There has been much alarm about the sell-off in US bonds: since the start of the year two-year Treasury yields have risen from under 0.8 per cent to just under 1.5 per cent, and 10-year yields from under 1.6 per cent to almost 2 per cent. For equity investors, such alarm is misplaced. These are in fact welcome developments.
I say so for a simple reason. History tells us that the higher bond yields are relative to the Fed funds rate, the better equity returns tend to be thereafter. Since 1990 the correlation between the gap between two-year yields and the Fed funds rate and subsequent three-yearly changes in the S&P 500 has been 0.4. The 10-year/two-year gap and 10-year/fed funds gap have also both been significantly positively correlated with subsequent returns.
The effect here is big. Since 1990 each one percentage point greater gap between two-year yields and the Fed funds rate has been associated with 25 percentage points higher equity returns over the following three years.
By contrast, inflation has no predictive power for returns. Since 1990 the correlation between annual CPI inflation and S&P 500 returns in the following three years has been an insignificant minus 0.11. Higher inflation therefore tells us nothing about future equity returns but higher bond yields do.
This ability of bond yields to predict equity returns is especially strong if we look simply at whether the yield curve is upward-sloping (with long-dated yields above short-dated ones) or not. Since 1990 the S&P 500 has risen by an average of 36.1 per cent in the three years after two-year yields have been above the Fed funds rate, but by only 8 per cent after they have been below it. And it has risen by an average of 36.3 per cent in the three years after 10-year yields have been above two-year ones, but fallen by an average of 23.7 per cent after they have been below them. For example, ten-year yields fell below two-year ones in 2000 and 2006, and on both occasions equities fell sharply in the following three years. But the 2009-2012 upturn and the recent bull market both followed 10-year yields being above two-year ones.
History therefore tells us that the rise in two-year yields relative to the Fed funds rate is good news, as it predicts better equity returns. What is concerning is that two-year yields have also risen relative to 10-year ones. But with the latter still above the former, the outlook for equities is OK.
There’s a simple reason for this. Bond yields should be equal to the expected path of short-term interest rates: if yields are higher than the Fed funds rate it is a sign that markets expect the funds rate to rise, and if they are lower than the funds rate it is a sign they expect it to fall. Bond yields have risen recently because markets have revised up their expectations for short-term rates.
And there is wisdom in crowds: these expectations are often correct.
But what are the circumstances in which short-term rates rise? Ones in which the economy is doing well and shares are rising: the Fed would not raise rates if it were otherwise. By contrast, a world of falling rates is one of a weak economy and falling share prices.
It is for this reason that economist such as Arturo Estrella at Rensselaer Polytechnic Institute and John Williams at the New York Fed have found that the shape of the yield curve is a good predictor of recessions: inverted curves predict recessions and upward-sloping ones predict growth. Because of this, they also predict equity returns.
In fact, if you want to know the chances of recession you should look at the shape of the yield curve and at nothing else. As the IMF’s Prakash Loungani has shown, economists are terrible at predicting recessions.
Which is why the sell-off in bonds is a good thing from an equity investor’s point of view. It is a sign that interest rates are expected to rise. But this will only happen if the economy and stock market are strong.
You can be forgiven for thinking this is weird. Efficient market theory – and indeed common sense – tells us that obvious information such as the level of bond yields should be immediately embedded into share prices and so have no ability to predict them. But they clearly have had lots of ability to do so.
This could be because investors have in the past paid too much attention to the bad news of rising interest rates (a higher cost of capital) and not enough to the good – the fact that rates rise when the economy is strong.
Which brings us to a perennial danger with inferring the future from the past. It’s always possible that investors have learned from their past mistakes and in wising up have eliminated predictability. Now, in this case there is no evidence this has happened. Quite the opposite. The fact that equities have fallen during this bond sell-off is consistent with investors focussing upon the bad news of forthcoming rate rises and not the good. But we cannot rule out with complete confidence the possibility that the historic ability of yield curves to predict returns has disappeared.
What we can say, though, is that insofar as history is a guide the sell-off in bonds – and especially the rise in two-year yields relative to the Fed funds rate – augurs well for shares. What we have to fear is not that bond yields have risen, but that past relationships have ceased to be a guide to the future.
