a history of heart disease is more likely to apply for insurance and/or apply for a greater amount of insurance cov-

erage than he would have otherwise done because he knows that he is likely to experience a claim in the foreseeable future. If he fails to mention this important information on his insurance application and the insurer does not otherwise become aware of it, the premium charged by the insurer will be insufficient to cover the risk involved. This premium deficit would be made up by the others in the pool who have paid their fair share. Adverse selection also occurs if the insurer is not permitted to obtain or use information that is pertinent to the risk being considered. In this example, the premiums charged would be insufficient to cover the risk involved if the insurer was not permitted to ask the proposed insured and his attending physician about the nature and severity of the heart disease or if this information could not be used in setting premium cost after it had been obtained.

What would happen if the insurance company were unaware of important, unfavorable information that was known to the applicant? In these instances, serious errors in risk classification would occur. Certain individuals would receive their insurance at unreasonably low cost. More claims would be filed than were expected, and if a significant number of these risk classification errors were made, the financial status of the entire insurance pool would be adversely affected.

But could premiums simply be increased across the board to cover the payment of these unanticipated benefits? Where permitted, an insurer could increase premiums to reflect these revised claim expectations. But this would encourage potential insurance applicants who are at lower risk to either buy from a different seller or exit the insurance market altogether. And for the individuals who had knowledge of their unfavorable risk status—individuals who had adversely selected against the insurance pool—further escalation of premiums would become necessary. More potential applicants would then decide not to apply for insurance.

Eventually, a point is reached in this upward spiral where the desired coverage becomes unavailable on any reasonable premium basis or the insurer becomes financially unsound. This "assessment spiral" is not merely a theoretical possibility. It actually occurred in some companies during the 1880s and 1890s because of poor risk classification practices. A more recent example of the effects of failing to classify risks properly is provided by the recent failure of a moderate-

sized casualty insurer located in Chicago.^[2] The company originally specialized in individual disability income policies. In the early 1970s, new management took over the company and decided to use its casualty authority to write auto insurance. They believed that people living in some of Chicago's neighborhoods were being charged auto premiums that were too high. Based on this belief, management ignored the actuarial statistics and evidence and wrote auto insurance for drivers in these neighborhoods at rates that would have been correct for a population with far fewer auto accidents. As a result, the company failed and everyone involved was hurt financially. All the company's lines of business were affected, including its disability income line. Many disabled individuals who had long depended on income payments lost those benefits.

The current risk classification system permits private insurers throughout the world to respond fairly to valid cost and experience-related differences among persons. To help guide actuaries in developing this system, the actuarial profession through the Actuarial Standards Board has adopted a risk classification standard of practice. This standard enumerates three basic requirements for an appropriate risk classification system. First, risk classification must be fair. Second, it must permit economic incentives to operate and thus encourage widespread availability of coverage in the marketplace. Finally, risk classification must do its part to keep the insurer solvent. To achieve these ends, a sound classification system should be based on at least four principles:

1. Risk classifications should reflect cost and experience differences. For example, employers of coal miners would pay more for their unemployment insurance than employers of computer technicians because coal miners historically have much higher rates of unemployment.

2. The system should be applied objectively and consistently. By this principle, for example, males of the same age with similar health histories would be charged similar rates for life insurance.

3. The system should be practical, cost effective, and responsive to change. This means that there would be limits on how much effort and money could be spent to classify a given risk and that risk

classification systems must be dynamic. For instance, when polio was eliminated as a public health hazard, the system changed to reflect that development.

4. Adverse selection should be minimized. As noted earlier, sound risk classification systems limit the ability of an applicant to take an unfair financial advantage at the expense of the insurance company or other policyholders already insured by the company.