Michael Strain pushes back against the emerging consensus that an economic 'soft landing' is likely in the US

Optimism is growing that the United States can avoid a recession. A Wall Street Journal survey of economists in July found that only 54% expect a recession in the next 12 months, down from 61% in April. Economists at Goldman Sachs have lowered their estimate of the probability of a recession to 20%. Following the release of encouraging consumer price index (CPI) data on July 12, investors have grown more confident that inflation can be tamed without sacrificing economic growth.

This emerging consensus may end up being correct. We are all rooting for a “soft landing” in which inflation continues to glide down toward the US Federal Reserve’s 2% target without the economy shrinking. But I worry that a recession in the next year is much more likely than not.

After all, underlying inflation is still double the Fed’s target, and its downward trend has not made meaningful progress in 2023. Given this reality, it was quite confusing to see so many commentators unfurl a “Mission Accomplished” banner following the latest data release.

True, it is good news that monthly core CPI inflation (excluding food and energy prices) fell to 0.2% in June, down from 0.4% or higher each previous month in 2023. But the Fed needs to be sure that inflation is trending down toward its target rate, and one month’s data does not make a trend. The core CPI index has experienced big drops in a single month before (including in July 2022), only to rise again. Though core CPI grew in June at half its 2023 trend rate, it still rose 4.8% – more than double the Fed’s target – over the past 12 months.

More to the point, core CPI is not the measure the Fed is most interested in. Rather, it focuses more on the core personal consumption expenditures price index, which has shown no improvement in 2023. And even if it does register an improvement for June, that, too, will be one data point, not a trend.

Though the Fed has hiked its policy interest rate by 500 basis points, much of the effect of those increases may have already hit the economy. Monetary policy slows the economy by tightening overall financial conditions – that is, by reducing stock prices and increasing longer-term interest rates and the value of the dollar. When the Fed started raising rates last year, financial conditions behaved as one would expect. But conditions have not tightened over the course of 2023, which suggests that those previous rate increases may have already been absorbed.

Despite the Fed’s rapid rate hikes from near zero to around 5.1%, monetary policy may still not be all that restrictive. According to my calculations, the real (inflation-adjusted) interest rate is around 1.5%, which is one percentage point higher than the Fed’s estimate of the neutral real policy rate (which neither stimulates nor reduces economic activity). Prior to previous recessions, the real rate has been higher.

While current market pricing suggests that the Fed will increase the federal funds rate once more this cycle, I think that is optimistic. Between stubborn and high underlying inflation, financial conditions that aren’t tightening, and real interest rates that are lower than is typical before a significant economic slowdown, there are ample reasons for the Fed to raise rates more than economists and investors currently seem to expect. If that happens, the risk of recession will increase.

Moreover, the lower the inflation rate falls, the more likely it is that the unemployment rate will climb into recessionary territory. So far, inflation has fallen without the unemployment rate rising. But standard models of the relationship between unemployment and inflation suggest that the costs of disinflation rise as the rate of inflation falls.

From April 2021 until May of this year, consumer prices grew faster than average wages, which meant that businesses could still raise prices faster than their labour costs increased. But that is no longer the case. Now, the businesses most exposed to labour costs will face increasing pressure to resort to layoffs if consumer price inflation continues to fall.

Finally, historical experience advises against predictions of a soft landing. In a soft landing, the unemployment rate would increase as economic growth fell below its potential rate, but it would remain out of recessionary territory. In the past, however, when the unemployment rate has started to rise a little, it then goes up a lot more. Since World War II, a 0.5 percentage point increase in the unemployment rate in a given year has been followed by a two-point increase. This could be because recessions are driven in large part by a loss of confidence among businesses and households about the near-term future. When managers see that other companies are laying off workers, they are more likely to worry about maintaining profits and lay off some of their own.

Yes, the odds of a soft landing have certainly increased in recent months. The outlook for shelter inflation will put downward pressure on core inflation, and demand could fall with the reduction in bank lending and the depletion of excess household savings. The labour market is indeed cooling. All these factors could well put underlying inflation on a downward trend. In addition, to the extent that labour-market tightness was driven more by high levels of job openings than employment, cooling could be achieved through relatively small increases in unemployment.

But the Fed should not stop raising rates until there is clear evidence that core inflation is on a path to its 2% target. That evidence does not exist today, and it probably will not exist by the time the Fed meets in September. The longer it takes for that evidence to materialise, the higher interest rates will go, and the smaller the chance of a soft landing.

Michael R. Strain, Director of Economic Policy Studies at the American Enterprise Institute, is the author, most recently, of The American Dream Is Not Dead: (But Populism Could Kill It) (Templeton Press, 2020). This content is © Project Syndicate, 2023, and is here with permission.