The trouble with size

Block trading is hardly new and has never really been conducted in the public gaze. But as the keyboard replaces the telephone, is the market in Europe keeping up? Richard Schwartz

A moneyless man goes fast through the marketplace, says an old French proverb. The problem facing a trader with huge orders to fill is the opposite: how to make it through the market and get your business done without being spotted and harassed.

Not too long ago, when exchanges were the dominant location for liquidity, the notion of a block trade was simple: any trade over 10,000 shares. While still broadly used, few people will accept that definition today without caveats.

“The standard market definition of block trading is still anything more than 10,000 shares,” acknowledges Gavin Little-Gill, senior vice president, product management and strategy, Linedata. “Any such trade is going to require a bit of effort; but how much effort depends on where you’re trading it and what the name is.” A better yardstick might be percentage of daily volume, adds Alan Price, head of front office sales at Linedata. “That would take into account liquidity of the individual stock. For example, 10,000 Google would take me about five seconds to work.”

Clearly the definition varies depending on whether the trade involves a very liquid stock or a small cap, says Rustam Lam, executive director, Townsend Analytics (TAL). Overlaying that, he continues, is a world in which there are different pools of liquidity from exchanges through MTFs and internal dark pools to crossing networks, each with their own peculiarities and constraints. There is also still the opportunity to send a block trade electronically to a sales trading desk. “With all of that in mind,” says Lam, “a block trade in the electronic environment really means that you start out with the intention of completing the trade as a block, but the actual execution of that block may take place in a number of venues.”

What is the likelihood today that a large order can be shifted from the blotter in one go other than by handing it off to a broker’s cash desk? Expectations have been talked up, in the US at least, by the proliferation of so-called dark pools, most of which claim to handle size without alerting the rest of the marketplace to an individual trader’s presence. Yet, says Harrell Smith, co-head, product strategy group, Portware, the size available in those pools is generally not of the order of magnitude required for today’s large blocks. “Many tout their ability to handle continuous crosses,” says Smith, “but, for real block trading there are far fewer than the figure of 40 or so dark pools and alternative liquidity venues that is generally cited.”

The term ‘dark pools’ has been widely used and is misunderstood, he contends. “Every broker out there claims to have some dark pool or internal crossing facility, but those are for all intents and purposes, simply links to the broker dealer’s internalisation engines. Only a handful of systems offer full block trading, with speed, anonymity and price improvement.” For Seth Merrin, CEO, Liquidnet, meanwhile, the term ‘dark pools’ simply brings a negative connotation to a market solution to fragmentation, which in other commercial domains, is regarded as healthy competition. Rather than ‘dark pools’, he believes, such liquidity venues should promote themselves as offering ‘quantity discovery’.

New paradigm, new problems
While differences exist as to the nature of the beast, there is general agreement that in today’s marketplace, it is difficult to feed. There is also a degree of consensus on why quantity is scarce. “The average size of trades on exchange order books has diminished,” says Roland Bellegarde, head of European cash markets and member of the NYSE Euronext Management Committee. At the same time, consolidation among both the buy-side and sell-side is producing larger and larger orders for traders to handle. Today, says Bellegarde, a significant portion of those large blocks ends up on order books being sliced and diced by the brokers and banks.

 With one eye on developments in the US market and another on MiFID, participants in European equity markets are looking with eager anticipation at the multiple initiatives that promise to alleviate the strain of pushing large orders though the market. These include, inter alia, Project Turquoise, which is finally gathering pace with the unveiling of a dedicated management team; the new joint venture between virt-x and NYFIX Euro Millennium aimed at bluechip Swiss equities; and Smartpool, a new pan- European block trading platform due to go live in mid- 2008 with the backing of NYSE Euronext, HSBC and BNP Paribas.

“We had long discussions with the main brokers and banks and pointed out what we saw as a paradox,” says Bellegarde. “We told them: ‘You would like to trade large blocks, but you end up with small trades.’ When we asked if they would find a large block trading platform useful, the answer was, ‘Yes, as long as it helps protect size and price’.”

The purpose of order slicing on the public exchange is indeed to protect size and price, says Bellegarde, “but if you can provide a similar service while still allowing large blocks to be traded electronically, you really have a service that adds value.” A large, dark blocktrading platform operating on a pan-European scale, but still leveraging existing connectivity to exchanges brings a time-to-market advantage, Bellegarde believes.

For Linedata’s Little-Gill, however, the question is not so much about where the liquidity resides, but about how you get there. “We have over 250 different liquidity destinations today,” he says. “At the end of the day, whether I’m sending an order to the Goldman Sachs cash desk, to a Credit Suisse algorithm or one of the block crossing networks is largely irrelevant,” he argues. “I still need to trade the order, I still need to pay attention to my broker list, and to allocate the executions back to the underlying accounts. The question then is how to access efficiently each of those components of the trading environment.”

The technology challenge
If this is correct, then the problem for buy-side traders is less one of strategy in dealing with multiple liquidity venues and more a question of appropriate technology and software. “The array of tools available to a buy-side trader now to try and trade size is wide and we find clients using a combination of tools,” says Paul Scott, director, FIXCITY.

 Some of these tools are not new, but have rather been adapted to meet current market conditions. Such, for example, is the indication of interest (IOI). “In the present climate, the IOI becomes a very good way at the start of the trading process for the buy-side trader to acquire information with no market impact as to where there might be an actual other side to their trade,” says Scott. Traders can look first at the advertised trade history to see which brokers have been most active in the stock concerned and then couple that with the IOI messages to see whose indicating that they may have liquidity. “The buy-side has not released any information at all in finding that out,” he suggests.

The indication of interest is only that though. It still leaves the trader the job of completing the order. Nevertheless, refinements in the IOI process have made it easier to distinguish exploratory forays from real trading opportunities. “Brokers are building more technology on their side to better distribute their IOIs,” says Scott. “We’ve seen a marked change in the market away from the days of fishing when IOIs were broadcast to the street and people didn’t feel there was anything behind them.” The use of FIX and advances in automation on the broker side, have enabled more targeted IOIs. Brokers are sending IOIs to particular buy-side firms that they believe might be interested in them. “They’ll look at trading patterns and try to attract the most pertinent parties first,” says Scott. They may therefore show slightly more of an order to a larger house with more funds under management. “They will tailor the IOI to the client they’re sending it to.”

Logical order “A lot of our clients will look at the IOIs first,” says Scott. “That could lead to a direct bit of business where there is a natural match to a negotiation around quantity.” Assuming that a certain portion of a large order can be completed in that way, where is the logical place to go next: one of the big four block matching services – Liquidnet, Posit, Pipeline or Millennium; a broker EMS or internal pool; a third party ATS or an exchangebacked platform like SmartPool?

 “There are a number of ways of analysing what type of venue would be most suitable to a particular order,” says Smith, One is obviously through integrating pretrade analytics into the EMS. “Any sophisticated trader or portfolio manager will know that small cap stocks are going to be more difficult to trade than some like the IBMs and Intels, but there is a large grey area in between,” he says. “Pre-trade analytics help firms gauge market impact and how difficult an order actually is.”

 Finding as much natural liquidity as possible at the start of the trading process would seem to be a good principle, but while a block crossing network may help to shift a sizeable portion of an order by locating a natural other side, it may also leave a residual quantity that is difficult to trade without resort to a broker algorithm.

 “Some venues have natural liquidity, some don’t,” says Little-Gill. A broker EMS may attempt an internal cross and then head out to a range of other liquidity destinations to which it connects. But routing out from any one liquidity pool may also be a hit and miss affair, since success may depend on pinging a pool at the moment when the liquidity one trader is seeking is passing through on a similar search.

The primacy of technology It is at this point that technology becomes paramount, since any of the panoply of trading options, if used in isolation, may result in missed opportunities – displayed or not – elsewhere in the market. Since FIX has simplified the process of connecting to new liquidity destinations, the question then becomes how the trading system interacts with the various trading destinations.

 For the trader, Little-Gill suggests, the key factor is how to access all available liquidity from the trading blotter. While connectivity to multiple venues is expected, the manner of access still leaves room for differentiation.

“The way people access liquidity today is through committed orders,” he explains. “So if I send an order up to Redi, Passport or any of the other broker EMSs, I’m trading in that particular destination; it’s a committed order and is no longer on my trading blotter.” What could then happen, at least hypothetically, is that a trader with 50,000 of a particular name on the blotter decides to put 10,000 through a particular broker EMS. “All of a sudden, Liquidnet comes in and scoops the 40,000 remaining, but it doesn’t get the 10,000 because the 10,000 is no longer there, it’s in the broker EMS being worked, ” says Little- Gill, “I may have 2,000 executed there, but that leaves 8,000 outstanding. There may be plenty more available at Liquidnet, but Liquidnet can’t see them.”

 Longview, the Linedata trade order management system, now treats such orders as uncommitted. “We continue to reflect the 50,000 shares on the blotter and we synchronise,” says Little-Gill. “That allows the trader to move, say, 10,000 shares up to their favourite broker EMS, but as soon as they’ve traded 2,000, that’s synchronised with the blotter, which now shows 48,000.” If Liquidnet scoops up the 48,000, the system will zero out the order on the broker EMS. This refinement allows a trader to run with an order in different venues simultaneously, including other block trading platforms such as Pipeline.

 Most desktop systems are now aware of the need to minimise the risk of missed opportunities and are introducing functionality accordingly. Portware allows for the integration of algorithms that act as smart order routers for block trading venues. “These will source liquidity from a number of block trading venues to see where to execute the entire order if there’s a natural other side,” says Smith.

 NYFIX Millennium meanwhile has added Millennium Plus, a feature allowing participants either on a pass-through or resident basis to generate a liquidity alert to a group of passive liquidity partners on the system. “These partners – other dark systems, internal pools, and institutions – systematically receive the alert and instantly respond with any standing orders they have,” explains Brian Carr, CEO, NYFIX Millennium.

Crossing the pond The US market provides a preview of some of the techniques that may help traders in Europe handle large blocks in markets with diminishing onexchange trade size, though conditions are never directly comparable between markets.

 In addition to partnering with virt-x, NYFIX Euro Millennium is set to launch in Q1 2008 as an MTF, bringing the NYFIX functionality to the European trading arena.

 Also crossing the pond is Townsend Analytics’ virtual Aggregated Book which initially focuses on major European stocks and which will be built out over time. For Lam, the role of TAL and its peers who provide the order and execution management technology to the buy-side desks is “to re-aggregate access to those fragmented liquidity venues.”

 An aggregated view allows the trader to get an idea of the liquidity dynamics between some of the execution points, says Lam. “When you look at trading a block, you want to have access to as much market information as you can,” he says. “You want to be able to access multiple execution points in that stock and to have the tools such as randomised reserve orders that will help you manage any market impact associated with that. The RealTick aggregated book shows not just NBBO, but every executable bid and offer.”

 In some respects, says Lam, the coming fragmentation will simply require traders to make greater use of functionality that already resides in the systems they use. “We are starting to see people taking advantage of some of the tools that had been available previously and probably hadn’t been used as much as they could have been,” he observes. He points to the ability to randomise a reserve order to minimise market impact or the idea of discretionary orders to link trades together so that one triggers the other.

 For Portware’s Smith, Europe is certainly heading in the same direction as the US in terms of the multiplicity of potential liquidity venues as a result of the removal of concentration rules under MiFID. However, he notes, Europe was significantly more advanced with regard to electronic trading and had a head start over the US in moving away from an open outcry model. “I doubt you will have the same proliferation of competing ECNs and block trading venues,” he comments. “That growth addressed a need that the specialists on the floor were not addressing.”

 An instructive example is Canada, which in some ways resembles a number of the continental European markets in levels of liquidity concentration. “In the Canadian market, there was only really one way to trade size until recently and that was picking up a phone and calling a dealer,” says Chris Jackson, executive vice president, sales and marketing at Perimeter Financial, founders of BlockBook, a consolidated pool of Canadian, TSX-listed, block liquidity. He describes Canada as “like the US in 1994,” with one big dominant exchange. “The first thing we learned was that there was no way in Canada that we could be an exclusive club, whether buyside only, sell-side only or hedge funds,” he notes.

 “If someone is looking to trade size in Canada,” Jackson comments, “they will still often go over the counter to the broker, to Liquidnet in some cases, or BlockBook. Those are really the only three places that you can trade for size. The fourth would be showing your order in a public marketplace and having a feeding frenzy.”

 Similarly, says Carr, “We are leveraging the aspects that you’d assume – the technology the architecture, but we’re not just taking Millennium to Europe and plugging it in.” One of the big differences, he notes, is that the bond between buyside and broker is still strong in Europe and, he adds, “We’re trying not to change their workflow.”

 NYSE Euronext has similarly taken differences in regulation and structure into account in its involvement with SmartPool on one side of the Atlantic and BIDS on the other. “The regulatory environment in the US is different,” Bellegarde points out. “BIDS is becoming a member of the exchange to be able to trade blocks. The model is different from SmartPool. One is compliant with the US regulation of providing a facility for trading blocks on the floor. The other is consortium- based, using our own infrastructure, not the facility of another market.” Nevertheless, he insists, the needs are the same. “We are designing services that meet the regulatory needs on both sides of the pond, but there won’t be a single block package across the Atlantic at this stage.”