The gap between two indexes

In the July manufacturing survey published by the Dallas Fed on July 27, the raw materials prices index eased to 41.3 from 42.4 and the finished goods prices index to 25.6 from 28.6. Both fell, but the gap between them held: firms report that the prices they pay are rising and the prices they can charge are rising markedly less. A printing manufacturer responding to the survey said tariffs and the war continue to create uncertainty and compress margins.[1]

In the same survey the activity side strengthened. The production index rose to 10.1 from 4.1, new orders to 6.4 from 2.3 and general business activity to 1.3 from 0.0, while the index of companies' own outlook gained 11.1 points to 13.4. Within that picture one heading moved the other way: employment, where the index fell to 12.2 from 13.9.[1]

Who carries the difference

A cost increase that does not pass into selling prices can go to three places: profit, hours or headcount. Production accelerating while the employment index eases suggests the first two are currently doing more of the work — the same or higher output produced with less new hiring. The manufacturers' own words point the same way: what is compressing is margin. Yet the employment index remains clearly positive at 12.2; that decline may equally be ordinary month-to-month movement, and reading a hiring slowdown out of a single month's change would be going too far.[1]

For the input side, relief appears to be coming in the months ahead. Crude fell sharply on Monday: Brent dropped 9.13 percent to $87.94 a barrel. Energy is both a direct input to manufacturing and a transport cost; if that decline persists, the raw materials prices index may ease further in the August survey. That the finished goods index would not fall as fast is the expected outcome, because price cuts follow cost cuts.[1], [2]

What this data does not show

The point to be careful about is the measure itself. This is a diffusion index: it counts how many firms report an increase and how many a decrease, not the size of either. It is therefore not possible to derive from it how far margins compressed, which pay band lost how many hours, or what happened to a household's income. The survey counts firms and does not count households; the answer to who the squeeze is booked against sits in wage and household data.[1]