Real personal income in May was recessionary (again); depleting savings (temporarily?) “saves” the day

 

 - by New Deal democrat

The third item from this morning’s torrent of economic data I want to address is personal income and saving. I’ll update the spending side tomorrow, including an updated look at motor vehicle sales, which were also released this morning.

On a nominal basis, personal income rose 0.7% in May, as did spending. But since prices as measured by the PCE deflator rose 0.3%, in real terms the former rounded to only a 0.2% increase (blue) and the latter a 0.3% increase (red). It is easy to see that the two metrics diverged sharply about a year ago:

Indeed, on a YoY basis, while real spending is up 2.1%, real income is *down* -0.2%:

As indicated above, I’ll withhold further comment on the spending side until tomorrow. But for real incomes to be negative YoY is dismal. As shown in the below graph, for the entire 60 year period between the inception of data and the pandemic, with only one exception (described below), real personal income was *never* negative YoY except during or immediately after the worst recessions:

The sole exception was in 2013, when a temporary Social Security tax withholding holiday that had been put in place for one year in 2012 as a stimulus measure expired. Note, however, that it was also negative in 2022 without a recession having taken place.

Similarly, real income less government transfer payments rose 0.3% for the month (blue, right scale), but is down -0.4% YoY (orange, left scale):

Similarly, before the pandemic except for 2013 and 2022, real income less government transfer payments was only negative during or immediately after recessions:

Similarly to 2012, in 2021 there were large direct one-time stimulus payments to households. Thus the 2022 income numbers suffered in comparison. 

How is it that spending can continue to be so positive, with no recession occurring, despite the actual decline in real inocme? Because consumers have dipped into their savings in a big way. The personal savings rate was 3.0% for the second month in a row:

As shown in the above graph, which subtracts 3 from the saving rate so that the current level shows at the 0 line, aside from 2022 and the 2005-07 period just before the Great Recession, the personal saving rate has never been this low during the entire history of the series.

Tomorrow I’ll dig deeper into the spending side, but for now the takeaway is that the US economy would almost certainly be in a recession except for the consumer spending spree, which is almost certainly in large part funded by stock market gains and the ensuing “wealth effect” which is in turn a byproduct of the AI data center boom - or Bubble.

Manufacturing sector continues to be positive through May

 

 -by New Deal democrat

Per my post earlier this morning, I am going to delay until tomorrow reporting on motor vehicle sales and an in-depth look at personal spending, but let’s look at the second significant data release from this morning: manufacturers’ new durable goods orders for May.

To cut to the chase, this was another good month, at least on a nominal basis, continuing the string of positive, even somewhat Booming reports so far this year. While total orders declined -4.5% for the month - still within the range of noise - core capital goods orders increased 1.6% to another all-time high. And even with the decline, the headline number was only lower than one month ago and one month last year:

Headline orders are up about 25% from their pre-pandemic all-time high in 2018, while core capital goods orders are about 37% higher. And as the below YoY% comparison shows, the rate of increase has been accelerating for core capital goods orders in the past few months:

I need to caution again that these are nominal numbers. But even if I were to adjust for CPI, PPI, or PCE inflation, the three month average of orders would be higher YoY, and core capital goods would still be about 6% higher YoY.

Meanwhile, real manufacturing and trade industries sales were also updated this morning for April, showing a -0.9% decline monthly to the lowest level in four months. But real sales remain about 13% higher than before the pandemic:

On a YoY basis, real sales are up 1.3%:

This is among the poorest showings since the pandemic, eclipsed only by the 2022 manufacturing downturn and several months last year. Further, a historical look shows that before the Millennium, a 1.3% increase in real sales only happened shortly before or during recessions. Since the accession of China to normal trading status, it has been lower in 2002, the industrial recession of 2016, and the 2019 period that had some pre-recessionary characteristics as well:

Keep in mind this is for April, while the durable and capital goods release was for May. Further keep in mind that new orders are a forward-looking, short leading indicator, while real sales are a coincident indicator. In other words, the main import of all the data remains positive for the manufacturing sector.

Jobless claims: seasonality returns, but remains very positive for the economy

 

 - by New Deal democrat

This morning a plethora of economic data was released, including personal spending and income, manuacturers’ new orders, motor vehicle sales, and jobless claims. Since tomorrow sees no significant data releases, I’m going to hold the in-depth look at spending and motor vehicle sales until tomorrow, and update the other releases today.

Let’s start with the typical weekly look at jobless claims, 1/2 of my “quick and dirty” forecasting method. To reiterate, the issue I’ve been looking at is whether post-pandemic seasonality is reappearing, or whether the “regime change” of signficantly lower YoY claims that started last July is intact.

And the apparent answer is: both.

Initial claims declined -12,000 to 215,000, while the four week average rose 750 to 224,250. With the typical one week delay, continuing claims rose 21,000 to 1.821 million:

Excluding the immediately preceding three weeks, this week’s initial claims number is the highest all year except for two week in February and one in April; and the four week average is the highest since last November. This very much looks like the return of post-pandemic residual seasonality.

But on a YoY basis, the very positive comparisons continue, as initial claims are down -8.9%, the four week average down -7.4%, and continuing claims down -7.1%:

This is in line with the excellent YoY comparisons we have seen almost all of this year.

To synthesize, it would appear that while post-pandemic seasonality has reappeared, it is at a lower level that from 2023-2025. Which is very positive for the near term economy.

Finally, let’s do our update of what this might mean for the June unemployment rate when that report is released next week. Unsurprisingly, it continues to show that downward pressure will continue to be exerted on that rate. I would not expect the rate to increase from 4.3%, and there is a very good chance it declines:

The positive news continues.

May new home sales: Another poor month for sales, prices stable, inventory increasing

 

 - by New Deal democrat

To start with the usual: new home sales are very volatile and heavily revised, which is why I pay more attention to single family permits. But they are the most leading of all the housing data; and averaged over three months, much of the noise goes away.

In May, new home sales declined -46,000 annualized to 580,000, just above their 3+ year low set in January:

This likely in part reflects the recent uptick in mortgage rates - but again it is within the range of noise. The three month moving average is virtually unchanged, but at the low level it has been for most of this year so far, which suggests that single family permits (red, right scale), which convey more signal, are likely to hold steady or even decline further from their recent range.

Meanwhile, the median price for a new home - which is not seasonally adjusted, so should be compared YoY - was almost exactly unchanged from one year ago (red, left scale):

The longer term trend over the past three years of slowly declining prices remains intact.

This is largely in accord with existing home sales, where the median price through May was up 1.3% YoY, and the Case Shiller and FHFA repeat home sales prices, which were up 0.7% and 1.7%, respectively. The difference is that home builders can change, and have changed, price points, not just by lowering profit margins, but also by building more densely, or smaller square footages, or fewer amenities - which they have done.

Finally, let’s look at the inventory of new houses for sale. As a refresher, here is the historical view of the leading/lagging relationship between the number of houses sold and houses for sale:

Inventory is generally the last shoe to drop in the housing market before recessions begin. But as I noted last month, there has been an interesting wrinkle this year, as inventory has turned back up:

This is all but unique. Historically a recession will not occur until inventory turns down again. But to reiterate, housing has been recessionary for a year, and yet no recession has occurred. The same has been true for motor vehicle sales. 

In summary: not a good report, it looks like housing is taking another step down. But no upward price pressure, and no indication of broader negative implications for the economy.

Consumer spending has turned red hot - expect no recession in the immediate future

 

 - by New Deal democrat

Well, I was working on one nerdy long-term historical post this morning, when I decided to check the weekly update on consumer spending from Redbook. And, as you’ll see below, that was the end of that! (I may yet follow up this afternoon. We’ll see.)

Last week, Redbook consumer spending was higher by 10.0% YoY! That’s the biggest YoY gain since the end of 2022:

You are simply *NOT* going to have a recession in the immediate future in the face of that kind of increase in consumer spending, especially when even the recent spike in CPI only brought it to a 4.2% YoY increase. I have seen commentary that the big pick-up in consumer spending is due to bigger tax refunds, which may be true given that the recent surge started right after April 15. If that’s the case, I would expect it to subside as the summer goes on.

This plays directly into the “quick and dirty” forecast method I highlighted yesterday: if stock prices are higher YoY (a proxy for the producer side), and jobless claims are not higher by over 10% YoY (a proxy for job security); and also if real retail sales (a proxy for the consumer side) are higher YoY - there has never been a recession, going all the way back to World War 2.

The “quick and dirty” forecasting method has been flawless

 

 - by New Deal democrat

On Friday I wrote about how I have been rethinking the long leading indicators — those that are useful for forecasting the economy 12-24 months out — because while when they are positive, the economy has been as well, but when they have been negative (twice) in the past 15 years, the economy has weakened but not gone into recession. They have functioned more like a “severe weather watch;” i.e., conditions are favorable for development, but by no means more likely than not.

By contrast, the “quick and dirty” short term forecasting system has been flawless.

What is the “quick and dirty” system? Simply look at the stock market and the four week average of initial jobless claims. If the stock market is lower YoY, and the four week moving average of new jobless claims is 10% or more higher YoY, the economy is likely to fall into recession in the next several months. Otherwise, the economy will remain in expansion.

Let’s take a look. The below two graphs track stock prices (gold), the four week average of initial jobless claims (red, inverted and adding 10 so that any YoY change of higher than 10% shows below the zero line); and also adds real retail sales YoY (blue), which has also had a very good long term record; first for the four years before the pandemic:

And here are the five years after the pandemic:

As you can easily see, at no time have both stock prices and the four week average of initial jobless claims been below the zero line together. In 2018, stock prices and real retail sales were negative for only one month, but jobless claims were doing very well. And in mid-2023, the system never quite signaled — stock prices went positive YoY one month before jobless claims turned sufficiently negative. And real retail sales likewise turned positive with stock prices.

Currently all three metrics are solidly positive, signaling economic expansion will continue for at least the next few months.

And what of the pandemic? Here is the close-up of 2020:

The pandemic lockdowns began roughly on March 9. Within several days, the stock market turned negative YoY, and jobless claims also did so by more than 10% on March 21 (in fact, there were such severe layoffs that I’ve had to cut the scale by 100!). By early May the stock market had turned positive again. Real retail sales also turned negative in March, and turned back positive in June.

LIke I said: quick and dirty - and flawless.

Weekly Indicators for June 15 - 19 at Seeking Alpha

 

 - by New Deal democrat

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

There was another air pocket this week in withholding tax payments, and some weakening in mortgage applications, but the overall tone remains positive, despite the ongoing international chaos emanating from Washington.

As usual, clicking over and reading will bring you up to the virtual moment as to the state of the economy, and help me with my lunch money.