Unlikely Estimates of the Ex Ante Real Interest Rate: Another Dismal Performance from the Dismal Science1

The ex ante real rate of interest is one of the most important concepts in economics and finance. Because the universally-used Fisher theory of interest requires positive ex ante real interest rates, empirical estimates of the ex ante real interest rate derived from the Fisher theory of interest should also be positive. Unfortunately, virtually all estimates of the ex ante real interest rate published in economic journals and textbooks or used in macroeconomic models and policy discussions for the past 35 years contain negative values for extended time periods and, thus, are theoretically flawed. Moreover, the procedures generally used to estimate ex ante real interest rates were shown to produce biased estimates of the ex ante real rate over 30 years ago. In this article, we document this puzzling chasm between the Fisherian theory that mandates positive ex ante real interest rates and the practice of macroeconomists who generate and use ex ante real interest rate estimates that violate this theory. We explore the reasons that this problem exists and assess some alternative approaches for estimating the ex ante real interest rate to determine whether they might resolve this problem.