February 07, 2014
Many people who should know better have been placing far too much emphasis on the weather as an explanation for weak economic data. Cold weather and snow do slow economic activity as people don’t like to go shopping or to restaurants in sub-zero weather or blizzards. But cold weather and snow are normal parts of a winter in the Northeast-Midwest. This means their impact is already included in the seasonal adjustment factors for December and January.
The weather will only have an impact on the data if this winter is notably worse than recent winters. I’m not a meteorologist, but that doesn’t seem so obviously the case to me. In other words, it’s not clear that the weather has had much impact on the data we have been seeing.
I’ll also add that it’s hard to understand the claim from Ian Shepardson that with last year’s seasonal adjustment factors (these change slightly year to year), we would have seen 265,000 jobs rather than the 113,000 reported by the Bureau of Labor Statistics (BLS). In the unadjusted data BLS showed a loss of 2,870,000 this year from December to January compared to a loss of 2,864,000 last year. Last January’s seasonally adjusted jobs number was 197,000.
Given the difference in the unadjusted numbers, at first glance that would look like we would have seasonally adjusted growth of 191,000 using last year’s factors. The actual number will not be simply additive because of differences in seasonal factors across sectors. Still is it hard to believe these differences would get us another 74,000 jobs.
Of course what seasonal factors give, they also take away. (On average, seasonal adjustments have to be zero.) In the seasonally adjusted data we created 149,000 fewer jobs in December of 2013 than in December of 2012. In the unadjusted data the difference was 194,000. If we want to say that we have the wrong seasonal factors so we should be happier about the January numbers, then we would have be more unhappy about weak December numbers.
Comments