We use FRB-US to assess how the September SEP can look like. While the incoming data have been bumpy, they did not trigger a significant revision to the model forecast compared to the previous run (here). Conditioning on our Q3 nowcast, the model expects about 50bps of FF rate cuts this year. Compared to the June SEP, the model expects slightly stronger real GDP growth, an unemployment rate a touch higher in 2024 (to 4.2%), and a similar path of core PCE (just a touch lower to 2.7% in 2024 and a touch higher in 2026). In other words, the model suggests that the new SEP could be something in between the June and the March SEP (possibly closer the latter). As for the FF rate, the model signals upside risks around the current market pricing (which is hard to validate for 2024, unless one assumes a much lower inertial coefficient in the Taylor rule or a judgmental accelerated Fed start (see below for details)).
We also use the model to run a set of scenarios, including: (i) an accelerated start of the Fed with 125bps of cuts in 2024, (ii) a non-inertial Taylor rule, (iii) a recession scenario, (iv) alternative R*, (v) a positive productivity shock, and (vi) a term premium shock.
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 new assumptions
We are assuming that real GDP growth (QoQ saar) will be 2.0% in Q3. This is marginally higher than what the model would forecast (1.7% QoQ saar), as the incoming data seem to point to another strong quarter. We are also assuming that the unemployment rate will average 4.2% in Q3, that core PCE price inflation will be 2.0% (QoQ saar), that the 10y yield averages at 4.0%, the 5y yield at 3.9% and that the 30y yield averages at 4.3% in Q3. Finally, R* is assumed at 0.8% in line with the latest SEP.
The updated baseline and the stochastic simulations
The incoming data since our previous note have not modified the baseline. As in recent quarters, the model is signaling that: (i) real GDP is expected to be broadly in line with the SEP, possibly a bit stronger, (ii) the unemployment rate is expected to trend higher in the near-term, as a result of a positive supply-shock, and (iii) core PCE price inflation can easily remain a bit above target. Putting everything together, FRB-US expects the FF rate at 4.85% in 2024:Q4 (for a cumulative 50bps of cuts in 2024), 3.8% in 2025:Q4, and 3.2% in 2026:Q4. Figure 1 shows the updated baseline, while Figure 2 shows the stochastic simulations around the 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 takeaway is the same of recent quarters. According to FRB-US, it is pretty unlikely to have a recession, and core PCE can easily remain above target at the end of the medium-term.
(The PDFs containing the numbers of Figure 1 and 2 below, and the path of the financial variables implied by the SEP and the baseline can be downloaded here, here, and here).
Figure 1. Latest SEP forecast (solid blue line) and FRB-US forecast (dashed red line)
Note: Real GDP growth and core inflation are expressed as YoY. Core inflation is core PCE price inflation. The solid blue lines show the latest SEP. The dashed 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.
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.
Scenarios
We have run a set of scenarios. We present the results of 6 of them. We remind the reader that we can simulate any scenario on-demand. The reader can also see the impulse response functions to a set of shocks here (“PING” program in FRB-US). As usual, in each figure below, the red dashed line is the current baseline, while the solid gray line is the simulated scenario.
- Scenario #1: accelerated start. In this scenario we forced the Fed to deliver an “accelerated start”: 50bps of cuts in September, 50bps in November, and then 25bps until we reach 3.25% next year. The impact is visible on growth which remains well above 2%. The unemployment rate (not shown) declines in 2025 and 2026 and core PCE price inflation remains above target. This is possibly the strongest argument the doves can use to push for a 50bps cut in September.
- Scenario #2: non-inertial Taylor rule. In this scenario we have lowered the inertial coefficient of the Taylor rule to zero. According to the model, in this scenario the FF rate would need to drop below 4% in 2024 and remains around this level throughout the forecast horizon. The result of this dramatic policy shift would be sustained growth, with the unemployment rate gradually trending down to 3.7% by 2026, and core PCE price inflation in modest reacceleration.
- Scenario #3: a recession. In this scenario we forced the YoY of real GDP growth to be (mildly) negative at some point in 2025. In order to achieve this goal, we had to fold-in a negative path for private consumption and investment starting from the current quarter assuming exogenous shocks. Even so, as mentioned, the resulting recession is very mild. In this scenario the unemployment rate increases to 5.6% in 2026 and the Fed cut aggressively until 2026.
- Scenario #4: different R*. In this scenario, we have assumed different R*s. Specifically, we have assumed a much higher R* (1.3%) and a much lower R* (0.3%) than in the baseline. The first scenario is shown with a dashed green line in the figure below, while the second scenario is shown with a solid gray line. As expected, assuming a higher (lower) R* results in a path of the FF rate above (below) the baseline. In any case, even assuming such different equilibrium rate, does not affect growth much. Indeed, real GDP growth remains above potential going forward in both scenarios.
- Scenario #5: a productivity shock. In this scenario we have assumed a higher productivity growth (50bps above the baseline). As expected, this scenario is more favorable: growth is stronger than in the baseline, the unemployment rate trends down and the FF rate remains above the baseline.
- Scenario #6: a term premium shock. In this scenario we simulated a one-time 100bps shock to the 10y yield (75bps to the 5y yield, and 35bps to the 30y yield). According to the model, this shock lowers growth, although there would be no recession. Core inflation is a bit lower and the unemployment rate increases to 4.5% in 2026. Because growth is lower and the unemployment rate is higher than in the baseline, the Fed cut more in this scenario.
