Thursday, November 8, 2012

Election Results and Employment Data.

The votes have been cast

This analysis was written before the election results were available.  Given the political silly season, it was deliberately not posted until after the election. That should make accusations that it was politically motivated less likely.  The media likes to portray the economy as being an appendage to politics.  A lot of that has to do with the fact that most of the reporters don't know how to analyze the economy.  That's even true of the financial press. 
Covering politics is a lot easier, especially when just picking out your favorite pollster and presenting his or her results is a substitute for analysis.  If that fails, just read the different candidates’ press releases.  However, there is an economy, and although affected by politics, the causality runs both ways.  Who is elected is as much a function of the economy as economic performance is a function of who was elected.

The posting entitled “The Day the Data Died,” noted it was more likely that the employment report for September was just wrong than that it was deliberately manipulated.  That probably was not a popular position among the most zealous Republicans.  However, it also pointed out that the employment report was well short of showing that the economy was recovering.  It was even more absurd for Democrats to view it as confirming the effectiveness of current policies.  If that were the case, it would have happened long before now.
All indications are that the assessment that the report was in error rather than fudged was correct.  More importantly, the combination of the September and October report make it fairly clear where we stand in the employment cycle.  Given that nothing stupid is done to reverse it, employment growth is now comfortably self-sustaining and should accelerate. 

This blog has generally avoided the practice of talking about employment reports each month.  Many commentators have to fill space, and the employment report provides them with an excellent opportunity.  However, this blog has taken the approach that if one doesn’t have anything to say, then don’t say anything.  Most of the time the employment report for any given month says nothing: what matters is the trend across multiple employment reports.
Thus, postings on this blog that have addressed the employment report have been about equally distributed between those posted before the report and those written after the report.  Often, knowing what to look for in the report ahead of time is a good way to identify an emerging trend.  Consequently, what to anticipate is sometimes more important than trying to interpret data after-the-fact.  However, “The Day the Data Died” actually did both.  It assessed the September numbers to the extent of commenting on whether the data was fudged or just wrong.  That's an issue that would not have existed but for this administrations acknowledgment that it viewed fudging numbers as an election tactic.  But, it also concluded by identifying what to look for in the October data as confirmation of an improving trend.

The reason for the curious posture of spanning two reports reflected the fact that interpreting the employment report for September without knowing the October results was impossible.  The September report was one of those anomalous reports that occur periodically during a business cycle.  The data anomaly is however a cyclical phenomenon.  It has to do with people finding jobs, starting businesses or creating jobs rather than finding work at the existing employers who are a part of the employer survey.  The magnitude of the anomaly would indicate that it was the result of both that cyclical phenomena and sampling error in the household survey.
The exact statement in the last posting was:
 “The bottom line, however, is that the employment report can contain data that is just wrong. It's much easier for one month’s report to contain a number that is wrong than for the report to show a false trend. Thus, next month’s report will eliminate the ambiguity introduced to the interpretation of this month’s employment report.”

Clearly, another drop in the unemployment rate without confirming data will only confirm that the administration is cooking the books. Confirmation should come in the form of a pickup in employment in the employer survey, a reduction in the rate of combined unemployment and underemployment, and revision to this and previous month’s anemic showing in the employer survey. Of less importance, but also telling, is whether the increases in employment in the employer survey disperses to include more industries.”
The report for October didn't show an unsupported drop in the headline unemployment rate.  To the contrary, the rate rose.  It rose for the right reason: increased labor force participation.  The combined unemployment and underemployment rate actually fell, and there were revisions upward to the September report and the previous few months.  The dispersion across industries wasn't dramatic, but that's a minor confirmation of the validity of the September employment gain.  Also, job growth reflected in the employer survey rose and job growth in the household survey declined.  That shrinking in the gap between the two surveys would indicate that the employment picture is still less robust than in other recoveries.  But, that is totally consistent with the anemic recovery in other indicators.

So basically, while someone now has reason to crow about the election results, there is no reason to crow about the employment results.  They are a continuation of weak recovery that can't get enough traction to be robust against political folly.  It may be sufficiently robust to reduce the likelihood of political folly.