If I had to rely on only two leading indicators of recessions, they would be initial unemployment claims and the yield curve (next in line would be housing). I talked about initial claims in the context of employment data in my last post. This post is about the yield curve.
An inversion of the yield curve has typically given a 12 month or better signal ahead of recessions:
Note also that it is the inversion that is important. The yield curve was fairly flat in the late-90's, a period of supercharged growth in the US economy. So when the Financial Times fueled the recession fears last week with this:
The US government bond market is blowing raspberries at the Federal Reserve. This could indicate trouble ahead for the American economy.
Last month, the Fed lifted interest rates for the first time in nine years, and short-term bond yields have duly climbed higher. But longer-term Treasury bonds have shrugged, with yields actually falling since the US central bank tightened monetary policy.
I was less concerned. In fact, I don't think the flattening yield curve should be any surprise as that is almost always the case after the Fed tightens policy:
The yield curve typically flattens to a 50bp spread between 10 and 2 year rates within a year of the initial Fed rate hike. Only the 1986 episode is unusual. Not only that, but the flattening begins immediately:
Even after the 1986 tightening the yield curve was flatter after the first 60 days.
Currently, the flattening of the yield curve - and the lack of any upward movement in 10 year yields at all - is consistent with my long-standing concern that the Federal Reserve's long-run projection of the federal funds rate - 3.5% as of December - is a pipe dream. Also why I was wary about the Fed's determination to raise rates. My preference was the Fed to wait until they were absolutely sure rates could be "normalized."
Optimally, my concerns will prove to be unwarranted. The economy may progress better than expected, productivity rises, the Fed pares down its stock of fixed income assets, the term premium rises, and the entire yield curve shifts up and the secular stagnation story dies. We are back in Kansas. No more flying monkeys. That is a perfectly acceptable, well-reasoned forecast and one I am sympathetic to, but I am not yet seeing it realized. What I am seeing at the moment is that the global pull of zero interest rates is sufficient to limit the ability of the Federal Reserve to "normalize" policy. We are stuck in Oz.
There is a school of thought that the yield curve is irrelevant now that we are near the zero bound. After all, you can't invert the yield curve very easily! And just look at Japan. Clearly the Japanese economy still experiences recession. If we are heading down the Japanese path, then I would expect longer term yield US yields to plunge below 1%. That is not my baseline, I don't think it is very likely, but I can't discount the possibility entirely.
Bottom Line: Don't discount the yield curve just yet. I think it is signaling something important about the limits of monetary policy "normalization." But it is also a signal that recession concerns are overblown. Even in a zero short rate world, the long end needs to plunge much deeper before the yield curve becomes a concern.