Did the Yield Curve Send a “False Alarm?”

"So much for the inverted yield curve."

Or so says a recent Bloomberg column. If I read this correctly, any inversion that fails to cause an immediate recession is proof positive that inversions are meaningless, the bond market clueless, and data analysis of little if any value. Therefore, as an economic indicator, this must be declared null and void and immediately disposed of.

To wit:

"Bond yields may have sent a false alarm. The government said today that U.S. employers added 211,000 jobs in March, capping the best start for hiring of any year since 2000. Gross domestic product last quarter probably expanded at a 4.7 percent rate, the fastest in more than two years, a Bloomberg News survey shows. The Federal Reserve is talking about the need to keep raising interest rates to make sure the economy doesn’t overheat."

Let’s reiterate some of our prior comments about the yield curve, and what it means for investors:

First, any serious student of the market needs to consider the depth and duration of the inversion; That impacts what it implies for the future. As we wrote last year:

"A brief, shallow inversion won’t signal any marked slowdown in the economy.

Over the past several decades, the yield curve has had to invert by two percentage points or more before a recession materialized."

The longer and more inverted an inversion is, the greater the slowdown it forecasts. A short, shallow inversion is a warning; The longer it stays inverted, the greater the impact. There is a nuance to the curve that seems to be getting overlooked.

Second, Yield curves do not take place in a vaccuum. The context and other factors involved need also to be considered, such as energy prices. And to reiterate this yet again,  one needs to consider at all times multiple variables when analyzing Markets. The context of rising oil prices and inversions is important (see the charts here for more).   

Its worth pointing out that the yield curve is actually more than one variable; its a combination of two factors:

First, Short interest rates, which reveal the Central Bank’s economic expectations.

And, second, its a function of long rates, which embodies the Bond Market’s economic expectations.

As we quoted in December:

"The historical record speaks for itself," said Merrill Lynch analysts in a report published Friday.

"In the past 30 years, the yield curve has inverted five out of the eight times the Fed has been tightening monetary policy. Each of those five times an economic recession has ensued one year later — our fear (though not our base case) is that this time will be no different."    – Marketwatch

That’s a year later — not 90 days. If you want to be even more precise, the historic lead time is 40 weeks prior to a slowdown.

Lastly, consider this chart:
(previously shown in September ’05)

click for larger graphic


The data looks fairly persuasive that any inversion is worth noticing, and that a longer deeper one is a broad warning sign that investors ignore at their own peril.

Yes, we live in a bumper sticker society, where nuanced is scorned, sound bites rule, and careful contemplation is a rarity. That doesn’t mean we all have to succumb to the philistines . . . 



Yield Curve Sent False Alarm as U.S. Economy Expands

Elizabeth Stanton
Bloomberg, April 7, 2006 09:44

Understanding the Inverted Yield Curve
August 20, 2005

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. thecynic commented on Apr 12

    i’m not exactly sure if this is related but pretty sure. if you track marshallian k, a measure of excess liquidity, it exploded from mid-2000 (when 2s/10s was inverted) into mid-2003 as the Fed was desperately flooding the sytem with liquidity and devaluing the dollar, trying to stop the japanese style carnage. over the same period Mk exploded, the 2s/10s spread went from -50bps to +260bps. that tells me the bond market was tightening because the Fed was super-easy. since that high in 6/03 when the Fed was finally finished at 1% FF, both Mk and 2s/10s peaked and began to trend lower. there has been a disconnect though, as the 2s/10s spread collapsed while the Mk has stayed relatively elevated, though well off the highs. i think this is why commodities, stocks and real estate also remain at recent highs despite 375 bps of tightening.
    so now the issue is whether the bond market (or macro hedge funds) eased too much thinking the Fed would choke off liquidity and now is staring at still a weaker dollar and high commodity prices indicating an elevated level of excess liquidity. if this is the case, the yield curve could begin to widen/tighten until the Mk comes back to more normal levels, especially if the Fed stops at 5%, which would put a lot of pressure on asset prices.
    i think we either get more tightening from the Fed moving over 5% or we get more tightening from the bond market through a wider 2s/10s spread. either way it’s bad for asset prices.
    anyway sorry for the babble, but BR, i’d be interested to see you thoughts on the matter.. bloomberg will allow you to chart Marshallian K (i got by dividing M2/GDP Cur $) over 2s/10s and you can see what i’m seeing.. i might be looking to much into it, but there are too many coincidences to ignore. at worst it’s another way of looking at the “conundrum”

  2. kharris commented on Apr 12

    Not to pile on, but…

    This link is to what seems a pretty respectable look at the predictive ability of the curve:


    The models used are probit models – yes/no dependent variables. Recession or no recession. We don’t need to be so crude in thinking about the implications of the curve. If we could guess on the strength of the curve that growth is going to go from 3.5% to some slower pace, wouldn’t we be able to use that information? Recession is, after all, just what a committee of guys says it is. My pay, my job prospects, the performance of my portfolio, the performance of my insurancer’s portfolio, the next election – knowing that growth might slow gives me information about all this stuff, without having to know that there will be a recession. Bloomberg’s suggestion that the cuve is a flop – even if there is no recession after 40 weeks – is pretty simple-minded.

  3. me commented on Apr 12

    “The government said today that U.S. employers added 211,000 jobs in March, capping the best start for hiring of any year since 2000.”

    Excuse me??? I thought Bill Clinton AVERAGED 240,000 jobs per month for 8 YEARS. These idiots think 211 is a great number??????????

    And why no one ever looks at the quality of those jobs created under Bush. 52,000 bartenders and waitresses. Construction? Now that is a job with a future.

    And dont’ forget $3 a gallon gas a month before the driving season and 2 months before hurricane season. No more refis to pay off those credit cards, this hurts now.

    Put me down in the yeild curve camp. Somehow Bloomberg or the government have little credibility.

  4. Detroit Dan commented on Apr 12

    Excellent, excellent comments by kharris and me. That Bloomberg headline & article were terrible, IMO…

  5. Bastiat commented on Apr 12

    Bill Clinton hired 240,000 people per month? Hmm, that’s pretty impressive.

    Oh wait, no, that’s not right. The Chief Executive can’t create jobs.

    Quality huh…

    It’s interesting because the MSM usually likes to dump on the Prez.

  6. me commented on Apr 12

    “Oh wait, no, that’s not right. The Chief Executive can’t create jobs.”

    You ever he the saying a fish rots from the head. Bush and his policies fo not foster the creation of jobs. They do foster pension theft, brook excessive and obscene CEO pay, well you get the idea.

    Its funny how conservatives like to take credit when Reagan created almost as many jobs as Clinton, but when Bush created none, then it is a different song.

  7. Ironman commented on Apr 12

    Via James Hamilton’s Econbrowser:

    Jonathan Wright, a brilliant research economist at the Federal Reserve Board, recently completed a very interesting paper titled The Yield Curve and Predicting Recessions. Wright’s research seems to have been influential in Fed Chair Ben Bernanke’s recent assessment that the current very flat yield curve does not signify a coming significant economic slowdown.

    In a nutshell, Wright finds that the two factors of yield curve inversion and the federal funds rate may be used together to better predict the likelihood of a recession occurring within a future twelve-month period.

    For the time period in question, the federal funds rate was low (by historic standards), leading the Fed to dismiss the yield curve’s “prediction” of recession.

    Political Calculations offers a tool for finding the likelihood of a recession occurring in the U.S. within a future twelve month period based upon Wright’s work.

  8. Idaho_Spud commented on Apr 12

    Perhaps the Fed is, as Dr. Bernanke might say, ‘managing expectations’. The expectations they are managing might not only be inflation, but also DGP growth.

    Of course it’s clear that they are unable to manage the real thing :)

  9. todd commented on Apr 12

    I think we would STILL be seeing an inversion had the long bonds not been spooked by inflation concerns. If long bond inflation concerns are indeed correct (which I believe they are), the Fed will have to ratchet short term rates at a faster pace which may re-invert the yield curve.

Posted Under