Michele Patron, Senior Quantitative Trader, AllianceBernstein talks to Stuart Baden Powell, Head of European Electronic Trading Strategy, RBC Capital Markets about sell-side algorithms, efficient sourcing of liquidity, the need for pre- and post-trade transparency and high frequency trading.
Recently, there has been much discussion about improvements in sell-side agency algorithms: some would argue that the core ‘building blocks’ of scheduled and opportunistic algorithms remain virtually identical, built around the same underlying models; others would point to a more radical shift away from mere incremental enhancements. Regardless of view, what is clear is that the buy-side is taking control of its execution destiny. Concerns about a reduction in trust, together with insufficient transparency of internal operations from many brokers have all contributed towards the shift. Whilst some buy-side firms will purchase off-the-shelf, canned algorithms from the sell-side, marginally tweak them and call them their own, other institutional firms are taking matters more into their own hands. Quantitative trading has been of huge importance to hedge funds over recent years. However, there are now a few select long only houses moving to incorporate quantitative trading in-house and link this to their own fundamental trading strategies. AllianceBernstein would fall into that latter bracket – Michele, you have worked at both CQS and BGI and now run European Quantitative Trading at AllianceBernstein. Could you talk us through what you are up to?
I think that the discussion about how much buy-side firms should rely on the sell-side for trading research should have a definite answer by now: it is well-recognised that saving transaction costs represents an important source of alpha – even for medium turnover strategies. In addition, the wealth of information that buy-side firms have about their own flow cannot be achieved by counterparties, especially in multi-broker interaction scenarios: an accurate estimate of an alpha decay profile, which could be based on simple internal factors (i.e. order reason or PM strategy), will give the buy-side an important trading advantage.
At AllianceBernstein, we see our counterparties as partners, both in the high and low touch space. Within Quant Trading – which is globally headed by Dmitry Rakhlin – we continually try to analyse and customize execution algorithms, after we have had open discussions with our key counterparties. The ‘building blocks’ that you referred to earlier, are in principle easy to understand, and all the algo offerings out there can be bucketed into a few categories: the sell-side can offer us a hedge with smarter technology and expertise around execution tactics. There are some very smart options available in the market to minimize the latency arbitrage effect on client flows. A good solution can be achieved without tweaking the most relevant variable for this problem – system latency.
Two key tasks for a buy-side trading desk are: sourcing liquidity efficiently – especially for high AD V orders – and managing momentum – for low AD V. Having the ability to provide electronic solutions to address the latter, gives traders the opportunity to concentrate on the first task.
SBP: Michele, I would like to pick up on your point about order flow information. I agree that the buy-side is best placed to handle execution metrics when using several brokers. However, there are also many questions around how informed the buy-side are with regard to flow intricacies. For example FIX Tag 29 (agency or principal), Tag 30 (last execution venue) and Tag 851 (making or taking) are all highly valuable but are still not uniformly implemented across the industry. For me, implementation should be a priority to assist the buy-side’s own best execution requirements. Having said that, we also need to take tags further still; for instance, I would like to see where the order has been routed/shown before it is finally executed. This is highly relevant in the ‘dark’ – both internal and external broker firewall – particularly internally where external electronic liquidity provider (ELP) programs operate, as well as both internal and external market making. Client discussions reveal the need for more transparency for the buy-side around the internal operations of bank dark books, particularly as average execution size is so small and thus something of a contradiction to the theoretical benefits of dark. The buy-side should have a choice of trading services, but it is important to understand exactly what is being chosen – it is difficult to argue for choice without having clear definitions of what is being chosen.
Ultimately, banks are under pressure to tilt their business model in favour of those clients providing the most upside to the bank commercials; if those are high frequency trading (HFT) clients as external liquidity providers (ELPs) ‘externally market making’ in a broker crossing network/over-the-counter (BCN/OTC) pool, HFT ELPs receiving blind immediate or cancel orders (IOCs), or internal prop trading taking precedence over the institutional investor, then that is the sell-side firm’s competitive choice and right. A downside of this is that tracking, even in relation to those simple top level questions, requires resource – an area that buy-side trading desks continually struggle with. Michele, is there a commercially justifiable business equation for the buy-side to conduct analysis in not just the dark space but also in the execution quality period? How does AllianceBernstein handle this and what are your views on dark today?
MP: It boils down to transparency, both pre- and post-trade.
All the FIX tags you mentioned – even if not harmonized – contain useful information for the buy-side. However, this transparency is limited at fill level and clients have no visibility on what has been done with their order before that. Which venues have been visited? How long have they been in the broker’s own dark pool before moving somewhere else? How has it been split across different venues?
Clearly, this introduces an important trade-off from the data and operational perspective. Trades are only the tip of the iceberg and the amount of data to be communicated from sell-side to buyside would increase exponentially. From an operational perspective, reverse engineering routing logics for each single broker’s algorithm is prohibitively time consuming and possibly adding little value.
Understanding brokers’ algorithmic processes and how they fit into our investment style and trading objective are important. Even in the short term, undesired behaviours in the routing logic will be evident in sophisticated performance metrics.
We are open-minded when it comes to liquidity and we read the endless debates about HFT. However, from a pure trading ‘toxicity’ perspective, trading in an ‘institutional buy-side only’ pool – if your (buy-side) orders are generally smaller than other players – should be equally worrying, given that being filled immediately against larger sized orders could increase your chance of being adversely selected. This is also a concern for HFTs. In current market conditions, it is worth looking at every liquidity pot available, provided that outcomes (i.e. performance, reversion and fill rates) are constantly monitored.
SBP: Some buy-side participants would struggle with the extra data and so tag condensing would be necessary to cover the audit chain.
You mentioned the difficulties of being a small buy-side order in a block pool. There are few genuine buy-side only pools: the average execution size in internalization engines and dark multilateral trading facilities (MTFs) is sub-optimal for many. Frequently broker BCNs now have ELPs (HFTs) as market makers in the pool providing a wall of very small-sized committed risk for institutional orders, each execution is signal capture for the HFT ELP and a cost saving for the bank. If, as a buy-side, you have seen a marked decrease in execution size in your broker dark books and small size execution(s) up front, perhaps with adverse interval tick reversions – you should ask the question about HFT ELP in the pool – these executions provide ‘synthetic’ blocks, however they come with increased signal risk per child order thus cumulatively increasing costs across the total parent. As many buy-side firms have mentioned to me – “dark today is basically lit”.
BCNs are financially important to the sell-side for cost reduction and revenue generation. Tighter market volumes and commission pressure have forced brokers to change their BCN models. Those firms without a hedged MTF platform have likely wriggled during the organised trading facility (OTF) discussions by stretching underlying definitions of an OTF to facilitate BCN participants and liquidity into an OTF. This may have played an important role in the removal of the OTF category in equities.
It is encouraging to see MEP Markus Ferber aiming to support the end investor. RBC has long conducted extensive neutral in-depth research around HFT, discussing the issues with clients and regulators alike. One of the Ferber metrics is the message to trade ratio also recognized in the market abuse directive – the 250:1 number we believe should be tighter still to reduce the signal generation capabilities of HFT and volatility inducing stochastic resonations.
The final point I would like to address relates to those banks with extensive HFT client bases looking to offer HFT execution strategies to their institutional clients. Would banks really take ‘cutting edge’ strategies from one client base and pass them straight to another? Would this not just be a straight transfer of wealth? When HFT as a client provides heavy commercial benefits to banks on the corporate side, PB or post trade for example, the algorithms are likely to be distinctly second hand where decay is pronounced. Furthermore, high frequency traders would know how to trade against those algorithms. We know of HFT-type strategies posting at multiple venues where one execution is achieved and ‘fading’ or effectively making the remaining orders disappear – does this provide false perceptions of liquidity on order books?
Michele, do you have any final points you would like to make?
MP: The focus on HFT has attracted attention all across Europe. Only recently, Italy’s leading newspaper La Repubblica1 , published a special report on the topic.
As regards multiple posting activity there are two points to remember. Firstly, it is normal that when you want to strike a trade, you need to advertise your goods, and more advertising will increase your probability of closing the deal. When it comes to HFT, this latency arbitrage you described has been qualified by some in the market as ‘painting’ or ‘manipulating’ the book. At a consolidated level, what gets shown in the order book by HFT is not what he or she is willing to trade, given that being ‘hit’ on one venue can cause the cancellations of the orders resting in others. Nevertheless, it is true that orders are in the markets for a very short amount of time, and if someone starts interacting with them using faster or smarter technology, they will increase the probability of accessing all the liquidity pockets before the cancellations happen. However, the second equally important point is that in any economic activity, the marginal improvements in performance have to be worth the additional cost.
Ferber’s effort to qualify HFT activity is taking us in the right direction, given that “you cannot regulate someone you cannot define”: the maximum messages-to-trade ratio looks a better constraint than the initially discussed minimum resting period, which is likely to impact the current market microstructure at a deeper level. Given the low volumes currently being experienced in the European market, excessive penalties/constraint on an agent category which is estimated to trade 40% of the daily turnover2 could have undesirable effects. At the same time, despite many HFT strategies being ‘flat’ on a daily basis, they still carry an element of risk up to settlement date.
As it stands today, there is no consensus in academia or the industry as to whether HFT is indeed damaging or positive to institutional trading or the market.
This can be explained, of course, by the lack of high quality data available. I hope to see the emergence of more convincing research before regulators make drastic steps in any direction to address the HFT issue – either to facilitate HFT into the small and medium enterprise (SME) space or to curtail existing activity in liquid names – regulators and the buy-side need to be engaged with the right market structure teams and research into an area where there is a material knowledge gap across the whole industry.
 http://inchieste.repubblica.it/it/repubblica/rep-it/2012/04/20/news/numeri_inchiesta-33637371 – in Italian
 Bank of England Report, 2010,http://www.bankofengland.co.uk/publications/Documents/speeches/2010/speech445.pdf
Encouraging Enhanced Transparency
FIX Protocol Ltd (FPL) has updated its recommended best practices for execution venue reporting, a set of guidelines designed to provide greater transparency for buy-side clients about where and how their orders are executed by their brokers.
The guidelines were originally produced to address the need for greater consistency and clarity in execution reporting. The revisions provide greater insight for firms trading on European equity markets.
The guidelines focus on requesting sell side firms to ‘tag’ their trades with specific details using FIX. These ‘tags’ are designed to provide the following information in a consistent format:
• Venue or exchange on which the order was executed
• Unique identification of markets, including market segments where it is appropriate to distinguish between lit and dark order books
• Whether the order was executed as a result of having traded passively by posting liquidity, or aggressively by taking liquidity
• Whether the trades were executed on an agency or principal basis
To view the guidelines please visit: www.fixprotocol. org/documents/6685/ FPLBuysideExecVenue_May2012_1.pdf
Commenting on the guidelines, Guy Sears, Director at the Investment Management Association said,
“We welcome all industry initiatives designed to provide the buy side with greater transparency over how and where their orders were executed. As the equity markets become ever more complex, algorithmic trading strategies are increasingly used by brokers in the quest for best execution. Buyside firms need to understand how their liquidity interacts in the markets. I hope that firms which adopt these best practices will find them of real value.”
Commenting on the enhanced guidelines, Brian Lees, Co-Chair of the FPL Buy-side Execution Venue Working Group and Manager of Application Development at The Capital Group Companies stated,
“The buy-side members of FPL have expressed a clear appetite for greater clarity and consistency in reporting from their brokers. Working with members in different markets globally we have been able to update the guidelines so they deliver even greater insight to the buy-side community.”
Jim Kaye, Co-Chair of the FPL EMEA Business Practices Subcommittee, and Director, Product Development, European Execution Services, BofA Merrill Lynch said,
“As a member of FPL, BofA Merrill Lynch, has been actively involved in developing the guidelines so they can be used more effectively for trading in markets such as Europe, and we feel the increased insight they will present to the buy-side will offer tremendous benefit. Not only will this enable firms like ours to have more strategic and compelling discussion about trading strategies with our clients, but also help us adapt our algorithms and trading strategies accordingly. Greater knowledge and understanding can only serve us all well, particularly in these times of lean liquidity.”