Retail's rebound is inflation in disguise

The week opens with what looks like good news for the American consumer and closes with a decision that may rest on wishing it were true. To open the week, days before the Federal Reserve's rate verdict, the Census Bureau is widely expected to clock a brisk rebound in August retail sales — a rise near 0.8% after July's 0.6% stumble. The verdict that follows on Wednesday is more fraught. Consumer prices rose at their fastest pace in months, and markets now place a near-90% probability on a quarter-point rate hike. The two readings are less independent than they appear. The rebound investors are told to expect and the inflation that is driving the hike are the same phenomenon wearing different clothes.

Start with July, because the arithmetic of the rebound is written there. Retail sales fell 0.6% on the month, the largest drop in more than a year and a solid miss of the small gain that had been predicted. Much of it was an accounting of timing rather than a collapsing shopper. Amazon had moved its Prime Day sale from July to June, pulling forward a haul of online spending that duly vanished from the July books, while the online retailing category fell 2.2%. Automobile dealers, giddy from June promotions, took a 1.8% "breather." Gasoline station receipts fell nearly one per cent as pump prices slid. Underneath all this, in the "control group" that feeds straight into GDP — stripped of autos, gasoline, building materials and restaurants — spending fell about 0.4%. Economists who treat the series as a proxy for the consumer's real pulse read the July report as a pause, not a panic: consumption is still marching at a roughly 2% annualised pace for the quarter.

August is expected to repair the headline. But look at what is built into the forecast of a bounce. Gasoline stations are a retail-sales category, and August was a month in which pump prices surged 3.9% on the month, with the broader energy index up 2.1% monthly and 16.3% on the year as Gulf tensions threatened the Strait of Hormuz. A chunk of the expected rebound is therefore not the resurgent consumer but the same energy shock, recorded twice — once in the inflation report, once in the cash registers of filling stations. The headline number and the inflation scare are two statistics with one cause.

That should puncture any relief the "rebound" label invites. The consumer's genuinely durable signal — real, inflation-adjusted spending and the ex-automobile, ex-petrol control group — is where July's weakness concentrated, and it is the measure that will tell the Fed whether its contemplated medicine is warranted or gratuitous. Elevated prices may mechanically lift the nominal tally while the volume underlying it merely stalls; retail sales data is not adjusted for inflation, so a headline gain can coexist with fewer goods actually moving.

That matters because the central bank finds itself in an awkward position. Inflation is back above target — 3.4% on the year, with "core" inflation of 2.4% — after the June moderation pause ended, and the new chairman, Kevin Warsh, has staked his credibility on price stability, insisting in his Jackson Hole speech that better summer readings had not shown underlying trends improving. Under his leadership the probability market attached to a hike at this meeting has swung from under 30% before his speech to near 90% after this week's consumer-price print. Yet the economists who study the institution are cooler: fully 70% in a Reuters survey expect the committee to hold rates at 3.50-3.75%, their conviction that a hike will come this year more than doubling from a month ago. Economists and markets, unusually, are pointing in different directions.

The structural problem is that a rate hike is the wrong tool for this particular inflation. The acceleration is largely a supply-side affair — fuel prices set by a war in the Persian Gulf, plus the lingering pass-through of trade policy — and raising interest rates cannot lower the price of petrol; it can only cool the demand that funds the spending the retail report measures. That is the uncomfortable arithmetic of the week: the Fed is being pushed toward a pre-emptive hike for inflation it did not cause, to be delivered into a consumer that is already souring. Confidence collapsed in July, with the University of Michigan's gauge down sharply to 51; employers shed 23,000 jobs in that month; credit-card debt sits above a trillion dollars and delinquencies are rising; the giants of retail are discounting heavily for back-to-school as shoppers trade down.

None of this settles whether the committee hikes, holds, or surprises by doing nothing. What it should settle is how a resolved investor reads the retail number when it lands. Read past the headline, into real volumes and the ex-fuel categories, and the "rebound" loses much of its glow: it is the inflation report wearing a retailer's uniform. If the durable measure disappoints, the market's near-certain pricing of a hike begins to look too confident — which would be a relief to rate-sensitive and discretionary holdings that have already sold off on the inflation scare. If it holds up, the Fed's dilemma merely sharpens: an economy too hot for comfort and a central bank that may have to choose between its credibility and the weakening demand underneath it.