Armstrong mentioned commodity indices that roll from one front month futures contract to the next as a particular example. “Date rape” is a term I have heard used to describe the way investors in certain passive energy trackers were treated last year, a topic I covered in a blog a few months ago.
Here’s a story of someone who’s made 218% in three years taking the other side of the trades that energy trackers like USO make every month. For anyone holding commodity tracker funds or notes without having a good long think about the roll policy that their investment uses, a report like this should make alarming reading.
You could argue that people who buy leveraged ETFs and hold them for more than a day fall into the same category of dumb money. Their investment is almost guaranteed to be eaten away over time by the combination of market volatility, exponential decay and trading costs.
But don’t capitalisation-weighted equity indices cause investors the same kind of almost guaranteed losses, albeit on a smaller scale?
I refer to the so-called index effect. Study after study – including this one, this one and this one, to give just three examples – shows that when capitalisation-based indices are rebalanced, there is a pronounced effect whereby equities joining the index rise in price between the date of the announcement of the index reconstitution and the date of the actual rebalancing, usually a week or two later. Equities leaving the index suffer the opposite trend between the two dates, often falling in price.
These price trends usually then reverse over subsequent weeks and months. The loser in all this activity is the index fund or ETF. The index being followed is assumed to sell the departing stock(s) and buy the incoming one(s) on the actual rebalancing date, typically at the closing price. This can result in a highly unfavourable price ratio for the funds involved.
A few months ago I looked in detail at one such example: the case of the terrible dealing terms suffered by index funds and ETFs in a stock called Sirius, which was kicked out of the Russell 3000 index during its annual rebalancing last June.
In the words of Ananth Madhavan, author of the first paper linked above and now a managing director at Blackrock, “passive index funds pay a steep price for minimising tracking error by rebalancing on the date of reconstitution.”
If you’re looking after a passive fund, you’re therefore probably better off getting involved in the games of bluff and counterbluff that characterise the activity of those specialising in index rebalancing trades. Banks even make a market in them for third parties, with ETF and index fund managers subcontracting to the banks the terms of the substitution trade (incoming index stocks for departing ones). The banks, in turn, offer to pay the funds enhanced performance; better, in other words, than the actual terms incurred on the rebalancing date. Incidentally, there’s no requirement to pass on any profits derived in this way to the index fund or ETF.
The mere fact that such trades exist shows that index funds and ETFs that follow cap-weighted indices are forced by their nature to sell low and buy high at each rebalancing date. Though you can argue that such trades are typically small by comparison to the overall fund size, investing in this way doesn’t sit easy with me. Aren’t those of us who buy such funds dumb money too?