Economists missed the construction spending mark by a mile this month.

The Econoday consensus was for spending to rise by 0.6%. Instead, construction spending fell by 0.6%.

Econoday finds strength in the numbers, I don’t.

Strength in residential building makes for a better construction spending report than indicated by the headline 0.6 percent July decline. Driven by single-family homes, residential construction rose a very solid 0.8 percent in the month for a year-on-year gain of 11.6 percent that contrasts markedly with the 1.8 percent overall rate. Home improvements, up 1.4 percent in the month, were also very strong. Spending on multi-family construction continues to moderate, down 0.8 percent in the month for only a 2.6 percent yearly gain.

The real weakness in the report is on the nonresidential side where private spending, reflecting weakness across all components and especially commercial building, fell 1.9 percent for a yearly decline of 3.6 percent. Public building is likewise soft with the educational category down 4.4 percent in July.

Though housing permits have been flat, the residential numbers in this report are solid. And strength for construction payrolls in this morning’s employment report might be hinting at better results for construction spending in the August report. Note that the unfolding effects from Hurricane Harvey will be difficult to gauge based on mixed results following prior hurricanes.

Strength?

Both new and existing home sales reports have been weak. Family formation is weak.

The alleged strength in this report comes not from building more houses, but from building more expensive houses.

Job growth came from building more retail stores and food establishments.

If commercial construction spending has turned, employment will follow.https://t.co/xcbHWN5ttw

— Mike Mish Shedlock (@MishGEA) September 1, 2017

However, construction spending is extremely volatile and heavily revised. All of the numbers in this report are suspect actually.