August Employment Report Preview

This Friday we have something a bit unusual — an employment situation report right before Labor Day.  Since everyone already views this as the most important piece of economic data, can the attention get any greater?

You bet!  Just watch this Friday and the rest of the weekend, including the Sunday morning talk shows.  We are in the political season where political partisans of all stripes want to make a point.  The "market partisans" are also aligned, with double-dippers, austerians, and happy campers all ready to make a point.

Today's strong ISM report is an encouraging economic sign.  Many will be looking to the jobs numbers for confirmation that all is well with the economy.  I expect disappointment.

 Let me start with some background on my approach.

Our Own Estimate

The non-farm payroll report is based upon a monthly survey, attempting to estimate the total of all payroll jobs for the week including the 12th of the month.  Each month my team asks the question, "What change in payroll employment would be consistent with other economic data from the same time period (the week including the 12th of the prior month)?

The answer to this question is not a forecast, per se, since we do not posit any causal relationship among these variables.  They are all concurrent 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 job losses.  This has increased sharply  — 484K versus 457K last month.  There is some discussion about seasonal adjustments, auto layoffs, and other factors.  I am always concerned when I have less confidence in an important variable.
  • We look at the University of Michigan sentiment survey, which we find to be 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 strongly influenced by employment.  When people have lost jobs, or know others who have, they get worried.  It is a very good concurrent indicator.  The Michigan index firmed up a bit last month (after a sharp decline in July) to 68.9, up from 67.8 last month.
  • We use the ISM manufacturing index.  This is 56.3, an uptick from 55.5 last month. This is the most bullish of the various indicators.  Most people do not know much about interpreting this index.  Here is what the ISM has to say about the implications for economic growth:

The past relationship between the PMI and the overall economy indicates that the average PMI for January through August (57.8 percent) corresponds to a 5.3 percent increase in real gross domestic product (GDP). In addition, if the PMI for August (56.3 percent) is annualized, it corresponds to a 4.8 percent increase in real GDP annually.

Our long-term preview record has been very good, especially when compared to the final revised data.  This makes sense because our model was derived from the final data.  Our approach also makes logical sense, because it involves some factors related to jobs lost, and some related to job creation.

How have we done?

The most recent report from actual data, covering the last quarter of 2009, shows total private job losses of 193,000.  Government job losses added another 20K or so.  The BLS initial estimated during this time frame showed losses of 286K and my estimates were 377K.  This is the only valid way to keep score — looking at the actual final data, nine months later.

I wish that others were so honest, especially those who make the mindless monthly attacks on the BLS and their methodology.  They did pretty well in Q409, beating our approach and doing much better than many skeptics.

Turning back to the current month, our estimate is for a loss of over 90K jobs.

Last month I noted that the preliminary reports have been running "hot" by about 100K jobs, a problem I discussed in this article.  As I showed above, this was not true in Q409.  You must also keep in mind that the BLS estimate has a 90% confidence interval (just for sampling error) of+/- 100K jobs.  There is a wide range of possibility for Friday's report, and  (for a change) I have no strong opinion.

Other Forecasts

It is always interesting to compare the job forecasts from different sources.  We follow several because of the 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.  Looking only at private sector jobs — no government, no census effect, ADP sees a loss of 10,000 jobs private sector jobs.

TrimTabs has another valuable approach — tax deposits.  Their forecast is a loss of 65,000 jobs. cites the consensus as a loss of 120K and their own forecast is a loss of 106K.

To summarize briefly, the market incorrectly focuses on predicting the BLS preliminary estimate — mostly ignoring the 100K confidence interval, the seasonal adjustments, and the benchmark revisions.  I am probably the strongest supporter of BLS methodology and integrity, but I still see their approach as only one method out of many.

Jobs data is so important — and we are all so interested — that we seize upon whatever information we have, even when we should accept the limitations.

Investment Take

I am not specifically trading around the number this month.  Given all of the other economic data, it is hard to believe that we will have brisk job growth.  If there is a negative print, it will make for a long weekend of explaining for President Obama's team.  They over-promised on the effects of the stimulus and over-emphasized the rebound.

This is a time of modest economic growth, and corporate earnings are great.  It is not yet translating into the brisk net job growth that we need for a robust economic recovery.

You may also like


  • SR September 2, 2010  

    I probably lost you somewhere but I didn’t actually see your estimate. One of your section is titled “Our own estimate”. But then I don’t see an estimate and I see that you have mentioned ” not a forecast, per se, since we do not posit any causal relationship among these variables.” So maybe you don’t give an actual forecast, and this is more about just tying the pieces of various relevant indicators together and comparing with some of the current forecasts? Sorry if I missed a number that you actually gave above. Could you please clarify?

  • oldprof September 2, 2010  

    SR — In the paragraph starting “Turning back to the current month” I note that we are looking for a net loss of over 90K. (It is actually 95K, but I do not like to imply a false sense of precision).
    I could have highlighted this better.

  • SR September 2, 2010  

    That is right! Sorry..I missed it. Very informative post, btw. Thank you.

  • Learner2 September 2, 2010  

    If I understand your estimate of -90,000, if the census bureau laid off 115,000, you would be expecting a net gain in the private and public sectors of +25,000. Is this correct? Thanks for your post.

  • stock market for beginners September 3, 2010  

    I also adopt that approach of comparing forecast from different sources, with this we are able to see clearly whats missing and what needs to be done.