In previous business cycles, an inversion of the yield curve as a precursor to recession was a niche issue.
Yes, the spread between the federal funds rate and the yield on the 10-year Treasury note was added to the index of leading economic indicators in 1996, where it maintains its starring role: first among equals. Even so, economists, both inside and outside the Federal Reserve, were always willing to overlook an inverted term spread — “This time is different!” — when the economy was looking hunky-dory.
It’s not obscure anymore. The subject, and the spread’s predictive power, have been discussed ad nauseum. The question remains: With knowledge aforethought, will the Fed elect to push its benchmark overnight rate above the yield on a long-term Treasury security?
The story so far: Starting in December 2015 and proceeding ever-so-slowly at first, the Fed has raised the funds rate from a range of 0%-0.25% to 1.75%-2%. In that same time frame, the yield on the 10-year Treasury note has moved up from about 2.25% to 2.85%.
The term premium — I use that phrase because “yield curve” means different things to different people — has been less than 100 basis points several times over the past year. There is nothing inherently contractionary about a 95 basis-point spread between banks’ marginal cost of funds (the overnight rate) and a long-term risk-free rate, a proxy for the return on assets.
With the economy charging ahead and the unemployment rate hitting an 18-year low of 3.8% in May, the Fed is going to confront the decision in the not-too-distant future of whether to willingly invert the curve or to reconsider its effort to normalize the funds rate.
At the Fed’s June meeting, the median forecast for the year-end funds rate was 2.25%-2.50%, implying two more 25 basis-point increases. The so-called “dot plot” projects three more increases in 2019, taking the funds rate to 3%-3.25%.
The fed funds futures market remains unconvinced that the Fed will stick to its agenda. And long-term Treasuries appear to be dancing to the beat of a different drummer, breaching and rejecting 3% in May.
This divergence between Fed and market expectations presents yet another “conundrum” for policy makers. (See Greenspan, Alan, Feb. 16, 2005.) If long rates refuse to rise — if they refuse to incorporate expectations for higher short-term rates and rising inflation, to reflect strong aggregate demand or to insist on a bigger risk premium — the Fed will be confronted with a choice of willingly inverting the term spread or standing down.
There are already an array of assertions as to why this time is different, some of which are outlined, and dismissed, in a San Francisco Fed Economic Letter by Michael D. Bauer and Thomas M. Mertens in March: “Economic Forecasts with the Yield Curve.” Some are retreads; others are new. Together they should be the “usual suspects” offered up if and when the spread inverts.
1. Interest rates are historically low to begin with, so increases in the funds rate won’t slow the economy as much as in previous cycles.
If this sounds familiar, it should. It is reminiscent of the argument made before the 2007-2009 recession about long-term interest rates, which were low and couldn’t possibly constrain economic output, even if the spread inverted.
2. The term premium, or the extra compensation that investors demand for holding a 10-year security compared to a short-term one, has been close to zero, with different implications for the spread and its impact on the economy.
While extrapolating the term premium in long-term interest rates has provided a new pastime for econometricians, Bauer and Mertens found no evidence that the reason rates are low — expectations or risk premia — makes any difference.
3. The neutral rate of interest has declined.
So what? (See No. 1 above.)
4. The Fed’s asset purchases have artificially depressed long-term interest rates, distorting the term spread and its implications for economic activity.
As a general rule, the forces driving inversion are irrelevant.
What matters is the difference between an artificially pegged short-term rate and a market-determined long-term rate, not their absolute levels.
Besides, if the Fed were really interested in limiting any yield-curve distortion as a result of its long-term asset purchases, it could focus on tightening policy by selling more long-term securities instead of raising the funds rate.
How or why the spread inverts (the cause) does not soften or negate the effect. The slope of the yield curve acts as an incentive (when it is steep) or disincentive (when it is flat or inverted) for credit creation, as banks generally borrow short and lend long.
While there may be reasons to think this time is different — no obvious ones come to mind — the Fed should heed the results and conclusion of Bauer’s and Mertens’ research.
“While the current environment is somewhat special — with low interest rates and risk premiums — the power of the term spread to predict economic slowdowns appears intact,” Bauer and Mertens write. “An extensive analysis of various models leads us to conclude that the term spread is by far the most reliable predictor of recessions, and its predictive power is largely unaffected by including additional variables.”
To answer the question posed in the headline: Will the Fed invert the curve? My best guess is no, with a caveat: My confidence level in my forecast is low.
