Description: This paper deals with the selection of appropriate interest and inflation assumptions, which is one of the most important considerations in the valuation of a defined benefit pension plan. The paper begins by describing a simplified model in which the only important consideration is the selection of an appropriate differential between interest and inflation assumptions. The discussion then considers some situations that might lead the actuary to attempt to select a realistic level of interest and inflation assumptions, in addition to an appropriate differential. The author's conclusion is that in some situations the absolute level of the interest and inflation assumptions can be all but meaningless, and the differential between the two assumptions is the important factor. In other situations the use of this technique will produce serious inaccuracies in the valuation results. As always, the actuary must analyze each situation in order to determine which of the available techniques will produce the best results with the least effort, given the degree of accuracy required. From Transactions of Society of Actuaries 1976, Vol. 28.

Description: This paper describes the actuarial valuation method for pension plans, the formulas and assumptions used, and the way these assumptions might be chosen. A case study is presented. Followed by several discussion papers submitted

Description: This paper examines tandem levels of salary and interest rates and finds that the absolute level of each assumption as well as the differential are significant cost determinants for defined benefit pension plans. In forming best-estimate salary and interest rate assumptions, the actuary must reconcile the difference, if any, between the currently observed rates and the valuation assumptions. Common practice is to choose expected long-term rather than observed rates for valuation purposes. The error in this procedure is analyzed and found to be tolerable for strictly temporary discrepancies between observed rates and valuation assumptions. For situations where these discrepancies are expected to persist for some time, a need is shown for nonuniform, calendar-year assumptions or their equivalent in order to develop best-estimate assumptions. Followed by several discussion papers submitted