November’s construction spending, et al

With the government still on a holiday schedule up until New Year's, the only major agency report released the past week was the November report on Construction Spending from the Census Bureau, typically scheduled for the first of the month...this week also saw the results from last regional Fed manufacturing survey for December: the Dallas Fed Texas Manufacturing Outlook Survey reported their general business activity composite index rose to +3.4, up from November’s –2.7 and its first positive reading since April 2022, meaning that a small plurality of Texas manufacturing executives saw improving business conditions for the first time in 32 months…

Privately issued reports released this week included the light vehicle sales report for December from Wards Automotive, which estimated that vehicles sold at a 16.80 million annual rate in December, up from the 16.50 million million annual pace of vehicle sales reported for November, and up from the 15.83 million vehicle sales rate reported for December of 2023, and the widely watched manufacturing purchasing manager’s survey from the Institute for Supply Management (ISM): the December Manufacturing Report On Business® indicated that the manufacturing PMI (Purchasing Managers Index) rose to 49.3% in December, up from 48.4% in November, with the below 50% reading indicating that a fairly small plurality of manufacturing purchasing managers are still reporting worse conditions in various facets of their business in December than in November...in addition, this week saw the Case-Shiller Home Price Index for October from S&P Case-Shiller, which doesn’t even report any prices of homes, but just a index generated by averaging relative sales prices of homes that sold during August, September and October against the sales prices of the same homes when they sold during prior 3 month periods going back to January of 2000….by comparing repeat home sales prices on that basis, the Case-Shiller report indicated that their national home price index for those 3 months averaged 3.6% higher than it did for the same homes that sold during the same 3 month period a year earlier, which was lower than the 3.9% year over year index increase indicated by their September report, which indexed sales prices from July, August and September, and also down from the 4.8% year over year increase indicated by their October 2023 report, indexing homes that sold during August, September and October of last year…

Construction Spending Flat in November After Prior Two Months Were Revised 1.0% Lower

The Census Bureau’s report on construction spending for November (pdf) estimated that the month’s seasonally adjusted construction spending would work out to $2,152.6 billion annually if extrapolated over an entire year, which was "virtually unchanged" (±1.0 percent)* from the revised October annualized estimate of $2,152.3 billion, but was 3.0 percent (±1.5 percent) above the estimated annualized level of construction spending in November of last year….for the first eleven months of this year, construction spending amounted to $1,986.8 billion, which was 6.5 percent (±1.2 percent) above the $1,866.0 billion spent on construction over the same period in 2023…

The annualized October construction spending estimate was revised nearly 1.0% lower with this release, from $2,174.0 billion to $2,152.6 billion, while the annual rate of construction spending for September was revised more than 1.0% lower, from $2,164.7 billion to $2,142.4 billion….since that’s an annualized figure, that $22.3 billion downward revision to September’s construction spending would be averaged over the 3 months of the 3rd quarter to produce an annualized quarterly change, and lower the annualized 3rd quarter construction figures used in the GDP report by more than $7.4 billion, before any inflation adjustment, which would imply that the 3rd estimate of 3rd quarter GDP growth was overstated by roughly 0.14 percentage points, a correction which will not be applied to published GDP figures until the annual revision is released at the end of next summer…

A further breakdown of the different subsets of construction spending is provided in a Census summary, which precedes the detailed spreadsheets:

• Private Construction. Spending on private construction was at a seasonally adjusted annual rate of $1,650.7 billion, 0.1 percent (±0.5 percent)* above the revised October estimate of $1,649.8 billion. Residential construction was at a seasonally adjusted annual rate of $906.2 billion in November, 0.1 percent (±1.3 percent)* above the revised October estimate of $905.1 billion. Nonresidential construction was at a seasonally adjusted annual rate of $744.5 billion in November, virtually unchanged from (±0.5 percent)* the revised October estimate of $744.6 billion.
• Public Construction. In November, the estimated seasonally adjusted annual rate of public construction spending was $501.9 billion, 0.1 percent (±1.6 percent)* below the revised October estimate of $502.5 billion. Educational construction was at a seasonally adjusted annual rate of $107.0 billion, 0.2 percent (±2.5 percent)* below the revised October estimate of $107.2 billion. Highway construction was at a seasonally adjusted annual rate of $142.9 billion, 0.2 percent (±4.3 percent)* above the revised October estimate of $142.7 billion.

As you can infer from that summary, construction spending is included in 3 subcomponents of GDP; investment in private non-residential structures, investment in residential structures, and in government investment outlays, for both state and local and Federal governments…however, getting an accurate read on the impact of the November spending reported in this release on 4th quarter GDP is difficult because all figures given here are in nominal dollars and as you know, data used to compute the change in GDP must be adjusted for changes in price…that’s a problem because the National Income and Product Accounts Handbook, Chapter 6, lists a multitude of privately published deflators that are used by the BEA for the various components of non-residential investment, ie, such as the Handy-Whitman construction cost indexes for electric light and power plants and for utility construction…in lieu of trying to find and adjust for all of the obscure price indices the BEA uses separately, we’ve opted to use the producer price index for final demand construction as an inexact shortcut in an attempt to make a reasonable price adjustment on the total..

That price index showed that aggregate construction costs were unchanged in November, after being down 0.2% in October, but after rising 0.1% in September and being unchanged in August…on that basis, we can estimate that construction costs for November were down 0.2% from September, down 0.1% from August, and down 0.1% from July, while they were obviously unchanged from October…we then use those percentage change differences to adjust the spending figures for each of those 3rd quarter months against October and November, which is arithmetically the same as adjusting lower priced October and November construction spending upward, for purposes of comparison…annualized construction spending in millions of dollars for the third quarter months is given as $2,142,427 for September, $2,162,132 for August, and $2,143,139 for July, while it was at annual rates of $2,152,250 in October and $2,152,581 in November…. thus to estimate the difference between our “inflation adjusted” construction spending of the two recent 4th quarter months and those of the third quarter, our calculation would be ((2,152,581+ 2,152,250 * 1.000)/2) / (( 2,142,427 * 0.998 + 2,162,132 * 0.999 + 2,143,139 * 0.999) / 3) = 1.00281695, meaning average real construction over the months of October and November was up 0.282% vis a vis the 3rd quarter….in NIPA terms, that means real construction for the 4th quarter has risen at an annual rate of 1.13% from that of the 3rd quarter so far…

To estimate the effect of that change on GDP, we figure the difference between the third quarter ‘inflation adjusted’ average and that of October and November and take the annualized result of that as a fraction of the inflation adjusted 3rd quarter GDP figure, and find that real construction spending is rising at a rate that would add about 0.12 percentage points to 4th quarter GDP, should real December construction continue at the same pace as that of October and November…

 

 

(the above is the synopsis that accompanied my regular sunday morning news links emailing, which in turn was mostly selected from my weekly blog post on the global glass onion…if you’d be interested in receiving my weekly emailing of selected links, most of which are picked from the aforementioned GGO posts, contact me…)