The "Fed reaction function" is the implicit rule that maps incoming economic conditions — chiefly inflation and labor-market slack — onto the Federal Reserve's policy decisions (the federal funds rate, and in crises, the balance sheet). It is not a published equation the FOMC obeys; it is a model of the Fed that markets infer and continuously update. The core tension is that the Fed's actual behavior is discretionary and judgment-driven, yet markets must compress that judgment into a tradeable forecast — so a data release matters less for what it says about the economy than for how it shifts the market's estimate of how the Fed will respond. Understanding the reaction function is therefore the difference between knowing what the data was and knowing what it means for rates.

How it's modeled / formed

The canonical formalization is the Taylor rule (John Taylor, 1993), the benchmark reaction function. Its original form:

> r = r\ + π + 0.5(π − π\) + 0.5(y)

where r is the prescribed nominal funds rate, r\ the real equilibrium ("neutral") rate (Taylor assumed 2%), π trailing inflation, π\ the inflation target (2%), and y the output gap (actual minus potential GDP). Both the inflation-gap and output-gap coefficients are 0.5 in the original (Taylor 1993, per Wikipedia and the Atlanta Fed's Taylor Rule Utility). A critical property is the Taylor principle: the response to inflation must exceed one-for-one (the implied coefficient on π is 1.5 here) so that real rates rise when inflation rises — otherwise policy is destabilizing.

The Taylor rule is a descriptive scaffold, not the Fed's operating manual. The real reaction function is built from the Fed's dual mandate — maximum employment and price stability (2% PCE inflation, formalized in 2012) — plus the relative weight placed on each at a given moment, the estimate of the neutral rate (r\*), and a "gradualism" preference for smoothing rate changes. Markets reverse-engineer this from FOMC statements, the quarterly Summary of Economic Projections / dot plot, Chair press conferences, speeches, and — most quantitatively — from how fed funds futures and the 2-year Treasury yield jump on data surprises.

How it's used in practice

For a trader or analyst, the reaction function is the transmission gear between the economic calendar and asset prices. Practically:

• Trade the surprise relative to the reaction function, not the absolute number. A hot CPI print only sells off bonds if the market believes the Fed responds to it. The same inflation surprise produces a larger rate reaction when the Fed is perceived as hawkish/data-sensitive than when it is in "looking through" mode.
• Map which data the Fed currently cares about. When the Fed is inflation-fighting, CPI and PCE dominate; when the labor market is the risk, nonfarm payrolls, the unemployment rate, and JOLTS move rates most. The weighting shifts over the cycle, and the calendar's "high-impact" ranking shifts with it.
• Watch the neutral-rate and dot-plot anchor. Repricing of r\* or the terminal rate moves the entire curve even without new data.
• This is the mechanism behind the "reaction function channel" of monetary policy (Filardo, Hubert & Rungcharoenkitkul, BIS Working Paper 816, 2019): asset prices depend not just on the current policy stance but on the perceived responsiveness of the central bank, so forward guidance works by shaping beliefs about the reaction function itself.

Adoption, debate & evidence

The concept is foundational and uncontested among economists and rates desks — every sell-side rates strategist runs some Taylor-rule variant. What's contested is whether any fixed rule describes the Fed. Bernanke (Brookings, 2015) argues policy should be "systematic, not automatic," and that mechanically applying the Taylor rule fails because the output gap and r\* are extremely hard to measure in real time and the rule gives no guidance at the zero lower bound. A 2025 CEPR paper (DP21473) goes further, arguing the Fed has used discretion rather than the rule.

There is solid empirical evidence the reaction function itself changes. The Cleveland Fed (Economic Commentary 2025-12) found markets expected policy to react more aggressively to inflation news in 2022–2024 than pre-pandemic, and that the 2-year yield's sensitivity to non-inflation news fell over that period — direct evidence the perceived reaction function shifted. The clearest regime change: the Fed adopted Flexible Average Inflation Targeting (FAIT) in 2020 (deliberately tolerating inflation overshoots to make up for past undershoots), then scrapped it in August 2025, returning to symmetric Flexible Inflation Targeting after the post-2020 inflation surge (Federal Reserve framework review; Brookings, 2025). So the reaction function is real and inferable, but it is non-stationary — a model that worked in 2019 misfired in 2021–22.

A related, well-documented hazard is the "Fed information effect": a hawkish surprise can sometimes raise equities if markets read it as the Fed signaling confidence in the economy, complicating naive "hawkish = risk-off" trades (Bauer & Swanson critique notwithstanding).

Strengths & limitations

When it works: as an organizing framework, it is indispensable — it explains why identical data prints produce opposite market reactions in different regimes, and it disciplines event-trading around CPI/payrolls/FOMC. The Taylor rule as a benchmark (Brookings' framing) reliably flags when policy is unusually loose or tight.

When it fails: (1) regime breaks — the function's coefficients and mandate-weighting shift, often abruptly (2020, 2025), and the change is only obvious in hindsight; (2) at the zero lower bound, the funds-rate rule breaks down and the balance sheet/QE becomes the operative instrument; (3) real-time mismeasurement of r\ and the output gap means even the Fed doesn't know its own "correct" rate. The #1 misuse is treating the reaction function as fixed — assuming the 2018 Fed and the 2022 Fed respond to a hot CPI the same way. The second is conflating the economic read of data with the policy* read; markets trade the latter.

Sources

• Taylor rule formula & coefficients — Wikipedia: Taylor rule; Atlanta Fed Taylor Rule Utility
• Reaction function as benchmark, limitations, "systematic not automatic" — Bernanke, Brookings: The Taylor Rule: A benchmark for monetary policy?
• Discretion vs. rule — CEPR DP21473: The Taylor Rule: Did the Fed use Discretion Instead?
• Reaction function channel — BIS Working Paper No. 816: The reaction function channel of monetary policy and the financial cycle (Filardo, Hubert & Rungcharoenkitkul, 2019)
• Shifting perceived reaction function since COVID — Cleveland Fed EC 2025-12
• Dual mandate — Federal Reserve Education: The Fed and the Dual Mandate
• FAIT adoption (2020) and 2025 framework reversal — Federal Reserve: 2025 framework review roadmap; Brookings: How the Fed revised its framework
• Fed information effect — Bauer: An Alternative Explanation for the "Fed Information Effect"

Dispute flags: (1) Whether any fixed rule describes Fed behavior is genuinely contested (Bernanke/CEPR vs. Taylor). (2) The sign of equity reaction to hawkish surprises is ambiguous due to the information effect. (3) The reaction function is empirically non-stationary — treat all coefficients as regime-dependent.