"When To Worry?": How Long After The Curve Inverts Does The Recession Begin

The recent (bear) flattening of the US yield curve to levels not seen since before the GFC, a move which has only accelerated in recent weeks as the stock market hit all time highs, has prompted some to question the strength of the US economic cycle, and others to ask outright how long before the curve inverts, signaling an imminent recession. Here, as Citi's Jeremy Hale notes, just as "Dr. Copper" can sometimes be viewed as a stock market precursor, so "Professor Curve" (particularly when inverted or aggressively flattening) can be viewed as a signal of that policy is too restrictive relative to economic fundamentals (especially when using term premium suggests the curve should already be inverted). That said, during an expansion it’s generally normal for the curve to flatten, as the economy expands and the output gap closes, as shown in the chart below. This can be attributed to expectation of a higher Fed funds rate, but also a lower term premium, or more ominously, an inability to pass through inflation to the broader economy, leading to tighter financial conditions which ultimately manifest in an economic contraction.

Putting the recent 2s10s flattening in context (blue line on chart above), assuming this cycle started at the December 2013 steepness of 264bps, the curve has flattened for the past 60 months. On average, historic flattening cycles last for 2-2.5 years and flatten ~270bps from peak to trough. As Citi notes, we have flattened three quarters on the way there, or roughly 204bps so far in this cycle, therefore in comparison to previous episodes; perhaps this flattening dynamic is growing grey hairs... but it’s certainly not finished yet.

Indeed, if history repeats, then another 67 bps flattening is implied before we see an inversion of the curve. And while this cycle of curve flattening has been particularly slow, assuming a runrate of 40bps of flattening per year, then we could see a flat curve within 18 months (Figure 3, left). So "should we be worried?" Citi asks and answers that, rightly or wrongly, market participants with grey hairs would preach that all is well until the curve begins to invert. Ah yes, but that's not the full story, as Citi explains below:

Sometimes inversion provides a timely signal for the economic cycle a la 2000, where Professor Curve predicted almost the ding-dong high in the SPX. However the 2006 episode of inversion dished up 7 months of pain for equity bears, with 18% further upside for the SPX. Ditto for the 1989 episode where equities continued to rally 22% into the 1990 recession (Figure 3, RHS). For now, we’re comfortable with the flattening dynamic with regards to other markets but would become increasingly cautious as the curve approaches zero.

In other words, once the curve inverts, it could either mark the top-tick of the market right there... or leave up to 22% more in equity upside before stocks finally crash.

Citi's optimism - for now - aside, one notable distinction about the current flattening is that, unlike much of the curve move in 2016, this one has been driven by the front end, i.e. a bear flattening.  The front end of the US curve has significantly re-priced since September with the extension of the debt ceiling and the realization that fiscal easing could be achieved by the Trump administration.

Also worth noting is that while the short end has been driving curvature, the long end has been relatively rangebound, at least over the past year. What Citi finds particularly interesting is that even with ‘impending’ fiscal expansion in the US and balance sheet normalization by the Fed, term premia are actually still negative and suggest that the nominal 10y UST should trade closer to ~1.6% if the priced in forward short rate was at end pre-Election levels (Figure 6, LHS). This would suggest that there is, of course, a risk that the unusually low term premium - pushed to near record low levels by foreign central banks QE and NIRP, herding investors into long-term US duration - could suddenly rise; of note, perceived inflation risk could reverse its course quickly if inflation
suddenly trended up.

Some Fed estimates suggest that term premium is ~0.9% lower than it would be without the Fed’s large securities holdings, and that this term premium effect will gradually diminish with the reduction of the Fed’s balance sheet. But the Fed’s normalization has been well telegraphed; therefore the market has had the opportunity to anticipate this for several months (this goes back to another point made by Citi's Matt King that the market has lost the ability to discount the future). In fact, the central bank depresses the term premium by limiting the uncertainty surrounding monetary policy. In short, Citi is skeptical of the material  impact that BSA may have on nominal yields given a relatively hawkish Fed. Furthermore, as the Fed continuing to tighten, it is possible the US economy is ‘locking in’ any gains that may be passed through to CPI. It is worth noting that in the last two cycles, US firms (and others) have had trouble - if not found it impossible - passing wage costs through to prices.

So even if a curve inversion does not spell imminent recession, what is next for the (shape of the) curve? Well, more of the same flattening it appears, as the curve takes more aggressive steps to flatten.

As Hale explains, the term premia and the forward curve suggests further flattening ahead (Figure 10 bottom LHS and top RHS). It’s also worth noting that in the past throughout Fed hiking cycles, the curve flattens on average between 100-125bps (Figure 10  bottom RHS), we’re currently around half way through on that basis assuming the Citi Fed call is right. But relative to other cycles at this stage, perhaps we have moved far enough for the time being. Citi's fair value model using ACM term premium, the breakeven curve and a proxy for the r* also suggest flattening is close to fair value for now (Figure 10 top LHS).

But more medium term as the Fed keeps tightening, the curve will likely continue to flatten, and may even begin to bull flatten, should inflation expectations fall further. Ironically, Citi concludes, if the Fed wants higher long-term rates, and with them a steeper yield curve, it may need to hold back on further interest rate hikes until inflation surpasses the target/ backward driven inflation expectations rise.

Until then, however, expect people to keep talking about the flattening yield curve.


Crazy Or Not conraddobler Sun, 11/26/2017 - 10:06 Permalink

as Max 2205 comments below FED will do all it can to encourage the consumer economies to limp on til sanctions ramp up with China Russia and the full pre-war preps get underway. Which likely by current narratives will unfold during 2018 -2019. First shots will be fired and US successes will be rolled out just prior to the re-election campaign. Likely targets:DPRKChina Base in DjibuitiRussian Base in SyriaSpratley IslandsLonger oddsSomaliaMining oriented targets interrelated with Chinese African rail extension to silk road 

In reply to by conraddobler

Consuelo Crazy Or Not Sun, 11/26/2017 - 16:42 Permalink

+1 China, and to a less-patient extent, Russia, have so far been willing to 'keep stacking', whilst absorbing the sanctions, threats, insults, etc., because they both know what the writing on the wall says.    I wouldn't doubt for a moment however that as the PNAC push-for-war accelerates, so too will the acceleration of the aforementioned nations to ramp up their capabilities to trade outside the $USD sphere, while securing ever more 'partners' to advance along with them on the new silk road.The War Pig and it's task-master there on the Mediterranean absolutely will not stop until the $grease burns away and metals fuse together in a complete budget-funding lock-up. 

In reply to by Crazy Or Not

Arnold Sun, 11/26/2017 - 10:01 Permalink

I know it's just the funnel of the venue,
but there seems to be a lot of confusion,
now that Subscription news letters and Analysis is giving its' last gasp.

Hedgies standing pat with current holdings?
Cats and dogs living together?

red1chief Sun, 11/26/2017 - 10:00 Permalink

The central banks are not stupid, they know a recession will result from the coming yield curve inversion. I think the crypto bubble is the one that finally convinced them to end the party. Cheap real assets await those who are patient, as it could take a couple three years to materialize.

Shibumi2 Sun, 11/26/2017 - 10:00 Permalink

Good analysis...but all the charts, verbiage and bullshit don't address that what will happen ultimately will be decided by a handful of men who will profit handsomely from whatever decision they make...

buzzsaw99 Sun, 11/26/2017 - 10:14 Permalink

artificial inversion. artificial. everyone talks about curve inversion as if it is a natural phenomena.  it isn't.  also, saying that inflation is a problem is bullshit too.  we have biflation now, where guys like bezos are worth twelve figures on paper and shitty art sells for nine.  does anyone see long term rates rising as a result?  as for duration risk the funny thing is that if the central bankers didn't jack with everything on a daily basis some investors might actually be willing to pay a premium for a longer term non callable note.  we'll never know though because they will never ever stop fucking with things.  1989 and 2006 probably just reflect a lag in market response to the fed fucking with shit.

SuperRay Sun, 11/26/2017 - 11:14 Permalink

Nobody knows what's going to happen.  To use past data in an environment that has never existed before (massive manipulation by central banks) is like predicting the price of apples based on how many refrigerators were made last year.  It's comforting to believe we can predict anything at this point, but the only thing I predict is that the Zionist will blow up the world if they don't come out on top at the end of this shitshow...

CRM114 Sun, 11/26/2017 - 12:32 Permalink

"Indeed, if history repeats" Nope. This level of artificiality, worldwide, has no precedent. Real markets are dead, it's all just Monopoly money. No one has a f#cking clue When, Why, Who or How for the trigger. But we all know What.