Stephen Gilmore – CalPERS' Total Portfolio Approach (EP.486)

27:50That has implications for liquidity because in both Australia and UCLA and Canada, for that matter, when there's a negative shock, equities are going to fall, but those currencies will also weaken so there's a liquidity consideration. That's different if you happen to be a US-based investor or it has historically been different. That's one consideration. You've got to look at the big factor, which is equities. And you're normally looking at regression analysis, but then again, you've got to think about What is the market value at a point in time? And these things are infrequently marked. There can be a lot of discussion and debate.

28:26In the end, you want to get something that's reasonable. As you go to implement, how do you think about, in the context of funding some risk budget in the reference portfolio, directly managed versus outsourced to managers? From a COPA's perspective, we do manage some of the public liquid markets internally, but it's more difficult to do that in the private markets. It relates to skill sets, the size of the team, the breadth of expertise. I don't anticipate that we will be particularly active direct privates apart from co-investment. But I expect that we will become more active in the public markets given the balance sheet management and our improved liquidity management.

29:14As you're looking at making these trade-offs and particularly making these shifts over time, what does the data and information that you need to aggregate look like on a dashboard on your desk so that you can make an informed decision? The data and analytics are hugely important. One of the things we have been doing at Kelpers is embarking on an effort to simplify some of the systems we're using to get that better whole of portfolio view. That's a multi-year exercise. You essentially want to be able to aggregate in a common language. historically we've tended to have best of breed your applications by asset class that can be great for a single asset class but it's not so good when you want to combine everything so you've got to be thinking about the right trade-off between that asset class functionality and the whole portfolio bias is to try and have a better view at whole portfolio what is it that you're looking at to understand what you own so that you know how markets might be impacting what's in your portfolio?

30:25A lot of the portfolio risk is going to be dominated by equity risk. The 75 equity, 25 bond reference portfolio, that simple construct will do a good job of approximating what our actual portfolio looks like. So you can stress test it. It may be that some things have a stress beta, which is higher than what you might get with the 75, 25, some things might be lower. What's most important is to be thinking about how the portfolio performs under different scenarios. People have looked at historical scenarios. They've looked at actual events. But of course, any of these shocks that you're going to experience is probably going to be different from the past.

31:09There may be similarities, but you're not going to get an exact repeat because people learned market structures are different and so on. Quite important to understand how the portfolio performs given a growth shock or given an inflation shock or given a real rate shock or a risk premium shock, other structural changes. Scenarios will play a more important role as we go forward. The reality is you cannot immunize the portfolio to all these different shocks because then you won't generate any decent return. It's more about understanding what might happen, being prepared for those. And if there are some outcomes that are unacceptable, then you can do something at the portfolio level to mitigate those risks.

31:50We're going to take a quick break in the action to tell you about Ridgeline. Imagine starting your day with reconciliation already done. No spreadsheets, no breaks to chase, no duct tape holding systems together. Ridgeline is the first front-to-back system of record built for investment managers. One platform, one real-time data set, embedded AI. Investment firms are replacing a patchwork of isolated and dated order management systems, accounting systems, reporting add-ons, and client tools with this fundamentally new operating model. Automating complex workflows, scaling personalized client experiences, and unlocking the full value of AI.

32:33If this is the year your firm is ready to modernize operations and harness AI, you can learn more at ridgelineapps.com. And now, back to the show. So at any point in time, once you have your portfolio built out, the marginal investment you might want to make needs to be additive, marginal contribution to the portfolio. Conceptually, yes. The practicality, of course, is difficult to compare every single investment with every other one. But if you have that common language, you can approximately do that. You've also got to look at the investments that are already in the portfolio. There's not just the new ones.

33:12Even the ones that are in the portfolio now continue to have to earn their place. In the strategic asset allocation model, that incremental investment is probably someone's assessment of better alpha. If it's a new manager and public equities, we think that manager is better than the manager we have, maybe in the construct of what we're trying to find. What might the similarities and differences be in that incremental investment in a TPA approach? With an SAA, the asset class is probably thinking about how additive that investment is given the asset allocation. Let's say it's an asset class with a 10 % allocation.

33:51They'll fill the bucket up to that 10%. Now, it could be that it's suboptimal. It could be that the return from the marginal investment in an asset class is less than it could be in another asset class. Or possibly it could be that it's a lot better and the team should be doing a lot more. You could be under-invested or over-invested depending on the relative attractiveness. If you've got a 10 % allocation, you're probably going to look to diversify that portfolio. You don't want to have a too concentrated an asset class portfolio. But when you're thinking about its contribution to the whole portfolio, you should be much more comfortable in having a more concentrated asset class portfolio because it gets diversified away at the whole portfolio level.

34:38Those are some of the differences. It's also one of the reasons why it's hard to hold an asset class as accountable in a total portfolio approach because the asset class may have been asked to do something for whole portfolio considerations. The assessment of the contribution is really the contribution to the whole portfolio rather than just looking at the S-class on a standalone basis. If at any given time you identify an opportunity set that is a good fit for the portfolio and seems particularly attractive in that current environment, how do you think about sizing? You want the sizing to relate to your degree of conviction in the investment.

35:19You're also going to be looking at the overall portfolio characteristics. Typically, any individual investment isn't going to make that much of a difference at the whole portfolio level. How do you think about how to measure appropriate diversification in a TPA approach? Scenario analysis is important. Let's take a simple example. Equities and bonds. Do they diversify? Is a bond exposure going to diversify and equity exposure? Well, it may. It depends on what's happening. Take the example of an inflation shock. If inflation goes up, nominal bonds are going to be hit. Equities are probably going to be hit as well.

35:57If you have a growth shock, it's going to be the opposite. Equity is going to benefit. Bonds are probably going to be hit. So bonds are diversifying there. You need to look at what's driving the event. Rather than simply looking at historical correlations, one needs to be thinking about multidimensional scenarios and to think about how the portfolio behaves given those scenarios. Among the tried and true principles that have worked for a long time for some of the strategic asset allocation models, rebalancing and private market exposure always comes up. How does the concept of mean reverting rebalancing work within a TPA approach?

36:35I don't see any difference. Future Fund, New Zealand Super would be rebalancing. Typically what you would do with the forms of total portfolio approach that I'm familiar with is you would have a target portfolio anyway. You would be aiming for something where you've got a reference portfolio, you're rebalancing the risk back to the reference portfolio level. If you're the Future Fund and you don't have a reference portfolio, you'll still want to think about equity equivalent exposure, and they'll want to rebalance back to that. It will be similar. In terms of the privates, that's an interesting one because with private market exposure, you can't move that anytime you want because they're illiquid.

37:19And the relationships involved, the investing teams will need to have clarity over a multi-year runway. In the organizations I've been in, there's usually some sort of runway or plan over multiple years. The target portfolio has to take that into account so that the teams have decent planning horizon and so they can manage relationships. How have you thought about the trade-offs of active and passive within the portfolio? In terms of reference portfolio example, that's passive. you will want to take active risk when you think you're going to get paid for it. It's as simple as that. How has that come through about where you're choosing to invest actively?

38:07It helped us in the past. The equity team, the fixed income team, have been good at generating good information ratios, but it hasn't been scaled, which I found interesting because there have been constraints on the active risk that can be taken and sub-constraints. I look at this and think, well, these are great information ratios in some of these strategies. Why aren't we doing more? Then there's a question as to how far these can be scaled. One of the big differences in the seat today from where you've been is the size of the asset pool itself. What are some of the areas where scale helps? Size helps a lot when thinking about the cost of transacting.

38:50You can improve your negotiating position because of your size. You can have very strong partnerships with economics works for both parties. I see the benefits to that. We've been taking advantage of that in recent years in our private equity portfolio. There was a period where we were under-invested in private equity and the strategy has changed. It's been particularly successful since 2022. The focus there has been on relationships, having that partnership, having that alignment. A key part of that has been getting more co-investment. There are economic benefits to that. That partnership works when you get better information flow.

39:32We've done a good job of selecting managers. Those things are more likely when you've got a somewhat larger team and asset pool because you can cover more ground and you can get access. You can't necessarily scale in the same way. On net, it is an advantage in an area like that. There are going to be some places where it's a disadvantage because you can't scale. What are some of those? If I was thinking about hypothetically, Stata Hub or something, how much could we do? When you've decided there's a particular opportunity where it makes sense to pursue active management. How do you think about the size of an active manager, their assets under management as a fit for the CalPERS portfolio?

40:14Given our scale, it's hard to make small investments because they don't move the tile. Some of these emerging managers can be outperformers. You want to take advantage of that. So you need to find the right framework and mechanisms for getting access to those smaller emerging managers. The reality is that the bulk of capital is going to be invested through larger managers because of the size of our portfolio. The governance boards, guardians, New Zealand, Australia, are thought of as very sophisticated investment pools of capital. And in the public pensions in the US, typically the people serving on the boards do not come from finance backgrounds.

40:54How has that changed how you thought about approaching the portfolio? It's true. The nature of the boards are different. The boards in Australia and New Zealand, the ones I dealt with comprise investors. There's a different type of conversation. But Kelp is the board. It was the ultimate governance body. Our move to a total portfolio approach pays attention to that. It's a management team that has the investment experience. We should be accountable for it. It becomes clearer. The board has the overall oversight, the asset liability management model remains the same. We've just moved to a reference portfolio and we define the active risk a little differently.

41:36Ultimately, there's more clarity around who is accountable. It's the management team that has the investment experience. The board has the governance experience. What are some of the subtle favorite features of yours that you've seen at TPA that you'd like to apply to CalPERS? one thing that stands out is aligning the active risk to a level of conviction. If you've got lots of silos by asset class, you don't necessarily do that. Back to New Zealand super, I used to describe the active performance of New Zealand super in baseball terms. If you look at all the different investment strategies, when I first looked at the analysis, I thought, well, actually the hit rate or the banning average is pretty ordinary.

42:26That's not that unusual. But the slugging average was amazing. The areas that it did best in, it had a lot of risk allocated. Those areas that it performed best in were the areas of highest conviction as well. That was something that has been a feature. That's something that's important for Kelp is to think about where our advantages are and to make sure we're allocating capital proportionately. When more and more people are thinking about total portfolio approach, where have you found that others are well positioned to do it that may not, or where they're really not going to be able to replicate it?

43:04It's probably smaller teams that are better positioned. In our case at CalPERS, yes, it's a larger team, but we had some of the enabling conditions. we had the alignment in terms of the compensation, the improvements in terms of liquidity management, and we've got a capable team. Those things are all helpful. The fact that we'd embarked upon a data and technology transformation was also helpful. To make it work, you need to have good collaboration. That's a key thing. Having people that speak with one another are aligned with the objective. How have you tried to bring that culturally from what you've seen to operate in that type of collaborative way?

43:45One of the things we've done at the level of the leadership team is to call out collaboration. When people are assessed for performance, collaboration is one of the key leadership competencies. We've heightened that in terms of how much focus we give to it. There's focus on communication, outreach. We've had lots of questions and answers and discussions. When you do that in the big forum, people don't feel that comfortable speaking up. There's been a lot of outreach at the team by team level. Early on, there was some discomfort in the private markets because people were thinking New Zealand doesn't have much exposure to private markets.

44:25The reality is Future Fund does. And if you look at the TPA adherents, they tend to have a bigger exposure to private markets. So there are a lot of misunderstandings because you've got to look at the organization and where its relevant advantages are when thinking about the asset allocation. There's been an education process, Q &A, listening, discussion, and it's ongoing. How do you anticipate making investment decisions. The key is to make sure that you've got the right blend of bottom-up, input, and top-down. When I think about the top-down, you're thinking about the overall risk levels of the portfolio.

45:06You're thinking of the active risk budgeting and the way that gets allocated. From the bottom-up, you're wanting to make sure that you see the ideas being socialized and being shared across the organization. In the end, if you are deploying capital, you want to be confident that it's going to beat the cost of what you're selling to invest it. You need that consistent framework for thinking about that. If someone wants to invest in infrastructure, we're going to effectively have to sell some equities and bonds to do that. If it's infrastructure, it's probably going to be illiquid. There's going to be some charge for that.

45:41The investing teams are going to have to be thinking about what's the opportunity cost of making this investment. And if everyone is looking at the same framework for assessing that opportunity cost, then you've got a model for making those investments. The teams will go out and try and find good opportunities. It may be that when you look across all the different opportunities that there are some relatively good ones and some areas and some that are less interesting. And we should be able to discuss that as a team and upsize those ones that look more attractive. When something's close on the margin, who ultimately makes the call?

46:19A lot of us delegate it down to the heads of the various teams. We raise the biggest questions to an internal committee of the various asset class heads. Ultimately, it typically is the head of the asset class that makes the call up to a particular size. Do you bring the heads of the asset class together so they're not just thinking about their asset class. Absolutely. Essential. Internally, we have what is called a total fund management committee. We meet fortnightly. We also have a similar committee that looks at the underwriting of deals, but those will only be the largest transactions. Those people get together fortnightly on these committees.

46:59As you think about stress test analysis and your risk assessments today, what most concerns you? When I look at our portfolio, you're exposed to growth. I would like to have a greater exposure to diversifying strategies. Those things are things in the back of my mind. In terms of what causes me to lose sleep, it's probably there being some unfortunate event and it's hard to forecast. You hope that you have the maturity to look through that and to take advantage of those unfortunate events. It's probably a function of behavior. How do we react when some adverse event occurs? I think about that.

47:44Do we have sufficiently strong governance arrangements to have gotten through that pro-cyclicality that's been a challenge for us in the past? Have you thought about increasing use of technology, maybe AI, in improving the efficiency or the output of the investment process? It's a continual process. We've spent a lot of time working on liquidity analysis, the analytics there. One of the things I did when I first came in to Kelpa's was to say that New Zealand, I could see the portfolio on my phone. I could see lots of different reports. We couldn't do that at Kelpa's when I arrived, but now I can see a lot of reports on my phone.

48:22The team have seized that. So really good initiative to make that work. With the breadth of asset ownership over a long period of time, imagine there's a lot of data that comes from that. Have you thought about the potential for AI in improving what you can know? It's a hard one because I don't know that I can necessarily access all the data that would be relevant. We were early investors in private equity. It'd be great to be able to look at the lessons that were learned early on, but I don't know how accessible that information is. there are probably missed opportunities. Going forward, we can build information systems to capture more than we have in the past.

49:06That's an area where we should have a true advantage given our scale, given the connections, given the access that we have. Given the seats you've sat in, you're particularly well-positioned to understand how large allocators might change their investing over the next five or 10 years. We'd just love to get your thoughts on there. Allocators often think about the past and project forward. There's always a theme. There was the Yale model. Maybe it doesn't work so well now. Then there's a Canadian model. Maybe there's some issues with that as well. Now people are talking about TPA. It has to be fit for purpose.

49:46Going forward, TPA will get more attraction, but it's hard because you've got to have that collaboration. There's going to be more thinking about the privates versus publics. Because you can see the efforts to try and give retail access. Interested in how that all plays out. Any other general thoughts? Right now, a lot of those entities, asset owners outside the US, are conscious of the great exposure they have to US assets because of the size of the US capital markets, the performance of the US equity market, and so on. There's a desire to diversify away from the US. That's probably going to be difficult because the other capital markets aren't as deep.

50:25the entrepreneurialism in the US is a standout, but it is going to be a theme over the next period of time. There's also going to be more thought about other regulatory interventions. I know that some regulations have been scaled back. Issues of national security become more relevant, whether that's security or supply chains. I can imagine there are possibly going to be more natural security related factors to be thinking about when it comes to the financial markets. One thing that I wouldn't like to see would be forms of capital controls, but that's quite possible. So we need to think about those sorts of things, things that haven't been in play for decades.

51:08They were in play back in the 80s. How do you incorporate that into the implementation of the investment portfolio? it's less of an issue for us in the U.S. than it is for others. It really is thinking about stress tests and scenarios. When you put together a series of stress tests, how do you think about what historical scenarios you want to populate it with or what future economic scenarios you put in that are enough but not too much? We used to do this at New Zealand super. I sprung a stress test on folks at Future Fund back in 2011. We've probably done less of that at CalPERS, but recently we had a great little exercise, which was initiated by the team, and this was a liquidity shock.

51:56When you design these things, you have to make them realistic. You want to get engagement from the team. I think back to the one we did recently. People came in on a Monday and they don't always do that because the core work days for us are Tuesday, Wednesday, Thursday. We had all these people in the room. We had our treasury team, the ops people, the various asset class heads and the total fund people. And we had the shock that was designed by the people overseeing some of our liquidity management. We made it so that certain key people were away. It worked really well. A lot of lessons from it.

52:33Those are the sorts of things you need to do fairly regularly. Exercise those muscles. I'd love to hear an example of an opportunity that you've pursued with the team at CalPERS that feels like just the right fit for this particular pool of capital. There are quite a few things that might be just the right fit in the sense that they are something that aligns with those advantages that we have. things like our time horizon, size, and so on. There are times when we can negotiate economics, which we think are favorable to both sides. We have done some of that in private equity. The thinking with the total portfolio approach is more around, is this a good investment?

53:20If we think it is, it's less important where it sits. If you're sitting there with a strategic asset allocation and you've got all these asset classes, it could be that you've got a hybrid. Where does it sit? Or there could be something that's new and you don't have a bucket for it and it's hard to do. With the total portfolio approach, you look at what are the return and risk characteristics of this investment and does it make sense to the portfolio? If it does, you're more likely to do it. Now, it could be in an opportunistic bucket in an SAA, but it fits more neatly in the total portfolio approach because you can think about cost of capital on a consistent basis across asset classes.

53:59those sorts of things that might have formed between the cracks are things that will be more relevant for us now given the way we're moving what do you hope the portfolio looks and feels like five or ten years from now five or ten years from now i hope we're more fully funded than we are at the moment a lot depends on what happens in the market i would like to think that we have improved information systems, those regular scenario tests, and that we're comfortable with the range of outcomes. I would like to think that we have a greater range of diversifiers within the portfolio. I would also like to have a more systematic dynamic allocation process in place.

54:45I would like to think we've got more of those things that would fit between the buckets in an SAA type environment because I think that's an advantage now. So after the end of whatever stretch of period of time, five or 10 years or longer that you're in this seat, what would feel like a success to you? It would be a success if we were more fully funded. It would be a success if people thought that Coopers was using its potential better in terms of the asset and the mandate, the talent we have. It would also feel successful if people thought this total portfolio approach has worked. Oftentimes, when you think of a public employee seat in the US, it's very prestigious.

55:31It's a big seat, may not pay that much. There's some intrinsic pull that someone may have, maybe their parents in your case, so as a teacher or whatever it was. I'm curious what that is for you in a seat in California, not your homeland. I like the purpose. It's quite meaningful. We're investing for 2.4 million members. There's something about these asset owner roles that I enjoy because you get to work in the markets, which is intellectually interesting. You get to think about the economics and the politics. You get a seat at the table. You get involved in interesting conversations. It's a privilege to be able to do this, particularly if you're doing it for a good cause.

56:14Intellectually, it's great to keep learning. When you're dealing in the market, you have to keep learning. It's also good to solve problems. It's that combination of things that appeals to me. Zim, before I let you go, I want to make sure I get a chance to ask you a couple of closing questions. Before we get to the closing questions, I want to tell you about one of our strategic investments. We've made a few and each are working on a product or service we think will be valuable to our community. One is Thema. For all the private equity managers out there, Thema uses AI to help map the landscape and source private businesses.

56:48It's incredible what a well designed AI tool can do to accelerate the discovery of businesses in private markets. There's a link in the show notes so you can learn more. And here are those closing questions. What was your first paid job and what did you learn from it? my first paid job was a student job. I was a laborer doing all sorts of manual work. That job brings back memories because I remember my first day on the job. A group of us started that day. We were given tools like crowbar, pick, shovel, and we were asked to dig up a road. It was a metal road. It was hard work because we only had these tools.

57:31At the end of the day, I went home. I had seven blisters on my two hands. My father looked at me and thought I needed toughening up. He got some denaturalized alcohol and just poured it on my hands because that toughens up the skin. It stung. From that time, I wasn't sure whether they were testing us as new joiners, but it got easier after that. I enjoyed doing that manual work. Which two people have had the biggest impact on your professional life? people who know me know that I don't stick to the script. So I'm not going to mention just two people. The people that have had the biggest impact have been folks who've taken a chance on me.

58:12I did a master's degree in economics. Normally, people in New Zealand who did that would go off and work in the treasury or the Reserve Bank. Did the Reserve Bank later, I was offered a job at the treasury if it didn't go there. I was offered a job at one of the universities in New Zealand to lecture on finance. I hadn't studied much finance, but they wanted an economist. I thought, oh, that's good. I'll do that. So that was Lyle McLean, who was the Dean of Accounting and Finance at Otago. He took a risk on me. I enjoyed that. And I got to spend some time with Simon Beninga. He wrote one of the standard texts on financial engineering.

58:53He was teaching us option pricing and that got me thinking more about finance. I went from there to the Reserve Bank. Then when I went traveling, I picked up a job at Chase Manhattan. That was a bit of a risk. I'd come from a central bank. I got parachuted into derivative structuring. It was Sykes Welford, who was one of the senior people at Chase who took that risk with me. That was great. IMF, same thing. Moving from FX options to IMF, I had that central bank background, so it wasn't such a leap. Going back to the market, same thing. People at Future Fund, Dave Neal, took that risk hiring me.

59:31It's those sorts of people who've been prepared to take someone who's come from maybe a non-linear role. What brings you the greatest joy? I really like walking in the mountains, in the bush, being with nature. From a professional perspective. I like seeing people grow. It's great seeing people you're working with or people who work for you do well. They make you look good. I really get a kick out of that. I like solving puzzles and of course spending time with friends and family. How's your life turned out differently from how you expected it to? It's turned out very differently. If you'd asked me when I was a teenager what I was going to do, I thought I would work in the wildlife service.

1:00:15I wanted to spend time outdoors. I didn't think I would be living in other countries, so quite differently. Last one. If the next five years are a chapter in your life, what's that chapter about? It's probably about what you've already asked. How have things gone at Kelpa's? I would like it to be a story of success. I would like it to be a story of fulfilled potential. I would also like it to be a story of using my leisure time better and doing the things I enjoy. Stephen, thanks so much for joining me. It's so great to hear about your latest seat. Thanks very much, Ed.

1:01:14and should not be relied upon as a basis for investment decisions. Clients of capital allocators or podcast guests may maintain positions in securities discussed on this podcast.