Memo: To Dan Okrent, NYTimes Ombudsman
From Jude Wanniski
Re Interpreting statistics
Welcome back from your August vacation, Dan. I'm sure your desk is piled high with complaints from readers about every issue under the sun, though I'm afraid I have to add one more. This has to do with an August 28 article by Timesman Timothy Egan titled, "Economic Squeeze Plaguing Middle-Class Families." I hadn't noticed it when it appeared, but Bruce Bartlett, did and wrote his syndicated column about it, with a headline: The Times has played fast and loose. Now I don't disagree that in many ways middle-class living standards have declined over the last 30 years of inflations and deflations. It does often take two breadwinners per family to sustain an equivalent lifestyle. But Bruce, who was a tax official at the Treasury department in the last Reagan years and first Bush administration, is right in faulting Tim Egan's interpretation of available statistics.
I'm sure Egan and his editors believe his report was objective. The problem is that this story seems to surface in presidential election years, often on the front page of the Times, which gets its information from liberal think tanks who are putting their own spin on government statistics. When Paul Krugman was at M.I.T., he was the source of stories at the Times about the rich getting richer and the poor getting poorer, which I had to point out resulted from a conflation of "income" and "wealth," when they are quite different measures of "rich" and "poor." What happens when the Times is not careful in presenting such analysis is that the Democratic Party takes the numbers as gospel and builds policy commitments around them. Senator Kerry has been doing that, using the term "Middle-Class Squeeze" in his speeches. As Bruce puts it in his column: "...It is doubtful that John Kerry's campaign staff would have written it much differently if it had been handed the assignment."
Here is his analysis, which you might ask the editors to answer, or at least do a follow-up piece that corrects for the impressions left by the first article:
"The central point of the article is that recently released Census Bureau data show the middle class is disappearing. The key data are presented in a chart accompanying the article, titled, "A Shrinking Middle Class." This chart shows that the percentage of those households with incomes between $25,000 and $75,000 have fallen from 51.9 percent in 1980 to 44.9 percent in 2003.
"The clear implication is that the middle class has suffered under Republican policies -- why else start in 1980, the year Ronald Reagan was elected? If the chart had started in 1992, the year Bill Clinton was elected, it would have shown the exact same trend. In 1992, those earning between $25,000 and $75,000 constituted 47.9 percent of all households. By 2000, this fell to 46.1 percent. I don't remember the Times calling attention to this fact.
"The reason is quite simple: This is actually good news, not bad news, as the Times report strongly implies.
"First, it is important to know that the data in the Times story are adjusted for inflation. This is mentioned in a footnote to the chart, but nowhere else in the article. It might be useful to know that those with an income of $11,825 in 1980 now make $25,000, or that an income of $75,000 last year is the same as an income of $35,475 in 1980.
"In other words, the data take account of increased prices on everything from gasoline to college tuition. Yet the article implies that increased costs for these things has taken place without a concomitant increase in household income. The effect is to make middle class families appear worse off, when in fact most are far better off than they were in 1980.
"The most egregious error in the article is the clear implication that the percentage of those defined as the "middle class" has fallen because many of those who used to be considered middle class have become poor. This is totally untrue. In fact, the ranks of the poor have fallen along with those of the middle class.
"Using the Times' characterization of any household with an income below $25,000 in 2003 as being poor, what do the data show? We see that this group fell from 33.1 percent of the population in 1980 to 29 percent in 2002. Looking at the data from the other end, we see that the percentage of those making more than $75,000 has risen from 14.9 percent of the population in 1980 to 26.1 percent in 2003.
"In other words, the ranks of the poor and middle class have shrunk for one reason only -- more of them are rich! How can it not be a good thing for society that fewer people are now making low incomes and more are making high incomes?
"Just to show that the income gains have not been confined to those who were relatively well-to-do to begin with, there has also been an impressive increase in the percentage of black families with middle- and upper-class incomes.
"In 1980, 53.8 percent of black households made less that $25,000 (in 2003 dollars), which fell to 43.4 percent in 2003. The ranks of the black middle class ($25,000 to $75,000) increased from 40.5 percent to 42.9 percent. And the percentage of black families falling into the Times' definition of rich (over $75,000) rose from 5.8 percent to 13.7 percent.
"The Times cites Factcheck.org, a website sponsored by the Annenberg Public Policy Center, for its analysis. But I could find nothing on this site with the same figures. The closest thing I could find is an Aug. 3 report that actually disputes Kerry's claim that the middle class is withering away under Republican rule.
"In short, the Times has played fast and loose with the numbers in order to turn good news into bad news."
* * * * *
"Fast and loose" may be too strong for how the story was presented, but the analytics should be slowed down and tightened up to present a clearer picture of what's been happening among income classes. Welcome back, Ombudsman.