Henry Smith: I’m going to put a couple of questions now to the asset managers around the table. What factors did you consider when shopping for an OMS?
Carl James: When we looked for an OMS, we wanted to capture everything from portfolio modelling and compliance to managing order flow and the execution end, and the ability to plug in to your settlement end of things. Then it was just a case of going through the usual suspects and having some discussions with them. Budget is a major factor in how you view OMSs and EMSs. It’s very difficult to get technology spend these days. The OMS that we’re using, they’re running quite fast to provide us with a lot of trading functionality. We’ve looked at a couple of EMSs and one of the reasons why we didn’t go ahead with it was how we were going to pay for it.
Chris Sims: Our selection process for an OMS happened eight years ago. EMS functionality and electronic trading didn’t really come into it at that stage. It was the ability to capture all the information, to do a certain degree of modelling and to display the current intraday value of the book and to do pre-trade and compliance. From a budgeting point of view, I can’t speak for the other buy-sides here, but the majority of the developments that happened a couple of years ago were always led off cost reduction and it was: ‘If we do this we will save money because we will be able to get rid of these staff,’ or: ‘We will save money because we will make fewer errors.’ Getting traditional buy-sides to spend money on a speculative business case from a technology point of view tends to be much harder.
Mike Roberts: We believe the buy-sides are looking for much more value in their OMS. The integration of pre-trade and post-trade analytics, the implementation of pre-trade compliance and integration of EMS capabilities – it’s much more for them than just about STP and operational efficiency now.
Lothar Cerjak: It’s already difficult to differentiate between OMS and EMS. When I look at Crédit Suisse, most of the OMS side has been built on our own. We spent quite a lot of money in the last couple of years to rebuild the whole infrastructure and platform. However, where the interface to the EMS comes in, there is a large part of grace – STP yes, but it’s not throughout all the products. Going then into the EMS world, there you connect to the system on the sell-side and the investment banking side but there is still sometimes a break for STP. We tried to have an A-Z system in place, building it on our own, having our own flexibility, but on the other side having a lot of issues with IT resources and spend to get the quality that you’d buy from a vendor. In a large company, it’s difficult to integrate a fully working system and adapt it to the interface and operations and risk systems we have. It’s a critical point.
Chris Sims: I agree with some of the different drivers on OMS selection. If you did a shortlist of OMS vendors five years ago, you’d have pretty much the same list as you would if you did it today. However if you went out and did it on EMS, five years ago there was nothing and now you have all these vendors out there.
Carl James: Eight to 10 years ago when we were looking, one of our criteria was how long these people were going to be around.
Henry Smith: I want to discuss the different needs of hedge funds. How do they differ from those of traditional asset managers? How does this difference affect respective demand for OMS and EMS?
Mike Roberts: From a vendor perspective there are two differences. One is clearly their appetite for different asset classes. There’s much more sophistication with hedge funds in terms of asset classes you cover and what level of functionality you provide around those asset classes. The other difference is all around the footprint you’ll have at a hedge fund. You can’t walk into a hedge fund with a monolithic OMS that’s going to take you eight months to get in and lots of integration and lots of effort from them.
Tim Wildenberg: I see the difference being on the OMS side. We are starting to see some of the bigger hedge funds becoming institutionalised and they’re starting to come to us for advice around OMS. Interestingly the advice and conversations we were having with Carl, Chris and Tony five or 10 years ago are now the conversations we’re starting to have with the bigger hedge funds. Often the fund manager and the dealer are the same person which means some of the functionality of the OMS is needed less. That’s not always the case until they get bigger. The EMS is typically much more of a hedge fund tool. A lot of the single-broker or non-broker neutral EMSs that exist have grown out of the prime relationships provided by some of the original broker deals, where the OMS or the back office is just handled away from everything else.
Henry Smith: How successfully has EMS been integrated with OMS? Is it more a case of closer convergence or full integration at some point in the future?
Paul Scott: One of the concerns is that if you’re integrating you’ve got two separate speeds. One concern is always that, rather than bringing one up to the level of the other, you end up taking it down so you’re operating in the lowest common denominator. If you’re loosely integrated, there’s only a certain amount of functionality you can give access to.
Mike Roberts: There are two very different demands from an engineering architecture perspective and EMS is all about speed and efficiency and getting to the markets as quickly as possible. With OMS it has an overhead in terms of operating against the transaction and the compliance features it includes and so forth, so I believe certainly that OMS will start offering more and more EMS-type features. But I don’t necessarily believe the day will come when an OMS absolutely replaces the requirement for an EMS.
Lothar Cerjak: I agree on that, but I do think there is also a lot of pressure on the street now to do something, as a lot of firms need to have it in place and you need to spend your money carefully. At that point, rather than going for one solution that is not there yet, the pressure will be there to push you in the direction of both.
Henry Smith: What about MiFID (Markets in Financial Instruments Directive)? Is that having an impact on EMS purchases?
Tim Wildenberg: The problem about MiFID is that it’s more about the downstream services provided by the broker, in terms of whether the algorithms will go to all the marketfront venues. I think as brokers and clients work together over the next nine months to define execution policy for MiFID, EMSs and how order handling is defined might well become affected. I don’t think people are buying EMSs because of MiFID.
Tony Whalley: One of the consequences of MiFID that we have seen already is that we had the big seven brokers getting together to produce their own trading platform. One of the side effects of MiFID is going to be that it will lead to a split in the number of liquidity pools. The more you get fragmentation in the market, and the more you get multiple liquidity pools, the more it becomes essential to have either an OMS that pulls them together or an EMS that will enable you to do that. From what we’ve said earlier, the EMS will enable you to do it very quickly, whereas the OMS will take longer to do that.
Tim Wildenberg: I don’t think that that’s happening yet.
Tony Whalley: Yes. I think coming back to your original question, one of the consequences of MiFID is going to be an increase in the number of liquidity pools.
Tim Wildenberg: EMS is a great tool for solving that problem.
Lothar Cerjak: One of the things about MiFID is this ability to prove best execution. At the moment everyone believes that proving best execution is going to be proving a process. What if six months or a year down the road, MiFID turns round and says: ‘No, it’s not just about the process, you’ve got to prove best execution’? What is best execution? You have to define it as well, so not only have you got to define best execution, you’ve got to prove best execution on top of that, and that’s very difficult.
Henry Smith: Will we see further consolidation among OMS and EMS vendors? What will this mean for buy-side and sell-side firms?
Paul Scott: I think there are some opportunities still, just a few, for consolidation, whether it’s OMS vendors buying the EMS vendors or potentially even vice versa, or whether it’s purely just from the broker side where brokers take the EMSs. There are opportunities where you will see the order management system vendors either creating their own tools and then potentially even spinning those off to be separate execution management systems in their own right, not just consolidation.
Mike Roberts: Consolidation is inevitable. The demand is there in the market for EMS features with the OMS. The OMS vendors have effectively two options. One is to build and one is to buy. Economically, to buy is the obvious choice.
Tim Wildenberg: The issue actually is whether it’s good for the buy-side and the sell-side. My personal view is it’s very important that the OMS and EMS marketplaces remain as free and neutral as possible.
Tony Whalley: Can I knock one nail on the head? We are one of the most vociferous in terms of broker neutrality in the market at the moment. I can absolutely assure you that the fact that ITG has bought McGregor made absolutely no difference to us. First of all, because of assurances we were given, but secondly we looked into it ourselves to make sure that we were happy with the situation. We’re talking about the fact that you have to get to a level where you trust people. I think if you go back in this market for five, 10, 15, 20, 50, 80, 100 years, the level of trust has always been essential to the way you do business. That hasn’t changed.
Henry Smith: Should a buy-side firm look to implement a single vendor system that combines the functions of an OMS and an EMS, or would they be better off to integrate systems from multiple vendors in a best of breed approach? Does embarking on the latter strategy give them more flexibility to integrate with other specialist products supporting the dealing process?
Mike Roberts: It depends on their requirements. I think they’re going to have a much higher level of integration and more sophisticated integration if they go with the best of breed, multiple platforms. It really boils down to what their requirements are.
Chris Sims: I would agree with that, but I tend not to see too much reason for going to single vendor systems. All of a sudden you say you want to upgrade something and something else and there’s another nine months gone in upgrading.
Henry Smith: According to research by Celent, some OMS vendors plan to launch embedded EMS in 2007/2008. Asset management firms with both an OMS and an EMS will then, says Celent, drop the EMS, provided the OMS version has the required functions. What’s your view?
Tony Whalley: I think it’s what we were saying earlier about constantly upgrading systems. It’s just the pace with which they do upgrade – by 2007, 2008, I would expect our OMS to provide a lot of the functionality that is currently available under an EMS umbrella.
Paul Scott: I think it actually requires the software delivery model from the vendor to change so your traditional OMS vendor, which has one release a year of its software and an additional three, four or five months of implementation cycle on the buy-side to put that in, is not going to be doing very well, even if it does put execution management functionality into it, if they continue that as their model. If in building the execution management functionality, they can separate to some degree the ability to roll out the execution management functionality apace with how it is now with a separate execution management system, then I think you’ll see a good thing. I think it will have to change to support that.
Mike Roberts: That model is changing because some of our client constituents and hedge funds demand that. Hedge funds require quick to market, small-footprint deployment. They better start responding to that.
Henry Smith: Are EMS platforms commoditised, or fast becoming so?
Tim Wildenberg: I think they are. I think there are 100 functions you need in an EMS and I would guess most of them provide 80. They don’t all provide the same 80 but most of them do more of the things. EMS is a big space. What we’re talking about at this meeting is buy-side EMSs, and I think most buy-side EMSs are subsets of sell-side EMSs with more order handling or integration with OMS functionality. It’s not rocket science. It’s all about usability, reliability, integration.
Mike Roberts: Any technology that’s brought to market to address inefficiencies will eventually become commoditised because of the technology deployed. Inefficiencies have been addressed and the technology itself becomes a commodity.
