What the yield curve tells us

The US yield curve is looking very interesting indeed. In normal times, yield curves slope upwards: investors are compensated with higher returns when they lock away money over the long term. This is because of the risks they incur – after all, 10 years is a long time in economics or anything else for that matter. 

But recently, the US Treasury yield curve has ‘inverted’, with the gap between the 10-year and two-year yields turning negative, as the chart shows. In other words, the yields investors earn on short-term Treasuries exceeds the return on long-term ones. And as well as being unusual, this is troubling.

Yield curve inversions can be driven by changes at the short end and long end of the curve. And the past month has seen movement at both ends. At the short end, yields are sensitive to interest rate expectations. With US inflation at 9.1 per cent, more rate hikes are expected to follow. This makes locking in rates with a two-year note less desirable, reducing demand. And as prices of two-year notes fall, yields are pushed up. 

Longer-term yields, on the other hand, tend to be driven by expectations about economic performance. If market participants expect a downturn, they bet that the central bank will later be forced to cut rates to stimulate the economy. This decreases expected borrowing costs in the future, and depresses the 10-year Treasury yield.

Yield curves are also influenced by market attitudes to risk. As concerns about recession escalate, investors look for a safe haven: Treasuries can provide this by offering refuge from choppy equities markets. Higher demand for 10-year notes increases prices, further depressing yields.

No surprise then, that inverted yield curves are seen as a signal of an impending recession. And a seemingly reliable one, too: according to research from the Federal Reserve Bank of San Francisco (FRBSF), yield curve inversions have preceded every US recession since 1955. On this side of the pond, there is evidence that the inverted yield curve’s power to anticipate a recession dates back as far as the 1800s. 

So the track record is pretty impressive. But how useful is it really as a recession predictor? Firstly, heed the maxim that ‘correlation doesn’t imply causation’. The evidence doesn’t imply that an inverted yield curve causes a recession. Rather, its slope fluctuates to reflect changing expectations about the economy – and it is these expectations that prove useful predictors of economic downturns.

If the yield curve is a useful recession indicator, it is also worth remembering that there are many others, too. Oren Klachkin, lead economist at Oxford Economics, said last week that "upbeat spending backed by a strong labour market and as consumers dip into their savings should keep the economy out of recession". The yield curve may well signal a downturn, but other indicators suggest a more optimistic economic outlook. 

What’s more, Fed economists Eric C Engstrom and Steven A Sharpe assert that the 2-10 spread offers a ‘particularly muddled view’ of impending recession. Instead, they argue that yields on Treasury debt with maturities of less than two years have better predictive power, by mirroring market participants' expectations for economic performance and interest rate policy. 

We also tend to be rather generous when ascribing predictive power to inverted yield curves too. The FRBSF found that between 1955 and 2018, an inverted yield curve did correctly signal all nine US recessions. But it also generated one false positive and couldn’t tell us much about when the recession would hit. In fact, it found that a yield curve inversion could precede a recession by anything from six to 24 months. 

And a lot can happen in that time. After all, the curve briefly inverted in 2019, and the US economy did enter a recession in 2020. But as Engstrom and Sharpe wryly note, "pure statistical analysis will give credit, where credit is not due" to the inverted yield curve for predicting the Covid recession.