The monthly employment situation report is the most watched economic release. The attention has never been greater.
- Everyone is interested in the unemployment rate, especially when it is approaching 10%. It is a natural topic for the media.
- Partisan politics has a jobs focus. While the Obama Administration may be claiming jobs "created or saved," this is of little interest to most. The bottom line for most voters will be whether the net job change turns positive and the unemployment rate declines. Few care about hypothetical results.
- Investors and traders question the economic recovery, despite improving data. As long as unemployment remains high, people will question the economic recovery as artificial and temporary.
The massive attention is no surprise.
Lack of Timely Data
Most casual observers take the reports at face value. Those watching closely understand the challenges in trying to do contemporaneous estimates of employment changes. This is not like an election, where we can have a "polling place" to determine who is working and who is not.
We actually have excellent employment data, a hard count from state employment agencies. The problem is that it takes six to nine months to get the solid results. I wrote about this three years ago in my article, Data We Have vs Data We Need.
I am going to repeat a careful explanation that I suggested two months ago. There are many sources trying to measure monthly employment changes. Eventually the state data will show who was right.
Each of the sources we cite is attempting to measure the actual net job change. A wise stat prof once said, "Suppose God whispered into your ear and told you the TRUTH."
The BLS is attempting to do the same thing, with dramatically different methods. The BLS result is not TRUTH. It is a statistical estimate. Actual TRUTH for a specific month will not be known for many months, when the state employment data are analyzed. The BLS tries to count all of the jobs in one month, all of the jobs in the next month, and then report the difference. They do this very well, but it is inherently difficult. It does not focus directly on the actual changes, as other methods attempt to do.
Meanwhile, the forecasters will all be graded by how well they predicted the BLS number -- the BLS estimate of TRUTH.
That is the wrong attitude. The BLS number is just another estimate -- and one which will not be official until all of the revisions are in. Despite this, the market will trade on the preliminary estimate revealed Friday morning.
Briefly put, everyone is trying to estimate monthly changes in a work force of over 130 million. The error band is small. The BLS -- and all of the other sources -- are doing a great job with various differing approaches.
Market participants would like to have more data, faster data, more accurate data.
It is my long-held position that too much emphasis is placed upon the BLS estimate as "official." The BLS does excellent and honest work, but other approaches deserve attention.
Few observers understand the nature and sources of error in the BLS estimate. Pundits routinely draw major conclusions from discrepancies of 50,000 jobs, when the sampling error alone is +/- 100,000+ jobs. And this does not include assorted non-sampling problems, like dealing with new job creation.
The BLS versus the Critics
The BLS approach is to make an estimate of the total payroll jobs in one month, make another estimate for the next month, and subtract the two to determine the change. They use an excellent and sophisticated survey technique to do this. Their historical record, judged by the eventual count from the states, has been very good -- until quite recently.
Some readers have asked me why the BLS even attempts to include job creation in their estimates, using their much-maligned Birth/Death adjustment. I will attempt the simplest possible explanation.
Any time you do a survey, there will be non-respondents. When the question is something like "How many people favor health care with a public option?" the non-respondent problem takes a simple form. You need only ask whether the non-respondents are similar to those who actually answered. Most polls make this assumption.
The employment question is qualitatively different. We are not asking the opinions of non-respondents. We are asking whether they are even still in business. If the BLS were to assume that non-respondents had all ceased operations, they would seriously underestimate total employment. Historical data conclusively show that the non-respondents are split between those who did not answer and those who are out of business. The data also show that new job creation, running at about 2 million jobs per month even in recessions, are a predictable function of dying businesses.
Because of this, the BLS employs a two-step process. The imputation step forecasts job creation from job destruction, and includes a cyclical component.. The Birth/Death adjustment, (the only thing cited by most critics, who ignore the more important imputation step), is a residual. For many years this residual was stable. The most recent test against the state data indicated a significant error, showing that the BLS estimates have been wrong for nearly a year, especially starting in Q1 09.
Were the Critics Correct?
The BLS critics do not look at the ultimate scorecard, the state data. I expect that most of them will "claim victory" in their criticisms, even though they never looked at the many years when they were wrong and the BLS was right. We should note the following:
- There was no intentional distortion of data or results, dictated by whomever was President, as many allege.
- The method worked well for more than five years, as measured by the final data.
- The breakdown did not come at a "turning point" as many suggested it would. That would have happened at the end of 2007.
- The breakdown did not occur because the model worked in expansion and not in recession. The big divergence happened in the first quarter of this year. The big job losses right after the Lehman downfall were not that far from the truth. Job creation seemed to stall a few months after the major credit contraction in September, 2008.
The actual breakdown in the BLS method happened one full quarter after the biggest job losses in many years. It means that job creation was seriously impaired from January to March of this year. At the moment, this is all we really know from hard data.
There may be some critic who predicted the timing of this breakdown, but I have not seen one -- and I follow this carefully. Whether or not the long-term BLS relationships resume the multi-year pattern is yet to be determined.
To repeat my main theme --- there are several methods of estimating the monthly job change. No one has a monopoly on the truth. I invite reader comment on this point. Any critic who actually looked at the state data, acknowledged the BLS when it was accurate, and forecast when their model would break down deserves some recognition.
Each month we ask the question, "What change in payroll employment would be consistent with other economic data from the same time period (the middle of the prior month)?
This is not a forecast, per se, since we do not posit any causal relationship among these variables. They are all concomitant indicators of economic activity.
- We use the four-week moving average of initial unemployment claims, culminating in the week of the employment survey. This is the best direct indicator of new lob losses. This has improved in the last two months to a loss of 532K
- We look at the University of Michigan sentiment survey, which we found more useful than the Conference Board's sentiment index. Michigan uses a panel, where some families are carried over from month to month. This is a good technique. Sentiment is influenced by employment. When people have lost jobs, or are worried about losing jobs, it shows up in sentiment. It is a good concurrent indicator. The Michigan index is now at 70.6, about the same as last month.
- We us the ISM manufacturing index, which showed improvement from 52.6 to 55.7 This is solid expansion in the manufacturing sector, it is quite bullish for the overall economy. The ISM's research shows that this rating, if annualized, corresponds to a 4.5% increase in GDP, better than the preliminary report for the 3rd quarter, and occurring after the cash for clunkers and homeowner stimulus boosts.
Our long-term record has been pretty good, especially when compared to the final revised data. This makes sense because our model was derived from the final data. In recent months we have been too bearish. The BLS benchmark revisions suggest that we have been much better than first thought.
This Month's Prediction
Our indicators suggest a net job loss of about 140,000, a little better than the Street estimates of -175K.
Our estimate reflects two contradictory forces. The initial claims are still elevated, not indicating economic expansion. The ISM data are strongly positive. Our forecast suggests that we are still a few months away from real job growth. We need to see lower unemployment claims before there will be net job gains.Other Forecasts
It is always interesting to compare the job forecasts from different sources. We follow several because of the interesting and widely varying methods they use. A wise interpretation would be to consider all of these disparate sources of information.
ADP has proprietary data because of its payroll management business. ADP sees losses of 203K.
TrimTabs also uses real time data. Their estimates are based upon tax deposits for salaried employees. They see a net job loss of 284K.
WANTED Technologies, a relatively new entrant in this field, has a model based upon online help-wanted advertising. This is an innovative and different approach to real-time data. They see a net job loss of 224k.
While we are (for a change) less bearish than other forecasters this month, we do not see it as the basis for a change in our trading outlook. We expect the usual full assault on the integrity and methods of the BLS, with the spinmeisters out in full force. The short-term risk/reward still leans to the short side.