The bond market is screaming recession

The gap between two-year and ten-year US Treasury yields reached minus seventy-five basis points in late November 2022, the deepest inversion since Paul Volcker was crushing inflation in the early 1980s. For more than four decades the slope of the yield curve has been the markets' most reliable guide to the business cycle. Central bankers can talk up a soft landing and chief executives can promise earnings growth, but a curve this far below zero is the bond market's verdict on both. It expects a recession, and it is saying so loudly.

An inverted curve is an unnatural state. In a free capital market someone lending for a decade wants a higher return than someone lending for twenty-four months, as payment for the duration risk, the inflation uncertainty and the loss of liquidity that come with tying money up for longer. When ten-year paper yields seventy-five basis points less than the two-year note, investors are saying they expect the Fed's rapid tightening to cause an accident, one that forces it into emergency rate cuts in the medium term.

From a blip to a chasm

The inversion has deepened in stages. The two-year yield first crept above the ten-year briefly in the spring, as markets priced the start of the tightening cycle, and the gap has stayed negative since early July. Each 75-basis-point rise from the Fed has pushed the front end up faster than the long end, as investors priced higher short rates now and lower ones later.

Fed officials have argued that the 2s10s slope is the wrong measure. Chairman Jerome Powell has said he pays more attention to the near-term spread, the gap between three-month bill yields and the bill rate expected about eighteen months ahead, which Fed staff research found to be a better predictor. That defence has worn thin in recent weeks, because the three-month bill now also yields more than the ten-year note, so the inversion no longer depends on the choice of the two-year as the short end.

Short end against long end

The inversion comes from a tug-of-war between policy today and the economy tomorrow. The two-year yield is tied closely to where the Federal Reserve is taking its policy rate. The Fed is committed to lifting the federal funds rate past 4.5 per cent, and two-year yields sit near 4.5 per cent as a result. The front end is pricing the Fed's determination to crush demand now.

The ten-year yield rests on long-run growth and inflation expectations. Investors have pushed it down towards 3.75 per cent, and they are betting real money that the US economy cannot live with borrowing costs above 4 per cent for long. Long-term investors are buying ten-year bonds on the view that credit stress, a run of corporate defaults and rising unemployment will soon force the Fed into a full retreat.

A record of warnings

The curve's reputation rests on a long run of correct calls. Every American recession since the late 1960s has been preceded by an inversion, including those of 2001 and 2007 to 2009, and the relationship was first documented in academic work in the 1980s. The lead time has varied from roughly six months to two years, which is why investors who acted on the signal immediately have often been early, and why officials find it easy to dismiss in the months after it appears.

The one episode that looks like a miss is recent. The curve inverted briefly in the summer of 2019, and the recession that followed in 2020 was caused by a pandemic that no bond market foresaw. Sceptics count that as luck. Defenders note that the economy was already slowing before the virus arrived, and that the curve's job is to flag fragility, not to predict what finally breaks it.

Banks stop lending

The trouble with an inverted curve is that it also helps bring a recession about. Commercial banks make their money from maturity transformation, borrowing short-term deposits to fund long-term business loans and mortgages. An inverted curve takes that business apart.

Bank net interest margins get squeezed to almost nothing. Funding costs rise fast as corporate depositors demand yields that match Treasury bills, while the return on new commercial mortgages and term loans is held down by the flat long end. With lending margins negative or negligible, banks do what any sensible business would do and lend less. They tighten standards, cut credit lines and shut weaker corporate borrowers out altogether, and a line on a chart turns into a credit crunch in the real economy.

The process has begun. The Fed's own survey of senior loan officers, published earlier this month, showed banks tightening standards on commercial and industrial loans and reporting weaker demand for them. Lenders tend to tighten gradually and then abruptly, so the next quarterly survey, due early next year, will be one of the more important releases of the winter.

The benign reading

The best case against alarm is that this inversion partly reflects falling inflation expectations, not collapsing growth. If investors expect inflation to return towards 2 per cent, long yields should fall even without a recession, and years of central-bank bond buying may have compressed the term premium that normally keeps long yields above short ones. Banks, for now, are also earning wider margins, because the rates they pay depositors have lagged well behind the rates they charge borrowers.

Each point has some force, and none disposes of the signal. The margin cushion lasts only as long as depositors stay put, and money-market funds paying close to Treasury bill rates are already drawing cash away. A curve inverted by expected disinflation still tells borrowers that the Fed's policy rate is above where markets think the economy can sustain it.

The Federal Reserve's habit of waving away the 2s10s inversion as an irrelevant quirk is institutional hubris, and a risky kind. At minus seventy-five basis points the curve is telling the Fed that its policy is now tighter than private borrowers can bear, and the likeliest result is a recession, probably within the next year or so, that starts with banks refusing credit. Every modern inversion this deep has been followed by recession, and betting that 2022 will prove the exception means ignoring how credit actually gets made; small businesses and lower-rated companies that rely on bank lines will feel it first, likely by the middle of next year.