Mention the term "subsector" to anyone in my household (especially my sandwich- for-every-meal son) and visions of sharing a foot-long deck of salami and cheese on a French roll begin to appear. Real salivation, however, is induced by the mention of that term to indexers.
We're indexed and subsectored to death, it seems. Not that there's anything wrong with having benchmarks, mind you. But must we have so bloody many? A quick callup of my database renders 329 U.S. equity sector benchmarks alone-and this f rom only two providers. Dow Jones bestows 111 of 'em while Standard & Poor's unloads its truck with 218.
If the typical attribution analysis slices U.S. equity into ten market sectors, it's easy to see that the bulk of these benchmarks are subsectors at best. At worst, they're sub-subsectors. From this, I've decided to shun any dinner invitations to the homes of the index provider honchos. My dessert dollop is likely to be sliced so razor-thin that I'll be able to view my dinner partner through it. S&P, apparently taking in its indexing cues from the old Wonder Bread ad agency, carves the healthcare market into-get this-TWELVE slices. Are there even twelve ways to get sick nowadays?
Subsectoring turns out to be discriminatory as well. Possessed with the mindset of a Florida swampland developer, S&P decided the "Air Freight & Couriers Industry" needed to be further divided into subgroups. We get the "Air Freight Carriers Sub-Industry" as the result, but no benchmark for couriers. I'm trying to keep this information from the bike guy who delivers packages 'round these parts. He's paranoid enough already, but if this becomes known to him he'll go positively-ahem-postal.
Drunks, depending upon the imbibe of their choice, should feel put upon by these indexing behemoths too. Unsatisfied with the specificity of its "Food & Beverage Index," Dow Jones cleaves the guzzling side of the standard into a "Soft Drinks Index" and a "Distillers & Brewers Index." So, messieurs Dow et Jones, what are vintners- chopped liver? Suds lovers, on the other hand, find no succor in S&P territory. The alcoholic side of S&P's Beverage Index is divvied only into a "Distillers & Vintners Sub- Industry," leading rather startled tavern denizens to wonder how all those Budweiser brewing plants were vaporized overnight.
If this denotes a trend, the smelly sequelae of any subdivision of S&P's "Fertilizer & Agricultural Chemical" benchmark will be staggering.
The willy-nilly propagation of sectors isn't, of course, illegal. In fact, two "laws" seem strongly to support it. That British historian C. Northcote Parkinson's eponymous principle quickly generalizes to the form "a structure will expand to fill the resources devoted to it," leads to all sorts of mischief. Parkinson's Law, for example, perverts the rationale behind building more hospitals into nonsense: "If we build them, more people will then get sick." That said, if index providers create more healthcare subsector indexes, will they be used?
Murphy's Law also plays a role in subsectoring. While Murphy gives: "If things can go wrong, they will," a Parkinson's overlay yields: "When things go wrong, they'll go wrong in the worst possible way."
It can't get much worse than juggling 329 slices of U.S. equity. Or can it?
To slow this freight train (er, maybe that should be "Air Freight") down, I'd like to see index providers perform a simple economic utility test before slicing-and-dicing the market further:
Let N = the total number of sector benchmarks in an index provider's universe
T = maximum number of hours in a U.S. market day to figure out how to use N
U = economic utility of adding another subsector
Then U = (T-6.5/N)
Gotta run. My courier's here.
