A plain-English guide to restarting a hiking cycle.

Wednesday's analysis piece is the live example. The FOMC voted 12-0 to raise the federal funds rate by 25 basis points to 3.75-4.00 percent, the first hike since July 2023, alongside a Summary of Economic Projections that pushed the 2026 year-end median to 4.10 percent and the 2027 year-end median to 4.10 percent (from 3.60 percent). This reference piece is the framework for reading a Fed decision that restarts a hiking cycle after a long hold, and specifically for the four historical parallels the current move rhymes with: 1994, 1999-2000, 2015-2018, and now 2026.

Why restart-hikes are different from mid-cycle hikes

A hike inside an existing hiking cycle is a repricing of the trajectory. A hike that restarts a cycle after a long hold is a repricing of the entire policy regime. The market has to update three things simultaneously: (1) the current policy rate, (2) the direction of future policy, and (3) the confidence that the Fed is willing to accept the growth-and-employment costs of pressing on inflation. The third variable is where restart-hikes deliver their biggest surprise: the market usually discovers, one dot plot at a time, that the reaction function has genuinely shifted.

Restart-hikes also carry more information about the Fed's inflation-tolerance band than any single hike inside a cycle does. In an existing cycle, the market can attribute the hike to the trajectory. In a restart, the market has to attribute the hike to a specific reading of the current dashboard, and the language the Chair uses to justify the reversal becomes the operative signal for the next 12 to 18 months of pricing.

The four historical parallels

1994. Greenspan Fed hiked 25bp in February 1994 after a 21-month hold at 3.00 percent. The move was widely anticipated but the pace of the following seven hikes (300bp over 12 months) surprised the market. The specific footprint: 2-year yield rose 300bp over 12 months, the 10-year yield rose 200bp, DXY rose 6 percent, gold fell 15 percent. The takeaway: the market underprices the pace of cycle restart, and long-duration assets suffer disproportionately.

1999-2000. Greenspan Fed hiked 25bp in June 1999 after a 12-month hold at 4.75 percent. This restart came in the middle of a productivity boom and rising asset prices. The Fed hiked 150bp over the following 12 months. DXY rose 3 percent; gold fell 8 percent; the S&P was up 12 percent through the hiking cycle before rolling over on the tech-cycle rather than on the Fed. The takeaway: restart-hikes do not necessarily crack risk assets in the short-run, but they set the pre-conditions for a later re-rating.

2015-2018. Yellen Fed hiked 25bp in December 2015 after a seven-year hold at 0.00-0.25 percent. The pace was famously slow (only one hike in 2015 and one in 2016) before accelerating in 2017-2018. DXY rose 3 percent in 2015 on the hike, then fell 9 percent in 2016-2017 as the pace disappointed, then rose 4 percent in 2018 as the pace picked up. The takeaway: the initial move matters less than the subsequent dot plot revisions.

2026. Warsh Fed hiked 25bp on September 16 to 3.75-4.00 percent after a 26-month hold at 3.50-3.75 percent. The 2026 median in the SEP moved to 4.10 percent (one more hike this year); the 2027 median moved to 4.10 percent (no cuts). This is closest in pattern to the 1994 restart in its unanimity and its SEP-driven upward revision, but closer to 2015 in its pace projection. The framework's read: the pace will depend on the trajectory of inflation expectations more than on the trajectory of realised inflation, because Warsh's stated reaction function weights expectations more heavily than his predecessors did.

The three signals the market reprices in the first two weeks

Signal 1: the front-end. The 2-year yield is the specific instrument that carries the "policy path" signal. On Wednesday September 16, 2026 the 2-year moved 12.1bp higher on the decision; the historical range for a restart-hike day is 8-20bp, so the 12.1bp move is inside the typical band. If the 2-year continues to grind higher through the two weeks after the meeting, the market is confirming the SEP trajectory. If the 2-year retraces back toward pre-meeting levels, the market is reading the September move as one-and-done.

Signal 2: the term-premium proxy. The 30-year yield's move above the 2-year's move is the term-premium signature. On September 16 the 30-year moved 8.4bp versus the 2-year's 12.1bp; the term-premium footprint is not extending on this decision. That is consistent with the market reading the hawkish surprise as "policy path, not persistence risk": the reaction function moved, not the market's view of long-run inflation.

Signal 3: the dollar's second week. DXY typically extends 60 to 120 pips in the week after a restart-hike, then consolidates or partially retraces in the second week. A DXY that keeps extending in week two is confirming the trajectory read; a DXY that retraces in week two is signalling that the market is fading the hawkish signal.

The four questions to ask on a restart-hike day

1. Was the vote unanimous? Unanimous restart-hikes (like 1994 and 2026) carry more signal than split votes. A dissent on a restart is a specific sign the committee is not fully aligned on the diagnosis.
2. How did the SEP move relative to the prior projection? The forward path shift is more consequential than the current move. A 30bp median shift like 2026 September is a decisive signal; a 10bp shift is a signal-preserving hedge.
3. What did the statement drop? Restart statements often drop forward-guidance language because the reversal itself is the strongest possible guidance. Watch for "additional policy firming" language and "the extent and timing" language; both dropped on September 16.
4. How does the press conference frame optionality? A Chair who commits to a path (as Bernanke did in 2013) locks in the trajectory; a Chair who preserves optionality (as Warsh did on September 16) leaves room for the trajectory to be shaped by data.

Related reading

• Wednesday's analysis piece: the live example this framework describes.
• SEP and dot plot: the projection mechanics and where the surprise typically lives.
• FOMC day tape: the two-hour repricing mechanic on decision day.
• Forward guidance: the language hooks in the statement that carry the signal.
• Monetary policy frameworks: the FAIT vs implicit-IT distinction that underlies Warsh's stated approach.