Net Interest Rates: History and Measurement

Johnson and Gelles (1996) observed that starting in about 1980 the interest rates on U.S. Treasury securities rose sharply, and argued that because of the increase, forensic economists should not use low interest rates based on earlier experience for calculation of present values. Interest rates did indeed rise sharply. But by 1996 interest rates were already falling; rates have continued to fall so that the status quo ante has been essentially restored. This bit of forensic economic history would be of little interest except that Johnson and Gelles are still cited to help justify net interest rates more appropriate to the 1980s than to 2015. This note updates their work, using their same sources and method; it is intended in part to show that the high interest rates described in their paper are not now relevant. The note continues, to discuss how to measure net interest rates. There are two issues: (1) The net rate for an n-year note should use the n-year rate of wage increase; alternatively it should compare the 1-year rate of return on the n-year note (with capital gains from resale) with the 1-year rate of wage increase. (2) Interest rates are measured daily, and wages monthly. Does the use of annual averages change the results? Answer: the overall picture given by alternative definitions is not very sensitive to the definition of the net interest rate, except that when 3- or 10-year notes are held for just one year and re-sold to give a total annual return that includes interest and capital gains or losses, volatility of returns is considerably higher than when the securities are held to maturity; the result suggests that the return on Treasury notes for one year with resale is not appropriate for discounting earnings losses to present value.