Finance Highlights 2006 0503 - Laserfiche WebLink

Home

INTEREST RATE FORECASTING OR RELATIVE VALUE ANALYSIS: WHAT WORKS BETTER FOR THE INVESTOR? Some fixed income portfolio managers forecast interest rates as part of their investment strategy, and others employ a more duration neutral approach (duration is kept in a close range around the index) while focusing on relative value among securities and sectors. Because of the divergence between these two distinct approaches one has to ask, is time better spent predicting interest rates or, rather, controlling interest rate risk? To find out we reviewed current research which concludes that it's difficult to achieve consistent returns through interest rate forecasting. Initially, we wanted to find out how successful professional economists are at forecasting future interest rate changes. Studies published by the Federal Reserve Bank of St. Louis', economists at North Carolina State University', and the Journal of Applied Statistics' indicate that most professional interest rate forecasters fail to outperform a random walk approach. In other words, despite years of experience and access to the most sophisticated tools available, the professional economists' forecasts are less accurate than those obtained byflipping a coin. A closer look reveals that forecasters' results are even worse than they initially appear. Even the forecaster with the best previous track record failed to consistently outperform a coin flip when attempting to predict interest rate changes. "Success in predicting future interest rates depends upon predicting both future changes in the information and the market's reaction to such news."' In order to be successful, portfolio managers must predict both future economic output and the manner in which other market participants will interpret this data. They must then forecast not only the direction of subsequent interest rate moves, but also the magnitude and the timing of the move. This is difficult to do correctly even once, and virtually impossible to accomplish consistently over a long period of time. An additional problem with forecasting interest rate changes is that different incentives exist for investors and professional forecasters. The professional forecaster is heavily rewarded for short term success, while for the investor, short term gains are insufficient for generating superior long term risk-adjusted returns. This problem is compounded by the fact that predicting interest rate changes becomes more difficult as the time horizon increases. "Studies have shown that error statistics often double in size when the forecast horizon is extended as little as from one to two quarters ahead."' The risk in trying to time interest rates is that once a forecaster makes a mistake, it becomes extremely difficult to recapture that lost performance. If investors do not want to assume this risk, are they then restricted to average returns? We do not think so. There are effective alternatives that do not depend upon interest rate predictions. By employing yield curve analysis, diligent security selection, and rigorous risk control, we believe a portfolio can consistently outperform its benchmark while maintaining an appropriate portfolio risk profile. This eliminates the possibility that an investor will underperform the benchmark due to erroneous interest rate forecasts. -Brian Perry 1"Predicting Interest Rates., A comparison of Professional and Market - Based Forecasts", Michael T Belongia, Federal Reserve Bank of St. Louis 2 "Professional Forecasts of Interest Rates and Exchange Rates! Evidence from the Wall Street Journal's Panel of Economists", Karlyn Mitchell & Douglas K. Pearce, North Carolina State University 3" Combination Forecasting for Directional Accuracy, an Application to Survey Interest Rate Forecasts", Mark R. Greer, Dowling College Page 2 0 2005. Chandler Asset Management Inc, A Registered Investment Adviser.