Housing continues to slowly weaken and completions surge relative to starts and permits.
Building Permits
- Privately-owned housing units authorized by building permits in September were at a seasonally adjusted annual rate of 1,428,000.
- This is 2.9 percent below the revised August rate of 1,470,000 and is 5.7 percent below the September 2023 rate of 1,515,000.
- Single-family authorizations in September were at a rate of 970,000; this is 0.3 percent above the revised August figure of 967,000. Authorizations of units in buildings with five units or more were at a rate of 398,000 in September.
Housing Starts
- Privately-owned housing starts in September were at a seasonally adjusted annual rate of 1,354,000.
- This is 0.5 percent (±13.0 percent) below the revised August estimate of 1,361,000 and is 0.7 percent (±16.1 percent) below the September 2023 rate of 1,363,000.
- Single-family housing starts in September were at a rate of 1,027,000; this is 2.7 percent (±13.4 percent) above the revised August figure of 1,000,000. The September rate for units in buildings with five units or more was 317,000.
Housing Completions
- Privately-owned housing completions in September were at a seasonally adjusted annual rate of 1,680,000.
- This is 5.7 percent (±19.9 percent) below the revised August estimate of 1,781,000, but is 14.6 percent (±11.9 percent) above the September 2023 rate of 1,466,000.
- Single-family housing completions in September were at a rate of 1,000,000; this is 2.7 percent (±7.8 percent) below the revised August rate of 1,028,000. The September rate for units in buildings with five units or more was 671,000.
Details from the Census Bureau New Residential Construction report.
Housing Starts Single Family vs Multi-Family
The trend on starts is negative led by multi-family.
Since May 2023, single-family starts have been steady on average, but in a very choppy fashion. The average over this period has been 1,006. Vs May 2023, single-family starts are up 2.8 percent.
Total starts dropped from 1,583 to 1,354. That’s a decline of 14.5 percent.
Multi-family starts from 584 to 327. That’s a plunge of 44.0 percent. Wow.
Housing Starts 1959-Present
To put the housing report numbers into proper perspective, starts are 18.3 percent below January of 1959.
Recent Economic Reports
- Retail Sales: Consumers Spend More on Food, Less on Gasoline and Cars
- Import and Export Prices: For the 2nd Month, Export Prices Decline Much More than Import Prices
- Industrial Production: Industrial Production Unexpectedly Declines 0.3 Percent With Huge Negative Revisions
Addendum
A reader asked about population-based comparisons. I replied …
Existing home sales are a much better model compared to prior years than housing starts and new home sales.
Housing starts are population based but one also has to factor in teardowns, a surge in immigration, the birth rate, and the rate of change in prime homeownership age say (25-40 or 28-48) or whatever.
I do not know how to adjust for all of those things. But a surge in immigration of 15 million in the past x years does require a lot of shelter.
That shelter has to come from somewhere. Is apartment construction enough? The data in this report does not imply anyone buying a home. At least multi-family doesn’t.
If we have not added enough apartments to accommodate immigration, teardowns, etc., my expectation of falling rents will be wrong or won’t last for long.
i’d suggest reading the wonderful easy read, “zillow talk”. probably the best book on how different markets in usa regions behave. in the places i’ve been an investor, and restorer of dozens of old houses in 4 distinct different regions of usa, i smell a nice pullback in 2 of those regions. in another few years, it might be time to buy back the places i sold 2 years ago.
When you (Mish) state “To put the housing report numbers into proper perspective, starts are 18.3 percent below January of 1959″ what you’re really saying is the relative increase in population and desire for new houses is 18% lower than in 1959. Once a house is built, it can be used for dozens (hundreds?) of years. New supply is intended to meet new supply in population.
In 1959, there were about 175M people growing by about 1.5M per year. So in 2024, the desire for new houses is 18% less than 1.5M or ~1.25M. I suspect the millions of border crossers are not likely immediately looking for a house to buy (they need to scam a bank into giving them a loan without a SSN). But current population is north of 330M, so 1.25M represents only 0.4% which is extremely low and will eventually cause prices to rise with such low supply.
I suspect normal population growth in the US averages 1% per year (3M), so should we expect these numbers to more than double to return to a normal housing starts rate figure?
Existing home sales are a much better model compared to prior years than housing starts and new home sales.
Housing starts are population based but one also has to factor in teardowns, a surge in immigration, the birth rate, and the rate of change in prime homeownership age say (25-40 or 28-48) or whatever.
I do not know how to adjust for all of those things. But a surge in immigration of 15 million in the past x years does require a lot of shelter.
That shelter has to come from somewhere. Is apartment construction enough? The data in this report does not imply anyone buying a home. At least multi-family doesn’t.
If we have not added enough apartments to accommodate immigration, teardowns, etc., my expectation of falling rents will be wrong or won’t last for long.
In construction, new privately owned 5+ units are down nonstop from 1,004K in July 2023 to 825K in Sept. Started are down again. Completed were up sharply from 91K in Jan 2011 to 735K in Aug, the largest move ever, before a slight decline in Sept. Completed inventory is the highest since 1974. Prices are too high. Rent is too high. Mortgage is too high. And the cost of labor and Land doubled since zero rates. Homebuilders are waiting for Kamala, an enigma.
Big corporate builders like Lennar have been subsidizing mortgage rates for a long time now. If they pump 5 or more points into buying the rate down, that means artificially inflating the sale prices. They can only do that for so long, since if recently-built homes come on the market (due to owners divorcing or being transferred), it will undercut values in the development.