AC: Klaus, could you broadly talk about your role and what you aim to achieve in it?
KP: I’m head of currency and overlay strategy for the EMEA region. What that really involves is educating both prospects and clients on the implementation services that Russell provides, helping our clients with currency, derivative and other implementation strategies that we could do from them. Everything that we do at Russell is bespoke to our clients’ needs.
On the foreign exchange side, Russell provides currency hedging overlays, where we do bespoke hedges. We also have an index-based overlay strategy called Conscious Currency that we implement for a lot of clients. But also we do a lot of individual FX execution as part of our services. Much of our flows are from our own fund managers and for our transition management and derivative overlays. Then we also have directed execution from our third-party clients, either managers themselves who outsource all their FX execution to us, or pension schemes that will direct their managers to have their FX execution directed through Russell’s desk.
AC: As far as Russell’s overall philosophy, how would you sum it up?
KP: For everything we do we act as a fiduciary and an agent. What that really entails us to do is put our clients’ interests first. As a pure agent, we have no book of assets we can trade in or out of. We’re not a principle, so for everything we simply get a fee for our services rather than receive remuneration off of the spread. We seek to get the best price and reduced risk and cost for our clients. Our interests are very much aligned with our clients in not only our FX execution but all of the different strategies we do, whether it’s physical execution – equity, fixed income, even commodities derivatives etcetera – or other transitions assignments, derivative and currency overlays.
AC: Returning to FX execution, those two words cover a wide range of styles and methods for achieving execution, what are the salient features of how you handle execution and work with other parties to service the clients?
KP: We do cover the range, obviously mostly forwards and spot trading. We do some FX options but there hasn’t been as much interest in our client base for that. We can trade most currencies, however with some of the restricted currencies of course we have to go through the client’s custodian or local banks, as everybody else does; but even in that sense we’re more working more and more to try to execute some of the restricted currencies more competitively.
The way we work is we have a centralised currency trading desk, that’s basically a 24 hour a day desk based in Seattle, Washington. We receive orders in various ways. We also have our electronic platform called the RFX network that we’ve developed with Integral Technologies and some clients can send trades to us directly through that network. We occasionally still get orders from emails and spread sheets, but the majority are electronic files. Those trades get centralised and, depending on the clients’ needs, we can go out various times a day to execute.
We have a lot of clients that still want to trade at the WM or try to get the WM pricing as close as possible, due to benchmark or other needs. But generally we try to go out to the market several times a day, as well as market orders. We try to net and match as much of the activity as possible. If clients are relatively agnostic as to when they go out it increases the chance of matching up activity, not only within their own legal entities – say for a pension scheme with multiple managers, but we also do a lot of peer to peer matching. Especially on our new RFX network. What it does is it allows both our own trading desk and outside managers that can get the RFX network on their desks to submit trades, to try to match off first internally as much flow as possible at the mid, and then do a peer-to-peer match which if that’s done at the WM we can actually get the mid. Basically going out to get mid on your own with a counterparty is all but impossible now. So it aims to be on our clients’ side and get back to getting mid-price on WM. And then any residuals that don’t get netted off and matched the system uses an algorithm to try to execute the residuals at as close to a mid-price as possible.

AC: If we can zero in on that, what sort of algorithms are you using? Are they supplied by banks or developed in house?
KP: The technology we use is Integral Development Corporation and then we’ve added other features that Russell is able to bring to the table. It’s their technology, going through various trading techniques, using over 50 counterparties to try to get streaming prices and try to get the best prices right around the WM and to try to minimise the spread. Integral has had long experience in being a FX technology firm and because we have a lot of interests aligned with them, we’ve come together to try to develop this technology to not only use their algorithms and their systems but also provide Russell’s knowledge base and our ability to do peer to peer matching. So it becomes the best of both worlds.
AC: It sounds like your client base is very much weighted towards their deals being done at or as close as possible to the WM. Why is that? In some situations, if there’s a large exposure to take care of, why not just go out at whatever time and try to get the best execution possible as opposed to the WM?
KP: It’s a good question. A lot of clients are still benchmark targeted, so their aim is to reduce tracking error. And as most index providers still use the WM pricing as the way to calculate the local index levels, if clients, especially asset managers, are concerned about any tracking error against an index they’re going to want to trade as close as possible to the WM. Now that said, we are – especially with what’s happened to the WM being in the news and the changes around that – we are having many discussions about moving away to see if it’s worth it, because WM trading, unless you’re able to get a match or get mid price somehow, is increasingly expensive.
Given that there’s a cost now, which means there’s going to be tracking error one way or another, some of our client base are starting to think, ‘Well, perhaps I’m not as concerned anymore and I can do a TWAP trade or something of that nature, to try to get better execution or different execution, because it may not be as relevant anymore’. That said, we still do have a large number of clients that, even though there is a cost now of trading at WM, still want to trade at the WM, which is also why we’ve developed this RFX network as well, to help try to claw back the mid.
AC: In terms of the ones starting to think of, let’s call it life beyond WM, what would you be looking to offer them?
KP: The RFX network can do a net or match at other times during the day. It doesn’t have to be at the WM – that’s just been what’s demanded right now. Then you’d either do an algorithm throughout the day to get best pricing. Or, if there’s a specific time during the day, or a TWAP or VWAP, the algorithm can do that. That said, that’s really what that technology does. Our model though is really bespoke about what our clients need. We have clients, if they want a best ex, where we’re going to go out and bid out that trade against our various counterparties and get them the best trade, not necessarily needing an algo. If you get some big orders – and we get some very big orders – using an algorithm may help break up that trade and reduce the footprint in the market in various ways. Algorithms are one way we do it, but we do telephone trading, we use other electronic systems, plus we have a lot of relationships with many different counterparties.

AC: The RFX network, what’s the background on that?
KP: We’re rolling that out globally. We’re talking to a lot of individual asset managers about it. But we’re also discussing it with some large schemes that do a lot of FX execution themselves. We use it in our own desk as part of our trading. People can outsource all of their FX to us, for which we’ll use RFX plus a lot of other trading resources we have, or the RFX network can sit on a client’s – say an asset manager’s – desk and they can submit some or all of their FX activity through that as well. It’s very flexible. It really is bespoke.
AC: It sounds like you’re actually competing with the sell side in that regard.
KP: I wouldn’t say we’re really competing on the sell side. We provide a service that we feel is relatively unique in that it’s a pure agency-only execution desk, and RFX network is an execution outlet that aims to try to match as much activity as possible, both internally and peer to peer. And then it allows us to get as close as possible to any residuals that have to be executed in the open market to a particular time, WM being the one in demand. But that can be at any time of the day, depending on our clients’ needs. So, I wouldn’t necessarily it’s competing per se. It’s just an enhancement to a service we already provide.
AC: The last thing I wanted to talk about was the future in terms of the take-up of algorithms in the market. You’ve witnessed the changes in the market over quite a long time. There’s been a lot of discussion about how FX algos are gaining traction and how various buy-side players, particularly on the real money side, are starting to get more comfortable with them. What’s your broad take on all that and some of the benefits algos might offer?
KP: I think what is changing is really clients’ awareness of what’s happening in the foreign exchange market. It previously was seen as an operational activity without a lot of view into the potential cost and the implications of currency execution and pricing and spreads. I think that has changed a lot, especially with the press around the WM and other trading issues. That I see as a positive thing. In terms of where I see algos could be used in the future and would be used, if people think that it will reduce a footprint in the market and take some of the power out of some large counterparties that they have concerns about, I can see that as a potential benefit. For us especially, it’s all around finding ways to get the best execution, to reduce footprint in the market, to reduce spreads. And algos could help with that.
