July 28, 2012 1 Comment
In keeping with my earlier post explaining what money actually was, I am going to use this post to explain insurance.
Insurance is simply the transfer of risk from one party (the insured) to another (the insurer) for a payment. The payment is called the premium. This premium is equal to the expected loss during the insured period plus some additional amount to pay administrative and operational costs of the insurer and build available reserves. The process of accessing the risk, determining the price of the risk, and deciding if to take on the risk exposure is called underwriting. When a loss occurs, the insured makes a claim and the insurer then indemnifies, or makes whole, the insured.
For insurance to “work” there need to exist insurability. There are several characteristics which will exist if a risk in insurable. If they do not exist then the risk must be transferred either through a compulsory insurance program such as workers compensation or social insurance such as social security disability insurance or through other techniques which I will address in future posts. I will use an example for each of these with the insurance company insuring $200,000 homes which have a 1% chance of a total loss due to fire and a 99% chance of no damage.
- Definite loss: The loss has to have a clear cause that took place at a particular time and place. In our example the cause of the loss would be the fire which occurs at the home’s location and we could clearly have a time of the event.
- Accidental loss: The loss has to be fortuitous to the beneficiary. That simply means that the loss needs to be outside of the insured control. In our example this would mean for instance that the fire was not arson by the beneficiary or at the beneficiary’s request.
- Calculable loss: The loss needs to be both of a calculable chance of occurence and of an estimated size. In our example we have a 1% chance of a loss and that loss is of $200,000.
- Large loss: The loss has to be economically significant. If the loss potential is too small then it is not worth the effort to underwrite the policy and administer it. In our example we have a total loss so this is not a concern because it is clearly economically significant.
- Affordable premium: The premium for that loss calculated above cannot be so great that the insured will not pay (for the risk of occurence is too high as an example) or so great that the insured is not at risk (because the cost is too high for the risk insured). We do not state a premium in our example but an actuarialy fair premium would be $2,000 for a year. The premium would be higher than this because of loading. This loading is how insurance companies pay for the administrative costs and generate an underwriting profit.
- Large number of similar insurable interests: This means that the insurer needs to insure a pool of homes that are similar in their risk exposure. This is because as the number of units grow we can use something called the “law of large numbers” to assume a normal distribution of losses. With one home the expected loss is $2,000 and the standard deviation of the loss is almost $20,000. If this were a pool of 10,000 homes, the expected loss per unit is $2,000 but now the standard deviation is just over $2,000.
- Limited risk of catastrophe: The individual trials should be independent. This means that there would not be a giant conflagration which burns a large number of the pool. Catastrophic losses defeats the advantage of pooling.
Role of Insurance
Insurance is one of the techniques to manage a risk. As the list above shows, insurance works well when you have rare events which result in large losses and those losses are independent of each other. If the size or frequency do not fit this pattern, then risk management is still important but other tools must be used instead.
In my next post I will look at the example of the health insurance industry in the United States and discuss if it is appropriate to consider it to be an example of insurance in our description above.