Income And The Interest Rate Game by Gene Tannuzzo, Columbia Threadneedle Investments

  • An investment strategy based purely on guessing where interest rates are headed is destined to miss the mark.
  • We still expect the Fed to raise rates soon (but probably not in October) and we think longer term rates should drift higher (but we wouldn’t want to bet the farm on that).
  • Investors should focus on areas of the bond market that offer attractive income relative to the risk, and retain the flexibility to adjust that positioning as the market evolves.

This week, the Federal Reserve will convene for another policy planning session. Once again, investors and the media will actively debate the merits and probabilities of raising short-term rates and what it means for the bond market and the world as a whole. Data-minded economists will estimate how much rate increases will affect the level of GDP and the value of the dollar. Meanwhile, yield-starved retirees, pension funds and 401k plans will yearn for the day they see higher rates of income. And so begins another playing of the interest rate game in which investors and observers submit myriad prognostications on where the 10-year Treasury yield is headed. This raises an important question: what is the value of an interest rate forecast?

To explore this question, we used information from the Survey of Professional Forecasters (SPF) published by the Federal Reserve Bank of Philadelphia. The SPF is the oldest quarterly survey of macroeconomic forecasts in the United States. In the most recent survey conducted in the third quarter of this year, 42 forecasters submitted their estimates for a number of variables including growth, unemployment, inflation, and interest rates. In order to better understand the value of interest rate forecasts, we compiled the historical median forecasts for the yield of the 10-year Treasury note four quarters ahead. Next, we compared these forecasts to the actual Treasury yield one year after the survey date (i.e. the number the forecasters were trying to predict). The comparison of these numbers contains some fascinating results (Exhibit 1).

Exhibit 1: Median forecast error of 10-year Treasury yields 12 months ahead

Interest Rate

Source: Federal Reserve Bank of Philadelphia, Columbia Threadneedle Investments, as of September 30, 2015.

Since 1993, the median forecaster was wrong about next year’s interest rates by 0.95% on average. That’s a pretty wide margin considering the absolute level of yields in the Treasury market. Second, the average forecaster predicted the wrong direction of interest rate changes 53% of the time. So more often than not, a professional forecaster would have pointed you in the wrong direction with regards to marginal changes in rates. Lastly, the median forecaster would have improved the accuracy of his or her forecast by 15 basis points (from 0.95% error to 0.80% error) by simply guessing the current market yield at the time of the survey rather than trying estimate a change of any magnitude in either direction.

The bottom line of this simple analysis is that investors should put little weight on the value of a single point forecast for Treasury yields, and an investment strategy based purely on guessing where rates are headed is destined to miss the mark. However, this does not alter the fundamental challenge that investors still need income. While we may not know with certainty where the 10-year Treasury yield is headed, we do know that it only yields about 2%. In contrast, we do find more attractive yields available in the marketplace that can help investors meet their income needs. For example, investment-grade telecom companies yielding 5% seem attractive to us, even in an environment of slow global growth. In the non-agency MBS market, 4% yields seem reasonable on assets that continue to be insulated from global influences such as currency volatility and lower commodity prices. In the high-yield market, we remain fairly cautious overall as the energy, metals, and mining sectors are likely to drag overall default rates higher over the next two years. However, excluding these industries, yields are above 6.5% on average and look compelling for a number of companies.

As for the Fed meeting, we still expect the Fed to raise rates soon, but doubt it will be in October. Meanwhile, we think longer term rates should drift higher, but we wouldn’t want to bet the farm on that. Instead, we think investors should focus on areas of the bond market that offer attractive income relative to the risk, and retain the flexibility to adjust that positioning as the market evolves.