The CPI Said Hot. The Payrolls Said Cold.

September real hourly earnings fell 0.5% on the month and 0.7% on the year, the BLS reported on 14 October, against a 0.1% gain in nominal pay. Payrolls added 29,000. The first figure is the tax; the second is the reason it cannot be passed on as wages - and Wednesday's CPI piece named the labour market as the channel that would turn an energy shock into a price spiral.

Real earnings are wages after prices: what a pay packet buys once the shopping is done. The Bureau of Labor Statistics reported on 14 October that they fell 0.5% in September, with prices up 0.6% and pay up 0.1%, and that payrolls added 29,000. The saving rate is 4.1%, and August real consumption rose 0.6% against real disposable income that did not move. Those numbers describe a squeeze with a missing ingredient. Inflation taxes the household, and the household is paying - out of savings rather than out of wages, because a labour market adding 29,000 jobs gives a worker no leverage to ask for the difference back. Wednesday's CPI piece named the labour market as the channel that makes an energy shock second round. This piece tests whether that channel is open, and what the credit tape says while it is not.

What printed

The Bureau of Labor Statistics reported on 14 October that nominal average hourly earnings rose 0.1% in September and the all-items consumer price index rose 0.6%. Real hourly earnings fell 0.5% on the month and 0.7% over 12 months.

The same release put the month's payroll gain at 29,000.

Two lines carry the argument. Nominal pay is still rising and prices are rising faster, so the household's real claim on output is shrinking, while an employment report adding 29,000 jobs leaves the worker with no leverage to ask for the difference back. The tax is being levied. The collection mechanism is missing.

The household is paying out of the buffer

The Bureau of Economic Analysis reported on 26 September that real disposable income was unchanged in August while real consumption rose 0.6%. Personal income rose 0.2% and outlays rose $190.7bn, and the personal saving rate stood at 4.1%.

Spending that outruns income is funded from somewhere, and in this release it was funded from the buffer.

A saving rate of 4.1% is not an empty tank, so this is not distress yet. It is the mechanism a later quarter tests: consumption that does not track income has to stop when the buffer stops, and the pullback lands on restaurants, travel, durables and debt repayment before it lands on the price index.

A rate is not a level

The 0.7% annual figure is a direction, not a pay cut. A wage growing 0.1% against prices growing 0.6% loses ground without falling; what it loses is buying power, and the loss shows up in the volume of goods bought rather than in the wage itself.

A slowing car has not gone backwards.

The distinction sets the timetable. A cut in the wage forces a cut in spending now. A lost rate of growth forces it when the buffer is drawn down, and the interval between the two is the only reason consumption is still rising against flat income.

The line the post leaves out: this is a transfer

The United States is a net exporter of total petroleum. In its Weekly Petroleum Status Report for the week ending 2 October, the Energy Information Administration put total petroleum net imports at -4,535 thousand barrels a day - a net export of about 4.5 million - against net crude imports of 2,075 thousand barrels a day.

For a net importer a higher barrel is a leak: the money leaves the country. For a net exporter of the total barrel most of it stays, so the household's larger fuel bill is matched by larger receipts at domestic producers, refiners and royalty holders.

The barrel is a distribution event wearing a macro costume. The aggregate absorbs less of it than the headline suggests, and the household absorbs the same amount as ever.

Demand destruction is the cure

The International Energy Agency's October Oil Market Report, released on 14 October, cut its 2026 demand forecast to a fall of 2.6 million barrels a day, said diesel would account for 25% of the decline, and put observed inventories 509 million barrels below pre-war levels, with a diesel rebuild unlikely before the second half of 2027.

Falling demand is the price doing the rationing. High prices are supposed to cut consumption, and cut consumption is what brings a tight market back to balance; the pain is the mechanism working, not the mechanism failing.

What would make it outlast the market is a wage response - a worker facing a bigger fuel bill, in a labour market short of hands, asking for the difference and being paid it. That needs leverage. Payrolls added 29,000, so the loop has no engine.

Where it lands on this column's tape

High yield OAS was 3.16% on 13 October, from 3.11% on 12 October and 3.03% on 6 October. Investment grade OAS was 0.83% on 13 October, unchanged across the week.

The funds split on the same days. HYG fell to 76.81, down 0.37 on the week, and JNK to 92.25, down 0.51, while LQD rose 0.56 to 102.66 - the equity-like tier selling and the investment grade tier bid.

Wednesday's CPI piece set two tests: a breakeven that rises on a hot print, and the 10yr real yield against a 2.95% invalidation. The breakeven was 2.33% on 13 October and 2.35% on 14 October, and the real yield fell from 2.887% to 2.84%. The benign leg printed, and the invalidation is further away than it was.

The fiscal and credit transmission

A borrower rolling a low-rate loan faces arithmetic that never appears in the deficit headlines. A property or business refinancing at 7% instead of 4% pays 75% more annual interest on an unchanged interest-only balance.

That cost lands on cash flow before it lands anywhere else.

The labour market is where it becomes visible. Initial claims were 197,000 in the week ending 3 October, from 199,000 the week before, and continuing claims were 1,716,000 in the week ending 26 September. Neither is a break. The squeeze reaches the price index only if the labour market transmits it, and this is the series that would show it starting.

What would settle it

One: real earnings. The decline is the mechanism in motion. A turn to positive real earnings while payrolls stay near 29,000 would say the squeeze is being settled at the wage desk after all.

Two: the saving rate. At 4.1% there is room left. A print under 3.5% says the buffer is close to spent and consumption has to track income.

Three: the credit wedge. High yield at 3.16% against a 3.24% peak, with investment grade pinned at 0.83% - the widening that matters is the one where investment grade follows.

Four: claims. Back under 190,000 would say the labour market is still tight enough to pass the tax on.