Liquiqui

Fed policy

Sixteen readings of what the Fed looks at, combined into one number: −1 when everything argues for a cut, +1 when everything argues for a hike

Thesis

Is the Fed's next move more likely to be a cut or a hike?

Index

Sixteen indicators, standardized and equal-weighted, updated weekly — from −1 (cut) to +1 (hike), 0 when conditions are at their average since 2003

Fed Policy Index
0.24
1 wk
↓ -0.00
1 mo
↑ +0.01
1 yr
↑ +0.07

Source: Liquiqui, from FRED / BEA / BLS

Contributions by block

Each block's share of the index — the three lines sum to the index above

Source: Liquiqui, from FRED / BEA / BLS

Where each constituent stands today against its target or equilibrium, in standard deviations of its own history since 2003 — positive pushes towards a hike

Inflation

  • Inflation breadthMonthly · Sep 5, 2026+1.4
  • Inflation outlook (1-year expected)Monthly · Oct 3, 2026+1.0
  • PCE inflationMonthly · Aug 1, 2026+1.0
  • Core PCE inflationMonthly · Aug 1, 2026+1.0
  • CPI inflationMonthly · Sep 5, 2026+0.7
  • Inflation persistence (sticky CPI)Monthly · Sep 5, 2026+0.3
  • Inflation expectations (10-year breakeven)Daily · Oct 10, 2026+0.1

Employment

  • Jobless claims (4-week average, inv.)Weekly · Oct 10, 2026+0.5
  • Wage growth (Atlanta Fed tracker)Monthly · Sep 5, 2026+0.4
  • Unemployment rate vs NAIRU (inv.)Monthly · Sep 5, 2026+0.1
  • Unemployed per job opening (inv.)Monthly · Aug 1, 2026+0.0
  • Payroll growth (3-month average)Monthly · Sep 5, 2026-0.1

Growth, financial conditions and policy stance

  • Financial conditions (NFCI, inv.)Weekly · Oct 10, 2026+0.9
  • Policy stance, real rate less FOMC neutral (inv.)Monthly · Aug 1, 2026+0.3
  • Consumer demand vs potential growthMonthly · Aug 1, 2026+0.1
  • Housing startsMonthly · Aug 1, 2026+0.0
Notes
What is happening
The index stands at +0.23 on 3 October 2026, against +0.26 in mid-September and +0.28 a quarter earlier: the evidence leans towards holding or hiking rather than cutting, but not strongly — and the FOMC came down on that side on 16 September, raising the target range by 25 bp to 3.75–4.00% (12–0), its first hike since 2023. Its high is +0.84 in June 2022, mid-hiking cycle; its low −0.80 in April 2020, and 2009 stayed negative all year. Prices are what pull up — breadth at +1.7 standard deviations, headline and core PCE at +1.0 each from their 2% target after the BEA's annual revision of 30 September, which took core from 3.3 to 3.0% — with financial conditions far looser than average (+0.9) and a real rate 0.5 points below the FOMC's neutral (+0.3). The labour market, by contrast, is at balance: 4.2% unemployment against a 4.4% NAIRU, one unemployed per opening, payroll growth at 50,000 a month (−0.1). No constituent pulls clearly towards a cut. The real-rate reading is August's: the hike will lift it a quarter point.
Why it matters
The Fed does not decide on one number but on a body of evidence: two mandates, inflation and employment, read through a dozen series every meeting reviews. This index gathers them into a single reading between −1 and +1: each constituent is measured against its own history since 2003, the labour-slack measures and the real rate enter inverted, and the average is put on a scale where ±1 is two standard deviations. It is not a forecast of the next decision; it is the weight of the evidence.
What's driving it
Inflation climbing since February 2026 on tariffs and energy; a labour market loosening from the bottom — less hiring — rather than through layoffs; financial conditions among the loosest since 2021.
What to watch
A fall below −0.3 would need unemployment above 4.5% and claims rising: the scenario in which the Fed cuts despite inflation. A rise above +0.5 would need core inflation through 3.5% with wages accelerating: the hike scenario, already under way at the FOMC — after 16 September, twelve of eighteen participants pencil in one more by year-end, for a 2026 median of 4.1%.

Raw data

The sixteen series behind the index, each charted on its own page — the latest reading and the sign it enters with

Inflation

Employment

Growth, financial conditions and policy stance

Methodology

  1. 1Sources

    FRED (PCE and CPI prices, breakeven, payrolls, unemployment, claims, JOLTS, wage tracker, real spending, NFCI, housing starts, sticky CPI, fed funds, Cleveland Fed expected inflation), BLS (the 21 CPI components behind the breadth measure).

  2. 2Normalisation

    Inflation measures year on year; payrolls as a three-month average of the monthly change, claims as a four-week average; openings as unemployed per opening; the policy stance as the fed funds rate less core PCE inflation, less the FOMC's longer-run projection net of 2% (the first projection, January 2012, carried back to 2003). Monthly and daily inputs read as of each claims week.

  3. 3Z-scores

    Each constituent minus its target, divided by its standard deviation since 2003; capped at ±3. The targets: 2% for PCE inflation, 2.4% for the CPI measures (the CPI runs about 0.4 points above the PCE), 2.3% for the breakeven and the one-year outlook, half the CPI's categories above 2% for breadth, 3.5% for wages (2% plus productivity), one unemployed per opening, the CBO's NAIRU for unemployment, the CBO's potential growth for demand, zero for financial conditions and for the stance (the FOMC's neutral rate). Payrolls, claims and housing starts have no natural target and are measured from their own average. Unemployment, claims, unemployed per opening, financial conditions and the stance enter inverted, so that higher always means more pressure to raise rates.

  4. 4Weighting

    Sixteen constituents, equal weights: seven on prices, five on the labour market, four on demand, financial conditions, housing and the policy stance.

  5. 5Index

    Average of the sixteen z-scores, divided by 2 and capped: ±1 when the constituents average two standard deviations from normal. Not rebased — 0 is the 2003-2026 average, not a starting date. Updated weekly.

  6. 6Interpretation

    Above 0 the balance of evidence argues for tighter policy, below 0 for easier. Beyond ±0.5 the case is strong; the sign of the move since a quarter ago says whether it is building or fading.

Related