An article on the front page of the Washington Post business section today was headlined, "Retail Sales Numbers: Which Ones Do You Buy?" The mystery described by the article stems from the fact that the June retail sales reported by the Commerce Department had fallen sharply from the May level. This negative picture supposedly presented a stark contrast from the data reported by department stores on year over year sales, which the article asserts propelled the markets to record highs. As the article eventually points out, the 3.6 percent year over year gain reported in the Commerce Deparment data is virtually identical to the gain reported by the department stores. So the problem is simply comparing year over year data with monthly changes in sales. And that is pretty much where the Post article leaves it. But, since they took the time to write the article, let's look a hair deeper. First, we are comparing nominal sales data that are not adjusted for the effect of inflation. If we assume that the goods in question rose in price by 2.0 percent year over year, then we are looking at just a 1.6 percent growth in real sales. That is not very good. Now, we have to be a bit more precise about the recent changes. Real consumption expenditures grew at an average rate of 3.3 percent for the last 3 quarters. This means that if real consumption expenditures are just 1.6 percent higher in June than they were a year ago, then real consumption expenditures have been falling in the last three months. To be sure, there are some differences in coverage between retail sales (primarily goods) and total consumption which includes services like rent and medical care, but the June data suggests consumption spending has likely been very weak lately, and that conclusion doesn't change regardless of which data series you use.
--Dean Baker
