Employment - Daily Treasury Statements

[Beet]

After Charles Biderman of TrimTabs Investment Research put out an estimate of 472,000 jobs lost in June last month, ahead of the BLS report of -467k (consensus -350k) which caused a huge sell-off in the markets, I inquired into his methodology. Biderman uses something called the Daily Treasury Statement (DTS) to predict monthly job losses.

DTS consists of a daily statement of the actual deposits and withdrawals of the US Treasury of its own accounts at the Federal Reserve Bank of New York and the other places where the US government keeps its actual liquid assets. One of the items in the report are deposits from "Withheld Income and Employment Taxes" collected each day.

Compared to the BLS Household survey and Establishment survey, the DTS is an actual number, not a survey. Furthermore, it is not based on incomplete information. The BLS Establishment survey, regarded by the markets as the best government survey of the employment situation, is based on an incomplete sample and subsequently revised for 2 additional monthly releases; finally, it is benchmarked to state tax receipts for the March of each year. It is subject to birth-death adjustment model, seasonal adjustment, and excludes those who work from home as well as others. Many of these adjustments are high controversial, and they often account for a high proportion of the report. Finally, even if you accept the government's modeling, the actual 90% confidence interval is very high. For example, according to the BLS, a report of 100k jobs created could actually mean anywhere from -400k to +600k.

DTS, by contrast, represent actual numbers. The government cannot lie about these numbers because its deposits are stored at other institutions. There is no statistical modeling or adjustment. The process is highly transparent. The numbers are released promptly at 4 pm the following business day, each day. I do not know how Biderman gets his numbers, but I decided to take a peek into the DTS data myself and look at its receipts for income and employment taxes myself.

I simply totaled the government's monthly receipts and averaged it by the number of working days for each month. I then did an amateur control for seasonal adjustment by using annual YoY numbers. Unlike the government's releases, you can double check my work at the DTS website by clicking on the final day of each month and looking under "Withheld Income and Employment Taxes", "This Month to Date" for each month. This data shows that the employment declines are accelerating. No green shoots.



Update:
How this measure has tracked the S&P 500 historically
http://financialsight.blogspot.com/2008/03/withheld-income-and-employment-taxes.html

But currently the two measures seem to be diverging.

UPDATE for August:

Withheld Income and Employment Taxes month to date 126,389.
Working days in August 21.
Average daily rate is 6,019.
Average daily rate a year ago was 6,498.
YoY change is -7.4%.

[Verily]

According to your data, the decline is decelerating (although not reversing), not accelerating. The year-over-year change is the data point to look at because monthly employment data are not directly comparable unless adjusted, which this is not. Year-over-year, the decline from June to July is slight, only 0.74%, against a substantial decline of almost 6% in June. Of course, the reversal from May to June shows just how volatile--and therefore unrepresentative of the economy--these numbers are, but generally speaking they do not tell the tale you want them to.

[Beet]

Bump

Biderman: -488k
Consensus: ~ -350k

Deja vu?

[Sam Spade]

Note that I'm not disagreeing with you... 

[○∙◄☻¥tπ[╪AV┼cVê└]

If we have another 400k+ job loss, the people who are talking about how the recovery is happening need to STFU.

[Swing low, sweet chariot. Comin' for to carry me home.]

If we have another 400k+ job loss, the people who are talking about how the recovery is happening need to STFU.

looks like I'm still talking! 

-247k July
+43k revision in May and June
------
-203k

9.4% (Previous 9.5%)
+0.2% hourly earnings
increase in average work week (i don't have number)

[Beet]

Given the margin of error, this could mean anywhere between +250k to -750k.

And I love how this was leaked to Goldman Sachs last night. (not)

[Swing low, sweet chariot. Comin' for to carry me home.]

Given the margin of error, this could mean anywhere between +250k to -750k.

you are not going to get a big bad adjustment to this July number, the final number will probably be -150k to -350k.  too many indicators in the report confirm it (seperate unemployment report showing 9.4%, average hourly earnings increasing, avg workweek improved, etc, etc), and ADP payroll numbers also confirmed it.

as it stands now, it looks like it is very unlikely unemployment will rise above 10.5%

[Swing low, sweet chariot. Comin' for to carry me home.]

Beet,


where do you get idea that the margin of error in this report is +/- 500k?!  I've been listening to these reports for 30 years, and from my experience, the margin of error is +/-100k with a revision in the range of 100k being rare.  So, whenever one of these reports come out, I simply assume +/-100k, not +/- 500k

[Beet]

I have to apologize on this one-- the Labor department states that the margin of error on the household survey is +/- 430k, but the household survey has a smaller sampling base.

"For example, the confidence interval for the monthly change in total
employment from the household survey is on the order of plus or minus
430,000.  Suppose the estimate of total employment increases by 100,000
from one month to the next.  The 90-percent confidence interval on the
monthly change would range from -330,000 to 530,000 (100,000 +/- 430,000)"

http://www.bls.gov/news.release/empsit.htm

I am not sure what the margin of error is for the establishment survey. I'll admit, this is a good report. The main wild-card that I can see would be the extent to which the June/July auto layoff stuff is affecting the seasonal adjustments.

[Swing low, sweet chariot. Comin' for to carry me home.]

I have to apologize on this one-- the Labor department states that the margin of error on the household survey is +/- 430k, but the household survey has a smaller sampling base.

"For example, the confidence interval for the monthly change in total
employment from the household survey is on the order of plus or minus
430,000.  Suppose the estimate of total employment increases by 100,000
from one month to the next.  The 90-percent confidence interval on the
monthly change would range from -330,000 to 530,000 (100,000 +/- 430,000)"

http://www.bls.gov/news.release/empsit.htm

I am not sure what the margin of error is for the establishment survey. I'll admit, this is a good report. The main wild-card that I can see would be the extent to which the June/July auto layoff stuff is affecting the seasonal adjustments.

what I am saying is that I've never seen adjustments greater than +/-200k for this report and a 100k adjustment is rare....so, my experience tells me that this report should be taken with a +/- 100k grain of salt.

[CARLHAYDEN]

Beet,

As I believe I pointed out in May, summer unemployment numbers are highly unreliable as there are so many adjustments.

You really need to look in a couple of months.

[Richard]

If we have another 400k+ job loss, the people who are talking about how the recovery is happening need to STFU.

They need to STFU right now, but let them talk.  We can laugh at them later.  Cash for clunkers has done a great deal of pulling forward demand.  And so did the stimulus.  Give it a few more weeks, perhaps even months... the demand will come crashing down and *then* there will be trouble.

[Person Man]

Well. We do need some sort of artificial demand...but that usually leads to inflation. The question is- are we permanently screwed? I mean, we could just artificially raise the supply, like the Republicans want, but that usually creates its own problems, especially if only certain people are actually participating in the economy (then you get inflation..and growth based on a smaller market).. That's why we developed the cheap loan tactics of the Bush Administration. Demand couldn't keep up with Supply...I mean, if Reagan was right, we could of built and they would have all come, whether or not there was hard and fast credit.

I mean, the entire idea that we can give people who are good at money more money so that they can give money to other people just didn't work. What else do you want me to say? People who are good at making money try to find a way to make more money without sharing it. How do they do that, you say? Easy. Hire just a few guys and make it easier for them to produce more. The guys that are giving you the automation equipment are doin the same thing for themselves. Eventually, you do have money running around, but it stays very concentrated. What's the use of having a sh**t load of money if I am in the poor house? It just doesn't make any sense...and at this point, people go negative and blame the victim. But what if the victim has more training than 95% of the population and still cannot access the market?

Although, I will be fair here. If things start working out for me because any new Republican plans, I'll let you know....but in the mean time, I will give you an oppurtunity to convert me- whether to the left or to the right-

What do all of these policies and proposed policies do for me? I think each one of you should ask others (and yourselves) the same exact thing. I mean, do people make these statements and vote the way they do because its a form of sex or jacking off?...or are they actually getting something of tangible benefit for what they are doing to our country?

[CARLHAYDEN]

Weasel,

The so-called 'stimulus' bill passed earlier this year was deliberately and dreadfully malstructured.  Had much of the money gone for infrastructure (highways, bridges, border protection, etc.) that are really needed and will in the long term benefit the economy, it would have been a benefit.

Unfortunately, the bill, accurately called 'porkulus' was primarily loaded with ultraliberal spending,

Saving Nancy Pelosi's rats does not benefit the people of the United States.

[Swing low, sweet chariot. Comin' for to carry me home.]

Weasel,

The so-called 'stimulus' bill passed earlier this year was deliberately and dreadfully malstructured.  Had much of the money gone for infrastructure (highways, bridges, border protection, etc.) that are really needed and will in the long term benefit the economy, it would have been a benefit.

Unfortunately, the bill, accurately called 'porkulus' was primarily loaded with ultraliberal spending,

Saving Nancy Pelosi's rats does not benefit the people of the United States.

^^^^^^^^^

[Person Man]

Weasel,

The so-called 'stimulus' bill passed earlier this year was deliberately and dreadfully malstructured.  Had much of the money gone for infrastructure (highways, bridges, border protection, etc.) that are really needed and will in the long term benefit the economy, it would have been a benefit.

Unfortunately, the bill, accurately called 'porkulus' was primarily loaded with ultraliberal spending,

Saving Nancy Pelosi's rats does not benefit the people of the United States.

What should we have done instead? More of a stimulus? People would whine even more and if we just had a Republican-lite stimulus, it probably would only work for like half a business cycle...if we were lucky.

[Swing low, sweet chariot. Comin' for to carry me home.]

Weasel,

The so-called 'stimulus' bill passed earlier this year was deliberately and dreadfully malstructured.  Had much of the money gone for infrastructure (highways, bridges, border protection, etc.) that are really needed and will in the long term benefit the economy, it would have been a benefit.

Unfortunately, the bill, accurately called 'porkulus' was primarily loaded with ultraliberal spending,

Saving Nancy Pelosi's rats does not benefit the people of the United States.

What should we have done instead? More of a stimulus? People would whine even more and if we just had a Republican-lite stimulus, it probably would only work for like half a business cycle...if we were lucky.

it should have focused more on infrastructure needs:  repairing our bridges, moving to higher usage and production and distribution of natural gas, building more nuke power plants...basically it needed to be geared to removing infrastructure inefficiencies and foreign dependencies.  Repairing bridges would save fuel, converting to nuke power would free up more domestic natural gas for transportation use, and we could use that domestic resource to start to convert auto production to natural gas.

Just think how much less oil we would consume if we would commit to building 50 more nuke plants every 10 years and mandate the of conversion to auto production to natural gas (and/or electric) for all new cars by 2020.

That’s the kind of things governments needs to be spending public funds on.

instead of cap and trade on carbon, we should have cap and trade on oil.  and if you're concerned with carbon emissions, you could even stagger the conversion so that by 2020, all cars are run by natural gas and/or electricity, then by 2035 have all new cars run by solely electricity for the first 100 miles (or whatever distance the technology allows at that time), and after 100 miles have then run on natural gas (or liquidified coal, or whatever).

these types of common sense approaches would be favored by 75% of Americans

[Person Man]

Weasel,

The so-called 'stimulus' bill passed earlier this year was deliberately and dreadfully malstructured.  Had much of the money gone for infrastructure (highways, bridges, border protection, etc.) that are really needed and will in the long term benefit the economy, it would have been a benefit.

Unfortunately, the bill, accurately called 'porkulus' was primarily loaded with ultraliberal spending,

Saving Nancy Pelosi's rats does not benefit the people of the United States.

What should we have done instead? More of a stimulus? People would whine even more and if we just had a Republican-lite stimulus, it probably would only work for like half a business cycle...if we were lucky.

it should have focused more on infrastructure needs:  repairing our bridges, moving to higher usage and production and distribution of natural gas, building more nuke power plants...basically it needed to be geared to removing infrastructure inefficiencies and foreign dependencies.  Repairing bridges would save fuel, converting to nuke power would free up more domestic natural gas for transportation use, and we could use that domestic resource to start to convert auto production to natural gas.

Just think how much less oil we would consume if we would commit to building 50 more nuke plants every 10 years and mandate the of conversion to auto production to natural gas (and/or electric) for all new cars by 2020.

That’s the kind of things governments needs to be spending public funds on.

instead of cap and trade on carbon, we should have cap and trade on oil.  and if you're concerned with carbon emissions, you could even stagger the conversion so that by 2020, all cars are run by natural gas and/or electricity, then by 2035 have all new cars run by solely electricity for the first 100 miles (or whatever distance the technology allows at that time), and after 100 miles have then run on natural gas (or liquidified coal, or whatever).

these types of common sense approaches would be favored by 75% of Americans

Of course I favor something like this! ...but it would be virtually impossible to get any of this stuff through...whether its a liberal government or a conservative government. The problem is that the Oil lobby in this country will destroy any real chance at change. We need nukes. We need solar power. We need natural gas and perhaps even some coal....Like I said, we might have to wait until our immediate survival depends upon it.

[Swing low, sweet chariot. Comin' for to carry me home.]

Weasel,

The so-called 'stimulus' bill passed earlier this year was deliberately and dreadfully malstructured.  Had much of the money gone for infrastructure (highways, bridges, border protection, etc.) that are really needed and will in the long term benefit the economy, it would have been a benefit.

Unfortunately, the bill, accurately called 'porkulus' was primarily loaded with ultraliberal spending,

Saving Nancy Pelosi's rats does not benefit the people of the United States.

What should we have done instead? More of a stimulus? People would whine even more and if we just had a Republican-lite stimulus, it probably would only work for like half a business cycle...if we were lucky.

it should have focused more on infrastructure needs:  repairing our bridges, moving to higher usage and production and distribution of natural gas, building more nuke power plants...basically it needed to be geared to removing infrastructure inefficiencies and foreign dependencies.  Repairing bridges would save fuel, converting to nuke power would free up more domestic natural gas for transportation use, and we could use that domestic resource to start to convert auto production to natural gas.

Just think how much less oil we would consume if we would commit to building 50 more nuke plants every 10 years and mandate the of conversion to auto production to natural gas (and/or electric) for all new cars by 2020.

That’s the kind of things governments needs to be spending public funds on.

instead of cap and trade on carbon, we should have cap and trade on oil.  and if you're concerned with carbon emissions, you could even stagger the conversion so that by 2020, all cars are run by natural gas and/or electricity, then by 2035 have all new cars run by solely electricity for the first 100 miles (or whatever distance the technology allows at that time), and after 100 miles have then run on natural gas (or liquidified coal, or whatever).

these types of common sense approaches would be favored by 75% of Americans

Of course I favor something like this! ...but it would be virtually impossible to get any of this stuff through...whether its a liberal government or a conservative government. The problem is that the Oil lobby in this country will destroy any real chance at change. We need nukes. We need solar power. We need natural gas and perhaps even some coal....Like I said, we might have to wait until our immediate survival depends upon it.

no, I work for the oil companies, and I can tell us with 100% confidence that the oil companies would LOVE to see greater demand for natural gas because it much less risky to explore for gas in North America.

for example:

http://www.latimes.com/business/la-fi-chevron1-2009aug01,0,822005.story

[Person Man]

Then, why aren't they pushing it? I mean, if they are really willing to do something about our problems, I'd listen to them....but will we have to wait until after the fact before things start to get better?

[CARLHAYDEN]

Then, why aren't they pushing it? I mean, if they are really willing to do something about our problems, I'd listen to them....but will we have to wait until after the fact before things start to get better?

Weasel,

The Obama administration is NOT listening to any ideas about producing more energy from either 'fossil,' hydroelectric nor nuclear sources.

Essentially they are into 'conservation,' i.e. make do with less.

[Person Man]

Then, why aren't they pushing it? I mean, if they are really willing to do something about our problems, I'd listen to them....but will we have to wait until after the fact before things start to get better?

Weasel,

The Obama administration is NOT listening to any ideas about producing more energy from either 'fossil,' hydroelectric nor nuclear sources.

Essentially they are into 'conservation,' i.e. make do with less.

Well, we need both...but conservation seems to be the easiest thing to do right now that will produce the greatest long-term effects. Drill, Drill, Drill? It probably WONT hurt the environment that much. Will it help us much? Probably for no more than 5 years...and we will be exactly where we are today...except for wasting our money.

[Beet]

Withheld Income and Employment Taxes month to date 126,389.
Working days in August 21.
Average daily rate is 6,019.
Average daily rate a year ago was 6,498.
YoY change is -7.4%.

[Sam Spade]

Withheld Income and Employment Taxes month to date 126,389.
Working days in August 21.
Average daily rate is 6,019.
Average daily rate a year ago was 6,498.
YoY change is -7.4%.

Do you have August 2007 or 2006 numbers, because that's pretty bad in comparison to August 2008?

I scanned through the employment report today and it looked very weak.  In fact, I'll go out on the record and say that July looks like "seasonal adjustment" games.  Birth/death model added 118K jobs.  Both July and June numbers were revised down about 25K.

Also, the big number that I watch - average workweek - stayed the same at 33.1 hours.  Not good, especially if it goes negative again.