September jobs report: a very positive report within a framework of continued deceleration

 

 - by New Deal democrat

As I have written many, many times, consumption leads employment; and the near stagnation in real sales and spending signaled that we should expect weaker monthly employment reports, with both fewer new jobs and a higher unemployment rate. In September, the former happened; the latter did not.

The three month average in employment gains since February has continued to decelerate from over 500,000 to 372,000. But this month the unemployment rate declined back to its post-pandemic low.

Here’s my in depth synopsis.

HEADLINES:

• 263,000 jobs added. Private sector jobs increased 288,000. Government jobs decreased by -25,000. 
• The alternate, and more volatile measure in the household report indicated a  gain of 204,000 jobs. The above household number factors into the unemployment and underemployment rates below.
• U3 unemployment rate declined 0.2% to 3.5%.
• U6 underemployment rate declined 0.3% to 6.7%.
• Those not in the labor force at all, but who want a job now, increased 285,000 to 5.834 million, compared with 4.996 million in February 2020.
• Those on temporary layoff declined -24,000 to 758,000.
• Permanent job losers declined -173,000 to 1,181,000.
• July was revised upward by -105,000, and July was unchanged, for a net increase of 11,000 jobs compared with previous reports.

Leading employment indicators of a slowdown or recession

These are leading sectors for the economy overall, and will help us gauge whether the strong rebound from the pandemic will continue.  These were all positive:

• the average manufacturing workweek, one of the 10 components of the Index of Leading Indicators, increased +0.1 hour to 41.1 hours.

• Manufacturing jobs increased 22,000, and are at a level higher than before the pandemic.
• Construction jobs increased 19,000, also at a level higher than before the pandemic. 
• Residential construction jobs, which are even more leading, rose by 2,300.

• Temporary jobs rose by 27,200. Since the beginning of the pandemic, roughly 300,000 such jobs have been gained.
• the number of people unemployed for 5 weeks or less declined by -69,000 to 2,154,000, about equal to its pre-pandemic level.

Wages of non-managerial workers

• Average Hourly Earnings for Production and Nonsupervisory Personnel rose $0.10 to $27.77, which is a 5.8% YoY gain, a further decline of -0.3% from last month and its 6.7% peak at the beginning of this year.

Aggregate hours and wages:

• the index of aggregate hours worked for non-managerial workers increased 0.5% which is above its level just before the pandemic.

•  the index of aggregate payrolls for non-managerial workers rose by 0.9%, a very strong increase compared with the last several months’ outright declines in prices.

Other significant data:

• Leisure and hospitality jobs, which were the most hard-hit during the pandemic, rose 83,000, but are still about -6.7% below their pre-pandemic peak.
• Within the leisure and hospitality sector, food and drink establishments added 60,700 jobs, but are still about 560,000, or -4.5% below their pre-pandemic peak. 
• Professional and business employment increased by 46,000, over 1,000,000 above its pre-pandemic peak.
• Full time jobs increased 326,000 in the household report.
• Part time jobs declined -7,000 in the household report.
• The number of job holders who were part time for economic reasons declined -312,000 to 3,763,000.
• The Labor Force Participation Rate declined -0.1% to 62.3%, vs. 63.4% in February 2020.

SUMMARY

This was a very good report with just a few blemishes. On the plus side, the entire slew of leading indicators in the report advanced. Total payroll gains were very strong. The unemployment and underemployment rates both declined. Part time employment was more than replaced by strong full time employment. The rebound in leisure and hospitality employment continued. Wage increases moderated YoY, but on a monthly basis advanced well.

The few weak points included the decline in the labor force participation rate, which is why the unemployment and underemployment rates declined as they did. The gains in professional and business employment were weak.

In summary, we have a very positive report, completely inconsistent with any idea that we are currently already in a recession, but within the larger framework of an economy which is decelerating.

Signs and portents of an employment slowdown and a near-term recession

 

 - by New Deal democrat

I continue to believe that a recession - possibly a deep if relatively brief one - is likely to start early next year. As I’ve mentioned before, this isn’t just an academic exercise; recessions by definition feature jobs and income losses, which is my primary interest.

With that introduction, let’s look at one overall metric, and several which will be updated tomorrow as part of the September jobs report.

First of all, if there were such a thing (ex- pandemic lockdowns) as a one quarter recession, courtesy of last week’s income revisions we almost certainly had one during Q2. Here’s a graph of most of the metrics relied upon by the NBER in dating recessions, plus several variations. Note that almost all of them went down during the April-June period - and I haven’t even included GDP:

The only recession-related metrics that improved during Q2 were nonfarm payrolls and the mining and utilities portions of industrial production. Manufacturing production, income ex-government transfers, real retail sales, real manufacturing and trade sales, as well as real GDP, all declined during the second quarter. In the first two months of Q3, all of them improved from their June levels, and I expect that to continue in September.

Next, as I have said many times, consumption leads employment. Here’s the YoY% look at real retail sales (blue), real personal spending (black), vs. nonfarm payrolls (red). Because spending is more variable than jobs, note that both sales and spending metrics are reduced for scale:

As I wrote the other day, real sales are flat YoY. Real spending is up less than 1% YoY. As a result, we should expect gains in nonfarm payrolls to decelerate to virtually nothing in the coming months.

Here’s a more granular look a the month over month change in each for the past 15 months:

Note the outright slight decline in sales since February, and the sharp deceleration in spending. Job gains have been slowly decelerating since, and because consumption leads employment, we should expect that to continue. While any month can be an outlier, the likelihood is that tomorrow’s report will show gains of less than 300,000 jobs for the month of September.

Next, initial jobless claims lead the unemployment rate. Here’s a graph of the 4 week average of new jobless claims through this morning’s report (right scale), compared with the unemployment rate (left):

Last month the unemployment rate rose from its expansion low of 3.5% to 3.7%. While again any given month might be an outlier, I expect the unemployment rate to remain above 3.6% going forward in the next few months.

Finally, as I wrote last month, real aggregate payrolls of non-supervisory workers have an excellent record as a coincident to shortly leading indicator of recessions. Previously I showed YoY changes. Here are the quarter/quarter % changes for the past 60 years:

Occasionally there will be a single negative q/q change during slowdowns. The first negative q/q change has tended to happen 0-2 quarters before the onset of recessions during that period.

Now here is the close-up since the beginning of 2021, shown both quarterly through Q2 (blue) and monthly through August (red):

We’ve had two positive months so far in Q3. Tomorrow’s report, once adjusted for inflation, will complete the picture. Needless to say, a positive number will be good and a negative one bad.

Watching for a continued slowdown in jobs created, a continued slight elevation in unemployment over the bottom, the change in aggregate payrolls, plus whether wage gains for non-supervisory workers moderate, is my primary focus in tomorrow’s jobs report. Especially with OPEC having now joined the Fed in apparently trying to deliberately cause a recession.

 

Jobless claims rise; the gas price low is probably in

 

 - by New Deal democrat

Initial jobless claims may have ended their recent downtrend.

Initial claims rose 19,000 to 219,000 from last week’s 5 month low. The 4 week average rose 250 from its 4 month low to 206,500. Continuing claims, which lag somewhat, increased 15,000 to 1,361,000:

The downtrend of the past 2 months was almost certainly a positive side-effect of lower gas prices. But in the past 2.5 weeks, according to GasBuddy, gas prices have risen over $.20/gallon. With OPEC deliberately cutting back production in order to cause a shortage in Europe and the US this winter (to aid Russia), it is very likely that gas prices will continue to rise again. If so, I expect jobless claims to rise again as well.

In the meantime, I have more I want to say about the economy in advance of tomorrow’s job report, which I’ll post later this morning.

Coronavirus dashboard for October 5: an autumn lull as COVID-19 evolves towards seasonal endemicity

 

 - by New Deal democrat

Back in August I highlighted some epidemiological work by Trevor Bedford about what endemic COVID is likely to look like, based on the rate of mutations and the period of time that previous infection makes a recovered person resistant to re-infection. Here’s his graph:

He indicated that it “illustrate[s] a scenario where we end up in a regime of year-round variant-driven circulation with more circulation in the winter than summer, but not flu-like winter seasons and summer troughs.”

In other words, we could expect higher caseloads during regular seasonal waves, but unlike influenza, the virus would never entirely recede into the background during the “off” seasons.

That is what we are seeing so far this autumn.

Confirmed cases have continued to decline, presently just under 45,000/day, a little under 1/3rd of their recent summer peak in mid-June. Deaths have been hovering between 400 and 450/day, about in the middle of their 350-550 range since the beginning of this past spring:

The longer-term graph of each since the beginning of the pandemic shows that, at their present level cases are at their lowest point since summer 2020, with the exception of a brief period during September 2020, the May-July lull in 2021, and the springtime lull this year. Deaths since spring remain lower than at any point except the May-July lull of 2021:

Because so many cases are asymptomatic, or people confirm their cases via home testing but do not get confirmation by “official” tests, we know that the confirmed cases indicated above are lower than the “real” number. For that, here is the long-term look from Biobot, which measures COVID concentrations in wastewater:

The likelihood is that there are about 200,000 “actual” new cases each day at present. But even so, this level is below any time since Delta first hit in summer 2021, with the exception of last autumn and this spring’s lulls.

Hospitalizations show a similar pattern. They are currently down 50% since their summer peak, at about 25,000/day:

This is also below any point in the pandemic except for briefly during September 2020, the May-July 2021 low, and this past spring’s lull.

The CDC’s most recent update of variants shows that BA.5 is still dominant, causing about 81% of cases, while more recent offshoots of BA.2, BA.4, and BA.5 are causing the rest. BA’s share is down from 89% in late August:

But this does not mean that the other variants are surging, because cases have declined from roughly 90,000 to 45,000 during that time. Here’s how the math works out:

89% of 90k=80k (remaining variants cause 10k cases)

81% of 45k=36k (remaining variants cause 9k cases)

The batch of new variants have been dubbed the “Pentagon” by epidmiologist JP Weiland, and have caused a sharp increase in cases in several countries in Europe and elsewhere. Here’s what she thinks that means for the US:

But even she is not sure that any wave generated by the new variants will exceed summer’s BA.5 peak, let alone approach last winter’s horrible wave:

In summary, we have having an autumn lull as predicted by the seasonal model. There will probably be a winter wave, but the size of that wave is completely unknown, primarily due to the fact that probably 90%+ of the population has been vaccinated and/or previously infected, giving rise to at least some level of resistance - a disease on its way to seasonal endemicity.

August JOLTS report: the game of reverse musical chairs in the jobs market is ending

 

 - by New Deal democrat

Since early this year I’ve been making the point that, because of the pandemic, there have been several million fewer persons looking for work, leaving a huge number of unfilled job vacancies, particularly in the face of a roughly 10% higher jump in demand. This has given employees the upper hand, as there are almost always higher paying jobs on offer for which they can apply. I‘ve also posited that the dynamic would only slow down once some employers throw in the towel, and the number of job openings signficantly declines. I’ve called this “reverse musical chairs.”

This morning we got some very potent news that the game is entering its closing phases, at least for this cycle. 

Almost certainly openings peaked in March. In August alone, openings declined -10% to 10.053 million, for a total - and accelerating - decline of 1.8 million, or over -15% since the peak:

At their average rate since March, openings will return to their pre-pandemic level by next April. At the rate they’ve declined in August, they’ll be at that level by November.

Meanwhile actual hires, as shown in the above graph, increased slightly m/m, but the decelerating trend since the beginning of this year is clear. This is simply more confirmation that consumption leads employment, and we should expect monthly jobs numbers to decelerate further, and go to virtually zero m/m by early next year.

Voluntary quits, which are an important measure of employee confidence in finding another job, increased 2.5% for the month, but again the overall declining trend since last autumn is clear:

The story is the same for total separations. There was a 3.1% increase for the month, but a clear decelerating trend since the end of last year.

Finally, layoffs and discharges increased 5.0% for the month, to their highest level since March 2021:

The game of reverse musical chairs is slowing down. It is most likely within months of ending. 

Because consumer spending as measured by real retail sales has been flat for over a year, and real personal consumption expenditures are up only 0.7% in the past 10 months, monthly jobs gains in the coming months are going to slow down or even stop. Here’s the graph of monthly jobs gains for the past 2+ years:

While any month’s number will be variable, in general this Friday I would expect a gain of less than 400,000 for September, and most likely below 300,000 - possibly below 200,000. We’ll see then.

September manufacturing new orders and August construction spending both turn down

 

 - by New Deal democrat

As usual, we begin another month and another quarter with important manufacturing and construction data.

The ISM manufacturing index has a very long and reliable history. Going back almost 75 years, the new orders index has always fallen below 50 within 6 months before a recession, and in three cases did not actually cross the line until the first month of the recession itself - although the recession did not begin until after the total index fell below 50, and in fact usually below 48.

In September the overall index declined to 50.9 - just slightly expansionary - and new orders declined to a new post-pandemic lockdown low of 47.1:

This is consistent with readings right before the onset of the Great Recession, but also with several slowdowns that did not quite turn into recessions.

Construction spending, both in total and residential, declined nominally for the seond and third month in a row, respectively:

Again, this is consistent with a recession, but also a slowdown as in 2018.

Adjusting for inflation using the construction materials special index, total construction spending is down about -17%, and residential construction spending down -8% from their respective peaks at the end of 2020:

Note the more leading residential measure has been flat for nearly a year. 

Because construction spending is the “real” economic activity, as is the metric of “housing units under construction” from last week’s permits and starts release, below is a comparison of the two measures measured YoY:

I mentioned last week that housing units under construction looked like it was peaking right now. Since construction spending seems to be coincident with or slightly leading units under construction, this is more evidence that the real economic activity in the leading housing market is at or just past peak. In other words, maybe not in Q3, but from here on in we should expect housing to subtract from real GDP.

August personal income and spending: major downward revisions overwhelm modestly positive monthly grains

 

 - by New Deal democrat

This morning’s personal income and spending report for August was positive month over month both in nominal and real terms, but the major story was in the revisions.

Personal spending is the essentially the opposite side of the transaction of retail sales. Both have been tracking relatively closely since the end of the stimulus-fueled spring spending spurge of 2021, as shown in the m/m% changes below:

Real personal spending was up +0.1% in August, compared with +0.2% for real sales. 

So far, so good. But as you can see from the above graph, real personal spending has all but stalled since April.  Compared with the average spending in Q2, the first two months of Q3 are only up +0.1% - which while positive is seriously weak.

Real personal income also increased less than +0.1%, rounding down to unchanged. Since May 2021, real spending has increased +2.8%, but real personal income is *down* -2.2%:

This reflects a major downward revision in income for the past year. Here’s last month’s graph, showing real personal income only down -1.0% since May 2021:

With this revision, we just got a big part of the explanation for why consumer confidence (and President Biden’s approval ratings) took such a hit earlier this year.

The same revision seriously affected the personal saving rate, which was unchanged for the month at 3.5% (shown as zero in the graph below for better historical comparison):

As revised, the saving rate has been close to all time lows for most of this year. Only 2005-08 were lower.

Here’s the same graph from one month ago, showing July’s saving rate as 5.0%, generally equivalent to the early 2000s:

But no more!

Usually the personal saving rate declines progressively during expansions, leaving consumers more and more vulnerable to negative shocks. With the revisions to the past year’s data, we are very much in that territory. If there is , e.g., another gas price spike, consumers’ luck will probably run out.

The positive trend in jobless claims continues

 

 - by New Deal democrat

For still another week, initial jobless claims continued their recent downtrend.

Initial claims declined -16,000 to 193,000, a 5 month low. The 4 week average also declined -8,750 to a new 4 month low of 207,000. Continuing claims, which lag somewhat, declined -29,000 to a 2.5 month low of 1,347,000:

The downtrend of the past 2 months is almost certainly a positive side-effect of lower gas prices. According to GasBuddy, gas prices have increased in the past 10 days. If gas prices stabilize, I expect jobless claims to do so as well.

But this is good news and very much at odds with the idea that the US is currently in a recession.

Interest rates, the yield curve, and the Fed chasing a Phantom (lagging) Menace

 

 - by New Deal democrat

There’s a lot going on with interest rates in the past few days.

Mortgage rates have increased above 7%:

This is the highest rate since 2008. Needless to say, if it lasts for any period of time it will further damage the housing market.

The yield curve has almost completely inverted from 3 years out (lower bar on left; upper bar shows a similar curve in April 2000, 11 months before the 2001 recession):

As of this morning, the curve is normally sloped from the 3.12% Fed funds rate up through the 3 year Treasury, which is yielding 4.22% (which, as an aside, is a mighty tasty temptation to buy medium maturity bonds). Beyond that, with the exception of the 20 year Treasury, each maturity of longer duration is yielding progressively less. If this is like almost all recessions in the last half century, the short end of the yield curve will fully invert (i.e., Fed funds through 2 years as well) before the recession actually begins. Although I won’t show the graph, the yield curve *un*-inverted before the last two recessions even began, immediately or shortly after the Fed began to lower rates again.

On the issue of rents, house prices, and owners equivalent rent, Prof. Paul Krugman follows up on the fact that OER is a lagging measure. Today he touts the monthly decline in new rental lease prices as possibly signaling a downturn in inflation:

He’s referring to the “National Rent Index” from Apartment List, which Bill McBride has also been tracking. Because it tracks rents in only new or renewed leases, it picks up increases or decreases more quickly than those indexes that measure all rentals (including those that were renewed, e.g., 9 months ago).

I don’t think the index is quite the signal Paul Krugman does, because it is not seasonally adjusted, and rents typically decrease in the last 4 months of each year:

Here is the cumulative yearly index for each of the past 5 years:

The -0.1% non-seasonally adjusted decrease in September this year is on par with that of 2018, and less of a decline in September 2019 or 2020. For the first half of this year, rents were increasing at a faster, and accelerating, rate compared with 2018 and 2019. Since June have rent changes been comparable with (and not more negative than) those two years.

I thought I would compare Apartment List’s with with the Case Shiller house price index, below:

Note that house prices broke out to the upside YoY beginning in late spring 2020, while apartment rents did not do so until early 2021. There were rent increase moratoriums in place during the pandemic, which may have affected that comparison. Still, it is cautionary that for the limited 5 year comparison time we do have, house price indexes moved first.

Finally, what would the Fed have done if it had used the Case Shiller index instead of owners equivalent rent in its targeted “core inflation” metric?

Via Mike Sherlock, here’s what the “Case Shiller [total, not core] CPI” looks like through last month:

Here’s another way of looking at the data, comparing the monthly % changes in the Case Shiller national house price index (blue), owners equivalent rent (red, right scale), and core CPI (i.e., minus food and energy) (gold, right scale):

Rent + owner’s equivalent rent are 40% of core inflation. Unsurprisingly, core inflation tends to track similarly to OER. But between May 2021 and May 2022, OER only averaged +0.4% monthly, whereas the Case Shiller index increased 1.5% on average monthly. If 40% of core inflation increased at 1.5% monthly instead of 0.4% monthly, core inflation would have on average been +0.4% higher each month for that entire year.

In other words, the Fed would have had a much earlier warning that an upsurge in core inflation was not going to be “transitory.” 

By contrast, during the last 3 months of the period through July that we have house price index data, OER has averaged +0.4%, whereas house prices have increased on average +0.6%. This would have brought core inflation down by -0.1% each month. If we use the last two months, OER is +0.6% and house prices have been unchanged. Core inflation would have been -0.3% lower in June and July.

In fact, if the trend of the last several months continues, by year end OER is going to be higher than house price appreciation on a YoY as well as m/m basis. And while OER has been increasing, house price indexes have been decelerating. 

In other words, if the Fed keeps raising rates, it is most likely chasing a phantom menace, a lagging indicator the leading measures for which will have already peaked and come down sharply.

House price indexes: more evidence of a summer peak

 

 - by New Deal democrat

The Case Shiller and FHFA house price indexes were updated through July (technically, the average of May through July) this morning. Ordinarily I do not pay them too much mind, but this year they are very important in confirming a peak in house prices.

Although the FHFA index is seasonally adjusted, the Case Shiller index is not, so the best way to show them in comparison is YoY. Here are YoY% changes for the last 2 years of each through June (FRED has not yet posted today’s numbers):

Remember, my rule of thumb for non-seasonally adjusted data is that the peak is most likely when the YoY gain declines to only 1/2 of its maximum in the last 12 months. the YoY peak in the Case Shiller index was +20.6% in March and April. The peak for the FHFA index was 19.3% in July 2021. By that standard, although both decelerated to 12 month lows, at +15.8% and 13.9% respectively, neither has actually turned down, although the FHFA index is closer. And since the FHFA has a tendency to turn slightly ahead of the Case Shiller index, this strongly suggests that a sharp deceleration in the Case Shiller index YoY will start within a month or two.

As I said above, however, the FHFA purchase only index *is* seasonally adjusted, and that index, after increasing consistently by 1% or more from June 2020 through May 2022, declined -0.1% in June, and -0.6% in July:

So the seasonally adjusted data in the FHFA Index indicates that prices did peak in May.

New home sales were also reported this morning. I’ll comment briefly on sales below, but let’s stick with prices first. These are also not seasonally adjusted. The median price of a new home YoY increased “only” +8.0% as of August, down from +24.2% in August of last year:

This confirms that new home prices have probably peaked as well.

Additionally, the maximum median YoY% change for existing homes in the past 12 months was last December at 15.8%. As of August the YoY increase was +7.7%, meaning those prices also probably peaked.

Summarizing the four measures of house prices, median prices of both new and existing homes appear to have peaked this summer. The FHFA Index suggests a peak either also happened this summer or is happening imminently. Only the Case Shiller index is probably several months away from peaking. 

Finally as to prices, as I have written many times over the past 9 months, the CPI measure of housing, “owners equivalent rent,” - which is about 1/3rd of the entire index -  lags actual house prices by about a year or more. Here are the YoY% changes of the house price indexes vs. OER (*2 for scale) over the past 20 years:

Through August, YoY% increases in OER have continued to accelerate. Since in the past episodes of significant downturns - 2001-2, 2005-6, and 2019-20, OER peaked about a year after the house price indexes, I suspect we will continue to see acceleration in OER through winter, or at least to the end of this year. 

Further, in several months I expect the YoY increase in the FHFA index to be less than that of OER - which means if the Fed continues to raise rates at that point, it will be chasing a receding phantom.

Turning briefly to new home sales, they increased sharply from July’s 6 year low of 532,000 annualized to 685,000, a 4 month high - but still well below the general pace of the past several years:

This data series is heavily revised, so we will see if the increase survives next month. There is no reason to suspect any change in the overall downward trend.

Gas and oil price update: good news and bad news

 

 -  by New Deal democrat

We’ll get some important house price information tomorrow, but there is no economic news of significance today, so let’s update gas and oil prices.

As indicated in the title, there’s good news and bad news. I’ll start with the bad news first.

According to GasBuddy, gas prices have not declined in over a week:

They have bounced off $3.64/gallon and stabilized at $3.66-7/gallon. Which still is only about $.20 higher than they were back in February.

But here’s the good news, in the form of a screen shot of oil prices this morning:

I have no idea whatsoever why oil prices suddenly declined further, or whether this will last, but as of now, oil prices are only about $3/barrel (or roughly $.06/gallon) higher than they were 1 year ago.

A comparison with the last year in gas prices:

suggests that, if oil prices continue in the $78/barrel range, gas prices will decline to roughly $3.25/gallon in the next several weeks.

Needless to say, if that happens, that is more money in the pockets of consumers to be saved, or spent on other things. It also means both consumer confidence and, politically, approval of President Joe Biden, will also increase - right before the midterm elections.