## Monday, May 07, 2012

### April Job Gains Could Range from 15k to 215k

As Dennis Gartman reported in The Gartman Letter last Thursday, here's how the BLS describes in a Technical Note how sampling error affects the monthly estimates of changes in nonfarm employment levels (emphasis added):

"Statistics based on the household and establishment surveys are subject to both sampling and nonsampling error. When a sample rather than the entire population is surveyed, there is a chance that the sample estimates may differ from the "true" population values they represent. The exact difference, or sampling error, varies depending on the particular sample selected, and this variability is measured by the standard error of the estimate. There is about a 90-percent chance, or level of confidence, that an estimate based on a sample will differ by no more than 1.6 standard errors from the "true" population value because of sampling error. BLS analyses are generally conducted at the 90-percent level of confidence.

For example, the confidence interval for the monthly change in total nonfarm employment from the establishment survey is on the order of plus or minus 100,000. Suppose the estimate of nonfarm employment increases by 50,000 from one month to the next. The 90-percent confidence interval on the monthly change would range from -50,000 to +150,000 (50,000 +/- 100,000). These figures do not mean that the sample results are off by these magnitudes, but rather that there is about a 90-percent chance that the "true" over-the-month change lies within this interval. Since this range includes values of less than zero, we could not say with confidence that nonfarm employment had, in fact, increased that month. If, however, the reported nonfarm employment rise was 250,000, then all of the values within the 90-percent confidence interval would be greater than zero. In this case, it is likely (at least a 90-percent chance) that nonfarm employment had, in fact, risen that month."

MP: Last Friday, the BLS estimated a payroll job gain of 115,000 in April, but the actual employment gain could be anywhere between 15,000 and 215,000.   Dennis Gartman questions why we should pay attention to these employment reports when they have such a huge margin of sampling error?

At 5/07/2012 9:16 AM,  AIG said...

This is known, and most people who pay attention to these numbers, know this. However, that doesn't mean that the estimate is "useless".

Although, I'm sure we'll hear how this is all due to "government propaganda", very soon ;)

At 5/07/2012 10:12 AM,  rjs said...

the BLS household survey is even worse; it comes from a sample of just 60,000 households...

but reports from census are the worst; for last months housing starts, for instance, the 90% confidence level was (±15.6%)

At 5/07/2012 11:40 AM,  Tom said...

Charles Biderman at TrimTabs market research points out that real-time payroll data is "readily available" from Treasury, which he compiles. All employers must report payroll counts and submit withholding taxes to Treasury.
He wonders why we accept survey data. It is an obsolete system.
http://trimtabs.com/blog/2012/05/02/bidermans-daily-edge-522012-a-reason-not-to-trust-fridays-bls-numbers/

At 5/07/2012 11:44 AM,  bart said...

... why we should pay attention to these employment reports when they have such a huge margin of sampling error?

Questions like that most frequently have only two possible answers - money or sex.

At 5/07/2012 12:20 PM,  juandos said...

Gallup seems optimistic regarding job creation: U.S. Job Creation Nears Four-Year High
May 2, 2012

'The April Job Creation Index of +20 is based on 36% of workers nationwide saying their employers are hiring workers and expanding the size of their workforce, and 16% saying their employers are letting workers go and reducing the size of their workforce'...

Yet as tom noted regarding to TrimTabs' Biderman: "The truth is that the initial BLS new jobs number is a joke even in the eyes of the BLS. How do I know that? I just reread some BLS footnotes for the first time in several years"...

At 5/07/2012 1:10 PM,  Ed R said...

Well since the low job gains for April was consistent with the similar low job gains for March, the 'confidence margin' for two months should be reduced to about 1/4 of the same margin for one month. (????)

My stats & probabability background is a little rusty; maybe someone better informed can help us out with this one.

At 5/07/2012 6:39 PM,  rjs said...

ed, all govt stats are revised at least twice, & then revised annually...both revisions on friday were positive; february was revised from 240,000 to a gain of 259,000, and March was revised from the originally reported 120,000 to 154,000..

At 5/08/2012 9:59 AM,  VangelV said...

MP: Last Friday, the BLS estimated a payroll job gain of 115,000 in April, but the actual employment gain could be anywhere between 15,000 and 215,000. Dennis Gartman questions why we should pay attention to these employment reports when they have such a huge margin of sampling error?

You shouldn't pay any attention to the government reported data. It overestimates job creation and serves a political purpose. If you count everyone who wants a job but can't find one we would be looking at an unemployment rate of more than 20%. And as I look around the situation seems to be getting worse, not better.

By next year you will see a major decline in shale drilling activity as the bubble that you failed to see starts to collapse. At that point most of the 'good news' will have been found to be self-serving hype and the recession that has not yet gone away will continue down its expected path. The Dow/Gold and Housing/Gold ratios are ready to resume their moves once again.