August new home sales consistent with a housing recovery. Unfortunately . . .

 

 - by New Deal democrat

In the past few months, I have described the housing market as being in a subpar equilibrium, with sales, construction, prices, and finally inventory moving more or less sideways - but simply not at an affordable level. This month I think I need to amends that, because unlike existing homes, new home prices probably *have* declined to an affordable level.

Let me start by repeating that new home sales are perhaps the most leading of all the housing sector metrics. But they suffer from being very volatile and heavily revised. So their three month moving average is more reliable as a signal.

With that said, this morning’s new home sales report for August was yet more evidence for both of the above theses. Sales, prices, and inventory all generally stayed on their recent trend level.

Sales (blue)increased 41,000 annualized to 684,000. This was the highest single month for sales this’d year. More importantly, the three month moving average was also at its highest level this year:

Further, a bottoming in inventory (red) is something we typically see towards or even after the end of a recession. And inventory, after bottoming at the end of last year has trended sideways to slightly higher this year. Were it not for the sharp upwards trend in interest rates, this would signify a recovering housing market. Keep in minds that these sales were for August, before the latest upward push in rates, so it is likely this trend meets its untimely demise in the next month or two.

Prices also continued their sideways to slightly declining trend, increasing $1,500, or 0.4%, for the month. But since they are not seasonally adjusted, the YoY% change ids most important - and it is obvious from the graph below that, July excepted, prices are at their lowest level in nearly five years:

Although I won’t bother with the graph, the YoY% price decline was -5.8%.

Where I am revising my opinion is in terms of affordability. The below graph deflates the median price of a new home by average weekly earnings, normed to 1 as of August:

Deflated by wages, the price of a new home is at its lowest level except for one month during the COVID lockdowns in almost 15 years! Were it not for 7%+ mortgage rates, the outlook for housing would be very positive. Unfortunately . . . 

In summation, housing - and in particular new home sales - is not currently forecasting a recession. Last month I closed with “With sales relatively stable and prices slowly deflating, the new home market is meeting the existing home market in a equilibrium, which is likely to remain unless something significant happens upstream, like an increase in mortgage rates back to 7%, possibly driven by a Fed rate hike.” Both of those “significant things” have since happened. 

We live in interesting times.

Extremely low jobless claims forecast unemployment rate declining below 4%

 

 - by New Deal democrat

Let’s take our regular weekly look at jobless claims. As a reminder, I look at these because they are an excellent short leading indicator for the economy.

Last week new jobless claims declined -1,000 to 197,000, continuing its streak of number close to all time historical lows. Thew four week moving average declined -1,750 to 202,250. With the typical one week delay, continuing claims rose 2,000 to 1.719 million:

Note that initial claims in particular are close to their previous all-time post-pandemic lows set in 2022.

On the YoY% basis more important for forecasting purposes, initial claims were down -10.0%, the four week moving average down -14.6%, and continuing claims down -10.3%:

All of these are extremely positive for the economy - indeed the most positive in thew past three+ years.

Lastly, the last time initial and continuing claims were at these levels, the unemployment rate was at 3.7%-3.9%:

In other words, this forecasts that the unemployment rate will continue to decline in the next several months not just to 4.0%, but likely even lower. We’ll find out a week from tomorrow.

The rebound in manufacturing employment is likely real, if fragile

 

 - by New Deal democrat

Over at Econbrowser, Prof. Menzie Chinn asks whether the gains in manufacturing employment as measured by the monthly payrolls report might be illusory, based mainly on a sharp deceleration in the ADP employment reports during July and August. While the current Administration in Washington is doing just about everything it can to sew chaos and throw monkey wrenches into the economy, I think the manufacturing employment recovery is real, if most likely narrowly based on AI-related spending.

Let’s start with Prof. Chinn’s contrast between the monthly ADP report (dark blue) and the payrolls report for manufacturing (red, right scale), together with ADP’s weekly data (light blue):

There certainly has between a stark contrast this year. Prof. Chinn also shows the QCEW comparison through March, which is also significantly weaker than the payrolls numbers, and will be used to revise them next February.

If the payrolls report were the only contrast, I would be inclined to be more concerned. But let’s take a look at some other manufacturing data from other sources.

First of all, both manufacturing production (red) and core capital goods orders (blue) for firms have been on a sustained upturn this year. And goods-producing employment as measured by the NY and Philly regional Feds (gold, right scale) has also picked up since spring:

Additionally, here is a comparison of the ADP figure with the monthly ISM manufacturing subindex for employment:

So we have four other sources - the Fed, several regional branches, the Census Bureau, and ISM - all showing an upturn in manufacturing employment this year, or preconditions for such an upturn, all independently. That suggests to me that it is the summer downdraft in the ADSP numbers that is anomalous.

For what it’s worth, the increase in manufacturing payrolls internally has followed an increase in the average weekly hours for manufacturing personnel, a long-time leading indicator, which started turning up early in 2025:

Again, I suspect this is a by-product of the $Trillions sloshing around in AI related building, so it rests on a fragile basis. But nevertheless it is a real underpinning for the recent employment growth.

Yes, Virginia, the k-shaped economy is real

 

 - by New Deal democrat

Financial pundit Lance Roberts recently wrote an article entitled, “K-shaped economy: reality or media driven perception,” in which he argued: 

Everyone “knows” wealth concentration is worse than ever. As I laid out in my earlier piece on the K-shaped economy and why the middle class moved up, the income story runs in the opposite direction from the coverage.

The wealth story is stranger still. Pull the Federal Reserve’s Distributional Financial Accounts and compute it yourself, and the top 10% share of household net worth peaked at 70.3% in the first quarter of 2019. It sits at 67.9% today. The bottom 50% share bottomed at 0.4% in late 2011, was 1.7% at the end of 2019, and is 2.5% now.

When looking at wealth concentrations, it is very easy to blame those at the top of the wealth pyramid. Yes, the top 10% of the population held a 31.8% share of economic wealth in the fourth quarter of 2025. Yet, the bottom half gains since 2019 came almost entirely from the 90th to 99th percentiles, which fell from 39.7% to 36.3%. In plain English, the professional class lost relative ground, not the working class. Such is a detail that changes who you think is complaining.

Furthermore, the recovery that no one called K-shaped was far worse. Between 2007 and 2016, median wealth for the bottom 30% of families fell 31%, while the top 10% fully recovered.⁷ Saez found the top 1% captured 91% of real income growth from 2009 to 2012. Nobody ran a K headline in 2013. The data was uglier then.

Since my approach to Roberts, who typically writes from a right-wing perspective, is not “is he wrong,” but rather “*how* is he wrong,” I checked his work.

First, as I recall, while they may not have called it a “K-shaped” economy, there were plenty of articles in thew first five years of the last expansion about the inequality of wealth and spending. But as per my usual practice, I went back ands independently looked at the numbers.

Roberts says that “the middle class moved up.” That’s certainly true if we look at real median incomes, which increased 42% from $61,910 in 1984 to $87,460 in 2025, as per the Census Bureau data that were just updated last week:

But that data only takes us up to 9 months ago, whereas most of the “K-shaped” commentary is from this year. And according to Motio Research, on a YoY basis real median household income in August was actually down -0.1%:

Ands yet according to the weekly Redbook Index, as well as the monthly data, retail sales YoY growth has actually accelerated this year:

As I’ve pointed out a number of times, the stock market has been up about 15%-20% this year, driving a lot of “wealth effect” spending. And owns stocks? The top 1% own an outright slight majority, with another 27% being owned by the 90th to 99th percentile, for a total of 88%:

And Roberts’s story on wealth distribution is problematic as well. While it’s true that the combined bottom 90% have a bigger share of total wealth than 2019, as confirmed in the below graph:

The year 2019 is not that relevant to the spending pattern in the past several years. If we look at the changes in wealth shares since July 2023, a very different picture emerges:

The top 1%, and even more drastically, the top 0.1% of the wealth distribution have been running away with the growth, while the bottom 50%, and even more the 50th to 90th%, have been losing ground. And that divergence accelerated this year.

This is the continuation of a long term trend that goes back at least until the early 1990s:

The real post-pandemic tale is told by the following two graphs, of the absolute nominal levels of wealth held by each percentile. Thew first norms each level to 2019:

The share of net worth held by the bottom 50% increased sharply in the immediate post-pandemic aftermath, buoyed by stimulus payments and big increases in wages supported by the white hot labor economy.

But since mid-2022, the gains have been totally lopsided: 

The nominal wealth of the bottom 50% is up 17.0%, and the next 40% up 23.1%, while the 90th-99th% is up 30.5%, the top 1% is up 42.4%, and the top 0.1% is up 46.6%, vs. 8.8% for the CPI. And we know that real income has declined in the past year, with real aggregate payrolls barely up at all. Since house price growth has stalled, and as per the above lower income households own very few stocks, the likelihood is that this year their amount of real net wealth owned has stalled if not declined as well.

 In short: yes, Virginia, the K-shaped economy is real.

Weekly Indicators for September 14 - 18 at Seeking Alpha

 

 - by New Deal democrat

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

The two big driving forces at the moment - increased interest rates and spiking gas prices - continue to be the most salient datapoints, along with the inflationary pulse that continues throughout the economy. Nevertheless, just like the monthly data, the high frequency data also continues to indicate that the economy is growing, with little imminent danger of a downturn.

As usual, clicking over and reading will bring you up to the virtual moment as to the state of the economy, and bring me a penny or two for my efforts.

Three reports on the state of goods production show rebound continues - but maybe with a wobble

 

 - by New Deal democrat



This week we got three significant updates about the state of goods production in the economy: industrial production this morning, and total business sales as well as the New York and Philly manufacturing indexes earlier this week. Let’s look at each of them.

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 on an upward trend since last year, and that continued in August, although the monthly numbers were decidedly mixed. Headline production (blue) was unchanged at its post-pandemic high, while manufacturing production (red) declined -0.3%. Meanwhile, electric and gas utility production (gold, right scale), which is most closely aligned with AI data center construction, increased 1.8%. 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 last month. Indeed, on a YoY basis, while headline production was up 1.7%, but its manufacturing component only up 0.3%, utilities have increased sharply, by 10.9%:

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.



A similar continuation of trends, both good and bad, was apparent in the NY and Philly manufacturing indexes. Here is the average of the headline number for both indices (blue) and the more leading new orders component (red):

This month the headline average was 22.7 and the new orders average was 15.6 [note that these are diffusion indexes where any value above 0 indicates expansion]. Although there was significant month over month deceleration, the three month averages continue to be close to their averages during the post-pandemic Boom in 2022. For further context, here is further context with the long term historical graph:

In particular the leading new goods component indicates that we should expect the manufacturing expansion to continue for the next several months.

If the good news continued this month, so did the bad news; namely, that the inflationary pulse continued as well. Here’s the historical graph of the diffusion indices of both prices paid by producers (blue) and prices received by them downstream (red):

This month prices paid showed further widespread increases, at 55.8, while prices received were also close to their highs over the past 16 months, at 29.7. As with all months so far this year, the current widespread pricing pressures on the incoming end are only partially being passed on downstream. Further, as shown in the below longer-term historical graph, this is also as bad as at any time since the turn of the Millennium, with the exception of the post-pandemic period and the gas price-driven spike during the first seven months of the Great Recession:

If our first two metrics showed a continued inflationary industrial expansion, the third, total business sales (blue)and inventory (red), updated through July indicates there may be some wobbling:

Note that these are nominal values. While sales increased 0.3% in July, inventories increased 0.8%. If we apply the PCE price deflator for July, which increased 0.2%, these are up 0.1% and 0.7% respectively. But note that sales remained below their recent May peak for the second month in a row. This is significant, because as the below historical graph (in log scale) shows, sales turn down first before inventories do:

In fact it is likely the downturn in sales and accumulating inventory which causes companies to cut back.

This may be just noise, or it could be more. I would need to see evidence of a sales slowdown in other metrics, like personal spending and the ISM reports as well as the regional Fed reports, to believe that this is more significant.

So, for now, the manufacturing rebound continues, although it is inflationary and mainly driven by AI related spending.