Raymond Richman       -       Jesse Richman       -       Howard Richman

 Richmans' Trade and Taxes Blog

Investors Are Entitled to Better Reporting of Unemployment Insurance Claims
Raymond Richman, 1/12/2012

While no one ever went broke underestimating the intelligence of the American voter, why should the media underestimate the intelligence of the American investor by reciting statistically manipulated data rather than the original data when the latter is readily available. Readers of this blog were made aware months ago that the figures issued by the US Bureau of Labor Statistics of the number of initial claims for unemployment insurance made during the preceding week include one doctored set of data called “seasonally adjusted” (SA)and another set of unadjusted data (NSA), i.e., the actual number of claims. In its notice to subscribers of the availability of the Report for the week ending January 7, 2012, the BLS  included this single paragraph:

"In the week ending January 7, the advance figure for seasonally adjusted initial claims was 399,000, an increase of 24,000 from the previous week's revised figure of 375,000. The 4-week moving average was 381,750, an increase of 7,750 from the previous week's revised average of 374,000."

Most reporters evidently did not go one to read the report itself. If they did they would have been amazed to find that the actual number of initial claims filed amounted to 642,381, 143,381 higher than the seasonally adjusted number. The difference is so great that it requires an explanation from the BLS. But none has been forthcoming.

If the Labor Department had no reason to mention the actual number, it would not be a serious matter. But the Labor Department's monthly report released Friday, January 7 showed employers added 200,000 jobs and the unemployment rate fell to 8.5% in December. This appears to be inconsistent with the large number of unemployment claims reported for the weeks ending December 31, 2011, and December 24, 2011, given the fact that the claims can be filed only after two weeks of unemployment.

Here is some of what the BLS included in its report:

SEASONALLY ADJUSTED DATA

In the week ending January 7, the advance figure for seasonally adjusted initial claims< was 399,000, an increase of 24,000 from the previous week's revised figure of 375,000. The 4-week moving average was 381,750, an increase of 7,750 from the previous week's revised average of 374,000.

And three paragraphs later:

UNADJUSTED DATA

The advance number of actual initial claims under state programs, unadjusted, totaled 642,381 in the week ending January 7, an increase of 102,314 from the previous week. There were 773,499 initial claims in the comparable week in 2011.

The advance unadjusted insured unemployment rate was 3.3 percent during the week ending December 31, an increase of 0.3 percentage point from the prior week's unrevised rate. The advance unadjusted number for persons claiming UI benefits in state programs totaled 4,139,988, an increase of 360,884 from the preceding week. A year earlier, the rate was 3.8 percent and the volume was 4,790,155.

The total number of people claiming benefits in all programs for the week ending December 24 was 7,333,213, an increase of 111,010 from the previous week.

As the reader can see, the actual number of claims filed was 642,381 and not 399,000, a week-to-week increase of 102,314, not 24,000. As an economist, I dislike the seasonally adjusted figure because the words “seasonally adjusted” tend to be omitted. CNBC, Bloomberg, the Wall Street Journal, the Investor’s Business Daily, the New York Times,  Associated Press and all of the media, printed, or televised (e.g., Fox News) omit the qualifying words. Not one reporter, to my knowledge, reported the actual number of claims. Indeed all those that I read, except Reuters failed to mention that the number quoted was “seasonally adjusted”.

The following table shows the initial claims for unemployment insurance, seasonally adjusted and actual (NSA) as reported by the BLS for the weeks ending January 7, 2012, December 31,2011, and December 24, 2011:

UNEMPLOYMENT INSURANCE DATA FOR REGULAR STATE PROGRAMS

Good reporting requires that the actual data, not seasonally adjusted (NSA), be reported. The NSA data should appear before any statistically manipulated data (SA) is reported. I (a one-time productivity division employee of the BLS) emailed Labor Secretary Hilda Solis, making that suggestion. I also notified the WSJ, CNBC, and others months ago to at least report the NSA data. While most acknowledged receipt, none took any action. Investors are surely entitled to better reporting of initial claims for unemployment insurance.