I’m sure we are all aware of how the yield curve acts when the credit markets see a slowdown in the economy coming. I came across this article this morning; http://www.tradersnarrative.com/yield-curve-model-zero-probability-of-new-recession-4449.html What question I have is; how can you rely on this type of analysis/leading indicator, when the curve is being anchored on the front end?
I’ve been wondering about this too. It depends on what you think the causal reason is for the yield curve effect. The traditional, convoluted explanation for the inverted yield curve is that the long end drops below the short end because investors expect a recession (omniscient guys that they are) and therefore expect the Fed to reduce short term rates. Since the long rates can be interpreted (noisily) as an average of the short rates from now until maturity, the long rates will go down. Causal Mechanism: Expectation of Recession -> Expectation of lower short rates -> lower long rates If this is the explanation for the yield curve effect (correct anticipation of recession leads to the inversion signal), then the implication is that a ZIRP will basically mean that the yield curve can’t invert. However, we may not be safe, because the only reason it inverts is because of expectations of recession. If we expect recessions for other reasons, but the yield curve can’t invert because of ZIRP, we may get a recession anyway. — Personally, I prefer a more simplistic explanation of inversion. When the yield curve is inverted, banks find it harder to make money by lending. The contraction of long-term credit reduces investment and provokes a recession. This is a similar to monetary tightening, but it happens at the long end, and not the short end. Causal mechanism: Inverted yield curve -> reduced lending -> reduced investment -> recession If this is the explanation for the yield curve effect, then the link between inverted yield curves and recessions should be stronger (though flat yield curves should create slowdowns too). The implication is that the chance of a double-dip recession is substantially reduced with ZIRP because the yield curve essentially can’t invert (at least in nominal terms). The two explanations are compatible, since you can have explanation 1 create the inverted yield curve and explanation 2 create the recession. If that is true this case, fear of recession can create a recession, if the fear affects the yield curve sufficiently. — Remember, though, the yield curve can invert in REAL terms, even with a ZIRP, if you expect sufficient inflation in the future. That would be a VBT (very bad thing), since the implication is that you will have recession AND inflation (stagflation), which is slightly better than outright deflation, but only just. — The assumption however, is that financial conditions are the main reasons for recession. This is almost certainly not the case (structural effects can cause recession too). To the extent that our economic problems are structural rather than cyclical, the yield curve can misjudge the probability of recession.
I was excited when I saw this topic, and more excited when I saw bchadwick as the last responder. Nice.
The front end of the curve is always anchored, or at least tethered. Short term rates will never deviate that much from what the Fed wishes them to be. The main point is in what direction the yield curve is pointing and the slope of the curve. Since the curve is at historical levels of steepness that would point to a coming expansion. Generally the yield curve has been pretty good at predicting the direction of the economy too. Anyway, to directly answer your question, the front end of the curve being dictated by the Fed is the whole point. That’s monetary policy in action for you.
DId you just write this? Jeebus, you’re a beast.
I often write in order to figure out what I think.
I agree with bchadwick with it being two combined factors driving it. However, I would put both 1 and 2 in terms of monetary policy and the expected path of nominal GDP growth. If monetary policy is tight, then the expected path of nominal GDP growth will fall due to the restricted lending and lower investment. Further, if monetary policy is tight now, then people will expect short-term rates to fall in the future as the central bank reacts to the weaker nominal GDP growth path it has effectively set the economy on. The proposition relates to the yield curve solely due to how people view the yield curve as reflective of monetary policy. If the yield curve is not reflecting the tightness/looseness of monetary policy (as you could argue now that it is), then it really isn’t that good of an indicator. Another thing I would point out is that when looking at the yield curve in real terms, you shouldn’t just subtract out the current inflation rate from both of them. Otherwise you’re left with the same curve. You would have to use the expected inflation rate over the respective period on the curve which means you may as well use the TIPs yields (only problem is they have a short history so no long comparisons to history). The models themselves are based on probit regression and if you have a package like stata it’s pretty easy to build one yourself. In these models yield curve dynamics generally come out as one of the best predictors of the probability of a recession. The issue is whether what made it a good predictor previously will make it a good predictor in the near future.
Interesting article (a bit dated though): http://seekingalpha.com/article/23870-an-80-year-yield-curve-history-and-its-implications
So how would the YC point to stagflation? Would you even be able to call it stagflation if we are in “muted” economic growth or even treading water as far as GDP is concerned? —>My opinion is that inflation is a risk in 4 to 5 years. There is way too much false liquidy in the system and too much debt needs to be paid off, a majority of which is implied on the GOV’T. The best way to attack this debt is to basically inflate your way out of it, thereby reducing the cost of the debt. BChadwick, I resound with this; “Causal mechanism: Inverted yield curve -> reduced lending -> reduced investment -> recession If this is the explanation for the yield curve effect, then the link between inverted yield curves and recessions should be stronger (though flat yield curves should create slowdowns too). The implication is that the chance of a double-dip recession is substantially reduced with ZIRP because the yield curve essentially can’t invert (at least in nominal terms). The two explanations are compatible, since you can have explanation 1 create the inverted yield curve and explanation 2 create the recession. If that is true this case, fear of recession can create a recession, if the fear affects the yield curve sufficiently.” However, I am not making the connection between fear causing the connection, yet credit markets still being theoretically easy. I can understand how prolonged pessimism can push the economy back into low or no growth, however, maybe I am too young to realize if there is a disconnect between fed rate and private lending.