Skip to content

Recession Ledger

An independent record of the downturn that began in 2008 and the years after it. Every forecast here is dated, and checked at the foot of its page.

Economic predictions

The next recession is here: reading the fourth-quarter 2021 GDP report under the hood

You cannot see the intruder, and you do not think he is in the house yet. But you can smell the cigarette smoke through the bedroom window.

Economic predictions · · 728 words


The fourth-quarter GDP report for 2021 arrived at the end of January 2022 with a headline of 6.9 per cent annualised growth, above the 5.5 per cent economists expected, and a full-year figure of 5.7 per cent that MSN called the fastest since 1984. On its face it made a fool of anyone who had predicted, in October, that the economy would be in recession by the winter. This page had. So it owed its readers a look under the hood, and this is the record of it.

The prediction

The October forecast was that shortages, then spreading like a tide across a shallow beach, would combine with the Federal Reserve’s tapering of bond purchases to tip the economy into a stagflationary recession over the winter: the rare and worst kind, in which prices keep rising while output falls. It added, as a separate and explicitly weaker statement, that fourth-quarter growth might prove negative or close to it. The second statement was wrong. Whether the first was wrong depended on what the 6.9 per cent was made of.

What it was made of

The Bureau of Economic Analysis’s own summary said the acceleration was led by exports and by inventory investment. Exports were washed out by an equal rise in imports, which subtract from GDP. That left inventories. Private businesses, seeing empty shelves and long lead times, had bought everything they could get and find room for, and the change in private inventories contributed close to five percentage points of the 6.9. Personal consumption grew, at 3.3 per cent, from a third quarter in which it had barely grown at all, and part of that was households doing the same thing businesses were doing. This site had told its readers in October to lay in a winter’s worth of whatever they normally bought, on the grounds that shortages and inflation would make the stock worth more than the cash; it is fair to say the whole country took the same advice.

A quarter’s growth that consists of goods moving into warehouses is growth that will be subtracted later, when the warehouses are drawn down rather than restocked. Zero Hedge, reading the same table, called it a very low-quality print and expected future quarters to pay for it. Even on the headline, the economy had not returned to the trend it was on before the pandemic.

The Fed and the curve

The other half of the argument concerned the Federal Reserve, which was preparing to raise rates and end its bond purchases into a quarter whose real momentum was close to zero. Goldman Sachs cut its 2022 growth forecast again that week; JPMorgan wrote of a sharp slowdown and said no further hawkish surprises should be expected; the Atlanta Fed’s GDPNow estimate for the first quarter was already near nothing. The Fed, on the evidence of its own statements, did not see this, and the expectation recorded here was that it would hike into a recession it would not recognise until later.

The yield curve, the Fed’s preferred warning, could not help it. Two years of bond purchases across every maturity had set the curve where the Fed wanted it, and a gauge held by its reader shows nothing. The prediction was that when the purchases stopped in March, the curve would flatten and invert at unprecedented speed, showing all at once what it had been prevented from showing gradually. That is the “curve ball” of the original heading, and the following page records it happening.

Where the argument stood in February

The claim, then, was not that the fourth quarter had been bad. It was that the fourth quarter had borrowed from the first, that the borrowing was fear of shortages, and that a central bank about to tighten was reading the loan as income. The recession, on this view, was already at the window.

Later note. First-quarter 2022 GDP came in at minus 1.4 per cent annualised, later revised to minus 1.6, with the inventory swing doing most of the damage as this page expected. The second quarter was also negative on first print. The Fed raised rates for the first time on 16 March 2022 and did not stop for sixteen months. The NBER never declared a recession, because employment kept rising throughout, which is the part of the picture the October forecast did not foresee.