by Lee Adler, The Wall Street Examiner
The Labor Department reported today that seasonally adjusted (SA) first time claims for unemployment were up by 34,000 to 386,000. Unfortunately, both the total and the change are categorically and demonstrably false, just like last week. The seasonal adjustment process applied to weekly data is the problem.
As I reported last week (False Claims and Absurdities of Mainstream Media Reports On Initial Unemployment Claims, “To accurately reflect the actual year to year to year change of a drop 6.4% in new claims, the SA number should be 6.4% less than the SA number for this week last year. That would be 384,000.” On that basis, this week’s change would have been a “jump” of 2000. But even that would not be accurate.
On a not seasonally adjusted basis (NSA)– in other words, the actual count–the year to year change from the same week last year was a decline of 2.8%. Applying that to the SA claims number of 418,000 for the same week in 2011, the current SA number would be 406,000, not 386,000, significantly larger than reported. It’s all very confusing, largely a product of the large variance in the SA adjustment for the same week in any given year ( see last week’s post).
To avoid the confusion, I take the simple approach. Ignore the fictitious SA data and analyze the actual numbers of claims as counted. Extract the trend from the actual data (NSA) and compare the latest week’s actual performance to the trend, to last year, and to the average performance for the week over the prior 10 years. Looking at that graphically it’s easy to see whether the trend is accelerating, decelerating, or about the same. This week’s data suggests that the improving trend of the past two years is on the verge of decelerating. But it’s not definitive. We’ve been here before. Next week could determine whether there’s real slowing or whether the trend has just had a normal fluctuation with the parameters of past oscillations over the past couple of years.
In the advance report for the week ended July 14, actual claims (NSA) were 460,000 (rounded) including the addition of 4,000 claims to adjust for incomplete state counts that do not include interstate claims at the time of the advance release. This week was better than the week ended July 16, 2011 when new claims totaled 470,000. Total new claims were also better than the average of the last 10 years’ claims for this week of 468,000. Approximately 13,000 (2.8%) fewer people filed first time claims this year than in the same week in 2011.
Seasonal adjustment factors for this week have varied widely. Over the last 10 years, the factor applied to the actual NSA number for this week has ranged from .826 to .998. The range of variance was 20%, which is absurd. We’re talking about the same week every year! While the factor applied could have been anywhere from .826 to .998, this week was .852.
Looking back 10 years, the current week is a swing week, sometimes showing increases and sometimes showing declines in claims. The current week to week increase totaled 14,800 (rounded). That compares with a prior 10 year average of +7,000 for the week. Last year claims fell by 3,800, and in 2010 they fell by 8,700 during the same week.
The net change for the past two weeks was an increase of 87,000. That compares with the prior 10 year average of an increase of 84,000 for the same 2 week period. Last year, the 2 week increase was just 44,000. Therefore on the basis of the change over 1 and 2 weeks, this year shows some weakening. But is it significant, or just noise?
Overall, this week’s data shows that the trend is still improving (chart below), but that the rate of improvement is slowing versus last year. The slope of the year to year line for this week is declining at a slower rate than the slower 52 week moving average. Such variances are normal.
The annual rate of decrease in new claims continues to oscillate around the -10% axis. The latest data was down by just 2.8% which is the upper limit of the rate of change over the past 2 years. We have hit this level 6 times in the past two years, and things subsequently improved. If the difference shrinks to less than the current level that could be the first clear warning of a decline in employment ahead.
Anything between a decline of approximately 3% and 20% year to year suggests that the improving trend is on track. So far, the Fed has no reason for additional QE (more in depth analysis in the Professional Edition The Wall Street Examiner
Plotted on an inverse scale, the correlation of the trend of claims with the trend of stock prices over the longer term is strong, while allowing for wide intermediate term swings in stock prices. Both trends are driven by the Fed’s operations with Primary Dealers (covered weekly in the Professional Edition Fed Report; See also The Conomy Game, a free report.) This chart suggests that the as long as this trend in claims is intact, the S&P would be overbought at approximately 1425-50, and oversold at roughly 1200. The market is approaching the overbought parameter. I cover the technical side of the market in the Professional Edition Daily Market Updates.
As the number of workers eligible for unemployment compensation has trended up since 2009, the percentage of workers filing first time claims has continued to decline. Comparing the current week yearly line to the 52 week moving average, the trend of improvement has slowed although not quite as much as the raw totals.
The chart below gives a longer term perspective on claims. The trend has been improving while remaining above the bubble years with their 10 million fake jobs taking orders for new and unneeded condos and houses, building them, permitting and inspecting them, and taking and processing mortgage applications.
Lately, conomists have been arguing about the “natural” unemployment rate. I think we’re at it now. If we recognize that the bubble period with its millions of fake jobs was abnormal, then the low level of claims during those years was also abnormal. Where we are today is probably normal and the expectation that the US will ever get back to 6% unemployment is a false hope.
The problem the conomists have is that they think the bubble rates with their 10 million fake jobs were normal. That was abnormal. Today is normal.
The following chart is a picture of reality versus the the Impressionist art of seasonal adjustments. Sometimes the SA data represents reality to some degree, and sometimes it doesn’t. If you are following only that data, at any given time you have no way of knowing which it is. One thing is certain– it is not photo-realism.
There are ways to measure trends using actual data. One way is shown on this chart, which is to show the year to year line as of the current and corresponding date. Another is to view the annual rate of change as shown in the first chart above.
The arbitrary seasonal adjustment process has raised a couple of false alarms over the past 2 years with big counter trend pops early in the second half of 2010 and in the second quarter last year. But in the July 7 week it gave a false positive reading, which conomists furiously tried to explain away, when the explanation was mundane. The SA data was just wrong.
The SA data for the July 14 week partly compensated for the prior week’s misadventure, but not completely. The actual performance of the labor market was a little worse than the SA data depicted. It wasn’t a trend break yet, but it’s close. That makes next week’s data an inflection point that will either signal that, yes, the economy is getting worse, or that it isn’t. Based on the withholding tax data, I’ll have to go with the latter view for now.
Stay up to date with the machinations of the Fed, Treasury, Primary Dealers and foreign central banks in the US market, along with regular updates of the US housing market, in the Fed Report in the Professional Edition, Money Liquidity, and Real Estate Package. Try it risk free for 30 days. Don’t miss another day. Get the research and analysis you need to understand these critical forces. Be prepared. Stay ahead of the herd. Click this link and begin your risk free trial NOW!
GEI analysis articles on employment