The yield curve: a crystal ball for the next recession?

Ask an investor what they would most like to know in advance and the timing of the next recession might be high up their list. The correlation between the ups and downs of the stock market and the ebb and flow of the economy is not exact but there is clearly a link. Recessions are bad news for investors. Seeing them coming would be very useful.

One of the crystal balls that investors use to identify turning points in the economy is found in the bond market. Specifically, in the ever-shifting relationship between the yields offered by long and short-dated government bonds. Put together on one chart, these trace out what is known as the yield curve. And it’s the shape of that curve that interests us.

There’s a debate about how useful the yield curve really is when it comes to foretelling the future, but even more sceptical observers would admit that it has had a pretty good track record. Afficionados claim that it has predicted every recession in the past 50 years. So, what is the yield curve, what determines its shape and what is it telling us today?

The yield curve measures the gap between the yields on different maturities of government bonds. Most of the time, this curve slopes upwards, which is another way of saying that the yields on longer-dated bonds, say those maturing in ten or thirty years’ time, are usually higher than those on shorter-dated issues – a two or five-year bond, for example.

The reason for this is that an investor requires more encouragement to lend money to the government for a longer period of time. Given the greater risk that inflation will eat into the value of their money, they quite rightly demand a higher income, or yield, as compensation. Because there is less uncertainty with a shorter-dated bond, they will accept a lower yield.

The relationship between these different yields fluctuates. Sometimes the normal, upward sloping yield curve flattens out, and more rarely still it turns downwards. It becomes inverted, in the jargon. That’s because an investor who is relaxed about the outlook for long term inflation, or worried about growth, will settle for a lower yield on longer bonds. And one of the key reasons why they might expect lower growth and inflation in future is because a central bank is taking steps to slow the economy today.

The principal way this is done is by raising short term interest rates. And because short-term bonds are more influenced by interest rates than by the growth and inflation expectations which drive longer bonds, rising interest rates will tend to push the yields of shorter maturity bonds up more than those of longer ones.

The yield curve, in other words, will tend to flatten when central banks are starting to raise rates. And that, of course, is precisely what is happening today. Indeed there is growing speculation that the Federal Reserve (the Fed), in particular, is about to ratchet up the pace of tightening, both in terms of rate hikes and by gradually pushing back into the market the US$9trn of bonds and mortgage-backed securities it has accumulated in recent years.

Having signalled its intent with a quarter point rate hike last month, the Fed is now expected to switch to a series of half percentage point hikes throughout the rest of the year. This might not sound like much of a difference but if it were to double the rate of increase next month it would be the first time it had done so in 20 years. The Fed watchers at Citi think it might deliver a half point rise at the next four meetings and comments this week from Lael Brainard, the soon to be vice-chair of the Fed, suggest they are right.

Having been stuck at zero since the start of the pandemic, US interest rates are now forecast to rise rapidly to 3pc or more over the next year or so. And this is why the yield curve has flattened and indeed, briefly last week, inverted. The yield on 10-year bonds actually fell below that of 2-year bonds. Cue gnashing of teeth and talk that recession is on its way.

But is it? The trouble with the yield curve’s recession indicator is that it is a notoriously imprecise tool. The lag between the yield curve inverting and growth turning negative can be anything between six months and three years. Once or twice it even delivers a false positive. When it briefly turned down in 2019 the signal was only confirmed by the arrival of the Covid pandemic. No-one in the bond market was predicting that, I don’t believe.

Another reason investors are questioning the message is the unprecedented impact of massive central bank bond purchases since the financial crisis and more so during the pandemic. It seems implausible that the weight of this money has not distorted signals to some degree. Nor that the reversal of the quantitative easing programme that’s just about to start won’t skew it in the opposite direction.

A third reason to be sceptical is that different relationships are telling us different things right now. While five-year bonds briefly yielded more than 30-year issues last week, and two-years exceeded the tens, another pair that’s starting to be watched closely, three-month bills versus ten-year bonds, is still upward sloping.

The equity market also seems relatively unfazed by the bond market’s negative message. Shares have largely shrugged off the triple worries of inflation, rising rates and war in Ukraine. Perhaps investors are remembering that the period between the inversion of the yield curve and the eventual arrival of recession has in the past been a rewarding final hurrah for the stock market.