US: FRB and Q3 – Large Upward Surprise

In this note, we provide an update of the FRB-US model based on the newly released dataset. The analysis shows that both the model and the September SEP (and thus the FOMC) were surprised on the upside by the data (the one we got and our nowcast). In particular, GDP growth has exceeded expectations by more than 2 percentage point (annualized, quarterly terms) for two consecutive quarters relative to forecasts made only a few months ago.

The only way we could previously rationalize the September FOMC’s projected sequence of rate cuts was by assuming a non-inertial Taylor rule. That justification has now effectively disappeared. Under both the inertial and non-inertial Taylor rules, the model no longer supports rate cuts in 2025. Despite the meager job creation in recent months unemployment rate remains low, inflation stays meaningfully above target, and it has therefore become increasingly difficult—within the model’s structure—to justify the policy easing path signaled by the Fed (which to us is set to cut in October no matter what).

Updated baseline

The new baseline incorporates our Q3 GDP nowcast (at 3.3% QoQ), updated information on long-term rates, as well as new labor market and price dynamics data (including our Q4 nowcast/forecast for core PCE price inflation). Figure 1 shows the real-side dynamics: the red line represents the baseline under an inertial Taylor rule, while the blue line is the September SEP.

As shown in Figure 1, the model now points to a significantly stronger growth outlook compared with the Fed’s September projections. In particular, GDP growth is expected to remain above trend, reaching 2.3% (Q4/Q4) in 2025 and staying above 2% through both 2026 and 2027. This upward revision reflects the substantial upside surprises from second-quarter data and, even more so, the robust nowcast for the third quarter.

The model projects the unemployment rate to move broadly sideways through the first half of 2025. In other words, not only does the model rule out a recession, it also fails to show any meaningful increase in unemployment—contrary to the recent FOMC/staff forecast. Core inflation remains broadly consistent with the Fed’s projections for 2025–2026 but stays materially above target through 2026–2028, in line with the signal from all major models at present. As a result, the implied path for the federal funds rate is markedly more hawkish than the Fed’s September baseline.

We do not display the non-inertial Taylor rule in this update, as its trajectory is now virtually identical to the inertial version. This alignment underscores a key point: absent a recession, the balance of risks for the federal funds rate remains firmly tilted to the upside.

For reference, the path of selected financial variables consistent with Figure 1 can be seen here.

Figure 1.  Update FRB-US forecast – real-side variables.

Note: Real GDP growth and core inflation are expressed on a year-over-year basis. Core inflation refers to core PCE price inflation. The red lines are the model forecast under an inertial Taylor rule, while the blue lines are the September SEP. The baseline has been updated as outlined in the text.

The output gap

To illustrate the model’s internal dynamics, Figure 3 plots the evolution of the output gap—that is, the difference between observed output and the model’s estimated potential for the U.S. economy. The current baseline is shown by the red line, the September SEP-consistent by the blue line, and earlier SEP-consistent datasets by the other lines.

Figure 3. Evolution of output gap in FRB-US – baseline and SEP-consistent databases.

Before the latest revision, the baseline and SEP-consistent projections were broadly aligned. Starting with the current dataset, however, the baseline now projects a positive output gap—already positive at present and accelerating further into the fourth quarter. The gap remains positive throughout 2026–2027 before turning negative only in 2028.

By contrast, the SEP-consistent dataset continues to show a negative output gap, as indicated by the blue line in Figure 3. This discrepancy explains much of the divergence between the policy rate path implied by our baseline and that embedded in the September SEP. To further clarify the underlying mechanics, Figure 4 plots the level of real GDP—observed and projected by the model (red line)—against the model’s estimate of potential output (blue line). As the chart shows, the model suggests that the U.S. economy is currently in a phase of reacceleration and is expected to remain above potential until the end of 2027.

Figure 4. Real GDP level and estimated potential output.

In short, despite recent softening in nonfarm payroll data and signs of cooling in the labor market, the model finds little evidence of an impending recession. On the contrary, it now points to a renewed acceleration in U.S. economic activity.

Our comment and judgemental calls

The first point to keep in mind is that none of our models currently show inflation converging to target within the medium-term horizon. The reason the Fed staff—and consequently the FOMC—project convergence is purely mechanical: the staff assumes that the tariff shock is temporary and that underlying inflation (pi*) is anchored at the 2 percent target. Both assumptions are generous. They are not implausible, but they remain optimistic relative to the model evidence. Moreover, while the probability of a recession is not zero—and may indeed be somewhat above its unconditional mean— the FRB model struggles to produce a recession scenario. On the contrary, the latest runs suggest a reacceleration in activity rather than a slowdown.

That said, market sentiment tells a different story. From our conversations with clients, it is evident that bond traders remain concerned about the risk of a meaningful slowdown in the U.S. economy, or even a potential recession. At the same time, the equity market appears to be discounting a very different outcome—one consistent with sustained nominal growth and the absence of a downturn.

We therefore remain somewhat puzzled by this divergence between bond and equity market narratives. It is not yet clear which of the two—growth or inflation—will dominate in the coming months. For now, we remain conservative in our calls, as additional data and time are needed to assess which side of the mandate will ultimately prevail. In any case, to us the Fed is set to cut in October no matter what.

Contact

Want something more tailored?

We provide tailored consulting on ad-hoc projects.