The Trouble With Numbers: Complaints on a Much Anticipated Jobs Report

September 7, 2012, 4:18 pm

By MARGARET SULLIVAN

There's a drumroll beforehand and a great crashing of cymbals as it happens. It's the federal government's monthly report on the nation's employment â€" a matter that's given huge political weight in the current presidential campaign, as Jim Rutenberg's news analysis on today's front page makes clear. The Wall Street Journal and many other news organizations also gave it front-page treatment, before the fact.

It's also a matter that is open to all kinds of interpretation.

The Times gave the report major display on its home page this morning in a story by Nelson D. Schwartz, beginning with a headline that emphasized both the positive and the negative: Hiring Slowed in August; Jobless Rate Fell t o 8.1%. Later, the political element was added in a revised headline: Hiring Slows in August, Adding to Pressure on Fed and Obama.

With the importance of jobs to Americans, many of whom have been out of work for long months and years, it's no wonder this report gets so much weight. But some experts, and some readers, think it gets too much emphasis, or the wrong kind.

Betsey Stevenson, former chief economist at the Department of Labor and now an academic economist at the University of Michigan, put it this way when I asked her about it this morning, having seen one of her Twitter messages on the subject.

The media tends to overreact and overinterpret the employment number. For instance, the (Times's) first line said, “Job growth slowed substantially.” Given that it isn't a statistically significant change, it's pretty hyperbolic to say “substantially.” If the number is revised up 40K when revisions are release d next month, will The Times lead with the statement “August job growth was revised up substantially”? I don't think so.

I'm not trying to downplay today's disappointing number, but we are measuring change on a base of 133 million jobs and telling the difference between a change of 96K versus 141K is a small change which may reflect measurement error or a slowdown. Since we can't tell the difference yet, calling it a substantial slowdown is wrong.

I asked Deputy Business Editor Winnie O'Kelley to respond. One can quibble with the wording, she said, “but we try to get the tone and tenor right, and I think we did.” There is little question, she said, “that the picture is disappointing.” She noted, too, that later versions of the story on the Web, and in Saturday's paper, would delve into the numbers more deeply.

But she rejects the notion that the numbers are overemphasized. “This was the most important jobs report of the year â€" n ot just politically, but because the Fed meets next week and will make decisions based on it about stimulating the economy,” she said.

For the average American, Ms. O'Kelley says, “it's an important barometer.”

A Times reader, Alex Dering, has strong feelings on the subject, for different reasons.

In an e-mail today, he wrote to me about what he considers an important missing element:

Although Mr. Schwartz's article makes very clear that the drop in unemployment is simply a mathematical sleight-of-hand as more people simply cross into the gray zone of no longer counting as a statistic, The New York Times still resists even a hint of a whisper of a murmur of the U-6 statistic, which puts the number of people who are not employed or who are underemployed - underemployed is a technical term meaning “starving to death more slowly than the unemployed” - at 14.6%: One in seven.

Unemployment - in the genuine, I-can't-pay-the-rent sense of the term - is very high. Media complacency (or complicity) in selectively presenting only some of the facts allows politicians to either ignore the issue outright or manipulate it to divisive ends as an endless parade of stats are presented to shore up both, contradictory, positions on the issue.

Why won't The Times clearly state the U-6 figure on a regular basis? This isn't a partisan issue; millions of people of all political stripes are going to the poorhouse, and The Times stands there like Officer Barbrady - “Move along, nothing to see here. Move along.” Unless someone with the authority and clout to do so starts shaming our elected officials into addressing the issue, it will simply go around and around while what's left of the middle class disappears.

Ms. O'Kelley responded that she and her staff were well aware of the U-6 number, and that “we do look at it,” whenever The Times is reporting on unemployment.

However, “c ontinuity over time is important. We use the official number of unemployed as the benchmark.”

It's a number people understand as a basis for comparison, she said. “Generally speaking, the trends are similar.”

In the case of both of these issues, context is important. We're likely to see more of that â€" and the more the better - in the in-depth treatment of the Friday's jobs report in Saturday's paper.