This post presents an update of the economic forecasts generated by the Federal Reserve Bank of New York’s dynamic stochastic general equilibrium (DSGE) model. We describe very briefly our forecast and its change since December 2025. To summarize, growth in 2026 is expected to be more robust, and inflation more persistent, than predicted in December. Stronger investment is the main driver for higher growth, while cost-push shocks, possibly capturing the effects of tariffs, are the key factors behind higher inflation. Projections for the short-run real natural rate of interest (r*) are the same as in December.
Note: The DSGE model forecast is not an official New York Fed forecast, but only an input to the Research staff’s overall forecasting process. For more information about the model and variables discussed here, see our DSGE model Q & A.
The New York Fed DSGE model forecasts use data released through 2025:Q4, augmented for 2026:Q1 with median forecasts for real GDP growth and core PCE inflation from the March release of the Philadelphia Fed Survey of Professional Forecasters (SPF), as well as the yields on 10-year Treasury securities and Baa-rated corporate bonds based on 2026:Q1 averages up to February 25. Starting in 2021:Q4, the expected federal funds rate (FFR) between one and six quarters into the future is restricted to equal the corresponding median point forecast from the latest available Survey of Market Expectations (SME) in the corresponding quarter. For the current projection, this is the January SME. Note that the DSGE forecasts were produced before the start of the Iran war and therefore do not incorporate its economic impact.
Once again, the economy turned out to be more resilient, and inflation more persistent, than the DSGE model had predicted in December. GDP growth in 2025:Q3 turned out to be about 1.5 percentage points higher than anticipated in the November SPF (which the December DSGE forecast used as a nowcast for 2025:Q3, as the Q3 GDP data were not available at the time due to the government shutdown). Moreover, growth in 2026:Q1, at least according to the current nowcast, is also more than one percentage point higher than the model predicted in December. The model attributes these upside surprises mainly to shocks that drive up investment. These shocks, which in the DSGE lingo are known as MEI (marginal efficiency of investment) shocks, arguably capture the strength of AI-related investment in the second half of 2025 and the beginning of 2026.
In light of the forecast misses, the model revised upward its projections for growth in 2026 by nearly half a percentage point (1.0 versus 0.6 percent). GDP growth projections are lower than they were in December for the remainder of the forecast horizon, as the level effects of the shocks on output fade (2027, 2028, and 2029 GDP growth forecasts are 0.2, 0.9, and 1.3 percent in March versus 0.8, 1.3, and 1.8 percent, respectively, in the December forecasts). The probability of a recession, defined as four-quarter output growth falling below -1.0 percent over the next four quarters, is 35.8 percent, lower than in December (37.5 percent).
Core PCE inflation in 2025:Q4 was slightly lower than expected, but the nowcast for 2026:Q1 inflation is almost half a percentage point higher than the model predicted in December. The DSGE attributes this forecast error to cost-push shocks, which possibly capture the effects of tariffs, as well as other idiosyncratic factors affecting inflation. As a consequence of this forecast error, the model revised upward its projections for core PCE inflation (2.4, 1.9, 1.9, and 2.0 percent for 2026, 2027, 2028, and 2029, versus 1.9, 1.6, 1.7, and 1.8 percent, respectively, in December).
The model’s predictions for the short-run real natural rate of interest (r*) are essentially the same as in December (1.9, 1.6, 1.3, and 1.1 percent for 2026, 2027, 2028, and 2029, versus 2.0, 1.6, 1.3, and 1.2 percent, respectively, in December). Since projections for the path of the nominal FFR are also unchanged relative to December, but inflation projections are higher, the model views the current path of policy as slightly more accommodative than it was in December.
Forecast Comparison
| Forecast Period | 2026 | 2027 | 2028 | 2029 | ||||
|---|---|---|---|---|---|---|---|---|
| Date of Forecast | Mar 26 | Dec 25 | Mar 26 | Dec 25 | Mar 26 | Dec 25 | Mar 26 | Dec 25 |
| GDP growth (Q4/Q4) | 1.0 (-3.4, 5.5) | 0.6 (-4.6, 5.9) | 0.2 (-5.0, 5.4) | 0.8 (-4.5, 6.0) | 0.9 (-4.7, 6.4) | 1.3 (-4.3, 6.8) | 1.3 (-4.3, 7.0) | 1.8 (-3.9, 7.6) |
| Core PCE inflation (Q4/Q4) | 2.4 (1.6, 3.2) | 1.9 (0.8, 3.0) | 1.9 (0.8, 3.1) | 1.6 (0.4, 2.8) | 1.9 (0.6, 3.2) | 1.7 (0.4, 3.0) | 2.0 (0.6, 3.3) | 1.8 (0.4, 3.2) |
| Real natural rate of interest (Q4) | 1.9 (0.6, 3.3) | 2.0 (0.6, 3.4) | 1.6 (0.0, 3.1) | 1.6 (0.0, 3.1) | 1.3 (-0.4, 2.9) | 1.3 (-0.3, 2.9) | 1.1 (-0.6, 2.7) | 1.2 (-0.5, 2.9) |
Notes: This table lists the forecasts of output growth, core PCE inflation, and the real natural rate of interest from the March 2026 and December 2025 forecasts. The numbers outside parentheses are the mean forecasts, and the numbers in parentheses are the 68 percent bands.
Forecasts of Output Growth
Forecasts of Inflation
Real Natural Rate of Interest
Marco Del Negro is an economic research advisor in the Federal Reserve Bank of New York’s Research and Statistics Group.
Ibrahima Diagne is a research analyst in the Federal Reserve Bank of New York’s Research and Statistics Group.
Keshav Dogra is an economic research advisor in the Federal Reserve Bank of New York’s Research and Statistics Group.
Elena Elbarmi is a research analyst in the Federal Reserve Bank of New York’s Research and Statistics Group.
Donggyu Lee is a research economist in the Federal Reserve Bank of New York’s Research and Statistics Group.
Michael Pham is a research analyst in the Federal Reserve Bank of New York’s Research and Statistics Group.
How to cite this post:
Marco Del Negro, Ibrahima Diagne, Keshav Dogra, Elena Elbarmi, Donggyu Lee, and Michael Pham, “The New York Fed DSGE Model Forecast—March 2026,” Federal Reserve Bank of New York Liberty Street Economics, March 20, 2026, https://libertystreeteconomics.newyorkfed.org/2026/03/the-new-york-fed-dsge-model-forecast-march-2026/
BibTeX: View |
Disclaimer
The views expressed in this post are those of the author(s) and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author(s).
You can follow this conversation by subscribing to the comment feed for this post.
Facts Only
The Federal Reserve Bank of New York’s DSGE model updated its economic forecasts in March 2026.
The model projects 2026 GDP growth at 1.0%, revised up from 0.6% in December 2025.
Inflation (core PCE) for 2026 is forecast at 2.4%, up from 1.9% in the previous forecast.
Higher growth is attributed to investment shocks, possibly linked to AI-related spending.
Inflation revisions are tied to cost-push shocks, potentially including tariffs.
GDP growth projections for 2027–2029 are lower than in December (0.2%, 0.9%, 1.3% vs. 0.8%, 1.3%, 1.8%).
The probability of a recession in the next four quarters is 35.8%, down from 37.5% in December.
The short-run real natural rate of interest (r*) remains nearly unchanged (1.9% for 2026 vs. 2.0% in December).
The model uses data through 2025:Q4, augmented with SPF and market expectations for 2026:Q1.
The forecasts were finalized before the Iran war and do not include its economic impact.
The model is not an official New York Fed forecast but an input for research staff.
Authors include Marco Del Negro, Ibrahima Diagne, Keshav Dogra, Elena Elbarmi, Donggyu Lee, and Michael Pham.
Executive Summary
The Federal Reserve Bank of New York’s DSGE model has updated its economic forecasts, projecting stronger GDP growth and more persistent inflation for 2026 compared to its December 2025 predictions. The model attributes higher growth primarily to increased investment, potentially driven by AI-related spending, while inflation pressures are linked to cost-push shocks, possibly including tariffs. GDP growth for 2026 is now forecast at 1.0%, up from 0.6% in December, though projections for later years (2027–2029) have been revised downward. Core PCE inflation is expected to average 2.4% in 2026, up from 1.9% previously, with slightly higher projections through 2029. The recession probability over the next four quarters has decreased marginally to 35.8%. The model’s estimates for the short-run real natural rate of interest (r*) remain largely unchanged, though the current monetary policy stance is viewed as slightly more accommodative due to higher inflation expectations. The forecasts do not account for the economic impact of the Iran war, as they were finalized before its onset.
The DSGE model incorporates data through 2025:Q4, supplemented by median forecasts from the Philadelphia Fed’s Survey of Professional Forecasters and market expectations for interest rates. While the model acknowledges recent economic resilience, it notes that growth surprises in late 2025 and early 2026 are temporary, with fading effects in subsequent years. Uncertainty remains significant, as reflected in the wide confidence intervals around projections.
Full Take
The strongest version of this narrative highlights the DSGE model’s adaptive response to unexpected economic resilience, particularly in investment and inflation dynamics. It credibly acknowledges upward revisions to growth and inflation while tempering long-term optimism with lower subsequent projections—a nuanced approach that avoids overconfidence. The explicit mention of AI-related investment and tariff-driven cost shocks provides concrete, testable mechanisms for the forecast changes, grounding the analysis in observable trends rather than vague speculation.
Pattern scan: The presentation leans heavily on technical jargon (e.g., "MEI shocks," "cost-push shocks") and confidence intervals, which could serve as a smokescreen for uncertainty (ARC-0024 Ambiguity). However, the disclosure of limitations—such as the exclusion of the Iran war’s impact and the model’s unofficial status—mitigates potential overreach. The framing of "more robust growth" and "persistent inflation" might subtly prime readers toward a specific policy interpretation, though the text stops short of explicit advocacy.
Root cause: The narrative assumes a neoclassical economic paradigm where growth and inflation are driven by exogenous shocks (investment, tariffs) rather than structural shifts. This echoes post-2008 forecasting patterns, where models struggled to account for non-linear disruptions like AI or geopolitical conflicts. The unstated assumption is that these shocks are temporary, despite historical evidence that technological and trade disruptions often have lasting effects.
Implications: If the model’s projections hold, policymakers may face a tighter balancing act between supporting growth and containing inflation, with AI investment acting as a double-edged sword. The slight increase in recession probability, though marginal, suggests lingering fragility. Human agency is implicitly framed within the constraints of monetary policy, with little discussion of fiscal or industrial policy levers.
Bridge questions: How might the model’s assumptions about the transience of AI investment shocks be tested? What alternative frameworks (e.g., behavioral economics, supply-side dynamics) could offer complementary insights? Would incorporating geopolitical risks like the Iran war materially alter the forecast, and if so, how?
Counterstrike scan: A coordinated influence campaign might exaggerate the model’s certainty (e.g., "The Fed predicts strong growth") while omitting its unofficial status or the wide confidence intervals. The actual content avoids this, explicitly noting the model’s limitations and the role of external surveys. No structural alignment with manipulation patterns is detected.
Sentinel — Human
While exhibiting some signs of human authorship, this text shows a slight lean towards being machine-generated. However, it is classified as likely human due to numerous indicators pointing towards its authenticity.