As the hedge fund industry matures and becomes more process-oriented, more and more hedge fund managers figure that if they can run their own back offices, why not run other peoples’ back offices? For example, Highbridge spun out Harmonic Fund Services in 2003 and Oak Hill spun out its back office into OpHedge in 2005. As various other managers enter this industry by spinning off their own administration functions, they bring with them a new language.
Nowhere is this more evident than in the recent announcement that hedge fund behemoth Citadel is launching an arm’s length hedge fund administrator, Citadel Solutions. Hedgeco.Net reports that Citadel caused a stir at SIFMA’s Annual “Technology Management Conference & Exhibit” in New York on Tuesday with news of their new initiative (which is slated to go live July 1):
“John Buckley, President of Citadel Solutions LLC said: ‘Approval by the BMA (Bermuda Monetary Authority) is an important step in the further development of our activities. With the addition of Robin to our leadership team, we are well-positioned to become a leader in offshore fund administration. Robin and the Citadel Solutions Bermuda team build upon our unique service offering, the delivery of Operational Alpha to our clients.'”
Whoa. Hold the phone. “Operational Alpha“? At first blush, this term smacks of marketing schlock. But then we recalled a presentation given by a BGI executive to a gathering of AIMA (The Alternative Investment Management Association) last fall that could give credence to this claim. It turns out that the BGI executive, Ananth Madhavan, has since published his research in the form of an article last month called “Transaction Costs Analysis as a Source of Alpha“. While Madhavan’s paper pertains to transaction costs – not administration costs – the same lesson still applies: every dollar of expenses is a dollar is taken directly from alpha.
When you think about, this makes a lot of sense. Most costs (save performance fees) are unrelated to performance. They are essentially a guaranteed negative return with basically no volatility. If you removed the negative sign, it would be quite apparent that this return would be tantamount to (precious) alpha. Our example from a few days ago of the hypothetical fund charging 2% in a year when gross returns amounted to 2%, clearly illustrates that fees basically net-out against alpha. The numbers may be small compared to overall returns, but are gigantic compared to the typical amount of alpha delivered. In other words, cutting fees amounts to a serious alpha “gimme“.
Madhavan has been examining the entrails of securities markets for years. So this article is a little technical. Essentially he and his BGI colleague Mark Coppejans say:
“Once largely overlooked, the impact of realized transaction costs on investment performance is now well recognized. Indeed, transaction costs can substantially reduce, or even eliminate, the notional returns to an investment strategy.”
“…we examine the impact of transaction costs on an active manager’s information ratio (IR), defined as the ratio of expected active return (or alpha) to active risk. We build on the deep insights in Grinold (1989), who shows that the information ratio is the product of skill, measured by the information coefficient (the correlation between the manager’s predicted and actual alpha) and breadth, measured by the number of independent active bets per year.”
In a nutshell, the authors add another term to the original Fundamental Law of Active Management…
…so that it reads…
They refer to that additional thingamajig term as the “endogenous transfer coefficient”. It’s mainly a function of the transaction costs and the manager’s ability to forecast those costs.
Why can’t a manager accurately forecast costs? Because “costs” can include a number of potentially unpredictable things (see chart at right from earlier presentation on the topic by Madhavan).
The article shows that a manager’s ability to predict transaction costs goes up, so does their information ratio. Conversely, the absolute inability to predict transaction costs can lead to a dramatic reduction in the information ratio.
The bottom line is that expense management isn’t just for the back-office any more. Apparently it can have a dramatic impact on a) the alpha-generating capacity of a manager’s strategy, and b) her information ratio. And with alpha reputedly in short supply, hedge fund managers are going to need to squeeze all they can out of a limited opportunity-set.
So maybe it’s not a bad thing that hedge fund managers themselves are moving into the fund administration space after all.