Frank Loughlin, senior vice president of Americas Equity Trading at money manager AllianceBernstein, spoke about his firm’s trading process on a panel at Traders Magazine’s conference on Sept. 24 in New York.
>On the trend in transaction costs
Our transaction costs-both implicit and explicit-have come down dramatically in this new, more competitive environment.
>On the “science” of transaction-cost analysis
In the last six months or so, we’ve gone a long way toward making it more science than art. We have a quant team that looks at things like short-term alpha or short-term momentum. We don’t have a sample-size issue; we have enough data to actually take a shot at accurately measuring this. So the quant team tends to look at things segmented by order size as a percentage of average daily volume. They spend a lot of time and effort trying to determine, first, for certain order types, what strategies tend to work best-generic strategies such as spread-capture strategies and implementation shortfall. Then, once they determine the strategy at the generic level, they go further and ask: Whose strategy actually works better? They are in the process of creating a trader cheat sheet.
>On trading in dark pools
We like to remove the choice of venue [from the process]. We use dark aggregators. Some of those strategies may interact with displayed markets, but in a non-displayed fashion. But we certainly subscribe to the view that if you can trade without displaying your intentions, you’re better off in the long run. We have had pretty good success with trying to take the choice of single-point venue out of it, using things where you have access to a number of different non-displayed venues simultaneously, so you don’t have to make the choice and possibly the wrong choice of Pool A versus Pool B.
>On trading blocks
In this environment, if you want to block-trade an order, then you are talking about certainty, so then you are talking about a risk trade with someone with a balance sheet. I don’t think in this environment that if you want to block-trade something that you will wait for a Liquidnet or Pipeline match-that’s for several reasons, including adverse selection and the uncertainty that you might not get a match [in a dark pool]. You need firms to stand ready to provide liquidity when you need it. That’s what it comes down to.
>On the use of TCA in selecting brokers
We are using it significantly to evaluate electronic strategies-the effectiveness of those strategies and the provider. In terms of broker selection, is it a day-by-day, minute-by-minute impact? No. What we are looking for is outliers. On a single trade there are many reasons why performance could suffer. What we are looking for is: Does a particular strategy or a particular firm’s high-touch desk seem to perform poorly on a large number of trades with different characteristics? We are not using it to review firms on a day-to-day basis.
(c) 2008 Traders Magazine and SourceMedia, Inc. All Rights Reserved.
http://www.tradersmagazine.com http://www.sourcemedia.com/
