Something odd has happened in the US bond market: real and nominal yields have moved in opposite directions. Since September the yield on conventional five-year Treasury bonds has risen from 0.3 to 0.8 per cent, while the yield on their inflation-proofed counterparts, despite a rise in the last few days, has dropped from minus 1.3 to minus 1.5 per cent.
You might think this is normal; when investors expect higher inflation they naturally dump conventional bonds and buy protection against inflation thus driving down the yields on inflation-proofed bonds. But in fact, this rarely happens. Yields on real and nominal bonds have usually risen and fallen together – more often falling in recent years. Since 2003 the correlation between the two for five-year maturities has been a whopping 0.89.
This isn’t a quirk of the US bond market. Here in the UK, index-linked and conventional yields have also been highly correlated.
And, in fact, there’s a simple reason for this. Normally, if investors expect inflation to rise they would expect not just a rise in nominal interest rates but a rise in real rates, too – that is, a bigger increase in nominal rates than in inflation. This is because if central banks are to reduce inflation they must reduce demand, which requires them to raise real interest rates. And expectations of high real interest rates should mean high real bond yields. And so real and nominal yields normally rise and fall together.
Which explains why the last few months have been different. The Fed won’t raise interest rates in response to higher inflation for some time. It has repeatedly said that it “expects to maintain an accommodative stance of monetary policy” until inflation has been “moderately above 2 per cent for some time”. Futures markets interpret this to mean that the fed funds rate won’t rise until at least late 2022. With the market expecting inflation to rise, this means it expects real interest rates to fall – hence the drop in real yields. But because higher inflation erodes the real value of nominal bonds, investors have dumped these.
But there’s a second reason for the divergence between real and nominal yields. It’s that inflation risk has increased. President Biden’s proposed $1.9 trillion fiscal stimulus reduces the chance of deflation but – to some minds – increases the risk of significant inflation. Demand for inflation-proofed bonds has increased as investors have sought insurance against this (perhaps small) risk.
There is, however, another interpretation of the drop in real yields. It’s a sign that, despite the fiscal stimulus, the market expects long-run secular stagnation to continue. In theory, long-term real rates should be high when real growth is high and low when growth is low. Negative real rates are therefore a sign that the market – the dispersed wisdom of crowds – expects continued weak growth. And not just for the next five years: despite a rise in the last few days, 10-year real yields are still far below their long-term average level.
All this matters for UK investors. For one thing, there has historically been a huge correlation between UK and US yields, both real and nominal. Whatever happens to US bonds is likely to happen to gilts – which is why nominal yields here have risen more than real yields.
Also, there’s a big impact of all this on gold. Because the metal pays no interest it becomes less attractive as bond yields rise. This is because the higher are bond yields the more income you are sacrificing when you own gold rather than bonds. For this reason, gold usually falls when bond yields rise.
But which yield matters most in driving gold’s price: the real or nominal yield? For a long time it’s been hard to tell simply because real and nominal yields have moved together. The last few months, though, have allowed us to find out. And the fact that gold has fallen 10 per cent since September tells us that it is rising nominal yields that are bad for the metal.
The big danger for investors is that both real and nominal yields could rise even more if investors were to expect the Fed to raise rates sooner than they currently anticipate: we saw such a move last week.
Personally, I suspect this danger is overstated, because inflation probably won’t rise far or quickly. One reason for thinking so is that the ratio of employment to population is still low by historic standards, which is a sign that there’s plenty of spare capacity in the economy. But my opinion doesn’t matter. At low yields, even small rises inflict heavy losses on bondholders: that’s the maths of bond duration. And these losses would probably be accompanied by losses on gold, too. They might even see losses on equities as well, if investors fear that higher yields will trigger a reversal of the 'reach for yield' that has seen investors buy shares out of despair on getting returns on other assets.
It used to be trivially easy to earn good returns on diversified portfolios. But this might not remain the case. Which is why there’s a case for us all to hold some cash.
