Three Soft Data Prints in One Week Cut the Odds of a September Rate Hike From Fifty to Thirty One

For most of the past three years the question about the Federal Reserve was how quickly it would cut. That question has quietly inverted, and a great deal of commentary has not caught up with it.

The debate now is whether the Fed raises in September. Last week the market put that at roughly even. After three data releases it is closer to a third.

What actually moved?

Jobs, consumer prices and producer prices, all in the same week, all softer than expected.

Any one of those on its own is noise. Three consecutive prints pointing the same way is a signal, and the repricing was immediate: hike odds fell from about 50 per cent to about 31 per cent, with subsequent estimates settling near 30 after weak retail sales and soft consumer sentiment.

The mechanism is straightforward. A hike would only be justified by evidence that inflation is reaccelerating or that the labour market is tight enough to force it. Soft jobs and soft prices remove the case, so the probability falls.

Why is a hike on the table at all?

Because the pressures on the long end have not gone away, and the Fed does not control them directly.

Long-dated government borrowing costs have been elevated for reasons that are largely fiscal rather than monetary: heavy issuance, persistent deficits, and enormous capital demands from the current build-out of computing infrastructure. Those forces sit outside the policy rate, which is precisely the difficulty covered in why the thirty year has stayed high while the Fed has not moved.

In that environment a central bank has an awkward choice. Cutting into a market already demanding a higher term premium risks steepening the curve further and looking as though it has given up on inflation. Holding is uncomfortable. Hiking is the option nobody wants and nobody can rule out.

What has it done to gold?

Gold is having a month that would normally require a crisis.