Preamble: we are not going to circulate a comment to the Governing Council meeting because there were no surprises at all.
In this note, we show that for the first time FRB-US is a bit more dovish than the latest SEP, and we provide stochastic simulations around the baseline. In part II of this note, we explain why the model is more dovish than the SEP. In a nutshell, this happens because the model revised up the level of potential output, driven by higher labor productivity growth and higher employment/participation. Consequently, the output gap is smaller, in fact negative even if there is no recession (a feature of the SEP-consistent database itself, although to a less extent). Finally, in part III we provide a set of simulations, including a recessionary shock and a less inertial Fed.
Recent FRB-US notes are “FRB-US and pi*”, “FRB-US, Q3 and the Term Premium”, and “FRB-US and the Fiscal.. Mess! – Part I and Part II”.
If the reader has questions, please direct them to our FRB-US specialist: Tilda Horvath (tilda.horvath@underlyinginflation.com). Tilda has programmed FRB-US at the Board and managed the model for almost 20 years. We remind the reader that we run ad-hoc scenarios using FRB-US free of charge. Do not be shy.
The update
This note is based on the newly released FRB-US SEP-consistent database; output variables have been updated through Q4, and inflation variables through Q1 using our nowcast. As usual, the “inconsistent” forecast removes the add-factors put by the Fed staff in order to match the latest SEP. All GDP components have been updated through Q4 following today’s BEA release. For the unemployment rate and government yields we have updated the series through Q4, while for core PCE we are nowcasting a 2.4% QoQ (saar) in 2024:Q1.
Note: in the December run of the model (here), we wrote: “according to FRB-US, growth can surprise again to the upside (the model now expects real GDP growth at 3.0% in Q4 QoQ saar), before decelerating in the second half of 2024.” Therefore, while the strong Q4 number was an upward surprise for the market, it was not for FRB-US. For this reason, the new baseline is broadly similar to the December one.
The new baseline
FRB-US is now a bit more dovish than the SEP. Figure 1 shows the December SEP (blue lines) vs the latest FRB-US model forecasts (red lines). There are three main takeaways. First, the model expects the US economy to remain stronger than in the SEP this year, a circumstance we had already flagged in the December and October updates. Second, the inflation forecast is nearly identical to the SEP; nothing has changed compared to the December update. Third, the model projects an increase of the unemployment rate which is a bit steeper than in December (the unemployment rate reaches 4.5% by the end of the medium-term). Consequently, for the first time, the path of the FF rate is more dovish than in the December SEP. According to the model, the Fed will cut the FF rate to 4.5% by 2024:Q4 and to 2.5% by the end of 2026.
Figure 1. SEP forecast (blue line) and FRB-US forecast (red line)
Note: Real GDP growth and core inflation are expressed as YoY. Core inflation is core PCE price inflation. The blue lines show the latest SEP. The red lines show the “inconsistent FRB-US” forecast (the current baseline), that is the model-based forecast removing the “add-factors” put by the Fed staff to match the latest SEP, updated as specified in the text.
How come the unemployment rate goes above the SEP (and the December run) if GDP growth is strong and revised up (marginally)?
The level of potential output (Y*) is higher, according to the model. We will discuss in detail this feature of the forecast in part II of this note. For the time being, we stress this happens because the model (and the SEP itself implicitly) revised up the level of potential output (Y*), driven by both an upward revision of labor productivity and the level of employment/participation. Consequently, (and this might be the counterintuitive part) the output gap turns negative sooner, including in the SEP-consistent database, and the unemployment rate trends higher. Having said that, while the unemployment rate is projected higher, the level of employment does not fall, consistent with an environment of positive but low growth (i.e. more people enter into the labor force, job creation is positive but anemic so the unemployment rate increases while the level of employment goes sideways).
(Note: for the record, the level of real GDP and Y* in history and FRB-US forecast can be seen here. The reader can appreciate why the current environment is often hard to interpret by looking at the evolution of the output gap in the figure – we will discuss it in details in part II).
What about the financial variables?
FRB-US expects the 10y yield at 4.3% in 2024:Q4. The forecast for the 5y, 10y, 30y yield, and the real expected return on equities can be seen here (as usual, the blue lines indicate the SEP-consistent forecast, while the red lines show the “inconsistent” forecast). The bottom line is that the model expects the 10y yield at 4.3% at the end of 2024. This forecast continues to be higher than in the SEP. We will explain in part II why.
Stochastic simulations
Figure 2 shows the stochastic simulations around the new baseline. Following standard procedures of the Fed staff we draw randomly from historical errors (1970q1 to 2017q4) for 54 variables and 5,000 replications. The simulations show that: (i) the probability of a recession is now quite low, and (ii) conditional on the estimated persistency, the model now sees a good chance of getting back to the inflation target by the end of the medium-term. The full set of stochastic simulations can be seen here (please, get in touch if interested in the details).
Figure 2. Stochastic simulations around the baseline.
Note: Real GDP growth and core inflation are expressed as YoY. Core inflation is core PCE price inflation.
Conclusion
For the first time, FRB-US is more dovish than the FOMC. The big news of today’s update is that for the first time in this cycle the model is now a bit more dovish than the FOMC. The good news is that according to the model the US economy is experiencing a positive aggregate supply shock. In the next part of this note, we will provide the details behind the model forecast and the evolution of Y*.