Two readings, one day
Williams told Reuters that the current stance of interest rate policy is well positioned to bring inflation back to 2 per cent, that his own forecast is for inflation to come down in the second half of this year and further next year, and that it would absolutely be appropriate to act if the economy is not on that trajectory. His supporting argument is specific: probably the vast majority or a large part of the effects of the tariffs already in place have now passed through into prices. The federal funds target range is 3.50 per cent to 3.75 per cent after last week's hold, and futures traders have priced in a decent chance of a rise by the end of the year.[1]
The manufacturing survey published the same morning gives that argument a rough edge. The headline index came in at 55.6 per cent, 2.3 percentage points above June's 53.3 per cent, with production at 58.5 per cent, employment back in expansion at 52.8 per cent and the backlog of orders at 55.0 per cent. The prices index eased to 71.1 per cent from 73.0 per cent, and it remained the highest component in the release by a wide margin.[2]
Where the cost is sitting
These two readings are not in contradiction, because they measure different things. A diffusion index of prices records how many firms report paying more for inputs, not how much of that increase reached a household. A figure of 71.1 per cent means the breadth of input cost increases barely narrowed. That is consistent with Williams's account, in which the tariff cost has already moved through to consumer prices and is now working out of the annual comparison. It is equally consistent with the opposite account, in which the cost has arrived at the factory gate and is being held on producer margins because demand will not carry another increase. A survey of purchasing managers cannot tell you which of those happened.[2], [1]
Which one it is depends on demand, not on the tariff schedule, and the rest of the survey speaks to demand. Production at 58.5 per cent and a backlog of orders that rose to 55.0 per cent from 50.5 per cent describe factories running against their order books rather than chasing work. A firm in that position has more room to pass a cost forward than one with an empty book. So the same release that leaves the pass-through question open also tells us the conditions under which it would be settled in the producer's favour, with the household paying rather than the margin absorbing.[2]
The question the interview leaves open
The interesting part of the interview lies in the premise underneath the conditional promise to act, rather than in that promise itself. If the tariff cost has genuinely finished moving through to prices, then the disinflation Williams forecasts arrives by arithmetic as last year's increases drop out. If instead a portion is parked on margins in a manufacturing sector that has just posted its strongest headline since May 2022, then the same demand strength that produced 55.6 per cent is what would release it. In that case the passing of the pass-through would be a timing statement rather than a completed event, and the trajectory Williams says he would act on is the one that gets rewritten.[2], [1]