Paris - Upcoming change to dark pools in the new regulatory regime was a key issue at the recent Trade Tech conference in Paris. In this latest edition of 'Sound Bytes', Automated Trader hears what industry figures and one regulator from the UK's Financial Conduct Authority had to say.
We also hear about views on trends in gaining access to liquidity and the popularity of emerging markets for the algorithmic trading community.
Sound Bytes lets you cut through the clutter to hear exactly what's on people's minds. The comments came from panel discussons and interviews by Automated Trader.
Tim Rowe, Manager, Trading Platforms & Settlement Policy, Financial Conduct Authority (panel discussion)
In terms of the dark space, in cash equity ... I think a key thing that stands out for everyone when I have this conversation is the 4% and 8% caps, which will be introduced on the amount of usage of the waivers. Of course one of the things that has sometimes got a little bit lost in the discussion about this is that the 4% and 8% caps ... don't apply to all dark trading. It's actually about, effectively, one and a half usages of the four waivers. So very briefly, the four waivers are: reference price waiver, where you trade on a system which derives its prices from another venue, the negotiating trade waiver, where you do a trade on an off-book basis, report it to a venue at or within the spread or because that trade is subject to conditions other than current market prices...; there's the order management facility waiver which is all about sending orders to a venue and those orders remain dark for some reason, it might be something like Iceberg functionality; and then there's the large-in-scale waiver. Of those waivers, it's only actually the first, reference price waiver, and half of the negotiated trade waiver... which are actually subject to the cap. All the other stuff, so for instance large-in-scale, is not subject to the cap, which is important because when you look at the 4% and 8% caps, they look pretty darn low... Of course you still have the ability to trade OTC to an extent, but there will of course be now on ongoing trading obligation to help draw OTC cash equity onto, supposedly, the lit venues. So that's very much fixed. We've got those 4% and 8% caps, we know what they relate to.
How the actual mechanics will work of course is completely different. How will we make sure we've got the right data to be able to judge where instruments sit vis a vis those limits? How will the industry go about managing trading versus those limits and how should that be organised? Those are all questions to be answered in due course. And quite aside from that, with respect to the large in scale waiver, they might not be subject, those trades, to the caps. But where the large-in-scale threshold sits for any given instrument is an open question and that's something that ESMA will be consulting on in due course. Should the thresholds go up, down or stay the same? So, there's a mixed bag at the moment in terms of what is currently fixed by virtue of the Level 1 legislation that was agreed a couple of months ago, and what is still in a state of flux. And an awful lot of this will be explored in the next couple of months as ESMA starts to produce consultation documents.
Brian Schwieger, head of equities, London Stock Exchange Group (panel discussion)
I think the issue that the industry has is that, if we look at dark pools today, most of them are operating under the reference price waiver. Most of the executions that we see going on in the dark would come under the reference price waiver. So the challenge for the industry is now, what can we do with that flow, do we simply put it on to the lit venues or can we find other ways of trading that which would be within the new cap regime?
Some of the ideas which have come from the industry... intraday auctions, that's obviously something which the LSE is currently consulting on. New order types, for example, conditional orders: so this is the idea whereby a tentative order is placed in a dark pool and where another order comes, which could potentially match with it, they're firmed up and brought together; the advantage of the conditional order type is that it allows for larger execution to take place, thereby hopefully increasing the number of large-in-scale executions that the dark pools might be able to have. Other execution types again will be things such as systematic internalisers. And I think really the solution for this is not going to come from the market venues themselves but actually from the brokers, and potentially also from the buy-side in terms of putting together algorithms which are actually able to use all of these various tools and come up with solutions for the investor.
Byron Griffin, head of equities, EMEA, Thomson Reuters (panel discussion)
An interesting way MiFID and MiFIR is going to tackle this is, through the waivers that they're introducing, and through the strict application of those waivers, it's going to ... increase the importance of a central order book tape. Now how that gets implemented, it could be implemented in multiple different ways. It could come from a regulatory (body), where it's mandated down. There's a lot of cross-structures involved in that and it's been done in the US -- it took seven years to do it in the US. It's quite burdensome now to manage and (there are) even some market participants looking to change that order book structure, in terms of that tape structure, to make it more flexible for what they're trying to achieve. So is that the right fit for Europe? Potentially not. Maybe the right fit for Europe is where you have a mix between regulation and market-led, so an example might be where we stipulate from a regulatory viewpoint what exactly is the central order book tape and how that should look... But there needs to be extreme clarity on what does and does not constitute a tape. Now, in Europe, it's more complex because you've got multiple currencies as well.
Michael Seigne, managing director, head of electronic trading Europe, Goldman Sachs (panel discussion)
(On the future of dark pools:) We're sort of thinking about it in terms of three sort of main levels. The first would be, we certainly need to work on, in the algorithmic layer, how do we improve the trial order size. So starting to aggregate trial orders, those sort of things. And you get into this in the world of trajectory crossing.... The second component, which I think is probably more interesting, is the component of the whole contingent order type scenario.... When we look at the breakdown in dollars traded between the electronic business versus the other businesses, roughly 55-60% of the notional dollars traded every day is self-directed by clients. Now it seems kind of crazy that, if clients are looking to trade bigger blocks, why that should always be walled away from the environment where larger blocks trade, which is typically the cash business. So the whole notion of contingent order -- you know I think it applies to brokers' business models too -- is that, in the sense that we have to really work to change our business models, so that we can make those two interact if the clients want it. That's a big if. So within Goldman, we certainly worked on what we called visibility, where we give clients the ability to ... make that information available for the cash world, where blocks tend to trade. We've had a lot of success with that. We're seeing about a 70% pickup in price improvement as a result of that.
Alexandra Foster, Global Head of Strategy & Business Development, Global Banking & Financial Markets, BT Global Services (AT interview)
The liquidity, at the moment, is around the same levels as 2004, but it's the methodology in how you source that liquidity that seems to be one of the issues that people are dealing with. A lot of people are talking about pure electronic trading, but that electronic trading doesn't answer all of the questions when you're looking to be able to source a big block in the market. You've got to use a range of tools and I think that one of the things that we're seeing is how all of those tools come together, you know, voice, integrated with collaboration tools.... So that high touch feel within a low-touch environment, to be able to source liquidity, seems to be a trend that we're hearing both in terms of the debate, plus also the technology roadmaps that we're seeing from our clients.
People talk about liquidity drying up, but actually if you have a look at historical levels around liquidity, those liquidity levels are around the 2004 levels, we're not at the sort of liquidity crunch that we saw during the crisis, but actually liquidity out there in the market, per se, isn't that bad.
Per Loven, head of corporate strategy and produce management, EMEA Liquidnet (AT interview)
I think there are two things that regulators try to tackle, and they're doing this on both sides of the pond. A, is the one of systemic risk, which is related to the Flash Crash, and the technological trading issues we've seen in the past. And B, is the cost of dark transparency, fragmentation and dark pools. Now, I think on the second topic, the regulators in Europe have been harder than the regulators in the US so far, but with one very important distinction. There is a common understanding in Europe that institutional block trading is necessary and should be exempted from further regulation. And that's part of the rules that have been suggested, meaning the volume caps does not impact large institutional block trading.
Having been very involved over a number of years, speaking to policy makers and regulators and politicians, both in the Continent, UK and in Brussels, we've got a very clear indication that they understand the need for an outlet for institutional block trading and we've also seen that in the proposed rules.
What needs to be understood when you talk about dark pools is we're not talking about one specific thing here. It's a range of different activities, with one thing in common: we don't post quotes on the lit market. On one hand you have institutional block trading, on the other hand you have small-sized algo trading, which is more suited for a lit book. And we group all of this together under one label to make it simple, but the world isn't always that simple.
Saurabh Srivastava, global head of electronic trading, Invesco (panel discussion)
One of the features of dark operating pools in general has been that ... buy-side trading desks, it's about not being able to tip your hand. You don't want to have information leakage and you don't want to display your order size out to the marketplace and signal to the marketplace that there's a large buyer or seller. And so, to the extent that dark pools help solve that problem, I think that's been one of the roles that dark pools have played.
Philippe Carré, global head, client connectivity, global trading, SunGard (AT interview)
We are seeing increasing client interest to go into emerging markets. The second thing is, emerging markets are changing very quickly. I think the accelerated pace of change means they are increasingly offering an experience similar to what our clients are used to doing on already developed markets. So our clients want to be able to experience the same set up as they are used to, so they might want a fast trading engine, they want access to market data, they want access to algorithmic trading, they want co-location facilities. And what's incredibly potent is to see how emerging markets have developed infrastructure in order to offer this.
Depending on where you are, it is still, I would say, a regional or a zonal strategy for a lot of funds. You will have very big funds who will invest worldwide, globally, multi-assets. But if I look at the shopping list of our clients over the last couple of years, it's been going, if you will, left to right. You have Latin America, and it's a picture which is beyond Brazil, which a lot of people have already invested into... Mexico, Chile, Columbia and Peru. It's going to be around central and Eastern Europe, Warsaw... It's also going to be about, increasingly, the Middle East and Africa. We see growing interest in people to going there, especially as a lot of exchanges in those regions, whether the Middle East and Africa, are equipping themselves with faster trading engines, with market data distribution channels, building co-location facilities... And then going further east, Southeast Asia, where we see a lot of interest for the ASEAN trading link and for those markets which are slowly but surely joining that area of the world.
You can see that, the second point is, there are a certain number of markets which are further down the road but which are joining those first pioneers. As technology becomes available, the adoption becomes much easier for newcomers, so we are beginning to see first client interest into places which I would say are very exotic, frontier markets. It's going to be around Bulgaria, it's going to be around Panama, Argentina, Uruguay, but also Nigeria, Ghana, Kenya, about Vietnam. So what we are seeing is that as people get more confident as to going into and doing their business in emerging markets in the same way they can do (in developed markets), they are increasingly vocal about the need for their brokers to be able to expand the coverage that they give access to those funds to.
Nicola Comninos, head of equity market development, Johannesburg Stock Exchange (AT interview)
We've seen quite a structural change, where our clients are just demanding faster access to the market. And there's more and more focus on technology improvements. So the old stock broker model has changed significantly to actually becoming far more of a technology house than a traditional brokerage house. And in response to that, we are now offering colocation services, which we are going to go live with at the beginning of May. It's a combination of both our brokers needing it for their own execution, but predominantly the broker needing it to service these new types of clients who have latency-sensitive strategies, and they would only employ those in a co-location environment.