May Employment Preview

We rely too much on the monthly employment outlook report.  It is a natural mistake.  We all want to know whether the economy is improving and, if so, by how much. Employment is the key metric since it is fundamental for consumption, corporate profits, tax revenues, deficit reduction, and financial markets.

Since the subject is so important, most people place too much emphasis on the official results, losing sight of the important elements of context.  I will first summarize the BLS official methodology.  Next I will review alternative approaches and those forecasts.  I will conclude with some ideas about what to watch for.

The Data

We would like to know the net addition of jobs in the month of May.

To provide an estimate of monthly job changes the BLS has a complex methodology that includes the following steps:

1. An initial report of a survey of establishments. Even if the survey sample was perfect (and we all know that it is not) and the response rate was 100% (which it is not) the sampling error alone for a 90% confidence interval is +/- 100K jobs.
2. The report is revised to reflect additional responses over the next two months.
3. There is an adjustment to account for job creation -- much maligned and misunderstood by nearly everyone.
4. The final data are benchmarked against the state employment data every year. This usually shows that the overall process was very good, but it led to major downward adjustments at the time of the recession. More recently, the BLS estimates have been too low. (See here for a more detailed account of this, along with supporting data).

Competing Estimates

The BLS report is really an initial estimate, not the ultimate answer. What we are all looking for is information about job growth. There are several competing sources using different methods and with different answers.

• ADP has actual, real-time data from firms that use their services. The firms are not completely representative of the entire universe, but it is a different and interesting source. ADP reports gains of only 133K private jobs. Steven Hansen at Global Economic Intersection endorses the ADP method over the BLS result. He has a strong analysis covering many nuances in the data. For those who really want to understand the jobs story, it is well worth reading.

• TrimTabs looks at income tax withholding data. The idea is that this is the best current method for determining real job growth. TrimTabs forecasts gains of about 124,000. They have been pretty bearish on job growth, but their approach is worth serious consideration.

• Economic correlations. Most Wall Street economists use a method that employs data from various inputs, sometimes including ADP (which I think is cheating -- you should make an independent estimate). I use the four-week moving average of initial claims, the ISM manufacturing index, and the University of Michigan sentiment index. I do this to embrace both job creation (running at over 2.3 million jobs per month) and job destruction (running at about 2.1 million jobs per month). In mid-2011 the sentiment index started reflecting gas prices and the debt ceiling debate rather than broader concerns. When you know there is a problem with an input variable, you need to review the model. For the moment, the Jeff model is on the sidelines.  This month would be impossible for my model anyway, since the ISM data will not be released until after the employment report.
• Briefing.com cites the consensus estimate as 150K and their own forecast is for 175K.
• Whisper numbers were reported to be revised lower today after the ADP report, with some predicting only a gain of 75,000 jobs.  Art Cashin reported that traders believe there is a 20% chance that the report will show fewer than 100,000 jobs gained.  If traders understood sampling error, they would realize that this statement would be true even if the actual job gain was 150,000.

Failures of Understanding

There is a list of repeated monthly mistakes by the assembled jobs punditry:

• Focus on net job creation.  This is the most important.  The big story is the teeming stew of job gains and losses.  It is never mentioned on employment Friday.  The US economy creates over 7 million jobs every quarter.
• Failure to recognize sampling error.  The payroll number has a confidence interval of +/- 105K jobs.  The household survey is +/- 450K jobs.  We take small deviations from expectations too seriously -- far too seriously.
• False emphasis on "the internals."  Pundits pontificate on various sub-categories of the report, assuming laser-like accuracy.  In fact, the sampling error (not to mention revisions and non-sampling error) in these categories is huge.
• Negative spin on the BLS methods.  There is a routine monthly question about how many payroll jobs were added by the BLS birth/death adjustment.  This is a propaganda war that seems to have ended years ago with a huge bearish spin.  For anyone who really wants to know, the BLS methods have been under-estimating new job creation.  This was demonstrated in the latest benchmark revisions, which added more jobs, as well as the most recent report from state employment offices.

Trading Implications

My experience with employment Fridays is that there is little benefit to being aggressively long before the report.  The spinfest usually provides shorts with a morning "dip to cover" when the number is surprisingly good.

The recent pattern has been a weaker-than-expected initial report, followed by upward revisions.  Check out last month's preview, where I covered this in more detail, including some great analysis from Scott Murray.

Adding to tomorrow's volatility we will have the ISM manufacturing report as well as data on personal income and spending.

Comments

[Octavio Richetta]

IMHO, the TREND not just in the US job data, but in the WW economic data is very concerning and should not be taken lightly.

Intelligent assessment of the economic situation and markets requires more than just looking at historical data, on which, even forward looking indicators, are based.

Fortunately/unfortunately forecasting/assessing the economic future is more an art than a science; thus cannot be 100% objective/rely 100% on data but also on assessing how these variables will interact in the future. Miracles do not happen.

So far I have been able to skip the US Tech bubble and the US housing bubble while also beating the S&P500 in the last 15 years. And now I am trying my best not to get run over by the aftermath of the 2008 crisis.

As I have posted at this site previously, the effect of the European Sovereign Debt Crisis, which I label the Euro Bubble, and the China hard-landing, on the fragile US economy should not be taken lightly.

[wkevinw]

I did a study of revisions of the GDP and the jobs data a couple of years ago. (Can't recall where I put the data). Anyway, in terms of being reliable enough to use as trading signals, three words: they are not.

The revisions are enormous. One example that is interesting is what happened in 2000-2001. I believe the first GDP estimate showed three consecutive negative quarters, and the revision showed the middle quarter to be positive. So, by the two negative consecutive quarters recession criterion, there wasn't one: after revision.

Thanks.

[Charlie]

Hi Jeff,

Always look forward to your posts on US employment.

I also like ADP's approach. However, they did change their methodology in 2008 (http://adpemploymentreport.com/PDF/REVISED_METHODOLOGY_DEC_2008.pdf) to include laggged NFP so I guess it is not as much cheating any longer in using ADP to forecast NFP since that seems to be what the ADP themselves are trying to do.

From the pdf:
"The BLS data (as currently reported) for growth of employment by industry is regressed on: (a) the matched-sample growth rates by industry based on the ADP data; (b) a weighted average of the historical average growth rates of employment in each cell based on QCEW data; (c) a weighted average of the historical average growth rates of employment in each cell based on the ADP data; (d) lagged values of BLS estimates of growth of employment by industry; (e) initial unemployment claims filed during the week immediately before the week that includes the 12th of the month."

Regards,
Charlie