The growing sophistication of the buyside was on display at a recent trading and technology conference. At Worldwide Business Research's (WBR) TradeTech USA conference in New York City last month, buyside traders spoke out on order management systems, FIX, algorithms, portfolio trading, ECNs and other tech topics.
Peter Driscoll of Northern Trust, Floyd Coleman of AXA Rosenberg, Richard Tsai of Barclays Global, Curt Engler of BlackRock, Gary Chropuvka of Goldman Sachs Asset Management and many others, held their own with the sellside at TradeTech's inaugural U.S. conference. For the past five years, the international WBR has hosted a
TradeTech conference in Europe. However, the three-day New York talkfest last month, sponsored by Wall Street's major brokers, was the first conference in the U.S. Brokers came to give algorithmic trading the hard sell. But the buysiders attending were not all lambs for the slaughter.
Strong voices from fund management firms spoke of a new era: Trading power is shifting from the sellside to the buyside because of technology and commission compression. And the buyside trader who doesn't cotton to technology will be hurt.
Below are edited excerpts from three presentations at the conference by a trio of the Street's largest money management shops. These buysiders outline their use of technology, offer advice and make forecasts.
Peter Driscoll, senior equity trader at Northern Trust, a $570 billion index player. He is also a former Chicago Stock Exchange floor trader.
Technology will continue to develop. It will continue to make us more efficient. Order managements systems, the Liquidnets, Harborsides, Pipeplines, PBTs. These are all trading tools. They are all things that a trader on the buyside is going to need. He's going to have to know how and when to use them.
As the evolution continues, demands will be placed on the traders to keep up. The days of pulling a body off the reception desk to fill in on the trading desk are gone. Buyside trading desks are going to have to begin to pay for talented people who are up to date on the use of these tools of automation.
Pre-trade and post-trade analysis. These services are going to become user friendly. Pre-trade analytics is going to become less hypothetical and much more understandable. They are going to be used as tools to develop traders and to get what is expected out of the trader.
Block trading, program trading, algorithmic trading. It's all about the money. Block trading costs four cents to five cents. Program trading is three cents. DMA, one cent.
'The days of pulling a body off the reception desk are gone.'
Peter Driscoll, Northern Trust
The other side of the equation is that we are under pressure on the buyside to reduce our fees to our investors. There is very little room for us to be paying more. And we are being squeezed. So we are just passing that down. Commission compression is here to stay. The blended rate of many firms now is down to about two-and-a-half cents to three cents. This rate isn't enough to support the sellside infrastructure. So, they push us to these algorithmic trading tools. These less costly tools. You have to keep a couple of things in mind. These trading tools have been around for a couple of years. The hedge funds have started to hire out some of the talent that developed them. They are reverse-engineering these things to be used against the buyside community. We have to be on our toes. We have to watch what is going on. The algorithmic model is difficult to brand. I see us being moved to a less personal relationship with the sellside. And I don't understand the strategy behind it. The high-touch, low-value offering being pushed towards us seems to be destroying the relationship. SRO trading venues will recognize the shift. They will start offering trading services directly to their members' customers. Nasdaq is coming out with some experimental tools on its Website. They are going to be offering them directly to the buyside. Floyd Coleman, head trader at AXA Rosenberg, with $56 billion in assets under management. Coleman encouraged buyside traders to kick the technology tires. His shop beta tests new systems, uses a proprietary OMS with IOI matching capabilities, and uses FlexTrader as a staging area for some orders. [Buying technology] is similar to choosing a spouse. You want something that is compatible. Has some attributes that you crave or desire. But also has some drawbacks that you can live with. I encourage you to try these systems. And just experiment and see what you like and what you don't like. The only caveat I would make is that you should have a checklist. Here's my checklist.
'[Buying Technology] is similar to choosing a spouse.'
Floyd Coleman, AXA Rosenberg
I have multiple systems. In terms of checking market impact on trading costs, I use more than one system. And more than one measure. You could be misled if you are just using one measure. And it looks great on one measure, but you look at some other measures and it is not so good. You want something that has grassroots support. Not to rule out new products because there are a lot of good new products, but at least you should start out with something that has grassroots support, a large installed base. Or just something that hits the ground running. That helps you in terms of the weak spots of the newer systems. Sometimes they are not quite fully developed and they have hiccups and you have to find the hiccups. You want something where the problems have been solved. The kinks have been taken out. You need something that is easy to use. This should be something that is intuitive. When you get the system you should just be able to eyeball it, and at least know how to use its rudimentary functions. My rule of thumb is if I have to use the manual for anything that is not sophisticated, I don't really want to use it. You should be able to do a simple trade without having to reference the manual. Next, you need something that is consistent with your trading style and philosophy. In our case, someone offered me a VWAP engine which I definitely wouldn't use because we don't use VWAP in our trading. We use target prices. Something that can help us find liquidity – that would be useful to us. Also, you need something that complements your existing or your upcoming order management system. You want something that will connect easily to your OMS, so that you can send orders back and forth. And, in the future, hopefully send more information such as target prices or alpha information. [Data] should be able to go back and forth between these systems very easily. Finally, you need something that is adaptable and easy to customize. And hopefully get a quick turn-around with the customization. The system will probably not fully meet your needs out of the box. It may do 90 percent of it. But for you to get that last 10 percent for the specific way that you trade or use it, you will probably want the vendors to tweak it. And it is important that, one, they do what you ask them. And, two, give you a fast turnaround. A good test of this will be when Reg NMS information comes out. A lot of these systems will have to be re-jiggered to take that into account. The best of the breed will give you the fastest turnaround. Curt Engler is the head quantitative trader at BlackRock Financial Advisors. BlackRock is primarily a bond house, but also manages $20 billion in equites in quantitative fashion. Engler's group trades lists. The trader is a disciple of the mathematicians and computer scientists Robert Almgren and Neill Chriss and their "Efficient Trading Frontier" hypothesis. Their work looks at the trade off between trading costs and timing risks. It is behind the newly- popular arrival price algorithm. We will periodically rebalance and I will inherit a two-sided list. It has a varying number of names, depending on size, depending on the product. In order to maximize the value of the investment decision, the key is to minimize the shortfall between the buy and sell portfolios. The implementation shortfall. I definitely believe that the trading strategy is driven by the investment process and pre-trade analysis. Gaining popularity is the notion of the efficient trading frontier where just as in the portfolio construction process where you give up some risk and return, there is an optimal asset mix that will maximize your return for a given level of risk or minimize your risk for a given level of return. It's the same for maximizing alpha per targeted tracking error. But when I get these lists, there is a definite trade off between the cost and the risk of the trade. And associated with that an optimal trade schedule can be generated. One of the products that I use is ITG's ResRisk+ where you can run multiple optimizations given the level of urgency. You can run these optimizations and see your optimal trade schedule broken up into various bins and you can change it on the fly. And then from there, their limit order model will break it up and algorithmically trade the list throughout the importance of the optimal bins selected.
'The efficient trading frontier is gaining popularity'
Curt Engler, BlackRock Financial Advisors
CSFB's Portfolio Hedging Device is another system that can take the algorithms with a layer of risk control on top of them and push you on to that efficient trading frontier. The future of the desktop…I think the key is in the configuration of the desktop. I don't see a lot of new algorithm tools being generated. Right now PCs are getting a little cluttered with multiple broker front-end systems. Hopefully, there is an OMS or front-end where you will have your optimizer integrated with your pre-trade analytics. Pre-trade is going to be pretty important because of these costs and risk models that are going to determine your level of optimality to put you on that efficient trading frontier. There are hopeful signs that pre-trade is definitely going to become more of an area of intense research. With my optimizer I can generate my optimal trade schedule. But I think the configuration is a little more difficult. I envision that platform allowing me to select any strategy and suggesting the algorithm that would make the most sense. Say Broker "A"s percent of volume, Broker "B"s VWAP, Broker "C"s implementation shortfall, or something we developed in house. We are not at the point yet where you have access to every possible strategy or destination. And where you are not limited by the provider of either the algorithmic strategy or the optimization strategy. I am sure some shops are configured this way, but it is certainly easier said than done. There is, of course, short-term alpha associated with our trades. Having better indicators of what that actually is that can be keyed into that optimizer. So, if you had cost, risk and alpha then you definitely get a much better idea of the level of urgency.
