(Click on image to enlarge)

The Census Bureau’s New Home Sales Report shows a surge in sales for July with a huge positive revision to June.

New Home Sales Key Points

• New Home Sales Sales of new single-family houses in July 2024 were at a seasonally adjusted annual rate of 739,000.
• This is 10.6 percent (±16.5 percent) above the revised June rate of 668,000 and is 5.6 percent (±21.3 percent) above the July 2023 estimate of 700,000.

Note the astonishing margins of error in this report.

The revised June rate was 668,000 up from 617,000. From the previous report, sales are up ((739,000 – 617,000) / 617,000) * 100 = 19.8 percent.

I doubt that sticks, but assume it does. How much would that add to 2024 Q3 GDP? Ponder that, and I will discuss below.

Median and Average New Home Sales Price

(Click on image to enlarge)

If you are a homeowner and wish to thank someone for the above chart, including property taxes, insurance, maintenance, etc., please thank the Fed.

Sales Prices

• The median sales price of new houses sold in July 2024 was $429,800.
• The average sales price was $514,800.

New Homes For Sale By Stage of Construction

(Click on image to enlarge)

For Sale Inventory and Months’ Supply

• The seasonally-adjusted estimate of new houses for sale at the end of July was 462,000.
• This represents a supply of 7.5 months at the current sales rate.

Those numbers are from the Census Department. They are bogus.

The department counts homes that have not been started and may not be started for months or even years as homes for sale.

Of the allegedly 467,000 homes for sale, only 102,000 are complete. 100,000 have not even been started. 362,000 are started or completed.

Supply of Homes

(Click on image to enlarge)

From a fictious number of new homes for sale, we can calculate a fictious supply of new homes at 7.5 months.

Mish Headlines Three Prior Months

(Click on image to enlarge)

For starters, anyone who has any faith in today’s report, shouldn’t. But’s let’s do some what if calculations assuming they are.

Impact on GDP

The numbers look huge but that is the result of annualizing sales.

The BLS revised the June NSA new home sales numbers for May to June from 53,000 to 58,000. Today we are at 64,000. I failed to capture the April to May NSA revision but let’s assume the total increase is ~14,000 NSA.

14,000 homes * $514,800 average price = $7.21 billion. But that is spread out over 7-12 months. Much of the value added happens towards the end, especially completion.

If we divide $7.21 billion by four, the impact to GDP will be $1.8 billion.

However, I did note seasonally adjusted completions fell from 96,000 to 93,000 in May and from 102,000 to 99,000 in June. They are now back up to 102,000 in July.

A Big Nothing?

Since completions went nowhere, I am not convinced there is any positive impact from this report, as remarkable as that may seem.

In the spirit of being positive, I will take a stab at +0.05 percentage points contribution to residential construction in the GDPNow forecast on Monday.

I have never tried this calculation before, so take that under consideration.

Also, much depends on what the model expects. If the model expected more completions we could easily see a decline. And if the model expected less we could see a bigger jump, perhaps to 0.20 percentage points.

A big part of the problem here is figuring out what the model expects. My bottom line assumption is that it all adds up to barely anything.

No Impact on Recession Call

Given my projections, this does nothing to my recession forecast.

August 20: Improving the McKelvey Recession Indicator, No False Negative or Positive Signals

August 22: A Breakdown, by Sector, of the Negative 818,000 BLS Job Revisions

It’s important to note my above call pertains only to the contribution to residential construction.

I have no idea, literally none, what the -818,000 negative job revision will do to GDPNow.

More By This Author: