29,000 Jobs and a 10-Year Up 27bp Since the Hike (Fed Minutes Preview, 7 October 2026): What to Expect and What It Means for the Dollar

The minutes of the Federal Reserve's 15-16 September meeting — the one where it raised rates for the first time since 2023 — are released at 2:00 p.m. Eastern on Wednesday 7 October. They describe a committee that has since been overtaken twice: payrolls rose just 29,000 in September and core PCE inflation came in at 3.0% against a 3.3% forecast, while the 10-year Treasury yield climbed from 5.01% on decision day to 5.28%. The question for the dollar is not whether the document reads hawkish. It is whether the conditions it attaches to the next hike have already been met, or already been ruled out.

When the minutes land, and what they cover

The Federal Reserve's meeting calendar schedules the minutes of the 15-16 September meeting for 2:00 p.m. Eastern on Wednesday 7 October — 18:00 UTC, 19:00 in London, 02:00 Thursday in Hong Kong. The next decision follows three weeks later, on 28 October.

The meeting itself is well documented. The Committee voted 12–0 to raise the target range by a quarter point to 3-3/4 to 4 percent, saying the move "will support a timelier return to the Committee's 2 percent goal." The Summary of Economic Projections lifted the median 2026 rate to 4.1% and — the bigger change — the 2027 median to 4.1% from 3.6% in June. CNBC's read of the dot plot found 16 of 18 participants expecting at least one more rise this year. We covered that meeting in detail in three rate decisions in 72 hours.

So the minutes cannot surprise on direction. A committee that just hiked unanimously and has a median dot one step higher is not hiding an easing bias. What they can surprise on is breadth, urgency and conditionality — and those are exactly the dimensions on which the world has moved since.

What has changed in the three weeks since the meeting

Minutes are a photograph of a room, published after everyone has left it. This set was taken before three developments that would each have changed the conversation.

The labour market data softened. The Bureau of Labor Statistics' September employment report showed nonfarm payrolls up 29,000, against the 84,000 economists surveyed by Dow Jones had expected, with August revised to 133,000 and July to a loss of 10,000. The unemployment rate rose to 4.2% from 4.1%. Average hourly earnings rose 0.1% on the month and 3.0% on the year, which CNBC notes is the slowest annual pace since May 2021.

Inflation printed below forecast — on a rewritten index. August core PCE rose 0.2% on the month and 3.0% on the year, against 0.3% and 3.3% expected; headline was 3.4% against 3.7%. But the Bureau of Economic Analysis changed its methodology for legal services, software and portfolio management at the same time, lowering July's core level by 0.36 percentage point. A miss produced partly by a measurement change is weaker evidence than one produced by prices. The mechanics are in our August PCE preview.

The deputies spoke. New York Fed President John Williams said on 29 September that "with the policy action we took at our September meeting, there is no need for urgency, and we have time to gather more information," adding that one further increase "may be appropriate late this year" if the economy evolves in line with his forecast (New York Fed). Vice Chair Philip Jefferson said on 1 October that "yields across the term structure have increased further" since the meeting and that the Committee's judgment "may take more time" (Federal Reserve).

Put together, the market's reading shifted from "October or December" to "December". After the jobs report, CNBC reported that the CME FedWatch probability of a hold on 28 October jumped to 82.8%.

The bond market tightened anyway

Here is the part of the picture the jobs headline hides. Using Treasury's daily par yield curve, closing levels moved like this:

Two things stand out. First, the long end has done three times the work of the front end. The 2-year, which summarises the expected path of policy, is up 9 basis points; the 10-year is up 27. Second, on the day of a 29,000 payroll print, both the 2-year and the 10-year closed higher than the day before. Yields fell straight after the release, then reversed.

That shape — a steepening led by long yields — says the move is mostly not about the next Fed meeting. A 10-year yield is roughly the expected average of short rates over a decade plus a term premium: extra compensation for locking money up through uncertain inflation and heavy government borrowing. When the long end rises while the front end barely moves, it is the term premium doing the lifting. We traced the same mechanism in August's long-end selloff.

For a central bank, that matters directly. Higher long-term yields raise mortgage rates and corporate borrowing costs and lower equity valuations — the tightening a rate hike is meant to produce, delivered without one. Jefferson's choice to mention the term structure in a short speech is a sign the Committee has noticed.

What to read in the document

There is no consensus forecast for minutes — they are text, not a number — so the useful preparation is knowing which sentences carry information that is not already in the price.

1. The quantifier on the next move. The Fed's minutes use a fixed ladder: "a few", "some", "several", "many", "most", "all". The projections tell us 16 of 18 pencilled in another increase, so "most participants" expecting further firming would be confirmation, not news. The news would be any language attaching timing — "at upcoming meetings" versus "over time" — or a group that wanted 50 basis points rather than 25. The August minutes showed "several" participants wanted a hike when only three dissented; the gap between the vote and the room is where surprises live.

2. The conditions. In August the market traded one clause: tightening "would likely be necessary if inflation did not decline." Look for the September equivalent. If the room tied the next move to inflation expectations, wage growth or the labour market, then September's data has partly answered it — wage growth is at a five-year low and unemployment ticked up. If it was tied to energy pass-through into core prices, the condition is still open.

3. Financial conditions and the long end. Any discussion of whether higher long-term yields could substitute for policy tightening matters most for the 2-year, because it would make December conditional on the bond market as well as the data. The 10-year was already at 5.01% at the meeting, so the staff and participants likely discussed it.

4. The balance of risks. Watch how participants characterised downside risks to employment. A committee that saw the labour market as a minor concern on 16 September has since had a 29,000 print and 60,000 of downward revisions. The minutes cannot reflect that, but they will show how much weight the room put on the employment side before it arrived.

Three scenarios and what each means for the dollar

The middle row is effectively what the market has priced, because the leadership has already described it publicly. That makes the outer rows the only ones with real price content, and the asymmetry is worth noticing. The hawkish tail has to beat sixteen dots and a unanimous vote. The dovish-conditional tail only has to show that the room's trigger was something the data has since moved against.

For equities the channel is the same yield, read from the other side. With the S&P 500's earnings yield already level with the 10-year Treasury — the comparison we set out in 5.21% against 5.26% — any minutes-driven move in long yields lands directly on index valuations, most of all on long-duration growth names in the NAS100.

The dollar through the five factors

The meter scores eight currencies on five factors: interest rates, growth, positioning, risk sentiment and commodities. The minutes act almost entirely on the first. The interest-rate factor reflects where US rates sit against the other seven, and the Fed has moved — the European Central Bank's deposit rate is 2.50% and the Bank of Japan's call rate 1.25%. What the minutes can change is the expected next step, which feeds through the 2-year.

Growth is the cross-current. The Atlanta Fed was tracking third-quarter GDP at 3.7% as of the jobs report, according to CNBC, even as payrolls stalled — a low-hire, low-fire economy with strong output. Risk sentiment is where the long end bites: a 5.28% 10-year is a headwind for equities, and episodes of equity stress have tended to favour the yen and Swiss franc over the dollar. Commodities enter through oil, which keeps energy in headline inflation and in the conditions the minutes are likely to describe.

The takeaway

The September minutes will almost certainly sound hawkish, because the meeting was hawkish: a unanimous hike, a higher dot plot and no easing anywhere in sight. That is precisely why the tone carries so little information. Since the meeting, payrolls slowed to 29,000, core PCE came in four tenths below forecast on a revised index, the Committee's two most senior deputies said there is time, and the bond market delivered 27 basis points of long-end tightening on its own.

So read the minutes as a test of conditions. If the room's trigger for the next hike was something the data has since moved against, December weakens, and the dollar has less rate support than the dots imply. If the trigger was core inflation that energy is still feeding, nothing in the last three weeks has defused it.

To learn how Pip Theory builds its fundamental currency-strength scores, see the methodology overview, or track the dollar's factor breakdown on its USD page and the yen on the JPY page. For the earlier stages of this storyline, see what July's minutes said on 19 August and our September jobs report preview.

Educational macro context only — not investment advice.