Clouds in the Crystal Ball: Has the Yield Curve Lost its Predictive Power?
April 2006
Between late February and early March last quarter, the spread between two-year Treasury notes and ten-year Treasury notes reversed from -16 basis points to +6 basis points as market participants rejected the idea that the curve should invert more deeply. As a result, the inversion of the benchmark yield curve lasted just over one month; this was a very brief occurrence in a historical context (Figure 1). Inversions in the Greenspan era generally lasted seven to thirteen months and peaked closer to 50 basis points. Thus, history suggests we may be in for a deeper and more prolonged inversion than we have experienced so far in the current cycle.
The good news is that inversions, by their nature, are temporary events. The normal slope of the yield curve is positive because of the risk of unexpected, value-erasing events taking place as time passes—the longer the time period, the greater the risk. Yield curves typically invert when monetary policy is viewed as too tight, and there is an expectation that the fed funds rate will be reduced. Once the Fed starts to ease monetary policy, the curve starts to normalize and eventually regains a positive slope.
The bad news is that during this current cycle, the flattening of the yield curve has occurred for atypical reasons, which makes historical comparisons tenuous. In the past, yield-curve inversions have been the result of a restrictive monetary policy, which is why they have exhibited a high correlation with economic downturns (Figure 2). Monetary policy today is not tight by historical measures, and futures contracts do not reflect an expectation for a substantial easing of monetary policy. In our opinion, the curve is currently flat because of a collapse in bond-risk premiums, which is apparently the result of an excessive amount of capital investment (demand) in assets with duration.
Surprisingly, investors have been quite willing to purchase bonds with longer durations, despite getting little or no additional compensation for taking on unexpected risks. One explanation for the decline in the risk premiums is the belief that bond investors are no longer worried about future inflation eroding their coupon; the market believes that the Fed can and will prevent inflation from rising above a nominal level. Thus, the risk has shifted from the back of the curve to the front of the curve, which is dominated by Fed activity. While this theory is credible, inflation risk is not the sole driver of risk, and market participants are still subject to other risks. In our view, the reason that risk premiums have been removed is because there is a lot of competition for fixed income bonds with long maturities (duration). This is a function of excess global savings, particularly in Asia, and also of the ease with which capital can now flow across borders. Too much money chasing too few bonds does not mean that capital is being invested wisely, but it does imply that long-term interest rates could remain low as long as these conditions persist.
Investors should therefore be very cautious of basing their forecasts on historic precedents; the old rules might not apply here. Instead, the investment landscape should be evaluated with the understanding that new forces are at work, some of which are likely to persist well beyond the current economic cycle.
Investors should note that the yield curve does not currently possess the predictive powers it once had. Those who believe that the current term structure of interest rates implies a strong likelihood of rate cuts by the Fed in the next twelve months might be disappointed. Also, trying to predict the demand for dollar assets coming from foreign central banks and other overseas investors is not a viable investment strategy. These flows are difficult to predict, and any attempt to form investment policy around such predictions would be highly speculative.
Perhaps the most prudent strategy is to go back to basics. Fundamental risk-return analysis, not speculation, should remain the basis for rational fixed income investors. The global demand for yield has squeezed risk premiums across the fixed income universe to extremely low levels, leaving very little margin for error. In addition to low inflation risk, credit spreads in the corporate bond market are near their tightest levels in years, while option-adjusted spreads and implied volatility in the mortgage markets are also near historic lows.
Despite what we believe to be a favorable economic outlook, the fixed income markets currently provide minimal reward for the assumption of risk. This asymmetric risk-reward profile argues strongly in favor of high-quality assets until the balance tips in the other direction.