Matt Trivisonno, a software developer in Miami Beach who follows the market, has come up with a neat way to track changes in the Federal Government’s witholding tax receipts.
After being rangebound for the prior 4 years, the year-over-year growth in withholding began rolling over in Q4 of 2007 — a possible date marking the beginning of the current recession:
>
click thru for larger graph
Source: Matt Trivisonno’s Blog
>
Note that withholding comes directly from Wages and Income, and would
likely precede major shifts in consumer spending or credit.
>
See also:
In Debt Crisis, Uncle Sam Is Piling It On
Mark Gongloff
WSJ, March 24, 2008; Page C1
http://online.wsj.com/article/SB120632203228958435.html
>
~~~
What do the numbers on the side of the chart represent?
“What do the numbers on the side of the chart represent?”:
The percentage change from the same period a year earlier.
Read the title of the graph => year-over-year percentage change in withholding taxes.
Not like it would dramatically change the conclusion derived but I wonder what the graph would look like if it was adjusted for the decrease in marginal tax rates that have occurred over the last few years.
In addition, I would be a little wary of this graph if there was an increase in marginal tax rates. Initially, you would likely see a short-term increase in withholding tax although increased tax rates would be detrimental to the economy over a longer time frame.
Have you noticed that the only way to get this year’s Federal Stimulus check is by ‘filing your income tax’? I’ve been hearing all the mainstream media outlets push the idea that you have to file this year, and there’s even a new simple form for those people who haven’t filed in a long time (i.e. Social Security & Welfare recipiants).
It sounds to me like they’re trying to bribe some of those non-filers into getting on the books… finally. I didn’t even know you could get away with that until I heard a bit on Marketplace last year when they interviewed non SS & Welfare dependents who ‘just don’t file.’
this is one of the best new indicators to come along in a long time. great catch.
You get much the same thing just by watching nominal income growth although that is only reported once a month and late in the month at that.
But couldn’t it also suggest that wages and income are rising?
1) If the % remains fixed but taxes are rising it could mean wages are rising.
2) Or the composition of employment is changing. Self employed are becoming employed by others, or 1099 employees are becoming reg. employees
1 and 2 are good news
3) Alternatively you are saying the % is changing. You’d hear about that and it hasn’t happened.
4) Inflation. Minimum wage hikes kicked in big time in January, so wages are up but so are taxes.
4 is sorta good news
Truthfully, it’s hard to see how this chart is bad news.
One last clarification, the slumping taxes from December of 2007 to now is bad news, but a long way from recession. (see Matt’s site and take a look at 2002 it hit -3%)
The very last data point (the small blip up) is the good news I was talking about.
With a drop in tax revenues, we’re going to get inundated in Treasuries.. The capital junkie needs a bigger capital fix each week just at a time when supply is drying up.
MD,
This probably should be considered to be one of many indicators as opposed to something to base all decisions on.
If you look back at the last recession which started around Q1/Q2 of 2001, withholding tax change was still greater than +6%. The change in withholding tax went negative around the time that the US economy officially came out of the recession!
If there a correlation implied by this, it is between the change in withholding tax and the markets, not between the change in withholding tax and the economy.
I’m on record that the recesssion started in the fourth quarter of 2007 (late October is my best guess)
So I guess I’m surprised that Matt’s chart isn’t worse.
I looked at the Census B. data for national totals of state government tax revenue for current and prior quarters. The last quarter available is 3Q2007. The year over year change in sales tax revenue for all states is quite interesting. For example:
1Q2000 11.95% – local peak
2Q2001 -0.84%
1Q2002 -1.40%
4Q2004 10.73% – local peak
3Q2006 6.24% – last local peak
4Q2006 4.28%
1Q2007 3.39%
2Q2007 2.35%
3Q2007 0.43% – last data point
Lets see when we get the data for 4Q2007. Not hard to predict that it would be negative.
This is a useful new metirc. Nice work, Matt, and thank you Barry.
One quick comment. A detailed read of Matt’s blog entry indicates that what is really being charted is equivalent to:
the Year over Year change
in the One Year Moving Average(total).
With such a long period moving average you will get a signficant lag at the turning points. It would be great to see what the same data would look like with a 30 day or 90 day moving average.
My guess is that we would be seeing a much steeper decline with the short period moving average.
An additional twist of interest would be to adjust the results to account for cpi inflation.
For reference, here’s Matt’s description for how he calculates each data point.
“Each dot on my charts represents two years of digested data. The number is calculated like this: All of the withholding-tax revenue over the past year is totaled. Then, another total is calculated for the prior year, and the two numbers are compared to see what the percent change was.”
Even with the slowdown, withholdings are still increasing over 5%, which is much greater than inflation. Social security tax receipts were up 5% in 2007. Every measure of wage and income tax receipts is growing much faster than the bulls**t wage inflation numbers coming out of the BLS. Yet everyone wants to take the bulls**t wage inflation numbers as gospel when comparing them to other inflation numbers to claim that real wages are falling.
TrimTabs tracks this type of data closely. Check out this post on Barron’s tech trader blog from January 3, 2008:
“Trim Tabs Investment Research asserted today that the U.S. economy added zero jobs in December, and that “a consumer recession has already begun.” Charlie Biderman, who runs Trim Tabs (and who once upon a time worked at Barron’s) says jobs data from the Bureau of Labor Statistics are inaccurate. He says that based on tax withholding data, “the U.S. economy added no jobs in December and an anemic 49,000 jobs in October and November combined.”
Trim Tabs says its own online job postings index fell 7.9% in December and 11.4% in the fourth quarter. Biderman says that consumer spendable cash – after-tax wage income, net equity extraction from real estate and tax refunds – will be down 4% in the first quarter versus a year ago. And he says a consumer recession “means that stock prices are likely to suffer more losses early this year.”
TrimTabs tracks this type of data closely. Check out this post on Barron’s tech trader blog from January 3, 2008:
“Trim Tabs Investment Research asserted today that the U.S. economy added zero jobs in December, and that “a consumer recession has already begun.” Charlie Biderman, who runs Trim Tabs (and who once upon a time worked at Barron’s) says jobs data from the Bureau of Labor Statistics are inaccurate. He says that based on tax withholding data, “the U.S. economy added no jobs in December and an anemic 49,000 jobs in October and November combined.”
Trim Tabs says its own online job postings index fell 7.9% in December and 11.4% in the fourth quarter. Biderman says that consumer spendable cash – after-tax wage income, net equity extraction from real estate and tax refunds – will be down 4% in the first quarter versus a year ago. And he says a consumer recession “means that stock prices are likely to suffer more losses early this year.”