Dude, you got your causality totally backwards. Yield curve inverts because everyone expects recession and the resulting easing of the monetary policy. A recession is a normal part of the business cycle. Everyone wants to go to the party, but nobody wants to stay and clean up. I posted a link to the 3m/10y spread above, it has the recessions shaded in grey. The only one that was a complete false signal was the inversion in 1998 during the EM turmoils. Most of the time it took a while to materialize (or to report, depending how you want to explain it) - from May 1989 to June 1990 for the recession of the 90s, July 2000 to Mar 2001 for the Dot Com bust and from Jun 2006 to Jan 2008 for the GFC. That's in the modern period, the ancient history like the 60s and the 70s it took even longer. This said, equity markets reacted to these inversions within a much shorted timeframe, following the usual waterfall - bond market are smart, equity markets are dumb and economists are complete idiots. An additional consideration should be that the higher the overall level of interest rates, the easier it is for the curve to invert. So an inversion with 3m bills at 2.5% is much more significant than an inversion at 6.5%, for example.
As @sle explained 1998 was false signal because of LTCM/Emerging markets scenario and some consider Feb 2006 was also false signal or very early warning. It dipped again to negative area later. There were some false signals in 1960's followed by the real one. But those were in different economic regime. Few economists consider signal has to persist for one quarter to be valid. Any way equities are not automatic sell because of the warning. They usually have much more room to run before they fall.
As you approach lower and lower rate, the term premium increases (because people rightfully ask how low can the rates stay for a long time) and the attractiveness of the longer term debt decreases (because nobody wants to own long-term a low-yielding asset even if it's safe).
Shouldn't this be "decreases"? When the yield curve is flat, there is no need to hold the long term debt. If you look at ACM term premium, it has been downward slope for long time. Too much government intervention at the long end after the financial crisis has affected this premium.
I think we might be using different definitions of "term premium", you are probably thinking about some output of a term structure model. My view is relatively simple. Imagine that you have two riskless curves, one has a 0.1% overnight rate and the other has a 7% one. For the former, it's rational to assume that interest rates are not going very negative (after all, you can keep cash in other vehicles like your mattress), so the expected distribution of the lower-yielding bond is very skewed to the downside. On the other hand, the higher yield is probably much more symmetrical in terms of expectations. Predictably, on a lower-rate curve, the investors will demand a higher yield to hold a bond for a long period of time. The result is usually an upward sloping curve that's much steeper in geometric terms. Economists usually break this steepness into "expectations" and "term premium", but I personally think it's a fools errand as the two are closely intertwined.
Yes. I am no economist. This is what i was referring to. Bond yield = Expected short term rate + Term premium But back to inverted yield curve, since Fed's are tinkering both short and long term rates, how valid is the yield curve signal? Recent fed comments about stopping rolling of long term debt has some thing to do with the recent inversion along with some softness in the data.
I think, personally, that short rate expectations get fuzzy past 2 years or so and most of the curve is shaped by the risk preferences. This inversion is very typical in that the Fed raised the short term rate, but the long-term rates stayed the same. So I personally think it matters, but who knows, of course.
As an aside, there are a lot of things happening around the Fed that are very worrisome. My primary concern is the attack on the independence of the Fed by the current administration. First, jawboning, now nomination of various stooges. Granted, this was bound to happen from the moment the Fed decided to do a monetary offset against Trump’s fiscal stimulus (or, at least, from the moment they said something along these lines). Now it is the reality and the Great Leader is not even being as subtle as Abe was 5-6 years ago. While at least one of the nominees is truly terrifying and the other one is simply scary, the real problem is where it ends. If Trump manages to put a bunch of political partisans on the Fed board, what's to prevent the next president to do so and then the next one? Then there is the hollowing of the fed fund rate "user base" - most banks are not really using the rate any more, it's down to about 25% of what it was rather recently (has to do with various structural changes). So Feds rate announcement is tending towards symbolic, while the real actions are going to be manifested via the OMOs. Combine this with the push towards "forward guidance" as a policy tool, add my prior point and you can see how that can be a bit scary.