Securities lending and tail risk

Hello and welcome to another episode of Peer Connections by Global Peer Financing Association, GPFA. I'm Brooke Gilman, Secretary for GPFA, and I was able to convince our prior guest to return for a second podcast. So again, today I have with me Matt Brunette from Norges Bank Investment Management, who is Head of Global Financing at Norges Bank and has been overseeing the securities lending program for Norges for the past 19 years when he first started the program. So Matt, thanks for returning. I appreciate that we didn't scare you away after the first episode. So thank you. No, thank you. Happy to be back. Good. So we covered a lot of topics in the first episode that we did. And one that you spent some time on that I thought would be great to go into more detail is Norges' approach to securities lending risk management. And you guys, I know that's sort of the core of what your team and your expertise you've built up over many years. And that's how you've been able to get comfortable being a first mover and being an innovator in some of the areas, whether that's collateral, whether that's market risk, whether that's different types of transactions and how you approach counterparty risk as well. So I thought we can maybe talk in more detail about your approach to securities lending risk. And maybe if you want to just start high level for folks and maybe those that aren't as in tune as a market practitioner for securities lending and give us a sense of how you view securities lending risk from a high level perspective and then your approach to actively managing that. Yeah, of course. So starting quite generally and reflecting a bit onto our last conversation, this is something that has evolved over time. Securities lending development takes time. There's both building a strategy and the expertise that comes along with that strategy. But like most of us in this forum, we have a, whether it be a board or a management group or an executive body that we need to anchor these things with. So that's a process that takes time. And we built this up incrementally over time. But the main point here is there is risk in securities lending. It's definitely there. We would classify that as tail risk, but it's securities lending or risk in the financial industry in general, it's these hundred year events that happen every 10 years, right? So there's a mix of theory, modeling, and preparation that goes into this. But yes, for better or for worse, we spend a lot of effort and a lot of time on securities lending. And I say sometimes for worse, because having an active view on risk means we actively have to say no to some trades that would have a positive P&L in them. And that's part of what we do, is doing a lot of work and saying, sorry, at the end of the day, we just don't think we're being compensated in the right way for this type of risk, or we're not comfortable that this type of collateral will be liquid and tradable in the case of a counterparty default. And if I go back a second, just for clarity, we have a bit of a division of responsibility within the fund on how we manage and approach this risk. So we do have an internal group that's responsible for our counterparty credit risk, and they do a lot of work on analytics of mostly bank counterparties and due diligence into what the actual counterparty risk is. So that's a group that has specialist professional competencies that doesn't exist in my group that we lean quite heavily on. And then my group focuses 100% on the actual portfolio or asset risk underlying securities lending transactions being the loan securities or the collateral. So Matt, because you have the counterparty analysis separated from the view that you're taking on the lent assets versus the collateral being pledged, and it's your team's responsibility to ultimately construct a risk neutral collateral portfolio. Do you ever vary some of your parameters in terms of acceptable collateral or even margining at a counterparty level? Or do you really look as though all counterparties are the same when you're doing your collateral portfolio construction? Yeah, well, we do a bit of all of the above. So I think it's important to look both on the micro and the macro level. So yes, our starting point is in a default, you're typically exposed to the legal entity and the contract under which that legal entity is contracting with you. A lot of the counterparties we deal with have several entities under a parent and we're transacting with the US entity and then their international entity. And these days, the trend for banks is to segregate even more. So EU entities are creating Australian entities, Japanese entities, Singaporean entities, where it's more efficient for them to finance their business that way. So we have to look as a starting point on if one of those entities could default independently, then we have to be able to manage that risk in isolation. But it matters also when you aggregate that up to a parent level of a company. And also when you aggregate different counterparties or different banks together, do you start having large concentrations of risk? So we want to manage the overall portfolio level. But specific to your point, yes, we differentiate everywhere we can. So the issue is if we could have a prime brokerage type discretionary model where we take in every day, we analyze the portfolio and the collateral and we say, okay, well today with what you've given us, the haircut is 11%. And yesterday it may have been eight or it may have been 13. We can't really do that. We're sort of stuck in the triparty world where we, for practical reasons, we have to have somewhat static portfolios and we change those quite actively, but it takes time. So we have to look at different counterparties because they tend to have different concentrations. They don't necessarily know on a day-to-day basis what they're going to be able to give us as collateral. Most of these are global prime brokers that are operating in pretty much all the markets we deal with. So we want to give them a flexible schedule where there's a lot of places they can give us collateral. So we accept, as an example, equity collateral in 25 different markets. So in a perfect world, we'd say, well, when we lend you something in Australia, when we lend you an Australian bank security, we would like an Australian bank security as collateral and create a perfect hedge, but the world doesn't really work like that. They're subject to their client positions and they want to pass those through. So we say, okay, we have this big pool of 25 countries that we can offer, but we never know on a day-to-day basis what we're going to get. So we have to make some assumptions that there's going to be some concentrations in there. So we charge, I think, a slightly higher haircut to compensate for that than we otherwise would if we had known on a pre-trade basis exactly what we were going to get. And to your point, we do differentiate across counterparties because they never know what they're going to get, but they tend to specialize or have sticky exposures in certain places. And we will adjust haircuts and collateral sets to accommodate, but to be compensated on a risk basis for those exposures. So let's jump into, and it might be slightly out of order, but I feel like there's a lot of questions that I have that probably relates back to what your experiences were during the Lehman default. So maybe let's jump to just that. What was your experience during the Lehman default? Because for our listeners, if I'm correct, prior to Lehman, you obviously, if not a handful of years prior, had adopted an equity collateral model where you were taking an equity collateral versus equity loans. So there's nice correlation there. But what was your experience leading up to Lehman? Because it's not as though you were lending out equities and accepting in government bonds, where inevitably, if equities fall and government bond prices increase, you're going to be very comfortable in that right-way risk. You instead probably have a risk-neutral approach. But as you just note, depending upon the market that you had out on loan and the collateral being pledged across all the markets that you would accept, you could lack a little bit of correlation at a market level, perhaps. How did that go? And how did you manage through that? This is the fun part, right? As we started getting into in our last conversation, about the evolution of our risk management, how we went from collecting data to analyzing data to eventually getting comfortable with equity collateral and the lack of indemnification. And then Lehman gave us the opportunity to test it out in the real world. So one thing that I don't want to leave out on this is there's a process behind this as well. So one thing is making really nice in the modern day world sort of Tableau dashboards and analytics and making time series and sensitivity analysis is on your portfolio, going back to the 90s and the 2000s and the financial crisis and the Russian debt crisis, et cetera. But there's a process that I'm thankful that we put a lot of time into and what actually happens in a default and how do you do it? How do you deliver notice of default and who's responsible for that? Is it the agent? Is it yourself? How do you actually get a hold of collateral? How do you transfer it to a portfolio that's tradable out of Tri-Party, et cetera. So we're fortunate to do that work. To your point on risk, we have a lot of conversations about this internally about right-way risk, wrong-way risk. My starting point is we just want to be hedged. We want to be risk neutral. I don't want to try to predict which way it's going to go next time. I think in this day of government support, we can imagine a scenario where you have a long, brutal market slow meltdown that ends in a default and government support and the market rallies after that. And I don't want to be in the right way risk category if that happens, right? So back to your question about Lehman, it was a nerve wracking time, but it proved the model. Lehman was announced, their insolvency was really the event of default on the Sunday, 15th of September, I think. And we had notice of default delivered to Canary Wharf at just past nine on Monday morning. We had global exposure to Lehman. So we missed trading in Asia, but we got about half of the European day and all of the US trading day. And fortunately, there was somebody standing at the front desk of Lehman and Canary Wharf just taking delivery of these lines of couriers from law firms in London delivering these defaults. But it was huge to be actually in the market and trading, even though the market was chaotic. And like I said, we didn't do everything right, but we got there in the end. And if I'm fessing up, one of the first lessons learned earlier in the year after the default of Bear Stearns, we took Lehman off our list because they had a very similar business model, we thought, very similar exposures. Well, we didn't think, I think the whole market thought that. We took them off our list. We came back. We did quite a bit of due diligence. We had a relationship, obviously, through equity execution. We'd kick the tires on their prime brokerage business. They were very open with us. We saw that and we said, this is a great company with great technology, great people, good risk management. But the lesson learned is we only met the equity people. There's more to their business than just prime brokerage on the equity side. And we added them back onto the list prior to their default. So there you go. And then the second thing is that we didn't quite appreciate as much. And we talked about we got live trading Europe was the daylight risk. So we had a bit of lopsided portfolios between loans and collateral between regions. So we were selling the US and then buying Japan the next morning. So massively exposed to market movements overnight between the gap between US and Japanese trading hours. But yeah, we got there in the end and we returned excess collateral to Lehman at the end of the day. Great. I think that's a huge testament to yourselves, but also securities lending and how it functions as a collateralized transaction and that orderly process, even though it's highly stressful when it occurs, but ultimately it's a relatively orderly process to unwind and to liquidate collateral buyback in the positions. And when you're collateralized with sufficient haircuts as you were, then ultimately you're left then probably for years figuring out how to get that money back to the Lehman administrators, I'm sure, right? Exactly. Yes. So obviously that was a very useful experience to go through in terms of managing a securities lending program, even though it would be nice if you didn't have to live through the Lehman default, but I'm sure that's informed many of your other decisions. I know when we had the last podcast, you talked a lot about thinking differently about your gross exposures at a counterparty level and such, but how do you approach today and your thoughts around current market pricing of risk? And what are you doing today that maybe differs than what you were doing back then? Well, one thing we're doing different, we talked in the previous conversation about post the financial crisis, we merged the equity and fixed income products. And we did that in securities lending as well. So in 2010, I got responsibility for fixed income lending as well. And the nice thing about that is it always helps to be able to merge more pools of risk into the same big pot. So you get some diversification effects and you get some netting effects in there. So we did that and we developed a fixed income strategy in 2016, which is mostly when we think about equity lending, we think about short interest and we think about fixed income lending. And I'm talking governments here. We think about funding and that's a very different type of risk. So that's really on the equity side, you're supplying hedge fund shorts on the fixed income side, you're financing their leveraged longs. So for the most part, it's a lot of lending high quality liquid assets to fund equity positions on term. And that's why the market would refer to as wrong way risk. But we think that's very manageable if you do it in the right way and critically, if you're compensated for it. So we, again, spend a lot of time looking at that type of risk and how it fits into the overall portfolio of risk with the counterparty. So if we have a base of lending business that's well diversified, you can add some wrong way risk on top of that, that I think makes sense from a risk perspective, as long as you are getting paid for that. And so what I think we really kind of peaked out that strategy, probably with the rest of the market in 2018. And then in 2019, we started saying no to more trades than we say yes to because pricing really changed. And we didn't think that where some trades were getting priced, especially on longer and longer terms, that you could really say at the end of the day that I think I am accumulating enough revenue to compensate myself for future losses if there is a default here. So our exposure started winding down with that pricing. That sped up significantly in 2019 as pricing deteriorated even further. And then a year ago when the pandemic crisis hit the market, we thought that was a turning point and we're quite optimistic. One, it was quite nice because we were at low levels of risk at that time because we just hadn't been putting on new trades or rolling existing trades. And so we had oodles of capacity and the market woke up and we could finally fulfill our role as a liquidity provider in that environment. So we were on the offer there. Volatility in the market, of course, spiked. We significantly increased haircuts. We significantly increased fees, even on what we were previously transacting on. And we were getting hit on those. The only problem is that it only lasted a few weeks. And then every central bank in the world was kind of sitting in front of us on the offer. So I think it's hard. I think we're in a pretty tough spot for the industry where across both, I don't want to focus too much on fixed income lending, but equity and fixed income lending, where there's a lot of deals or a lot of trades getting done on terms that we just wouldn't consider and we can't make the math add up that you're being compensated for risk. And I don't know what eventually causes that paradigm to change, but I think it's getting to a point where you're cutting towards the bone that even there's externalities in lending as well. It's not just about revenue and absolute risk. It's also, we have to think about, I think as ESG is in focus now, people start saying, I'm also losing some of my shareholder rights in some of these transactions. Am I being compensated for that? Am I being compensated for eventual fails and claims, et cetera, that you probably didn't consider much before when spreads were wider? Sorry, that was a bit of a long-winded answer to your question. No, no, no. I have lots of things to say in response, but I'll be efficient as well with our time. But one question I have, so again, this is a GPFA Peer Connections podcast. And as you know, one of the many focuses of the GPFA is not just to bring members together for education and best practice discussion, but also potentially creating a network for them to be able to meet each other and talk about potential opportunities if members are interested over time to link up and potentially do business together. But I'm curious about your thoughts as it comes to fixed income and more the funding type trades or the collateral type trades, what your maybe future vision might be as you think about potential peers. Everything you were talking about in terms of some of the bank risks and taking off a lot of those exposures for a period of time and just the pricing of all of that. If you think that that changes much when you then think about potentially doing fixed income type collateral or liquidity type trading with other peers. Yeah, I think the fundamentals are the same for us. Our approach to risk overall wouldn't change much. So all we're saying at the end of the day for fixed income type funding trades or any secured financing trade is we have to assume that any counterparty could default. And at the counterparty defaults, we just have to say, one, can I actually trade this asset in what's likely to be quite a volatile, illiquid market. If you go back to March last year, where we were kind of heading in that direction and say, could I trade my corporate bond collateral portfolio? It was kind of hard to trade boons at the time, much less corporates, right? Can I trade SPACs or convertible bonds, et cetera? And then if you can trade it, is the haircut enough to compensate for the volatility. And again, if you go back a year, some of the standard haircuts in the securities lending market, if you just accept what we call industry standards, don't fly. I mean, you had, I don't remember off the top of my head, but in the month of March alone, I think you had eight or nine days when the S&P traded in excess of 6% or 7%. And you had a 10% and 11% movement intraday on the S&P. That's massive. And not a lot of people have haircuts that are going to cover that type of risk. But back to your question, the counterpart, one diversification and counterparts would be a wonderful thing because right now we all face banks and they are extremely correlated risks. So if you flip it around and you look at the hedge fund model, So big hedge funds in 2008 going into the crisis probably had three prime brokers and unlucky if one of them was Lehman, right? After that, they said, okay, well, we need to diversify our prime brokers. And now they have six, seven, nine prime brokers. But if you step back at that and you say, well, they're all kind of the same. They're all facing the same risk factors. Some may be incrementally better rated or better managed than others, but they're the same risk factor, if you will. So having some diversification in there would be great. You can probably make the argument that there's a lot of pension funds that are better credit risk than some bank. Yourself included, not being a pension fund, but being a much better credit than most of your banking counterparts, I'm sure. So other peers would look fondly upon that. Well, I think we'll probably look to wrap this up just to keep our listeners anxious for more. But Matt, this has been really useful. And there's still so many topics that I would love to explore. And I know GPFA members would be interested to hear more about and things that we didn't hit upon. And Sam, you touched very briefly on ESG and your voting stance and sort of how you manage that. We also didn't get into and all of this risk management views. I know you have views as well on the pledge model, on what you guys think about central counterparties. So I thought I was lucky getting you back for a second one, but maybe we can convince you to even do a third at some point. So again, thank you, Matt. Really appreciate both your involvement with the GPFA group. I know that that membership is growing nicely and now has an asset base with yourself included and with Norges Bank included. Gosh, I think it's over six and a half trillion now in terms of the breadth of the asset owner size of that group. And so a small group, but mighty probably is a good way to think about the GPFA at this point now and just over eight months in probably to its founding. But thank you again and really appreciate you doing this. You're very welcome. And thanks again for the invite. My pleasure. Great. And so to our listeners, this has been another episode of Peer Connections by Global Peer Financing Association. And as we always say, we love to hear from you. We love the communication and we'd be very interested to hear what other topics of interest that you'd like to hear about on these podcasts, or even if you are a GPFA member at upcoming member meetings, or whether it's a quarterly meeting or some of the subgroup discussions. So thank you again for listening. And please don't hesitate to reach out to GPFA through the GPFA website or LinkedIn. Thank you so much. Have a great day.