Sorry!! The article you are trying to read is not available now.
Thank you very much;
you're only a step away from
downloading your reports.

Don't Trust Those Seasonally Adjusted Jobless Numbers


Knowing what's happening with employment in real time seems to be of little interest to economists and reporters. To investors, however, getting accurate data is pretty important.

With the mainstream media reporting the seasonally adjusted first-time unemployment claims down by 6,000, it's time for a reminder that this number is fake, and may or may not give an accurate representation of the trend. If it does, it's purely a coincidence. Economists and financial journos love the seasonally manipulated number because it gives them a smooth curve over time. But watching the jagged edge of reality is what gives us an edge in seeing what's really going on over time. When the trend changes, we'll know it a couple of months before those who are watching only the seasonally fictionalized version. Knowing what's going on as it's actually happening seems to be of little interest to economists and reporters, but to us as investors, it's a bit more important.

The problem with any form of statistical smoothing is that it obscures the actual situation in the present, and it only reflects real trend change well after the fact. It is particularly absurd to do this with the weekly claims data because the government does provide us with the actual counts, which it collects from the 50 state bureaus with a lag of just one week. Since it is based on actual tabulations rather than tiny survey samples, the data is somewhat more likely to be reasonably representative of reality than the numbers the government builds from a sample of one tenth of one percent of the population.

This weekly tabulation is as opposed to the Bureau of Labor Statistics' (or BLS) monthly sampling of total employment, which relies on extremely small samples and statistical manipulation such as the business birth-death adjustment, even before applying seasonal factors, which weaken the reliability of the reported numbers. The final numbers are based on assumptions derived from past history. When the underlying conditions undergo real change, these assumptions are often a year or more behind that change. Adding seasonal adjustment factors on top of that can delay recognition of trend change by several months.

The actual initial claims data suffers from no such manipulation, yet nobody in the mainstream media reports the actual data. Economists don't like it because the jaggedness of the data makes their heads explode. Market technicians don't suffer from that malady. They are accustomed to looking for patterns in data, and changes to those patterns which give early indications of trend change.

The actual weekly initial claims data exhibits week to week patterns each year that are consistent, as certain industries tend to add and subtract workers at the same time each year. Rather than smoothing the data to obscure what really happened last week, we can compare the numbers directly with prior years' performance during the same weeks to get an accurate reading of the current trend. Like an optometrist, we can look at small changes and ask whether they are better, worse, or about the same as last year. By carefully evaluating subtle changes, we gain clarity of vision.

The first thing to notice about this chart is the consistency of the yearly pattern. Every year the low in claims is just after Labor Day and the high is just after New Year's Day. Zooming in to October, there's always a zigzag between the first and second weeks of the month -- up the first week, down the second. So it's no surprise that this year in the week ended October 15 first time claims fell by 47,229 after a gain of 71,660. The two-week net is an increase of 24,431.

The question is, how does that performance stack up with last year and prior years. Last year was a "recovery" year. The two years before that were recession years. Last year claims fell by 71,303 in the comparable week. Suddenly, this year doesn't look so hot. In 2009, first-time claims were down by 49,281 in the like week. Uh oh. This year didn't even beat that. However, it did beat 2008's drop of 37,954, but not by much. This year is barely better than the worst year of the "recession."

The two-week performance, including the prior week where claims jumped, shows an increase of 24,431 this year versus 18,056 last year, 8,566 in 2009, and a decline of 10,678 in 2008. By that standard, this year is by far worse than the last 3 years.

All of these numbers suggest that employment momentum has deteriorated. While we see first-time claims falling year over year by 8.9% from 391,737 to 356,825, in the most recent week the rate of improvement slowed. However, this change is not yet material. Claims have been falling on a year to year basis by between 3% and 15% for the past 2 years. That hasn't yet changed.

While the number of new claims is diminishing at a fairly constant rate, they remain high by the standards of the bubble years of 2005-2007, when millions of fake jobs were constantly being created. However, the current number also remains somewhat above the recession years of 2002 and 2003, when claims totaled 350k and 328k respectively, and they are much higher than the recovery year of 2004, when claims were 280k in the mid-October week.

This is only half the picture, however. It tells us only how many people are losing jobs each week. It doesn't tell us how many are finding work. For that we have to wait from the less than reliable monthly data from the BLS.

What we know from the weekly data is that over 350,000 people lost jobs in the week ended October 15. To put that in perspective, the total for the past year is 21,691,397. That's a lot less than 2009's 30 million, and better than 2010's 24.2 million, but it's still worse than October 2007-2008's 19.6 million when the economy first collapsed. The rate of job loss will need to slow a lot more before the economy can truly be said to be recovering. The risk here is that this could easily tilt the other way.

By tabulating and tracking the actual data instead of the seasonally smoothed fictional data, we should get an early heads-up on which way the trend is tilting. And unlike economists and mainstream journalists, our brains won't explode.

Editor's Note: This article was originally published on Wall Street Examiner.


No positions in stocks mentioned.

The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.

Copyright 2011 Minyanville Media, Inc. All Rights Reserved.

Featured Videos