We define risk as a condition of the real world in which there is an exposure to adversity.
First that in this definition risk is a condition of the real world; it is a combination of circumstances in the external environment. Also that in this combination of circumstances, there is a possibility of loss. When we say that an event is possible, we mean that it has a probability between zero and one; it is neither impossible nor definite. Also that there is no requirement that the possibility be measurable; only that it must exist. We may or may not be able to measure the degree of risk, but the probability of the adverse outcome must be between zero and one.
The undesirable event is described as "an adverse deviation from a desired outcome that is expected or hoped for." The reference to a desired outcome that is either expected or hoped for contemplates both individual and aggregate loss exposures. The individual hopes that adversity will not occur, and it is the possibility that this hope will not be met that constitutes risk. If you own a house, you hope that it will not catch fire. When you make a wager, you hope that the outcome will be favourable. The fact that the outcome in either event may be something other than what you hope constitutes the possibility of loss or risk.
In the case of an insurer, actuaries predict some specified number and amount of losses and change a premium based on expectation. The amount of predicted losses is the desired outcome that is expected by the insurer. For the insurer, risk is the possibility that losses will deviate adversely from what is expected.
Because the term uncertainty is often used in connection with the term risk (sometimes even interchangeably), it seems appropriate to explain the relationship between the terms risk and uncertainty.
The most widely held meaning of uncertainty refers to a state of mind characterised by doubt, based on a lack of knowledge about what will or will not happen in the future. It is the opposite of certainty, which is a conviction or certitude about a particular situation. A person says "I am certain," which means the same as "I am positive." Both statements reflect a conviction about the outcome. Uncertainty, on the other hand, is the opposite mental state. If one says "I am uncertain", the statement reflects the lack of knowledge about the outcome. Uncertainty, then, is simply a psychological reaction to the absence of knowledge about the future. The existence of risk - a condition or combination of circumstances in which there is a possibility of loss - creates uncertainty on the part of individuals when that risk is recognised.
The individual's conviction or lack thereof (certainty or uncertainty) about a specific fact or situation may or may not coincide with the conditions of the real world. The person who says "I am certain" may actually get another outcome. Uncertainty varies with the knowledge and attitudes of the person. Different attitudes are possible for different individuals under identical conditions of the real world. It is possible, for example, for a person to experience uncertainty in a situation in which he or she imagines that there is a chance of loss, but where no chance of loss exists. Similarly, it is possible for an individual to feel no uncertainty regarding a particular risk when the exposure to loss is not recognised. Whether or not a risk is recognised, however, does not alter its existence. When there is a possibility of loss, risk exists whether or not the person exposed to loss is aware of the risk.
It is intuitively obvious that there are some situations in which the risk is greater than in other situations. Just as we should agree on what we mean when we use the term risk, we should agree on the way(s) in which risk can be measured. Precisely what is meant when we say that one alternative involves "more risk" or "less risk" than another?
It would seem that the most commonly accepted meaning of degree of risk is related to the likelihood of occurrence. We intuitively consider those events with a high probability of loss to be "riskier" than those with a low probability. This intuitive notion of the degree of risk is consistent with our definition of risk. When risk is defined as the possibility of an adverse deviation from a desired outcome that is expected or hoped for, the degree of risk is measured by the probability of the adverse deviation. In the case of the individual, the hope is that no loss will occur, so that the probability of a deviation from what is hoped for (which is the measure of risk) varies directly with the probability that a loss will occur. In the case of individual, we measure risk in terms of the probability of an adverse deviation from what is hoped for. Actuarial tables tell us, for example, that the probability of death at age 52 is approximately 1 percent and that at age 79 it is about 10 percent. At age 97, the probability of death increases to nearly 50 percent. Using the probability of an adverse deviation from the outcome that is hoped for, we view the risk of death at age 79 as greater than that at age 52, but less than that at age 97. The higher the probability that an event will occur, the greater is the likelihood of a deviation from the outcome that is hoped for and the greater the risk, as long as the probability of loss is less than one.
In the game of Russian roulette, there is more risk when there are two bullets in a revolver's six chambers than when there is one bullet. Adding a third bullet increases the risk, as does adding the fourth bullet and fifth bullet. Adding the fourth and fifth bullets increases the probability of a deviation from a desired outcome. If a sixth bullet is added, the player can no longer expect or even hope that the outcome will be favourable. The sixth bullet makes the outcome certain, and risk no longer exists. If the probability of loss is 1, there is no chance of an outcome other than that which is expected and, therefore, no hope of a favourable result. Similarly when the probability of loss is zero, there is no possibility of loss and therefore no risk.
In the case of a large number of exposure units, estimates can be made about the likelihood that a given number of losses will occur, and predictions can be made on the basis of these estimates. Here the expectation is that the predicted number of losses will occur. In the case of aggregate exposures, the degree of risk is not the probability of a single occurrence or loss; it is the probability of some outcome different from that predicted or expected. Insurance companies make predictions about losses that are expected to occur and charge a premium based on this prediction. For the insurance company, then, the risk is that its prediction will not be accurate. Suppose that based on past experience, an insurer estimates that 1 out of 1000 houses will burn. If the company insures 100000 houses, it might predict that 100 houses will burn out of the 100000 insured, but it is highly unlikely that 100, and only 100, houses will burn. The actual experience will undoubtedly deviate from the expectation, and insofar as this deviation is unfavourable, the insurance company faces risk. Therefore, the insurance company makes a prediction not only with respect to the number of houses that will burn, but also estimates the range of error. The prediction might be that 100 losses will occur and that the range of possible deviation will be plus or minus 10. Some number of houses between 90 and 110 are expected to burn, and the possibility that the number will be more than 100 is the insurer's risk. When standard deviation is used, risk is measurable, and we can say that more risk or less risk exists in a given situation, depending on the standard deviation.
At times we use the term more risk and less risk to indicate a measure of the possible size of the loss. Many people would say that there is more risk involved in a possible loss of $1000 than in that of $1, even though the probability of loss is the same in both cases. The probability that a loss may occur and the potential severity of the loss if it does occur contribute to the intensity of one's reaction to risk. It seems, therefore, that a measurement of risk should recognise the magnitude of the potential loss. Given two situations, one involving a $1000 exposure and the other a $1 exposure, and assuming the same probability in each case, it seems appropriate to state that there is a greater risk in the case of the possible loss of $1000. This is consistent with our definition of risk, since the loss of $1000 is a greater deviation from what is hoped for (that is, no loss) than is the loss of $1. On the other hand, given two situations where the amount exposed is the same (for example, $1000); there is more risk in the situation with the greater probability of loss.
While it may be difficult to relate the size of the potential loss and the probability of that loss in the measurement of risk, the concept of expected value may be used to relate these two facets of a given risk situation. The expected value of a loss in a given situation is the probability of that loss multiplied by the amount of the potential loss. If the amount at risk is $10 and the probability of loss is 0.10, the expected value of the loss is $1. If the amount at risk is $100 and the probability is 0.01, the expected value is also $1. This is very useful concept, as we shall see later.
It is not uncommon for the terms peril and hazard to be used interchangeably with each other and with risk. However, to be precise, it is important to distinguish these terms. A peril is a cause of a loss. We speak of the peril of fire, or windstorm, or hail, or theft. Each of these is the cause of the loss that occurs. A hazard, on the other hand, is a condition that may create or increase the chance of a loss arising from a given peril. It is possible for something to be both a peril and a hazard. For example, sickness is a peril causing economic loss, but it is also a hazard that increases the chance of loss from the peril of premature death. Hazards are normally classified into three categories:
Physical hazards consist of those physical properties that increase the chance of loss from the various perils. Examples of physical hazards that increase the possibility of loss from the peril of fire are the type of construction, the location of the property, and the occupancy of the building.
Moral hazard refers to the increase in the probability of loss that result from dishonest tendencies in the character of the insured person. More simply, it is the dishonest tendencies on the part of an insured that may induce that person to attempt to defraud the insurance company. A dishonest person, in the hope of collecting from the insurance company, may intentionally cause a loss or may exaggerate the amount of a loss in an attempt to collect more than the amount to which he or she is entitled. Fraud is a significant problem for insurance companies and increase cost of insurance.
Morale hazard , not to be confused with moral hazard, acts to increase losses where insurance exists, not necessarily because of dishonesty, but because of a different attitude towards losses that will be paid by insurance. When people have purchased insurance, they may have a more careless attitude towards preventing losses or may have different attitude towards the cost of restoring damage. Morale hazard is also reflected in the attitude of persons who are not insureds. The tendency of physicians to provide more expensive levels of care when costs are covered by insurance is a part of the morale hazard. Similarly, the inclination of injuries to make larger awards when the loss is covered insurance is another example of morale hazard. In short, morale hazard acts to increase both the frequency and severity of losses when such losses are covered by insurance.
Regardless of the manner in which risk is defined, the greatest burden in connection with risk is that some losses will actually occur. When a house is destroyed by fire, or money is stolen, or a wage earner dies, there is a financial loss. When someone is negligent and that negligence results in injury to a person or damage to property, there is a financial loss. These losses are the primary burden of risk and the primary reason that individuals attempt to avoid risk or alleviate its impact.
In addition to the losses themselves, there are other detrimental aspects of risk. The uncertainty as to whether the loss will occur requires the prudent individual to prepare for its possible occurrence. In the absence of insurance, one way this can be done is to accumulate a reserve fund to meet the losses if they do occur. Accumulation of such a reserve fund carries an opportunity cost, for funds must be available at the time of the loss and must therefore be held in a highly liquid state. The return on such funds will presumably be less than if they were put to alternative uses. If each property owner accumulates his or her own fund, the amount of funds held in such reserves will be greater than if the funds are amassed collectively.
Furthermore, the existence of risk may have a deterrent effect on economic growth and capital accumulation. Progress in the economy is determined to a large extent by the rate of capital accumulation, but the investment of capital involves risk that is distasteful. Investors as a class will incur the risks of a new undertaking only if the return on the investment is sufficiently high to compensate for both the dynamic and static risks. The cost of capital is higher in those situations where the risk is greater, and the consumer must pay the resulting higher cost of the goods and services or they will not be forthcoming.
Finally, the uncertainty connected with risk usually produces a feeling of frustration and mental unrest. This is particularly true in the case of pure risk. Speculative risk is attractive to many individuals. The gambler obviously enjoys the uncertainty connected with wagering more than the certainty of not gambling - otherwise he or she would not gamble. But here it is the possibility of gain or profit, which exists only in the speculative risk category that is attractive. In the case of pure risk, where there is no compensating chance of gain, risk is distasteful. Most people hope that misfortunes will not befall them and that their present state of well-being will continue. While they hope that no misfortune will occur, people are nevertheless likely to worry about possible mishaps. This worry, which induces a feeling of diminished well-being, is an additional burden of risk.
Although it is theoretically possible to insure all possibilities of loss, some are not insurable at a reasonable price. For practical reasons, insurers are not willing to accept all the risks that others may wish to transfer to them. To be considered a proper subject for insurance, certain characteristics should be present. The four prerequisites listed next represent the "ideal" elements of an insurable risk. Although it is desirable that the risk have these characteristics, it is possible for certain risks that do not have them to be insured.
There must be a sufficiently large number of homogeneous exposure units to make the losses reasonably predictable. Insurance is based on the operation of the law of large numbers. A large number of exposure units enhance the operation of an insurance plan by making estimates of future losses more accurate.
The loss produced by the risk must be definite and measurable. It must be a type of loss that is relatively difficult to counterfeit, and it must be capable of financial measurement. In other words, we must be able to tell when a loss has taken place, and we must be able to set some value to the extent of it
The loss must be fortuitous or accidental. The loss must be the result of a contingency; that is, it must be something that may or may not happen. It must not be something that is certain to happen. If the insurance company knows that an event in the future is inevitable, it also knows that it must collect a premium equal to the certain loss that it must pay, plus an additional amount for the expenses of administering the claim and operation. Depreciation, which is a certainty, cannot be insured; it is provided for through a sinking fund. Furthermore, the loss should be beyond the control of the insured. The law of large numbers is useful in making predictions only if we can reasonably assume that future occurrences will approximate past experience. Since we assume that past experience was a result of chance happening, the predictions concerning the future will be valid only if future happenings are also a result of chance.
The loss must not be catastrophic. It must be unlikely to produce loss to a very large percentage of the exposure units at the same time. The insurance principle is based on a notion of sharing losses, and inherent in this idea is the assumption that only a small percentage of the group will suffer the loss at any one time. Damage that results from enemy attack would be catastrophic in nature. There are additional perils, such as floods, which, while they would not affect everyone in society, would affect only those who had purchased insurance.
