The NBIM Model

Hello, everyone, and welcome to another episode of Peer Connections by the Global Peer Financing Association. I am Brooke Gilman, Secretary of GPFA, and very excited to have one of our new members of GPFA joining me. Matt Brunette from Norges Bank is with us today. So, Matt, thank you for joining. Hello. Thank you for the invite, both to GPFA and to the podcast. Definitely. Is this your first podcast, I think, by the way? I know you have a lot of information out there and with recent articles and everything else, but is this the first time doing a podcast? It is. Big fan, long-time listener, first-time caller. Good. Well, great. Well, we'll make it easy and seamless and show everyone else how easy these things are to do. So great. So Matt is the Global Head of Financing at Norges Bank Investment Management and has been with Norges for probably, what, just under 20 years or so, I think? It'll be 19 years next week. Yes. So I started Easter 2002. Wonderful. Good. And so you've been in Oslo for quite some time, but I know the past had been at other organizations. And I was looking at your LinkedIn before, and it reminded me that when I first met you probably years ago, I realized that we do have something in common, which is we're both from Southern California originally, and we actually went to rival high schools. And I don't know if you remember that, but I went to Villa Park High School and you were at El Maldino, which was where we would play all of our games because you guys had the stadium. So anyway, small world. Indeed, it is. All right. So what we've been doing with GPFA podcast is trying to offer other members educational insights into securities lending. and through that have found it really useful just with some members that were interested to give an overview of how you approach your securities financing activities. And so I thought today we could start out with this podcast and maybe we can convince you by the end of it to do a second just to hear from you around how Norges runs your securities lending program. And I know you've had a significant amount of evolution since you started lending probably in the late nineties to where you are today. So I thought maybe we could just first talk about your model and get a sense of your team and how you approach securities lending and just start with the basics and we'll go from there. Yeah, absolutely. So I can probably say at the beginning that for a very detailed version, we did recently produce a history of our equity lending anyways, that's available on our public website as part of our investing in equities history. So it's a pretty good time. I'm pretty fresh off the subject, but starting a little bit with me, I have a background in securities lending from Barclays Global Investors, what's now BlackRock, followed by Morgan Stanley in London. And I was hired into Norges Bank, like I said, 19 years ago for a project by our former CEO, Inge Schlingstad, to look at securities lending. We had a standard custody model where we signed up for custody on the equity side in 1998. And they just kind of said, hey, we have this thing, securities lending. It operates in the background. You won't see or hear it, but it basically covers your cost of custody. And he had the idea of maybe there's more to it than that. We should look into it, understand it a bit better, understand what the risks are in it and what the potential is for a program like that if we take a more active role in it. So he hired me to do that project, but securities lending wasn't really considered a job. So I was fortunate to try out a few different things in the fund. I worked my first four years as an equity trader, learning about market structure, execution at a pretty interesting time when the fund was growing massively from the government's ownership of or taxation on North Sea oil sales. So I kind of did it as a side job, as a project, and that sort of evolved over time, as you say. But I think I was fortunate that he was smart enough to realize that maybe this belongs in the front office. Maybe this is a real investment strategy and we should look more into it. But a bit of background on the fund as well. We get a bit of publicity, but we're a pretty vanilla fund. We buy and sell bonds and equities, and there's a bit of real estate in there. So we use almost no leverage. The fund has a maximum leverage of 5%, of which we typically use 1% to 3%, something like that. We're very light on derivatives. We have a handful of listed equity in fixed income derivatives to manage exposure, a handful of IRS, but compared to the size of the fund, which is about a trillion dollars, the US are a bit over, it's tiny really. So we're really looking as a lender, just how do we create a revenue stream off of our inventory of assets? Great. And so going back to when you first started this sort of as a side project, as you say, being on the equity team and in the investment space was a bit unusual because I was sort of early 2000s. And for many beneficial owners in that timeframe, securities lending was just starting to be thought of in transitioning as an investment discipline and really historically still was sort of stuck in the back office. And as you noted, closely tied to custody. In terms of Norges taking an investment discipline approach, what did that mean from a culture and recruiting standpoint as you grew your program over time? And how has that probably influenced what some of those earlier decisions were that you made back in the early 2000s? We were lucky enough to have a visionary that as a CEO, Ingva Anders' predecessor, that turned a disadvantage into an advantage. We're a government-owned fund, so we couldn't just go out and hire resources. And they weren't very well-experienced resources to just come and bring a strategy to us. And they weren't really here anyways. You can get deep pools of securities lending resources in the financial hubs of New York, Hong Kong. We talked about Boston earlier as well, but not in Oslo, right? So turning a disadvantage into an advantage, we kind of had developed a culture for hiring people with potential, I think is what we always like to say. And there's something to that I learned over time as well, that you can take people that are bright and inquisitive and don't necessarily have a predetermined notion on what is risk and what is the right compensation for risk. And especially in our space in securities finance where everything's OTC, it's much more interesting to take someone without experience and say, how should this transaction be priced? What is the risk behind this instead of this is the way it's always been done. And this is GC goes for X basis points and collateral haircuts are Y. And here's a formula to how you implement a securities lending program. And on top of that, we got and give a lot of freedom to young people with potential to experiment in this space and to take some risk. But I probably caveat that by saying that that risk is also pretty tightly controlled. So like everything, I would say the NBIM culture is we're not afraid to go into new areas, but we start quite small. We take our time, we learn things by doing, and then later on, we make a scaling decision on that. And I think that's pretty true for our securities lending business as well. So you obviously then, once you started in the early 2000s, started to grow a team to layer in that control and oversight over lending, but you've always felt, I know that the securities lending function because of the intense operational activity has been best suited to remain with your agent. Can you talk a little bit about that and with the growth of your team and the oversight and the ownership over risk that you've taken with securities lending, how you've coordinated that with your agent over the years and how maybe that's evolved as well? Yeah, I think that's a great example, actually. So yes, we've always had an agent lender on both the equity and the fixed income side. As a bit of background, equity and fixed income grew up as two different business lines up until English Linkstown became CEO in 2008. It was really run as two different businesses and he brought those together. So prior to 2008, there was a fixed income business that had a fixed income securities lending product. There was an equity business that had an equity product, but we both used agents and did some small incremental transactions ourselves that we can get into. But yeah, we definitely piggyback the agency model from the beginning and continue to use that as the backbone of our program. So we're a very lean organization and securities lending is a very operationally intense business. There's a lot of transactions. We own 9,000 unique equities across the world, across 70 some countries. We lend in 35 of those countries. On any one day, we have over 5,000 unique equities out on loan. And there's settlement and asset servicing on the back of all of those across multiple counterparties. But back to your question on risk and evolution. So this is something that started quite small, where it was me, started as a project, as officially 25% of my allocation of time as a resource internally. After a few years, I was able to hire somebody to help me out, but it still became 50% of each of our jobs that we were also doing index portfolio management at the time. But it gave us a lot of insight into how the portfolio was run and also the risk metrics that we could apply to that. So we started looking at number one was how do we get data? When I started in 2002, we got hard copy reports at the end of every month and it gave you a general idea of who was doing what, but it was really nothing that you could do any analytics or any time-based analytics on. So we spent a lot of time on just collecting data and getting better at cleansing data and how we could view and analyze that data. And then we started to apply some ways of some portfolio analytics that we look at internally that eventually evolved into kind of a concept that we still use today on our securities lending risk is it's not quite the same as owning a portfolio because it's what we would call a default contingent portfolio to say we have a long portfolio that's our collateral. We have a short portfolio that's all our loan securities. And that's what we're going to own when a counterparty defaults. And so what we like to say is how good of a hedge is that long collateral portfolio for the loan portfolio? And more importantly, what's the risk neutral haircut required for us to accept that risk? And that kind of becomes the way that we look at it. That was kind of the starting point. And we spent two or three years just really creating a data model. And this is going back to 2004, 2005, 2006, just to be able to take in CSV files at the end of the day and build up a database. And then we spent another few years just creating really that were analytics on that, being able to view those portfolios, being able to view sector exposures in those long, short portfolios, being able to view liquidity metrics, et cetera. And then And eventually that led into us becoming more comfortable with that risk that, okay, we now have a history that we can look back on. We can start to do sensitivity analysis on it, et cetera. And that led to us being able to manage equity lending versus equity collateral on an indemnified basis. So the main point was, this is quite a long process to where we got to the point that I think has been discussed in recent years with other lenders to say, can we live without indemnification and what does it mean? And it's not a decision that we made overnight. Right. No. And back then, I mean, it's still unique today to live without indemnification or to choose to do so. Obviously, most of your peers probably are still utilizing agent indemnities in some form, and some would probably still feel that it's an absolute requirement. But back then, it was probably even more of an outlier. So again, I think it speaks to, as you know, the comfort and the education that you all undertook to gather all that data, model it out for a period of time and to really get comfortable and to understand where those risks are. Because you're right. I mean, indemnification, it hasn't been proven to be necessary really in many, if any, scenarios over the years. And so I think your sort of risk model approach. Maybe that's something we can get into in another podcast, just even diving deeper into that and going back in time and talking about what happened during the Lehman default and all that. So maybe we can do that in more detail. But one question I had as well. So you've talked about your use of an agent and your involvement in lending out across 35 markets and the level of involvement that you have and overseeing that from an operational risk perspective too. Can you talk about how you entered into new markets, because I know you guys are also definitely, in addition to sort of being maybe at the forefront of equity collateral and going unidentified, you've also really been at the forefront of opening up new markets and thinking about those markets, not just from a traditional SBL perspective, but also synthetic lending too. So can you speak to how you've approached that and how that's worked with your agent over the years also? Yeah, sure. Maybe it helps to start with how we view the agency lending model and the distribution of responsibilities between us and themselves. So we have a few more resources now. So the securities lending team has and has had for the last probably five or six years resources in each region. So in our offices in New York, London, Oslo, and Singapore, they liaise on a daily basis with the agency lending desks in those regions. But again, we see the agent as our infrastructure, but also our administrator and trading functionality, right? So we don't claim to have any real competency on pricing individual securities. We have very strong views on how we price risk. And I think, as you said, we'll get into that maybe in another podcast in detail. But the agent provides the infrastructure, they provide the operations, they provide trading services, securities pricing. The securities lending traders on my team are the liaison with the internal PMs. So a lot of what we can do in lending is affected by how the PM wants to manage their portfolio. If you think about corporate actions, term transactions, et cetera, especially in emerging markets where the inventory needs to be managed quite closely in markets that you really don't want to fail or can't fail in, right? So in addition, we have some operational controls, but we have mostly, to your point, risk management and strategic development. So the biggest part of strategic development is just getting markets or inventory available for loan. So in the early days, it was just saying, we have inventory in these markets, we would really like to make some yield rather than no yield on them. So I think to your point, there's a process behind taking in some of the earlier markets where they develop securities lending infrastructure. I think Korea is the first one that I looked at in like 2003. And the agent does 95% of the work, right? And ensuring that they can send free of payment transactions, that this is within the scope of regulation and within the law. And then internally, we do a legal tax and light regulatory review and just verifying the things that the agent has proposed. But then the interesting stuff got into other markets where there was quite clear demand as we started meeting with prime brokers in different regions to say, where is demand? What markets are you interested in? it was quite clearly markets where they can't access supply. Taiwan was the big watershed moment for us where there was tremendous demand from hedge funds for short interest in the Taiwanese market, but there wasn't a functioning stock loan program. So we piggybacked. At the time we traded TRS or single stock swaps on the equity portfolio management side. This is back when we had long, short portfolios. We don't have any more, but we were able to piggyback that infrastructure to do synthetic lending in Taiwan. And it was complicated. We were kind of on our own in how to set that up from a lending perspective rather than a portfolio management perspective. And we didn't do everything right initially, but we got it right in the end. And it ended up being quite a lucrative product for us. But the fun part is we didn't necessarily like doing that. Again, that wasn't our expertise in pricing individual securities and managing an infrastructure for securities lending. So once the Taiwanese market did develop, we rolled that all into the agency program, not overnight, of course, but we're quite happy to pay the agent for their services to manage that product going forward. And we did exactly the same thing for Brazil and for Greece around the same time period or in subsequent years. So once those markets established the proper SBL infrastructure, you then sought to transition away from synthetics when you could over time. Really, it sounds like mainly for resource reasons and where you want to be spending your time prioritizing your business versus what the agent can do for you, it sounds like. Yeah, absolutely. We now view the peer-to-peer space in the same way, in that you have securities lending and prime brokerage not only requires a lot of infrastructure, but requires a lot of networks as well. And banks are really good at that in bringing two sides of a transaction together and providing the infrastructure, both the legal infrastructure, but the operational infrastructure as well, putting that together. And we like to use that. We don't necessarily need to reinvent that. So even when we look at peer-to-peer transactions, which is, you know, in my view, kind of the modern version of synthetic lending, we're quite happy to use prime brokerage infrastructure that's already in place. We don't necessarily need to rebuild that. If you can partner with somebody on those transactions, we're quite happy to do that. So you've been in this space at Norton just nearly 20 years. If you have another 20 years in you, I don't know if you do or not, but if you were to, Do you think that automation and technology advancements that are happening already and are bound to happen over that period of time, do you think that would change your perspective over time where you might look to do more directly yourself to support that operational administrative infrastructure if automation was there for you to do so easily? Yes, absolutely. And that's coming too. So I'm not saying that this is the model that's going to work forever. We definitely are happy to look at development in the market. But to answer your question, yes, it's finally happening. I always kind of joke around that the securities lending space is kind of the last place to see technology, but we implemented a repo platform last month, you know, so it is happening. Right. It's coming to where we've, I won't say fully, but significantly automated our repo transactions that started off as Bloomberg chats and punching tickets into an OMS is now kind of a drag and drop is caught up to where equity electronic trading was when I did it 15 years ago. So absolutely, we definitely embrace technology automation and change in the industry. Great. And maybe just one other topic before we can wrap up this podcast. And I think that I've already convinced you during this conversation to maybe continue on to a second podcast where we can talk about your approach to risk management in more detail. But maybe before concluding this one. So one other area, you know, we've talked new markets, we've talked about your approach to synthetics, we've talked about the adoption of equity collateral early on and how you got comfortable moving to an unindemnified program. But you also, in those early years, one other area that you were sort of an early adopter was around exclusives. And I know maybe you're no longer active in that space today, but can you talk to us about what your experience was back then in the early 2000 and why you went the approach of exclusives and what your learning experience was over time with exclusives? Yeah, absolutely. So this was kind of in the very early days when I was just going around the market and kicking the tires and seeing two things, how other people were approaching lending and also on the prime brokerage side, where their demand was coming from, what they were interested in, and what are the assets that could attract a premium. So we did a lot of those meetings back in the day. We did a lot of due diligence and looking at different agents and different models. I probably owe a lot to Dan Kiefer at CalPERS, who's also part of GPFA, who was kind enough to share his model with me at the time and got me onto the idea of exclusives and how his experience was and success with that model. So we looked at those, we made an idea again to sort of experiment with it. So we started with a small portfolio, but it was a global portfolio, but it was maybe 3% of the assets of the entire fund that were captured in one portfolio. And we auctioned it off on a regional basis and we were happy with the results, but it was marginally better than we were achieving in agency lending. And then we did quite a bit of work on actual auction theory and how that works. And going back to, I guess, one of the cultural principles here is we try to incorporate a lot of academic work into how we develop investment strategy or use that as a starting point, at least. And auction theory kind of told me two things. One is if you want to get the best price for your assets, whatever it is you're selling. One thing is you chop it up into as many pieces as possible and invite as many people as possible to bid on it. And there were, of course, some practical implications to that. But for us, that meant everybody on our counterparty list would be invited and we would chop it up into country portfolios, which was kind of the smallest pieces we could conceivably manage. And then number two was whatever model you use for an auction, whether that's a Dutch auction, whether it's a first price open auction, whatever, should theoretically at least give you the same result as long as the rules are clear and bidders trust the process. So we spent a lot of time on communication with the borrowers that were bidding on this to get their input on how to structure this in a good way, in a fair way, and give them clear feedback that if you gave a competitive bid, we will give you, I think we did it in quartiles where they were so they could see how far off and ensure that the best price would clear the market. So we did expand that over time. It was very successful. We wound it up shortly after the financial crisis for a few different reasons. The short version is the PB model was pretty murky at the time. our portfolios have gotten huge to where country portfolios, I don't think any PB wanted to take an entire country from us. They were a bit unwilling to allocate capital to a business model that was under threat at the time. At the same point, from our side, if I'm being honest, we got a lot more conscious about gross counterparty risk. So we thought we were pretty good at managing net counterparty risk, but the size positions involved in exclusives where people just had really unrestricted access to assets in there was quite large. But to your point, we haven't used it in a while for different reasons, but it's definitely still on the menu somewhere. I mean, synthetic lending is something that came back. We did that initially for access markets. We do it now for GC markets. So after an eight-year hiatus from synthetic lending, it's now a strategy that accounts for 10% to 15% of our program probably. So I certainly wouldn't rule it out. Well, I think one theme just from listening to the past half an hour or so is really just that you guys continue to pay attention and adapt and evolve. I think the biggest thing that maybe other members of GPFA or other beneficial owners could glean from this is that this is a constant moving target that you need to continue to educate yourself, continue to stay informed. And if it didn't work before, it doesn't mean it's not going to work now. So revisiting topics and doing as you did as well, speaking with peers and getting a good sense of what the counterparts in the market are interested in, what other peers are doing, you know, all of that I think is excellent and obviously speaks well to the mission and the focus of GPFA as well in terms of information sharing. So this has been wonderful. I've really appreciated this conversation. I know that other GPFA members listening and other listeners probably really appreciate hearing from you on how the Norges Bank model works for securities lending. And definitely we're going to do a second one, I think. So maybe we can find time in your calendar to get back on the mic and go for a round two, if you're willing. Sounds pretty good. So thank you Brooke. I appreciate the compliment. I appreciate the invite. We also are quite excited about GPFA. I think there's a lot of innovators in this group and we want to be parts of those conversations. So much appreciated. Great. Well, thank you, Matt. And to our listeners, thank you for listening. This has been another episode of Peer Connections by the Global Peer Financing Association. And as always, do not hesitate to reach out to GPFA through LinkedIn or through the GPFA website. we'd love to hear your feedback we'd love to hear ideas for upcoming episodes other topics that you'd like to discuss if you're a member out there listening and you have something to share as well we'd love to hear from you so you all know and you've heard it here that we hopefully will hear more from matt in the coming weeks but please don't hesitate to reach out if you have other ideas or topics that you'd be interested in thank you everyone have a great day