Apparatchik Press: AP Vastly Exaggerates Modest to Non-Existent Improvements in Unemployment Claims, Housing Data

Thursday morning, initial weekly unemployment claims as reported by Uncle Sam's Department of Labor came in at a seasonally adjusted 414,000. It was 16,000 lower than the previous week's upwardly revised (as usual) number, but certainly no indicator in and of itself of meaningful improvement.

The housing industry data really wasn't any better. True, the seasonally adjusted figures from the Census Bureau for building permits issued and housing units started were somewhat improved, but the raw data still had several examples of record weakness.

Wait until you see the headline the Associated Press applied to a story covering the DOL and Census reports by Derek Kravitz and Christopher Rugaber:

Economy gets lift: Layoffs ease, home building up

The AP pair's actual report was a bit more measured, but still pretended there was a sunny side:

Fewer Americans applied for unemployment benefits last week and builders broke ground on more homes in May. The latest data offered some hope that the economy may be improving after hitting a slump in late spring.


Unemployment benefit applications fell to a seasonally adjusted 414,000, the Labor Department said. It was the second drop in three weeks and a positive sign that layoffs are slowing.


Still, applications have been above 400,000 for 10 straight weeks, evidence that the job market is weak compared to earlier this year.


Home construction rose last month to a seasonally adjusted annual rate of 560,000 units per year, the Commerce Department said. Economists say the pace of construction is far below the 1.2 million homes per year that must be built to sustain a healthy housing market. Many credit-strapped builders are struggling to compete with low-priced foreclosures.


The modest improvements in two of the economy's most troubled areas were enough to give Wall Street a lift after a major sell-off the previous day.

The item's time stamp is 4:52 p.m., after the market's Thursday close, making the final excerpted sentence from Kravitz and Rugaber just plain wrong. In the real world, "Wall Street" hardly got a "lift," guys:

The Dow Jones Industrial Average rose 64.25 points, or 0.54%, to 11961.52, recovering part of a 179-point tumble Wednesday to a three-month low. The Standard & Poor's 500 stock index gained 2.22 points, or 0.18%, to 1267.64. The measure's defensive utilities and consumer staples components advanced, but the materials sector fell sharply. The Nasdaq Composite slid 7.76 points, or 0.29%, to 2623.70, pushing the technology-heavy index deeper into the red for 2011.

The Dow, the least representative index, went up a bit, but the other two combined ended up as about a wash. Further proof: The NYSE Composite Index lost 4.21 points (-0.05%), and the Russell 3000 gained 0.94 points (+0.12%). Lift, shmift; the markets were basically flat as a pancake yesterday.

"Lift shmift" also applies to the data which drove the markets to their non-existent good day.

The reporters described initial unemployment claims in an unexcerpted sentence as being in a "steady decline." Horse manure. Two weeks in a row does not a "steady decline" make. Moreover, yesterday's 414,000 claims will, based on DOL's track record, almost certainly be revised up to 417,000 or so next week. If/when that happens, it will be higher than the week of April 9, when claims began their current 10-week string of 400,000-plus results after six of the seven previous weeks (mid-February through early April) were below 400,000.

As to the housing data, it's hard to get excited when:

  • Actual (i.e., not seasonally adjusted) housing starts trailed May 2010 by a tiny bit, and single-family starts came in 6% lower.
  • Actual single-family building permits of 39,000 were lower than any previous May going back to 1959.
  • Actual completions were lower than any previous May going back to 1968, both overall and for single-family units.

This appears to be another example of how, as has sometimes been the case with employment data, seasonal adjustments of data in volatile and atypical situations can mislead. Shame on AP and the rest of the establishment press for never, ever digging deeper, and in AP's case for pretending that yesterday's unemployment claims and housing data represented anything about which to get excited.

Cross-posted at

Tom Blumer
Tom Blumer
Tom Blumer is a contributing editor for NewsBusters.