Yield curve inversion happens when short-rates rise above longer ones. Inversion happened in the U.S. in March of this year and has broadly continued since. This is a bad sign. The Federal Reserve (Fed) has expressed some concern about an inverted yield curve. For one thing, it can discourage banks from longer term lending, and projects that need long-term finance can boost growth. There are other reasons to worry too. Fortunately, the Fed has an important weapon in its toolkit, it can control the short-end of the yield curve with reasonable accuracy. So can the Fed simply end inversion by cutting rates? In theory, yes. Though the futures market suggests whether or not we see such an outcome in 2019 is finely balanced.
Cutting Rates Are Only Half The Story
The main aim of the Fed for 2019 so far has been to move from the prospect of raising cuts to a neutral stance, to enacting the first rate cut in over a decade. That should be good for avoiding yield curve inversion, right? If short-term rates are above long-term rates, then cutting short-term rates could solve the problem directly. As you can see from the graph below, it hasn't worked. In fact, longer rates are moving down slightly faster than the Fed is cutting. Therefore the yield curve is becoming slightly more inverted in recent weeks despite July's cut.
Since inversion the yield curve model from Estrella and Mishkin suggest that we're looking at a probability of a U.S. recession within 12 months of around 30%. Now, that's actually not quite as worrying at is may sound, the U.S. economy hits a recession around 1 in every 5 years historically, so we might consider a 20% chance of recession 1 year out to be relatively normal regardless of what the yield curve is telling us.
Nonetheless, if the yield curve becomes more inverted then those recession probabilities may rise steeply. For example, should the 10-year dips 1% below 3-month rates for an extended period, then the odds of recession rise to 50% and if the gap widens to 2% then a near-term recession becomes highly likely on their model. So with 3-month rates at around 2% today, look for the 10-year to fall below 1% and then to turn negative for the recession signal to materially increase.
Can the Fed stop inversion? It's not clear, current market expectations expect more cuts. Somewhere between one and three further cuts this year based on current interest rate futures. However, that's not much in absolute terms. At the low-end of that range, rates are still above 1.25% at the close of 2019. If that happened and the 10-year remained at its current level of around 1.7% then yes, inversion would end. However, it's going to take another at least two cuts and flat 10-year yields to get there. Furthermore, the way the yield curve signal works, history matters just as much as the future, the fact that the yield curve has inverted is what generates the warning signal. A bounce in rates in future doesn't undo it. Therefore, the Fed may not be able to act fast enough to eliminate the signal that inversion creates.
