The 1994 Trade

I was twenty-six the first time a Fed chairman convinced himself he was ahead of the bond market. Greenspan hiked in February 1994 off a fed funds rate of 3%, and the long bond did something nobody at the desk had a playbook for — it didn't calm down, it accelerated. By November the 30-year had gone from roughly 6% to nearly 8%, Orange County had bankrupted itself on the wrong side of the move, and every hike that year turned out to be chasing a market that had already decided rates were going higher regardless of what the FOMC said on any given Wednesday. The lesson, which took me a decade to actually internalize instead of just repeat at dinner parties, is that the Fed doesn't lead the long end. It ratifies it, badly, on a lag, and gets blamed for the overshoot either way.

I keep thinking about that November because Wednesday felt like the same joke wearing a different suit.

Warsh hiked 25 basis points to 3.75%-4%, first increase since July 2023, and the vote was 12-0 — no daylight, no dissent, the kind of unanimity a chairman wants when he's about to tell you the destination is further out than you'd hoped. It was. Sixteen of eighteen participants see at least one more hike this year. Four see two. The median 2027 dot jumped from 3.4% to 3.9%, and the first cut doesn't show up on the committee's own grid until 2028. Warsh didn't submit a dot himself, which I choose to read less as humility and more as a man who doesn't want a paper trail if the committee's own median turns out to be wrong in either direction.

Here's the 1994 part. The 30-year yield touched 5.37% on Monday, before the meeting even started — a level this market hasn't traded since 2007. The 10-year cracked 5% the same week for the first time since 2023. None of that happened because of Wednesday's vote. It happened because Saudi Arabia's East-West pipeline is still down after the drone strike, Libyan production is still offline, Russian refining capacity is still getting chewed up by Ukrainian strikes, and the bond market did the arithmetic on sustained $100-plus oil weeks before the Fed's models caught up. Warsh hiked into a long end that had already hiked itself. That's not policy leadership. That's a committee formalizing a decision the market made for them, the same way Greenspan's committee did in the back half of 1994 — except in '94 the tightening was actually working against demand-side inflation, and this time the thing pushing yields is a supply shock sitting in the Strait of Hormuz and the Red Sea, which a fed funds rate has never once fixed in the history of central banking.

What worries me isn't the hike. It's what showed up quietly the same morning and got buried under the FOMC headline: the Empire State Manufacturing index printed 7.6, badly missing the 14.1 consensus and falling hard from 20.6 the month before. That's not noise. That's the first visible mark of what 5%-plus long rates do to a regional manufacturing base that was already running on fumes from tariff-driven input costs. The VIX had already jumped nearly 8% to 17.10 by Monday, ahead of the decision, which tells you positioning was defensive before Warsh said a word. Then Thursday the S&P ripped back to 7,596, up 0.59%, as if none of that mattered, because oil pulled off its highs on a surprise US inventory build and gold slipped back under $4,310 before catching a bid again above $4,300. Three different tapes, three different stories, and not one of them is pricing what an Empire State miss actually implies about what happens to the rest of the manufacturing surveys once the fourth or fifth hike shows up on the dot plot instead of the first.

I watched the Orange County treasurer testify in 1995 about how a portfolio built for a rate environment that stopped existing eighteen months earlier destroyed a county's finances in about six weeks. Nobody in that room in February 1994 thought the Fed was going to keep going as long as it did. Consensus never does, because consensus is built by extrapolating the last data point, and the last data point on Wednesday was a stock market shrugging off a hawkish SEP because oil inventories ticked up for one week.

The Fed just told you, in writing, eighteen dots wide, that this is a multi-hike cycle running into a supply-driven inflation problem it structurally cannot solve. The bond market told you that back in August, before the meeting, at 5.3% on the long end. The equity market hasn't told you anything yet, because it hasn't had to. It will, the way it always eventually does, usually about two quarters after the manufacturing surveys start rolling over and about one quarter after everyone stops calling it transitory.

I've seen this movie. I just can't remember anymore whether it's the one that ends fine or the one that ends with a county selling itself into a fire sale on national television.