Goldman's Take On The ISM Number, Which Leapfrogged The Top Of The Economist Expectations Range
The ISM Number of 56.3 came higher than the top of the range of what every single economist had been predicting, which topped at 56.0. But at least the administration works in mysterious, if not so subtle ways. Here is Goldman's explanation for what caused the unexpected surge.
ISM Up on Output/Employment; Construction Falls from Downward-Revised Base
BOTTOM LINE: While we and almost all other forecasters had looked for a sizable decline, the ISM manufacturing survey rose slightly in August. The headline increase was chiefly driven by a strong read in the production index while new orders fell slightly, almost closing a gap between the orders and inventories indexes that was more than 20pts just three months ago. Construction outlays fell, from a base that was revised down sharply, with weakness tilted toward residential spending.
ISM mfg index +10 (5, +2), with -1 judgmental adjustment for composition.
Construction outlays -3 (1, -3), with -2 judgmental adjustment for large downward revisions.
ISM mfg index up 0.8 pts to 56.3 in Aug vs. GS 52, median forecast 52.7.
Construction outlays -1.0% in July (mom, -7.9% yoy) vs. GS -1.6%, median forecast -0.5%.
1. The Institute for Supply Management (ISM) surprised on the upside, rising 0.8 point to 56.3 in August. This increase was mainly driven by stronger production and employment indexes (which rose by 2.9 points to 59.9 and 1.8 points to 60.4, respectively). New orders and supplier deliveries, however, both declined (from 53.5 to 53.1 and from 58.3 to 56.6, respectively). As the inventories index rose 1.2 points to 51.4, the difference between orders and inventories fell by 1.6 points to 1.7. Barely three months ago, this gap stood at 20.1. It is largely on this basis that we have applied a -1pt judgmental adjustment to the US-MAP reading, which still leaves +10 in the aggregate index. Regarding the components that do not enter the headline, prices paid rose by 4 points to 61.5 and customer inventories rise by 4.5 points to 43.5.
2. Construction outlays fell in July - more than anticipated by the median forecast but less than we had expected. However, data for May and June were both revised down significantly. The level of outlays in June appears to be about 2¾% below its previous level, prompting the -2pt judgmental adjustment noted above. The weakness, both in the preliminary gain reported for July and in the revisions, was concentrated in the residential component. Overall, the report confirms weak paths into Q3 in both residential and private nonresidential spending on structures.