US Economy Tanking- 9/26/22

[DrStool]

I am really in the weeds, as I continue to deal with the fallout of trying to live in a renovation site. My publication schedule is in chaos. I've been working on a Liquidity Trader Money Trends update, which will be posted hopefully within the hour. Then next a Technical Trader update, hopefully before the NY open. Finally, I will almost certainly have to delay the swing trade screen picks until Tuesday morning. 

Meanwhile, this comment drew my attention:   

On 9/24/2022 at 2:39 AM, sandy beach said:

Federal withholding & employment taxes up $16.9% in September from last year. Wow!

That brings up the fact that I track and report this very important data every month virtually in real time. So I'm surprised to see that someone already knows what September will bring. I wasn't aware of the existence of future data.  Curious as to the source. 

Anyway, the above report is FALSE! I don't like to leave misleading data out there without setting the record straight. 

The current month to date through September 22, the last posted data for September is +3.9% nominal, or around negative 1% real, suggesting recession has started. This is a rolling 4 week average of 2 weeks of collections. So it is smoothed and lagged a bit, but even comparing wildly volatile day to day, its barely positive nominal, and negative real. The trend is clear. 

Let the record stand corrected. 

I will be updating this report when the month end data is available. The Treasury will post it Monday evening, and I will attempt to post an update on Tuesday or Wednesday if all goes well. 

My last look. 

Withholding Tax Collections Collapsed in August But BLS Data Didn’t Show It – Part Two

I shall return in an hour or two to post the hourly ES chart update. 

[DrStool]

Markets Face Catastrophe as Dealers Mitigate Too Little Too Late

[fxfox]

1 hour ago, DrStool said:

Markets Face Catastrophe as Dealers Mitigate Too Little Too Late

Given that, it is quite remarkable that the Put/Call ratio is as high as never before. At least that is said in the media. I have doubts that that Put/Call ratio number is correct, smells like something isn't correct with that reading, like if something extraordinary or some important fact or whatever is not considered. Is the ratio really higher as around the Lehman Event in 2008? 

[DrStool]

Another irrelevant sentiment indicator. Public sentiment follows the trend. What was extreme in a bull market would not be extreme in a bear market. 

I am only interested in Primary Dealer positioning.  There are no puts/calls on their bond positions. They hedge with futures. 

[DrStool]

Could be in a 5 day cycle up phase here. I'd expect it to be weak, followed by a crash in a day or two. Still a 5 day cycle low projection of 3620. Although we got close enough for a DCB. 

• Meanwhile, back at the big picture:

 

• Stock Market in Crash Mode September 26, 2022
• Markets Face Catastrophe as Dealers Mitigate Too Little Too Late September 26, 2022
• Swing Trade Screens – Restrained Bearish Instincts September 20, 2022
• Small Comfort for Gold Holders September 23, 2022
• Fed Speeds Into Dead Man’s Curve, More Black Tuesdays Ahead September 15, 2022
• There Will Be More Black Tuesdays September 14, 2022
• Withholding Tax Collections Collapsed in August But BLS Data Won’t Show It September 2, 2022 
• Warnings of August Liquidity Crash Come to Fruition – Here’s What to Do August 28, 2022
• Has Rule Number One Been Repealed? August 18, 2022
• “As Good as It Gets” Was Good While It Lasted August 6, 2022
• Treasury Confirms Supply Tsunami We Expected – Will Obliterate Everything August 3, 2022

If you're serious about the underlying forces of supply and demand that drive the markets, join me! 

If you are a new visitor to the Stool, please register and join in! To post your observations and charts, and snide, but good-natured, comments, click here to register. Be sure to respond to the confirmation email which is sent instantly. If not in your inbox, check your spam filter. 

[Jorma]

Flash crash in the GBP last night?  

[fxfox]

22 minutes ago, DrStool said:

Another irrelevant sentiment indicator. Public sentiment follows the trend. What was extreme in a bull market would not be extreme in a bear market. 

I am only interested in Primary Dealer positioning.  There are no puts/calls on their bond positions. They hedge with futures. 

Makes sense. Thanks.

[Jorma]

Fascism returns to Italy on the 100th Anniversary of its arrival. News wise It doesn't get any worse than that. 

[DrStool]

They got 25% of the vote. Italy's electoral system is a mess. DeGaulle was smart in the way he set up the French constitution.  For practical purposes an extremist minority can't win. That doesn't guarantee good government, but it largely keeps the wingnuts out. 

[fxfox]

6 minutes ago, DrStool said:

They got 25% of the vote. Italy's electoral system is a mess. DeGaulle was smart in the way he set up the French constitution.  For practical purposes an extremist minority can't win. That doesn't guarantee good government, but it largely keeps the wingnuts out. 

and pretty much of those 25% were from former Salvini and Berlusconi voters (both parties lost votes) and yes the Italian Electoral System is a catastrophe.

[sandy beach]

4 hours ago, DrStool said:

Meanwhile, this comment drew my attention:   

That brings up the fact that I track and report this very important data every month virtually in real time. So I'm surprised to see that someone already knows what September will bring. I wasn't aware of the existence of future data.  Curious as to the source. 

Anyway, the above report is FALSE! I don't like to leave misleading data out there without setting the record straight. 

The current month to date through September 22, the last posted data for September is +3.9% nominal, or around negative 1% real, suggesting recession has started. This is a rolling 4 week average of 2 weeks of collections. So it is smoothed and lagged a bit, but even comparing wildly volatile day to day, its barely positive nominal, and negative real. The trend is clear. 

Let the record stand corrected.

My apologies doc! In my haste I didn't accurately describe what I was comparing to derive the percentage change. It was not data from the future and was not exactly year-to-date change either. I take row "Withheld Income and Employment Taxes” and column “Federal Tax Deposits Month to Date” as of 9/20/2022 the latest data available at the time of post which is $168,602. I then count the number of reporting days for the month which was 13 and divide $168,602/13 = $12,696 for the average tax withholding per reporting day. I then take the last reporting date for September last year which was September 30, 2021 and the 21st reporting day of the month and use the same row and column ($232962) and divide $232962/21=$11,093 for the average tax withholding per reporting day for the *entire* month.  And that gives me a 16.9% increase. My method is fairly simple and is prone to bias if there is an annual trend towards higher tax collections later in the month (I haven't checked this). I also don’t do any moving averages or anything comparable to what you do. I just occasionally check the numbers to see what the temperature of the economy is looking like. I apologize for not accurately reporting what I was seeing. I’ll try to be more careful in future. I much appreciate the work you do.  

[fxfox]

 

[DrStool]

Payrolls are typically paid either weekly, bi weekly, or twice a month. The largest collections are at the end of the month. Likewise, in between, day to day, the collections are volatile. So if you compare a partial month to a complete month, you may get a whacky result.

There's also the problem of collections that differ in the same month year to year because they both may not end on a weekday. And using reporting days vs. calendar days also can cause a skew since a huge percentage of US workers have 7 day work weeks. 

Last year, weekly payers would have made their last deposit for September on October 1. So that would reduce the September total below what was actually earned in September. Also, this year by only comparing collections through 9/21, you are including 6 weekend days. 6 * 30/21 = 8.6. Last year's full month would have included earnings for 8 weekend days. 

So there are a lot of ways for there to be a haywire result using your method. Even a direct month to month calendar method often has wacky results. A little smoothing using fixed time periods corrects for that, with the downside being that it does result in a lag of a week or so. That's usually not a problem if you apply TA to the chart. 

In the end I have always felt that good analysis requires a visual. Numbers alone just don't do it for me. TA works just as well on economic series as it does on prices, maybe even better. 

The recent trend on that chart is pretty clear.  That's why I immediately saw a red flag in your post.  

When analyzing data, in comparing time periods always be sure to compare like to like, and make a visual. Seeing is believing.  That's why I put tons of charts in my reports.  I need to see it. 

[DrStool]

This morning's rally fell short of a parallel made by connecting the two most recent lows and mirroring it from the last high. So either the rally isn't finished yet, or things are about to get really bad. Or good, if you're short. 

[DrStool]