The July inflation report, released this morning, carried a number the White House will like and a set of reasons to hold the champagne. Consumer prices rose 3.4 percent over the year through July, down from 3.5 percent in June, and up just 0.1 percent on the month [1][2]. That is the second straight month of deceleration, off May's three-year high of 4.2 percent. Core inflation, which strips out food and energy, eased to 2.5 percent from 2.6 percent [1][2]. Both readings matched what forecasters expected almost to the decimal, which means the market, the Fed, and anyone who had been paying attention already knew roughly what today would say. A note on sourcing before the numbers do any work: the Bureau of Labor Statistics page returned a 403 error and could not be read directly, so every figure here is attributed to Trading Economics, which carries the official CPI series, and to CBS News, which reported the release [1][2].
Start with what a 0.1 percent month actually means at a kitchen table, because that is the number doing the least talking today. A household spending $5,000 a month across its whole budget paid about $5 more in July than in June on the same basket [1]. That is the entire monthly story: not a windfall, not a squeeze, a rounding error you would not feel. The year is where the weight sits. That same 3.4 percent, compounded over twelve months, is what turned last summer's prices into this summer's, and none of it reverses because one month came in quiet.
The part that fell is the part that always moves the fastest. Gasoline dropped 2.9 percent on the month, and the pump showed it: the average gallon ran $4.06 in July, roughly 10 cents cheaper than June [1][2]. That is welcome for anyone with a commute, and it is also a story about crude, not about the underlying trend. Gasoline is still up 24.6 percent from a year ago, and at $4.06 it sits about 35 percent above the roughly $3.00 average of February, before the Iran war pushed oil higher [1][2]. What looks like disinflation at the pump is mostly a war premium bleeding back out. Energy did the visible work here, the way it did in June, and energy is the one component that can hand the number back next month.
Now the part that did not move. Shelter rose 3.2 percent over the year, barely down from 3.3 percent, and it remains the single largest sticky weight in the index [1]. Rent and the cost of owning do not swing with a strait or a truce; they grind, and they are still grinding at better than three percent while the headline advertises 3.4. That gap is the whole reason core sits at 2.5 percent rather than down near two. When the cheap energy comparison rolls off, shelter is what will be left holding the average up.
Here is the ratio that matters to a paycheck. Wage growth reached 3.2 percent in July, and prices rose 3.4 [2]. A worker earning $60,000 a year who got the average raise took home about $1,920 more; the same year of inflation cost roughly $2,040 in purchasing power on that income, leaving them about $120 behind before the year is even out. 'Inflation is wiping out wage gains for many,' the economist Heather Long noted [2]. A falling headline and a shrinking real wage are not a contradiction. They are what happens when prices slow down but still climb faster than pay.
That leaves the question of what the Federal Reserve owes this report, and the honest answer is close to nothing. The Committee has held the funds rate at 3.50 to 3.75 percent, and Chair Kevin Warsh told Congress on July 14 that its members have 'no tolerance for persistently elevated inflation' [3]. A print that arrives exactly on consensus does not vindicate a cut or force a hike; it confirms the path everyone already priced. No same-day Fed reaction to this number exists, and none is required, because a number nobody was surprised by does not move a decision by itself. The September meeting will be argued over jobs and over whether shelter finally cracks, not over a tenth of a point that landed where the forecasts pinned it.
One bookkeeping correction, since our own archive is where the confusion started. A month ago this desk reported that June CPI fell to 3.5 percent on gasoline [1], and that figure holds: July's 3.4 percent extends exactly that trend. A separate piece two weeks later flagged June PCE, the Fed's preferred gauge, printing 4.1 percent. Those are two different rulers measuring two different baskets, and the 4.1 belongs to PCE, never to CPI. Any follow-up note that carried June CPI at 4.1 percent conflated the two; the CPI number for June was 3.5, full stop.
The record for July reads like this. Inflation is cooling, and the cooling is real. The visible relief came from a pump price giving back a war premium, the sticky cost of keeping a roof overhead is still running above three percent, real wages slipped again for the average earner, and a report that surprised no one handed the Fed no new marching orders. The headline fell a tenth. The reasons it may not keep falling were printed in the same table.
