The Bureau of Economic Analysis reported Thursday that the economy grew at a 3.4% annual rate in the third quarter, up from 2.2% in the second and above the 3.2% Reuters poll. In the same release, the PCE price index, the Fed's preferred inflation gauge, rose at a 2.1% rate for the quarter, down from 5.0%.

Read those two numbers alone and it looks like a soft landing: strong growth, inflation near the Fed's 2% goal.

Released alongside it, September's monthly data showed PCE prices up 0.5% in the month and 3.6% from a year earlier, with core inflation at 3.1%. The day before, the Fed had held rates at 3.75% to 4% with two officials voting to raise them.

Nobody misreported anything. The release covered four different time windows and several different ways of slicing the economy. Here is how to read each part.

The economy did not grow 3.4% between July and September. It grew 0.8%, and BEA states that number in its technical notes. The 3.4% is that quarterly change compounded over a year, the convention every GDP headline uses.

That matters for comparisons. Your raise is a 12-month number, and so is year-over-year inflation. Line those up against an annualized quarterly GDP figure and you are comparing a year with a quarter.

GDP is a sum. Here is what added up to 3.4 percentage points, from BEA's detailed tables as compiled by Global Economy Charts:

Three of those lines need context.

Inventories and imports go together. Businesses rebuilt stock after five straight quarters of running it down, which added 1.61 points. Much of that stock came from abroad: imports rose 19.7% and subtracted 2.64 points. Netted, the restocking and the imports that supplied it come to about −1.0 point. Both lines are among the most heavily revised in GDP, so read them as a pair and expect them to move.

The government line is mostly accounting. Most of the 0.90-point government contribution came from the Strategic Petroleum Reserve selling less oil than it did in the second quarter. BEA counts those sales as a reduction in government spending, so smaller sales show up as an increase. BEA's own note: "there is no direct effect on GDP," because the oil shows up elsewhere in the accounts.

The equipment boom is concentrated. Business equipment spending rose 20.9%, the fastest since 2020. BEA says it was led by "information processing equipment (mainly computers and peripheral equipment)," and Reuters ties it to AI infrastructure. That spending is real, but it comes from a small number of very large buyers. Residential investment, the part of the economy tied to mortgage rates, fell 3.3%.

BEA publishes a cleaner measure that leaves out trade, inventories and government: real final sales to private domestic purchasers, meaning consumer spending plus private fixed investment. It rose 4.9%, up from 4.6%, and faster than GDP itself.

That is the best single reading of underlying private demand, and it was strong. Consumer spending rose 3.8%, the same pace as last quarter, and accounted for about three-quarters of the headline. BEA says services growth was led by health care (hospitals and outpatient care) and restaurants, and goods growth by prescription drugs.

The unflattering reading is that demand is strong because spending on health care and AI hardware is strong. Housing is shrinking, and exports fell 5.7%, led by petroleum goods.

This is the gap that produced the confusing headlines.

A quarterly average can look calm while its last month speeds up. The quarterly number says the second-quarter spike faded. The September number says prices picked up again at the end of the quarter, and the year-over-year number says the past 12 months have been expensive.

For your budget, the year-over-year number is the one to use. For a sense of direction, the latest monthly number is the best indicator, and it pointed up.

The Fed's October 28 statement said: "Economic activity is expanding at a strong pace... Productivity growth is strong, and capital investment is robust." And: "Inflation remains elevated." Governors Lisa Cook and Beth Hammack dissented in favor of a quarter-point increase.

Read the GDP report the way the Fed does and the dissents make sense. Private demand grew 4.9%. Nominal GDP, before adjusting for inflation, grew 5.6%. Monthly inflation sped up in September. Nothing in that combination calls for a rate cut.

If you're a worker: compare your raise to the 3.6% year-over-year figure, not the 2.1% quarterly one. If your raise is below 3.6%, your pay bought less this September than last.

If you're buying a home: the economy's strength is part of why mortgage rates stay high. Freddie Mac's 30-year average was 7.24% last week, up from 6.17% a year earlier. A 3.4% growth headline does not bring rates down.

If you run a business: the inventory rebuild is done or close to it. If your customers restocked in the third quarter, expect fewer reorders in the fourth.

If you invest: growth is concentrated in AI equipment and health care. Strong overall GDP does not mean a broad expansion. Housing-linked sectors shrank.

If you follow the data: BEA's second estimate comes November 25, with gross domestic income alongside it. When GDI and GDP disagree by a lot, one of them gets revised toward the other.

The skeptical reading may be too skeptical. Final sales to private domestic purchasers is designed to strip out noise, and it rose 4.9%, which is strong by any standard. Business investment growing 12.7% overall, not just in equipment, is the kind of spending that raises productivity later, and the Fed called productivity growth "strong."

The 2.1% quarterly inflation figure is also real progress against the second quarter's 5.0%. If September's 0.5% was a one-month energy bump, the pickup may not last.

Claim: BEA's advance estimate for fourth-quarter 2026 GDP, scheduled for late January 2027, will show private inventories subtracting from growth after adding 1.61 points in the third quarter.

Stated confidence: 65% · Verification date: BEA Q4 2026 advance estimate (late January 2027) · Status: OPEN

What would make us wrong: businesses keep building inventory ahead of expected tariff changes or supply disruptions, or the second and third estimates revise the third-quarter rebuild down enough that the fourth quarter has room to keep adding.

This newsletter is the free version of what I do: take the headline number, break it into its parts, and show which parts carry the weight.

If there's an economic figure you keep seeing quoted, reply with it. I'll tell you what window it covers, what it leaves out, and whether it supports the claim attached to it.

cokas.io — no sales call, ever.

Know someone who read "3.4% growth, 2.1% inflation" and stopped there? Forward them this. They can get the next edition free at thepatternbrief.com.

Every prediction in The Pattern Brief carries a verification date and is revisited in a future edition, including this one.