Extra Credit: Algorithmic Advantage - Systematic Investing in Credit

The world of credit is evolving. Gone are the days when investment grade bonds were only sold over the phone and the world of direct lending was a little-known niche. Corporates now have a multitude of ways to access capital and investors can diversify their credit allocation more than ever before. Extra Credit, the fourth season of our Long Story Short podcast, will explore the ways credit is changing and how it fits into your portfolio. Hosted by Danilo Rippa, Head of Global Credit Multi-Strategy and Global Convertibles at Man Solutions, this season will feature experts from across the credit investing spectrum.

In this episode of Extra Credit, Robert Lam and Paul Kamenski, Co-heads of Credit at Man Numeric, join Danilo Rippa, Head of Global Credit Multi-Strategy and Global Convertibles at Man Solutions, to discuss adopting a quantitative approach to investing in credit markets. In their wide-ranging conversation, Robert, Paul and Danilo discuss the rise of quant credit, the alpha opportunities it presents, as well as what the future of trading credit systematically might look like.

Recording date: March 2024

Episode Transcript

Note: This transcription was generated using a combination of speech recognition software and human transcribers and may contain errors. As a part of this process, this transcript has also been edited for clarity.

Danilo

Welcome to Extra Credit, the latest season of our investment podcast. Long story short, I'm the Danilo Rippa, head of global Credit Multi-strategy and global convertibles at Money Solutions. And joining me today are Robert Lam and Paul Kamensk, Co-heads of Credit and Man Numeric. Rob Paul, thanks for joining me today. I want to start with a very simple question, what is quant credit?

Danilo

Can you please explain the mechanics of quant credit investing?

Robert

Yeah. So that's a great question, Danilo. There are many pieces of quant credit investing and in many ways, if people are familiar with the quant process, it actually has very strong parallels to how we think about systematic investing in other asset classes. At its core, it's about exploiting new and innovative data sets in order to capture key drivers of quantitative or corporate credit spreads.

Robert

It's about trying to find data sets that allow us to capture those key drivers. And then finally, it's about trying to find robust quantitative techniques in order to isolate that intended inefficiency that we're looking to capture. And then, of course, the hard part, in my opinion, is going from the expected returns that we can research in a theoretical state, then bringing in the other practical pieces of the investment process, which include risk modeling and portfolio construction and liquidity and transaction cost analysis.

Robert

And ultimately, finally, it's about creating the strategy that tells you what Q6 we're looking to buy and sell that lead to our final strategy.

Danilo

Thank you. That was very clear. We all know that credit trading at being conducted primarily through telephone calls or even electronic messages, since I dealers and after a slow start, actually after a very slow start for bond credit, particularly compared to systematic equity, we seem to be making good progress. I have two questions for you. Number one, why did it take so long relative to equities to get to this point?

Danilo

And number two, what has contributed to this recent shift?

Paul

Yeah, so it's a great question. It's near and dear to my heart. It's honestly the piece that I think has made systematic credit, as hard as it is, and as Rob was just alluding to before, particularly the portfolio construction side. I think we're in a really interesting time where the methods of execution that existed for decades still exist within credit.

Paul

Voice trades are still a big fraction of the market. But from what we've seen, there's barely been a big tipping point. And over the course of a few years, we've seen a decades plus worth of evolution that we saw in other asset classes like equities. And in many ways, the stories of very much gone in parallel, very much a similar story.

Paul

And think credit, obviously it's a hard asset class. You have a lot of securities from single issuers, many different risk characteristics, whether it's subordination, whether it's different sort of locations on the curve, whether you're shorter or longer duration and also just the market as a whole has generally have been quite a bit less liquid than equity markets. And I think that lack of liquidity actually makes it very difficult and has made it difficult for years for market participants to participants to be able to come in and provide markets electronically in a very efficient way.

Paul

Thankfully, I think we've cracked the not a lot of sell side dealers. Back in 2017, 2018, very much grew their algorithmic market making capabilities as we saw a lot of growth there. But it really didn't hit that tipping point until we saw some evolution in 2021. For us, that was the year of portfolio trading, where all of a sudden not even just primary dealers, even secondary third tier dealers started to be able to make good markets for broad sections of investment grade and high yield corporate credit.

Paul

And so we were able to see improved sell rates, improved certainty of execution with really attractive transaction costs. However, fast forward into 2022 that if you remember, lots of rates, volatility was spiking up. There were some changes in that ETF create redeem process that made it a bit harder for for those desks to recycle risk. And what we saw was transaction costs generally went up substantially, not just for us, but across the market as a whole and single in electronic execution held up best.

Paul

So I think we've hit that.

Danilo

Let me stop you there. Sorry. I believe that as recently as 2022, you just mentioned it, we saw 15 billion of AIG and paper trading over a 12 month period. How are you seeing the liquidity still the biggest hurdle to credit taking bigger market share?

Paul

I think at this point, the tipping point has happened such that we're in the sweet spot for quantitative approaches. To be clear, it's still very difficult. I think you still have the problem where there's venue selection. There is deciding whether you want an order to be high touch or low touch. We're in that transition period where there are absolutely still circumstances where the best course of action is to trade voice.

Paul

But more and more, we are seeing that slicing up orders, being able to enter into positions and exit positions in a more fluid way on a daily basis as quant strategies always try to slice into that closer efficient frontier portfolio. We're really hitting into that sweet spot in my point of view.

Robert

Yeah, I mean, Danielle, you bring up a great point. 15 billion. I actually think that's a very big number. And year by year, we continue to see it grow. We continue to see electronic venues to continue to take market share. And I think the trend is clear in my opinion, that electronic trading is here to stay at this point.

Robert

When I look at that kind of venue, all the venues within the electronic trading space and all the different protocols, it's really become time tested. And not only that, it's become volatility tested. We've gone through periods of significant volatility over the last three years and they've stood up with very high liquidity numbers. So taking a step back, I really do think that there's two barriers in my mind of how to reach the next level of liquidity within the corporate bond space.

Robert

On one side, I view by side participants as having to navigate this transition where the traditional block trades that we see within the market, where you're looking to sell 10 to 50 million, you have to work with a principal or agency market maker that actually might not be the optimal way to execute corporate bond transactions. The alternative is being able to to slice it and to be able to optimize your pre trade routing to the best venue and the best protocol.

Robert

So that's a big change for the buy side participants. And then on the other side of things, you have the liquidity providers. Now, liquidity providers have historically clamped down on how much liquidity they're willing to provide during periods of volatility. We saw that by voice traders. We saw that during certain periods of time for electronic traders. But I actually think that what's become clear over the last few years is that the liquidity providers that have been able to stay in and be sophisticated around how they're providing liquidity are really reaping the benefits.

Robert

It's only going to become more competitive in that space.

Danilo

Understood. Understood. Rob Paul, you made your point. This area is growing, but let's make the conversation a little bit less friendly now. I'm a discretionary investor. Why would someone listening to this episode choose to allocate to credit over traditional discretionary credit strategies? Where is the alpha?

Paul

Yeah, I think no disrespect. I actually very much appreciate discretionary approaches. I think one thing that sticks out to me at least, is that no one in the fixed income markets is using enough data. I think no one's able to respond to things quickly enough. And I think those are real sweet spots for quantitative approaches, being able to marry the breadth and depth of data science with a fundamentally driven investment process, I think really does put you in a position that you can take advantage of human bias in what we view as an otherwise untapped market.

Paul

There's a lot of opportunity there. There are concepts where from a network point of view, you have a supplier company reporting before a customer company and there's obviously some fundamental read through that we can all think of economically, fundamentally, but multiply that by hundreds of issuers as reporting and in the heat of the earnings cycle and the complexity of all of those relationships, being able to tap into a lot of that information flow in a timely manner, not just for alpha considerations, but even from risk considerations, is, I think, key.

Paul

And that frankly, why allocate diversification is one of the only free launches out there. And I think from a process point of view, it's different and we're able to tap into those insights and leverage data, as we've been discussing before, even from an execution of portfolio construction point of view, I think we've really hit that tipping point.

Danilo

I agree about can we get a little bit deeper? I understand systematic approach to trading, but how about the asset allocation? Can it be improved?

Robert

Yeah, absolutely. I think that there is opportunity to exploit the idiosyncratic drivers of returns, as we've been talking about today, where we're using single name data around balance sheet income statement, cash flow statement, alternative data to be able to now cash KPIs, consumer behavior, competitive dynamics, those are all at the single bond level or the single company level.

Robert

And I think that there's a real opportunity to take that a level higher and understand how group behavior works, how group concepts, how you forecast sectors or ratings, or how you forecast between corporate credit, IG and high yield, or even taking that even a step further and thinking about asset allocation from high yield to sovereign to quasi sovereigns to securitized products.

Robert

I think all of those are real opportunities for us to apply the same things that have worked for us and the idiosyncratic part and be able to apply that same time tested technology at the asset allocation level.

Danilo

So based on these, how do you expect credit to look in five years time?

Robert

Danielle I think we're just getting started here. I think the quant credit space looks drastically different today than it did five years ago. Alphas have gotten more sophisticated. Asset allocation, as you've highlighted, has become a core component in the mix. And when I think about what quant credit will look like in the next five years, my base case is that the alphas that we're going to be using will look drastically different.

Robert

In fact, I think they'll look drastically different in the next three years. I think the revolutions and the evolutions that we're going to continue to experience within quant credit is going to accelerate. And a big part of that is because the market structure is changing so quickly. As we've discussed, and then we're able to leverage time tested technologies that have been developed for decades in other asset classes.

Paul

Yeah, and I think I would just add to that that we've heard conversations from Allocators in the past with with perfectly valid good questions. Should I have a bucket of allocation to quant credit, or is this still something that's to be proven? I think what we know on the equity side is that there absolutely are distinct application buckets.

Paul

There's a lot of diversification benefit from a return point of view, from a process point of view. And we're still in a very low single digits of market share when it comes to fixed income assets. Broadly, quantitative approaches still are very small and I think what should absolutely be the case five years from now is we absolutely should be much closer to that equilibrium point where it's not that quants take over, it's not discretionary, maintains its near 100% market share for forever.

Paul

I think there really can be a good equilibrium between these two where allocators see the benefit of both approaches and try to take advantage of that. And if anything, we've seen so much evolution and being able to implement strategies and strategies like this over the last several years in that that process this happen quickly, I actually suspect the allocation process in that equilibrium point could happen quite quickly, too.

Danilo

Let me ask you a more pragmatic question here. Is this strategy a way of investing in credit suited for particular part of the cycle?

Robert

Yeah, I think I think that the way that we think about Alpha within the corporate credit space is to remove a lot of the unintended risk exposures. And when we think about unintended risk exposures, it's things like loadings or unintended loadings on things like ETFs, which can induce market betas, things like durations as well, and those really large tilts that tend to flow through in, you know, less sophisticated portfolio construction techniques can actually create or induce betas that don't make it as consistent throughout various market cycles.

Robert

So in our opinion, the way that we try to create this kind of all weather approach is to really target that idiosyncratic alpha, which can really tip the scales when you think about it as an asset allocation framework. And in terms of, let's say, investment grade yielding, you call it six or 7% over long market cycles, you know, you can really tilt the scales in your favor when you have that idiosyncratic alpha of that 1 to 2% above the L, above the benchmark.

Robert

And I think that that's important characteristics when you think about, you know, what is the right asset allocation mix over over a full market cycle.

Danilo

We briefly touched upon this earlier on, but I'd like to get a little bit deeper on your conversations with investors. Basically, I'd like to understand how people should think about it from a portfolio structure point of view. Is this just a diversifying tool? Is this the risk management? Is this reduction of cost? What aspects are people taking into consideration when they allocate to this strategy?

Paul

Yeah, you've touched on several key points, I think, and it will depend on the investor in their unique circumstance. But I think really a belief that leveraging more data, more data science, trying to codify a lot of the fundamental concepts that we're well familiar with from a relative valuation point of view all the way through to cross asset information, through to defensiveness, characteristics and even new concepts.

Paul

When you think about trends within marketplaces, that resonates with people where they see this has worked in other asset classes, we think it should work within corporate credit as well, and I think we're seeing the fruits of that now. So it's definitely an opportunity, definitely a diversifier across other discretionary approaches. And frankly, with annuities from what we see, even diversifying within other systematic credit approaches because it still is relatively early days.

Paul

But a nice thing too, is that when we have the tools at our fingertips from a portfolio construction point of view, from a signal point of view, we're able to tune things efficiently and be able to do quite a lot. And so we have absolutely seen customization by way of trying to plug holes for certain regions or characteristics that maybe it's hard to find a certain discretionary exposure that can sort of fill that plug.

Paul

That whole we've found it across pretty severe universe reduction, being able to focus, for example, on our eye considerations, responsible investing, ESG considerations. A lot of allocators have unique criteria when it comes to that, when it comes to even insurance companies with various regulatory implications, being able to have scores, alpha scores, expected returns, risk characteristics available for the broad universe means that we can tailor those things to fit people's needs.

Paul

And I think that's another key point where having those at your fingertips lets you customize, but also lets you do it in a scalable way. And I think obviously that has fee implications and who doesn't want to do do more with less?

Danilo

Is it fair to say that basically what you said because worked in equities, investors expected to work in fixed income as well? I know it's an oversimplification, but.

Paul

I as well, I think it's I don't think it's an oversimplification, but that was the hope. I think the reality and the slap in the face is there's a lot of things that you can take for granted in systematic equity approaches that that reality hits you when you're trying to trade the credit markets. I mean, anyone trying to source liquidity for many bonds struggle in regular markets, let alone when when there's volatility, you can have a whole capital structure where you have to ask yourself the question, can you just trade the most liquid bond?

Paul

Can you trade the top half? Can you trade all the bonds in the capital structure? There's a lot of considerations there, including transaction costs, knowing how much it will cost you to trade once you do make a decision that make this really hard, it sounds.

Danilo

Like we're going back to the first few minutes of our conversations. Liquidity is the biggest hurdle and it will disappear over time.

Paul

I think that's the bit that we've hit the tipping point that's easier. But to me that's the piece that if you just try to apply equity approaches in credit, you're in for a rude awakening.

Robert

Yeah, I mean, I think I think you both bring up really good points about the difference between equities and credit, and you're absolutely spot on in the fact that, you know, oftentimes we get questions about how is it even possible that two asset classes could develop so differently within the quantitative space? And I do think that one of the points that is central to this, this topic is the unique credit market microstructure, so somewhat related to the liquidity.

Robert

But I think that within the credit markets, one unique aspect is that runs and prices are not firm and actionable, and that's actually quite unique relative to other asset classes On top of that. Yeah, as you noted below, you know, on top of that, I'd say that because the credit market has developed and grown to be so fragmented across venues and protocols and each one are designed to handle a different nuance or a different way of interacting with the credit markets, I think that actually introduces a host of quantitative complexities that are actually quite difficult to address when you want to create a fully systematic strategy.

Robert

So that, I believe, is actually one of the largest barriers to entry to the corporate credit space that doesn't really exist in equities. Equities markets has kind of developed to a point where a transaction cost model can be some sort of closed form model that's actually relatively simple, that actually works quite effectively. And that doesn't that certainly doesn't happen within credit.

Danilo

Yeah, I would doubt that equities have got only one izing the one stock for every year and the credit has got several I think. Yes, please.

Paul

Yeah. And I just want to add because I think we've been we've been focusing on why it's been different and harder. I think it's to me a really key point, though, is that part of this hypothesis or hope or story as to trying to really grow into systematic credit that both credit equity are claims on effectively the same assets.

Paul

I think a lot of the insights you can we do see that there is some some information that is applicable to both. I think when we think about again, starting with a blank sheet of paper of what really are the drivers of credit spreads, what are the what are the recurring contributions that you get from defensiveness, characteristics or relative valuation or, or again, cross asset signals, sentiment, signals, flow signals, etc..

Paul

Many concepts that we that we have great experience with across asset classes do apply. The trick is one, you need to be very careful how you apply them to the credit markets. One gotcha is definitely that it's duration, time spread volatility characteristics vary tremendously across the corporate credit space. If you're comparing a short duration versus a long longer duration or a new issuance triple C that the volatility character characteristics on those two assets is is hugely different.

Paul

And so but being able to leverage a lot of the insights has panned out well and frankly what we've seen over the over life periods when we think about raw model efficacy or return efficacy to be able to predict returns for these securities. If anything, we're seeing things even more robust within the credit space than on the equity side where things have been really panning out quite well.

Danilo

That's a personal curiosity. Do you see it expanding from a junior yield to other credit asset classes that are not in mainstream today?

Robert

Yeah, we have our eyes set on continuing to expand beyond IG and high yield. As as as you've highlighted, I think that there is an opportunity to be able to apply this to basically all fixed income assets that have some sort of a coupon, some sort of a yield that we can that we can trade in a liquid fashion.

Robert

Now I think there are parts of our process that people find very attractive to the less liquid parts of the credit market or potentially the more stressed or distressed parts of the credit market. And I think that those are absolutely valid hypotheses. And I very confident that, you know, some of these advantages that we bring to the table around of analysis and data, data science, I think that that will actually be a new frontier into an application into the fixed income market in our world because we run fully systematic processes is one of the keys for us is that we run very high transfer coefficients.

Robert

And when I think about a transfer coefficient, it's the ability to translate your research into a life portfolio. So we've proven that we can do this over a number of years within the IG and high yield credit space. But I truly believe that we can do this within sovereign, within quasi sovereign, within securitized products. And I think that those will be areas of significant growth for our business.

Danilo

Thank you for joining me, Rob. Paul, it's been a real pleasure.

Robert

Thank you. To know. I really appreciate the conversation.

Paul

Yeah, thanks for having us.