By Matthew Yglesias Feb 06, 20149:09 AM
The New York Times did an earnings release this morning, and it’s a little bit confusing. The company keeps emphasizing that numbers being slightly down in 2013 from where they were in 2012 is actually good news between the NYT’s fiscal 2013 contained fewer weeks than the NYT’s fiscal 2012. As a casual reader just looking to ascertain whether or not the paper is going out of business (the answer is “no”), it’s hard to do this math in your head. But the surprising conclusion of academic research by Rick Johnston, Andrew Leone, Sundaresh Ramnath, and Ya-wen Yang (PDF) is that even supposedly sophisticated financial markets get confused by this.
Here’s how it works. For accounting purposes, a firm is allowed to define a “quater” as either a span of 3 months or else a span of 13 weeks. A “year” is then four quarters. Since a year contains exactly 12 months, that means that if you define your quarters in terms of months everything adds up. But if you define your quarter in terms of weeks there’s a problem—13 x 7 x 4 = 364 and a calendar year contains 365 (or sometimes 366) days. Consequently, a firm that defines its quarters in terms of weeks needs to hold a 14-week quarter every four or five years. That creates a situation in which you need to do a little math to get a proper comparison. These aren’t nice round numbers that are easy to do in your head, but making the adjustment doesn’t require you to do anything conceptually difficult. What’s more, the occurrence of 14-week quarters is entirely predictable. So the impact of the extra week on earnings should be fully priced into financial markets in advance.
But it isn’t:
Many firms define their fiscal quarters as 13-week periods. For these firms each fiscal year contains 52 weeks, which leaves out one/two day(s) a year. To compensate, one extra week is added to every fifth/sixth year; consequently, one quarter therein comprises 14 weeks. We find evidence of predictable stock returns and forecast errors in 14-week quarters, which suggests that investors and analysts do not, on average, adjust their expectations for the extra week. The ease with which 14-week quarters can be predicted, and expectations adjusted, suggests a surprising lack of effort on the part of investors and analysts.
The last time this happened in a newsworthy way was with Apple’s 14-week Q1 2012 followed by its 13-week Q1 2013, leading to Philip Elmer-DeWitt’s memorable headline “Apple analysts: Stupid or lazy?” But apparently Apple analysts are neither especially stupid nor especially lazy—they’re typical. And yet when you see that kind of error being made about one of the biggest, most famous, and most closely watched companies in the world you can only imagine what happens with lesser firms.
Long story short, this is yet another reason why the Jacobins were right and we should adopt the Republican Calendar.