The “gold standard” QCEW through Q2 suggests little if any employment growth this year

 

 - by New Deal democrat

The Quarterly Census of Employment and Wages (QCEW) is “the gold standard of US employment measures. It is an actual census of 95%+ of all employers, who must report new employees for purposes like unemployment and disability benefits. Because of this, it is used for the final revisions, a/k/a benchmarks, for monthly jobs numbers, which are estimates based on surveys. Its drawbacks are that it is not seasonally adjusted, and is delayed months after the end of the quarter.

The important news is that, after a considerable delay due to the government shutdown, it was released for Q2 this morning.  And there was good news and bad news.

The good news is that, on a non-seasonally adjusted basis, 2.3 million jobs were added in the 2nd quarter of this year, the same number as was the case in for the NSA private monthly payrolls number. After seasonal adjustment, that translated into just under 200,000 jobs net for the three months, as per the below graph of SA and NSA private nonfarm payrolls over the past four years through June:

 https://fred.stlouisfed.org/graph/?g=1OXVJ 

The bad news is that, in contrast to the 1.286 million jobs added in the 12 months after June 2024, only 237,000 were added per the QCEW for the entire 12 months period. In other words, beginning in late 2024, the jobs sector has been dead in the water.

Per previous releases of the QCEW, even at the end of 2024, on a year over year basis employment grew by about 1.4 million, or 0.9%. Then in the first quarter of this year, comparisons fell off a cliff again. Based on the most recent updated QCEW data, only about 427,000 jobs were added NSA YoY through the end of March. So there was further deceleration in Q2. 

These are not seasonally adjusted numbers, so it is entirely possible that the NSA gain turns into an actual decline. In my August review of the Q1 QCEW, I concluded that it wss “suggesting there might not have been any job growth at all this year.” Updated through Q2, it seems likely there was a very small gain, but not even keeping up with prime employment age population growth, i.e., firmly supporting the increase in the unemployment rate this year. And it is still possible that there have been no net employment gains whatsoever this year. 

Consumer inflation went to sleep in October-November, driven mainly by a steep deceleration in shelter price increases

 

 - by New Deal democrat

A truncated version of the CPI was reported for November this morning. Since no data was collected for October, the monthly change in those values was not calculated, the Census Bureau apparently having elected not to publish with the cumulative two month changes since September. Nevertheless, the two month numbers are easily calculated, and the important ones are all the same: headline, core, shelter, and headline minus shelter all rose 0.2% in the last two months.

All of which means any calculations except for YoY ones have to be taken with numerous grains of salt. The only exceptions are for data that was privately collected and then integrated into the calculations, which were not affected by by the government shutdown.

With those caveats out of the way, by far the most important component of the CPI this month was shelter, and in particular the sharp deceleration in its YoY increase. Actual rent YoY declined from a 3.6% increase in September to 3.0% in November. Imputed Owner’s Equivalent Rent decelerated from a 3.8% YoY increase in September to 3.4% in November. The total for shelter decreased from 3.6% to 3.0%. Note that in the graph below the November readings do not appear:

 https://fred.stlouisfed.org/graph/?g=1OVTF 

All of these are the lowest since summer 2021, and are in line with their pre-pandemic ranges. And they are very big declines for only two months, so I am treating them with extra caution. It would not surprise me at all to see a signficant revision of these numbers next month.

This big decline in the shelter component of CPI, which is over 1/3rd of the entire calculation, is what drove the headline and core numbers down. YoY headline inflation was only 2.6%, the lowest since July, and core inflation YoY was 2.6%. YoY inflation ex-shelter was 2.5%:

 https://fred.stlouisfed.org/graph/?g=1OVSF 

All of these are equivalent to their readings back in July, i.e., lower than either August or September. 

Collection of data for new and used vehicles was not affected by the shutdown. New vehicle prices rose 0.1% in October and 0.2% in November. Used vehicle prices rose 0.7% in October and 0.3% in November. As a result their YoY increases were 3.6% and 0.6% respectively:

 https://fred.stlouisfed.org/graph/?g=1OVUz 

Another recent problem child for inflation was transportation services, mainly vehicle parts and repairs as well as insurance. These also have gone somnolent, with an incrrease of only 1.7% YoY, the lowest since early 2021:

 https://fred.stlouisfed.org/graph/?g=1OVUT 

Finally, prices for gas home service, and electricity have also become a problem, the latter likely a side effect of the building of massive data centers for AI generation. Electricity prices were up 6.9% YoY in November, the highest increase since 2008 except for the post-pandemic inflation:

 https://fred.stlouisfed.org/graph/?g=1OVV5 

This has already created a backlash, and I expect that backlash to intensify.

In summary, November’s CPI report covering two months showed a big slowdown in all major components of consumer inflation, and in particular that for shelter. This in turn drove the YoY comparisons back down to where they were last summer. That being said, treat this month’s report with extra caution, and look out for significant revisions next month.

Jobless claims continue to paint a much more positive picture than the unemployment rate

 

 - by New Deal democrat

[Note: An update to my OS the other day has nuked my ability to post graphs. For now, I will post links to FRED graphs that you can access. For this post, I am only using one such link. If I am unable to resolve the problem, drastic action may be reqeured.]

The return to normalcy in jobless claims after a skewed reading for Thanksgiving week due to unresolved seasonality continued. Initial claims declined -13,000 to 224,000, very close to the midpoint of its range since the beginning of July. The four week moving average, which still includes the outlier Thanksgiving week, rose 500 to 217,500. Continuing claims, which are delayed one week and thus only one week out from the Thanksgiving skew, rose 67,000 but were still below 1.9 million at 1.897 million.

On the YoY% basis which is more important for forecasting purposes, initial claims were up 0.9%; the four week average still down, by -3.5%, and continuing claims higher by 1.9%.

It will be two more weeks before the skewed week is out of the four week average, so I would discount that reading.

Still, very slight YoY increases in new and continuing jobless claims are neutral and do not portend an imminent recession.

Last week I noted that the unresolved post-pandemic seasonality that was so apparent in 2023 and 2024 has been much more muted, especially in the second half of this year. Claims did rise into June, but then sharply declined in July, and have generally remained in that range since. I further noted that in the first half of this year, jobless claims typically were in the +10,000 range YoY. That all changed since the end of June. In the 23 weeks through one week ago, jobless claims averaged just under -4,000 lower YoY.

This week’s initial claims number was only 2,000 higher than last year’s for the equivalent week, which adds another week of evidence for the idea that there has been somewhat of a change in regime for jobless claims since the middle of this year. Significantly, it also suggests a serious disconnect between what has been happening with both initial and continuing claims vs. the unemployment rate, which as you recall increased to a multiyear high in the jobs report released for November the other day. Here is the update on that comparison:

 https://fred.stlouisfed.org/graph/?g=1OVPM 

An unemployment rate 0.4% higher than one year previous has in the past almost always meant that a recession has begun. The exceptions were one month in the 1950s, two months in 1963 - and five months last year. The jump in the unemployment rate is consistent with the significant increase in continuing claims that started in June. 

It is also possible that this is a reaction to the anomaly last year in the unemployment rate, which was put down (rightly I think) to a miscalculation of the impact of immigration on the population numbers. But this year all the evidence is that has reversed. The answer apparently lies in the 0.6 million surge in the unemployment level as well as a 1.2 million surge in the civilian labor force calculation in the Household Survey since July, especially in comparison with very low increases calculated last November. I suspect we are going to have to wait for January for YoY comparisons to be more valid, since they will not be against the big immigration undercount of 2024. Since that undercount does not affect jobless claims numbers, at very least we have to take the alleged triggering of the “Sahm Rule” in November’s jobs report with extra grains of salt.

Real retail sales contract; depending on inflation report may signal further job losses

 

 - by New Deal democrat

[Note: An update to my OS the other day has nuked my ability to post graphs. For now, I will post links to FRED graphs that you can access. If I am unable to resolve the problem, drastic action may be reqeured.]

Real retail sales, one of my favorite broad-economy indicators, was finally updated yesterday, although as per most government data releases, it was still somewhat stale, being for September and October. Nevertheless, it does give us some new information, so let’s take a look.

In nominal terms retail sales rose 0.1% in September, and were unchanged in October, but since consumer prices rose 0.3% in September, real retail sales declined -0.2%. CPI for October hasn’t been reported yet, but another 0.3% increase would mean a further -0.3% decline in real sales for October. The below graph, through September, shows real retail sales (blue) and the similar measure of real spending on goods (gray), both normed to 100 as of their peaks in December of last year:

 https://fred.stlouisfed.org/graph/?g=1OTNM 

Both have only exceeded that peak by at most 0.1% this year, and in September both were below it by -0.1%, although the 3 month moving average continued to increase slightly.

When real retail sales turn negative YoY, going back 75 years it has almost always meant a recession (with the very notable exception of 2022-23, which was countered by a steep positive supply shock). Here is what the YoY comparison looks like for both of the above metrics:

 https://fred.stlouisfed.org/graph/?g=1OTNG 

Neither are negative YoY, but both have decelerated sharply since their YoY peaks in early spring. Should the trend continue, they could be negative YoY by December or January.

Finally, because consumption leads employment, here is the update of YoY real sales (/2 for scale) together with employment (red), updated through yesterday’s report:

 https://fred.stlouisfed.org/graph/?g=1OTNB 

The sharp deceleration in YoY growth in consumption has forecast the slide in employment. Should the October CPI release mean a further deterioration in sales for that month, that would forecast even more deceleration - in fact a downturn,- in employment in the immediate months ahead.

Combined October and November jobs report: a hairs-breadth from recessionary, at best

 

 - by New Deal democrat

At Last this morning we got some up to date labor data from the federal government, but only partially. The Establishment Survey was updated for both October and November, while the Household Survey was not conducted at all for October, and so jumps from September to November.

In November virtually all of the other reports, including from the regional Feds and ISM, as well as others, indicated an actual decline in employment. While that didn’t occur, as we will see below, the general tenor was negative, albeit nuanced.

Below is my in depth synopsis. Note that for the Establishment numbers, I give both October and November reads, as well as the two month net change.

HEADLINES:

• -105,000 jobs lost in October; 64,000 added in November for a net change of -41,000.
•  Private sector jobs increased 52,000 in October and 69,000 in November, for a net gain of 121,000. Government jobs declined -157,000 in October (these are mainly the delayed DOGE layoffs), and another -5,000 in November for a total loss of -162,000.
• September was revised downward by -11,000 to +108,000. 
• The alternate, and more volatile measure in the household report, rose by 96,000 jobs since September.
• The U3 unemployment rate rose 0.2% to 4.6%, since September, the highest since September 2021.
• The U6 underemployment rate rose 0.7% to 8.7% since September.
• Further out on the spectrum, those who are not in the labor force but want a job now rose 203,000 since September to 6.136 million, aside from August the highest level since September 2021.

Leading employment indicators of a slowdown or recession

These are leading sectors for the economy overall, and help us gauge how much the post-pandemic employment boom is shading towards a downturn. For the second month in a row they were mainly negative:

• The average manufacturing workweek, one of the 10 components of the Index of Leading Indicators, rose 0.1 hour in both October and November to 41.2 hours, but remains down -0.4 hours from its 2021 peak of 41.6 hours.

• Manufacturing jobs decreased by -9,000 in October and -5,000 in November, the sixth and seventh declines in a row. This series declined sharply in the second half of 2024 before stabilizing earlier this year. It is now at a 3.5 year low.
• Truck driving, which had briefly rebounded earlier this year, declined -1,300 in October and another -4,400 in November, for a total decline of -5,700.
• Construction jobs declined -1,000 in October, but rose 28,000 in November.
• Residential construction jobs, which are even more leading, rose 900 in October and another 3,400 in November for a total of 4,300, making three increases in a row.
• Goods producing jobs as a whole declined -9,000 in October, but rose 19,000 in November, for a net gain of 10,000, after declining earlier this year for 4 months in a row. 

• Temporary jobs, which have declined by over -650,000 since late 2022, declined again in both October, by -12,700, and November, by -5,000, for a total decline of -17,000, a new post-pandemic low.
• The number of people unemployed for 5 weeks or fewer rose 316,000 from September to November,  to 2,543,000, the highest number since the end of 2020.

Wages of non-managerial workers 

• Average Hourly Earnings for Production and Nonsupervisory Personnel increased 0.4% in October and another 0.5% in November, for a net gain of 0.9%, with a YoY gain of +3.9%. This continues to be significantly above the 3.0% YoY inflation rate through September.

Aggregate hours and wages: 

• The index of aggregate hours worked for non-managerial workers was unchanged in October, but increased 0.1% in November, and is up 1.3% YoY, its highest rate since January.

• The index of aggregate payrolls for non-managerial workers rose 0.4% in October and another 0.5% in November, and is up 5.2% YoY, its highest rate since April.

Other significant data:

• Professional and business employment declined -7,000 in October, but rose 12,000 in November, for a net change of 5,000. These tend to be well-paying jobs. This is the fifth decline in a row, and is the lowest number in over 3 years. It is also lower YoY by -0.2%, which in the past 80+ years - until now - has almost *always* meant recession. This is vs. last spring when it was down -0.9% YoY.
• The employment population ratio declined -0.1% to 59.6% from September through November, vs. 61.1% in February 2020.
• The Labor Force Participation Rate rose +0.1% to 62.5% from September through November, vs. 63.4% in February 2020.

SUMMARY

Looking at this from the perspective of the meet two month changes, it was a poor report with only a few bright spots.

Let me note the bright spots first: construction employment, and wage growth. Despite the pounding that the housing market has taken, residential construction employment made a new post-pandemic high in November. And the residential sector was relatively the weakest one within construction. Non-residential construction employment rose sharply - I suspect due to AI-data center building.

Additionally, wages grew sharply in both October and November, which also powered a sharp rise in aggregate payrolls, a very good sign. Additionally, manufacturing showed signs of life as the average workweek in that sector rose.

But these were outweighed by all the negatives. On net, jobs contracted by -41,000 since the last report. Unemployment rose 0.2%, and underemployment rose by 0.7%, both to multi-year highs. Both short term new unemployment and those who want a job are not even looking rose to close to post-pandemic highs. Employment declined in manufacturing, trucking, and temporary help - all leading sectors. And while labor force participation rose, the employment to population ratio declined.

Right now the only sectors holding employment afloat are health care, which on net accounted for all of the gains in the past few months, as well as construction (likely mainly related to AI data center spending). In the seven months since April, only 119,000 jobs have been added, an average gain of only 17,000 jobs per month. And that’s before the likely downward revisions which will be made once the QCEW is integrated into the results.

In summary, the jobs market is either a hairs-breadth above contraction, or actually in contraction.

What do vehicle miles traveled and gas usage tell us about the economy?

 

 - by New Deal democrat

Today is the last day of our data drought before the onslaught of delayed reports that begins tomorrow with the November jobs report and CPI. 

In the meantime, there is a commenter on another economics site who generally believes that everything is OK as along as vehicle miles traveled YoY stay positive - and as of October, the rolling 12 month average was higher by 1.0%. Others robustly disagreed. So I thought I would take a look.

In the first place, here is the 12 month rolling average of total vehicle miles traveled for the past 50+ years in absolute terms:

[I am having a problem with Google giving me access to my photos, so instead here is a link to a FRED graph of the YoY% change in the 12 mile moving average of vehicle miles driven:

 https://fred.stlouisfed.org/graph/?g=1ONkm ]

It’s pretty obvious that total mileage traveled turned down, or at least decelerated markedly, at the onset of or shortly before all of the recessions in the past 50 years except for the pandemic.

But of course that just tells us that it is a *coincident* indicator, best for confirming in the rear view mirror what other data has already been suggesting. Notably, only in 1979 and 2000 did it turn down or decelerate in advance; in 1981, it did not turn down until months after the recession had already started.  Further, there was a significant deceleration beginning in the summer of 2005 that did not correlate to any recession in the next year; and for 4 entire years after the Great Recession miles traveled were completely flat, punctuated with several periods of small declines that did not coincide with recession, and so would have been very wrong signals had they been followed at the time.

Instead, I think that over the long term total vehicle miles traveled have told us that almost all US recessions in the past 50 years have at least in part been due to oil price shocks. And indeed the deceleration in miles traveled in 2005 coincided with gas prices hitting $3/gallon for the first time in the wake of Hurricane Katrina, and the four year period after the Great Recession was characterized by what I called at the time the “oil choke collar”; i.e., every time the economy would pick up, gas prices would shoot to $4, cooling the economy down, which would result in gas prices sinking back to $3, and the cycle would repeat.

In support of this, here is the YoY% change in vehicle miles traveled compared with the YoY change in the price of gas (red, /10 for scale) and for the period predating that statistic, the price of oil (orange, /20 for scale):

[To be supplied]

The oil price shocks stand out, as the YoY change in prices coincides with a sharp downturn in miles traveled. When the shock ends, vehicle miles recover. By contrast, in the 2001 recession and the pandemic there were no spike in gas prices, and the downturn or deceleration in usage was caused by other things.

Similarly, when we look at vehicle miles traveled compared with real GDP (red), we see a strong although not perfect (nothing ever is!) correlation:

[To be supplied]

 

In general, this suggests that we should expect at least a deceleration if not a downturn in vehicle miles traveled roughly coincident with the onset of recession. So here is the close-up of the past four years:

[To be supplied]

Through October, YoY vehicle miles traveled were generally steady, suggesting no recession had begun as of that time - but giving us no information with which to forecast.

But there is a very similar metric, which is gasoline product used, which is updated every week, and thus is current through the first week of December, and here is the story it tells:

[Liink to E.I.A. Data and grap of 4 week average of gasoline usage]

The four week average was significantly negative YoY through most of the summer before recovering in September into October. But for the past five weeks since the beginning of November, it has been very negative.

So, if the signs were positive into October, the decline in usage since then does not bode well for the November data were are about to be deluged with starting tomorrow.