The subject of how data is used to conduct Transaction Cost Analysis (TCA) formed part of a lively debate at Profit & Loss Forex Network Chicago. Galen Stops moderated.
Paul Aston, the CEO of Tixall Global Advisors, proposed the motion that “TCA Is Just a Morality Carwash”, effectively arguing that many buy side firms are simply handing over their fiduciary duty to ensure best execution for their investors to their sell side counterparts, while using TCA to justify the trading decisions that they make as a result.
Opposing the motion was Isaac Lieberman, CEO of Aston Capital Management, who formed his argument around the proposition that market participants who are trading risk do so against a set of metrics by which their success is measured and therefore they are conducting TCA to show performance against these metrics.
“The morality carwash comes down to legitimising activity. If you look at TCA led by the sell side, which is producing the prices in the first place, and you have a TCA process that simply checks a box and says, ‘Did I get a good price?’ and your liquidity providers says ‘Yes, you got a good price’ that is really just a compliance check box that is carwashing the morality of whether you really got best execution.
“If the best execution and the transaction cost analysis you’re performing is not fully ingrained in your particular investment process, then you end up with a box ticking exercise and it doesn’t get you anywhere in terms of what truly is best execution, it’s just a validation of pricing that you’re getting from liquidity providers, which is validated by the liquidity providers themselves.” – Paul Aston
“Anyone who is in the business of trading knows that there’s no whitewashing of the key metrics of touching risk. You can forget about TCA as morality car washing because a person who did that would not be able to do their job. What we really need to understand is why TCA is being used and why it has become so popular in this day and age.
“The reason is that in the past couple of years the availability of data, and the availability of tools to analyse data, to sort data and to tag key metrics around your business execution have become more readily and commercially available. When you look at TCA and why it’s so popular now, it is because it means that not only can we look at the metrics of performance risk, but also the key metrics of entering and exiting risk, and this allows firms to look for new ways to improve their business model and boost the returns of their trading activity.” – Isaac Lieberman
Paul argued that not enough buy side firms are ingraining TCA into their decision-making processes and then modifying their behaviour or altering their strategies as a result of that TCA. Instead, he said that many buy side firms are abdicating their fiduciary responsibilities by letting their sell side counterparties provide validation that they achieved best execution by using their TCA.
Although Paul said that most sell side institutions will try and offer the best TCA services that they can, there is an inherent conflict of interest involved in this arrangement.
“You can’t act as a principal and an agent and a fiduciary all at the same time,” he commented.
Paul added that, even if the TCA methodology being used by sell side firms is very accurate and quantitatively sound, if it doesn’t form part of the decision-making process, then it isn’t actually of any value.
“It just ends up being a compliance check box and again, it’s just a validation that prices were at a certain point, that this was the best price that I could have gotten at that time in the competitive panoply of where I could have traded, that it was top of book or close enough, and therefore that I got best
execution when in reality, that’s just the best price of liquidity,” he explained.
Isaac responded by stating that the best way to conduct TCA is for firms to have their own infrastructure and data to ensure that it is representative of actual liquidity, spread, credit and market access that is available to them.
But, he pointed out, a lot of buy side firms might not have the tools, infrastructure, resources or talent to run an internal quantitative desk that can do this effectively. Therefore, said Isaac, these buy side firms naturally look to their sell side partners, with whom they have long-term established relationships with, to provide this TCA service.
“If you’re working with a sell side liquidity partner that has a large market footprint, a global footprint, and they run internalisation and very big metrics, then the TCA from them is going to be pretty accurate in terms of what your best execution costs would be on both a pre- and post-trade basis and they’re going to provide this because of the longevity of that relationship,” he added.
Although Paul acknowledged that the metrics used by TCA providers has improved, and that third party TCA can be helpful because it can allow a bigger cross-section of liquidity providers to be analysed, he continued to criticise what he described as the “self-fulfilling” and “capricious” benchmark aspect of TCA.
To explain this comment in more detail, Paul presented a scenario in which a firm has a choice between using an arrival price and asking for risk transfer for immediacy or using a TWAP to execute that order over an hour-long period. If the firm was buying as the market was rising, traditional TCA might show that the TWAP pricing did very well because it was filled competitively within a micro-pip of top of book, making it look like the firm got best execution. But in actual fact, said Paul, the risk transfer price might have been much lower than the amount of drift that the TWAP produced, meaning that this would have actually been a better execution choice.
“What is the benchmark by which you define ‘best’? I’m troubled because it seems like we keep circling back to where is the bid/offer spread at a certain point in time and whether or not you’re getting the most competitive spread. That is important, but it seems to me to be a criteria that is very market maker driven, it is not liquidity consumer driven,” said Paul.
Isaac countered that Paul was getting bogged down in “theoretical fictions” regarding the TCA process.
“Everyone who touches risk, anyone who actually executes risk, knows that each trade has its own individual footprint. It’s almost like stars or snowflakes, they look the same, but when you get into the nuts and bolts of it, there’s individual characteristics. And when you go to trade or execute, many different moving parts are happening which leads you to make judgments to get the best execution and this is why the data and the systems and the relationships and understanding how you’re trading is the most important thing,” he commented.
Paul agreed that proper TCA is a very personalised process, but stated that this doesn’t change the fact that buy side firms are abdicating their moral responsibility to their investors by enabling other firms to decide how TCA should be conducted.
Giving an analogy, Paul pointed out that he could look to Vogue or GQ magazines as arbiters of style and, subsequently, go to a high-end fashion store and buy lots of very expensive clothes. However, he said that if he were to do this, it would likely result in ridicule from friends and family because they know that this wouldn’t fit his style.
“What I’ve done there is punted my decision to some other compass telling me what the best execution really is and that compass is pointing towards factors that, yes, it is best execution in a localised micro sense, but in the overall style of who you are and what you’re about and what your mandate is, I still think that it has to be customised. TCA can be a morality carwash if you don’t take it and incorporate it into your process and add it to a more holistic sense of what you’re trying to do,” argued Paul.
By contrast, Isaac claimed that, rather than using TCA to justify their execution decisions or punting away their fiduciary responsibilities, buy side firms are using the data analytics associated with TCA to optimise their businesses and create better practices. He said that if firms do this it will be reflected in their P&L, which Isaac insisted is ultimately the “judge and jury” of TCA.
Isaac: Why do we want to use TCA, how did it come about? Because data is available and systems are available for analysing data which show us a clear metric for what our transaction costs are. But the end-all is not to know what it’s costing, the end-all is to use that data to grow and optimise your business process and your execution process, which is where the customisation comes in. It can’t be a morality carwash because the tool is being used to grow and optimise your business.
Paul: I would agree if it’s used that way, but what happens is that in a lot of cases it isn’t. What happens is that the TCA process becomes a surrogate for other responsibilities or actions. When I see practitioners using TCA, they’re using it in lieu of risk taking and skilled decisions that need to be made, not in addition to them. That’s where the carwashing comes into play.
Isaac: That doesn’t add up, because if someone is in the business of trading and taking risk, then they need to have clear metrics on their performance or they wouldn’t exist, they’d get carried out. Simply using TCA to show your fiduciary responsibility means that you’re not running an optimal business and if you’re not running an optimal business and you’re trading risk and you’re running a P&L, you won’t survive.
Paul: That’s not true, for example, you have something like 11 or 12 trillion being managed at the WMR Fix, which has a systematic bias in terms of the price spiking anywhere from 10 to 40 basis points on average on any particular day. You could trade a TWAP or a VWAP through the bulk period of the day and generally outperform the WMR Fix, and yet you see a lot of the index mandates not doing that.
So when you’re talking about an active trader who is in the job of generating P&L from active FX trading I would agree with you, but there’s a lot of people using TCA as a check box because it is not part of their core decision making, they’re making a decision on equities or fixed income and saying “FX is kind of complicated, let me just get the best price, let me just get a benchmark, and as long as the TCA procedure says that was good execution, that’s good enough for me”.
Isaac: That’s very historical because what’s happening today, especially with MiFID II coming out, is that everybody is accountable and everybody has to explain themselves, not through a report, but through facts on the ground and actions on the desk. And it doesn’t stop there, it’s a constant business of optimisation and growth through accountability and better practices.
In terms of WMR Fixings and execution, I don’t see how there could be any desk, no matter how institutional, no matter how removed from the core alpha of the execution model, that can simply do an old-style execution, get a report, check a box and walk away. That’s not happening, it’s all about the post-TCA process of optimisation and bettering your business execution process.
Paul: MiFID II is the most excruciating tick-box exercise ever concocted. Under MiFID II procedures you have to ping the market prior to trading – which is going to signal – but then after you’ve traded you have to submit the transaction to this massive ocean of a database that is going to become unruly.
Isaac: But that’s the downstream stuff, that’s the byproduct. The core of MiFID II is best execution, showing that even on complex products like options and swaps, that you had a variation of value that you went and took a risk action on. Anyone who is in the business of trading and has capital and has performance metrics wants to have multiple liquidity sources to get best execution. The fact that ultimately it goes downstream and gets reported and goes to compliance, that’s just the way to keep it all accountable. But the core of MiFID II is to do the right thing and get best execution for your investors.
Paul: But it’s on a transactional basis. I don’t think that best execution has anything to do with a transactional basis. Transactions accumulate to an end result, that’s where you enter and exit, but in a lot of mandates you see where they’re instructed to trade at a certain price, they’re not compensated or rewarded for deviating from that price.
So when you use TCA, it becomes very revealing on an annual basis that your NAV is off, let’s say 200 or 300 basis points, because you could have traded a very naïve passive alternative. We’ve proved with analysis that there’s alternatives that you can use to re-pocket massive amounts of alpha that are in some cases three times what the underlying asset markets are delivering.
Isaac: But that’s an execution strategy, that’s like saying “Cable is illiquid in Asia so I’m not going to trade it, I’m going to wait until the London opens”, but by the time the London opens it’s 150 points higher. Guess what? I’d rather trade Cable in Asia, pay the spread, go through the spread, get my execution and be in the money by the time London opens. You have to understand how markets work, this is a dynamic process.
Paul: Of course, but you’re talking from a skilled FX practitioner’s point of view. I think where the whitewashing of TCA is coming in as a check box is for someone for whom FX is not a core competency, it’s a peripheral and ancillary activity. And to be continually trading on a price that can be shown as sub-optimal, and use the TCA process to say that it was best execution when it was actually just best execution at that time under the transactional approach, that to me is misleading to the asset owner. That’s the morality carwash.
Isaac: But TCA is educating those folks who are not in the front lines to make the practices of those that are better. So it’s a good thing because it’s teaching them how to be smarter, not in the reporting, but in the actual acquisition of liquidity.
Paul: But that’s what I’m saying, I don’t think that the actual asset owner is being educated by the TCA process right now.
At the risk of sounding like this is a cop-out, the answer almost certainly lies somewhere between the two positions being advocated here. On the one hand, anecdotal evidence strongly suggests that TCA is being viewed by market participants as a way to appease investors, internal committees or justify their FX execution in some cases.
Although Isaac’s confidence in market forces to weed out inefficient players is probably justified amongst FX specialists, firms for whom FX is merely a by-product of their larger business are unlikely to suffer significant consequences of ongoing poor execution in this market, within reason, obviously.
And in private conversations with some bankers it becomes evident that their desire to offer third party TCA to their clients is driven by a desire to offer a compliance and audit function rather than a genuine concern that this TCA will turn up something that their own lacks.
Certainly the wider market thinks that TCA is, at the very least, in danger of becoming a compliance function. In a recent survey by Profit & Loss, 72% of respondents said that TCA is in danger of becoming a tick box exercise, in comparison to 19% that said it isn’t and 9% that said they are unsure whether it is or not.
And yet on the other hand, to say that TCA is just a morality carwash certainly seems to overstep the mark. There are clearly numerous firms that use TCA in a serious and effective manner in order to, as Isaac points out, optimise business and execution processes.