Written by Steven Hansen
The headlines say construction spending was unchanged and below expectations. Our view is worse than the headlines..
Analyst Opinion of Construction Spending
There was significant backward revision starting in January 2015. The rolling averages declined. Also note that inflation is grabbing hold – and the inflation adjusted numbers are worse than the headlines – but still better growth than real GDP.
The employment gains year-over-year are now better than the year-over-year improvement in construction spending..
Econintersect analysis:
- Growth decelerated 0.8 % month-over-month and up 4.7 % year-over-year.
- Inflation adjusted construction spending up 3.6 % year-over-year.
- 3 month rolling average is 5.6 % above the rolling average one year ago which is a 0.4 % deceleration month-over-month. As the data is noisy (and has so much backward revision) – the moving averages likely are the best way to view construction spending.
- Backward revision for the last 3 months were strongly upward..
- Unchanged month-over-month and up 4.5 % year-over-year.
- Market expected from Bloomberg / Econoday 0.0 % to 0.9 % month-over-month (consensus +0.5) versus the 0.0 % reported
Construction spending (unadjusted data) was declining year-over-year for 48 straight months until November 2011. That was four years of headwinds for GDP.
Indexed and Seasonally Adjusted Total Construction Spending (blue line) and Inflation Adjusted (red line)
This month’s headline statement from US Census:
Construction spending during May 2017 was estimated at a seasonally adjusted annual rate of $1,230.1 billion, nearly the same as (±2.5 percent)* the revised April estimate of $1,230.4 billion. The May figure is 4.5 percent (±2.5 percent) above the May 2016 estimate of $1,177.0 billion. During the first 5 months of this year, construction spending amounted to $469.2 billion, 6.1 percent (±1.3 percent) above the $442.4 billion for the same period in 2016.
PRIVATE CONSTRUCTION – Spending on private construction was at a seasonally adjusted annual rate of $943.2 billion, 0.6 percent (± 0.7 percent)* below the revised April estimate of $949.3 billion. Residential construction was at a seasonally adjusted annual rate of $509.6 billion in May, 0.6 percent (±1.3 percent)* below the revised April estimate of $512.7 billion. Nonresidential construction was at a seasonally adjusted annual rate of $433.6 billion in May, 0.7 percent (± 0.7 percent)* below the revised April estimate of $436.7 billion.
PUBLIC CONSTRUCTION – In May, the estimated seasonally adjusted annual rate of public construction spending was $286.9 billion, 2.1 percent (±5.3 percent)* above the revised April estimate of $281.0 billion. Educational construction was at a seasonally adjusted annual rate of $74.3 billion, 5.1 percent (±3.3 percent) above the revised April estimate of $70.7 billion. Highway construction was at a seasonally adjusted annual rate of $90.6 billion, 0.9 percent (±16.9 percent)* below the revised April estimate of $91.5 billion.
Unadjusted Total Construction Spending Year-Over-Year (blue line) and Month-over-Month (red line) Change
Unadjusted Total Construction Spending Year-Over-Year (blue line) Growth
Unadjusted Private Construction Spending Year-Over-Year (blue line) and Month-over-Month (red line) Change
Unadjusted Public Construction Spending Year-Over-Year (blue line) and Month-over-Month (red line) Change
Private construction had been fueling construction growth.
Caveats on the Use of Construction Spending Data
Although the data in this series is revised for several months after issuing, the revision is generally minor. This series is produced by sampling – and the methodology varies by sector being sampled.
The headline data is seasonally adjusted. Econintersect uses the raw unadjusted data.Econintersect determines the month-over-month change by subtracting the current month’s year-over-year change from the previous month’s year-over-year change. This is the best of the bad options available to determine month-over-month trends – as the preferred methodology would be to use multi-year data (but the New Normal effects and the Great Recession distort historical data).
The data set for construction spending is not inflation adjusted. Econintersect adjusts using the BLS Producers Price Index – subindex New Construction (PCUBNEW-BNEW). However in the inflation adjusted graph in this post, FRED does not have this series – andEconintersect has used Producer Price Index: Finished Goods Less Energy (PPIFLE), Monthly, Seasonally Adjusted which has similar characteristics.
Construction (which historically is an major economic driver) is a literal shadow of its former self. Its contribution to GDP is down $400 billion from its peak level in 2006. The main driver of construction spending is the private sector. Here is the historical breakdown. The graph below uses US Census seasonally adjusted data.
Obvious from the above graph that public spending on construction is falling off, while private spending is slightly trending up. The overall effect is that construction spending is near the same place it was in early 2010.
include(“/home/aleta/public_html/files/ad_openx.htm”); ?>
