A Note On Employment

Includes: DIA, SPY
by: Kevin Flynn, CFA


Tracing reality through the fog.

Weather, numbers and relative strength.

The trend is still intact (but getting older).

Employment is good. It is not great, not as good as some recent reports (and subsequent fawning) made it out to be, yet not as weak as the March report appeared. This matters for two reasons, one perhaps partly behind the Fed's recent decision to lower the threshold for the non-accelerating inflation rate of unemployment (NAIRU) to 5%, and the other being the market obsession over the wrong thing.

The monthly jobs reports in the first quarter receive heavy doses of seasonal adjustment. Every January, about 2% of the workforce disappears from the December establishment payroll total. The large turnover is calendar related, as employers and employees make changes: Layoffs and terminations probably make up the lion's share of the total (the first two weeks of January are the heaviest periods for actual jobless claims), with the rest being due to employees changing jobs, retiring, starting their own companies and whatever else might lead to their being missed in the payroll surveys.

The two percent figure has been surprisingly tenacious over time. In severe downturns, the number does rise higher, with the recent 2009 peak of 2.72% being the highest since the 1975 downturn (2.84%), though rarely does it extend much below 2% (if you're curious, January 2015 came in at 1.99%). The estimate of the actual count returns to "normal" levels over the next few months. One indicator of the strength of the labor market is how long the unadjusted sample estimates take to return to December levels: April in normal years, May (or even later) in less good years.

The Bureau of Labor Statistics (BLS) does not publicly detail its seasonal adjustment methodology (and for good reason, I might add, one can only imagine the back-seat driving and complaining), but one can reasonably assume that factors such as year-over-year comparisons in the raw count, both for counted jobs and for weekly claims, play a substantial role in arriving at the headline totals. February 2015, had a year-on-year increase in the unadjusted count of 2.38%%, much better than the 1.6% of the year before. But that was largely due to a strong January comparison, a 2.33% rate that benefited from an easier weather comparison. It may be hard for us in the Northeast to remember now, but January 2015 was not as cold as the polar vortex January of 2014 that affected a larger swath of the population. Since then the year-year comparison has flattened (now at 2.27% pre-revision), also influenced by the weather.

I am not quarreling with the BLS numbers, nor do I have any axe to grind about them trying to portray the numbers in one way or another, nor am I looking for easy excuses for a weak report. The real point is that the numbers are estimates, and that estimates by nature are going to be better in some years than others, the first few months of the year involving more informed guesswork than the others, and so the series can at times produce weather surprises in the February and March reports, and springtime surprises in the April or May reports, when incoming data refines the initial estimates. In short, the data catches up.

Last year, we had the polar vortex, and so 2015 got off to a fast start comparison-wise. Then came record-breaking snow in the Northeast in February and March of this year, and now the estimates are flattening out. The last two years have seen weaker job recapture rates (i.e., March raw totals compared with December numbers) in the first few months, but those also have seen tough winters in much of the country - the overall differences have still been small. We also had a weather comeback in the second and third quarters of last year, with the result that 2014 posted a 1.94% increase in payroll jobs, compared with 1.71% in 2013.

As an aside, I would like to address "the strongest job growth ever" claims parading across the screens in January and February. The population and civilian workforce are both growing over time, and 250K increases aren't what they used to be. The latest unadjusted payroll estimate of 140.3 million jobs is 10 million higher than March of 2000, and 40 million higher than 1987, but does anyone think that the current job market is more robust than those two years? Keep the percentages in mind - we have yet to top 2% annual growth in this recovery (2000 was the last time) a level that was common in the 1980s and 1990s, and we have a participation rate of only 62.7%, the lowest since the stagflation misery of the late 1970s.

A nugget that may surprise you, surely one of the reasons the Fed keeps saying the labor market isn't as strong as it used to be, is that the payroll count shows 140.3 million jobs, but the claims database showed only 133.4 million insured workers at the end of 2014. The latter number is still lower than the pre-crash peak of 133.9 million insured workers, despite weekly claims data that's running at levels of peak cyclical strength. We will most likely pass that total in this quarter, a little more than six years later, but it does imply that a lot of people are working multiple, part-time jobs. The monthly jobs report counts jobs, the claims insured database counts people.

Another lesson from the data: the last peak in the insured database was the fourth quarter of 2008, a year into the recession. Employment is a lagging indicator, and peak levels of employment are just that - peaks, not some kind of magical springboard to an era of endless prosperity that the Street habitually dangles in front of our eyes. Jobs grow into downturns, shrink into recoveries. Internet or no Internet, low interest rates or high, the business cycle has never been repealed, despite the promises of permanent plateaus, one-decision stocks, new economies and ZIRP. Wage levels tighten the most at the end of recoveries, not at a mid-point that launches another cycle of prosperity.

I don't like the recent downward revisions to previous months in the job reports, traditionally a negative sign, but I'm still not ready to admit them as official evidence. As the weather improves, inventories will need to replenish and the port strike will have its rebound. Those factors will mean upward pressure on economic activity in the coming quarters, and we will again hear talk about the economy reaching liftoff. But the underlying trend has remained completely intact throughout. Not only that, the cycle will still end, with or without the Fed tightening rates, and without the economy ever reaching 3% annual real growth in this cycle. If you're waiting for 2000-style valuation levels, 5% short rates, inflation doubling or negative jobs reports, you're waiting for the wrong things.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.