The numbers

The US Census Bureau's advance estimate landed at 8:30 am ET. Retail and food services sales came in at $773.9 billion for August, up 1.2% on July and 6.0% on August 2025. Consensus was +0.8%. July, which first printed at −0.6%, was revised up to −0.5%.

Take cars out and it improves rather than fades: sales excluding motor vehicles and parts rose 1.4%; excluding autos and petrol both, 1.2%. The control group — the slice the Fed and the GDP statisticians actually use — rose 1.4% against a forecast near 0.4%, after a 0.4% fall in July.

Month on month, by category (Census Table 2):

• Nonstore (online) retailers +2.6% — the biggest single contributor, and +9.9% YoY
• Gasoline stations +3.1% (+21.0% YoY)
• Electronics & appliances +1.6%; restaurants and bars +1.2%
• General merchandise +0.7%; car dealers +0.6%; groceries +0.5%
• The only fallers: building materials & garden −0.2%, department stores −0.8%

What it means

Start with "control group". It's total retail sales minus four noisy categories: cars, petrol, building materials and restaurants. Those four swing on things that aren't consumer choice — fleet deliveries, crude prices, the weather. What's left is the clean read on discretionary goods spending, and it feeds straight into the government's GDP arithmetic. A headline beat driven by expensive petrol would be noise. 1.4% against 0.4% expected isn't.

Second, the part the headline hides: these are nominal dollars. Census doesn't adjust for prices. August CPI ran +0.4% on the month and +3.4% on the year, per the Bureau of Labor Statistics. Petrol is the cleanest example — station sales are up 21.0% YoY while the BLS gasoline price index is up 27.4%. Americans are paying more and buying fewer litres. Net 3.4% inflation out of the 6.0% headline and you get roughly 2.5% of genuine extra stuff sold. Still growth, just not six percent of it.

The timing gives this teeth. The FOMC announces at 2:00 pm ET — 11:30 pm IST — with a fresh dot plot, and futures had about a 93% chance of a 25bp hike to a 3.75–4.00% range priced before this print, per CME FedWatch data cited by Kiplinger. A consumer this strong removes the last argument for going gently. The 10-year had already crossed 5% for the first time since 2007 on the way in.

Who it touches

• Online and big-box retail. Nonstore was the engine, which reads well for Amazon. General merchandise rose more modestly — relevant to Walmart, Costco and Target.
• Home improvement. Building materials was one of only two declining categories. With the 10-year at 5%, mortgage-linked demand is the soft spot — read across to Home Depot and Lowe's.
• Restaurants. Food services +1.2% on the month, +5.8% on the year — the demand backdrop names like Chipotle trade on.
• Rates, the rupee and Indian IT. Hot US data plus a hiking Fed lifts the dollar and US yields, pressures emerging-market currencies, and raises the discount rate on long-duration assets. Strong US demand helps the client end for Infosys and peers; a hiking Fed is the offset.

What to watch

Tonight's FOMC statement and, more than the rate itself, the Summary of Economic Projections — whether the dot plot shows more hikes after this one. Tomorrow 8:30 am ET brings August housing starts and jobless claims (forecast 209K, previous 206K), a test of whether the building-materials dip is a real housing crack. Then the Bank of England Thursday, Bank of Japan Friday. September retail sales land 15 October — and that release revises today's number, built from a subsample of roughly 4,800 firms.

Full schedule: MarketChacha calendar

As of 9:15 am ET / 6:45 pm IST, 16 Sep 2026. Sources: U.S. Census Bureau, Advance Monthly Retail Trade Survey (Aug 2026), U.S. Bureau of Labor Statistics CPI (Aug 2026), Trading Economics, Kiplinger (CME FedWatch). For discussion and education only — not investment advice. Verify before acting.