August 03, 2012
Binyamin Appelbaum has an interesting piece reminding readers of the importance of the seasonal adjustments in the job numbers that are released each month. It points out that July has the second largest positive seasonal adjustment, after January, of any month. (I don’t quite understand the chart, which seems to show positive seasonal adjustments for every month.) For example, last year the seasonal adjustment added 1.3 million jobs to the raw data, turning the seasonally adjusted number into a gain of 96,000 jobs, despite an unadjusted loss of 1.3 million jobs. Appelbaum’s point is that a small error in the size of the adjustment would have a huge impact on the jobs number reported for July.
The point is actually even more important than Appelbaum suggests. There are very large changes in employment month to month for seasonal factors that have nothing directly to do with the state of the economy. When these patterns change then the jobs numbers will give us a misleading picture of the state of the economy.
That is why the unusually warm winter weather gave an overly optimistic picture of the economy. Since this job growth was borrowed from the spring, the subsequent months gave us an overly pessimistic picture of the economy. Seasonal patterns can also change for reasons not directly related to weather. For example, stores now start holiday sales as early as October and the auto industry no longer shuts down all their factories in July for retooling.
Anyhow, Appelbaum is right to remind us about the importance of seasonal factors in the jobs numbers. We might not need a weatherman to know which way the wind blows, but we do need one to know how the economy is doing.
Comments