Like a patient with reverse-SAD, the economy has followed strong winter months with consecutive, wobbly summers. Is it coincidence, or is the weather playing tricks with the data?
In the autumn of 2010, things were finally rounding into shape for the U.S. economy. Between October and April, we added more than 100,000 jobs for seven consecutive months -- our best streak since 2005. In the winter, Goldman Sachs boldly declared 2011 the Year of America and many investors agreed.
We were wrong. The economy wilted in the summer of 2011. Suddenly economists were talking about a double-dip recession. Job creation averaged less than 100,000 for four consecutive months until September.
In the autumn of 2011, things were finally rounding into shape for the U.S. economy. Between September and March, we added more than 100,000 jobs for seven consecutive months -- our best streak since, well, exactly a year earlier. In the winter, optimism reined on Wall Street as stocks soared in anticipation of a big 2012.
We were wrong, again. The economy has wilted in the warmer months of 2012, again. Economists are fretting about a double-dip, again. With preliminary estimates in for May and June, job creation has, once again, averaged less than 100,00 for three consecutive months. For the third time in two years, economic prognosticators are getting a bad case of whiplash.
What the heck is going on here? It's like economists have always said: Fool me once, shame on you. Fool me twice, shame on me. Fool me thrice, shame on seasonal adjustments.
Okay, economists haven't always said that. But they're starting to wonder whether the Bureau of Labor Statistics is exaggerating both the cold-month jobs surge and the summer doldrums. Every month, BLS tinkers with its jobs statistics by applying a statistical technique called "seasonal adjustment" to account for seasonal events that would otherwise make jobs data fluctuate wildly (i.e.: weather, holidays, the opening and closing of schools). If the BLS is over-adjusting, that means that the bad months (like the last three) aren't as bad as they seem, and the good months (like the previous seven) weren't as good as we thought.
When Floyd Norris of the New York Times looked at annual change in actual private sector employment since January 2000 ("looking at 12-month changes means you can ignore seasonal adjustments") he saw a recurring weakness in summertime jobs reports. In fact, the last time we had a June, July, or August with more than 100,000 jobs created was back in 2006. On the other hand, the months between October and March have each added more than 110,000 jobs for two consecutive years. Could it be a coincidence? Sure. The Libyan Revolution helped drive up oil prices in 2011, and one year later the Euro/China slowdown hurt global growth.
But if the seasonality theory if correct, two things are true. First, the recovery has been more of a steady muddle than a roller-coaster over the last few years. Second, we'll have two more disappointing jobs reports for July and August, followed by a happy September surprise. The October report drops on November 2. That's four days before the 2012 election.