The Arbitrary Calendar Year

An article in the WSJ the other day, on a subject completely unrelated to manufacturing, got me wondering.  And you know what happens when I start to wonder…!

Winning streaks are tough to keep going in the funds world. As of Dec. 31, just nine U.S. diversified stock and stock-and-bond funds had beaten the Standard & Poor’s 500-stock index for nine straight years, the longest active streaks around, according to a year-end search by Morningstar Inc. of about 2,000 funds with track records at least that long.

Winning streaks can provide investment lessons, so we wondered: Would there be more winners if we measured 12-month streaks using June 30, instead of Dec. 31, as the end point?

Interesting question.  With regards to mutual funds, I wonder what they discovered…

We crunched more Morningstar data and came up with a list of managers who weren’t on our calendar-year winners’ list. We also found a couple of slightly longer active streaks: Phillip Perelmuter at Hartford Midcap and a team led by money manager Mario J. Gabelli at Gamco Westwood Mighty Mites AAA have outperformed the S&P 500 for 10 straight years. In addition, managers at eight other funds have nine-year active streaks.

This reminded me of a presentation at last year’s Lean Accounting Summit that admonished the traditionalists in the audience for "budgeting to the wall."  Budgets are created the year before, then variances are measured periodically throughout the year… right up to the end.  Then wham!  Everything gets reset in January again.  What good is that?  He counseled that budgeting itself is wasteful and forces a restrospective analysis instead of prospective, and suggested that budgets should be eliminated and financial decisions based on the current reality.  I was one of those that took that lesson and ran with it, and my company no longer has budgets.

Just as with mutual fund performance and traditional accounting practices, arbitrary time intervals can mask the current reality.  Sometimes we try to "level" ("fake?") by taking moving averages, and sometimes we apply any one of a number of statistical techniques to draw meaningless trend lines.  I shouldn’t bash all of them; for some applications it may make sense.

But be careful.  What are your year-over-year financial, quality, and other metrics really telling you?  And what decisions are you basing on that information gleaned from arbitrary time periods?  Is it really an accurate basis for those decisions?