Loss Reserve Testing: A Report Year Approach

During the nineteenth century, the functions of the actuary were twofold: the calculation of premium rates and the setting of reserves. While reserves for life insurance may be set by actuarial formulae, casualty reserves are more frequently established by claims adjusters on an individual case basis. One of the tasks of the casualty actuary is to test the adequacy of reserves. Such tests are generally carried out on a statistical basis, reviewing a whole portfolio of case reserves at one time. As a result of the actuary’s test, it may be necessary to increase or decrease existing reserves, add special bulk reserves, or issue new instructions redirecting the claims adjusters in the setting of reserves. The significance of these steps for the financial solvency of the company cannot be overestimated. As Balcarek has clearly shown, changes in the degree of reserve adequacy have a very substantial impact on the earnings of casualty insurance companies. Thus, reserve tests should be, and usually are, the concern of senior management. The most familiar reserve tests are those incorporated in the annual statement. Currently, Schedule P is organized on an accident year basis and provides a means of performing a reserve test. However, it can be argued that Schedule P alone cannot be used to determine the amount of overstatement or understatement of the current reserves for a given line Skurnick includes more elaborate accident year tests in his survey of reserve calculations and, no doubt, the accident year approach is favored by many actuaries for reserve evaluation. This paper will discuss a reserve test which is based on an alternative scheme for organization of the data. Fundamental to this approach is the tabulation of claims (both reserves and payments) by report year. The latter is defined to be the year in which the claim was reported to the company, regardless of the accident date. In his survey, Skurnick mentions two methods of using report year data. The first is a projection, or loss development, approach dealing with total report year losses at various points of development. The second is a payment development method in which it is assumed that the percentage increase in the ultimate incurred value will be the same as the increase in average paid claim costs for claims of a similar age. In this paper, we will describe a third, more complex method utilizing report year data, and will then proceed to show how this approach is used in practice.