How is execution analysis impacting your portfolio?

Published on 2 May 2018

Analysis of trade execution directly impacts your equity portfolio, making the use effective of transaction cost analysis (TCA) vital to investment managers.

We get an expert insight into how best to use analytics from Teresa Woodard, head of trading analytics, Global Trading, at T. Rowe Price; Tom Kennedy, head of Analytics services at Thomson Reuters; and Ian Mawdsley head of ETI Trading EMEA at Thomson Reuters.

Dan Barnes: Welcome to Trader TV Equities – your insights into the trading climate for professional stock investors. I’m Dan Barnes. In April’s show, we’re going to be asking how effectively equity execution can be measured and the extent to which end investors can use this information. I’ll be joined by Teresa Woodard, head of trading analytics for T. Rowe Price, to discuss the way execution is currently quantified, and from Ian Mawdsley, head of ETI Trading EMEA ,and Tom Kennedy, global head of Analytics Services at Thomson Reuters, to look at the options for improving the reports used internally and those seen by asset managers. Teresa, welcome to Trader TV.

Teresa Woodard: Thanks, Dan. Thanks for having me.

Dan Barnes: You’re welcome. To start with, I’d like to understand the extent to which transaction costs actually reflect best execution?

Teresa Woodard: Well, it’s pretty interesting, so here at T. Rowe Price, we believe that best execution is a process and transaction cost analysis or TCA, is simply one of the factors to evaluate how effective we are at executing that process. If we do it well and with care, we should be able to measure each part of the trading life cycle to measure how effective we were at doing what we intended to do in the market.

Dan Barnes: So which metrics might be included in both TCA and best execution analysis?

Teresa Woodard: Well, one of the things that I think is important when we’re evaluating TCA or our trading patterns is what our intent was. Some of the metrics that we use are, how close were we to the investment managers decision point or arrival price? Or did I have undue impact in the market? So did the stock or a reversion, did the stock revert back after I completed my execution? We also look at TCA as a way to measure our counterparty performance, so we may be looking at a VWAP or average price over the course of the order lifecycle.

Dan Barnes: And so it sounds like you use a lot of this in-house. What are the pros and cons of using an in-house measure vs an external measure, and do you use both or just one?

Teresa Woodard: Here at T. Rowe Price We have a comprehensive TCA effort, and so we use both third-party analysis, along with our in-house trading analytics team. One of the benefits of having someone doing this work in-house is that you can really provide trading insights directly to the traders on a realtime basis. And so being embedded with trading and trading management, we can really understand the nuances of how trading exists here at T. Rowe Price. But I think it’s very important and I’m sure our clients think it’s particularly important, that we also have a third-party analysis, an independent analysis of how we’re executing, how we compare to our peers, and if we’re investing as effectively as we can.

Dan Barnes: Sure. And so you mentioned the clients there, so obviously, the trading desk wants to understand the metrics around trading. Clients could also see performance. Who uses these measures, then, within the firm and externally?

Teresa Woodard: One of the best things or most interesting aspects of TCA now is everyone wants to see a piece of it. And so both internally and externally, we may be providing trading data to traders, trading management, some of our fund boards may need to see transaction cost analysis as a part of their governance effort. Obviously, our traders internally and again, as we’re evaluating our counterparties’ performance, our portfolio managers are interested to know how they are impacting how we execute in the market. And, of course, our T. Rowe Price investors; we have a best execution mandate. And as I mentioned earlier, TCA is just a factor in how we’re executing on behalf of our investors.

Dan Barnes: And so do the analytics that you represent to different groups have to have different characteristics reflecting their use?

Teresa Woodard: Yeah, it’s actually one of the most challenging parts of TCA is how do you present this very nuanced data to different stakeholders? And of course, everyone has a different horizon when they’re looking at transaction costs. As I mentioned, if our fund boards are looking at our trading costs, they may be looking at it from a governance perspective, trying to see how we’re impacting overall performance. Of course, our clients may be trying to compare our trading with maybe another asset manager. And of course, there are challenges on how you present that data, and we want to make sure that we present it in a way that’s clear and actually informative to the stakeholder.

Dan Barnes: Is there a danger, perhaps, that somebody might look at one TCA metric and then perhaps misunderstand what’s happening in the execution process?

Teresa Woodard: I think you hit on a very interesting point. And one of the things that we’re seeing in the marketplace today is folks become much more interested in TCA, is when you hear transaction costs, what you are really when you hear cost, folks automatically think that we may need to reduce those costs. And so one of the things that I always point out is that you can always have a lower transaction cost, if you’re buying things that are going down. And I don’t think that’s what our investors want us to do. They want to make sure that we’re we’re making good investments in the market, so it’s important to me to present that data in a way that is both informative and not misleading, and at the right cadence for the different stakeholders.

Dan Barnes: Presumably there are different types of data that you have to bring together and technically, that could potentially be quite challenging to integrate. Is it?

Teresa Woodard: It is quite a bit. One of the things that we’ve taken a lot of care with is on the data side, and making sure that we’re capturing all of the relevant information throughout the order lifecycle. From the moment that the portfolio manager is placing the order all the way out to where we begin to send our executions out into the market. One of the biggest challenges that we’ve seen over a number of years is just the amount of data that we’re capturing. But also at the same time, when we’re sending out a piece of data, but a part of the lifecycle, when we’ve sent an order to a counterparty and trying to evaluate what that broker may be doing on our behalf. That’s a challenge there, because there’s a whole other set of data that we may not have access to.

Dan Barnes: OK, is there a way to resolve that challenge?

Teresa Woodard: Well, there’s been a number of different efforts in the industry over the past few years to start to begin to capture that data. One of the things that we hope for T. Rowe Price is that there’ll be a normalization of that. I think that’s primed for a third party to come in to start evaluating some of that data from the different counterparties.

Dan Barnes: So, TCA informs the trading desk, it informs the portfolio managers, it informs the end investors as well. Are there then any best practices, you think that could be shared in the industry when communicating between the trading desk and the investment process to make sure this information is drawn into a holistic process?

Teresa Woodard: I’ve been in this space for for quite a while, and I remember a few years ago, some of us in the industry got together just to have some common language on how we talk about best execution and how we talk about trading costs. I think one of the most important things, if we’re talking about communication between the portfolio manager and the trader, is intent. What is the intent of that investment? And then we as an asset manager should be able to communicate that to our end investor as well. When we’re looking at TCA, one of the things that we try and and caveat is that comparing transaction costs results from one asset manager to another is very difficult, because you’re not looking at an apples to apples comparison, unless those portfolio managers are transacting in the same stock, on the same day, in the same size, then it’s very difficult to compare. So myself as an analyst, I need to be prepared to to share with our end investors; this is the intent of the portfolio manager, this is his trading style and, these are the subsequent costs.

Dan Barnes: I understand. So you have qualitative as well as quantitative information to share.

Teresa Woodard: Exactly. I mean, one of the things that I always say is TCA is a story and it’s a part of the best execution process. So, it’s how you share that story and make sure that you’re giving informative information to our stakeholders.

Dan Barnes: That’s great. Teresa, thank you very much.

Teresa Woodard: Thanks, Dan. I appreciate it.

Dan Barnes: Next, I’m joined by in Ian Mawdsley and Tom Kennedy of Thomson Reuters, to look at the models for improving execution and transaction reporting. Ian, Tom, welcome to Trader TV.

Ian Mawdsley: Thanks for inviting us.

Tom Kennedy: Great to be here.

Dan Barnes: What sort of analysis do you think end investors want to see from their asset managers?

Tom Kennedy: So I think investors are looking for cross asset transactional reports. These reports in the equity markets have been sent around looking at slippages, broker analysis. MiFID II has really kind of upped the ante. There’s not one factor, there’s about six factors. The firms have to look at price, cost, speed, likelihood of execution, so looking at multiple metrics and different characteristics is very important.

Dan Barnes: What sources of data might buy-side trading desks be using to assess TCA, so transaction cost analysis?

Tom Kennedy: So you obviously, you know the ingredients that you need to be able to do TCA, so you need market data and that’s coming from regulated markets, MTFs, OTFs, and APAs now for some of the off-venue trading.

Dan Barnes: And what’s APA?

Tom Kennedy: so an APA is an approved publication arrangement. So it’s mandated now with MiFID II, that all trading is reported, post-trate reported and their obligations to do pre-trade reporting also across certain assets if they’re deemed to be liquid.

Dan Barnes: OK, that’s very interesting. We talked about MiFID II. If we take the conversation a bit more globally, sort of looking at the US markets and the Asia Pacific markets, presumably then the quality of data, the amount of data one can get, it’s quite varied.

Tom Kennedy: Yeah, I think, you know, we look at from a US perspective, Dodd-Frank. You know, it mandated best execution requirements over quite a number of years. But what we’re beginning to see is regulators in Asia now implement or start to go to the market and consult on its own best execution initiatives. The monetary authority of Singapore has issued a consultation paper around this very matter. And also the HSBC in Hong Kong have issued a thematic review. And some of those regulations have merged with what is happening in Europe, so we’re seeing the supranational kind of response to best execution and especially investor protection.

Dan Barnes: OK, how does the trading desk interact with that data? How does it bring it together and how does it use it?

Ian Mawdsley: Well, I think there’s a couple of ways, obviously. So outside of standard reporting, as Tom was alluding to there, so things like best execution policy and also investor reporting. I think there’s also an element where the trading desk itself is looking to measure its own performance, particularly around trader compensation. If you’re trying to work out how valuable your trading desk is as opposed to outsourcing to a high touch desk, maybe at a broker? I think it’s essential to be able to see what sort of alpha they’re adding to the orders, if you like. And in terms of then the traders themselves using that data, they want to be able to interact with the baskets of orders they’re receiving from the various portfolio managers. And you can see there’s outliers there, certain orders need of extra attention, or certain orders need to be a bit more aggressive in terms of getting these things filled. So there’s certainly a couple of ways there that they’re interacting on a real-time basis. I think one of the newer polls that we’re seeing at the moment is around sort of more pre-trade analysis, if you like, market impacts. So when I’ve got a large orer to do, I’ve got a lot more complexity in terms of market structure these days, it’s not just a question of, ‘I’ve got a larger Vodafone order to do, I’m going to come to the NLC. There are something like 19 points of liquidity. And that’s not including all of the banks that offer in terms of their size as well.

Dan Barnes: That system has internalized.

Ian Mawdsley: Exctly, yeah. So that’s added a lot of complexity to the trading world. I think really the only way to be able to look at how you can best execute an order now, is to be able to do some kind of pre-trade analysis and then let the system come back and make some kind of suggestion on how you should be trading.

Dan Barnes: OK, that’s really interesting. Thank you very much. What are the technical challenges then in pulling this data together?

Tom Kennedy: You know, there’s a lot of data to be modeled, so the information modelling challenge is really complex. So there’s a lot of pressure on in-house IT to aggregate data. So the lack of a consolidated tape provider in places like Europe has made it very difficult to do best execution. So you have to go out and you have to source this data that has to be integrated into internal systems, being able to capture timestamps all the way past the trade lifecycle; from portfolio management decision all the way through to execution on an exchange or marketplace.

Dan Barnes: So you can understand the entire lifecycle of when an order was conceived, to when it was actually executed.

Tom Kennedy: Absolutely. And then there’s the transaction data that needs to be then persisted. And you need to connect to your brokers, our venues and persist that

Ian Mawdsley: We’re certainly seeing a lot more scrutiny in terms of the messages that are going backwards and forwards to the market as well. So I think MiFID I introduced the idea of looking at the venue, if you like, to tag 30 as it’s called within Fix. We’ve got a lot more granular since then and people are genuinely interested now in terms of understanding, ‘why did I go to a particular venue? Was it because there was a rebate paid to the broker, for instance?’ You know, ‘how much am I operating in the dark pool? More than I should be? Am I interacting with a firms SI, more than I want to be?’ So I think that, that level of granularity has definitely improved and the buy-side has got a little bit more valuable data to be able to play around with in terms of having that deeper analysis.

Tom Kennedy: And just to add, I think when you look at it from a customer service perspective, spreads are getting tighter. You know, there’s a challenge on getting that best execution. I think firms want to be able to give their clients, whether it’s the sell-side to the buy-side, the buy-side to investors, better insights and have more transparency.

Dan Barnes: Is TCA used the trading desk? Could you explain where Thomson Reuters sees TCA being used within the buy-side?

Tom Kennedy: So I think when we look at the metrics that TCA provides, I think those metrics can start to actually be used further up the Decision-Making chain. And so I think quantitative portfolio managers will take an active interest in understanding, when they’re constructing their portfolio or either when they’re rebalancing their portfolio, when they trade into a position. What is the likely outcome the FCA have written thematic reviews around best execution, highlighting the basis points improvements and how much money the industry could save if people took an active interest in optimizing that execution. So I think those metrics will be factored in. It’s not primarily a quantitative portfolio managers’ job to do this, and it still is a trading function. Yes, but I do believe that they will begin to take more of an active interest.

Dan Barnes: Okay, that’s great. Gentlemen, thank you very much.

Tom Kennedy: Thank you.

Teresa Woodard: Thank you.

Dan Barnes: I’d like to thank our guests, Teresa Woodard of T. Rowe Price, Ian Mawdsley and Tom Kennedy of Thomson Reuters. And, of course, you for watching. To see our insights into trading fixed income and crypto this month or to subscribe, go to TraderTV.NET.