August 26, 2007

Top Six Errors Committed by David Cay Johnston and/or the New York Times in Their Income Growth Report

NOTE: I was tempted to open up real-time commenting at this post given the interest in it. But Mr. Johnston reports that he has received threats — “quoting your work ….. urging death to me and my family.”

Of course I condemn those threats unconditionally, and I encourage David to pursue those who have made the threats, and to contact appropriate authorities. I will co-operate in any way I can.

Yes, these posts have been contentious, as is this one. But how anyone could get so bent about posts relating to the accuracy and propriety of reporting on economic and tax data boggles the mind.

Therefore, I will not take a chance on anything remotely similar to what Mr. Johnston reports getting through. And people, don’t even THINK about writing a comment accusing Mr. Johnston of making the threats up.

_______________________________________

Okay, I’ve done my “I was wrong, I am sorry” thing over a net error of $38, as noted yesterday, and as documented in the original post.

Today David Cay Johnston and the New York Times owe their readers a number of “I was/we were wrong. I am/we are sorry” statements. I only have time to get to six.

And here’s the deal, David — Don’t waste your time commenting here until you refute all six of them.

Every. Single. One. (Update. 1:15 p.m. — You are welcome to cherry-pick via e-mail, if you must.)

You’re the one who has been baiting me about how I (from your e-mail last night) “used phony numbers and statements to attack my work and my honor, but have not shown a single error in (your) report.” Okay, fine. I’m about to show six. Given your statement, I don’t see why you’re entitled to come back with a response until you think you’ve knocked the number down to zero. I frankly don’t think it’s possible. Oh, and I have more errors and offsets in reserve if they’re ever needed.

So here are the Top Six.

6. The Times had to make a correction to the article as it originally appeared in one of its dead-tree editions.

Johnston’s article now has this correction at the end of it:

An article in Business Day yesterday about the average income of Americans from 2000 to 2005 misstated, in some copies, the nation’s total adjusted gross income in 2005. It was $7.43 trillion, not billion.

If a Times reporter is going to beat up on me over an error of $170 per person netting down to $38, it doesn’t seem unreasonable to call him (perhaps) and his paper (for sure) on failing to pick up a thousandfold error. Wouldn’t it be interesting to be a fly on the wall at the Times to see what Johnston’s original submission contained? It would also be useful to know whether the correction dated August 22 was caught before or after this post by James Pethokoukis at US News dated August 21 at 2:12 p.m. went up. Mr. P saw the billion-trillion error in the Times’s sister paper, the Boston Globe.

5. Johnston states that “Total adjusted gross income in 2005 was $7.43 trillion….” It was $7.42 trillion, as seen here (The $55,238 in the graphic is the 2005 “average income” figure Johnston reports earlier in the article; the IRS Excel document is at 2005 Table 1.1, which can be found at this IRS page; I changed the column widths so they would fit in the space available, but made no other changes):

IRS2005AGIlessDeficit

Paraphrasing Item 6, if someone’s going to quibble over a net $38 per person, I don’t think I need to be shy about noting a $10 billion bust due to improperly rounding down instead of up (Johnston rounded up by $7,504,336,000 instead of rounding down by $2,495,664,000).

4. That same sentence, “Total adjusted gross income in 2005 was $7.43 trillion….” incorrectly uses the term “adjusted gross income” (AGI), when, as seen in the graphic above, that amount is “Adjusted Gross Income Less Deficit.”

There IS a difference between the two:

  • AGI is commonly understood as the number at the bottom of page 1 of the long-form 1040. It’s specifically identifed by name on the shorter forms as well. It’s income before either the standard deduction or itemized deductions.
  • “AGI Less Deficit” is a very stangely named IRS term that adds some things back back AGI.

Without doing further study than I have anything resembling time for, I’ll just have to say that the two things aren’t the same, that there’s no indication in Johnston’s article that AGI Less Deficit is used, and that the reader has no reason to believe that what’s involved is anything other than AGI as commonly understood. I probably won’t dig into this point any more, because what I have seen indicates me that using AGI, or AGI Less Deficit, for the purpose Johnston employed has so many flaws that what results from it has little, if any, meaning. Sort of garbage in, garbage out, if you will.

But you work with the garbage you’re dealt, so let’s get to number 3.

3. Johnston’s opening statement (“Americans earned a smaller average income in 2005 than in 2000, the fifth consecutive year that they had to make ends meet with less money than at the peak of the last economic expansion….”) is a clear reference to something approximating “after-tax” or “disposable” income, and clearly NOT to AGI, or AGI Less Deficit, as discussed in the previous item.

2. Subtracting Federal Income Tax and Alternative Minimum Tax paid from “AGI Less Deficit” shows that the average tax return filer had real “disposable income” income gains from 2000 to 2005.

Here it is:

IRS2000and2005DispInc

The data came from IRS Historical Table 1, a spreadsheet you can download by going to this IRS page and clicking on the “1″ link near the bottom of that page. The bolded numbers near the top of the chart, “AGI Less Deficit Per Return,” are what Johnston and the Times used.

The term “disposable income” is in quotes, because true “disposable income” would also subtract out Social Security and state/local income taxes. But I don’t see how those taxes, which are relatively consistent year to year, could have increased by enough in real terms during the intervening five years to make up for the difference shown.

And as one example of holding something in reserve, I also did not include the impact of the Earned Income Tax Credit in any of the years presented above.

1. So the Number 1 error in Johnston’s and the New York Times’s report IS ….. the thrust of Johnston’s statement, namely that 2005 was “the fifth consecutive year that (Americans) had to make ends meet with less money than at the peak of the last economic expansion….”

In fact, thanks to the various reductions in federal income taxes since, 2005 was actually the first year since 2000 in which Americans had MORE.

__________________________________________________

UPDATE, August 25, 2008: Research I did in connection with this August 16, 2008 BizzyBlog post indicates that 2004 — not 2005, as indicated in the final sentence above, was the first year since 2000 in which Americans had more. The 2005 conclusion at this post was based on attempting to adjust Johnston’s work. The 2004 conclusion in the August 16 post is based on the use of more consistent and complete data from the government’s Bureau of Economic  Analysis.

Share

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.