FOCUS 2026: Systematic investing comes to fixed income

Glen Davidson: Welcome to Scottsdale. This is the first of this large format that you've done. I had the pleasure of speaking to you last year in Boston for a smaller group. It was a very, very interesting story talking about quantitative research and investing. That's why you're back. There were a lot of questions then as well so if you have any questions please do send them in. Now, quantitative research in investing, I'm glad I was able to talk to Andrew about quant research. That's a little team in Toronto but we should give everybody a background on how QRI came together in the United States.

Karishma Kaul: Definitely. QRI, Quantitative Research and Investments, it's a very unique group. It was created in 2020 with Fidelity making a very long term, and a big commitment to quant, with the context of what would really keep us relevant in the next 30 years. With that QRI came about. Neil Constable is the head of it, he established. My entire career has been in the quant space and the systematic fixed income space. I will say there is nothing like this on the street today. It's a group of 200 quants supported by about 250+ technologists and it covers every asset class within Fidelity. It's kind of sitting horizontal across various groups covering every asset class, different investment styles, fundamental, systematic, technical, various formats, long-only, long/shot so it covers the whole spectrum, and has a dual mandate of supporting fundamental businesses along with creating and building systematic or quantitative strategies. That's where my group comes in.

If you zoom into that our systematic fixed income group has a mix of portfolio managers and researchers. We're all coming from some of the top shops in there with a lot of expertise and experience in this space. PMs have an average of 17 years of experience managing systematic assets along with a lot of deep educational PhDs and that kind of background. We cover all asset classes within fixed income, the whole spectrum, long-only, long/short, as I mentioned so in terms of capability and scale it's a very, very powerful, well-resourced group.

Glen Davidson: My understanding was years ago, because you talked about that growth starting in 2020 to 200 people that are in QRI now, quantitative research existed for a long time and it was really a trilogy of quantitative, technical and fundamental. Portfolio managers could draw upon all of those. It was in 2020 that quant took off and it really has to do with moving beyond human bias, I would imagine. What was the catalyst?

Karishma Kaul: Quantitative investing in equities has been around for a very, very long time. I'd say quantitative fixed income is relatively new and that's because the barriers to entry have been generally harder and I think to get success there there are a couple of things that you need, and I feel like we're kind of equipped with all of them. Data is always a big issue in fixed income, getting access to clean data. Fidelity has an edge here with decades' worth of fixed income data within the platform. Trading is always painful because it's not like an exchange-traded asset class so transaction cost. Having best-in-class execution I feel is very critical in success of these systematic strategies. Again, Fidelity has an edge with the trading groups that we have that have been doing this for years and have relationships in the street going back in time.

I'll say implementation is a key point in terms of having the domain expertise. My team has a lot of expertise in this space but we also sit within the broader investment ecosystem which is Fidelity so we do draw on that to navigate various scenarios. Fixed income in itself has changed as well. When you look at how bonds are traded, the rise in electronification, portfolio trading, a lot of the elements are a lot more automated now. While the asset class is generally harder to get it right there are a lot of things moving in its favour so I'd say since 2020, post-COVID, trading improvements, it's all worked in our favour so here we are.

Glen Davidson: What a competitive advantage. QRI, quantitative research, is a really additive angle for fundamental portfolio managers but the I part of investing has to do with running portfolios. As you said there's equities but there's fixed income and that's what you're responsible for. There's the Fidelity Core US Bond ETF and there's the Fidelity Absolute Income. We'll get into those in a minute. First of all, you were talking about all these PhDs, in your case you have a BA in electronic engineering and you have a Master's in financial engineering. As a guy with a BA in economics it's pretty cool that you have all that stuff. What brought you into the financial world and, in particular, I know you've been to a number of different companies but a great career so far at Fidelity. Did you expect that you'd be in this sort of world as a portfolio manager?

Karishma Kaul: I did my BA in engineering so I do have an engineering background. I always liked quant math. In terms of gratification I feel like when you see actual impact, when you're solving real problems, that gets you excited. That was the motivation towards financial engineering. I thought you're kind of trying to model something that you can't always perfectly model but if you get it right you see a very big impact. That's the background. I've been in fixed income all my career, I've in quant systematic all my carrier. Fidelity as an opportunity was very, very exciting for me, just kind of getting inspired by the vision of what we were trying to build and what we've kind of built in QRI. As I mentioned early on it's pretty unique. As an org, it's almost the size of a small quant firm or a big quant firm.

In terms of what do quants need to succeed, data always comes first. I think about what Fidelity has built in here. They've got decades of data sets which are all proprietary. They have a lot of alternative data sources. They have big budgets to actually buy and ingest alternative data sets which the competitors don't. From a quant perspective t's like all shiny tools and you're like, oh, can I play with this? So 40+ data sets, we aim to test 50 data sets every year going forward which is pretty, pretty exciting from a technology point of view. This is top tier talent. We have a data science team, specialists in machine learning, knowledge graphs, there's a lot in there that you can play with. If you have these building blocks, data, technology, expertise, and you're working with a group of people that have great talent but they are collaborative and it's a very collegiate environment, it just feels like, yes, you can do something really great here. That's what we've been able to do over the last few years.

There's a lot of synergies that I see working across groups. My group is systematic fixed income but we work very closely with the equity guys because guess what, from an investment point of view the economic priors do overlap. There are synergies when it comes to the actual modelling of insights, combining of insights, trying to build prediction models, or even macro overlays. Coming to an environment which is no silos, open, collaborative with that kind of talent and a foundation of data and technology that I know is at the forefront of innovation, it was just very exciting to get in and build all of this in here.

Glen Davidson: You get a really big smile when you talk about the access to data. I think what all of us want to know is what kind of data do we provide as consumers through our phones? I just ask you this to talk as a QRI representative on stage today and then we'll get into fixed income. It's been fascinating to me to hear from your group what is derived through ...  that Fidelity will buy ... data from our usage of cell phones, credit cards, I don't know what else, which burgers we buy at a restaurant. There's an amazing amount of data that's aggregated and you have access to that, which makes you smile. Can you talk about that and how we're affecting that?

Karishma Kaul: Yeah, for sure. You touched upon a couple of data sets so credit cards, receipts. I have a great example of this one study that we did a few years ago. There's a lot of consumer data and addresses. We collated all of that and we used that data to create a migration data set. That essentially means understanding where people are moving to. That's such a unique data set, how can you use that insight? You can use it in something like munis. That's an investment vehicle that will help you understand where folks are moving in and out from...

Glen Davidson: Migration of people.

Karishma Kaul: Migration of people and what's driving that and the state fundamentals behind it. Something like that is a very unique example where I feel like you have an edge with a data set like that. Similarly, there are a couple of other examples that are so powerful. Planet Fitness, there were a lot of headlines around it. The portfolio managers were wondering is that real, are these headlines real, and is it really impacting their bottom line? Guess what, we have credit card data and we can slice and dice that by different states, by different demographics. It turned out when you looked at the receipts it was fine. When you translate that into an investment insight you have now something to act upon and that works from an investment or alpha perspective which is all of what we are working towards.

Glen Davidson: It's really exploiting information that's out there, it's just that not every firm has the ability to do that. The private ownership at Fidelity must have been a draw for you as well because that really means the sky's the limit as far as resources for all sorts of businesses but QRI in particular.

Karishma Kaul: Yeah, for sure. Even for my team, our core team is working on models, working on insights. We don't really have to worry about cleaning the data, validating the data, modelling, mappings and understanding how all that is. Once you take that away and you've created a central data platform now we can do thoughtful research and we can do it a lot quicker. We can do it a lot more robust. That's definitely been a big draw. We've had some great success here.

Glen Davidson: What would you like to have at QRI that you haven't seen yet? Where's the future?

Karishma Kaul: What's the future? I think we have a lot. We have a lot of exciting things going on. It's hard, if you would have asked me this question in Jan. I would have had a different answer. The last two months I've just been blown away. We spoke a lot about AI these two days. There's a lot of theme, AI trade, we're using it. In terms of productivity my team is two to three times more productive. I just cannot keep up with research because it's kind of unlocked a lot of tools for us validating research a lot better. I know Andrew touched upon domain expertise and asking the right questions, that's 100% true. It's going to make a lot of mistakes so you need the domain expertise to use it to your advantage, whether it's ingesting new data sets, whether it's going back to the models we built and validating them, working with different teams and just doing it a lot faster.

I do think AI integration, again, Fidelity is at the forefront of adopting a lot of these technologies and giving us access to that but I do in the next few years AI integration is going to be a big theme. Alternative data sources, especially with everyone looking at the same data and AI kind of helping sift through that a lot quicker,  a lot of the traditional data sources are gonna be priced in sooner, faster. Being on top of that alpha decay from traditional sources is gonna be there. Finally, quant fundamental, it's not quant versus fundamental, it's alpha that we're all looking for so more merging of that, I do see all of that coming.

Glen Davidson: You've done a great job at establishing what QRI is all about, how you ended up there, how everybody else ends up there, the data points that you have, but it also sounds like it's going to grow exponentially because of the use of AI to make things more efficient for you as well. This has been a great tool to fundamental portfolio managers. You work with them but it also has created QRI derived solutions. Let's talk about systematic fixed income investing and why that makes sense.

Karishma Kaul: By systematic fixed income investing, or quant investing, what we essentially mean is a data-driven research process, a model-driven investment process to create repeatable, consistent outcomes. Any insight that you have, any model that you've have, you backtest it. We usually try to make sure that there is an economic prior there but then we create backtests, and we test them, understand why a model works, when it works, why it works. Once we have that model that gets to become a strategy our portfolio managers would manage to that model. Now, since these strategies target different sources of return and have a different portfolio construction methodology than traditional funds they behave differently. They generally have lower correlations to traditional strategies, which is a good thing.

Glen Davidson: Which makes them complementary as well.

Karishma Kaul: Yes, that makes them complementary.

Glen Davidson: [indecipherable] it better.

Karishma Kaul: They're complementary. We've expanded the menu here because this lets you not only diversify across asset managers but then also diversify across investment styles.

Glen Davidson: Does it allow for greater diversity because of the efficiency of the way it's constructed?

Karishma Kaul: Yes.

Glen Davidson: Greater than a typical fundamental portfolio.

Karishma Kaul: Yeah. The way to think about it, what we're trying to do is make small bets. Since you have a model covering all securities instead of making big concentrated bets with very high conviction you're making smaller, broader bets. This allows you to expand your universe. This allows your portfolios to have consistent returns and you have good sense of when does an insight work. What is the insight? The insight could be a value-based signal, a quality-based signal, but you have a good sense of when it would work, when it wouldn't work. If there is like an idiosyncratic name that blows up or you got that one bet wrong, that's okay because it wasn't a big bet. Overall, that factor or insight should work, and it generally does work, so that does provide diversity. It also provides diversity ... we touched upon your positioning is different so since your positioning is different it expands your liquidity profile from a client perspective.

Glen Davidson: Is the trading frequency increased because of many small bets versus larger bets?

Karishma Kaul: Yes, they are generally higher turnover strategies.

Glen Davidson: What do you do to minimize costs?

Karishma Kaul: If you get it right, that's when you get systematic fixed income right and that's always been a hard thing for folks. We've done this forever. We've done it through our entire career. We work very closely with trading on execution, working with the Fidelity trading desks to minimize trading costs, to make sure we use tactical opportunities to get closer to the model with zero-T cost. We also have our own proprietary tradability model. We have a transaction cost model that we use when we build these backtests so that we have confidence in the stuff that we run. For example, we manage a bunch of strategies right now across the entire spectrum but every backtest that we've run we are conservative than actual realized T-cost. We are conservative when it comes to returns, we are conservative when it comes to transaction costs. That's exactly what you want. A lot of the overlays that we work with the trading desks on are additive. We call it execution or implementation alpha. You've got the model alpha and then we have the implementation alpha that comes from reducing transaction costs and in some cases also using short term trading signals to make the best use of the opportunities we have.

Glen Davidson: If I started a quant shop today, I'm not going to do that, and I started a systematic fixed income solution, that's just one little piece of the pie. What you've done, and the division has done, is figured out all the periphery as well because that was really interesting and thorough about reducing trading costs. Could you expand on how you use trading data as you just alluded to as a signal as well?

Karishma Kaul: That's one of the new signals that we're using. A good thing about being at Fidelity, so much data, so much trading data going back years so you can test this stuff and run it in with confidence. Some of the insights that we've looked at is just flows, flows into particular issuers, flows into bonds. Using those insights we have some flow signals that are very fast. We've tried to use them as either short term filters in slower funds or actual trading signals. The other insight that we've used is a toxicity signal which is essentially understanding how broker dealers are being on a, you know, what generally happens in the spaces you have an insight that comes through or people have a sentiment. They'd be posting these bid asks but there's a direction towards it.

When you look at it from a time series perspective you can kind of glean out, oh, there is this issue or a particular bond that the street doesn't really like and everyone is going on the other side. Guess what, even though the price looks really good the sentiment is really bad. Having access to that kind of data, being able to glean those insights and then actually turn them into investment positions or investment strategy that gives us a big edge. The trading signals we are pretty excited about in addition to a lot of the non-traditional signals that we're using from the equity guys.

Glen Davidson: I get the sense that every day there's some eureka moment where someone almost wants to ring a bell that they found another area that they can dig into to get a sense of direction. It's a fascinating group and there must be discovery on a regular basis. We should get into the solutions that are available now. We'll first talk about the Core US Bond Strategy, if you could let everybody know what that's all about.

Karishma Kaul: That Core US Bond Strategy is a core building block from a fixed income investor perspective. It's benchmarked to the Ag. Its goal is to outperform the Ag over a business cycle. We launched it last January so it's a little over a year. I think it's over 100 bips of alpha. The way we've approached it is the same systematic. Everything is model driven. We have three aspects of alpha models or alpha levers that we use. We start off with the sector allocation model which uses systematic signals like value, quality, momentum, and allocates across granular sectors, across rates, credit. Then we have a credit security selection model which is used across almost all of our strategies which uses multiple insights across alternatives, fundamental, relative value, momentum. We combine all of that to come up with an alpha that works in this fund in the IG space.

Finally, we have a macro risk timing model that essentially uses the high yield market. It uses a few other market components to come up with regimes and does allocation on a DTS or a duration level to add an overlay on top of that. This is a core building block, consistent, same investment philosophy as we've discussed and has performed really well over the last year.

Glen Davidson: It incubated for a while, I would imagine, before being launched so that you could feel it's perfected.

Karishma Kaul: Yes. We actually worked internally, within FMR we have another version of this strategy which was launched in 2023. We had a year or two years of like pilot, not pilot period but it was for essay. I think that fund is 1.2, 1.5 billion right now. It's really taken off. It's the same strategy just dialled up the risk a little bit.

Glen Davidson: The FMR Karishma was referring to is the US business of fidelity available down there. I said perfected but it's never perfected because you're constantly evaluating and making it better.

Karishma Kaul: Yes, yes, actually that's a very, very important point. Yes, it's rules-based, yes, there is a model but it's not like we build the model and wash our hands and we are done and just sit. We do a lot of rigorous research, model management meetings, we monitor what the model is doing, we understand what the market is doing and we make incremental improvements as we need to. All those improvements go through the same research approval process. There's a lot of thoroughness and robustness there but it's evolving, there's feedback loop so we try to understand whether it's improving from an insight or a signal perspective or it is just liquidity, tradability, those aspects, adding those insights and improving the model. Your job is never done.

Glen Davidson: Now let's talk about the Absolute Income Fund.

Karishma Kaul: The Absolute Income Fund, it kind of goes back to what is systematic good at. Systematic is good at repeatable, consistent performance and the flexible nature of it allows you to target a lot of objectives in addition to total return. Absolute Income is an income-oriented fund. It aims to generate at least 90% of high yield income with much lower vol and much lower drawdowns. It does that in a variety of ways, a similar three-level approach, I'd say. Very first  it has an income-focused sector allocation model that allocates and diversifies this risk across high yield, IG, EM, mortgages, Treasuries. It also has a security selection layer but the security selection layer in addition to all the systematic insights all incorporates income. That's how we incorporate income. The marginal income is incorporated in both security selection and sector allocation.

Then we have, similar to the Core Fund, we have a macro risk overlay on top that navigates different regimes. It's done really well since it's been launched as well. I think the last year it's outperformed the Ag by 5%. It's outperformed high yield by 40, 50 bips and has much lower vol than high yield, lower vol than drawdowns. It's high yield-like income but a more diversified fixed income fund that has lower vol. Given fixed income sectors can move around a bit it essentially makes use of that opportunity and allocates across a diversified set of sectors.

Glen Davidson: That's a lot of information. Someone that's saying, I have a sleeve of a portfolio that could really benefit from this removing the human bias that we talked about and using this strategy, if we look at Core US Bond and then we look at Absolute Income where do each of them fit within a portfolio?

Karishma Kaul: Core US Bond would be a core building block. If you're looking for IG exposure, if you're looking for Ag exposure this will give you that and add alpha on top. It should look very similar to the benchmark in terms of vol, in terms of constraints. There wouldn't be a lot of out of benchmark exposure. This is a core risk exposure that you're trying to...

Glen Davidson: It's core but it also has diversity that your fundamental portfolio probably wouldn't have. There's where the complementary nature...

Karishma Kaul: Actually, that's a good point. It's a core building block and then when you look at the correlation of returns it should look and feel different, both in terms of positioning and in terms of returns. On the Absolute Income strategy, that is an income strategy if you're looking for high yield-like income with lower vol, using again, a very diversified set of instruments. The other thing I'll talk about Absolute Income, generally when you reach for yield, reach for income, you go into the plus sectors and you're in the tail of credit risk. This is not a yield-seeking product. It does everything in a very risk controlled way. There is another added layer of up in quality. We have a lot of quality signals that work and that are very, I'd say, when it comes from a credit perspective there's a lot of value that you get from having an up in quality and understanding what your tail is. We have quality insights that go into this particular fund that help it with drawdowns and it, for the most part, avoids the tail of credit risk.

Glen Davidson: What would you like to see added from a solution standpoint, or what are you incubating that you can allude to as far as a complementary solution to these? A wishlist from Karishma, hypothetically.

Karishma Kaul: The very next thing that we're very excited about and we've been thinking about is a long/shot credit strategy. Similar what Andrew mentioned early on, once you are long-only there are only so many opportunities that you can get on the short side. Opening the opportunity set where you can take both long and short bets at different position sizes, that just increases your alpha potential.

Glen Davidson: You're covering everything, almost, both sides of the market, really.

Karishma Kaul: Yes, you're covering both sides of the market and you're making the best use of your alphas. Something like that, using our signals is next top of my mind. That should look pretty compelling. One of the future things that we're working on is a combined bond equity strategy where you could have similar insights that are working both on the equity side and the bond side and having them combined in one portfolio that could look pretty interesting as well.

Glen Davidson: That seems to be the future, to look at both sides of the market, look at long/short. We'll stay tuned and see if that's something that we have available for clients. What does the street look like? Is this untouchable as far as competition? Are you untouchable, Karishma?

Karishma Kaul: That's a tough question. I'm generally a humble person but I will say what we've built here, the resources that we have as a competitive platform, we're kind of quite a bit ahead from our peers. I say that because a lot of the team is coming from some of the top firms. I, personally, have a lot of experience in this space. We really come together and build on the data, the technology, the insights, the expertise that we bring to build something truly world class. I'm pretty optimistic. I think we're way ahead in terms of innovation, in terms of execution, in terms of just the products and the offerings and capabilities. I do think we are prepared.

Glen Davidson: I'm wondering if your competitive nature made you break your foot but we can get into that another time. Can you talk to the audience about where they should really be looking at quantitative research and investing within their portfolios?

Karishma Kaul: As I said, in terms of where would you put quant, if you're looking for consistent, repeatable, disciplined outcomes that's where quantitative comes in. If you are looking to diversify your outcomes in the sense that I have a core strategy in a particular product but I want to diversify the returns, if you look at the returns of quantitative strategies, generally flat to negative correlations. That's where they would fit in. If you're looking for certain outcomes, whether it's income, whether it's like, oh, I want a defensive profile, I want a custom outcome, I want certain filters or screens, all of that fits into quant very, very seamlessly because it's so flexible. Not only can you add it easily you can also quantify what is your alpha loss. If someone says I don't want to buy energy, well, let me tell you over the cycle you're going to lose this much from an alpha perspective. Stuff like that. We can integrate customization pretty easily.

Glen Davidson: Fascinating, and fascinating to talk to you again. It's going to be very, very interesting to see what the daily change is for QRI because it sounds like it's really on an amazing curve. We'll talk again and explore what that's all about. Thank you very much, Karishma Kaul.

Karishma Kaul: Thank you guys.