US: FRB-US Additional Scenarios

Preamble: We have taken note of Powell’s statement that “tariff inflation is likely a one-time price increase, adding 0.5% to 1.0% to inflation.” We respectfully disagree. In our estimates, tariff-related inflation would account for at most around 0.5 percentage points. A figure closer to 1 percent lies well outside any reasonable confidence interval and reinforces our view that the Fed staff remains, once again, overly optimistic regarding inflation dynamics.

We also heard Powell state: “You can have a series of these supply shocks and that can lead the public generally—businesses, price setters, households—to start expecting higher inflation over time. Why wouldn’t they?”. In this case, we agree. Indeed, we believe that the post-Covid environment is increasingly characterized by large supply shocks rather than demand-driven dynamics. For this very reason, we are concerned that the Fed staff may be underestimating the situation.

We have received several comments on our previous analysis based on the FRB-US model, along with multiple requests for additional simulations under a more constrained set of assumptions. In this note, we present a number of alternative scenarios.

The conclusion remains unchanged from what we have communicated in recent days: across a wide range of simulations, the Fed remains on hold—implying neither rate cuts nor rate hikes. A rate hike would only become plausible in the event of a meaningful acceleration in core inflation above 3.5 percent, approaching 4 percent, combined with a stable labor market. Should oil prices rise significantly further, the shock could become marginally dovish if accompanied by an appreciation of the dollar. Overall, nearly all scenarios point to a Fed that remains on hold in 2026.

Finally, we also show how the model behaves under modified Taylor rules, confirming that the overall conclusions remain unchanged.

(Riccardo’s view here) In a nutshell: we never bought the 3-4 cuts before the Iran war, we did not buy the possible hike and we do not buy the possible cuts now.

Another large oil shock scenario

In this case, we simulate two scenarios. The first (green lines) assumes oil prices peak at around $120 per barrel, while the second (grey lines) assumes a peak of approximately $180 per barrel. The red lines represent the current baseline of the model. (A natural caveat to these scenarios is that the FRB-US model is not designed to capture non-linear shocks. Therefore, pushing the oil shock to these magnitudes may result in dynamics that in the real world could be significantly more pronounced on both sides of the Fed’s mandate.)

The key takeaway is that, according to the model, if oil prices were to continue rising, inflation would clearly accelerate (and possibly by more than the model currently estimates). At the same time, a negative impact on the labor market could no longer be ruled out. As a result, the net effect on the federal funds rate remains modest.

It is worth noting that these scenarios are not inconsistent with what we have communicated in recent days. Specifically, the model also indicates that if the labor market were not adversely affected by higher energy prices, the Fed could consider a rate hike once core PCE rises above 3.5 percent, and would likely be compelled to act if it were to exceed 4 percent.

It remains true that a rate hike can be generated within the model also by assuming a non-inertial Taylor rule (results available here). That said, we continue to view this scenario as relatively unlikely at this stage, and we expect the Fed to remain broadly inertial in its policy approach.

Finally, adding persistence to the shock and assuming an appreciation of the dollar results in a more dovish Fed. The results are reported here (link), under the assumption that the shock—peaking at $140 per barrel—persists for one quarter (green lines), two quarters (grey lines), or three quarters (purple lines).

Figure 1.  FRB-US: The Fed hike scenario (green lines) vs current baseline (red lines)

Note: Real GDP growth and core inflation are expressed on a year-over-year basis. Core inflation refers to core PCE price inflation.

Using the Ugap

One of the features of the model at this stage—both in the SEP-consistent database and in the baseline—is that the output gap is estimated to be negative starting in 2025 and remains so throughout the forecast horizon (see here).

The reason for this is straightforward: over the past two years, the model has revised upward its estimate of potential output. (For reference, while the model incorporates population variables, the realized data on real GDP have driven these revisions.) This is the main reason why, in the baseline forecast, the federal funds rate declines even as the unemployment rate falls and core inflation remains above target.

(If the reader is wondering how the unemployment rate can decline while the output gap is negative, the answer is that the decline in labor force participation observed in the data more than offsets the effect of the output gap.)

For this reason, we have experimented with replacing the output gap with the unemployment gap in the Taylor rule. Figure 2 presents the results. The takeaway is straightforward: if the Fed were to base its decisions solely on the unemployment rate (or unemployment gap), the federal funds rate would remain on hold through the end of 2028.

Figure 2.  FRB-US: Ugap in the Taylor rule – Baseline (red lines) vs scenario (green lines).

Note: Real GDP growth and core inflation are expressed on a year-over-year basis. Core inflation refers to core PCE price inflation.

Conclusion

As always, we are not concerned with what the market is pricing or what the sell-side is suggesting. We follow the models. The conclusion is clear: the Fed is on hold. The only factor that could push the Fed toward a rate hike would be an acceleration in core PCE inflation above 3.5 percent (and toward 4 percent). On the other hand, the bar for a rate cut is equally high and would require the unemployment rate to rise by several tenths.

Contact

Want something more tailored?

We provide tailored consulting on ad-hoc projects.