Fed FRB/US, rebuilt
FRB/US
Traces how a US funds-rate or fiscal shock moves output, unemployment and inflation, quarter by quarter.
Good for
- Monetary and fiscal shocks through the Board’s own equations
- Quarterly paths against the Board’s October 2026 baseline
- Wage and hours paths for the incidence bridge (experimental)
Not for
- PolicyEngine reforms:
score_reformrefuses it (limits) - Forward guidance: no model-consistent expectations in the hosted tools
- Anything about the UK: OBR emulator
A worked example
The Fed raises the funds rate by one percentage point for one quarter (2026Q3, the first projected quarter of the October 2026 LONGBASE), under the default inertial Taylor rule.
- Deviations from the Board’s baseline.
- Against the Board’s published run (a 2020Q1 shock): core PCE 0.99×, output gap 1.03×, unemployment peak 1.28×, larger from later start dates (the comparison).
- Source:
results/window_2026q3_preview.mdat162cfe6; this 2026Q3 run is not cross-checked against pyfrbus (the solver match is on the 2026Q1 run).
How far to trust it
Three checks, weakest included; each opens its evidence.
- Solver, 100bp funds-rate shockFed’s pyfrbus 1.0.0 and 1.1.1; gate 1e−61.4e−8 and 1.8e−8pass
- 100bp shock, size of responsesBoard’s published run on its own design (2020Q1 shock); gate ±10%0.96–1.04× the Board’spass
- Inflation effect of the 2022–23 hikesFRBSF Economic Letter 2025-03, about 3pp (model economics, not implementation)−0.12ppfail
Trust the solver, not the economics. The match shows the Board’s equations are solved correctly, not that they are right (all limits).
Run it
The worked example above, three ways: a 100bp funds-rate tightening for one quarter.
In FRB/US, raise the federal funds rate by 100bp for one quarter from 2026Q3 under the inertial Taylor rule, and show real GDP, unemployment and core PCE inflation each quarter.The hosted server calls frbus_shock; no install. Connect
pe-macro frbus-shock --var rffintay_aerr --shock 1.0 --start 2026Q3from frbus import Frbus, load_data
start, end = pd.Period("2026Q3"), pd.Period("2031Q2")
# load LONGBASE, set the demo fiscal switches, init_trac: see /frb-us/code#code-5
with_adds.loc[start, "rffintay_aerr"] += 1
sim = model.solve(start, end, with_adds)With the microsimulation
Its labour-market path tells the microsimulation who bears a shock.
frbus_shock_incidence scales US employment income with the wage and hours deviations, so federal and state taxes, SNAP and the EITC show who bears it (every join in the suite).
Uniform scaling is an experimental illustration. Real incidence concentrates in job losers.
Model cardinputs, outputs, horizon, data, access
- Answers
- How US output, unemployment, inflation and the funds rate respond to a monetary or fiscal shock
- Inputs
- One reviewed add-factor shock: a lever, a size and a policy rule
- Outputs
- GDP, unemployment, inflation, prices, federal funds rate
- Horizon
- Quarterly paths, read as deviations from the Board’s baseline
- Data
- The Board’s
model.xmland October 2026 LONGBASE - Uncertainty
- Not comprehensive: the hosted shock returns one deterministic path
- Access
- Hosted raw shocks, CLI, Python package (Code)
- Runtime
- About 3 seconds cold; under a second warm
- Joins the microsimulation
- Wage and hours paths scale US employment income (the join above, experimental)
- Status
- Replicates the Fed's pyfrbus (1.0.0 and 1.1.1) on one 100bp funds-rate shock, within a 1e-6 tolerance; a solver check, not a test of the economics
- Cannot answer
- PolicyEngine reforms; model-consistent expectations, so no forward guidance; anything about the UK