I was speaking recently with the head trader at a large global asset management firm and he said something that made me pause. "I think I hired my last trader about five years ago," he noted.
At first I misunderstood him. I thought he was referring to the fact that assets under management were down, turnover was down and, as a result, he probably wasn't going to need any more traders on his desk anytime soon. But that wasn't what he meant.
He went on to say that everyone he's added to the trading team in the past five years has either been a quantitative analyst or a developer. His ability as head trader to contribute to and preserve alpha is increasingly dependent upon his firm's ability to quantify, customize and optimize its equity trading solutions, he explained. Ten years of active U.S. equity management performance numbers have been blown to bits since the credit crisis hit, and the only way a buy-side trader will survive is if he closely and proactively guards every basis point of positive alpha that his portfolio manager can uncover.Today's buy-side trader bears little resemblance to the order clerks of the last millennium. Traditional asset management firms now execute 48 percent of their U.S. equity orders electronically. There are more than 65 trading venues, more than 600 broker-built algorithms, thousands of smart order-routing logic solutions and more reams of post-trade TCA data than any analyst could ever learn to love. The complexities of the buy-side trading decision require an entirely new set of quantitative and analytical skills that once existed only in the nether regions of sell-side proprietary trading teams and quantitative hedge funds.
Yet today there is an additional component of the buy-side trading process that increasingly is quantitatively driven: the closer coordination and alignment of inputs into the investment decision with inputs into the trading strategy.
In a Perfect World
It makes perfect sense that a portfolio manager would use a transaction cost estimate as one of the variables in his stock selection model. If the expected alpha on his mid-cap stock is 600 basis points over the life of the position, and the expected trading costs on the way in and the way out add up to 450 basis points, then the PM might well look for substitution candidates with comparable expected returns and lower trading costs.
It also makes sense that a trader would trade more effectively if some of the input variables and alpha expectations that went into the decision to buy a given stock were reflected in the algorithm the buy-side trader uses to execute the order. A PM's projected alpha, the timeframe over which alpha is to be captured, and the price momentum and sector characteristics of the security, as well as the expected correlation of prices with other securities and other assets -- these are arguably critical inputs into the decision as to how and where to trade.
Further, individual PMs have their unique behavior patterns. One may tend to be overly aggressive, consistently believing that a shorter execution timeframe is optimal, when indeed he may be bleeding alpha by pushing traders to cross spreads, spray liquidity sources and leak information to sniffer algorithms.
Conversely, there are some PMs who are ultra-price sensitive and consistently miss out on liquidity opportunities. They may be gun shy at responding to liquidity at just the time when it might reduce opportunity costs to step up. As buy-side trading desks capture and analyze trade data for each of these PMs, they may develop a more accurate feedback loop to adjust and compliment the PM's instincts and behaviors alongside the input variables and underlying assumptions behind his stock selections.