The Fed's October decision hangs on two data points

The most important central bank on earth is about to set policy by two numbers — and one of them might not even be published, because the government that produces it has shut itself down.

Citi's weekend note cut through the noise with unusual bluntness: forget Wednesday's PCE, forget the speeches — the October 28–29 decision comes down to the September payrolls report due Friday and the September core CPI due October 14. Everything else is commentary. The market agrees: LSEG's pricing puts 64% odds on a second consecutive hike, an extraordinary wager just weeks after the Fed lifted rates to 3.75–4.00%.

Citi's own forecast — roughly 85,000 new jobs, unemployment nudging from 4.1% to 4.2% — sketches the knife-edge. Stronger, and the hawks have their mandate for back-to-back tightening; weaker, and the doves can argue September's hike is already biting. In an economy where Lisa Cook is warning about oil and AI as twin inflation horsemen, as we reported this morning, the labor market is the last supposedly clean signal.

A central bank that demands data-dependence, denied its data by the government it serves.

Then Washington intervened in the data itself. The federal shutdown that began September 29 has shuttered the Bureau of Labor Statistics' normal operations, and Friday's jobs report — the single most market-moving release in American economics — may not appear on schedule. The Fed would then be flying into its most divided meeting of the year with one instrument out.

The irony is precise: a central bank that demands data-dependence, denied its data by the fiscal chaos of the government it serves. Chair Powell's press conference on October 29 would become an exercise in conditional tenses — what the Fed would do if it knew what it cannot know. Markets price 64% on a hike; they should price the growing chance of a decision made half-blind.

Wednesday's PCE — the Fed's preferred gauge, due September 30 — still matters, but Citi's framing demotes it to scene-setting. The October CPI on the 14th is the decider because it captures the tariff pass-through and the oil-price surge in one print. Two numbers, one shutdown, and the most leveraged fortnight in global finance.

Wall Street's lens is probability and plumbing. The 64% pricing is a bet that inflation's second wave — oil near $97.50, tariffs passing through, AI capex juicing demand — forces the Fed's hand regardless of labor softness. The Western read is unsentimental: the Fed hiked in September because it saw this film before, in the 1970s, and it will hike again for the same reason.

The subtler Western read is institutional. A Fed that hikes into a shutdown-delayed data calendar is asserting independence theatrically — showing Congress that monetary policy does not pause for fiscal dysfunction. Every delayed BLS release becomes, paradoxically, an argument for the Fed's autonomy.

Beijing's lens is the dollar. A Fed that hikes twice into autumn strengthens the greenback against every currency Beijing manages — the yuan, and every Belt and Road borrower's dollar debt. The Eastern read: American inflation fighting is exported as emerging-market pain, and the 64% is a weather forecast for everyone else's storm.

Tokyo's Eastern lens is more immediate. With the BoJ contemplating its own back-to-back hike, a Fed that keeps climbing narrows the rate gap from the other side — the yen trade becomes a race between two central banks. For Asia's finance ministries, October is not one decision but two, and they interact.

The Global South's lens is the cost of money. Every Fed hike reprices dollar debt from Buenos Aires to Nairobi; a back-to-back sequence — 64% priced — would be the fastest tightening impulse the developing world has absorbed since 2022. The Southern read: America's inflation is fought with the South's interest bills.

There is a second Southern reading, about the shutdown itself. If the world's reserve-currency central bank can be blinded by its own government's dysfunction, the South's long-standing complaint — that the global financial architecture rests on American political stability — stops being theoretical. The delayed jobs report is a metaphor with a Bloomberg ticker.

Reading the two numbers: what payrolls and core CPI actually say

Citi's framing — two numbers decide the October meeting — sounds like punditry. It is closer to mechanics. The September payrolls report and the September core CPI are not merely important releases; they are the two halves of the Fed's legal mandate made legible. The payrolls report answers the employment half: are businesses still hiring, is the 4.1 percent unemployment rate a floor or a ledge? Core CPI answers the price-stability half: is the tariff pass-through and the oil surge showing up in the prices households actually pay, stripped of food and energy noise? A central bank with a dual mandate does not get to pick which half it reads. It reads both, and hikes only when both halves point the same way — or when one half screams.

The payrolls number deserves a closer look, because Citi's forecast — roughly 85,000 new jobs — sits in the most treacherous zone of labor-market reading. Below 100,000, payroll growth is flirting with stall speed: enough to keep the unemployment rate from spiking, not enough to absorb new entrants to the workforce. The tick from 4.1 to 4.2 percent unemployment that Citi expects is the market's tripwire. A tenth of a point sounds like nothing; in the Fed's reaction function, it is the difference between "normalization" and "deterioration." Labor markets do not send engraved invitations to their turning points. They send tenths.

Core CPI is the harder read, and the more consequential one. The October 14 print will capture two forces the September data could only hint at: the tariff pass-through working its way through import prices and retail margins, and the oil-price surge feeding into transport and production costs. Core inflation is designed to look through energy volatility — but when energy stays elevated long enough, it stops being volatility and starts being the price level. That is the 1970s lesson the hawks keep citing, and it is why Citi demotes the September 30 PCE to scene-setting: the Fed's preferred gauge tells you where inflation was. The October CPI tells you where it is going.

Then there is the shutdown — the wild card that turns data-dependence into an epistemological crisis. The Bureau of Labor Statistics does not merely publish the jobs report; it is the jobs report, and a shuttered BLS means the October 2 release may not exist on schedule. The Fed has tools for this — private payroll estimates, jobless claims, the anecdotal Beige Book — but every substitute is a downgrade in authority. A central bank that built its credibility on "we follow the data" does not easily pivot to "we followed our best guess." If the payrolls report slips, the October 28–29 meeting becomes the purest test of the Powell Fed's doctrine: does data-dependence mean waiting for the data, or acting on the data you wish you had?

The 1970s in the room

Every mention of the 1970s in a Fed debate is doing the same work: it is the hawks' founding trauma, the memory that justifies preemption. The history, stripped of legend, is this: the Fed of the 1970s eased too early, repeatedly, declaring victory over inflation that had only paused — and each premature easing let expectations reset higher, until taming inflation required the brutal medicine of the early 1980s, double-digit rates and a deliberate recession. The moral the institution drew, and has never unlearned, is that the cost of hiking too much is a recession you can reverse, while the cost of hiking too little is an inflation psychology you cannot. Credibility, once spent, is not restocked.

That is the lens through which the 64 percent pricing makes sense — and through which it should worry the doves. The market is not betting that the economy needs a second consecutive hike. It is betting that the Fed fears needing one later more than it fears the damage of one now. Back-to-back hikes are a statement of institutional memory: we saw this film, we know how it ends, and we will not be the committee that let the sequel happen. The September hike to 3.75–4.00 percent was the first installment of that statement. October would be the second — and second installments are where doctrines get tested, because the first hike can be called prudence while the second must be called conviction.

The counter-history deserves its hearing, because the 1970s analogy has a blind spot: the shocks are different. The 1970s inflation was entrenched by wage-price spirals, cost-of-living clauses, and a decade of fiscal accommodation — a whole economy organized around rising prices. Today's pressures — oil near $97.50, tariff pass-through, AI capex — are supply shocks and relative-price shifts, the kind that monetary policy can only crush by crushing demand itself. Hiking into a supply shock does not produce more oil or fewer tariffs. It produces fewer jobs. The doves' case is not that inflation is harmless; it is that the medicine's side effects are being prescribed for a disease the medicine cannot cure.

Which is why Lisa Cook's warning — oil and AI as twin inflation horsemen, reported this morning — lands with such weight inside the building. It reframes the debate from "is inflation back?" to "what kind of inflation is this?" If the pressures are structural — energy transition costs, deglobalization's price tag, the AI buildout's demand pull — then the Fed is not fighting a cycle. It is fighting an era. And central banks do not have tools for eras. They have a single interest rate, and the question of whether to raise it in October is really the question of whether a single tool can discipline a structural repricing. The 1970s say yes, eventually, at terrible cost. The doves are asking whether "eventually" is a strategy.

Who pays for American data-dependence

While Washington debates tenths of a point, the rest of the world does the arithmetic in its own currency. Every Fed hike reprices the dollar, and the dollar reprices everything: the cost of servicing dollar debt from Buenos Aires to Nairobi, the price of imported energy for countries that buy oil in greenbacks, the carry trades that fund Asia's afternoons. A back-to-back sequence — the 64 percent the market prices — would be the fastest tightening impulse the developing world has absorbed since 2022, delivered not by any decision in their capitals but by two data prints in Washington. American inflation is fought with the South's interest bills. That was true in 2022. It is truer now, because the debts are larger and the buffers thinner.

Tokyo's position captures the developed world's version of the same bind. The Bank of Japan is contemplating its own back-to-back hike — an extraordinary posture for an institution that spent a generation at zero — and a Fed that keeps climbing narrows the rate gap from the other side. The yen trade becomes a race between two central banks, each hiking into different domestic logic, their decisions colliding in the currency market. For Asia's finance ministries, October is not one decision but two, and they interact in ways neither central bank controls. Coordination is a word used in communiqués. Collision is what happens in markets.

And then the deepest cost, the one priced in legitimacy rather than basis points. If the world's reserve-currency central bank sets policy half-blind — because its own government's shutdown delayed the very data the decision depends on — the episode becomes evidence in a case the Global South has been making for years: that the global financial architecture rests on American political stability, and American political stability is no longer a safe assumption. The delayed jobs report is not just a scheduling problem. It is a demonstration, with a Bloomberg ticker, that the system's foundation has a crack in it. The Fed may hike anyway, asserting independence theatrically. But every hike made without the data is a small withdrawal from the credibility account the next crisis will need.

The fortnight ahead, then, is leveraged in every sense. Two numbers — one possibly unpublished, one possibly decisive. A market pricing 64 percent on conviction the Fed may not feel. A world repricing its debts on a decision it does not get to vote on. And underneath it all, the oldest question in central banking, asked again at 3.75–4.00 percent: is the rate the economy needs the rate the Fed dares to set? On October 29, we will learn the answer. The bill, as always, will arrive everywhere else first.

The art of flying half-blind: the Fed's fallback toolkit

Central banks have been here before — not with a shutdown, exactly, but with the data going dark. The playbook for a blind meeting is well thumbed. First, the jobless claims: published weekly by the Labor Department's employment office, which historically keeps running on skeleton staffing even during shutdowns, claims are the highest-frequency read on layoffs available. They do not measure hiring — the half of the labor market the Fed most needs to see — but a spike in firings would show up here first, and its absence is itself information.

Second, the private estimates. Payroll processors publish their own employment counts days before the BLS, built from actual pay stubs rather than surveys. The Fed's staff economists track them closely in normal times; in a shutdown they become the closest thing to an official series. The problem is authority, not accuracy: no chair wants to stand at a press conference and cite a private vendor as the basis for moving the world's most important interest rate. Markets would hear the number. They would also hear the asterisk.

Third, the Beige Book — the Fed's own anecdotal survey, compiled from business contacts in all twelve districts. It is qualitative by design: "contacts reported," "several districts noted," the careful hedges of an institution that knows anecdotes are not data. But in a data vacuum, the Beige Book's texture becomes disproportionately powerful, because it is the only read on the economy the Fed itself produced. Expect the October edition to be parsed like scripture — every adverb weighed, every regional divergence treated as signal. That is what happens when the official statistics go quiet: the whispers get amplified.

The deeper question is what all this improvisation does to the doctrine. Data-dependence was adopted as a discipline — a promise that policy would follow evidence rather than ideology or politics. A meeting that hikes without the payrolls report, or pauses without the CPI, does not technically violate the doctrine; it reveals what the doctrine always was, which is a preference, not a law. The Fed will make its decision on October 29 with whatever it has. The honest press conference would admit the size of the "whatever." Whether Powell gives that press conference is the real data point of the fortnight.