Mika Kastenholz - Global Macro Series - June 7th, 2023

[00:00:07] Mika

At the end of the day, it’s really much about the volatility of inflation going forward and the level of inflation. So if we continue to be in this elevated inflation environment for much longer, I think the rates volatility that we see is going to persist. But at some point, if that’s the case, then indeed, the expectation of the market that we’re going to get rate cuts between 1% or 2% over the course of the next 12 months cannot play out.

[00:00:38] Intro

Imagine spending an hour with the world’s greatest traders. Imagine learning from their experiences, their successes and their failures. Imagine no more. Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world so you can take your manager due diligence or investment career to the next level.

Before we begin today’s conversation, remember to keep two things in mind. All the discussion we will have about investment performance is about the past, and past performance does not guarantee or even infer anything about future performance. Also understand that there’s a significant risk of financial loss with all investment strategies, and you need to request and understand the specific risks from the investment manager about their products before you make investment decisions.

Here’s your host, veteran hedge fund manager, Niels Kaastrup-Larsen.

[00:01:33] Niels

Welcome and welcome back to another conversation in our series of episodes that focuses on markets and investing from a global macro perspective. This is a series that I not only find incredibly interesting, as well as intellectually challenging, but also very important given where we are in the global economy and geopolitical cycle.

We want to dig deep into the minds of some of the most prominent experts to help us better understand what this new global macro-driven world may look like, and we want to explore their perspectives on a host of game-changing issues, and hopefully dig out nuances in their work through meaningful conversations. So please enjoy today’s episode hosted by Hari Krishnan.

[00:02:20] Hari

Thanks very much for the introduction, Niels. My guest today is Mika Kastenholz, cross-asset derivatives trader. Until recently, Mika was a Managing Director and Global Head of Structured Macro at Credit Suisse in Hong Kong. Notably, he’s the author of Trading Derivatives: The Theoretical Minimum, which I think is a great title. He holds a PhD in computational biophysics from the ETH in Zurich. It’s a pleasure having you on the show, Mika.

[00:02:54] Mika

Thank you very much.

[00:02:56] Hari

Yeah. Maybe we can start with just a little bit about your background. I tend to do it in reverse chronological order, but it’s totally up to you.

[00:03:05] Mika

Right. Yeah, thank you very much again for the invitation. It’s a pleasure. We have, I think, a very short window of opportunity where I’m still maybe a bit more free to speak my mind without getting preclearance from let’s say internal entities of where I work. So I’m going to start maybe still in the beginning because as you mentioned before, I don’t have a background in finance or economics.

I did a PhD in computational biophysics, which kind of means that I know probably a bit about math and computers. I actually always had a good relationship with computers, which has turned out to be quite a valuable asset during my career, and I still like to tinker around with data and do some analysis here and there. After my PhD, I wasn’t really too sure of whether I wanted to stay in the academic realm.

It so happened that at the time, so this is probably 2004 or so, a good friend of mine worked as a quant at a hedge fund and he told me that look, before you go down the academic route, why don’t you interview with a couple of banks and insurance companies to see where there might be something interesting for you to do in terms of crunching numbers. So I ended up joining Credit Suisse in, I believe, 2005 or ’04, initially in structuring, equity derivatives structuring, equity exotic structuring, and then moved on to trading, I believe, in 2007 in London.

This was right at the onset of the 2008 great financial crisis, which to be honest with you, I was in equity exotics at the time. It was clearly a formative year, I think in my experience as a trader, because you were sometimes a guest to experience the kinds of volatility that happens maybe once every 20 years back in those days. So that was a formative experience. If you experience things like gap risk and drying up liquidity firsthand instead of just reading it in a book, that’s clearly a different experience, and clearly, it also determined how I look at risk going forward.

So I started in equity derivatives, later hybrids, and also corporate derivatives. I left Credit Suisse in 2012 to co-found a fintech. At the time, it was a good opportunity to raise funds and maybe move more towards the product side of things also because the industry was probably a bit in decline at the time following GFC in 2008. We are probably going to talk about the changes that the GFC did to our industry in terms of regulatory impact and whatnot a bit later. We were doing things in the domain of wealth management, developing tools for relationship managers but also for partially algorithmic trading or CTA type of wealth management approaches.

But honestly, by 2019, I kind of missed trading. It’s a fantastic job in the sense that you have almost immediate feedback on your decisions, maybe not on a day-to-day basis depending on the type of business you run, but clearly, that’s quite unique, I think, in our jobs. So I rejoined Credit Suisse in 2020, this time in rates, initially for local rates in Hong Kong, and then later I was given the mandate to run what we call structured macro global at Credit Suisse, so that’s structured rates and hybrid effectively.

[00:06:51] Hari

Was your first tenure at Credit Suisse also in Hong Kong?

[00:06:56] Mika

No, that was in London. Yeah. Correct.

[00:06:59] Hari

So what induced you to go to Hong Kong then?

[00:07:02] Mika

Well, on the one hand, I’d never worked in Asia before, and still, I feel that way. To be in Hong Kong specifically, it’s a great place in terms of what is going to be playing out in the geopolitics between the US and China. It’s also a very exciting place to work. It’s very entrepreneurial there. Those were my key motivations to go there. But of course, what went wrong a bit is COVID, which is something that none of us really foresaw playing out the way it did. So that put a lot of challenges on that.

But otherwise, it’s been a great experience. Indeed, until very recently, which is April 2023, I was taking this global role for structured macro, and well, I will soon be joining another firm, more on the asset management investment side of things.

[00:08:00] Hari

Great. I can say that there are some at least vague parallels between the two of us because I had no formal training in economics either. For me, it was basically, I didn’t know what people were talking about when I started. So every day I’d go home, I was single back then, and at least read up what it was that they had said and try and piece it together. I found it was a very useful way to build my own framework for understanding things. So I presume there are some parallels there.

[00:08:31] Mika

Yeah, look, I think maybe if you enter our industry as a bit of quantitative background, you’re naturally drawn to maybe the more quant stuff, even if you work in trading or structuring. Until you experience a situation, maybe, where you cannot necessarily derive the future of the economic cycle by just looking at price and maybe volume data or implied volatility, you start to think that maybe there’s something else I need to catch up on.

In my case, that was clearly an understanding of how economic cycles work, which until then, I’d obviously never studied in university. So that was my way into then also thinking more in terms of global macro and how the whole interaction between, let’s say, central banks and the underlying economy to play out.

[00:09:25] Hari

Perfect. Now, when you were in Zurich originally, I presume that you like many of us were at least exposed to econophysics, which I think was started in the 1990s and then was going strong, well into the early 2000s. Did you do any work in that area and what, if any, were your findings?

[00:09:48] Mika

Yeah. So during my PhD, not. However, this was kind of in between co-founding the fintech and leaving Credit Suisse for the first time. I did a short stint with Didier Sornette at ETH, as I guess a research assistant to some extent. Of course, Didier is not only predominantly in finance known for the LPPL or log periodic power law type of models that try or aim to predict market crashes.

So I worked a bit with Didier on the model, because at the time, he wanted to have a bit of input of maybe somebody who came from a more trading background, which was quite interesting and it’s certainly a very interesting model. Sometimes it works and a lot of times, it doesn’t really work. But it was an interesting experience and Didier is obviously a brilliant mind on his own, so I enjoyed that stint before co-founding the fintech.

[00:10:52] Hari

Absolutely. Crisis prediction is almost the holy grail. If you could do that, then you could run masses of risk most of the time and then wait for the LPPL signal to flash and then you’d be golden. But I guess it’s a very, very hard problem to solve and probably given your experience in 2008, you’re well aware of the risks that can emerge even from a low volatility backdrop. I know 2020 was like that too. So that much is clear.

[00:11:25] Mika

Yeah. If you think about it, crisis or crash prediction is, in a way, a subset of market timing, and market timing is inherently difficult. What is much easier is to get your position sizing to a level of risk that you’re comfortable with. So instead of maybe pinpointing the correct moment to enter and exit, try to develop something that sizes your positions accordingly to your perception of risk and your risk capacity. I think that’s a much better way for most people to approach the problem than trying to find a particular holy grail formula that’s going to give you the answer.

[00:12:05] Hari

That’s very well said. What’s your view on just naive sizing where you scale positions according to 1 over the historical volatility or some expectation of vol?

[00:12:18] Mika

Honestly, I’m not a big fan because it can lead to issues where your position size might still be very large because vol is very low. We are right now in this interesting period where if you compare fixed income markets to equity markets, there’s a clear disparity between how fixed income market prices vol, if you look at simple metrics such as the MOVE, and then in the equity space if you look at the VIX.

I’m not saying these two indicators should trade at a certain ratio to each other because they’re quite differently defined and they look at different things in the options space in terms of implied volatility. But there’s clearly a disconnect as we speak. So either the fixed income market is completely paranoid or the equity market is too complacent about potential growth or slowdown in the economy and etc. So, yeah.

[00:13:19] Hari

So if I saw something like that where the MOVE was high and the VIX (to use a simple proxy) was low, is your thinking putting your macro hat on, I should be a seller of rates vol and a buyer of equity vol or is that far too dangerous given the spread risk?

[00:13:38] Mika

Well, it depends on the positions you’re running first. If you start with an empty book, I think nobody’s going to criticize you for taking that type of spread risk. My personal view on macro right now is the following. I think there’s three approaches you can take in the current framework, where, let’s be honest, a lot of us have been quite surprised how this particular year 2023 has played out so far.

Even if we understand that in the equity space, for example, a lot of key performance is in the mega-cap space and semiconductor space, it’s still quite astonishing how resilient or range-bound the market has been between, let’s say, 3800 or 4200 in the S&P space. That’s fairly clear. So the first approach you can take is if you try a traditional or neoclassical view of how market cycles play out, that is to say, okay, first, the central banks tighten, then maybe between 6 and 12 months, leading indicators peak, earnings decline or compress, margins compress, employment declines, and then suddenly, we hit a recession and inflation subsides and then we can actually cut rates again.

Maybe in this particular classical cycle, we are somewhere in between earnings compressing maybe not as much as people have feared and it just doesn’t still yet trickle down to the employment numbers as we would expect. The way to square that circle is to basically say, well, there are just delays this time possibly because we have put so much money into the system that the savings are still kind of carrying the economy through it. That’s sort of the classical way of looking at it.

Now, the other view you can take is to say that, well, things have changed quite a lot following 2008, which means that it’s more about liquidity or flow of funds; the argument that Michael Howell puts forward. So taking a broad liquidity metric, which is not something as simple as M2, for example, but a much broader way of looking at, for example, how collateral is used to actually fund and roll the debt cycle in our global economy. that’s one way of looking at it.

If you do that and construct indicators in this business type of mindset, then you can indeed make an argument that maybe we’ve already kind of bottomed in terms of the liquidity compression that we have had over the last 6 to 12 months, which means that we might be more in a better place in terms of also equity returns going forward.

[00:16:29] Hari

Now, I have a few questions though because I agree with everything you’ve said broadly speaking. But if liquidity is abundant, why are certain commodities so depressed? That’s point number one. Point number two is, why is the yield curve so persistently inverted? Because I presume that in all of these collateralized or in many of these collateralized trades or structures, the most liquid asset available, which may be the US 10-year note, is posted as collateral which suggests that if the yield curve is severely inverted, there is an excess bid for the liquid asset which is suggestive of maybe not great liquidity in the market. How do you square the circle?

[00:17:19] Mika

Yeah, it’s a very good question. But I think that’s part of the argument to some extent. We’ve gone through a period of austerity, especially in Europe until the COVID episode. There are actually collateral shortages in the market. One of the indications, as you pointed out, are inverted yield curves, but also negative term premium, in particular, in the US. So people are bidding that collateral. Also, obviously, banks who need it for HQLA purposes and whatnot.

The second part of your question, in terms of commodities, well, that clearly points more towards the first model of looking at the economy, which is to say that, well, actually that slowdown is happening. In particular, if you look at data in China, even if you might not fully believe all the PMI numbers that come out of China, but they have been maybe a bit weaker than people would have expected. So clearly, the China engine is not yet starting as we would have hoped. That clearly fuels big commodity demand globally.

So that is maybe the answer to your first part of the question about why you don’t see that on the commodity space, because we might still be in that first model where it’s more led by your traditional neoclassical way of looking at the business cycle.

[00:18:47] Hari

Yeah, this is an excellent point that you’re raising because I historically thought of the problem a bit like this. First rates move in FX perhaps then equities, then real assets, and then the cycle turns upside down. So half a cycle is bonds up, equities up, real assets up and then the cycle flips. But that seems to be broken, as you pointed out, post-2008. You can even see that statistically if you look at trend measures.

So if you say what’s the probability of a sequential move of the type I described, historically it was quite high. Now it seems to be broken or severely extended. How does one deal with that when thinking across asset classes in terms of trading? It’s a very broad question so please answer it how you see fit.

[00:19:43] Mika

If you ask me personally, I’m probably more within the first camp, as in we are still in a fairly conventional business cycle episode. But as we know also from the past, there can be significant lags between the eventual points in the cycle that you effectively get that hit on unemployment. So I’m probably more inclined to the first model. But I think you need to respect Michael Howell’s work because clearly the observation that we moved from an unsecured funding environment pre-2008 to an environment where you need collateral to roll that debt that we have obviously created all over the place in order to cope with the aftermath, in order to keep the system intact.

So, for me, being more that first camp, clearly, if you need to be in equities right now, don’t be in cyclical equities. That’s fairly obvious, right. At these low volatility levels, it’s probably an opportunity to load up on Vega than to sell it.

[00:21:01] Hari

Did you think the post-2008 regulatory changes impacted banks very negatively in a number of ways? One way which is as we discussed, if the yield curve is inverted, that takes away the basic banking model to some degree. Secondly, the increased funding requirements probably make banks more reluctant to take on and warehouse trades and also to offer the same range of structures that they may have previously because they have to think about whether the new customer demand actually lowers the risk they already have on their books. Do you think this is really a negative structural trend for banks? Do banks always find a way, so to speak?

[00:21:52] Mika

What is challenging with respect to investing in bank equities; I think there’s two issues the way I see it. The first one is that the introduction of risk-weighted assets and gauging your capital requirements of that is obviously, let’s say, a setup that might still lead to a case where banks are not capitalized enough. Maybe some more savvy investors actually know that, depending of course what type of bank you run, whether it’s a globally diversified play or just a custodian, that depends, but a bank that takes substantial amount of risk and has teams that optimize RWA usage calculation, etc.

Maybe I’m not going to say there’s some arbitrage in that, but possibly, banks still have quite a lot of leeway to work around the system. So that’s, I think, one issue. The other issue is I think more structurally in the sense that for banks, for a long time, it’s hard to see a significant growth model in terms of a strategy for growth other than saying, okay, I will go to where people need my services more in the future, which might be Asia, for example, or Africa. But that’s more a plan and not a strategy in my view.

[00:23:29] Hari

Yeah. People occasionally ask me, well, why are these autocallable structures offering massive yields? So people, various banks, offer term sheets to high-net-worth individuals and so on. Is it fair to say that these sorts of structures are being peddled by banks even though they seem to have remarkably high yields Because they help the banks offset risks in the rest of their books? In other words, there’s kind of a short vol profile if you’re a buyer of such a structure. At least you’re sure to deep out of the money put on some bespoke basket of names. Is that the reason for a lot of these things coming to the market that superficially seem so attractive but may fill a risk hole in bank balance sheets?

[00:24:18] Mika

Okay, so there’s a couple of things to unpack here. I think in general, you’re right that structured notes issuance, whether it be on the equity side or fixed income side of things, is a source of funding that can be valuable to banks. I think that’s your question in terms of sourcing cash and capital. I think that’s a fair point. What I disagree with is that the auto-callable structure or the trigger redeemable structure is a trivial structure to hedge and replicate. Any kind of worse or forward type of structure is very tricky to manage, especially with its back to your Vega profile and whatnot.

As you also know, there have been many small and big accidents within the autocallable space, so probably as a source of funding unless you do something very short-dated, but then it doesn’t help you so much in terms of stable funding. Probably auto-callable is the wrong way to go about funding. I think you also need to distinguish a bit in terms of the investor type. The structured note, equity investor, it’s more of a beta play. Maybe you pick a couple of stocks where you believe that there’s a bit of alpha.

[00:25:31] Hari

They’ll hang in there.

[00:25:32] Mika

Yeah. On the fixed income side, it’s more like fixed income structures and outside, it’s more a bond replacement type of strategy where you maybe take a bit of risk on top in terms of an exotic component in that particular note and get a yield enhancement on top of what you would have for a, let’s say, conventional corporate bond yield. So I think there are two different investor types. Yeah, I just want to make sure that the listeners understand and I guess most of them do, that the auto-callable structure is not easy to hatch.

Also, what people forget is that there is a significant amount of computational expense as well because it’s a path-dependent structure typically with a daily observation, multiple underlyings. So you need to obviously calculate your PBs and your Greeks, not only overnight, but intraday, and you need to run various stress tests. I’m not sure whether we’re going to go into the XVA topics and potentially FRTB, but even if you move your computation to the cloud, you’re fully loaded cost in a market where margins are obviously also compressing over time, because, imagine when we kind of started in the early 2000s, not everybody could price everything.

Today, if the two of us wanted to establish a broker-dealer with respect to trading exotics and even taking some risk on our books, we can probably start in two months and be operational by just buying a third-party system. It doesn’t even need to be a spreadsheet, it can be like a proper system. But still, there is a computational expense to this. I’m not even talking about support functions, which is obviously market risk, product control and whatnot. I think you need to really look at the fully loaded cost of this type of business in relation to the risk to see whether that actually makes sense for you or not.

[00:27:32] Hari

On a slightly tangential note, is it fair to say that the attractiveness of these high-coupon paying instruments, whether in the right side or in the equity side, is lower now with T bills at 5%?

[00:27:46] Mika

It should be.

[00:27:49] Hari

But is it? I’m not in that market so I don’t know what the attractiveness is for 10% yielding autocallable structure or a yield pickup structure in REITs land or something like that?

[00:28:03] Mika

Well, I guess it depends how you look at the auto-callable. If you look at it as a kind of market-neutral type of instrument or play without any directional component, and if you believe the market is just going to hover in this range-bound area where we are, yeah, maybe that’s a safe play to do, depending also a bit on the maturity of that structure. Or if you believe that directionally, the market is going to go further up, then that thing is going to knock out, you get paid your coupon and you can move on and reinvest.

So I think there’s a directional component in it that we probably don’t have that much in, let’s say, structured rates, which is more a bond replacement type of strategy instead of a beta or directional play.

[00:28:51] Hari

Which is somewhat similar to the thought process behind convertible bond arbitrage. Assuming that you buy or convert where you think the bond floor is hard, you’re basically getting a reasonable coupon and you also participate on the upside. Now, I think you mentioned convertible bond arbitrage in your book. Is that something you’ve done a lot of? Have you followed it over the various cycles? In the late ’90s, I think it was hugely successful and profitable for hedge funds and then had a hiccup in 2002. Every four to six years, it has a hiccup. Is that an area that you’re involved in and do you have a current perspective on it?

[00:29:31] Mika

Not anymore, to be honest with you. We need to distinguish. I guess there’s two forms of arbitrage. One indeed, the one you hinted at which is you kind of try to identify some hidden asset in the corporate bond that is maybe mispriced. But to that point (and I think this is something that you also alluded to in your second book), in these types of plays, it’s very important to understand who owns that convertible bond. It might just be that there is one large player in this network of market participants that owns 80% of the issuance and that’s why the price is not moving, because they’re not selling.

So then this mispricing or potential mispricing that you see maybe on the volatility component of the convertible is actually just an artifact. There’s of course also prospectus arbitrage, where you have somebody going through 50 or 100 pages of paper trying to find some hidden clause or some hidden optionality, which is why you might want to own that corporate debt convertible or not. I think these are the type of plays you can do there.

[00:30:46] Hari

On the buy side, maybe my knowledge is not as deep as it could be. But I would have thought that prospectus arbitrage is not a major component of convertible bond arbitrage because you’re paying through the spreads. You need to take more of a broad stroke view about things. But I presume on your side, which is partly flow driven as well, if not largely flow-driven, you can focus on perhaps more minute mispricings.

[00:31:16] Mika

I think that’s a fair comment. Having said that, converts, it’s clearly a very complicated instrument, hybrid instruments by construction. So you need to pay a lot of attention to respective details in the pricing and potential arbitrage that you might be supporting or not.

[00:31:42] Hari

Now, speaking of complexity, looking through your book again, you have a section on swing options, let’s say, in the energy space. I cannot imagine a more complex balance sheet intensive structure than say an option that gives the buyer of the option flexibility as to how much natural gas they’re going to take within a threshold and when they’re going to take it. Is that the sort of stuff that banks still do or is it just so much of a balance sheet drain that it’s a recent relic of the past?

[00:32:17] Mika

I would say in general, anything that is heavy on the balance sheet is something that only very large players can still do. In this particular example, for sure, it’s something also from the past. We have clearly moved, over the last 15 years or so following GFC, into a world where capital is very precious and constrained even for the larger players. So the capital hurdles that you need to overcome or maybe express more in XVA terms, the KVA that you need to kind of monitor or price in over the lifetime of a transaction is so punitive that these type of plays or levered plays is a relic of the past, as you correctly point out.

[00:33:10] Hari

Are these trades still done by non-bank entities then?

[00:33:13] Mika

I wouldn’t know to be honest with you. But the question is, you need to have both sides on that trade. I’m not sure that anybody would still be keen to put on this type of leverage.

[00:33:30] Hari

So if I had access to the data for common structures that have gone through the markets for years, say, interest rate swaps or swaptions and things like that, would I see wider bid-ask spreads or is it simply now that the depth of the market has gone down?

[00:33:46] Mika

Yeah. I would say that the depth of the market has clearly gone down in liquidity profile, even for things like index options as you probably know. That’s clearly, to some extent, an outcome of the regulatory framework, which is more punitive for banks in that space. It’s more difficult to warehouse risk. It’s more in the hands of, I would say, market makers who are obviously more keen on entering and exiting their respective positions as fast as possible because running large overnight positions is not something that is in your business model. So in that sense, clearly, the liquidity profile of the market has gotten weaker, and that’s cross-asset.

[00:34:37] Hari

Do you see things statistically as well, where markets tend to close at or near their highs or lows more often than they used to, let’s say, particularly on a Friday or preceding a holiday?

[00:34:49] Mika

Honestly, I haven’t seen that too much. I haven’t done that analysis myself in that sense. Also, I guess if you work on the more non-linear side of things, our books tend to be rather slow or the positioning of our books tend to be… I’m using the wrong expression here, but it’s not really day trading per se where these types of seasonal effects are important to us, I would say. But if you have seen that yourself or measured that yourself, I’m inclined to believe you.

[00:35:24] Hari

I can’t say for sure. I won’t comment on that. Maybe yes, maybe no. Yeah. Well, you had mentioned that banks are more reluctant to take on risk. How does cross-margining work nowadays? Let’s say that you, how should I put it, maybe you’re trading in two different markets, you’re trading futures against cash in some market and they trade either on different exchanges where one is listed and requires official posting of margin and the other one isn’t. Is that something that you have to deal with in risk land nowadays given the changes in regulations?

[00:36:06] Mika

Yes and no. Part of that discussion is to some extent taken care of by people who work more let’s say in the XVA space, which is often more segregated compared to the let’s say primary business that we run on the desk. But for sure, CVA pricing or initial margin pricing (which maybe is what you’re alluding to) is something that clearly tier one banks, maybe even tier two banks need to price in, for sure. But it’s part of the mechanics already. It’s almost a default.

[00:36:50] Hari

Yeah, I’m being on the outside here. If a big customer trade goes through, a big deal is going through the market and your bank is thinking about taking the deal, is there an XVA desk that preempts everything or does it go straight to a trading desk? In other words, is the funding desk overseeing and controlling everything that goes through or is it more on a trading desk by trading desk basis?

[00:37:25] Mika

Yeah. I think it’s a joint effort. So if there is a larger client trade, or it could even be something smaller on the OTC side like, I don’t know, a cross currency swap or whatnot, then for sure, the mechanics is such that the trading desk would price up the trade and then either immediately, because we have access to the models that spit out the respective XVA charges, whether that’s CVA, of course, I’d also FVA in case we’re talking about a non-collateralized transaction which is obviously nowadays rare. But if you deal with corporate clients, that can still be the case obviously. Or if you deal with a government entity or quasi-sovereign entity, that can still be the case.

On the KVA side, which is a bit more, I guess, subjective or bespoke, because if you want to do KVA or sort of try to estimate the capital requirements for a particular trade over the lifetime, you implicitly also would need to make assumptions about changes in the regulatory environment going forward, which is challenging, plus your own funding conditions point in time or forward funding conditions point in time. That’s obviously very challenging.

So typically there, it’s more things like capital hurdles that you would look at, which is something obviously we’ve done all the time trading, even pre-2008, in order to assess whether a particular trade actually is profitable with respect to return or risk-adjusted return on capital.

[00:38:59] Hari

Pardon me. Are the regulators a bit like central banks where people are trying to hear their communications and infer what their future course of action is likely to be? You mentioned this idea that KVA, you don’t know what the regulations will be in five years, what the funding requirements will be in the future. So if there’s a long-dated bilateral agreement, how can you forecast the capital requirements not knowing what the regulatory regime will be in the not-too-near future?

[00:39:31] Mika

Yeah, that’s a fair question. What you can do is obviously assume that it might be a bit more punitive and maybe punitive by a certain percentage on, for example, your tier-one capital going forward. Having said that, another thing that has obviously changed over the past 15 years is that we don’t have so many long-dated trades anymore on the books for exactly those reasons. That obviously makes the problem a bit easier along those lines. I think I just mentioned KVA because it’s probably the XVA adjustment that is more discretionary and where maybe the industry is not 100% aligned even at the tier-one space, how you look at that.

[00:40:20] Hari

So what’s to prevent one bank from being far more aggressive and taking all of these deals on their books, and at least in the short term, doing more business than a more conservative bank?

[00:40:33] Mika

Sure. Look, I think it boils down to the type of risk limits and culture that you have within your bank. The word culture obviously is being thrown around quite a lot recently all the time. We all understand it’s kind of an umbrella term for many things, from prudent risk management to alignment of interest to having clear accountability and responsibility on your respective business line and how that can affect the whole bank going forward.

But here, I think we have reached the stage where probably in the next phase, we need to make sure that we can quantify that, what we actually mean by culture. Because otherwise we’re almost making it an HR problem in the sense that, okay, please don’t hire bad people or don’t take stupid risk and that has a very subjective element to it. Plus, people might change. You might get the nicest graduate and once he gets paid or she gets paid a bit more money, they change. So I think we need to express culture more quantitatively. But it’s tricky, right?

Coming back to capital hurdles, if you think about it, if you are the CEO or part of the executive board for a large diversified global bank, then one of your key responsibilities is to make sure that capital and resources is allocated to businesses that have the best risk-adjusted return in the future. But getting that right cross-asset, cross-region with also different regulatory environments is not trivial.

So how do you compare, let’s say, a financing business (which is effectively akin to writing puts, so, that’s great until it doesn’t) to a market making business or to let’s say, a structured notes business in terms of risk-adjusted return in the future? It’s very tricky. But I think that’s where people need to spend the brainpower on in order to get that right. If you get that right, then I think you put a quantitative framework around what we currently call culture.

[00:42:51] Hari

There’s a lot in that comment, a lot of great stuff. It seems to me or maybe again, I’m speaking as an outsider that the bank scene is perhaps overly focused on short-term dynamic management of risk and not focused enough perhaps on generating long optionality in their business lines. Is that a fair comment? If so, do you have any suggestions as to what lines of business actually do increase the positive convexity of payouts within the overall business?

[00:43:26] Mika

Yeah, that’s interesting. To be fair, I’ve never thought about it this way of trying to build portfolios and businesses that have, I guess, as Nassim Taleb would call it, anti-fragile properties or positive convexity. One of the problems with that as well as that, think of a large derivatives book in a, let’s say, structured notes business, for example. Of course, your dream scenario is that you would have 20 different trade types, with offsetting risks, that you can trade every day in order to keep the second and third order and nth order risk on the book flat.

But in reality, especially in the retail space, it’s not that investors want to do 20 things, they typically want to do one or two things. In the equity derivatives space, for example (as you put out correctly before), it’s worse off down on input of type of structures typically. So how do you recycle that? Can you build a portfolio out of that flow business that has positive convexity or great properties in a severe market crash? No, you can’t. The only thing you can do is you can limit the positions.

So I think there’s this dream of building positive convexity in your portfolio or similar to what you might do in asset management or in a hedge fund ideally. I think that’s very challenging simply because the market flows typically don’t allow you to do so.

[00:45:06] Hari

Perhaps you’re too big to do it as well. You can’t just go out and buy a bunch of teeny options in some market where you think vol is cheap and just stick it in the cupboard while you continue doing the rest of your business.

[00:45:20] Mika

Yeah, correct. For sure, if you run a business where you know that in your stress scenarios, you can drop a couple of bucks or millions, you will certainly run this with a breakout skew bias typically, as a hedge overlay all the time. Maybe you even have some macro hedges on top of it just because you cannot hedge out those flows all the time.

[00:45:48] Hari

Indeed. That kind of brings me to a complex systems question and I know you’re going to have lots to say. But how about if I set up an idealized scenario where every bank had exactly the same risk limits, the same basic recipe for modifying positions or reducing risk over time and traded in largely the same markets, wouldn’t there be a crash that would be imminent because everyone would be de-levering or reducing risk at the same time? Isn’t there a risk that if these things become too standardized, it actually increases the fragility in the financial system?

[00:46:28] Mika

Yeah, I guess in the asset management space, similar arguments have been put forward with respect to ETFs or passive investments in general. It is true. I can follow the argument, let’s put it this way, and for sure, as market liquidity has decreased given some of the regulatory impacts and the fact that we cannot warehouse as much risk anymore in certain market-making operations, for example, given, for example, balance sheet constraints.

Indeed, if everybody’s selling at the same time and there is no counterparty willing to pick up the flow, then that market is going to crash for sure. Ideally or hopefully, you will have participants in the market, at that time, from entities such as asset managers, hedge funds, insurance companies that are willing to pick up this collateral or the hedges that you’re selling.

[00:47:26] Hari

At some price, perhaps. But yes.

[00:47:28] Mika

At some price. Unfortunately, and I guess this is maybe to some extent the moral hazard that we’ve created post 2008, very often, it is the central banks that step in to stabilize the system again. We’ve seen that very recently with the turmoil on the regional banking sector in the US and we’ve also seen that during the COVID episode, where maybe, we had this V-shaped recovery following March 2020 or April 2020 and it was all over with because central banks stepped in with swap lines and liquidity facilities in order to calm markets down. But the question is, what type of system have we created following that?

[00:48:16] Hari

Indeed. As a kind of off-the-cuff comment, I would have thought that some randomization in the system is healthy. So for example, if instead of having the S&P 500 with its cap-weighted index and passive flows implicitly propping up the prices of the biggest names perhaps under certain price impact models, if you just had a random index, the weights would randomly be changed every so often.

One could imagine that in the same way that an engineering system can be stabilized sometimes by the introduction of random noise, the financial system might be stabilized by the, maybe not unpredictability necessarily, but just the lack of rigidity in the system. Does the banking system think at all about this sort of problem?

I know that the Bank of England had certain research reports that they wrote about thinking of networks and stability of networks and contagion risk and so on and so forth. Is that something that you’re just too busy as someone who works or has worked at a major investment bank to consider or is it something that is vital to understanding the future of how the network will evolve over time?

[00:49:34] Mika

Well, I believe we clearly think about trade flows, for sure, and client flows because that’s obviously what is driving asset price dynamics quite often. In terms of I think discussing network effects in the domain of complex systems research, as you point out, I think some of the more quant-focused colleagues will do that. But I think it doesn’t go beyond let’s say armchair philosophy, over copied and actually to be implemented in our daily work.

[00:50:13] Hari

Do you think that the complex system’s push in certain parts of finance is a healthy one? Is it overstated a bit like machine learning can be in terms of, at least in the financial markets? Or is it something that is undoubtedly true and will find its way more and more into market practice?

[00:50:36] Mika

That’s a good question. I’m probably biased because as you can imagine, I like thinking about these types of topics. To some extent, maybe we’ve also reached a limit of… If you go back to physics, I think what you’re comparing here, in a way, is kind of a mean field approach, that we have typically in economics and finance, where we average out in order to make things more tractable. But clearly, I think we’ve reached a limit of how far we can go with this type of, let’s say, non-agent-based models, if you want to put that into this term.

So especially in academia I would say, but of course, there are some industry practitioners as well who are deeply involved in this research, should if you think of Boshoff, for example. Yeah, I think this the way we need to push. I don’t think honestly it matters too much as well whether, within five years, we come up with a perfect theory that can maybe help us stabilize markets in ways that we haven’t done so before. But just the fact that we are looking into this realm and trying to move away from a pure mean field approximation is very valuable. But can I see something tangible right now? My answer would be no.

I see something much more tangible with respect to machine learning because that’s what we use every day, whether that’s a regression or something more sophisticated. But in terms of complex system research, tangible on the trading desk, I don’t see that happen yet.

[00:52:24] Hari

I would agree. I’m sort of disappointed that you mentioned the second book that I wrote, I was the lead author on, people often ask me, where do you apply this? Perhaps some very specific cases, but it’s not broadly applicable enough to really be the foundation of an asset management strategy or a strategy within asset management or something like that. It’s a purely episodic way of identifying excess concentration in markets.

Whereas I think the leverage question perhaps is more all-pervasive, the Jeff Snider, Michael Howell world where maybe you can’t say who’s owning what, but you can say how much cash and credit is broadly available in the system. If you make the leap, if you have the leap of faith or make the leap of faith that at the margins, these changes in credit flows, have large impacts on major asset classes, maybe you can do something with that.

[00:53:30] Mika

Yeah, I typically look at it (and I say all the time), I typically look at it when my sort of traditional framework of looking at global macro is not really playing out to my expectation, which is right now as we speak. I think in order to get an idea about especially maybe short-term or tactical directional flow, I think these tools are very good. But I’m still not at a stage where I would be willing to abandon the traditional way of thinking about the interaction between central banks, the banking sector, and then ultimately, the business cycle.

[00:54:11] Hari

Let me give you one example in that direction. So if I look at the short end of the US yield curve, I might say that market expectations are that the Fed funds rate will be 2% lower than where it is today within 18 months. I know the MOVE index is high, and it has been very high recently, but it’s lower than it was, significantly lower than it was a few months ago. I might scratch my head and say, well, look, how can the market be that confident that that aggressive an easing cycle will take place within the next 18 months? Is there an explanation one can give in terms of these more generalized liquidity measures or is it just an anomaly that potentially could be exploited?

[00:55:04] Mika

I’m not sure whether the liquidity measures… The liquidity measures will actually, right now, tell you that the recession that we might be entering into in the US is clearly not going to be as severe as you might want to expect if you look at your traditional ISM measures with respect to economic activity. What is happening in the rate space right now (and I’m just throwing this into the picture) is that there are clearly also many market participants in the space right now who have not experienced this type of rate environment, maybe ever.

So that might be an explanation as to why the volatility is as high because people can just not make sense of what that path is going to be going forward. That’s one way of looking at it. Then, of course, at the end of the day, it’s really I think much about the volatility of inflation going forward and the level of inflation. Today we’re recording this on the 24th of May, I was about to say that inflation is obviously coming down a bit if you look at core inflation in the US, which is great, maybe also not too unexpected given where we started. But we just had a pretty big print again in the UK and that obviously sent yields soaring again.

So if we continue to be in this elevated inflation environment for much longer, I think the rates volatility that we see is going to persist. But at some point, if that’s the case, then indeed, the expectation of the market that we’re going to get rate cuts between 1% or 2% over the course of the next 12 months cannot play out.

[00:57:06] Hari

There are two points that come from that in my head. The first one is that I was taught, in the days when I did learn some quant finance, that the short rate drives the rest of the yield curve, A, and B, yields never go negative. Now, both of those seem to be violated. The second one was violated just numerically, factually, a few years ago.

The first one that I raised, though, is more complex and more in line with these credit / liquidity issues that one thinks about because it seems to me that if the most liquid asset is the 10-year notes (just to pick one), then that should be driving, in some sense, the rest of the yield curve because it’s the thing that trades the most, it’s the thing where people can express their views on the state of the world most directly and effectively with the least risk of not being able to unwind the position. So that seems like a perplexing thing there.

Do you broadly agree with that or is it that actually it’s not that the short end drives the long end but that the yield curve has its own different players and different reasons for moving at different points?

[00:58:23] Mika

No, for sure, I think the yield curve always has had different players playing the respective terms structured differently, that’s for sure. The distortion we see in the front end, given that we’ve raised rates at this rapid pace over the course of 12 months, can certainly explain some of these dynamics that you see. On the 10-year, many lives have been lost trying to predict where that 10-year rate is going to go similar to where the US dollar is going to go. I don’t think that has necessarily changed.

But going forward, you can probably extract more information from the respective slope and curve inversion or curvature for that matter, instead of maybe looking at individual points. Clearly, the market right now, to be honest with you, is confused. It’s not really quite clear where things are going to go.

[00:59:20] Hari

You mentioned the negative term premium which has been persistent. My knowledge was that the term premium was one of the few mean reverting, truly mean reverting quantities in the rates world other than maybe volatility. Do you see that reverting at any point in the near future or is this just a persistent, stubborn feature of modern collateralized markets?

[00:59:46] Mika

I think there, I’m pretty much in Michael Howell’s camp. I think the negative term premia, that we currently see, is clearly maybe not predominantly driven, but there’s a large component due to collateral shortages. So I completely agree with him there. I don’t really see that going away unless w start issuing more bonds which depends how the US deals with the debt ceiling now. So we’ll see which points they will choose on the curve. I guess it will be more on the short duration metric in order for the market to absorb that. But unless we have more collateral available, I don’t see those negative term premia going anywhere to be honest.

[01:00:33] Hari

Fair enough. I’ve sort of reached a gap in the discussion. Was there anything you’d like to talk about or add before we wrap it up?

[01:00:41] Mika

No, I think it’s fine. It’s been quite broad. We covered quite a lot of topics to be honest and I enjoyed that very much. Hopefully, your listeners could extract some nuggets of wisdom here and there.

[01:00:56] Hari

So have I. I’m very impressed by the breadth as well as the depth of your knowledge on these things, so it’s been a real pleasure. With that, I hand it back to Niels.

[01:01:10] Niels

Thank you so much Hari and Mika for a delightful conversation about some of the inner workings of the markets we trade and the banks that we interact with when doing so. There were quite a few takeaways that I thought were very interesting. In particular, to what extent markets can be modeled as complex adaptive systems, as well as to what extent regulation constrained bank activity in the markets, and of course, not least, what to make of the apparent distortions of the markets like the yield curve and the current disconnect between bond volatility and equity volatility.

I also enjoyed learning about the LPPL model that is trying to predict market crashes, something that a lot of people seem to be focusing on especially if you read the headlines on YouTube. Make sure you go and follow Mika’s and Hari’s work because as you can tell from today’s conversation, there are many exciting facets to learn from those who have been in the trenches for many years, and we really look forward to exploring many more of them as our series continue.

From Hari and me, thanks so much for listening. We look forward to being back with you with the next episode. In the meantime, take care of yourself and take care of each other.

[01:02:25] Ending

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today’s episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you’ll be sure to get all the new episodes as they’re released.

We have some amazing guests lined up for you and to ensure our show continues to grow, please leave us an honest rating and review on iTunes. It only takes a minute and it’s the best way to show us you love the podcast. We’ll see you next time on Top Traders Unplugged.