Insurance is the pooling of money by a company from a group of people or organizations, to pay for the fortuitous losses that any of them may suffer. The money that the people pay to the insurance company is called the premium, and for this premium, the company promises to indemnify any of its customers for covered losses. The company may also provide risk management services.

Another way to look at it is that insurance involves the pooling of losses among its customers. In return for losing a little money in the form of a premium, the customers are indemnified against a large loss. Thus, the average loss among a group of people is substituted for actual losses.

To illustrate, consider a sample of 1,000 farms, and suppose that each farm were identical, and had the same value of $300,000 for the house and barn. Over the years, it was discovered that about 1 farmer would have a total loss every year. If a farmer's buildings burned completely, the farmer would be out of $300,000, but if each farmer paid into a fund every year to pay for the fires, then each would only have to pay 300,000/1,000 = $300 per year. Thus, for $300 per year, each farmer can avoid a loss of $300,000 if a fire destroys both house and barn. The Amish, a religious sect, do something similar, but instead of pooling money, they pool labor. If a farmer's barn burns down, for instance, then other members of the community pitch in and rebuild the barn. Thus, in exchange for providing some labor periodically, each farmer is protected against major losses.

Insurance is the transfer of risk to an insurance company, which pools the losses of many people to provide indemnification for any who suffer covered losses. An insurance company also invests the premiums, since otherwise it would lose the opportunity cost of the money, usually in the form of interest, while it holds it. For a private insurance company to be a going concern, it must earn enough money from premiums and investments to pay not only for the losses of those covered, but also to pay to operate the company and to earn a profit. However, for any private insurance company to be able to insure a risk, the company's payout must be fairly predictable and the premiums charged must be affordable.

Note that some transfers of risk do not depend on the pooling of losses. Futures and derivatives are financial instruments that transfer the risk of one party to that of another party, referred to as the counterparty. Hence, if a wheat farmer wants to ensure that he receive a minimum payment for his wheat, then he can sell a futures contract before the wheat is even planted. If the price of wheat declines by the delivery date, then the farmer loses on the sale of the wheat, but gains on the futures contract. If the price of wheat increases, then the farmer loses on the futures contract, but gains on the sale of wheat. Either way, he is guaranteed to receive a minimum payment, but only at the opportunity cost of not making a higher profit if the price of wheat exceeds the market price that was expected when he agreed to the futures contract.

Paying an insurance premium may seem akin to gambling, since a small payment is made for a possible large payment in the future. However, insurance differs from gambling in that the premium covers losses from a pure risk, whereas gambling is paying a small wager in the hope of earning a large profit in the future. The risks that insurance premiums cover would exist whether the insured had insurance or not. On the other hand, there is no risk before a wager is placed. However, when placed, a speculative risk is created in the amount of the wager.

Insurable Risks

Because private insurance companies are businesses that want to make a profit, there are only certain risks—known as insurable risks—that private insurers are willing to cover. Almost all risks insured by insurance companies are pure risks, which are risks where there is no possibility of profit. Additionally, since insurable losses can only be compensated by the payment of money, only risks involving financial loss are insurable. Insurance companies also do not insure against predictable losses, such as wear and tear, and inherent vice, which is the self-destruction of property caused by the nature of the thing itself, such as the rotting of fruit.

So that an insurance company can remain a viable concern, any risks that it chooses to cover must have a predictable payout and an affordable premium that will cover the costs of the payout and the insurance company's operating expenses, and still yield a reasonable profit.

Affordable Premiums

Several requirements must be satisfied for an insurance company to be able to cover a risk. First, the premium for the insurance must be affordable; otherwise few people would buy it, thereby spreading the risk among fewer people. With fewer people insured, the objective risk of the pool would be higher, because actual losses would be less predictable, more variable. Therefore, higher premiums would have to be charged to cover the increased objective risk.

Secondly, the premium must be considerably less than the policy coverage; otherwise, people would simply self-insure.

Losses must be accidental and unintentional. If they were not, then a moral hazard would be created, where many people would cause losses to collect insurance. This would raise premiums to unaffordable levels, and make payouts by the company unpredictable, since the losses would not occur by chance. People would stop paying premiums if they knew that the premiums were used to enrich dishonest people. For instance, one reason why insurance companies rarely cover mental illness is because it is easily feigned, and thus, it would be difficult to prevent dishonesty.

Another requirement to keep premiums affordable is that the loss must be determinable and measurable. The loss must be determinable because an insurance company will only pay for a loss if the loss was for a covered reason. The loss must be measurable to know how much to pay, because an insurance company will only pay to cover the actual loss, and no more. If the insurance company paid more than the actual loss, then the insured could make a profit by causing partial losses, then collecting the entire amount from the insurance company, thereby violating the principle of indemnity. This would not only increase premiums because of the larger payout, but it would also create a moral hazard by motivating some people to cause losses to profit from insurance, raising premiums even more.

Predictable Payouts

Another basic requirement for a private insurance company to insure a risk is that the total payout would have to be fairly predictable in a large sample of the population. Although there would be some variation in total losses from year to year, these losses should not deviate too much from the mean over time. To be predictable, both the frequency and the severity of losses should be calculable. Since these calculations rely on the law of large numbers, there would have to be a large number of exposure units, the basic items that are insured. Without accurate forecasting, an economical premium cannot be calculated.

The law of large numbers is based on the observation that the greater the number of events that have a given or observed probability, the less the observed frequency will deviate from the expected frequency. As a simple example, consider the tossing of a fair coin. The chance, in any 1 toss, that it will come up heads or tails is 50/50, but if the coin were tossed 10 times, there is a small possibility that all 10 flips will yield heads. Even if it is not all heads or all tails, there is a very good chance that the observed frequency will deviate significantly from the expected 50/50 chance. However, if the coin were flipped 1 million times, then heads and tails will be more nearly equal.

Because most catastrophic losses, such as floods, earthquakes, and war, occur erratically, and usually with great destruction, both the frequency of occurrence and the severity of losses cannot be accurately forecasted, and, thus, catastrophic losses are usually not insured by private companies. Moreover, catastrophes frequently cause losses for many members of a group, thereby negating the benefit of pooling losses, since many members of an insured group may suffer extensive losses at the same time.

Insurable Types of Risk

There are generally 3 types of risk that can be covered by insurance: personal risk, property risk, and liability risk.

  1. Personal risk is any risk that can affect the health or safety of an individual, such as being injured by an accident or suffering from an illness.
  2. Property risk is any risk that can cause a partial or total loss to property, such as theft, fire, or so-called "acts of God".
  3. Liability risk is the personal or business risk associated with being found liable to another because of negligence or willful acts that caused a loss to another's property or person, such as injuring someone while driving under the influence of alcohol, or because the insured failed to perform a duty, such as performing contractual obligations.

Most types of insurance cover all 3 types of risk to some extent, depending on the type of insurance. Auto insurance, for instance, can cover damage to the insured's motor vehicle, personal injuries to the insured or any passengers in the vehicle, and the cost of liability that arises from the negligence of the insured because of damage to other people's property or because others were injured or killed.

Underwriting and Adverse Selection

Insurance companies, like most companies, have competition. The better that an insurance company can calculate risks, the lower the premiums that it can charge and still make a profit. One way to do this is by underwriting, which is the selection and classification of insurance applicants according to the probable payout for that class. Actuaries set the insurance rates, while the underwriter decides which class the insurance applicant belongs.

For instance, it is observed that some people have more auto accidents than others. By classifying people according to the observed risk of accident, a lower premium can be charged to safe drivers and higher premiums to reckless drivers. If the same premium was charged for both types of drivers, then the premium charged would be higher for the safe drivers and lower for the reckless drivers than it would otherwise be. In other words, the safe drivers would be partially subsidizing the reckless drivers. In a competitive industry, subsidization will be eliminated because if 1 insurance company were charging the same premium for all drivers, then another insurance company would start selling cheaper insurance only to the safer drivers. Then the safe drivers would choose the cheaper insurance, and the company that was charging the same rate would lose its source of subsidy, and would be forced to charge a higher premium to the reckless drivers. Subsidization can only exist if the law forces insurance companies to charge the same premium to different classes with a different risk profile. Such is the case with sex discrimination. Although young men have more accidents than young women, the insurance companies cannot use sex as a factor in calculating premiums—thus, young women partially subsidize the premiums for young men.

Insurance companies have devised various methods for more accurately rating the risk profile of certain classes of people. Thus, factors used to determine premiums for automobile insurance include the number of points on a driver's record and the number of accidents caused by the driver in the past.

Another controversial factor for underwriting is the use of insurance scores, which are based on credit scores, because there is a large correlation between people with low scores and the number of claims that they file. This, too, is limited by law in some states.

Many of the factors that are indicative of someone's risk are determined from the information provided by the applicant in the insurance application. Since many high risk people know that providing truthful information will increase their premiums, they provide false information to get a lower premium. If the insurance company provides the coverage for a lower premium, then the company, because of adverse selection, will incur more losses than it expected. Adverse selection results from the tendency of some people with high risk profiles to provide false information to get standard premiums.

To prevent adverse selection, insurance companies verify the information, especially if a claim is filed. If material information known to the insured at the time of the application was false, then the insurance company does not have to pay the claim. Various clauses in its contracts also provide specific conditions under which a claim will not be paid, such as the presence of pre-existing conditions for a health insurance applicant. For instance, if the claimant files a claim for a health condition that the claimant would have known about when he applied for the insurance, then the company will not have to pay the claim. (Update: Obamacare has made it illegal to base health insurance premiums on pre-existing conditions.) Similarly, an insurance company will not pay if a claimant commits suicide shortly after buying life insurance.


Generally, retained losses are small losses that can easily be paid by the individual or organization suffering the loss. However, for large organizations and companies, losses that would be catastrophic to individuals are small losses for the organization. Hence, self-insurance is simply a form of retained risk. Additionally, some types of losses, such as illnesses and accidents that require healthcare, are fairly predictable among a large number of workers. In these cases, many large organizations decide to self-insure. Some have argued that because there is no transfer of risk to a third-party or a pooling of losses, self-insurance is a misnomer. However, there is some pooling of losses. Many large businesses, for instance, self-insure for healthcare and workers' compensation. The employees accept lower pay in exchange for the payment of healthcare, if an illness or accident should strike them. Because only some of the employees will incur large healthcare costs in any given year, the lower pay accepted by all the workers is used to pay for the larger costs of healthcare for the few. However, many of these organizations hire insurance companies to manage their self-insured plans.

Economic Benefits of Insurance

Insurance does not prevent losses. Indeed, because of moral hazard, morale hazard, and fraud, losses in a society are actually increased because of insurance. However, insurance does provide some economic benefits over just covering financial losses. It alleviates some anxiety about losses to both individuals and organizations by limiting losses to a certain, known amount, thereby allowing better planning. Additionally, it allows individuals and organizations to use capital for projects that may yield a higher return, thereby increasing the allocative efficiency of the economy. For instance, without insurance, a business would have to save much more money to cover possible losses, thereby lowering the amount available to invest in its business, especially for riskier, but potentially more profitable, projects.