How seasonal swings in jobs lead to some unsexy figures
Photo courtesy of Reuters/Mike Blake
It’s a good thing The Atlantic’s Matthew O’Brien stuck around for the last presentation at the Brookings Institution’s Panel on Economic Activity. The topic was, as Brookings warned reporters, the soporific seasonal adjustments to the monthly jobs numbers. But as O’Brien and a few others gathered from the presentation, these adjustments may have made much more of a difference in economic policy than we realized. These adjustments — how the Bureau of Labor Statistics adjusts monthly employment data to control for seasonal swings (summer and the holiday shopping season, for example) — actually have a very large impact on the whole economy.
O’Brien explains how the adjustments work:
“So to give us an idea of how good or bad each month actually is, the Bureau of Labor Statistics adjusts for how many jobs we would expect at that time of year. This doesn’t change how many jobs we think have gotten created over the course of the year; it changes how many jobs we think have gotten created each month of the year.”
These adjustments are calculated based on the past three years of data. To understand why that could be problematic, think back to late fall 2008 and early 2009 and the collapse of Lehman brothers. The seasonal adjustment model recognized massive layoffs from those periods as seasonal incidences.
We spend a lot of time fussing about BLS numbers on our Business Desk page. (Check out our own monthly calculation of their unemployment data: the Solman Scale).
Thankfully, the financial crisis of 2008 and 2009 is no longer within the three years the BLS is now using to seasonally adjust the current numbers.
But as O’Brien points out, the complete employment picture, including seasonally adjusted numbers, influences the all-important Federal Open Markets Committee’s quantitative easing policy. And those post-crisis seasonal errors may have doubled the typical margin of error for the number of jobs added each month from 90,000 to about 170,000, which means that the level of job growth reported each month may be less accurate.
As O’Brien concludes, “There’s nothing sexier than getting the jobs numbers right.”