There's been a great deal of sound and fury over the jobs report. Here are two thoughts that come to mind.
Against the bulls: An increase in 245,000 payroll jobs was about 100,000 above expectations (at 150,000). But exactly a year ago, the experts were calling for a January 2011 increase of about 140,000 jobs and the initial number was just 36,000, or about 100,000 below expectations.
A year ago, the bulls dismissed this as a seasonal issue, due mostly to the freakishly bad January weather. And they turned out to be correct ? later revisions were upward and the average through the first quarter of the year ended up being close to the 2011 trend.
Well, they need to keep this in mind now. We have had a freakishly good January in terms of weather, which might be giving a boost to the report. And we see particularly in strong gains in restaurant workers and construction workers, the exact kind of jobs that would be salvaged by an unseasonably dry and warm January.
Remember, January is actually a month where the economy sheds jobs ? as the holiday surge finally comes to an end. So this print of 245,000 is the government's effort to see through the holiday effects to divine the underlying trend. But if the holidays had a different effect this year than they normally do, then the number will be off. In particular, it might be the case that businesses kept workers on through January because good weather boosted activity, and the government interpreted fewer seasonal layoffs as an indication of a better underlying trend.
My guess is that just as the bulls were right about the weather last January, one should consider the weather a factor this January.
Against the bears: Drudge had a headline indicating that 1.2 million people dropped out of the workforce this month alone. Huh?
That's not what really happened. In fact, the government revises its assumptions about the population every January, but there is a catch: It does not go back to adjust the old unemployment numbers. So, there was not a "drop" in the workforce ? because that requires what is essentially an apples-to-oranges comparison between December and January.
My bottom line: This was a good report, no doubt. But we saw some equally bad reports not that long ago ? because the monthly jobs report is an inherently bouncy metric. There are the typical problems of sampling error that come with any survey, and on top of that there are non-random errors that can creep up because of issues like the weather. The average job growth in 2011 was 152,000 per month, but the standard deviation was a whopping 70,000. And we had a few months last winter when things looked as good as today's number, only to taper off later.
On a political level, call it a kind of sugar high for the Obama administration. The president has been able to tout a drop in the unemployment rate for several months. However, the foundations still suggest a staggeringly large number of people not in the workforce. That is biasing the unemployment rate downwards, suggesting more improvement than there has really been.
Put another way, the final quarter of 2011 saw an increase in 683,000 jobs, according to the household survey, but there was also an increase in the population of 513,000. In other words, we really only added jobs to keep up with population growth, so the big 0.5 percent drop in the unemployment rate was mostly due to potential workers leaving the labor force.
If people become encouraged that there is work to be found, and start returning to the job market, the unemployment rate is either going to stall or increase. So, the high-fives for Team Obama in February might be headaches in October.