Fed’s Rate Hike Based on Flawed Jobs Data as BLS Cuts 60,000 Positions

The numbers were wrong. The Federal Reserve raised interest rates on September 16 — its first hike in three years — and it did so leaning on employment data that has since been revised significantly downward.

The US Bureau of Labor Statistics just cut its July and August job gain figures by a combined 60,000 positions. July’s initial reading of a 21,000 job gain got flipped entirely — it’s now a 10,000 loss. August got trimmed by 29,000. That’s not a rounding error. That’s a meaningful swing in the picture the Fed was working from when it decided borrowing costs needed to go up.

The rate hike hit mortgages and credit costs directly.

What the Fed Thought It Knew

In its September statement, the Fed said job gains were aligning with workforce growth. Kevin Warsh, a Fed nominee at the time, pointed to rising job openings and longer weekly hours as signs the economy was near full employment. The Fed’s logic, basically, was that the labor market was strong enough to handle tighter monetary policy.

But job creation had actually been on a declining trend for five years before any of this. The August numbers, which looked decent on the surface, weren’t really what they seemed. Jefferies’ chief US economist called the apparent August strength just a rebound from weak prior months — not a sign of genuine momentum. And the initial August report, before anyone knew it would be cut by 29,000, pushed UBS to raise its rate hike probability estimate from 50% to 60%. So bad data fed into bank forecasts, which fed into market expectations, which fed into Fed confidence.

September’s actual employment data made things worse. Employers added just 29,000 jobs against a forecast of 84,000. Unemployment climbed to 4.2%. That’s a pretty stark miss.

BLS Revisions and the Accuracy Problem

The BLS said the downward revisions came from additional business and government reports and from seasonal adjustments. That’s the standard explanation, and it’s probably accurate — but it doesn’t make the timing any less uncomfortable. The Fed moved on preliminary data. The more complete picture came later.

See also: Bitcoin Surges to $86,000 as Fed Officials Signal Rate Stability

Financial commentators have been fairly blunt: the Fed’s decision might have looked different with accurate figures in hand. It’s hard to argue otherwise. A July that was actually a net job loss, combined with an August that was weaker than reported, and a September that badly missed forecasts — that’s a labor market that doesn’t obviously scream “raise rates now.”

And yet 16 of the 18 Federal Open Market Committee members, after the September hike, still expected another increase before year-end. Whether those revised figures change that math is unclear yet. No one from the Fed has come out and addressed the revisions directly or said anything about reconsidering the pace of tightening.

What Comes Next for Fed Policy

The FOMC’s next steps are genuinely murky. The committee was already working from a forecast framework that assumed a strong labor market. The revised data doesn’t fit that frame cleanly. Unemployment is up. Job creation is running well below what forecasters expected. The BLS’s own numbers now show the summer was softer than it looked in real time.

There’s a real question here about how much weight the Fed puts on revised historical data versus the forward-looking signals it typically prefers. Central banks generally don’t love reversing course quickly — it sends a bad message about their own credibility. So even if policymakers privately think the September hike was premature, they’re unlikely to say so out loud anytime soon.

What the market is watching now is the next round of jobs data. If October comes in soft too, the case for another hike gets harder to make. If it bounces back, the Fed will probably treat the summer weakness as noise.

Read also: Pound Edges Up to $1.2274 as Dollar Stays Strong Ahead of Jobs Report

The BLS revisions have also reignited a broader debate about preliminary data and how much policy should rely on it. Monthly jobs reports are always estimates — they get revised, sometimes substantially. July’s swing from a gain of 21,000 to a loss of 10,000 is a 31,000-job reversal on a single month. That’s not unusual in the long history of BLS revisions, but it’s a jarring reminder that the first number isn’t always the real number.

For now, the Fed hasn’t commented further on how the revisions factor into its outlook. The FOMC’s next move will likely depend on what the labor market looks like in the weeks ahead. Sixteen of eighteen members still expected another hike as of the September meeting — but that was before the revised figures landed.

September’s 29,000 job gain, against an 84,000 forecast, with unemployment at 4.2%.