Historical Comparisons between Various Interest Rates and Growth Rates in the CPI, MCPI, Average Weekly Earnings and Total Compensation in the Employer Cost Index
Ireland, Thomas R., Journal of Legal Economics
Forensic economists who calculate damages in personal injury and wrongful death disagree about appropriate discount rates to use for reducing future values to present value and about growth rates that should be used to project increases in wages, fringe benefits, the price level and the medical component of the price level. Those disagreements take two forms. First, there is debate over which interest rates to consider. Some economists advocate use of instruments of very short maturities only, such as the 9 1 day U.S. Treasury bill rate or the six month U.S. Treasury bill rate. Some advocate use of one year or three year notes or 10 year or 30 year U.S. Treasury bonds. Some advocate use of municipal Aaa rates because of their tax-free features. Some advocate use of rates on Treasury inflation indexed securities and some advocate blended rates. Likewise, there is some disagreement about whether to measure changes in earnings by use of the average weekly earnings series or the employer cost index.
The second form of disagreement focuses on the historical periods that should be considered when choosing both a discount rate and an earnings growth rate. This area of disagreement can be further divided into two issues, The first issue is whether the discount rate and earnings and price growth rates should be treated disparately or in the same manner. Some forensic economists argue that growth rates in earnings and prices should be derived from past data, but the discount rate should be based on current market data. Others argue that if past growth rates in earnings and prices are used, discount rates from the same period should also be used. The second issue is what the appropriate period should be if one is using past data for either growth rates or growth rates in earnings and prices. Should it be five years, ten years, twenty years, or some other period? There is indication of the scope of differences that exist in the 1999 survey of members of the National Association of Forensic Economics (Brookshire and Slesnick 1999).
Those issues notwithstanding, the purpose of this paper is not to argue for or against particular interest rates or historical periods (though this author has expressed his views on these matters elsewhere). Rather, the purpose is to provide broad based data to facilitate whatever historical comparisons any given researcher might wish to make between commonly used growth rates and discount rates. It is this writer's hope that the tables provided in this paper, particularly tables 7 and 8, will enable good forensic economists to quickly disprove factually inaccurate calculations they may confront when looking at the reports of other forensic economists. The rate comparisons provided in this paper include the 91 day Treasury bill rate, the 3 year Treasury note rate, the 10 year Treasury bond rate, the 30 year Treasury bond rate, the muni Aaa rate, the corporate Aaa rate compared with rates of growth in the consumer price index (CPI), the medical component of the consumer price index (MCPI), average weekly earnings, and the total compensation series in the employer cost index (ECI).
For convenience, all data in this paper are taken from the Economic Report of the President: 2000 (Council of Economic Advisors 2000). That convenience imposed limits on the length of the historical periods that could be considered. Table B-71 provides annual values for most of the interest rates back to 1953, allowing for 47 years of comparisons. The only exception is the 30 year U.S. Treasury bond, which only began to be issued in 1977, allowing for 23 years of comparisons. Rates of increase in the CPI and MCPI are taken from table B-3. Data in that table goes back to 1940, but only figures after 1953 were used, based on availability of discount rates from table B-71. Data for the Average Weekly Earnings series in table B-45 starts in 1959, which imposes a 4 1 year limit. Data for the ECI total compensation series is reported only after 1980, imposing a 20 year limit. …