Friday's Consumer Doubleheader: Retail Sales at 8:30, Sentiment Stuck Near 55 at 10:00

by Alina Castellanos · August 14, 2026

Economics Correspondent, US Market Current

Two readings on the American consumer arrive ninety minutes apart this morning. At 8:30 a.m. Eastern the Census Bureau publishes July advance retail sales, where Investing.com's calendar carries a forecast of a 0.1% monthly gain against a 0.2% rise in June. At 10:00 a.m. the University of Michigan releases its preliminary August sentiment index, and there the published forecasts do not agree with one another at all. With a Fed that spent July arguing over whether to raise rates, reconciling what households buy with what they say is no longer academic.

Start with the spending side, where the headline is the least informative part. The Census Bureau's June advance report, released July 16, showed total retail and food services sales up 0.2% from May but down 0.2% excluding motor vehicles, with sales running 6.7% above June 2025 and May revised up from 0.9% to 1.0%. Census flagged that the June monthly change was not statistically distinguishable from zero. In an Aug. 13 preview, Investing.com put the July forecast for core retail sales, its ex-autos measure, at 0.2% against that negative June print, and listed a 0.4% prior for sales excluding both gasoline and autos with no forecast attached. Continuum Economics looks for total sales unchanged and ex-auto-and-gasoline up 0.4%.

Those definitions are not interchangeable, and in a month when pump prices moved, the choice changes the answer. Continuum expects gasoline prices to remain a significant negative in July, if less so than in June, with food stores and eating and drinking places due modest gains and a modest correction lower in autos. If the headline disappoints because gasoline receipts fell while the ex-gas measure holds a 0.4% pace, the miss is a price artifact. If the ex-gas, ex-auto measure decelerates too, that is a real crack, and it feeds most directly into the consumption line of GDP.

The sentiment side has the stranger history. The Michigan index sat at 49.5 in June, printed 54.4 in the preliminary July estimate, and finished July at 55.2, an 11.5% monthly gain and, as Advisor Perspectives noted, the highest reading since March. Current conditions stood at 54.8 and expectations at 55.4. The Surveys of Consumers reported the improvement was broad-based across income, education, wealth, age and political party, but also that sentiment sat 10.5% below July 2025, with director Joanne Hsu describing a generally somber view of the economy amid five years of elevated inflation.

The August forecast is where readers should be careful, because there is no single settled number. Investing.com's calendar and its Aug. 13 preview both carry a consensus of 54.4 against the 55.2 prior — but 54.4 is also precisely the July preliminary reading, and the figure that calendar carried as the forecast for July's final. Trading Economics lists a 54.5 consensus alongside its own 54.6 model forecast. FXStreet's calendar page carries 55.2. The defensible frame is direction against July's 55.2 final: forecasters broadly expect the rebound to stall or give a little back, not to reverse.

The part of the release the Fed actually reads sits below the headline index. Year-ahead inflation expectations fell to 4.2% in July from 4.6% in June, while long-run expectations held at 3.3%. The five-year figure is the more consequential one, because anchoring at the long end is the argument policymakers use to tolerate an above-target headline rate. If the August preliminary shows the long-run measure drifting higher while the index falls, the committee gets the worst available combination.

Now the gap. Surveys have households near the bottom of their range; tracking data has spending accelerating. The Atlanta Fed's GDPNow nowcast for third-quarter real GDP growth stood at 5.8% as of its Aug. 6 update, the Atlanta Fed's own page shows, and that is still the current vintage — the next refresh is scheduled for today. FRED carries the same observation at 5.83%. The model's initial third-quarter estimate was 5.0% on July 30; an Aug. 3 revision lifted it to 6.2%, with InvestingLive reporting the third-quarter consumption nowcast rose to 4.6% from 3.3%. Two cautions belong with a number that large: GDPNow is a mechanical nowcast with wide error bands early in a quarter, and it tracks a quarter following a Q2 in which real GDP grew at a 1.5% annual rate, per the BEA's advance estimate.

The inflation backdrop explains part of why mood and money diverge. July CPI, published by the Bureau of Labor Statistics on Aug. 12, rose 0.1% seasonally adjusted after falling 0.4% in June, and was up 3.4% from a year earlier, with core prices up 0.2% on the month and 2.5% over twelve months. But the energy index was up 14.7% year over year, food 3.0% and shelter 3.2%. The categories consumers see priced most often are still rising fastest, while the disinflation sits in the core basket.

The counterweight is the labor market. The BLS Employment Situation released Aug. 7 showed July nonfarm payrolls falling by 23,000, with unemployment at 4.1%. The same release revised May down 66,000 to a gain of 63,000 and June down 37,000 to a gain of 20,000, leaving the two months a combined 103,000 lower than previously reported. Kraken's Aug. 12 brief notes the July figure landed against a forecast gain of roughly 83,000. Spending that outruns hiring has to be financed from somewhere.

Which brings the two prints back to policy. The Federal Open Market Committee left its target range at 3.50% to 3.75% at its July 28-29 meeting in a 9-3 vote, with Beth Hammack, Neel Kashkari and Lorie Logan all preferring to raise the range by a quarter point, according to the Fed's statement. Kraken's brief calls that the first time since September 2016 that three policymakers aligned on one directional dissent. Since then the market has walked back the hawkish case: FXStreet, in a piece published at 6:33 a.m. GMT this morning, reported the CME FedWatch Tool put the odds of a 25 basis point September increase at about 33%, against nearly 50% a week earlier. Those odds move daily.

The rates market is positioned for neither outcome in particular. The two-year Treasury yield closed Thursday at 4.140% and the 10-year at 4.641%, per Investing.com's quote pages, roughly 50 basis points apart. Both have drifted up in Friday's premarket, quoted at 4.154% and 4.664% in early trade — live quotes, not closes. The Fed's H.15 release puts the two-year constant maturity yield at 4.20% and the 10-year at 4.68% on Wednesday, Aug. 12, so the front end came down into Thursday. The front end is also where a surprise gets expressed.

Neither release settles anything alone. The minutes of the July 28-29 FOMC meeting arrive Wednesday, Aug. 19 at 2:00 p.m., per the Fed's own August calendar. Walmart releases second-quarter results at 6 a.m. Central the next morning, Aug. 20, with a call at 7 a.m. Central, according to the company's announcement. That sequence hands the macro reader a spending number, a survey, a policy transcript and then a corporate account of the same household inside four business days. The BEA's second estimate of Q2 GDP follows Aug. 26; the FOMC meets Sept. 15-16.

The frame for this morning is narrow. Watch the ex-auto and ex-gasoline measures rather than the headline, and check which definition a forecast used before scoring anything a beat or a miss. Watch the Michigan inflation expectations lines before the index level, and read the index against July's 55.2 rather than any one calendar's decimal. And watch whether the pair narrows the gap between mood and money, because a survey in the mid-50s alongside a consumption nowcast above 4% is not a stable configuration. One of those numbers eventually moves toward the other, and which one does it decides whether September is an argument about a hike or about nothing at all.