This is the last part of the note “FRB-US, Q4 and the Positive Supply Shocks”; part I is here, and part II is here. In this part, we assess the general equilibrium implications of a set of scenarios. The list includes: (i) a permanent productivity shock, (ii) a recessionary shock, (iii) a non-inertial Fed reaction function, (iv) a term premium shock, (v) a lower U*, and (vi) a higher core inflation forecast. As usual, we do not attach probabilities to each scenario and let the reader assess the risks.
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.
Preamble
At this link (here), the reader can find a document containing the FRB-US impulse response functions (IRFs) to a set of shocks. Each row shows three panels: the response of the output gap, the response of core PCE price inflation, and the shock itself. In the document, there are 8 shocks (FF rate shock, G shock, equity premium shock, oil price shock, productivity growth shock, productivity level shock, term premium shock, and the dollar shock) and 16 IRFs.
What will the Fed do?
All scenarios presented are in deviation from the baseline (see part I of the note here). The code color is as usual: the red lines show the current FRB-US “inconsistent” baseline, while the green lines show the specific scenario. We provide the intuition of each scenario; the details are available upon request.
Higher productivity as a way to rationalize the SEP. Figure 1 shows a scenario generated assuming a 1% permanent positive productivity shock. As expected, real GDP growth is way higher than in the baseline, and the path of the FF rate is more hawkish. (For the record: real GDP growth increases more than 1% w.r.t. the baseline because Y* itself is higher than in the baseline, and ultimately what matters is the output gap). Under this scenario, the unemployment rate (see here), remains flat around 4%. This scenario offers a hint: the model suggests that the path of the unemployment rate in the SEP requires higher growth than in the SEP, for instance via higher structural productivity. A PDF with all variables in this scenario is here.
Figure 1. Productivity shock scenario
Almost impossible to generate a recession. The news of this update is that it has become virtually impossible (at least using reasonable assumptions) to generate a recession in the model. In order to get negative (YoY) real GDP growth, we had to assume an overall negative demand shock (2pp drop in consumption, 3pp drop in investment, 1 percent drop in government spending) and a 100bps equity premium shock equivalent to a 20 percent drop in the stock market. Otherwise, the model projects positive growth going forward. Figure 2 shows the FF rate and real GDP growth in this scenario: growth goes negative in 2025, the unemployment rate (see here) reaches 5% in 2025, and the FF rate is cut to 4% by the end of 2024, and close to 2% by 2025:Q4. (for the record: as usual, the response of core PCE price inflation is muted in FRB-US, see here and here why). A PDF with all variables in this scenario is here.
Figure 2. Recession scenario
A less inertial Fed saves the labor market. Figure 3 show a scenario in which the Fed follows a non-inertial Taylor rule. The idea is that, as some people say, “the Fed can cut more aggressively that it raises rates”. In this case, the FF rate goes below 4% in 2024 (the “bizarre” look of the FF rate is precisely due to the non-inertial aspect of the Taylor rule), growth is stronger, and the unemployment rate (see here) remains low (3.6%-3.8%) throughout the entire forecast horizon. A PDF with all variables in this scenario is here.
Figure 3. Non-inertial Taylor rule scenario
Term premium shock implies lower growth, lower FF rate. A 100bps shock to the 10y yield (50bps to the 5y, and 50bps to the 30y) lowers growth and the FF rate w.r.t. the baseline, although the full impact is somehow delayed until 2025. The unemployment rate (see here) goes above 5% by the end of the forecast horizon. A PDF with all variables in this scenario is here.
(Our October 2023 note on the FRB-US and the term premium is here.)
Figure 4. Term-premium shock scenario
Lower U* implies lower projected U. In part II of this note, we showed that the model has revised down its estimate of U* (and up its estimate of Y*) to about 3½ percent. In reality, the model assumes a higher and fixed U* in the forecast (at 4.1%), similar to what happens in the SEP. Figure 5 shows a scenario assuming that U* does not increase and stays at 3½ going forward. The result is that the FF rate is a touch lower than in the baseline because Y* is higher and the output gap is more negative, which is essentially what happened in between September and December (the forecast of the unemployment rate is here). A PDF with all variables in this scenario is here.
Figure 5. Lower U* scenario
A 2½ percent core PCE price inflation does not alter the FF rate in 2024. In the final scenario (Figure 6), we assume that core PCE re-accelerates a bit and runs at 2½ percent going forward. The bottom line of this scenario is that, as expected, the FF rate is higher than in the baseline but only at the end of the forecast, while the difference in 2024 is negligible. A PDF with all variables in this scenario is here.
Figure 6. A higher core PCE price inflation forecast
