Wednesday, February 10, 2010

Trim Tabs and Withholding Taxes

- by New Deal democrat

Well, {sigh} Trim Tabs is at it again, and the usual suspects are biting . They're touting the following "analysis":
Real-Time tax withholding data shows that wages and salaries declined an adjusted 1.0% y-o-y. In January 2009, wages and salaries declined 5.0%. If the labor market were improving, we would expect a positive year-over-year growth rate. The fact that tax withholding data is still declining year-over-year suggests that the labor market is still contracting.
(my emphasis). No, it doesn't. How many times do I have to point out that simple YoY comparisons miss turning points?!? The fact that YoY data is still declining means ... we're not even with a year ago. D'uh!

YoY data of course is valuable, particularly where there are large seasonal fluctuations, as there are in items like housing. Indeed there is definite seasonality in tax withholding -- generally the period of December through March shows more payments than the remainder of the year.

While it would take averaging out the last 5 or 10 years to get a more exact picture, an easy way to look at the data is to see how the YoY comparison is changing. Are the YoY comparisons getting more amplified or smaller? If more amplified, the trend is accelerating; if smaller, the turning point may already have arrived.

With that in mind, let's look at tax withholding data since January 2008.

Year/monthWithholding ($Blns)YoY% change
Jan 2008 163.4 -4.1
Feb. 2008 156.1 +6.6
Mar. 2008 162.6 +3.6
Apr. 2008 146.9 +3.2
May 2008 140.8 0.0
Jun. 2008 145.5 +0.9
Jul 2008 143.0 +6.0
Aug. 2008 136.5 +2.5
Sep. 2008 142.8 +10.4
Oct. 2008 142.5 +0.4
Nov. 2008 135.6 -0.4
Dec. 2008 167.6 -3.6
Jan. 2009 151.8 -7.1
Feb. 2009 142.9 -8.5
Mar. 2009 156.0 -4.1
Apr 2009 133.2 -9.9
May 2009 126.0 -10.5
Jun. 2009 133.8 -9.2
Jul. 2009 131.4 -8.0
Aug. 2009 126.4 -8.0
Sep. 2009 125.2 -12.3
Oct. 2009 124.7 -12.5
Nov. 2009 127.7 -5.8
Dec. 2009 154.9 -7.6
Jan. 2010 140.4 -6.7

First of all, you can easily see the seasonal pattern. Tax collections are generally strong from December through March and generally weak the rest of the year.

Secondly, note how the YoY% change remained positive until almost a year into the recession. If you were simply citing a YoY change without noticing the underlying trend, you would have been completely wrong. Notice also that withholding tax collections held up for several months after payrolls started to be cut with wild abandon in about August-September 2008. Simply put, withholding taxes lag employment data.

Finally, turn your attention to recent months. You can see that the maximum YoY percentage decline was in October, which also happens to be the lowest raw number. Between May and October, each and every month was 8% or more lower than the year before. All 3 months since November have come in lower than that comparison.

In other words, while we can't say for sure, it looks like October was the actual, seasonally adjusted bottom in payroll tax withholding.

Unsing the same methodology, Trim Tabs said the recession was already over in April 2008, something Barry Ritholtz called "one of the weakest, most poorly reasoned and mathematically challenged analyses I have read in two decades on Wall Street."

Trim Tabs and their gullible sycophants are wrong again.