Can economists get anything right?

The economics profession in general did a terrible job of forecasting the persistent surge in inflation in 2021-2. As Powell pointed out at his latest press conference: “The good ship Transitory was a crowded one, with most mainstream analysts and advanced-economy central bankers on board.” Sitting in my lifeboat, I thought the Fed was about six months late in hiking rates—underlying inflation had been running well above the Fed’s target for about eight months before the Fed responded in March 2022 (Chart)

However, reading the business press one would think the economic consensus continued to misfire in the last two years. The main complaint is that these “dismal scientists” were too pessimistic, pushing back against a soft landing and arguing that getting inflation down would require a big rise in the unemployment rate.

The problem with this argument — like a lot of the “news” these days — is it simply is not true. Consensus forecasts for unemployment and inflation have been relatively accurate in the last two years. Consider the consensus forecasts from the Survey of Professional Forecasters. At the height of pessimism on the outlook, the median SPF forecaster expected the unemployment rate to peak at an historically low average of 4.4% in 2025. More pessimistic forecasters thought that rise to come with a brief, very mild recession.

The consensus also saw only about a 10% chance of the unemployment rate averaging above 6% in 2025. Indeed, the consensus saw a better chance of a great outcome, seeing about a one-in-three chance that the unemployment rate would remain below 4%. There were only two unemployment pessimists: one with a peak unemployment rate over 5% and one over 6%.

Equally striking was that this modest rise in the unemployment rate was expected to solve the Fed’s inflation problem. For 2023, 2024 and 2025 the median Q4 over Q4 core PCE inflation forecast was expected to fall from 3.7% to 2.3% to 2.0%.

Here we are two years later and, so far, the consensus looks remarkably prescient. The unemployment rate has moved up to about 4.2% and is expected to drift a bit higher. In July core inflation was running at 2.6% y-o-y. On Friday the consensus looks for a benign 0.15% increase in the core. With a few more benign readings, the consensus may get the 2.3% 4Q/4Q number it was predicting more than a year ago. What the consensus got wrong was that a lot of the rise in the unemployment rate would come for labor force growth rather than payroll weakness.

The Fed’s recent forecast record is almost as good. In the summer of 2023, the median FOMC member was looking for the unemployment rate to peak at a 4.5% annual average in both 2024 and 2025. Core PCE inflation was expected to dip from 3.9 to 2.6 and then 2.2. Not bad for government work.

Dismal science?

Why has the consensus been unfairly criticized? I think the alleged extremely negative view of economists was a strawman. Only a few high-profile economists expected a true hard landing. Exhibit A was former Treasury Secretary Larry Summers. At about the same time as the above SPF survey, he attracted a tremendous amount of press, highlighting a very high risk of a big rise in the unemployment rate. He said, “We need five years of unemployment above 5 percent to contain inflation—in other words, we need two years of 7.5 percent unemployment or five years of 6 percent unemployment or one year of 10 percent unemployment.”

It is worth noting that even Summers didn’t push this argument very hard. Rather, he pointed out that it is consistent with typical estimates of the “sacrifice ratio”—the amount of above-normal unemployment it takes to lower inflation by a percentage point. Still, the point is that no one in the Survey of Professional forecasters expected such a bad outcome.

Why did economists generally get it right? Again, the myth is that economists were divided into two camps—one seeing inflation as purely temporary and others seeing it as mainly persistent. In reality, most economists thought most of the inflation was transitory but some was not. The consensus was that fixing supply chains, getting people back to work and the Fed cooling—but not crushing—the economy would be enough to bring inflation under control.

Looking ahead, a relatively soft landing remains the most likely scenario. The main risk of a recession was always that inflation prove more persistent than expected, forcing the Fed to create a hard landing. The rise in the labor force has also helped. Now the Fed can focus on making sure that soft landing continues.

My advice to my fellow dismal scientists: take a bow, but be ready to duck the rotten fruit.