The Near-Term Forward Yield Spread as a Leading Indicator: A Less Distorted Mirror (Summary)

Long-term yield spread—notably, the yields on 10-year government bonds versus yields on 2-year bonds—is commonly seen as an indicator of impending recession. In particular, when the yield curve inverts and short-term yields are higher than long-term yields, fears of a recession intensify.

But is the long-term spread really the best predictor of recessions?

The authors consider whether near-term forward spreads provide more accurate forecasts of a coming recession and, by extension, of future GDP growth and stock market returns.

The authors define a near-term forward spread as the difference between the implied interest rate on a three-month T-bill six quarters ahead and the current three-month T-bill rate. They argue that this near-term forward spread measures the market’s expectations for near-term monetary policy and contend that forward rates should provide clearer signals of recessions than the classic 10-year minus 2-year spread because yields are an average of forward rates and thus blur the signals.

The authors compare near-term forward spreads across a 45-year period (1972–2019) against other common investment industry practices for forecasting recessions—namely,

• The 10-year minus 2-year spread

• the 10-year yield minus the 1-quarter bill rate

• The 10-year to 6-quarter forward spread, which is the complement to the near-term forward spread

They also test whether the model can outperform the sum of human capital in the investment industry by comparing it with the Survey of Professional Forecasters (SPF). The SPF invites respondents to predict the probability that the US economy will enter recession in each of the following four quarters.

The authors also consider how near-term forward spreads stack up against rival models in terms of predicting GDP growth and stock market returns.

The near-term forward spread is found to have more predictive power than the other variables—namely, GDP growth and stock returns—for forecasting recessions.

When near-term forward spreads price in monetary policy easing over the subsequent six quarters, a recession usually follows. The model is considerably more accurate than the 10-year minus 2-year spread. In particular, in the lead-up to financial crisis, the near-term forward spreads forecast a substantially higher probability of recession than the conventional long-term spreads.

Relative to the SPF, the model’s yield curve is better at predicting recession risk, although the SPF does add additional forecasting value when combined with the near-term forward spread.

The near-term forward spread also more accurately forecasts GDP growth over the following four quarters, outperforming the SPF consensus. Traditional long-term spreads are found to be poor indicators of GDP growth.

The near-term forward spread does not predict market returns well but has substantial power for predicting stock market downturns, particularly when combined with earnings yield, which implies a strategy for market timers.

Investors might want to rethink a reliance on the long-term yield curve when making macro forecasts. If near-term forward spreads are included in forecasting, yields on bonds maturing further out than six quarters have little additional value in forecasting recessions, GDP growth, or stock returns.

This finding is particularly relevant in an environment such as the one at the start of 2019, where the traditional yield curve is inverted and market participants believe a recession is on the way.

The results lend support to a simple market-timing strategy—untested in this paper—for equity investors. That is, in any given quarter, the investor should remain fully invested in equities if the near-term forward spread remained positive during the preceding several quarters.