Unraveling Steepener Trade Heightens Focus on Fed-Liftoff Timing

Wall Street banks are increasingly calling the end of a popular trade that dominated the bond market for the past few years.

Bets on a steeper Treasuries yield curve driven by the so-called reflation trade are crumbling, raising the crucial question for investors of what it means for when the Federal Reserve may start raising rates from near zero.

Traders are braced for hiking to start in early 2023, the same year most policy makers see it happening, and for a Fed announcement on tapering bond purchases to come as soon as the next few months. Citigroup Inc. analysis shows that before the 2008 financial crisis tightening came about 12 months on average after the 5- to 30-year curve topped out.

“The curve always reaches its steepest point well before the first rate hike,” said Jabaz Mathai, head of Group-of-10 rates strategy at Citigroup. “That’s because the market starts pricing hikes into the front end of the curve and slower growth in the back end.”

Unraveling Steepener Trade Heightens Focus on Fed-Liftoff Timing

That dynamic has been in play since the curve reached a roughly seven-year high in February. Five-year yields are about 25 basis points above mid-February levels, as the economy’s revival from the pandemic led traders to bring forward bets on when the Fed will hike. Rates on the long bond, meanwhile, tumbled Thursday to the lowest since Feb. 2. The result is that the yield gap is approaching the narrowest levels of 2021.

In the easing cycles that ended in 1992, 1996, 1998 and 2003, the spread between 5- and 30-year yields peaked from 7 to roughly 16 months before the first hike, Citigroup data show. But after the recession that ended in 2009, it took years for the Fed to tighten, as slower trend growth helped cause the central bank to consistently undershoot its 2% inflation goal.

Confidence Questioned

For HSBC Holdings Plc’s Steven Major, who sees lower yields ahead, the market’s confidence in the timing of tightening is misguided.

“I’ll believe it when I see it,” said Major, global head of fixed-income research at HSBC. “The timing is supposed to be contingent on the tapering being completed, but what happens if there’s some emergency during the tapering?”

Wednesday’s release of the minutes from the Fed’s June meeting showed officials saw progress toward tapering, although they weren’t ready to communicate a timeline.

Major, a long-time bond bull, warns investors not to be thrown by the fact that in past cycles the curve may have peaked at steeper levels. He says that after adjusting for the current “rate regime,” in which he predicts the Fed won’t be able to lift rates too high during the entire hiking cycle that may be ahead, the curve has already hit sufficient extremes.

He predicts 10-year yields will fall to 1% at the end of 2021, from around 1.3% now, and still be there a year later. Reflation has largely been priced in, and the market is now looking to what comes next, he said in a research note this week.

Economists surveyed by Bloomberg already predict the U.S. economy will slow from a growth rate of 6.6% this year to 4.1% in 2022 and 2.3% in 2023.

Signal Strength

The shift from steepening to flattening doesn’t, however, provide as significant a signal on future growth as an inversion of the curve does on a pending recession, according to Campbell Harvey, a professor at Duke University’s Fuqua School of Business. He’s credited with research in the 1980s that showed a link between curve slope and economic activity.

“A positive yield curve is predicting positive growth, and according to the model, the steeper the curve the higher the expected growth,” Harvey said. Yet “the maximum explanatory power the model got is when you are in the downstate, so when you get that inversion you got a high probability of a recession.”

The curve’s inversion in 2019 put the economy on recession watch. A decline in long-maturity yields below those on shorter-term obligations is considered a reliable harbinger of recession in the U.S. within the next couple of years.

To be sure, curve steepening still has its proponents. This flattening episode was “more of a corrective move,” said John Fath, managing partner at BTG Pactual Asset Management. “Those types of moves happen when you get closer to a Fed move. It’s a predictor, but it’s premature.”

A rate increase is at least a year and a half away, in his view. In the meantime, he says the cost of holding a flattener trade on the 5- to 30-year spread amounts to a bet that it will be at least 30 basis points lower in a year, flatter than in December 2015 when the last cycle of hikes was already under way.

Momentum Builds

The momentum toward flattening over the past couple of months came amid declining inflation expectations, in part as the Fed moved up projections for when it will start lifting rates.

The 10-year breakeven rate, a market proxy for the average annual inflation rate over the next decade, has dropped to roughly 2.2%, from an eight-year high in May at 2.59%. At the same time, 10-year yields have tumbled to the lowest since February.

Helping push down long-term yields, traders have lowered their view on how much tightening they foresee, Bank of America Corp. strategists say. BMO Capital analysts say investors are also buying into the Fed’s mantra that surging inflationary pressures will prove transitory -- and initially function as a tax on consumption -- and that if price pressures do trigger Fed tightening it may interrupt the recovery.

Bank of America sees the the 5- to 30-year gap narrowing to about 105 basis points by the end of March. The forecast reflects policy makers’ signal last month that they’re increasingly concerned about inflation risks, strategist Meghan Swiber said. The bank sees liftoff in the second half of 2023.

In their view, the bias is toward flattening “because the five-year is the component of the curve that’s most sensitive to the pulling forward of rate hikes,” she said.

©2021 Bloomberg L.P.

BQ Install

Bloomberg Quint

Add BloombergQuint App to Home screen.