Webinar Transcript
Webinar Date: June 30th, 2021
Click here to watch video.
Moderator:
Ariel Berezowski (Head of Treasury, Point72 Asset Management)
Panelists:
Ted Leveroni (Senior Product Manager, Bank of New York Mellon)
Joseph Spiro (Director of Product Management at Hazeltree)
Poseidon Retsinas (Founding Partner, HedgeLegal)
- Introduction
Ariel Berezowski: Good morning and welcome everybody. My name is Ariel Berezowski, I’m the head of treasury at Point72. Today we’ll be discussing topics related to uncleared margin regulations, but instead of covering the usual material that we have all seen lately related to UMR, we wanted to take a step further into this topic, call it UMR 102. Our topic today will include both legal and operational considerations. That is, it is time to go live. So for today, Hazeltree has put together a great panel of experts in the field that will help us onboard some challenging questions, some of which we will not have the final answers yet, but do help provoke some interesting conversations.
We will have 45 minutes to discuss some topics with the panelists and then we’ll leave 15 minutes for Q&A at the end. Throughout the presentation, we will present some polls for the audience to answer and we’ll provide the results as we go. All lines will be muted. If you have any questions for the panelists, please use the Q&A section provided by the app. At this time, I’d like to introduce the panelists themselves, starting with Ted.
Ted Leveroni: Ariel, thank you. So my name’s Ted Leveroni. I’m a Senior Product Manager within the segregation and collateral business at BNY Mellon.
Ariel Berezowski: Thanks, Ted. And we also have Joe.
Joseph Spiro: Thanks, Ariel. Hi, I’m Joe Spiro. I’m a Director of Product Management at Hazeltree. I’ve been in the collateral management industry for about 20 years, holding leadership roles at firms like Merrill Lynch, Deutsche Bank, BNY Mellon, Société Générale. I’ve contributed to ISDA and other industry working groups on the topic of collateral management. At Hazeltree, I’m focused on helping our buy-side clients to improve their collateral management process, to increase efficiency, reduce risk and increase liquidity. I’m honored to be on this esteemed panel today and looking forward to a great conversation.
Ariel Berezowski: Thanks, Joe. And lastly, Poseidon.
Poseidon Retsinas: Thanks, Ariel. I’m Poseidon Retsinas, the founding partner of HedgeLegal. We’re a boutique law firm, which is focused on the negotiation of trading documentation. Prior to HedgeLegal, I’ve been in the industry for about 15 years working at Clifford Chance in the derivatives practice, as well as at Innocap where I was heading up the onboarding and negotiation with counterparties. Our firm has been assisting firms, assessing the impact of UMR as well as negotiate documentation. There’s lots of great topics to get through today, Ariel, so back to you.
2. Choosing a Custodian
Ariel Berezowski: Thank you. So let’s get started. So let’s start with point number one. Essentially, one thing we know at this point is that we got to pick a custodian. It doesn’t have to be the same one between us and the other party, so each one gets their own. But the question is, and this is for you Poseidon in your opinion, what should people look for in a custodian? There’s plenty of options out there. What are your thoughts?
Poseidon Retsinas: That’s a great question. So I think the custodian question comes up very early on in the process. As soon as the manager knows that they’re in scope for UMR, they’ve done the ANA assessment, they’ve realized that they’re over the IM threshold of 50 million, with at least one counterparty, and will probably be so at the time of their either phase five or phase six implementation. It’s essential that they start discussions with custodians. If they’ve never worked with a custodian before, there’s a bit of a learning curve to go up both operationally, as well as some of the legal documents that are needed to set this up.
So I think it’s really important that they start the process early. I recommend that they speak to multiple custodians to understand their service offering, both along the lines of, do they have tri-party, do they have third-party? Has this custodian been involved in previous phases? Do they have experience? How are they in terms of their onboarding capacity to meet the deadlines for phases five and six? So I think these are really important questions, as well as assessing what level of connectivity does that custodian have with the counterparty or the swap dealer that the fund will be facing. So there’s also benefits to be had by using the same custodian or custodians that the counterparties are familiar with. If you’re bringing in a custodian that they’re not familiar with, that’s going to take more time and there may be even pushback from the swap dealer as to using them altogether.
Ariel Berezowski: Thanks. So you mentioned a couple of keywords there and generally some confusion out there on our experience. Point72 is a phase five UMR participant and some of the results that are coming here pretty divided. There’s a lot of attendees that are not hedge fund managers. The rest are phase five or phase six, as well. But you mentioned two keywords there and there’s been a lot of misunderstanding, third-party and tri-party. The buy-side has used the word tri-party for a very long time, but maybe it would be interesting to really understand the difference. So Ted, you’re at Bank of New York, you know this more than anybody else probably out there. You can give us a little bit more color about the difference between tri-party, third-party.
3. Custodian Model: Understanding the Difference Between Third-Party and Tri-Party
Ted Leveroni: Sure, sure. Thanks, Ariel. This question obviously comes up a lot with us. We’re one of the few custodians, there are others out there, we’re one of the few custodians that offers both the tri-party and third-party model. What you’ll find is most custodians or service providers offer one or the other. There’s a couple out there that offer both, like BNY Mellon, so we’re having this conversation a lot. Like you said, Ariel, the terms have been thrown around a lot. I came from the buy-side and we always thought of tri-party as really what now we’re calling third-party, which I realize is confusing, but I think the industry now is settled on these terms, right?
Third-party model is essentially a separate custodian account for every single pledging to counterparty pair. So if you are a hedge fund and you’ve got one legal entity that’s in scope, and you have five dealers, you’re going to open up a full custody account for each one of those pairs and then you’ll post to that. There’s a wrapper around it. For UMR, there’s a regulatory wrapper around it, the ACA has to be compliant. There’s default processing that goes to it and the like, but it is essentially a custody account and not a lot more. There’s not a lot of other functionality in that account. You instruct, you will determine what collateral is going to be moved, how much is going to be moved through your counterparty, you’ll instruct your existing custodian, or whoever’s holding the assets, to deliver it into the custody account and then you’ll instruct to have it sent back and then the secure party will release it. The third-party custodian doesn’t check eligibility, doesn’t check haircuts, doesn’t check concentration limits and the like. It’s simply an account that holds the assets and holds them safe.
The tri-party model is a little bit different, both in structure and then certainly in level of services. The way the tri-party model works is you create one custody account. That custody account is the pledgers own assets. They can put the assets in there, and that’s essentially what we might call long box, where those assets are sitting there, unencumbered, still fully owned by the pledger. What the tri-party agent then does is they add shells that attach to that single custody account and each one of those shells represents the counterparty pair. When the counterparties agree to the overall required amount, then they would message the tri-party custodian and the tri-party custodian has already, within their system, captured all the eligibility criteria, all the haircuts, all the concentration limits, maybe some optimization preferences and then the tri-party service provider or custodian will then move the appropriate amount of assets either from the long box into the individual shells.
Or if there’s overcollateralization, the collateral that’s segregated is greater than the required amount, then the custodian will pull assets from those shells back into the long box. So basically unsegregate those assets. So the third-party model I described, for anyone that might have experienced being a 40 act mutual fund, where they segregate VM, looks very, very similar. The tri-party model was born from the tri-party repo model. So if anyone has experienced a tri-party repo, they might be used to this model. So those are the two models, essentially, that the industry has settled on for the segregation of assets. The third-party, custody account and the tri-party adding in allocations, optimizations, eligibility checks, haircuts and the like. I think you’re on mute, Ariel.
Ariel Berezowski: So you’re the front row of the action here, right? The poll suggests that about 60% hasn’t decided yet which model to use. The rest is 50/50 down the middle. What have you seen at BNYM?
What do we expect most managers in Phase 5 and 6 to use?
Ted Leveroni: That’s interesting. I’m surprised to see that poll. I think it shows that we’ve got an educated audience, because, I’ll tell you, going into … phase one through four were primarily banks, broker dealers. They almost universally save one, that I know of, went tri-party. It’s automated, it’s easy, it’s efficient, it’s a model that the sell side fully understood and had used and, like I said, in the repo market. When phase five came along, obviously a lot of buy-side firms, we went into this thinking it would probably be 80/20, 80% third-party, because they’re buy-side, 20% tri-party. We were surprised to see it was really almost 50/50, which actually is what the poll said. So it tells me that people have been educated on it.
Because what happened was, going into phase five, the knee jerk reaction still on the buy-side was oh I’m going to go third-party, but tell me more about this tri-party stuff. I think as they educated themselves about the model, how it worked, some of the efficiencies, some of the assumptions around cost that may or may not have been accurate, a lot of them realized that tri-party model actually worked with them when they thought that it might not. So traditional asset managers, I know traditional asset managers that have gone tri-party, definitely third-party as well, but tri-party, as well.
Insurance companies, I know some that have gone both. Hedge funds, I know some that have gone both. So, you really need to go in, in an educated way to make this determination. There’s a lot of assumptions you may make going, before you learn more about the two models and speak to custodians that can offer them both, but you want to really challenge those assumptions and then make the determination in a fully educated way. You may fall on third-party, you may fall on tri-party, but what you think going in may not be what you end up with.
Ariel Berezowski: So Joe, from your side, you do a lot of servicing for a lot of people who had to manage collateral and you see custodians not BNYM but also additional custodians. What’s your experience? What are you seeing? How this simplicity to basically put in place a tri-party? I know a lot of people are afraid, oh tri-party, this seems like a bank thing only it’s complicated. What do you see on your side?
Likelihood of CP Custodian’s being Tri-Party – Matching RQV
Joseph Spiro: Yeah, no I think that’s a really, really interesting observation from Ted that we were expecting to get a large percentage, a very, very large percentage, the overwhelming majority, in phases five and six elect for a third-party account and maybe that’s not playing out quite as strongly as we had thought. Which, in my opinion, is great. I’ve always been a big proponent of the tri-party system. This conversation has been going on for a lot of years and I think a lot of people, just because of the familiarity aspect with the third-party model, because this is what people have always used, may have just had a knee jerk yes third-party, that’s it. But as you learn more about it, you realize that it’s actually a lot more efficient to go with the tri-party, so it’s nice to hear that people are keeping an open mind about that.
The other thing to consider, and Ted alluded to this, depending on which one you choose, whichever one you choose, there’s going to be a big impact on your operational process because the operational processes for the two are very different. If you have a third-party account, it’s the traditional model of collateral management that we’ve gotten used to where the secured party makes a call, I’m calling you for X amount, and then the pledging party says okay, I agree to X amount and here’s what I’m going to send and they define a particular asset, whether it be cash or a bond or whatever the case may be. Then they wire that particular asset into the custody account.
In the tri-party model, it’s completely different than that. Ted alluded to it. The custodian is the one in charge of selecting the asset based on an optimization ladder. He’ll consider the price and the haircut, et cetera., and they’ll allocate those assets accordingly. So the only thing that the two parties involved in the margin call have to do is agree on the amount. That amount is called an RQV, required value. Both parties have to communicate to the custodian, the tri-party custodian, what that RQV is so that the tri-party custodian can match it up and move the assets.
Big difference because, in the third-party model, you as the pledging party need to know what assets you have available, you have to track prices, you have to track eligibility, you have to track haircuts and you have to have systems and processes in place to track all of those things so that you’re pledging the right assets. Tri-party model is simpler, because you only have to communicate a number, but one thing that is not always considered, that is a very important consideration, is even if you as the pledging party are not selecting the tri-party model, if your counter party is selecting the tri-party model where they are the pledging party, where your dealer is pledging to you, both parties have to communicate that RQV number to the custodian. They have to match it from both of them. So even if you are not selecting a tri-party model, you still have to have the ability to calculate the RQV and communicate it to the custodian in a manner which they’re willing to accept, which is often Swift messaging. It’s a big consideration that people sometimes lose track of.
Ariel Berezowski: So let me just follow-up on that one for a second. You mentioned Swift messaging and, probably a question back to Ted, is there a possibility of doing tri-party without Swift? I know a lot of people don’t have Swift.
Ted Leveroni: There is. So we at BNY Mellon, and I don’t think that we’re unique in this, is we try to make things easier for our clients. The larger firms or firms that use administrators and the like, they typically can use Swift and that’s the industry standard. We also have a portal, I think most custodians do have a portal that you can go in and you can manually make RQV instructions, as Joe said, or movements, pull down reports, all of that.
Then also, linking to industry infrastructure is something that I think is important for any major custodian. It’s something that we take seriously at BNY Mellon and so creating links to firms like Hazeltree that have a group of clients that can benefit from that link. So you don’t need to … from the client side it’s great. They don’t need to create the new Swift messaging and they can enjoy, still, an automated messaging component because they’re on the right system. Then, for us, it’s great because we don’t have to work with … as much as we like working with our clients, we don’t have to work with onboarding all of these clients because they’re on a system, it’s fully integrated with us, already tested with us. We know the Swift messages are good, so for us it’s still Swift, for the client it’s not and it’s taken care of by a provider.
So there is that middle ground between creating your own Swift messaging and logging on to portals. Which, for low volumes, is fine but I think as the volumes go up, you’d like to enjoy that electronic connectivity that’s automated in a Swift format. So, a number of different ways to of it. Like I said, we take seriously at BNY Mellon to try to make sure that we integrate with the entire ecosystem. It just benefits everybody.
Ariel Berezowski: All right. Let’s move on to the next topic. So we pick our custodian, we have picked our model. Next big questions is who do we need to pay for what? Obviously you have all your counterparties that you need to go after and back in the days when it was free to establish any ISDA, you would go to a counterparty, you will pay for every single entity they had, but you never did a trade in the last 10 or 15 years with some of those counterparties. So Poseidon, you’re in the front row of the action on the legal documentation, what are your thoughts on this? Why would be your recommendation to a client on how to approach the massive amount of documentation that is going to come their way?
4. CP CONSIDERATIONS: Which Margin Approach will your Firm Use?
Poseidon Retsinas: That’s a great question and it’s a loaded question, but I think the discussion point … having discussions with your counterparties is paramount, just as much as you want to start talking with your … as choosing a custodian, approaching all of your counterparties, particularly those that you have a larger trading relationship with is going to be really important. For those that are under the 50 million threshold and, again, this is across the counterparty group, so if you’re dealing with multiple affiliates at one counterparty, that threshold gets applied across all of them or aggregates.
You need to essentially figure out which ones are potentially going to be in scope. Those ones that are in scope and, as you say, where they have multiple affiliates, you need to figure out where those trades are with the different affiliates. You probably will not need to repaper with all of those affiliates. You’ll probably only need to choose a smaller subset, maybe one of them. So, that’s an initial assessment that needs to get done.
I think more importantly than that, though, is we see managers be strategic about how they’re approaching UMR both from the portfolio composition perspective, but also as to how they spread their business across their counterparties. So what I mean by that is, on doing business with counterparties perspective, let’s say we have a manager that has five counterparties and when they project into the future what the IM requirements are going to be, let’s say four of them are going to be over the 50 million threshold or close to it. They would have to repaper with all of those counterparties. By changing the mix, either by concentrating more business towards a smaller subset, let’s say two out of the four, you can reduce the number that you’d have to paper, so you can concentrate your business at fewer counterparties and so only have to repaper with a smaller subset.
Or the opposite can also be true, where you can put in place more trading lines, other ISDAs, spread your business across more counterparties and that way, potentially never get over the 50 million threshold with any of them. So I think that’s something that should be looked at very early on, to be ready for that and it can be really beneficial. The other thing is, on the portfolio composition, let’s remember that a fund can fall into and out of compliance depending on what their aggregate, average notional amount of uncleared OTC derivatives is over certain periods of time over the year.
So there are certain ways that a portfolio can be modified so that you actually your total outstanding notional amount of uncleared OTC derivatives. You can do some portfolio compression or you can even move, if possible, can use other cards and you move to cleared OTC and you move to future products to actually get out of … essentially get out of being subject to the rules all together. So I think that’s another step that should be done as early as possible to see if there’s any way to work around those as possible solutions.
Another point on these counterparties, on the point with counterparties, is this 50 million threshold. So going forward in the future, this is a threshold that needs to get monitored. If you’re under the 50 million threshold, let’s say you’re at 25 million with a counterparty, you’re not going to need to repaper. The regulators have stated that, if you’re under the threshold, you don’t need to repaper. As you approach this thresholds, then you will need to repaper and you should repaper because if you go over it, you may have to put trades downs or you won’t be able to continue trading with that counterparty. So the point of monitoring is just to be aware of where you are with the different thresholds, have a conversation with the counterparties, the swap dealers are not going to want to paper somebody that’s at 20 million on the threshold. It’s a lot of work for them and it’s probably not worth it. If you’re at about 30, 35 or 40, then they’re probably going to say yes, let’s paper this and let’s be proactive. So that’s another point into the discussion.
We’re three months, what are we … we’re two months away from implementing phase five which is really soon. Phase six has a little bit over a year. I think it’s really important for phase six firms, at this point, to start getting ahead of this. I’ve heard from some phase six firms that counterparties are not ready to engage with them, they’re focusing on phase five. That is probably true and it is certainly is true, but I think it’s still really important for phase six firms to be at the front of the queue. There are going to be a lot more entities in scope for phase six than there were for phase five. It’s estimated it’ll be around double and each phase has seen more and more entities in scope.
So the number of entities that were in scope for phases one through four were somewhere around 50 or 60. We’ve seen probably about 100 here and there’s going to be at least, probably at least 200 in the next phase, based on various estimates that we’ve been hearing. So I think getting in there quickly, being proactive with counterparties is really important so that you’re not at the end of the queue come Sept 2022.
Ariel Berezowski: All right, so let’s touch base a little bit on a lot of information there. Touch base on a lot of people coming in. Phase five obviously, Point72 has seen a lot of the activity going on with the brokers and they wanted to be done with phase four before they pay attention to us. But a very good point that you mentioned, Poseidon, is you want to be in the front of the line. There’s a lot of implementation to be done, you want to attract attention to broker dealers, but not necessarily everything can be negotiated. A lot of things are established by the rules and there is, on the custodian side for example, they need to deal with everybody.
So sometimes there is some limitations of what is humanly possible. So Ted, back to you now, what approach is BONY taking, what’s the approach of the industry as they had to tackle hundreds of these agreements themselves? It’s like on my side, it is overwhelming, we have a person dedicated to negotiating UMR, we have people and lawyers dedicated to negotiating UMR and this is just us, but you guys face everybody, almost everybody. So what’s the approach BONY is doing here? Can you give some color to participants and some recommendations, as well?
Ted Leveroni: Sure, sure. First of all, it’s not hundreds for us, it’s thousands because for every large deal that’s a client of ours, we’re facing off and we have to deal with the ACH for their pledging, even if their pledging counterparties are not pledging through BNY Mellon. The lift right now is like nothing I’ve ever soon and I’ve been in the industry for 20 plus years.
A couple things that we’ve done. The first thing we did is that we set aggressive deadlines when we needed the documentation. Honestly, there was a push back on that and we got a lot of questions. Do you really need this much time, why is it that your deadlines are March or February for certain documents? Other custodians are later. But we had those deadlines for a reason, it was because of the volume of documents and the work that we were going to need to do to onboard all of these clients.
We also set up a receiver’s concierge group. So one of the things to remember is, BNY Mellon, we were a very large participant in servicing the phases one through four, all the broker dealers. So there’s a huge number of their counterparties that we have to onboard that we don’t know as a client. There’s a lot of work that has to be done there for the client. Joe mentioned the RQV has to be sent by both counterparties. This isn’t something that most buy-side firms have done in the past. Buy-side firms are also going to want reporting, they’re going to need to get that. That requires … it’s not a heavy lift, but it is a lift nonetheless.
We have to support it, so we actually set up a receiver’s concierge group so anyone that’s receiving collateral, be it BNY Mellon, whether they’re using us or not, to plug this, they should email us at marginside@BNYmellon.com to get this information to get set up. This is important. We’re telling you our brokers, we publish that with ISDA, we’re trying to make it easy for receivers to set up and we’ve set up a, I don’t want to call it a taskforce, a team to handle just receiver. These are non-pledging clients.
Something we also did, is we standardized our ACAs. This was an interesting point and there was a lot of discussion about it internally because phases one through four, they would be allowed the typical negotiate however you want the ACA and that worked. Because we had so many ACAs going in, what we did is we actually created a committee of buy-side representatives, sell-side representatives … by the way, not just one type of buy-side, we realize that buy-side is a monolithic entity, so we had all different types. We invited insurance companies, traditional asset managers, pension funds and hedge funds and the like. We took a look at the ACA that we had for phase four and we said how do we make this, essentially, more friendly to everybody, easier for everybody.
What we came up with was this non-negotiable ACA but with options. So when we have a body of an ACA where this is terms that are locked down that have been tested by the industry and we say okay, these are good, typically, for everybody. Might not be ideal for anybody, but good enough for anybody and then at the end of the ACA, we have options you can pick. So if you want the right to contest a default notice, you and your counterparty want that, check the box and you’re good. You want delay period on a default notification, check the box and you’re good.
So we ultimately, what we did was, we came up with this document that could very quickly be executed by our counterparties. Now, truth be told, a lot of counterparties struggled with this in the beginning and I understood why, because they’d never been told, especially large clients, they’d never been told no you can’t negotiate these items. But almost universally, once they started to get these executed, they found that it actually was a really good decision because it got to the point where everyone one their counterparties has already seen and agreed to these documents dozens of times. So, if you’re coming in near the deadline and you go to your dealer and you say, oh I’m going to pledge at BNY Mellon, they’ll say you know what, we already know that ACA. We don’t have to do a whole lot of time figuring this one out, we’ve agreed to it 80 times just last month. We can do that again. So, that’s something that we also did and I think that that helps.
Other things like outreach, education and the like are things that, I think, we’ve taken seriously. We had to do it because of the tri-party, third-party discussion that we had earlier. But those are some of the big things that we did, within our shop, to try to make phase five go as efficiently as possible, given the numbers of new firms that were being pulled in.
Ariel Berezowski: I appreciate that Ted. One more topic that was from things that Poseidon mentioned is how do we find ways to try to stay below thresholds? Some of the things that are out there, as well is … there’s different ways of posting collateral. Poseidon, you’re the expert on this matter, as well. Joe, you have seen a lot of activity, as well, on this side. So I’ll start with Joe … sorry, Poseidon and then we’ll go with Joe. Poseidon, if you want to go through the different type of collateral methods and Joe, I’d love to get your color, as well, of what are people doing? What are people electing?
Poseidon Retsinas: Thanks Ariel. Listen, I’m going to pass this one off to Joe. Joe, if you want to go first and discuss the margin approach because I feel like you’re … I want to get you into the action here, so go ahead.
Margin Approach – which method to choose?
Joseph Spiro: Sure. So yeah, the margin is approach is interesting in relation to what Ted was just discussing because Ted was discussing how, in order to get everything done, we really need to standardize. Well this is one aspect where the industry has introduced something that’s the opposite of that. It’s not standardized and it’s a term that didn’t actually exist before in early forms of the CSA because it wasn’t a relevant concept before. We have the notion of independent amount and initial margin. Well, isn’t that the same thing? It was always those terms were used interchangeably to mean exactly the same thing, similar to tri-party and third-party, and now they mean something different.
Used to be a broker, based on their risk appetite, would charge certain clients and certain trades with certain clients an extra added amount often expressed as a percentage of notional, but it would be done a number of ways, as a buffer above and beyond the mark to market they need to post. People called this independent amount, people called this initial margin.
Now with UMR, the regulated amount that must be posted according to the regulations is referred to as initial margin. So that amount from the broker, based on their risk appetite and their own discretion, has been called independent amount. Two different concepts. It’s an important one because if you’re subject to independent amount today, pre-UMR and you’ll be subject to initial margin post-UMR, what do you do? The new CSAs have introduced this concept of the margin approach where the industry settled around three different approaches, there are pros and cons of each of these.
The distinct approach basically means that if you’re subject to independent amount, you continue to pay it and then if you’re subject to initial margin, well now you’re going to pay that, too. So you’re essentially paying both. As you can imagine, the dealers who are currently receiving the independent amount would be in favor of this method because it meets the regulatory requirements and they continue to get use of those funds. From the buy-side’s perspective, it’s probably a little less appetizing to have to pay both.
The greater of method is another approach that’s used which basically says you look at the amount of the independent amount, you look at the amount of the initial margin, you see which one is greater and you post that amount in its entirety into the segregated account. From the buy-side’s perspective, this is great because they’re posting the least amount possible and operationally it’s efficient because you’re only making one flow of collateral into one place. From the dealer perspective, if they were getting that independent amount posted to them directly previously, they’re going to be losing the use of those funds so it’s a little bit less ideal.
A bit of a compromise is the third one, which is called the allocated approach, which essentially says you look at the amount of the initial margin, the regulatory amount, and that needs to be posted to the segregated account. You can’t get out of that. But, to the extent that the independent amount exceeds that, you would take the amount that it exceeds the initial margin and allocate that back to the broker. So that’s a meeting in the middle. Important to note that some buy-side parties don’t post their independent amount directly to the broker, they post it to a segregated third party account and in that case, from the broker’s perspective, it doesn’t really matter whether you choose allocated or greater of. It’s very similar from their perspective, so you might as well go with greater of in that case so you only have one flow of collateral.
Ariel Berezowski: Thanks. So, just in the interest of time, I’m going to move to the next topic but first some answers from the poll. Seems pretty divided down the middle, some people electing the distinct approach. That one is surprising to me just as Ted said, sorry Joe said, you’re double posting collateral. Probably on the buy-side, it’s the least preferred one, but obviously different people make different choices what’s better for each company. So now we have a custodian, we know who our counterparty is, now it’s time to post collateral. That, essentially, is defined in the ECSs or equity collateral schedules … eligible collateral schedules, I’m sorry.
One of the topics that has been a hot topic with the sell-side is, can we post cash? The answer is yes, the answer is no. Ted, you’re one of the, if not the biggest, custodian out there. What it’s going to take? Cash is eligible, it’s not eligible, depends on the regulation? Thoughts?
5. ECS: What Collateral types do you intend to use?
a. Cash as Eligible Collateral
Ted Leveroni: The answer is yes and no. That’s almost the best description of where, I think, the industry lies on cash. The thing to remember is, and I’ll speak US right now, because I think most of the audience is from the US, but I’ll touch on Europe, as well.
According to regulations, cash in a vacuum is eligible collateral. It’s part of the list of collateral that’s eligible under the regs. Now the problem is that, under the regs, you have to segregate the collateral and custodians don’t actually physically segregate cash, they’re in what’s called the “anonymous account” so the view from the regulators, and the industry, because they understand the way custodians hold cash is that segregation of cash doesn’t work. So the regulators decided okay, the cash has to be transformed into something else that is eligible. So what you have, basically, is … it comes down to be really labeling something, but you have cash that can be potentially, if the counterparties agree, can be used as a transfer asset, it can be delivered, if that’s what you want and the counterparties agree and the custodian can support it, but it can’t be ultimately held as what is collateralizing the deal. The exposure, the required amount.
So, what most firms that we have talked to have settled on is money market funds. Now, not all money market funds are eligible and, in fact, there is a significant issue that I don’t know we’re going to get into today, but I’m happy to talk about it other times, is just talking about how they use money funds across the border when there are multiple jurisdictions that are regulating an entity. But, in the US for instance, money markets can be used if it’s treasury only and there’s no SEC finance or repo on that money market funds. So there are a subset of money market funds out there that are eligible collateral.
Typically what we’ve seen is firms saying all right, if they agree the cash can be delivered but money markets are what’s going to be held, they put into their collateral schedule essentially that and the firms gets a call for a million, they send a million to their counterparty, if it’s a third-party model, cash is delivered into the third-party account and then transformed. It’s automatically swept into a money market fund, so now what’s actually collateralizing the exposure is that eligible money market fund. And when they want the return of that collateral, they agree to a return, the money market fund is liquidated to their proper amount and cash is delivered back. So cash just becomes, really, how the delivery occurs but what’s ultimately segregated is a money market fund. That would work with treasuries or any of the eligible collateral the firms agreed to, but money market funds is so easy to do with cash. So that’s typically what we see.
Now, there are other impacts and determinants to this because remember, just because something is okay per the regulations doesn’t mean that you’re automatically allowed to post that. There’s still a negotiation between the counterparties on what ultimately is going to be agreed to and delivered. So, you can’t go to your counterparty right now and say Ted Leveroni said cash is okay, so put it in the CSA. Doesn’t work that way. We see, as a service provider, so many eligible collateral schedules and we have discussions about clients and unlike … I’d say very few of them are accepting absolutely everything that’s eligible per the regs. Firms have to be comfortable with it because you ultimately might get stuck with it if there’s default, there might be balance sheet impacts, there might be processing challenges. So ultimately, there is that negotiation after you get the full universe of what is eligible. But like I said, for cash, technically it’s eligible, but it’s not segregatable and so it has to be transformed into something else.
Ariel Berezowski: Thank you. So, we’re going to see from the results of the poll that different participants are looking to post all different assets. Cash is one of them. One interesting thing about the ECS is, as well, that look at this typical CSA. You see cash and you might see money market funds, sometimes you might see treasuries. When it comes to an ECS, it could have pages and pages long of eligible collateral that could go in there and not necessarily have to be the same from one side or the other side. It could be completely asymmetrical, not only by what is eligible, also by the haircuts that will apply for each party, as well could be asymmetrical. So, question for you Joe now, what are you seeing as a holder of a lot of these collateral schedules? What is the market out there?
b. The Different Eligible Collateral Schuedules (ECS) Options
Joseph Spiro: Yeah, you’re right. The eligible collateral schedules, especially from a UMR perspective are typically defined separately. Now, as Ted mentioned, both parties could have the same custodian and you could have the same eligible collateral schedule, et cetera, but it’s not necessary, it’s not required to do that. So, if the documents are negotiated separately and executed separately, there’s every possibility that they could have completely separate schedules, especially if they’re with different custodians.
But I think what we have heard is that, even if some participants on the buy-side aren’t planning on posting the wide and vast of various types of collateral that their counterparties on the sell-side might be pledging, a lot of time they negotiate the same eligible collateral schedule. Basically it should be even, even if the haircuts aren’t the same, at least the asset population would be the same and then they just don’t necessarily use it.
I was actually encouraged by that poll that just came up. It’s not all cash. Historically, collateral for mark to market purposes has been mostly just cash and will probably continue to do so. But for IM, it’s going to not be all cash, partly because of what Ted just discussed about the regulations, but also just using the assets that you have on hand to be more efficient. So, firms need to have systems and processes in place to track their assets, the eligibility of the assets, the price, the haircut, et cetera and then be able to post them especially in a third-party environment. In a tri-party environment, you have to post the RQV so that the custodian can do all that. But it’s interesting to see that it’s going to be a lot more than just cash going forward.
Ariel Berezowski: All right. So we have a custodian, we know which model we’re going to we’re going to approach, what type of collateral we can post. The question then becomes how much? Generally when you do calculations mark to market, people look at their books and records, that’s something not really complicated to do. Initial margin, people have different methods to do it. Let’s call it independent amount, to follow Joe’s lead from before, but when it comes to initial margin, calculating the SIMM itself, it is not necessarily as simple as the models, definitely can be replicated, but the question is what are people doing? So Joe, once again with you, what are people doing? Are they buying an assistant to calculate SIMM, are they filling it themselves, getting regulatory approval? Doesn’t matter if it’s here or there. Same question, afterward, for Poseidon. Do you have any guidance about what people should be doing and just be careful out there to follow the letter of the law. Let’s start with market collateral. What are people doing? Are they buying? They’re replicating themselves?
6. Calculating SIMM
Joseph Spiro: Yeah, it’s a great question and sometimes this is the start of the conversation. I think this is a great panel because we’re covering the 102 topics, as you said, and not necessarily starting with this one. But it’s an important one because how much are you going to need to pay for initial margin is obviously the meat and potatoes of the whole issue. What we have seen from buy-side firms is, frankly, it will vary from one jurisdiction to the next.
If you’re going to calculate the SIMM, you need to first calculate your sensitivities which some buy-side firms are not prepared to do, so it could be an outsource solution where you can hire one of the solutions on the market to calculate your sensitivities for you and then calculate your SIMM. There are plenty of services out there that would be willing to do that.
In certain jurisdictions, that’s absolutely mandatory for all participants. In the US it’s a little bit different because, and Poseidon can talk about this a little bit more, but the buy-side is not necessarily subject to the rule that requires the to replicate the model. So in that case, what we have seen a lot of US firms doing, is electing not to independently calculate, but rather just to accept the dealer’s number. There’s precedent for this. If you think about the future’s market today, a lot of FCMs will calculate an initial margin in the form of span or something else and they issue those margin calls.
There are not many parties on the market who are independently replicating span to verify the future’s margin call. So a lot of US firms have taken the same approach to SIMM. Hey, I’ll let my counterparty calculate it and I’ll go with that number. It’s certainly a lot more cost effective to do it that way and then the downside of it is that you give up some of management of your own risk. Also, as I mentioned, it’s definitely different from one jurisdiction to the next depending on the regulatory regime. I know that there is some industry advocacy out there to potentially change that, but let me throw it over to you, Poseidon. What are you seeing as far as the differences from one regime to the next?
Difference between US and EU in SIMM calculation
Poseidon Retsinas: Yeah, thanks Joe. I think you touched on it there. The main difference is that the rules in the US apply to the sell-side firms. The buy-side firms are being drawn in by implication. The regulations don’t directly apply to the buy-side firms. It’s a very different scenario over in Europe where the rules there apply irrespective of whether or not you’re a buy-side firm or a sell-side firm.
The buy-side firms find themselves in a situation right now and there is advocacy going on and hopefully there will be some relief granted, but they do find themselves in a situation where that, if they have elected to use SIMM, they need to be subject to a certain governance of using SIMM. So governance as to documenting why they selected SIMM and the appropriateness of choosing SIMM, as well as monitoring it and testing on their own end. Finally, even in some jurisdictions, there’s a preapproval process needed where managers will need to see preapproval from the regulator to use SIMM. So, it’s a very different landscape with respect to the use of SIMM for the US firms and firms in the EU. That’s one of the big differences. Hopefully these gaps will get closed. There is an initiative to try and harmonize some of the cross border impact of these rules, but it’s still a work in progress.
7. Timing Onboarding and Documentation with Custodians and CPs
Ariel Berezowski: Thanks. So, let’s close it out with what’s the road ahead? So most phase five people should be, at this point in time, be in the final stages of dealing with documentation. If they haven’t started, I would highly recommend that they seriously get on it. But for phase six, let’s start with you, Poseidon. What’s your recommendation? How fast do people need to get started with this? Let’s say September that we’re done with all phase five and the brokers will pay their attention and BNY will pay their attention now to phase six.
a. Documentation Backlogs
Poseidon Retsinas: Yeah. So, the start of the documentation process, I would separate that from the start of the actual UMR process which I think is, even for phase six firms, if you’re not already starting … you need to have already calculated ANA and you need to have already been, in my view, estimating what your IM requirements would be. You need to have already started making decisions, whether you use SIMM or not.
But strictly just with respect to documentation, once you’ve figured it out, I think the best practice is to probably budget about a year to get it done. It seems like a lot, it seems … even as I’m saying it, it seems unreasonable like why would it take a year to agree to legal documentation? But from the experience of the previous phases, it in fact has taken that long. So, the documentation that’s needed here is a lot more complicated than, for instance, for VM rules which were rolled out about four years ago where there was a big scramble in the industry. Let’s update the CSAs and it got done in time and there was a scramble there.
Here it’s more complicated because of the tri-party or the three-way nature of it that you have to seg it, you have the custodians involved and anyone who’s every work on any of these documents, as soon as you have three parties in a negotiation, it makes it three times as hard to get something agreed. The back and forth time can really lag. So, to Ted’s point earlier, the fact that the Bank of New York Mellon has standardized the ACA, I think that has gone a long way to speeding things up with respect to that.
There are some pros and cons there, as Ted said. Some firms are not that comfortable with the non-negotiable nature of some of the points, even though I think that the major issues were covered in the elections, in those ACAs. So I think most firms could probably get comfortable with that.
I would say get started as quickly as possible, figure out which counterparties are in scope, get the documents moving quickly. There’s multiple documents needed here per set of counterparty. You’re looking at four or five, maybe even six documents per relationship. Whereas you’d usually think well, it’s just this CSA. Well, not it’s more, there’s the CSA, there’s the ACAs on both side of the equation, there may be some other security documentation that’s needed depending on the jurisdictions and the custody account, which … depending on third-party or tri-party would also have some of its own nuances. So having said that, I think it’s key to get started as quickly as possible.
Ariel Berezowski: Thanks. Ted, your thoughts?
b. Onboarding Capacity and recommended timing from BNYM
Ted Leveroni: Yeah, I think Poseidon really hit it. Just a few added comments. The way I view it is actually there’s three pieces to this. There’s education and decision making, there’s your bilateral documents, so getting set up with a custodian and all the documents there and then there’s the trilateral docs, the ACAs. Like Poseidon said, you need three parties there.
The education, it’s already starting. If you’re watching this webinar, you’re starting the education process. You’re not doing it too early, trust me, this is where you want to start educating yourself. Tri-party versus third-party. What are the operational impacts? What are the costs? Who can do that? Bucket are you going to be in scope, as education? Bucket how are you going to manage the thresholds, as education? Those things you can do now and there are service providers and there’s vendors and there’s industry associations that can try to help you with that today.
You can also start your bilateral documents because these are … you’ve got to set up your custody account. If this is the first time you’re setting up a tri-party, segregation model, there is new documents that you’re not used to. Security agreements, GCAs, global custody agreements, all that stuff. You have to deal with your custodian. Custodians are happy to start those conversations.
You’re not likely going to get a lot of attention from your dealers, and I can’t remember, someone else mentioned it earlier on the panel, I don’t know who it was. You’re not going to get a lot of attention from your dealers right now or, frankly, from your tri-party custodian on the ACAs right now because that is what’s consuming a lot of our time. There’s not a lot of bandwidth for that, but that can certainly happen September, October time once that September time is phased out.
Then once that happens, then it’s the actual physical implementation. The set up, the testing, all of that stuff. There’s a lot of operational documents that are actually needed to either set up an account later that aren’t contracts. SSIs, tax documents, certificates of incumbency for signers, all of that stuff needs to be collected before you go live. So there’s a lot of work, even on the back end. So education now, documentation for the bilateral documentation now and later into the summer and into the fall and then hit the ACAs as quickly as you can, but also be prepared for those other operational documents that you need in the run up to phase six.
Also, when you’re educating yourself, check with your custodians around deadlines. Because like I said, they’re not all the same, they’re not all the same for a reason. It’s not because one custodian is lazier than the other, it’s because of the book of business that they’re going to have to support going into phase six, the heavy life they’re going to have. So, don’t assume that they’re all the same in terms of deadlines and know them well in advance because you don’t want those deadlines slide by.
Ariel Berezowski: I appreciate it and you know you really don’t want those deadlines creeping up on you, it would be a bad thing. We had some questions come in already from throughout the day. We’re going to start some of them to the panelists, whoever wants to take it first. If not, I’ll take them myself, as well.
So, a question that came in is, should the SIMM be calculated pre-trade by the buy-side or just before selling? Any takers?
Joseph Spiro: Maybe I’ll comment on that one, Ariel. Calculating the SIMM pre-trade is a good 102 approach. First of all, it requires you to have a SIMM calculation method, so the whole idea of just relying on your dealers calculation is out the window at that point. But assuming you have that, calculating a SIMM pre-trade meaning if I do this trade, it’s almost the what if scenario, if I do this trade what is the impact on my SIMM, where is it most efficient for me to do this trade, with which counterparty, et cetera. That’s a great practice.
We do see a lot of phase five and phase six clients talking about that because the idea here is you’re going to have to post more collateral than you had to previously, so doing that as efficiently as possible and keeping that number as low as possible, is important. Especially to the extent that you have a 50 million dollar threshold on your SIMM. That’s the other thing, is you have a 50 million threshold, potentially, with each counterparty. It’s bilaterally negotiated, but the regs allow you to use that. If you can put the trade with a counterparty where you have a little bit of space, that could potentially keep you under the line.
Let me just, for a quick moment, expand on that. We were talking about timing. Obviously, you heard Poseidon and Ted talk about get your docs in, that’s important because you’re going to get caught in a bottleneck. But there’s a couple other aspects of timing that are not documentation related and that’s one of them. If you have your 50 million threshold, there was some guidance and relief given by the regulators a couple years ago that said if you’re under the threshold, then you don’t have to have your documentation finished. A lot of people said, whew that’s great, now I’m free.
There’s a small aspect of that, that really, really needs to be considered. Once you cross that 50 million SIMM threshold, you have to be prepared to move collateral on that day. You don’t start your negotiation of your documents at that point. You have to be prepared, on that day, to move your collateral. So being able to monitor where you are against that threshold is absolutely critical. If you’re approaching the 50 million threshold, you’ve got to get everything in order. Not only your documentation, but all your systems and processes for being able to calculate the SIMM, issue the margin call, respond to the margin call and move collateral. It’s all got to be ready on the day that you cross the line. So it doesn’t give you quite as much relief as some people originally thought.
One other thing to keep in mind is, we saw in an earlier poll, some of the participants said that they’re not in phase five or six, which is great, but keep in mind that even if you are not crossing that ANA threshold in time for phase six, you still have to maintain below that ANA threshold into the future, otherwise you’re going to get caught up in the regs post phase six. So just as much as it’s important to monitor your IM against the 50 million threshold, it’s also important to monitor your ANA levels against that eight billion threshold because, if you cross that line, you’re going to be subject to the rules.
Ariel Berezowski: Let me ask you just one follow-up question. Is it mandatory to calculate that SIMM pre-trade? It’s good practice. Is it mandatory?
Joseph Spiro: To calculate the SIMM pre-trade is not mandatory. In the US, as Poseidon mentioned, it’s not necessarily mandatory for the buy-side to calculate the SIMM at all, but pre-trade certainly not. In other jurisdictions, post-trade it would be, but not pre-trade.
Ariel Berezowski: All right. Another question, this is more market color, so probably more for Poseidon or Joe. Legacy trades, are those coming in? Is there any push back from the sell-side? Any color you can give to the audience today?
Legacy Trades
Poseidon Retsinas: Sure. So, I can take that one, Arial. I think from what I’ve seen so far, in terms of legacy trades, the consideration has been to the extent that it’ll depend on the turnover of the book. If you’re a manager that has a large set of legacy trades, you don’t want to bring them into scope and they’re long dated trades so that you think they’ll stay there for some time. I think that’s certainly a conversation to have with your counterparty, to keep those out of the rules. There’s an operational impact there, of course, because if you’re keeping legacy trades out, but you have new trades going in, then you’re repapering for those, there’s going to be a dual process there.
From what I’ve seen do far, at lest for phase six firms, the decision so far seems to be mostly moving towards bringing legacy trades into the fold and not keeping them outside. But certainly, that doesn’t have to be the case. Maybe I’ll turn it over to you, Joe. I’m not sure what you’re seeing on this one.
Joseph Spiro: Yeah, it’s a little bit of a split and understandably so because, as you alluded to, it’s a balancing act between cost and efficiency. If you bring all of your legacy trades in, it makes your operational process a little bit more efficient because you only have to calculate one portfolio. But then the downside of that is you’re going to have to post initial margin on the full portfolio trades, not just the trades that were executed since the effective date. Frankly, your counterparty is going to have something to say about that, too, because all of the dealers, they have plenty of systems and processes in place to do that, to segregate those portfolios because they’ve been doing it for a long time.
If you, on the buy-side, don’t want to do that and you want to include legacy trades, that’s going to have an impact on your counterparty because now they’re going to have to start posting initial margin to you earlier than they might otherwise have to. So, that’s an important consideration, too, when you’re having those conversations.
Ariel Berezowski: So in the interest of time, because people are probably now going to their 10 am meetings, parting thoughts, 30 seconds in a row. Poseidon, 30 seconds, parting thoughts?
Poseidon Retsinas: Parting thoughts. It’s a complicated area, there’s solutions out there. I think the lessons learned from the other phases are going to be really helpful, in particular phase five which is really where the hedge fund and asset managers have come in in earnest. I think there’s a process to be followed. If you’re on track and you stay organized, it’s not that scary. I don’t think it’s going to be as bad in terms of what people were thinking initially, like that you have to repaper all of your relationships and all that, so I think taking a level headed approach and finding out what’s most important and tackling it, I think, is definitely doable and something that people in phase six are in a position to do right now.
Ariel Berezowski: Thanks. Ted, parting thoughts? 30 seconds.
Ted Leveroni: Take advantage of your service providers, your custodians, your legal contact and industry associations. They’re all willing to talk to you. Talk to them and test every assumption you have about what phase six is going to be. Timing, document load, tri-party versus third-party. Every assumption that you have, test it with the people that are willing to talk to you that have gone through the earlier phases. Biggest challenges, biggest problem, biggest rollbacks of all the previous phases, I think came down to incorrect assumptions. So you test them, take advantage of the knowledge, you’ll be in a much better place for phase six.
Ariel Berezowski: … Ted. Lastly, Joe.
Joseph Spiro: Yeah. When we talk about UMR, it seems like SIMM always gets the headline, but as we’ve discussed here, it’s far from the only consideration. There’s lots of other things out there. When you are subject to UMR, you’re going to take a hit as far as having to post more collateral, but what you can do is, there’s plenty of things you can do to make yourself more efficient to limit that hit as much as possible. So get your systems and processes in place to do that and you can limit the impact on your liquidity, which is the goal.
Ariel Berezowski: My side, my parting thoughts. It’s a long process, people should start working on it as the panelists said already. Most importantly, have some patience, it’s going to be a lot of back and forth with your counterparties, with the custodian, tri-party conversations, what’s eligible, what’s not. It’s a patience game. You’ve got a year to get it done and I wish everybody good luck and God speed on this. At this point, I want to thank everybody for being here. I want to thank the panelists, as well, for their time and have a great day.
Poseidon Retsinas: Thanks Ariel.
Ted Leveroni: Thank you.
Joseph Spiro: So long everyone. Have a great day.
