What we're measuring, and why each works
The Fed Funds Rate is set by the FOMC, but it's not unpredictable — two very different things see it coming, and (the useful part) they don't overlap.
1. The 2Y−FedFunds spread — the market's forecast
The 2-year Treasury yield is, in effect, the market's expected average funds rate over the next two years. Subtract today's funds rate and you get a pure read on where the market thinks the Fed is going:
\[ S_t \;=\; y^{2Y}_t \;-\; \text{FedFunds}_t \]\(S_t<0\) (the 2Y sits below the funds rate, a short-end inversion) is the high-conviction cut signal; \(S_t>0\) leans hike. It works because the bond market front-runs the Fed — but only off the zero bound.
PCE momentum is the acceleration of inflation — the 12-month change in the year-over-year PCE inflation rate. First the YoY rate, then how much that rate itself has moved over the past year:
\[ \pi_t \;=\; \left(\dfrac{\text{PCE}_t}{\text{PCE}_{t-12}}-1\right)\times 100 \qquad M_t \;=\; \pi_t - \pi_{t-12} \]\(M_t>0\) means inflation is speeding up vs a year ago; \(M_t<0\) means it's cooling. Why it predicts the Fed where the inflation level doesn't: the Fed reacts to the trend, not the absolute number — and at a peak level the Fed is usually about to pivot, so level is contra-indicative. We use PCE specifically because it's the gauge the Fed's 2% target is written against (it beat CPI in the horse race). Think of it as the acceleration the Fed is chasing.
Fed Funds vs. its two predictors
Top: the Fed Funds Rate and the 2Y yield move almost as one (the 2Y is coincident, not leading — the gap is the signal). Middle: the 2Y−FF spread (below the dashed −0.25 line = cut-watch). Bottom: PCE momentum (above zero = inflation accelerating, hike pressure).
Watch the pattern: the spread rolls below zero and PCE momentum turns down before the Fed Funds line falls — and vice-versa before hikes.
How well each predicts the next 12 months of Fed Funds
| Signal | Hit-rate | R² |
|---|---|---|
| baseline (guess "up") | 47% | — |
| 2Y−FF spread alone | 68% | 0.105 |
| PCE momentum alone | 69% | 0.152 |
| both signals agree | 83% | 0.267 |
- They're independent — the spread is the market's view, PCE momentum is the inflation driving the Fed.
- R² nearly adds (0.105 + 0.152 ≈ 0.267), so each brings new information.
- When both agree (~57% of the time): 83% accurate.
- When they disagree there's no edge (43–57% — a coin flip).
- Both break at the zero bound — the Fed is immobilized by forward guidance, not market rates.
- The 2Y level is coincident, and inflation level is useless — it's the spread and the momentum that carry the signal.
The current read (2026-04)
2Y−FF spread
+0.16%
Inside the ±0.25 band → no high-conviction signal from the market.
PCE momentum
+1.49%
Inflation re-accelerating — hawkish, hike-leaning (PCE YoY 3.77%).
The two predictors disagree / are mixed right now → no high-conviction call; historically that's a coin flip, so the honest move is to wait for them to align.
To anticipate the Fed, don't watch one number — watch two independent ones: what the bond market prices (2Y−FF spread) and what's forcing the Fed's hand (PCE momentum). When they point the same way, you're right ~83% of the time a year out; when they split, wait.