Payrolls at 29k versus a 90k forecast: what the market is really pricing

The data itself: a clear cooling

US non-farm payrolls added 29,000 jobs in September on a seasonally adjusted basis against a consensus of 90,000. July and August were revised down by a combined 60,000, and the unemployment rate rose from 4.1% to 4.2%. By sector, leisure and hospitality and local government education were the main drags. Read on its own, the release points to a labour market weakening at the margin.

Yet the dollar did not trend lower after the release. Understanding that non-reaction means putting the data back into the chain in which it is actually used.

Link one: a single month is noisy

First prints of payrolls have always been revised substantially, and a combined 60,000 downward revision to July and August is itself evidence that a single month's reading is unstable. The three-month moving average is still near 50,000, jobless claims remain low, and the participation rate rose 0.2 percentage points to 61.8% — supply-side repair alone lifts the unemployment rate. Put together, these signals are not enough for the market to read 29,000 as the start of a recession.

Link two: where inflation sits decides the weight of policy

CPI for the same month rose to 3.7% year on year, with a 0.6% monthly print, the largest in five months. With inflation still above target and energy prices held high by Middle East tensions, a marginal weakening in employment is not sufficient to change the direction of policy. The New York Fed's Survey of Consumer Expectations showed one-year inflation expectations rising from 3.58% to 3.9% — the type of signal the Fed watches more closely.

Link three: pricing is about probability, not direction

The market currently prices the Fed as holding in October and hiking 25 basis points in December, with a cumulative probability above 85% for a hike before year-end. That means a single month of employment data would need to weaken persistently and across several releases to flatten that path. One data point cannot change the path; it can only change the confidence attached to it.

What a trader can do with this

First, judge "is the data good or bad" separately from "will price move" — a policy path sits between the two. Second, the first move after a release usually coincides with the worst liquidity of the session, when spreads widen and slippage is obvious; entering there is voluntarily raising your cost. Third, the genuinely tradeable signal is whether price and policy pricing diverge persistently after the release. Employment weakening while hike pricing stays put is the moment dollar longs should become cautious.