Fixed income
23 Aug 2019 | By Huw Davies

The US yield curve – an alternative view

Article by Huw Davies on the yield curve and economic recessions.

  35 viewed

The relationship between an inverted Treasuries yield curve, an economic recession and the real Fed Funds rate may be more nuanced than expected.

Discussions in the market and the financial media have been rife over the last few weeks about the US yield curve’s move to an inverted shape (the yield on 2yr US Treasuries now being higher than 10yr Treasury yields) and the alarm bells this signal is potentially giving for an approaching recession in the US.

Historically, the correlation between the yield curve inverting and a recession arriving within 18-24 months has been very strong, with this signal seemingly predicting the last three recessions in the US economy:

US 2s/10s

The curve decisively inverted in February 2006 and a recession arrived 21 months later; inverted again in February 2000 and a recession arrived 12 months later; and inverted in January 1989 and a recession arrived 18 months later.

However, as any good statistician knows, there is a huge difference between identifying a correlation and also establishing a causation. Famously, correlations exist between the sales of ice cream and the observed incidence of drownings, murders and forest fires. Plainly the determining effect in this correlation is not the sale of ice cream but higher temperatures, so it’s important to analyse whether there is causation in any relationship rather than just an observed correlation.

The difficulty with the debate around the shape of the US curve and recessions is that the market has identified the correlation and jumped immediately to the conclusion that the shape of the curve causes the approaching recession. But exactly how that relationship works appears less than obvious and is also rarely discussed!

It could be argued that a flattening yield curve indicates that investors are becoming ever more risk averse and buying longer dated government bonds as a rush to safe-haven investments, yet in the approach to the curve inversions mentioned above, 10-year yields were rising ahead of these recessions, it was just that 2-year yields were rising even more rapidly.

There is also often a link to falling bank profitability from a flattening yield curve, but low profitability for developed economy banks has been with us for over a decade now and if anything has been improving recently improving.

My view is that, historically, the reason for the flattening yield curve was sharply rising front-end rates and particularly, the real Fed Funds rates. In fact ahead of the three recessions mentioned earlier, real Fed Funds rates were very highly correlated to 2-year/10-year (to show this more obviously, the 2-year/10-year line in this chart has been inverted):

Fed Funds versus US 2s/10s

Interestingly, between the mid 1980s and 2008, “2-10s” and real Fed Funds rates were 89.5% correlated; they were effectively telling the same story of rapidly rising front-end rates ahead of recessions. To me, the level of real front-end rates would seem to have a far greater causational effect on economic activity. Is a regular, medium-sized US company going to alter investment decisions based on real short-term rates, or on the spread between 2-year yields and 10-year yields? In my mind the more obvious causation would be short-term rates. I’m not sure many finance directors would even be aware of the 2-year/10-year spread, and they certainly would not let that metric drive investment decisions.

This brings us to the recent period – short-term real interest rates have risen and the yield curve has flattened aggressively. But ahead of the last three recessions, real Fed Funds rates peaked at +5.2%, +4.2% and +3.1% respectively. Their recent peak was +0.5% and they are currently 0% (actually +0.04%), and the correlation between 2-year and 10-year yields and the real Fed Funds rate has dropped to 43.8% since the global financial crisis.

The pertinent question appears to be, is the real Fed Funds rate at effectively 0% going to cause a recession? A lot has changed in the global economy but to my mind the usual sharp rise in company defaults seen during previous recessions is unlikely with rates so low across the curve. Recessions are usually caused by over-tightening of financial conditions. It is difficult to argue that we are seeing that scenario in this present cycle.

Like ice cream sales and forest fires, there may be a correlation but in my view the causation between this yield curve inversion and a recession has been significantly altered.