The Fed Staff is the real hawk at the FOMC table right now. The December 2022 FOMC minutes reveal that the staff has revised up its estimate of the natural rate of unemployment. This is a hawkish signal that should not be ignored. We assess its implications.
Our comment
The Fed staff has assumed a higher U* in the Tealbook. According to the December 2022 FOMC minutes “[…] the staff assumed a slower pace of decline in the natural rate of unemployment over the near term in response to recent estimates suggesting that job-matching efficiency was not improving as fast as previously anticipated. With all these changes, output was expected to move below the staff’s estimate of potential near the end of 2024—a year later than in the previous forecast—and to remain below potential in 2025. Likewise, the unemployment rate was expected to move above the staff’s estimate of its natural rate near the end of 2024 and remain above it in 2025.”
The changes implemented by the Fed staff have hawkish implications. According to the new staff forecast, the labor market is not expected to put downward pressure on consumers’ prices before 2025. While the staff forecast for core inflation converges back to target by 2025, it does so mainly as a result of factors outside the Phillips curve framework rather than by the contribution of the labor market. In other words, the staff sent a message to the FOMC along these lines: “We think that disinflating the US economy is a bit harder than last round, as it requires a marginally higher unemployment rate (and therefore a higher path of the FF rate). Core inflation can converge back to target, but it requires help from supply-side factors outside the Phillips curve framework”. Conditional on this, it is easy to explain why the FOMC was very hawkish at the December meeting and why, in our view, it will remain as such in the next couple of rounds.
Model-based forecast marginally higher. We have run our “main” Phillips curve model conditional on the new path of U*. Figure 1 shows the new results. Unsurprisingly, the forecast is a touch higher than the previous run (December 23rd), as the downward pressure from slack kicks in only at the end of the forecasting horizon. The model expects core PCE price inflation at 3.4% in 2023, 3.3% in 2024, and 3.1% in 2025.
Figure 1. “Main” model forecast – core PCE price inflation
No pivot in January, very low chance in March. In our experience, when the Fed staff revises the stars (i*, U*, and especially pi*), the FOMC takes it seriously. As previously discussed, in our view the near-term incoming data do contain signs of disinflation (and more hope comes from soft data). But until the Fed staff will receive dovish signals from the models, we continue to think that it will be hard for the FOMC to pivot. At the moment, no model is suggesting inflation will quickly return to target and the models are unlikely to send different signals until 2023:Q2.
All told, very hard to see how the FOMC can pivot at the January meeting.
No matter how good next CPI print will be.
(For the record, today, Minneapolis Fed President Neel Kashkari has published an analysis titled “Why We Missed On Inflation, and Implications for Monetary Policy Going Forward”. The analysis contains nothing new on a technical level: Phillips curve models capture the demand-side of the economy only partially but, in any case, better than the supply-side. Because Covid triggered shocks on both sides, the models were able to capture/predict only part of the story. Adapting the models to improve the out-of-sample performance given the Covid-shock is the next challenge in the academic literature. Having said so, Kashkari’s note reinforces the messages contained in the minutes: FOMC members are committed to raise rates and keep them elevated, and they seem genuinely unsatisfied with the signs of progress in the data so far).