How useful is the yield curve as a predictor in recent cycles?
The ‘yield curve’ has flattened significantly. This is worrying both investors and some Fed members given its track record of anticipating recessions. But has the yield curve become a less timely predictor in recent cycles?
So… what is this 'yield curve' everyone’s talking about?
- It’s the difference between long and short-term interest rates, e.g. the gap between 10-year and 2-year US government bond yields
Why are investors and some Federal Reserve (Fed) policy makers so worried about it?
- When short-term interest rates rise above long-term rates, 'inverting' the typically upward-sloping yield curve, it has historically foreshadowed a recession
Yikes! So… what is happening now?
- The gap has fallen from a peak of 2.5% in 2014 to just 0.25% now
- With the Fed expected to keep on hiking interest rates, many investors fear the spread will fall below zero soon, signalling a recession
Are things “different this time”?
- Arguably, yes: the ‘term premium’ has fallen
- In the 1970s and 80s, the yield on 10-year government bonds was higher than the central expectation of short-term interest rates over the next 10 years
- This is because investors needed a ‘risk premium’ when lending money long-term to compensate them for the danger of inflation getting out of control
- That ‘term premium’ has collapsed in recent decades. Deflation, not inflation, is the biggest concern. And there is a shortage of AAA rated ‘risk-free’ assets
- With long-term interest rates lower than before, it takes less of a rise in short-term interest rates to cause the gap to close
How can we get around this problem?
- Fed economists have derived a new measure of the yield curve looking only at shorter-term interest rate expectations
- If the market expects interest rates to rise, that’s good news. If they expect interest rates to fall, this is bad news
- They show that this measure of the yield curve is statistically better at predicting recessions than the ‘traditional’ one
What is the Fed’s adjusted yield curve showing?
- Investors expect the Fed to keep hiking interest rates over the next 18 months, consistent with a growing economy
Does the yield curve have magic powers? How do investor’s rate expectations cause a recession?
- Correlation is not necessarily causation
- If you think it’s going to rain, you’ll take an umbrella to work. But taking an umbrella to work does not cause it to rain. Instead you take an umbrella to work when it’s cloudy and you think it’s going to rain
- Similarly, if bond market investors expect interest rates to fall, that doesn’t cause a recession
- Instead, investors expect interest rates to fall when the economic outlook is cloudy – typically after the Fed has hiked rates aggressively, pushing short-term rates up relative to long-term rates
How reliable is the yield curve?
- The gap has closed before every recession, but the time lag between it closing and recession has lengthened, particularly in the mid-1990s and 2000s
Why has the yield curve become less reliable?
- It probably reflects deregulation of the banking sector, reducing the link between changes in monetary policy and credit availability
- In the 1960s and 1970s, there was a very close link between changes in short rates (and therefore the yield curve) and credit conditions
- This is because interest rates set by private banks were regulated
- But that relationship broke down after the banking sector was deregulated in the 1980s (see chart)
- Banks have since been allowed to set interest rates according to ‘market conditions’
- Credit conditions have been a more reliable predictor of when a recession will occur
What are credit conditions showing?
- In contrast to the flattening yield curve, credit conditions continue to ease
- This is similar to what we saw in the mid 1990s and 2000s, suggesting the expansion has further to run