Thinking clearly about the yield curve

Barry is like the farmer who thought the sun would no longer rise after his rooster died.  He has observed a correlation and has inferred causation.  The yield curve is an indicator of something, not a cause.  Ask yourself this question:  Would U.S. economic prospects be higher if Asian banks were less willing to buy our bonds?

I’ll give a more complete explanation, but first look at the complete article.

Link: Inverted Yield Curve: Its different this time (not).

The yield curve, as measured by the ratio between the 10 and 2 year treasuries, is merely a few ticks away from inverting. This is something worth paying close attention to. What’s the significance of an Inversion? It reflects a decreasing demand for …

The slope of the yield curve is sometimes a precursor to a recession.  As Barry points out, and he is better than many others at this, it seems to be a necessary but not a sufficient condition.

Use of a simple indicator like this without consideration of the complete causal model is a heuristic, a simple rule of the sort extensively analyzed in the Behavioral Finance literature.  It is important to think about why the rule works, and that means identifying a causal model.

A typical causal interpretation might be one where the Fed was "behind the curve" in anticipating inflation.  High inflation expectations would be reflected in high long-term interest rates.  The Fed would be pushing short rates above the "neutral" level, and neutral might be 6 or 7%  — or even higher.

The data used for this hypothesis do not support the conclusion.  Try to find a prior inversion where the Fed increases started from 1%, stopped at around 5% and the ten-year stayed below 5%.

I realize that some who do not understand how inflation is calculated will say, "Yes, the Fed IS behind the curve.  There is inflation!  Look at energy prices."  But that is not the message of the market, and it is not the message from the economic data.  If long bond buyers really feared inflation, they would insist on higher rates.

Three final notes:
1)  I think that anyone predicting a recession from the yield curve inversion should explain exactly why it would be good for the economy if the ten-year yield moved higher.  Not why it would be good for banks, or their own hedge fund, or the carry trade, but why it would be good for the economy.

2)  Those making this argument should specify what interest rate move it would take to make them more bullish.

3) This is an example of how the financial punditry reads behavioral finance books and figures that the principles apply to the masses, not themselves.

You may also like


  • Barry Ritholtz December 27, 2005  

    I thought this was pretty clear:
    An inverted yield curve is not a guarantee of a recession, but it is nonetheless a worrisome thing. If it doesn’t foretell a recession, its not because “its different this time;” rather, its more likely because only some conditions precedent will have been met . . .

  • Barry Ritholtz December 27, 2005  

    BTW, an inverted yield curve coming contemporaneously with rising inflation and high energy prices has been a precursor to recessions.
    No guarantee it happens again, but historically, the odds favor it.

  • Jeff Miller December 27, 2005  

    Hi Barry,
    I think that readers will understand both what you wrote and what I wrote. You did not say anything about a guarantee of recession, but you say the odds favor it.
    I am hopeful that you will do more than the average person discussing it on CNBC today. Spell out your causal model. What are the historic precedents that you cite? How many of them are there? Did any of them start from a 1% Fed Funds rate?
    This is the difference between research and a heuristic. If you will not spell this out, I doubt that anyone will.