As 2s10s Inverts, Dudley Says US Recession ‘Inevitable’

Two days ago, in “Fatalism Sets In As Curve Starts Recession Clock,” I wrote that if your preferred section of the curve wasn’t already inverted, you wouldn’t have to wait long.

The wait on the 2s10s was less than 48 hours. Tuesday’s inversion (figure below) was the first since 2019. The bull flattening move came amid apparent receiving flows and cash-buying in fives and sevens. A tailing seven-year sale belied decent stats.

I’m reluctant to countenance the veritable deluge of excuses and belabored attempts to explain why this time is different vis-à-vis inversions and recessions. As regular readers will happily attest, I’m someone who adores nuance. So, if you’re inclined to venture a nuanced explanation for why this time is different, I’m all ears. But I’m not very receptive to the idea that the curve gets “too much credit.” With the caveat that the bond market isn’t self-aware, and thus can’t actually “predict” anything in the strictest sense of the word, the curve’s track record is sterling.

There is, of course, a vociferous debate about how long the lag is between inversions and recession. All we can say is that if past is precedent, an inverted cash curve is a bad omen.

One person who thinks a recession is now a foregone conclusion is Bill Dudley. In a somewhat abrasive Op-Ed for Bloomberg (and Dudley isn’t a stranger to abrasive Op-Eds), he blamed the Fed for inflation and suggested a soft landing is all but impossible.

Although he didn’t deliver a wholesale indictment of flexible average inflation targeting, he did indict the Fed’s application of the strategy, arguing that the Committee’s commitment to achieving full employment and ensuring the durability of inflation before tightening policy meant the Fed was destined to find itself behind the curve.

Dudley wasn’t in the mood to accept Powell’s “victim of circumstance” excuse. “Powell blames bad luck — surprises such as snarled supply chains that officials could not have anticipated,” he wrote. Although Dudley conceded that Powell may be right “to some extent,” he nevertheless said the Fed can’t deny “responsibility for being so slow to recognize the inflation risks and begin to tighten policy.”

As for the odds of a soft landing, Dudley all but suggested it’s a mathematical impossibility. He cited the Sahm Rule (figure above).

If the three-month moving average of the unemployment rate rises by a half percentage point or more, a downturn can’t be avoided. “To create sufficient economic slack to restrain inflation, the Fed will have to tighten enough to push the unemployment rate higher,” Dudley said.

He left no room for equivocation. The title of his piece: “The Fed Has Made a US Recession Inevitable.”

Speak your mind

This site uses Akismet to reduce spam. Learn how your comment data is processed.

16 thoughts on “As 2s10s Inverts, Dudley Says US Recession ‘Inevitable’

  1. The Sahm Rule is a new one to me. Certainly makes sense. Just had a look at the Taylor Rule, and if my mind math is correct, it says the Fed Funds rate should be north of 9%. That would make the Fed more than a little behind.

  2. The bit of math I’d like to see is the effect of the historically anomalous excess of job openings, which may act like a buffered solution in chemistry. Tightening conditions may need to warrant companies eliminating several million openings before the unemployment rate even starts to increase. Reductions in openings will reflect both hiring churn as well as reductions in growth and hiring plans. Intuitively, I suppose, one would expect the excess openings to increase the amount of time needed in this tightening cycle before seeing inflation begin to moderate, which in turn may increase the odds of policy overshoot.

  3. I think saying recession is inevitable is a bit strong- but I think Dudley’s observations are basically correct. The Fed had a number of difficult policy choices. They threw their lot to backstopping the economy until they were sure it could get past covid. In the context of risk management and policy choice- not really a bad trade off. In hindsight, which is always easy, they should have likely stopped Q.E. late last summer or early this fall rather than when they did recently. The earlier you go, the less you have to tighten later. Now it appears Powell & Co. are going to slam on the brakes. The economy can clearly take some tightening but not in big 50 bps bites- not in rapid succession, and not with all the hawkish claptrap coming out of the FOMC and Fed bank presidents loudspreakers. It is always this time is different. The yield curve is one of the great leading indicators of future growth. It is trying to send a message- if only folks would listen. Powell and his Board can still avoid recession if they back off the hawkish tone, and slow down (note not stop but slow down) their rush to raise rates. If they raise them cautiously they have a better chance of keeping the economy going and giving the economy time to adjust to some of the shocks. Otherwise Dudley is going to be proven correct.

  4. I like that a recession is imminent because the Fed “must” pull a Volcker. Like, ok, why do we need to intentionally put people out of work?? Because the arbitrary price on their jug of milk might be 20 cents higher next year? Clearly they’ll be better off if they have no job and can’t afford milk at all!

    Absolutely ass-backwards boomer ideology.

  5. Extend the chart backwards more. Didn’t the curve also invert in August 2019? I guess the pandemic crisis obscured the predictive merit of that particular dip.

  6. H-Man, the inversion is like hurricane season in the south. You know when it gets really hot in August, water temperature rises providing the fuel. Does it guarantee there will be hurricane, no. But have the chances substantially increased, yes. You need to seek out Marko or ME who can spit out the percentages and likelihood of a recession based upon inversion but it still won’t be a guarantee.

  7. Here then “all ears”,

    “Not All Yield Curve Inversions Are Fatal”
    By John Authers
    March 28, 2022, 9:29 PM MST

    This comes one day after “Fatalism set in …”, and one day before Bill Dudley brandishes Sahm’s Rule. Why Sahm’s Rule? Does Dudley not give much credence to the flattening 2s10s curve recession theory himself?

    “Claudia Sahm’s Big Idea Is Making a Splash in Canada’s Election”, by Shelly Hagan and Kait Bolongaro September 2, 2021, 6:00 AM MST, also at, is some curious “strange bedfellows” at first glance. Invoking a ‘rule’ usually (of late by both Canadian Conservatives and American Liberals[1]) suggested as a mechanism to trigger fiscal automatic stabilizers payments to households (yet another M2 cluster bomb? delivered into American male bank accounts … go long Bitcoin and Meme Stocks again?). Seems a tad rich but not quite a red-herring. In any event, after Authers and his curves within curves dissection it’s all back to being a muddle to me. At least, it seems reasonable to me that the unemployment line crossing it’s ~10-month average probably belongs on anyone’s “wake up and smell the coffee” dashboard. Anyway, invoking the rule didn’t work out any better for the American Democrats against US Senator Joe Manchin than it did for the Canadian Conservatives against Prime Minister Justin Trudeau. Besides, I think she only proposed the rule in 2019! What sell respecting male economist would use a rule that new and proposed by a woman at that!?! This is Economics after all! SNAFU.

    [1] “Senate Democrats Push White House to Tie Relief Payments, Jobless Benefits to Economic Conditions in Build Back Better Plan” —

    1. Good grief! Now my timelines above are getting knotted up. Bill Dudley said in a speech, “The Outlook for the U.S. Economy in 2018 and Beyond” dated January 11, 2018[1]:

      “Historically, the Federal Reserve has found it difficult to achieve a soft landing—especially when the unemployment rate has fallen below the rate consistent with stable inflation. In those circumstances, the Federal Reserve has been unable to both push up the unemployment rate slightly to a level that is consistent with stable inflation and avoid recession.[fn8] Now, I don’t want to imply that a recession is inevitable once the FOMC finds it necessary to nudge up the unemployment rate to a sustainable level. The starting point in terms of the inflation rate is also important because it will influence how far the FOMC is likely to go in terms of making monetary policy tight. Nevertheless, I think it is fair to say that the track record on this score is not encouraging.

      [fn8] Since the late 1940s, any time the three-month moving average of the unemployment rate increased by 0.3 percentage points or more, it was a sign that the economy was either already in or about to enter recession. The cumulative rise in the monthly unemployment rate from its low prior to the official start of the recession to its peak has been at least 2 percentage points.”

      Dudley didn’t precisely state Sahm’s Rule back in Jan. 2018, but, he has been saying fairly similar things since. What pricked the sides of my intent to wile a my time on this was how the “rule” became Sahm’s namesake? What Economist doesn’t want a Rule named after themselves? Better yet she didn’t even have to die first! … Bill may be abrasive but deep down maybe he’s a big softie! Or maybe, it’s just Politics? Sahm packaged the idea up as “automatic stabilizer” and put in front of the right constituency and tada, a Rule is born!

      Of course, I haven’t read the pertinent chapter in the book[2] that may yet untangle this Gordian Knot. That said, just looking at the two versions of the Sahm rule available on FRED[3] it looks to cut it kind of close for a sell-signal indicator, unless you really, really, really don’t like false sell signals.

      [2] Sahm’s rule source work (?) which just happens to be the only economists paper I’ve ever look at online that didn’t have a publication date on the top … aaarrgh! Turns out the source work was not a Fed Research Paper but rather a chapter she wrote for a book. Available for free at

  8. I continue to be entertained by the 3 month spread stuff. I guess my curiosity has to do with interpretations of what happens when the old libor referencing isn’t part of the stories out there in the media. I’m drawn to the ameribor as a substitute for libor. I was chasing that curiosity yesterday. It seems the ameribor does price in a different story, because it’s not securitized like a treasury bill, which is good, because it factors in realtime market premiums.

    Anyway, here’s a Fred link that tells that story. It’s odd looking, because the treasury value lage more than the better, more current values of ameribor, which adds clarity as to what’s going on out there:

    These words and links are based on things in my head and I totally understand if what I’m writing is confusing. I have no position in anything, except lying on the floor stretching my back

  9. I’m not hoping to punish anyone or cause stress, but here’s another Fred chart comparing, the prime rate, treasury rate and ameribor rate.

    This demonstration is intended to show the change in short term rates and differences between the ameribor unsecured rate, secured treasury and the bank rate that’s adjusted for a premium.

    The reason I think this is interesting, is because when there is talk of inversions and spreads, we’re living in a very new untested world, where the old fashioned libor referencing is dead. The replacement values that are being tossed around in media stories don’t seem to be adjusting their spreads. The Fed is pushing their own SOFR reference, but, that’s not incorporating adjustments like libor or prime, thus, some Fed speak related to these spread issues is essentially questionable.

    I’m not s voting member of FOMC and not a supercomputing bot, but this strikes me as a strange anomaly that’s like little nano sphere like sand pebbles grinding away inside the Fed plumbing, which may cause the toilet to either back up or explode, not unlike various bathroom accidents, which can be prevented.

    Bottom line, why aren’t financial media gurus investigating how the reference rates differ and then what that implies in terms of 75 bos hike in May? What if our Fed friends are way off, because of this moronic nonsense?

    These words came me out of the blue and as usual, I haven’t gone back to read anything that was tapped out.

    1. Gosh wise @oldbird, maybe cuz, the “financial media gurus” are more reactive than proactive? Maybe those getting gored by the ox of rising short-term rates aren’t screaming loud enough yet to be noticed? Maybe the early victims of the next credit tightening cycle won’t engender all that much sympathy (a modern day variation on the old moral hazard story or maybe the English deep down still feel debtors opened and closed the doors to their debtor-prison accommodations on their own accord)?

      If I find out anything I’ll send you a telegraph.

      Will check back tomorrow to see if my repy to myself regarding Sahm’s Rule survivied the mysterious moderation process. With that I’m tapped out tonight.

  10. I’ll look forward to your telegraph.

    I’m done looking into this, but it’s fascinating that libor has been dead for almost 4 months, without much of a wake or grieving period. I think people were somewhere aware of it’s death, but as we enter this new stage of QT, inversions and increased risks, I’m not sure anyone realizes that our libor friend was actually fairly stable.

    I’ve done a poor job beating this dead horse but I still believe that these 10/2yr spreads or anything related to short term bills are distorted. Amen, I’ll turn this over to Jennifer and say no more, say no more:

    – Jennifer Wasylyk


    “Although the ARRC-recommended spread adjustments are expected to result in a SOFR Term Rate close to USD LIBOR loans in “normal” times, as a result of near-zero interest rates the current spread differentials between USD LIBOR and SOFR are less than the ARRC-recommended spread adjustments (as seen in the table below)….”

  11. Yield curve inversions roughly mark (+/- several months) stock market peaks. That has been true as far back as I’ve looked, into the 1960s. The “base case” should be that the relationship will hold. The argument “it is different this time” has a heavy burden of persuasion. Economists seem unable to predict recessions, and strategists have institutional pressures.

    The last time the curve approached inversion in 2019, there was the usual litany of arguments why it was different this [that] time. Engstrom at the Fed produced his paper arguing the 3M/F3M curve was better than the 2Y/10Y curve, which is the same argument the Fed is turning to now. Shortly thereafter, the 3M/F3M curve also inverted, and the bulls stopped talking about it. I had moved portfolios to very defensive, with very large allocations to cash & equiv, gold, short duration UST, and defensive equity regions/sectors. The pressure from “on high” to adopt the official near-max-bullish positioning grew so intense by early 2020 that I resigned. Soon thereafter the market fell apart.

    We didn’t get to see if a recession would have developed absent Covid, but economic data (not just the yield curve) was distinctly weakening . . . regardless, the yield curve chalked up another win and the economists/strategists another loss.

    1. “We didn’t get to see if a recession would have developed absent Covid, but economic data (not just the yield curve) was distinctly weakening . . . regardless, the yield curve chalked up another win and the economists/strategists another loss.”

      The trouble with the arrow of time for decision makers is that once we have made a decision, as the Fed has over the last two years, we can’t go back and see what would have happened if we had chosen otherwise. We also can’t see what would have happened if other conditions were present because they weren’t present and won’t be again. We make choices every day that can’t be second guessed. Que sera, sera. A wise man once said if one wants to reconsider old choices one would find that if they had it all to do over again, they would do what they did. All choices taken are real. Simulations are not and neither is the prospect of a do-over.

NEWSROOM crewneck & prints