Episode Transcript
<p><!--block-->JC: Hello and welcome to Ahead of the Curve, the podcast from AcadiaSoft where we take time to get under the skin of the margin and collateral industry, to dig deep and present topical perspectives and insights on this hugely important sector. Now, here’s a question: what’s your view on Data Standardisation? How much time are you putting aside to think about it? And what could the impact be for your company and our industry more widely? To put it simply, the standardisation of the way the Over-The-Counter bi-lateral derivatives industry records and digitizes trade details is hugely important. But the process of achieving those methods of standardisation are anything but simple. In this edition of Ahead of the Curve, we’re going to examine what Data Standardisation really means in practice. And we’ll ask the crunch question: is our industry ready? To do this, I’m joined by an expert panel of guests. Donal Gallagher is Chief Executive of Quaternion Risk Management. Fred Dassori is Chief Product Officer with AcadiaSoft. And Richard Barton is Head of Product Management, also with AcadiaSoft. Gentlemen, welcome. Now, before we go any further tell me a bit about yourselves and your roles at Quaternion and AcadiaSoft. Donal, I’ll come to you first.</p><p><!--block-->DG: Thanks Jon. I’m Donal Gallagher, I’m one of the co-founders and the CEO of Quaternion, and for the last 10 years at Quaternion we’ve been helping clients to build solutions and optimise the complex regulations that came into force after the financial crisis, so things like the Uncleared Margin Rules, the LIBOR Transition, the new capital rules, the Basel Accords, and at Quaternion we’re very strong advocates for transparency and open source solutions and of course part of that is data standards, so delighted to be here today.</p><p><!--block-->JC: Well it’s great to have you on board, Donal. And Fred, Fred Dassori, if I could come to you next.</p><p><!--block-->FD: Thanks Jon. I’ve been at AcadiaSoft a little over 3 years, based in New York, and as you mentioned I’m Chief Product Officer responsible for product, for the development of AcadiaSoft’s community, as well as corporate development, which includes our partnership with Quaternion. Those areas, I think the one that’s potentially unique to AcadiaSoft is the idea of community development, where we have a dedicated team focused on producing services for the industry, rather than just for individual clients, and that fits well with today’s topic.</p><p><!--block-->JC: It certainly does. Thank you very much indeed, Fred. And Richard Barton, finally to you.</p><p><!--block-->RB: Yes, I’m Richard Barton and I head up the product management team. We basically take the strategic ideas, strategic direction, and are responsible for delivering that, so looking at those ideas and working out how to implement them. Part of the responsibility for my team is to understand what industry standards exist, determine if there’s a need for industry standards, and we contribute into existing standardisation processes and also initiate them as well.</p><p><!--block-->JC: Thank you very much indeed as well, Richard. So the three of you have got a lot of insight and knowledge on the topic that we want to discuss and talk about on the podcast, so let’s get to the issue at hand: Data Standardisation. Donal, talk us through this. The origins of much of the move towards standardisation lie in the great financial crisis of 2007/08. Why does standardisation matter? </p><p><!--block-->DG: That’s a great question. I think standardisation is part of the solution to the problem that was the great financial crisis and one of the reasons is that it helps us avoid some of the problems that came up. Let’s think a little bit about how we got here. If you remember, there was huge innovation in derivatives in the run up to 2007 and a lot of it had to do with disaggregating risk and being able to move it around the financial system using instruments, and that’s a tremendously beneficial thing from a risk management and balance-sheet management perspective for the banking industry. There were also some new instrument types and new asset classes, particularly around credit but not exclusively, and also around interest rate and equity, and what this allowed financial institutions to do was disaggregate risk and move it from one place to another where it was probably best managed, but it also gave rise to increasing complexity. I think everyone knows about CDOs, the famous Collateralised Debt Obligations, that became one of the proximate causes of the financial crisis of 2007, but that’s just one example. I’ve heard a number of something like 254 different derivative types, but in fact nobody actually knows because there is no comprehensive naming system for over-the-counter derivatives, which is another example of a lack of standardisation. So we found increasing complexity, layers of complexity, options sitting on top of multi-asset class and so on and remember distinctly being involved with a financial institution, which was actually very good at managing risk, but for CDOs in particular, in which they had a large position. They had three separate systems, three different ways of representing them and three different ways of pricing them, so even within a single financial institution let alone across the system there was no standardisation. So it really was a question of innovation running ahead of where the industry was. And this led to surprises; I think people failed to anticipate quite a few things, and that became very clear in the financial system in the great crisis. One of the big failures was the failure to anticipate the huge amount of counter-party risk that was being built up by the risk transfer. So if I transfer risk from my institution to another institution, in effect I’m transferring an obligation, and if that obligation gets very very large that has a significant impact on my balance sheet and potentially my solvency and ultimately the entire financial system, which is what happened as institutions started to fail in 2007 and brought us right to the edge of collapse, I would say, if it hadn’t been for the interventions of the authorities.</p><p><!--block-->JC: So the shock of the shock of that then, Donal, the shock of all of that, as you say the CDOs, the counter-party risk that was clearly at play there, that’s taken us on this journey towards a call for greater standardisation: that’s what you’re saying?</p><p><!--block-->DG: Very much so, because there was a big wave of regulatory response after that. If you think about some of the things that happened, the move to clearing and the uncleared margin rules for bilateral derivatives that weren’t going to be cleared; things like the SR 11-7 and TRIM, which were around standards for risk modelling, and the LIBOR Transition, which brings transparency and certainty to benchmark indices. They were all regulatory responses and standardisations, in a way. But there are still barriers out there, and we think that the standardisation in the data representation of derivatives is probably one of the last barriers to be overcome, and it really will allow a new wave of interoperability sharing, cost reduction and ultimately safety in the industry, which I think is the most important point. </p><p><!--block-->JC: And Fred, just pursue those thoughts a bit more for us, would you? Why standardisation is so important, why it’s essentially a good thing for our industry.</p><p><!--block-->FD: I’ll start with maybe an anecdote to show the other side. Before joining AcadiaSoft I worked in a fixed-income electronic trading business, and counter intuitively we felt in that business like the complexity of the market was really a competitive advantage for us, that the fact that we had the in-depth market knowledge, unique market knowledge, and the resources to exploit it made us able to operate in an environment that really kept others out, right, that stopped other firms from being able to compete effectively with us. And that is the opposite of what we are attempting to achieve with AcadiaSoft. AcadiaSoft is not about keeping competition out, it is about levelling the playing field to the extent we can. So for us, it’s simple. Standardisation reduces time spent on reconciliation between counter parties, standardisation eliminates disputes, and as a result standardisation drives down costs, which is ultimately what our clients like to see. </p><p><!--block-->JC: OK, so where are we now, then, in all of this process? Donal, Fred, you’ve taken us on a journey as to why it’s necessary and the potential benefits that’s there, but how close are we to reaching an effective form of standardisation that works across the industry in the way that you’ve just described, Fred? Richard, what are your thoughts on that?</p><p><!--block-->RB: I think if I was equating it to a transition from childhood to being a fully-functioning adult I would say we’re kind of in our teenage years at the moment.</p><p><!--block-->JC: Those are difficult years, Richard. They can be very messy.</p><p><!--block-->RB: I know. Having a couple of teenage kids myself is kind of why it’s fresh in my mind, but we kind of understand roughly what the rules are, but at the same time we don’t always apply by them, we don’t always follow them. So I kind of feel like we’re kind of at that stage, so we’re not quite there yet in terms of having a set of standard ways of doing things and following those in a responsible way. We kind of know they’re there but we kind of do our own thing a little bit, and that’s kind of what you see. So from the industry perspective, there’s a number of different initiatives that are out there that are aimed at standardising OTC derivatives data, so from a perspective of trades or legal agreement data. If you look at it maybe with the FpML – Financial Product Mark-up Language – and more recently with the Common Domain Model ISDA CDM, those probably have the most – and CDM in particular – have the most visibility and traction at the moment, and there’s definitely huge opportunity there with CDM and it’s definitely something that we’re collaborating with and helping to move in a positive direction. </p><p><!--block-->JC: And just to clarify on CDM, Richard, you’re referring to the Common Domain Model.</p><p><!--block-->RB: That’s correct, yeah. </p><p><!--block-->JC: So, Fred, we’ve got this teenager that’s trying to work things out – it’s a great analogy from Richard there – we’re starting to grow up, starting to take things more seriously, but where are we going to go from here? How close are we getting to our teenager to a respectable 20-something? </p><p><!--block-->FD: I think to continue that analogy, how do you get the teenager when they grow up, how do you get them to leave the house? It’s free to stay at home, probably in a comfortable bed, your mum makes you food. What gets you going? And for us, I think the exciting thing about where we are right now is that we are, in terms of standardisation, the industry is in the inflection point, excuse the cliché. But with the Uncleared Margin Rules, UMR, there’s an event that’s impacting all of the largest participants in the bilateral derivatives market. In order for firms to comply with UMR, they need to translate trade data into risk data, right, so there’s a cross-industry catalyst that is forcing firms to figure this out right now. And what’s different about this event, is that it affects so many firms. Many of those firms don’t have resources internally to deploy on this standardisation so there are going to be fewer bespoke homegrown solutions, and what that means is that more clients are going to turn to vendors to help them solve their problems and keep their costs low, and that by definition means standardisation. </p><p><!--block-->JC: Thank you, Fred. And Donal as well, from your perspective at Quaternion, is this what you’re seeing too?</p><p><!--block-->DG: Absolutely. Let’s put some numbers on it. When the uncleared margin rules are fully enforced there’s going to be something like 1500 of the world’s largest firms who will be obliged to reconcile not just the value of their derivatives, but the risk in those derivatives, so in the initial margin and on a daily basis, and unless these firms can agree on exactly what the economic terms are of the underlying contracts, this is something close to an impossibility. So even small differences in something like an option expiry date or just agreements on an option expiry date expressed in different data can give rise to huge differences in the risk and failed reconciliations, and that’s just not scalable to 1500 firms. So I would say for this particular teenager that standardisation is not a luxury, it’s a must-have. We’re obliged to go into adulthood when it comes to agreeing between all the firms what data looks like.</p><p><!--block-->JC: OK, Donal, thank you. Now in essence, standardisation isn’t actually necessarily a new thing. There have been moves in this direction for some time, but here we are now taking much bigger steps in that direction. Richard, talk us through what AcadiaSoft has already been doing in this space.</p><p><!--block-->RB: I guess I would class what AcadiaSoft does sort of like practical implementation of standards. So, if I kind of contrast back before I joined AcadiaSoft back in 2010, I worked on a standardisation project and we came up with a standard way of representing margin call exchanges between firms. It was a great definition, worked out very well, but no one uses it today. Why do people not use it today? Because there wasn’t really any practical implementation of that standard. So what AcadiaSoft does is work on standards, we look to see what’s in the industry and if there isn’t anything there we use our strong community, as Fred mentioned, to help build that. And where there is standards, we look to incorporate that into a practical implementation of it so that we can help grow adoption of it. So a lot of the places where we run and use standardisation is in where our sweet spot is, which is really the inter communication between firms within this industry. If we look at AcadiaSoft, what we’ve built standards around and that we’ve been very successful on is standardising the exchange of margin call messages and dealing with that you’re creating a mechanism for moving away from email-based communication, which was fraught with inefficiency and operational risk, and providing a more electronic way for firms to communicate around a standard representation. So we’ve done that with margin calls and we’ve expanded that into interest statements and had a lot of success with that; we have standardised around initial margin reconciliation related to supporting uncleared margin rules, and we’ve also recently been expanding that into the legal agreements space and initiated standards around agreements, and then also contributing into and helping to develop the ISDA CDM representation of a legal agreement. It’s to sort of backfill, backport any enhancements that come from that industry standard with our own definition. </p><p><!--block-->JC: So Richard it doesn’t sound like you’ve been standing still, that’s for sure. As we approach the delayed phase 5 of the requirement to exchange two-way regulatory initial margin under the Uncleared Margin Rules regulation – quite a mouthful – what does that mean for our industry? What actions will people need to take in respect of data standardisation? The people that you’re dealing with and working with and helping each and every day. Fred, I’m going to come to you for this.</p><p><!--block-->FD: Sure, thanks. I think first it’s worth highlighting again the sheer number of firms in phase 5 relative to the prior phases. As Donal mentioned, there are hundreds of firms between phases 5 and 6, potentially over 1000. The number 1500 is frequently thrown around for reference. So it is a scale that is in a completely different proportion from phases 1 through 4, where you had roughly 20 firms in phase 1; every phase to date has been effectively double digits, and now we’re entering new territory. And so the requirements for these firms fall into two categories, which are the operational requirements that a firms needs to meet to use the Standard Initial Margin model – SIMM – and then the regulatory requirements. So on the operational front what a firm is now doing is really generating the input that drives the calculation and reconciliation SIMM IM. That input is called the CRIF file – CRIF is the Common Risk Interchange Format – and what it contains is the clients’ positions that are subject to UMR and that those trades risk sensitivities. And if you look at your average buyside firm – even fairly large firms that are subject to these rules, the larger derivatives trading buyside organisations – they just do not have excess resources with capacity that can be diverted to creating this daily process. On the regulatory front, what firms need to do is demonstrate that SIM is appropriate for their specific portfolios. So really showing that that standard model, which has a number of benefits for the industry, fits specifically what their firm is doing. There is some potential that if a firm has a particularly skewed or concentrated portfolio, that the results of PnL moves day-to-day would exceed what the SIM model is calibrated to compensate for, and I will leave it there and let Donal dig into some more of those details.</p><p><!--block-->DG: Thanks Fred. I would say the watchword here is ‘prepare’. And prepare, and prepare, and prepare some more. What we’ve seen among clients is that it really takes time to bed down the processes that feed into a good, functioning daily margin exchange process. I think it’s really important to start early, to get to a place where you’re in a position to calculate that CRIF file that Fred talked about – that’s standard Common Risk Interchange Format, so that’s the file that contains the sensitivities of the trades – get that out early and then start the process of matching that and reconciling it with your counterparties long before you go live, because that flushes out all the things that you need operationally to be in a position to exchange margin. It’s really important to have a good checklist and experience around that checklist, and things like matching trades: do you see the same trades in the portfolio as your counterparty does? You’d be surprised how challenging that can be. Making sure that you’re actually comparing apples with apples, that for example down the details if you’re using crowdsourcing data and your counterparty isn’t you can disagree on the risk buckets. It could be that you’re using decomposition but your counterparty isn’t, and that’ll give rise to huge differences in calculation; and ensuring things like that you’re looking at the same risk classes and you assign things to the same risk classes. So gaining practical experience with all of those things is vital, and having a bit of runway so that you’re not doing that under pressure as you’re approaching the go-live date, and if you need additional expertise or additional help you’re in the position to have time to get that. And finally behind it all are the calculations that feed into the SIM model, the Standardised Initial Margin model, because they drive everything. So you need to gain organisational experience with what drives SIM: what drives the risk on an interest rate derivative, what drives the risk on an equity basket derivative? And finally I would say it’s exactly what Fred said: don’t forget you can get overly focused on getting operationally ready, but you have to have compliance, which means you need to initially validate the SIM model and then make sure it’s working on an ongoing basis, whether it’s by testing or monitoring processes depending on our regulatory requirements, but there are ongoing monitoring processes in addition to the daily exchange. And of course around all of these things standardisation in all its various forms really, really help. </p><p><!--block-->JC: So I think it’s fair to say that there’s a fair amount for any party in our industry to digest and think about. We’re talking about scale here, we’re talking about a degree of complexity, but also how important standardisation is. Looking forward, and finally now – last question to you all - what are the major challenges that lie ahead? Is our industry ready for standardisation? Donal, you said businesses need to start early, they need to be thinking about this now. When you’re talking to them at Quaternion, are they doing that? Is that a challenge that you’re seeing that’s being met or are we some way off? </p><p><!--block-->DG: I think the industry is ready, for sure. I think there’s a convergence of the regulatory waves that we talked about, the existence of utilities like AcadiaSoft, the centralisation in the community, aspects of organisations like AcadiaSoft is hugely important to facilitate that transition. And you don’t have to take my word for it: Risk magazine recently talking about standardisation, things like that ISDA CDM, and the things that will drive that, in the words of Risk magazine, are the large infrastructure providers, so it’s probably not any one or even a number of those 1500 firms getting together, it’s actually the central infrastructure providers that will drive standardisation. Go back to the main thrust of your question, Jon, which is are firms ready? Yes, they’re ready for standardisation, but in practise I think firms need to allow more time. I think they need to start earlier on this journey of preparedness to allow more time for testing, to allow more time for processes to bed in and to understand what standardisation needs to be put in place and what the benefits of it are.</p><p><!--block-->JC: Donal, thank you. Richard, what’s your view? What are the major challenges that you see from your perspective at AcadiaSoft?</p><p><!--block-->RB: I think the challenges really come down to what is the driver for moving towards standardisation? That’s one of the challenges that we see, is that we’re seeing great progress in terms of standardisation where there’s a regulatory push, where there’s a regulatory need; where there isn’t necessarily a regulatory stick there that’s kind of pushing people towards coordination or improvement, then there can be a bit of a slowness in taking up and contributing to standardisation. We’ve built up margin standardisation and interest over time where they haven’t necessarily, initially anyway, had a regulatory push. So we’re seeing that a little bit with some of the other initiatives we’re working on. With legal agreements, for example, there’s a recognition that the way things are done now is not efficient, but there’s lots of other competing things going on, like complying with regulation, that sort of mean firms don’t always get to working on these projects, so what’s the driver? What’s the availability of time and resources to work on it? As Donal said, it’s not easy, you’ve often got, as we’ve been growing up and heading towards our teenage years, we’ve been doing our own thing and it’s difficult to change, it’s difficult to make that move, and there’s an investment that’s needed to kind of move away from internal proprietary ways of representing data to them being able to do effectively the same thing but based on a standard, and that’s a challenge moving people away from that. I think where you’ll see the biggest benefit, and this is where AcadiaSoft has been successful, is in the interchange between firms or between functions or systems within your organisation. So where it’s easier to implement standards where they exist, is when there’s an exchange of communication between one party and another party, or one system to another system. If you can do that based on a ubiquitous language, a common way of looking at things, it’s a lot easier to move forward. So that’s kind of where I see it. What’s the regulatory driver and how can you budget and deal with a lot of competing interests?</p><p><!--block-->JC: Fred, I’ll give the last word to you on this. What would you add to Richard’s summary there?</p><p><!--block-->FD: I don’t know if it’s OK to continue beating the teenager analogy, but I’m going to go for it.</p><p><!--block-->JC: It’s always a good thing. They’ll only answer back.</p><p><!--block-->FD: Thinking about it this time from the parents’ perspective, what’s required here is really a leap of faith, and to me that’s one of the challenges with adopting standardisation. It’s giving up control, it’s seeing that your baby is ready to leave home. And you’re dealing right now with lots of firms, highly trained employees that have focused on really building out complex and therefore somewhat customised ways of representing bilateral derivatives, and you’re asking them to take a leap of faith, to adopt a common standard for their long-term good, really to save long-term costs, and ultimately potentially for benefits that are outside their immediate discipline. That said, you know what our answer’s going to be to this question of ‘is the industry ready?’ We are in a really unique position right now where we have on the one hand a well-defined model for OTC trade data; we also at the same time have what is an industry catalyst to spur that adoption in UMR, and we have the cost argument that standardisation around one solution is by far the most efficient path forward. And we at AcadiaSoft are seeing that play out in practice. Our goal in building these services was really to eliminate the barriers to firms using SIMM and a big one of those is cost. And we’ve heard it come up on every front, from regulators asking us what the solution’s going to cost, obviously in the client conversations, and our goal from the beginning has been to build out an efficient standardised process that would allow any firm subject to UMR to meet its requirements at the absolute minimum cost. And so far it looks like we’ve been successful, with roughly 50%, potentially more, of the firms in phase 5 that are expected to move margin in the first year using the standard model that we’ve implemented. So yes, I think it’s possible, and yes, I think now’s the time, and I think it’s exciting for all of us. We’re looking forward really to the next two years, to watching it play out in the industry. </p><p><!--block-->JC: Well as you say, Fred, exciting times and a fascinating subject. Unfortunately, though, we’re out of time. Donal Gallagher, Fred Dassori and Richard Barton, thank you all very much. And thank you for listening to Ahead of the Curve. We’d like to know what you think, so please do get in touch and share your thoughts. And you can find out more about AcadiaSoft by going to <a href="http://www.acadia.inc">www.acadia.inc</a>. </p><p><!--block--> </p><p><!--block--> </p><p><!--block--> </p><p><!--block--> </p>
