The March 2023 FOMC Minutes reveal very important developments: the Fed staff is now forecasting a recession in the second half of this year followed by a recovery in 2024 and 2025. The recession (in our estimates) raises the unemployment rate more than in the latest SEP. Also, the Fed staff raised its 2023 core PCE price inflation forecast but lowered it in 2024, as a reflection of the recession.
We take this development in a positive way. The Fed staff is finally confessing: it takes a recession to bring inflation back to target. (For the record: this is what all models have been saying for more than a year, so we are happy to see the Fed staff admitting reality.) We explain what, in our view, the Fed staff has changed in the Tealbook forecast and what are the implications for the model-based forecast and the path of the FF rate. Overall, there are still some upside risks around the Fed staff forecast but we are finally going in the right direction.
The FOMC minutes and what the Fed staff has done
According to the minutes: “[…] the staff’s projection at the time of the March meeting included a mild recession starting later this year, with a recovery over the subsequent two years. Real GDP growth in 2024 was projected to remain below the staff’s estimate of potential output growth, and then GDP growth in 2025 was expected to be above that of potential. […] The level of real output was projected to move below the staff’s estimate of potential output in early 2024, more than a year sooner than in the previous projection. Likewise, the unemployment rate was projected to rise above the staff’s estimate of its natural rate early next year.”
What we think the staff has written in the Tealbook
Given the upward revision to U* that the staff implemented at the December FOMC (see minutes here), in our view the staff is operating under the assumption that U* is around 5%. For the record, in the latest (unconstrained) run of FRB-US, U* is estimated at 4¾ percent, in line with our “OUI” model. This implies that the Fed staff is forecasting the unemployment rate to move up by (at least) 1.5pp in the next 8-10 months, way more than in the SEP. In fact, we suspect the forecast assumes the unemployment rate to go a bit above 5% in 2024. As for 2025, because the staff expects real GDP growth to be above potential, by Okun law we suspect the staff is forecasting an unemployment rate ticking down over the last year of the forecasting horizon (say, from a peak of 5.2% to 5.0% in Q4:2025). In a sense, this is non-surprising because it is in line with the “unconstrained” FRB-US forecast we have presented here and here. To the best of our knowledge and memory, this is one of the most aggressive calls of the Fed staff outside recessions.
Implications for the model-based forecast
A recession triggers a downward revision of the forecast. Still, not at target. We run our models under the same assumptions (as inferred from the minutes) of the Fed staff. The idea is to assess the risks around the Fed staff judgmental forecast using model-based evidence. In this case, rather that showing the usual “fan” chart, we show the evolution of the forecasts over time. Figure 1, 2, and 3 compare the Q4/Q4 forecasts for core PCE price inflation; our “main” model forecast is shown as a black line in each chart. We take our “main” model seriously given that the first time the model delivered a 3¼% -3½% forecast for 2023 (that is, what the Fed staff and the FOMC are currently forecasting) was in March 2022. Because we have now folded in a (mild) recession, our model has revised down its forecast in each year (caveat: what the reader sees is a first pass, the forecast can change in the next few days). The model now expects core PCE prices to expand 4.1% in 2023, 3.5% in 2024, and 3.1% in 2025. In other words, a mild recession helps brining the model forecast closer to target, but the Fed staff forecast remains about 1pp below the model in 2024.
Figure 1. Evolution of the 2023 (Q4/Q4) core PCE price inflation forecast (%)
Figure 2. Evolution of the 2024 (Q4/Q4) core PCE price inflation forecast (%)
Figure 3. Evolution of the 2025 (Q4/Q4) core PCE price inflation forecast (%)
Conclusion and implications for the FF rate
We expect the hike in May to be the last one. What happened at the March meeting is a very important development. For a long time, we have argued that it was virtually impossible to see inflation converging to target without a recession (“immaculate disinflation” has always been a dream more than a forecast..). If the FOMC suspects that a recession is on its way and that the unemployment rate can rise 1.5-2pp within 12 months, it will probably be very careful with rates. At the same time, inflation remains a problem (reminder: MoM of core CPI today was still close to 5% at annual rate). Whether a mild recession will be enough to be around target in 2024 remains a question mark in the models, but with the staff calling a recession, the tone at the FOMC has probably changed. All told, in our view, the only thing that makes sense given the constraints on both sides of the mandate is to raise the FF rate one last time and stay there.