September new housing construction: saying farewell to the late incipient housing recovery

 - by New Deal democrat

Last month I started my report on housing construction by writing that “This morning’s report on housing construction may be the last decent one for quite awhile.” And it was. 

Up until last month I had been writing that housing construction was most consistent with an economy coming out of a recession than going in to one. Needless to say, with the spike in mortgage rates, that is no longer the case. In September housing starts (blue) declined -5.2% to 1.221 annualized, while the less volatile and slightly more leading permits (gold) declined -0.2% to 1.400 million. The least volatile metric of all, single family permits (red, right scale) declined -4.0% to 845,000:

All three of these metrics had been trending essentially sideways since the summer of last year. As of this month, the YoY comparisons are now near the bottom of that range, with starts down -7.4%, permits by -3.0%, and single family permits by -4.8%:

This doesn’t mean housing is forecasting a recession yet. While I won’t bother with the graph this month, in the past these three metrics have generally been down -10% or more YoY to be consistent with the onset of a recession.

Now let’s update the “last shoe to drop” in this series, housing units under construction. This was unchanged on an annualized basis at 1.280 million units:

This series has been essentially flat - even rising slightly - as well since the beginning of this year, which is typical of an economic recovery that is beginning or has already begun:

Which, if it weren’t for rising mortgage interest rates would be good news.

So let’s close out with graph of mortgage rates (blue, inverted) on a monthly basis, showing that they rose to an average of 6.86% in September and 7.38% so far in October:

As the above graph shows, a similar increase in mortgage rates in late 2023 (noisily) led to roughly a -10% decline in permits issued (gold, right scale) over the next few months. With the further increase in mortgage rates this month, I expect further deterioration in new housing construction over the next several months as well.

One final note:  from a forecasting perspective, the housing market, along with corporate profits and real retail sales per capita, is one of the three nonfinancial, or “real world” long leading indicators that is of added importance for whether the US economy continues in expansion or falls into recession. Despite the renewed downturn in housing, as of now, both of the other two are still very positive. So - barring particularly strong new lunacy from Washington - the inflationary expansion appears likely to continue over the near term.

Consumer spending, income and wealth

 

 - by New Deal democrat

For awhile now I have been writing about how the wealth effect has been powering sales among upper income households. Earlier this year the Minneapolis Fed took a look at the phenomenon, comparing data from several different sources, including Moody’s:

The New York Fed:

And Bank of America:

They noted that “While Moody’s pegs a sharp break after 2020, Bank of America did not observe a widening gap in their data until mid-2025, when a large separation in spending growth opened between lower-, middle-, and upper-income customers—arguably more ‘E’ than ‘K’”

Interestingly, while they don’t supply any explanatory data, in their conclusion they note:

For each measure above, households are stratified by income. This is a natural consequence of the available data; from credit card issuers to market research firms to U.S. government surveys, income is the common denominator.

“However, wealth—rather than income—increasingly distinguishes the most well-to-do U.S. households. According to the Federal Reserve Board, household wealth has become increasingly concentrated among the top 0.1 percent. And there is ample evidence that wealth (independent of income) empowers spending, with asset-rich individuals able to borrow against their holdings and fund very high levels of spending.

This is particularly noteworthy because in the past four years there has been a huge divergeance between income and wealth, at least as measured by stock market gains. As shown in the graph below, total personal income has risen 23% since September 2022, while the stock market has doubled - it’s up 100%!

All of this adds to the evidence that at the moment the US economy *is* the stock market. Which also means that, if all else stays the same (or reasonably close thereto), a sharp selloff in the stock market is likely to prompt a sharp decline in consumer spending - which can easily become a vicious cycle, which would be broken only when sales come more in line with income growth (or, heaven forbid, decline).

Weekly Indicators for October 12 - 16 at Seeking Alpha

 

 - by New Deal democrat

My “Weekly Indicators” post is up at Seeking Alpha.

The tug of war between the AI-powered inflationary expansion, and the Big Bad Billionaire Bust-out Bill and Iran war-related increase in interest rates and gas prices continues. In the present, so far AI is winning. But out on the horizon, enough damage has accrued that (for this week at least) the long leading indicators have turned from positive to neutral.

As usual, clicking over and reading will bring you up to the virtual moment as to the state of the economy, and reward me with a penny of two in lunch money for my efforts collecting and organizing it all for you.

Insutrial production, primarily powered by AI-related utilities, continued its rebound in September

 

 - by New Deal democrat

One of the biggest, if not the biggest, trend I have been writing about is the rebound in manufacturing that has been driven by AI data center construction. In that regard, industrial production, one of the four important monthly series that the NBER looks at in determining whether the economy is expanding or contraction, has been an important indicator of that upward trend, and that continued in September.

Headline production (blue) rose a sharp 0.7% to another post-pandemic high, and manufacturing production (red) rose 0.3%. And electric and gas utility production (gold, right scale), which is most closely aligned with AI data center construction, increased sharply against, by 3.5%. The below shows all three normed to 100 as of just before the pandemic:

Utility construction has outpaced the manufacturing sector as well as the headline number in the past few years, and continued to do so in September. On a YoY basis, while headline production was up 2.2% and manufacturing production up 1.1%, utility production continued to be up sharply, by 8.7%:

So the big news from industrial production is that AI data center construction continues to be the centerpiece of the manufacturing rebound since last year.

Retail sales continue this year’s positive trend, suggest jobs market likely to improve as well

 

 - by New Deal democrat

As per usual, real retail sales is one of my favorite economic indicators, because it says so much about the health of the US consumer, and since consumption leads employment (noisily), also forecasts the trend in job creation.

Normally I don’t spend a lot of time on nominal retail sales, but this year that is more useful. That’s because the weekly Redbook index, which tracks nominal retail spending YoY, has been accelerating, up as much as 11% YoY (in fact this week it was 10.5% higher YoY). Here’s what the past five years of that index look like:

It was white hot in 2022, cratered in 2023, and has been accelerating back higher ever since, including this year.

But was it telling an accurate story? 

That’s where nominal retail sales come in. These were up 0.8% in September, and higher by 6.3% YoY. Here is what the YoY% changes in that metric look like over the past five years:

This is the exact same pattern as in the Redbook graph above. In other words, Redbook’s weekly updates have been telling us a true story (which is one reason why I track them every week as part of my Seeking Alpha weekly roundup of data).

But as useful as nominal information can be, the inflation adjusted data is more important. And since consumer inflation was higher by over 0.6% in September, real retail sales rounded to only a 0.1% gain. Still, note that there has been a sharp increase in real retail sales this year, although they remain lower than during the immediate post-pandemic Boom:

Note that this is telling a similar story to that of jobless claims that I just updated again this morning.

Although I won’t show the historical graph in this post, for about 70 years prior to the pandemic a decline in real retail sales YoY generally meant that a recession was close at hand. As you can see from the below graph, that wasn’t true for 2023-24. But YoY real retail sales have been positive, and improving, this year:

Further, real retail sales lead nonfarm payrolls (red in the above graph), albeit noisily. Sudden changes, such as caused by YoY comparisons with stimulus spending, take a longer time to get “digested” by the jobs market (a decent proxy is the 12 month average of the YoY% change in sales), but typically the lead/lag time is on the order of about 6 months. In that regard, here is the close-up of the last two years in the above YoY graph:

With the improvement in real retail sales this year, I expect the next few jobs reports to be positive, and perhaps more positive than we saw earlier this year.

So, to reiterate: like jobless claims, this is another positive for the economy.

Jobless claims continue near historic lows (yet again)

 

 - by New Deal democrat

Let’s take our regular weekly look at jobless claims. As per my periodic reminder, these are an excellent short leading indicator for the economy, particularly in conjunction with the other half of my “quick and dirty” forecast method, the YoY% change in stock prices.

These have been near historic lows much of this year, and continued in that vein this week. Initial claims rose 1,000 to 199,000, and the four week moving average rose 250 to 198,500. With the typical one week delay, continuing claims declined -4,000 to 1.710 million:

On the YoY% basis more important for forecasting purposes, initial claims were down -10.4%, the four week average down -11.7%, and continuing claims down -11.3%:

This metric continues to be very positive.

Since we’re still early in the month, I’ll dispense with what this means for the unemployment rate (but it certainly means I don’t expect the unemployment rate to increase). So here is the update on the “quick and dirty” forecast mentioned above. Note that since the metric for jobless claims is that they not be higher by more than +10% YoY, that line is inverted, and 10% added:

In contrast to the deteriorating real aggregate payrolls I discussed yesterday, this - like most current leading indicators - remains quite positive. To reiterate: when it comes to gauging the economy, leave your ideological predisposition at the door. In essence, the AI Boom (or bubble) is outweighing all the other economic determinants at the moment.