Hari Krishnan, Head of Volatility Strategies at SCT Capital Management

[00:00:01] Hello, this is Dean Curnutt and welcome to the Alpha Exchange, where we explore topics in financial markets associated with managing risk, generating return, and the deployment of capital in the alternative investment industry.

[00:00:15] It was a pleasure to host a conversation with Hari Krishnan, Head of Volatility Strategies at SCT Capital Management, on the changing nature of volatility markets, portfolio hedging, and why commodities may offer increasingly valuable diversification in today's environment. Hari reflects on his book, Second Leg Down, which explores practical approaches to tail risk hedging and the cyclical nature of volatility.

[00:00:44] He discusses how investors often ignore protection in calm periods, only to rush towards hedges after markets have already repriced risk. That dynamic leads to a broader conversation on planning, budgeting, and approaching hedging as an ongoing portfolio discipline rather than a reactive decision. We then turn to option markets more broadly, including vol risk premium, SKU, and the challenge of protecting against fat-tailed outcomes.

[00:01:11] Hari explains why moderately out-of-the-money options often embed persistent premium, while deeper tail risks can be difficult to price with confidence. The conversation then shifts to commodities, where Hari sees a differentiated opportunity set. We discuss how producer hedging, end-user demand, and forward-curve dynamics create a very different vol ecosystem than that in equities.

[00:01:35] He outlines a strategy focused on gaining long exposure to select commodities while using option structures to reduce carry costs and preserve upside convexity. We close with a discussion on cross-asset dislocations, the recent divergence between oil, gold, and equities, the role of commodities in a world where bonds may be less defensive, and how AI tools are accelerating research, customization, and hypothesis testing across markets.

[00:02:04] I hope you enjoy this episode of the Alpha Exchange, my conversation with Hari Krishnan. My guest today on the Alpha Exchange is Hari Krishnan. He is the head of volatility strategies at SCT Capital Management. Hari, it's great to have met you recently, and it's a pleasure to welcome you to the podcast. It's my pleasure, actually, Dean. We've been running into each other quite a bit recently. I'm enjoying it every time, so thanks for having me.

[00:02:32] Yeah, this is one of those instances where we're very much in the same sandbox. I think we know a lot of the same folks. It was a LinkedIn post where you were talking about some of the collaboration you've done with Mike Green on Passive, something he's focused on for years. So I took great interest in your piece, reached out to you, and here we are. We've gotten to know each other a little bit, and I am excited to host this conversation

[00:02:59] and to solicit some of your insights on all things vol, both in assets like the S&P, and then in very different assets like corn. So as we get started, tell us a little bit about your background. Well, once upon a time, I did a PhD in applied math. I specialized in chaos theory, which was a pretty good description for my life at the time, perhaps. And perhaps a description of today's markets. Yeah, I'm well prepared for it without realizing. And then the bulk of my career has been at three hedge funds.

[00:03:29] I worked as a global macro PM at Cross Border Capital in London. I worked as a vol PM, mainly in S&Ps and to a lesser degree rates, at Doherty Advisors in New York. And I've been working at SCT Capital more on the systematic slash machine learning side for the past five years. So that, altogether, that comprises about almost 20 years. They're all mid-sized hedge funds, and so I've learned a lot of different stuff going from one to the other.

[00:03:59] Well, if you've been looking at markets for 20 years, you've seen not everything, but just about everything. It's, again, perhaps a sign of the times and a sign of the chaos. You've written a book that has always gotten a lot of really good feedback from industry participants, and I've just picked it up and started to spend a little bit of time on it. That's almost where I wanted to begin. It's called Second Leg Down.

[00:04:26] And in some ways, it seems that it is a reflection of just how you have learned to think about markets and specifically about vol and using option-related instruments in a portfolio. Tell us a little bit about how you came to write the book and then the big picture thrust of what's in there. Well, I kind of stumbled into the tail risk space or the long vol space in 2007,

[00:04:53] which was an interesting time to stumble into it, so to speak. I ran a couple of substantial mandates. One was for a family office and one was for a charitable foundation, both in the UK. It was either boring or scary running long vol hedging programs because nothing would happen. I would bleed for a long period of time. People would lose interest. It was hard to market. And then all of a sudden, there'd be blood on the streets, for lack of a better phrase, and people would call me up.

[00:05:23] And they'd say, what's your P&L today? What's your P&L in the middle of the night? We need to put more hedging on. I feel scared. I'm underprotected. What should I do? And so it would go from a state of supreme calm and indifference to abject fear as markets changed. So I wanted to write a book that would address that issue by assuming that clients wouldn't be interested in hedging all the time. But when market conditions got worse, they suddenly would.

[00:05:50] And what could I do that would be reasonably efficient to protect capital, even when volatility had gone bad? So my approach to that was to think in terms of almost a volatility cycle, a bit like an economic cycle where everything is calm initially. So you can almost think as though as if you're a private equity investor and you warehouse long-dated insurance. And then if there is a bit of a blip in the market to the downside, the short end of the curve tends to go bid.

[00:06:20] So short-dated vol goes up. And so that lends itself to various spread strategies at the short end while maintaining long vega positions at the long end. And then as conditions get progressively worse, everything richens up. And so from there, I got into these very short-dated gamma hedges, high gamma, low vega hedges, which not by design, turned out to be well ahead of its time because I talked about intra-week options when there were only weeklies available,

[00:06:49] well before the zero DT mania. And I talked about the econophysics of fat tails. In other words, tails tended to be much fatter over one-minute horizons than over, say, three to six-month horizons. And there's a gradual transition from 10 standard deviation-type moves that occur every now and again on intraday horizons to very rare ones as you go out over a couple of weeks or so. And so I got into the econophysics of that, and I wanted to tell people about it

[00:07:19] and give practical takeaways as to how they could use these ideas in protecting their book, their portfolios. So that's how I got into it. From a sales standpoint, it was pretty shaky because it's not a textbook, as you pointed out. So I couldn't get guaranteed sales through that. The academics basically looked the other way. It seemed to be a strange beast for them. The RIA community or the more wealth management community back in the day was far less sophisticated than it is today.

[00:07:49] So they didn't have a clue what I was talking about. And so it was a slow hit, as my wife might put it, where the word got around. I didn't really market it. And thankfully, it does as well today as it did 10 years ago. I'm very happy about that. Yeah, I think that is probably some outcome of its relevance to folks like me in the practitioner space. I want to ask you a couple things that I picked up from the book.

[00:08:16] And I wanted to start out by just having you reflect a little bit on a couple of statements I'll make. The first is that in options, and we can use the S&P as an example, but you can go across asset classes and you can even include assets like car insurance and homeowners insurance. These folks are not selling insurance without the motivation to profit. And so what we observe in, again, whether you're looking at Geico,

[00:08:46] Allstate or Goldman Sachs, the seller of insurance is typically compensated. There is the vol risk premium. We also know, as you just referenced, that there is a fat tail, oftentimes a left fat tail, especially in financial assets like the S&P. And so you can win, win, win, win, and lose a lot. There is the skewness of negative returns.

[00:09:10] How should people think about the big picture of hedging set against the VRP and also set against the experience that the five standard deviation event in markets happens much more frequently than a model can really capture, a distributional model? Okay, there's a lot in there. There certainly is a VRP. So most options who cheat are expensive most of the time.

[00:09:37] So you can make money selling them if you're disciplined in the way that you do it. I would argue that options that have strikes that are moderately out of the money tend to be, on average, overpriced. And those options tend to correspond to scenarios which people think are possible and wish to defend against. So for the S&P, it might be a 5% to 10% one-month down move.

[00:10:03] In other markets, it might be higher or lower depending on the average tendency of that market to move. Now, someone like Nassim Taleb might say that, I don't want to quote him inaccurately, but he might say that the tails are underpriced. So very, very low delta options, either on the downside or the upside, are cheaper than the market thinks they should be. I wouldn't necessarily argue that. What I would argue is that they're unpricable.

[00:10:29] It's very hard to run the statistics to figure out what fair value is because these mega events don't occur very often. And so my view is that if you can combine those teeny options on the long side, if you can always own them, you can always own the wings and combine that with stuff that can actually be statistically modeled, you get interesting outcomes. And so one of the ideas that I presented in The Second Leg Down was the idea of selling

[00:10:56] one 25 or 30 delta put on the S&P and buying, say, two, three, or even more five or 10 delta puts. That's a hard position to manage, but it's a good expression of the idea that tails are either cheap or unpricable, and the stuff in between tends to be statistically overpriced. Now, with regards to the question of insurance companies, sure, they make money because statistically

[00:11:23] the insurance policies they sell are rich. They're overpriced. But markets are a little bit different because of feedback. So I would agree that over the market cycle, however you choose to define it, options are probably overpriced. So if you took a typical option in a typical market over a six to seven year period, so let's say you just track the performance of the 25 delta cull in crude oil over a seven year horizon, it would probably go down.

[00:11:53] And so you would lose money simply because options are over the cycle overpriced. But especially for financial assets, there are waves of euphoria or complacency as well as periods of fear. And in the complacent periods, it's sometimes the case that buying options can be really attractive or even cheap. So it depends. It would be my reply there. I saw this in your book.

[00:12:22] Some people call them backspreads. You called it a one by two put spread flipped upside down. I will say that I have done a lot in that space as well. We saw, gosh, this had to have been like 2010 or 11. And I think the S&P might have been at 900 or so. And we saw the 600, 500 one by two trade. I saw the same trade. Yeah. Yeah. And I looked at this thing and I said, if someone thinks that's a hedge, if someone bought

[00:12:52] one to sell to and they were told that that's a hedge, boy, that's not going to be, it might work, but it's not going to work for the right reasons. And we started looking at it and we said, what if we just flip this thing upside down? And we called it the teeny put reaper, which kind of the name really caught fire with us and modeled out the Greeks in the same way. You get this vega convexity, a really interesting way of implementing in some ways a carry neutral.

[00:13:21] Not always. Of course, the Greeks will change along the way, but I was really happy to see you cite that. I saw the same trade go through the market in 08. And I wondered about it too. That's how I learned about it actually. And I viewed it from a macro lens, through a macro lens, which was whoever put the trade on, if they weren't doing something else, if that were a standalone trade, which is a big they were expressing the view that the S&P is not going to stick here. It's either going to crash or it's going to recover.

[00:13:52] I can't afford for it to crash, but by the same token, I don't want to pay through the nose for insurance because vol has gone bid, tremendously bid. So I'm going to do a trade where if the market does break to the downside and does so quickly, I'm going to make a loss, but I'm not really going to pay for it. So if there's a rebound, I'm not going to pay for it. So the bet there, the embedded bet is there's not going to be a trend down.

[00:14:18] It's either going to shank big time or it's going to stabilize and or recover. And for me, that's a very attractive way to play insurance because you're not doing a calendar spread. You're not leaving all this open-ended risk, which I have done in my history, sometimes effectively, sometimes with shocking changes in risk profile over time. But you're not taking calendar risk here. And you're also expressing a view that this is not equilibrium.

[00:14:47] Whatever equilibrium might be, this isn't it. Things will stabilize higher where they'll just completely collapse. Right. And so your setup is if you break this thing apart, you've bought two very far out of the money puts and you financed some of it, maybe a good portion of it by selling same month, but a higher strike single put. As you say, if you get the giant shock down because you're long two, you're net long vol,

[00:15:16] you're going to get shorter delta on the way down. That can be attractive. There is management of it. So if you do, I don't know, a three or four month trade and two months go by, it starts to change. Its Greek profile starts to change. How do you think about the management part of it? Well, I said a few things in the book and I kind of stick to them. One of them is that the short ratio spread, even though it's a great emergency hedge, is even more attractive in quiet markets.

[00:15:44] Because if at the money vol is low and skew is not too high, those strikes that you're short and long respectively get much closer together. And so it's a much easier position to manage. If vol is very high, then you have to, in order to be premium neutral, let's say, you have to move those strikes pretty far apart. If you're selling one to buy two or even worse, selling one to buy three, that low strike is going to be way down there. Now that strike doesn't have to get hit.

[00:16:13] One of the big things about tail risk hedging is that you don't need anything to go in the money to make a lot. All you need is a dramatic repricing of the premium you paid to put the trades on. But having said that, if you're short one to buy two below on the put sides, let's say for the S&P, and the market trends down, so you get a 2022 type scenario where vol is pretty high going into the year, and then the S&P trends down.

[00:16:41] It's a tough position to manage because you're basically long a straddle that is running away from you in terms of time decay every day that nothing happens. It's a trade where you really hope something happens quickly, then you don't have to manage it. If it doesn't happen quickly, then you're playing the realized versus implied game, which may or may not be reflective of your initial best. Yeah. Stepping back, one of the things you say earlier in the book is having a plan.

[00:17:10] And you and I talked a little bit about this, and I said, you see so much focus, rightly so, on the long side of the ledger, the risk on side of the ledger. Folks have a process. They are interested in re-underwriting positions all the time. They think all the time about sizing. What you and I have both observed over time is that you get a quiet period in markets, and then suddenly something pops up.

[00:17:36] Maybe it's the last couple of months with oil, and everybody is rushing to implement protection, but they don't really have a game plan. They haven't really thought about budgeting. I'd love to get a little bit of your thinking on what you've seen over time and how you recommend your clients just try to come with a plan and execute on the plan. If I had to put it on the back of an envelope, because this can be pretty complex, what I'd

[00:18:02] say is if you're worried for structural reasons but realize vol is pretty low, it's not too hard a game to play. Just buy some protection. Buy some puts. Buy some calls. Try and find cheap pockets in the vol surface that are reflective of your macro view or your pressure points in terms of risk. Once things have gone south or north in the case of oil, as the case may be, spreads are a lot more attractive in my view, just as a rule of thumb.

[00:18:30] If oil's gone from 50 to 75 and you're worried it's going to go to 100, which it did, you're better off buying a wide call spread than just buying an out-of-the-money call simply because the price of severe upside risk has already gone up. You really don't want to be joining the crowd to buy an asset that's overpriced over the cycle and now hideously overpriced given what people have to do to manage their risk.

[00:18:58] So I think trading spreads is a very good strategy if you're reactive, if you're anticipatory and you're thinking about structural risks in the market. So if you take the view that I do that the market is nothing more than a set of positions and if prices reach certain levels then there can be follow-through, feedback-induced follow-through, then just buying options can be very attractive if things haven't repriced yet in the options markets.

[00:19:29] Some of the more savvy global macro funds do exactly that. So you could think of oil where if inventories are low, prices could still be low. But the sensitivity to a shock is very high. So if you want a not-too-risky way to play that, you can just outright buy a low delta call auction. Let it sit. Just put it away. Don't think about it. And then if there is a major move, you'll be heavily on site.

[00:19:57] One of the things you just alluded to was in some ways crowding. And you talk about this a little bit in your book as well is trying to understand who's alongside you. Where there's concentrations of overlapping ownership. And I think especially where the reaction function of the people that own the same thing is also very similar.

[00:20:20] So if a lot of people who are, let's say, vol-sensitive, mark-to-market-sensitive folks are short, I don't know, downside puts and hedging according to a similar protocol and vulnerable to mark-to-market risk, that can produce an outsized move. How have you thought about overlapping exposures across different constituents in the market?

[00:20:46] And how does that impact how you think about trade construction and positioning? Every now and again, there's an easy setup to analyze, even if it's ex-post. So the vol-mageddon in February 2018 is a good case in point. It was possible to pretty precisely map out how much the ETMs would have to trade, either the levered or the inverse ETMs would have to trade.

[00:21:11] Given an initial move in the front month VIX futures, if you had a price impact model, you could map the follow-through. So it was very attractive to buy way out of the money calls on the VIX as those ETMs grew in size where they dominated the market. In general, though, it isn't quite so easy to do. The way I think about it is just to have a series of agents. It's almost as though I have primitive versions of standard models on my desk. So I have a trend-following model.

[00:21:41] I have a vol-control model. A couple of other ones. I can't even remember them off the top of my head. But mainly trend-following, vol-control, and one or two others. Risk parity or something. And you want to see where they are going to have to rebalance. So if you can build a primitive version, not for alpha generation purposes, but for understanding positioning in the markets, you can go quite a long way with that.

[00:22:07] But it may be said that an even more direct and clean way to do it is just by looking at the price at full. So you can tell if there's buying or selling just by looking at the skew or the term structure volatility in various markets and seeing where the distortions are relative to history. So price is as good an arbitrary positioning as anything in the auctions markets. Druckenmiller said, the S&P 500 is the best economist I know. I'm a big fan of Druckenmiller, so yeah.

[00:22:37] The information content of price is a fabulous thing. It doesn't mean the markets are right in their predictions, but there is capital that's establishing those clearing prices that is meaningful. I want to get your take on what we're in right now in terms of market dynamics. So February 28th, the US and Israel bomb Iran.

[00:23:03] It leads to a pretty protracted spot up vol up in crude. Just following 10 delta call vol got to 120. Crude obviously got up to at least Brent to, I want to say 108, 110. And the market was pretty forward looking in terms of embedding risk premium. VIX got to 31, even without realized going that high.

[00:23:28] And then, of course, Trump epically climbed down and said, we're not going to wipe a country off the face of the earth. And the VIX went back down to 18 or so. And the S&P, of course, has rallied tremendously. I'd love to just turn it over to you. And what have you observed? You're in the commodity markets trading vol. And of course, in more financial assets like the S&P, what are your takeaways from these last 50 days?

[00:23:57] The pieces don't fit together in my head. So I'm going to make a little bit of a cop out. I would have thought natural gas would be a lot higher. I would have thought wheat would be a lot higher. I think they're all tangentially exposed to the same issues that crude oil is. I would have thought that risky assets, particularly the S&P, would be lower. Of course, who am I to make pronouncements about the S&P? It goes up a lot of the time. The median drift is higher than it's ever been.

[00:24:24] And it's more flow-driven and perhaps less fundamentally driven than it's ever been. I don't want to make that as a pronouncement because there have historically been bubbles and various things that occurred in the past. But certainly, it's more flow-driven than it's ever been. And so I do see a strong disconnect in terms of cross-asset volatility at this point. I'm pretty sure there are very rich opportunities in trading volatility in certain markets against vol in other markets.

[00:24:54] Structurally, I would be long S&P vol versus energy vol at this point, at least oil vol. But to do that requires a really good safety net, either in terms of buying extreme out-of-the-money options to hedge against unpalatable or to hedge against a severe continuation of what's going on, or through more dynamic strategies.

[00:25:18] I tend not to be a believer that an option can be directly replicated through dynamic futures trading with a T-bill on the side. I don't really believe that. I think options adapt more continuously and reflect more of the future and more about sentiment than futures ever can. But I would suspect that a very sophisticated, higher-frequency trading outfit could extract

[00:25:43] from the differential between implied and realized in markets such as crude oil. And in fact, we have seen a reasonable drop-off in implied volatility for crude. I think we're down to around 80 in the front month, low 60s in the second month, which I think is July now, and so on. Once things do stabilize, even though the price of oil is still persistently high,

[00:26:07] if realized vol isn't continuing to be high, implied does tend to come down over time. You know, each of these crisis episodes, there's certainly commonality. Vol goes up and risky assets go down. That's certainly the big picture, but there's always something, at least I found, at the center of the storm. 2011, the euro was kind of a vix. Certainly, Italian sovereign spreads were a vix.

[00:26:35] 2016, British pound vol, ahead of Brexit, was a vix. You could draw that. You could map those together. And now, most recently, crude is really at the epicenter of these correlations to the S&P, to the vix, to things like the move index, even to gold. And this is what I was going to point out. You would think that they would be positively correlated, but as of late, at least on a daily basis, they've been highly negatively correlated.

[00:27:03] And gold is behaving like the risk-on asset, and oil as the risk-off asset. So they're moving in opposite directions at this point. And you make a great point on crude is back to where it was a couple weeks ago. But when it was this high, vol was higher. The market was... It's almost as if, I'd be curious to get your take, the market is settling in to this. Look, we may not see a real escalation from here,

[00:27:30] but de-escalation, i.e., much lower crude prices, is going to be difficult as well. So we could be in this crude is high, but crude vol is not off the charts as it was two or three weeks ago. The market... The S&P reached a low when? In March 2009? Something like that? Yep. And vol was far lower than it was in October 2008. Great point. So the stickiness of the situation doesn't lead to higher vol.

[00:27:57] It's usually the initial astonishing move that drives vol higher. When we talk about different asset classes, some are just better linked than others. Certainly, the co-movement of the VIX and credit spreads, well-established, Merton model type stuff, rate vol and S&P vol related, maybe not as tight as credit spreads, the VIX. But you trade a lot of commodity vol,

[00:28:26] and not just crude, but in things like corn and wheat. And that's really where I wanted to go and spend some time because it's a very unique set of risk outcomes, different. And I think if there's one thing in today's environment for investing, the need to find things that are different is really important. Your work with Mike Green on passive,

[00:28:53] we've got seven stocks, 40% of the S&P. It's epically concentrated. The tightness of correlations between the S&P and some of its close sister assets like credit markets, and even, and you referenced this in your book as well, stocks and bonds are not zigging and zagging as they used to. The risk-free asset can be risky, especially in today's world of broken politics.

[00:29:22] But commodities, I think, are different. And I'd love to learn how you got to start trading both futures and options in things like corn and wheat. Tell us a little bit about what you're up to there, and then we'll drill down. No pun intended, yeah. I had to say that. I got into commodities because I've started to feel that's where the action is. I bought into the story told by many people, including Adam Rosenzweig and many others,

[00:29:50] that we might be going into a commodity super cycle. Now, many of these managers are in the space, so they're incentivized to say it, but there did seem to be a lot of geopolitical reasons and demographic reasons that commodities might become more expensive over time. And so I did what I always do, which is to start with what I know. I kind of understood, maybe not as well as certain people in the world, but I knew pretty well where positioning risk was

[00:30:18] in the S&P, how to model that, why there was a put skew, how to attempt to monetize it, and so on. And I figured that since volatility could be suppressed in the S&P and other financial assets, it might be interesting to go in the opposite direction and look at the call skew for various commodities. And when I started this voyage into real assets, a lot of commodities were very cheap. So all the grains were cheap. This was probably two or three years ago.

[00:30:48] Oil wasn't too expensive. Natural gas was very cheap. Metals were cheap. Even gold, surprisingly, was not expensive. But industrial metals were very cheap. Things like copper and so on. Things that were needed as inputs for AI. By the same token, when a commodity is cheap relative to history, the risk is really asymmetric. And the example I give every time, because I'm lacking in imagination perhaps, is that when corn is $3 a bushel,

[00:31:17] the probability that it's, the odds that it's going to go up by $2 versus down by $2 are much more skewed to the upside than when corn is $6 a bushel. So cheap commodities tend to have significant positive skew. But you pay a price to try and tap into that using futures, because that upside skew is partially offset by the steep contango in the forward curve. So if corn is trading at $3

[00:31:46] and you buy the front month futures, so if the spot's at $3, you're probably paying $3.50 for the front month futures with 30 days to go, let's say. So you have to keep buying and rolling, selling cheap and buying rich to keep the position going. So you're at an implicitly short time decay, short carry at least. And so what I did was I repurposed a lot of the hedging techniques that allowed me to not completely eliminate, but heavily reduce the cost of warehousing or carrying

[00:32:15] hedging positions into the commodities markets. So this is a bold claim, but I think at least statistically I'm able to cut about 50% to 75% of the roll costs from warehousing a long position to buying and rolling a long futures position in cheap commodities. And so that was the genesis of it. I figured that would be an interesting idea, an interesting concept for people who bought into

[00:32:44] the structural inflation or structural increase in commodities thesis, but didn't really want to have to mess with the physical or deal with managers who would go long and short or do complex relative value trades. So it's a slightly simpler strategy. It may have less alpha, but it's designed to give people exposure at low cost to cheap real assets. So that was the goal of it. And basically it involved re-engineering a lot of the hedging strategies

[00:33:13] to cover the call skew. And a big part of that was trying to figure out who the natural buyers and sellers of vol are in those markets. Now for the S&P, the world and everybody and their dog is long the S&P. So puts tend to be rich because people want to buy downside protection. Some income producing strategies sell colts. So you tend to have a lot of collars on the S&P, which leads to an exaggeration of the put skew.

[00:33:43] Fine. In commodities, it's a more complex setup. Because the natural thing for producers to do, at least what they're advised to do, is to sell futures on the way up to reduce sensitivity to the underlying commodity price for the thing they're producing and lock in revenues. So to immunize themselves against a sharp decline either in the price of a grain at the harvest or in the price of an energy or whatever at the point of sale. But end users,

[00:34:13] such as airlines, who are sensitive to the price of jet fuel or utilities that might be sensitive to the price of natural gas, tend to buy out of the money call options because they are in a different situation where a small increase in price doesn't directly hit them, hit their bottom line because they can adjust their pricing and they can use substitution and so on. But significant increases can be problematic. And understanding how those players

[00:34:41] were natural buyers of vol, along with the macro crowd, perhaps, who want to have access to non-recourse leverage and want to express high conviction views based perhaps on fundamental inputs, they might be natural buyers. But there are a slew of other people who are natural sellers of vol. QIS groups who are trying to extract, as we discussed, the volatility risk premium, hedge funds or higher frequency hedge funds that believe they can actively

[00:35:10] delta hedge their exposures, large trading desks, say at the big commodities houses that have a variety of positions and may be looking to arb one against another or to create income off the physical holding they may have. They tend to be natural sellers and I would argue that the buyers dominate the sellers on average in the commodities markets. So I think out of the money call options

[00:35:39] that the 25 to 30 delta strike type things tend to be a little bit rich in the same way that puts in the S&P tend to be rich. However, what I would say and I'm rambling on a bit is that the skew is a bit more like in rates than S&Ps. It's a bit more of a chameleon style skew. So if you look at the trend in price over the last six months, let's say, I'm picking a, I'm fixing my parameters a bit too much. If the trend

[00:36:08] is really severely up, you'll see a call skew. If the trend is really severely down, you'll see a flat skew or a put skew. So a lot of the hedging that probably goes through the market is reactive. Otherwise, that wouldn't be the case. Whereas for the S&P, you rarely see a call skew. You'll see it in individual names. And again, you would know more about this than I do. But for indices, you'll see a persistent put skew. For commodities, it's a bit more variable in terms of what the price action has been doing. You've talked

[00:36:37] a little bit about the folks who are operating in some portion of the ecosystem of supply and demand. There are sellers, there are buyers, and they are doing so at different strike points. If you were to step back, this is probably too broad a question, but I'm sure you've studied it in some way. If you were to help us understand the vol risk premium in the S&P as compared to some of what you've observed

[00:37:06] on the commodity side, the risk premium, what's similar? What's different? Is there the notion of, hey, I'm going to go into this market, all of them, and I'm just going to be a seller of vol, and I'm going to get nicked big time occasionally, but on average, I'm going to collect the commodity vol risk premium. Is there that notion or it's just too episodic? What are your conclusions

[00:37:36] having studied this? Well, there are two risk premiums. One is the forward curve and the other is implied vol. So for the S&P, the forward curve is boring. It's purely based on the differential between the risk-free rate and the annualized dividend yields. For commodities, it can be all over the shop. And what I can say that I've done is I've looked at, I think we discussed this line, the relative wildness of term structure movements across commodity market categories.

[00:38:05] So electricity has wild and crazy moves where one month can move down and another month can move up. And the variability is very high even if you normalize for standard deviation. Natural gas is also very volatile across the term structure. For storable commodities such as gold and copper, they look a heck of a lot more like the S&P in some ways. There are shortages, the curves can go into backwardation, but the cash and carry

[00:38:35] relationship is far more stable. So selling vol in and of itself is a bit tricky because as soon as you migrate away from the markets where the forward curve dynamics are pretty clean, you have to be very careful about where you're selling the vol. If you sell it in the wrong month, you might think you're playing it safe or you might think that you've extracted edge because say November corn has a higher implied vol than July, but the reason November corn may have a higher vol is because that's harvest season.

[00:39:06] So there's more uncertainty surrounding it. It's a bit like ignoring earnings when trying to buy or sell vol on individual names. If you didn't know what you were doing, you probably just sell vol willy-nilly before earnings because you wouldn't understand why the risk premium had gone bad. But here you have a similar situation where even if the events are pretty predictable, there are regular harvests, there are regular winters, there are regular summers, people tend to do the same

[00:39:35] thing when it's really cold and when it's really warm. Still, you need to be a lot more careful about playing the interaction between the risk embedded in the volatility surface and the risk embedded in the forward curve. And the interaction between those two is pretty complicated. I've tried to figure out which predicts the other. And here I can only supply a negative result, which is that it doesn't seem that either predicts the other reliably. But what is true is that the recent trend predicts both.

[00:40:05] So if the trend is sharp and up, as we've seen for crude oil, the front end of the term structure tends to rally. And implied vol tends to go up. Now that can create very interesting ways to go long. If the term structure is very elevated at the short end and volatility is very elevated at the short end, it makes some sense to buy further out goal spreads or further out futures because you get some roll up, assuming the prices remain

[00:40:34] sticky in the spot market, and you do it at lower implied volatility. So you're getting a discount in terms of betting going long in that market by deferring your bets. So lots of opportunities open up in that kind of multi-dimensional world that's based on a very simple asset at some level. It's very different from the stock market in that sense. The relationship between a time curve, just

[00:41:03] a futures curve, or even a yield curve, and vol seems pretty consistent. Inverted yield curves are higher vol periods. The market's trying to find a new home in terms of where Fed policy is going. We right now have a pretty inverted curve for crude, backward-dated curve for crude. That's associated with 80 to 100 in terms of implied vol. Those seem to be related.

[00:41:33] I want you to go through this in a little bit more detail because I think it figures into some of your trade construction process. And I just want to go through it because it is a little complicated. So what you've said is, number one, you typically have both of these things happen at the same time. The crude vol curve is inverted, and the futures curve is inverted at the same time. I'm betting that's reasonably consistent. That's just not an accident.

[00:42:02] That's realize vol explodes, people want to buy gamma, and they flock to the front, the shortest dated option. And of course, the time curve does the same thing, the futures curve. So walk us through this idea of owning these longer dated call spreads and trying to use both the vol curve and the time curve putting that in your back pocket and making that something that helps you from a carry standpoint. Let me take

[00:42:32] out the spread component and just say talk about buying a call. This is well known nowadays, but it probably wasn't that well known five or ten years ago, which is that although you probably knew, there are ways in which you can buy an option and actually experience little to no time decay. The option itself may decay, but if the forward curve is in severe backwardation for a market where you buy a call, the roll-up in the forward curve can pay the rent of

[00:43:01] time decay for the option. And that is a very attractive place to be if you think that the market is going to go up because you're effectively constructing a zero carry cost, long convexity play on a given market. And it is kind of weird, but the shape of the forward curve has no direct bearing on the price of an option. The only thing that matters is the volatility that the market

[00:43:31] assigns to that tenor and the price level of that tenor. Nothing else matters. So if the curve is in severe contango or severe backwardation, that should have no impact or no material impact on the price of the option. And the reasoning for that is a Mertens-style replication argument, that you can basically immunize yourself against shape changes by simply dynamically trading the futures against the options position. But the

[00:44:00] reality is for many of us who like to buy stuff and let it sit and don't delta hedge, either through laziness or a desire to express a macro view in an uncorrupted way, that can be very beneficial. And at the long end of when you start buying long dated options, you're way outside of the world of financial engineering 101 because you're making bets on sentiment, on forward curve dynamics, and try to align those with your view. And that's a very different place.

[00:44:30] You're basically long vega, very long vega, so you're long uncertainty or fear, you're not very sensitive to small price movements, and you're very sensitive to changes in curve dynamics. So you have a lot of row, and you have a lot of vega, and your investment style changes quite dramatically as a result. In terms of the call spread, that would be a very specific trade that you might do now for crude oil, where you might buy, say, a

[00:44:59] September 9100 strike. The call spread, and the idea there is that the call skew is steep, that steepness in the skew propagates out to September. The front month is trading above the 90 strike, so if oil stays sticky at our current levels, I think we're around 92 in the front month now, you actually get paid to hold a position because your long strike will increasingly go in the money if nothing happens to the flat price.

[00:45:29] And by the same token, the 100 strike vol should drop if the market stabilizes, so you get the benefit of carry, steep skew, and if the market stays sticky, you make money on both sides of the spread. So I floundered through the end bit, but if nothing changes, it's a strategy where both strikes should do well, and you're getting paid to have a long options position. That's a sort of attractive hack that you can use in various markets. It's much harder to

[00:45:59] do that in the S&P, where, as I said, the forward curve is relatively uninteresting. In equity derivative markets, anything longer than three or four days is considered long-dated vega. Nowadays, yeah. I want to talk just about your overall commodity trading strategy. We talked about this a little bit, but the ability to lean into the notion that, and you reference this in your book, Crisis Alpha, the

[00:46:27] defensiveness of U.S. treasuries. Perhaps that's an era that will come back, but it is not a consistent risk-off asset in any way like it was, especially in the post-GFC period for a decade, really. And so this need to find trading strategies or investment exposures that are different than things like the S&P, I think it's just more critical than ever.

[00:46:56] So you're doing a couple things. You're giving some of your clients exposure to commodities, but also overlaying a long vol focus, long vol exposure on top of that. Talk to us about what is in that. We talked a little bit about corn and beans. Just tell us a little bit more about how you're thinking about it and how the options side gets implemented. Conceptually, you can think of everything as almost a broken fly on commodities markets

[00:47:26] where at-the-money options tend to be relatively fairly priced. Out-of-the-money options tend to be expensive for two reasons. One is end-user hedging, which I mentioned. The other is dampening of moves by producer hedging, selling of futures and so on. But the moves that do follow through tend to follow through big time, some along the tail as well, the right tail. So that in combination with long futures is providing me with a lot of the long biased exposure. Now if a commodity is cheap

[00:47:56] and vol is cheap, then I'm outright long convexity. But you can think of everything else as a bounded risk trade that makes money all the way up. It focuses on cheap commodities and tries to monetize mispricings in the call skew to pay for carry. That's the basic idea. And it's something that people almost certainly do in individual markets, but to do it on a multi-market level I think is a bit, it's fairly unique.

[00:48:26] So that's the way I think about it. I don't think it's a good idea as a long commodities manager to cut off the right tail. When I initiate a position it might look like that, but I am a trend follower on the way up. So that futures position doesn't go away if the market rallies by a certain amount. I want to keep it on until there's a reversal. The futures piece is providing me the long gamma exposure, the tail risk piece is providing me with the extra kicker if something happens quickly, and the skew

[00:48:56] harvesting is paying for the roll costs. And so this is one of my clients would call it, I think is the You're selling one, maybe it's a 25 delta call, but buying multiples of 10 delta calls, is that the orientation? Roughly, yes. And how far out in terms of expiry? For most markets it's three months, two to three months. For oil I can go further out, it's partly a function of where the liquidity is. You know, if I

[00:49:26] trade anything other than corn or beans, it becomes a bit dicey in the ag space to go out for. So stepping back and just thinking about a portfolio that consists, let's say, of the S&P 500, the global risk benchmark that everybody's got to own some portion of, and then allocating something in the realm of commodities. How do you think about, and then we're going to add your options part

[00:49:55] two, but just commodities exposure. How should people think about how the S&P interacts with the S&P of commodities? An indexed position in a portfolio of commodities. What does that correlation look like between stocks and commodities? Well, it's going to depend on what the driver is. I'll take aim at the 60-40 portfolio first, and hopefully that will connect the dots, allow me to connect the dots. So the 60-40 portfolio is

[00:50:25] vulnerable if rates go up. So bonds go down mechanically, and we go into a kind of an inflationary or a headwind, a financing headwind where the S&P comes under threat. A position in commodities almost naturally offsets that. So if the driver is rising rates, possibly induced by inflation, you're very likely to get hit on the 40. You'll probably get, you'll be vulnerable in the 60, so you need something else.

[00:50:54] And so for me, the combination of long vol as a principle and commodities, which can both be implemented as overlays at low cost, can be attractive. You've said it very well, if you don't believe in bonds in the short to medium term, where can you go? The S&P, most people have to have it. I would hardly begrudge that, but where are you going to get your defense from? You either need to get it from convexity as a matter of principle, or an

[00:51:23] asset class that doesn't behave like the S&P or like bonds. Now, some of the commodities do behave temporarily or from time to time like the S&P, but certain other commodities don't. I don't think cotton bears any relationship to the S&P. I doubt that most of the ags do, or the softs. So you do get some really natural diversification there. And this is something that I'm hardly the one to come up with. This trend followers have been saying this for many years.

[00:51:52] That diversifying across markets can be very attractive, especially into some of the smaller markets, at least relative to financials, where the dynamics are totally different. And if I took from my playbook, I would say that that's really significant now because diversification is becoming increasingly hard to find in a world of volatility suppression in financial assets. So there's a lot of focus and there's a lot of ETFs now that

[00:52:21] try to shape S&P returns. You mentioned the collar. You limit some of your upside, but you sleep at night. We can argue all day long about whether that's efficacious. My own view is the market beats you to it by way of the skew. This trade construction struggles because it's trying to take something from the market that the market doesn't really want to give you. But that's the goal is, okay, you've got the S&P asset and we're going to shape the return and the risk characteristics of it.

[00:52:51] If we were to go to commodities, and this is where I wanted to dive into your options overlay, and we had some version of the spy of commodities, just an indexed exposure of commodities, and now I'm overlaying your option trading in commodities on top of it, talk to us about the risk return attributes that you're looking to change and obviously enhance. How should clients think about what you're doing there in terms of

[00:53:20] starting with a base exposure to commodities but then hopefully benefiting from your options overlay? Well, if I buy a commodity cheap and the client's willing to evaluate performance over a reasonably long horizon, let's say a year, I should only get about 20% of the downside and 60% to 70% of the upside. Now, if a commodity does rally, then the risk becomes more symmetric even in the positions I have because my options deltas go up.

[00:53:50] But the goal is really to have an asymmetric payout profile where on a position by position basis rather than on a day by day basis, the risk is asymmetric and heavily skewed to the upside. That's the way I try and think about it from a very high-level perspective. So let's finish off with AI as it relates to your investment process. These are fabulous tools in our conversation. You mentioned using Claude. How has the emergence of some of these new

[00:54:20] data-oriented tools enhanced your process? And where are you taking some of these things in terms of your own research? The easiest thing to say is hypothesis testing is almost a median now. So I don't really sit in my chair and think, what's the greatest idea that I can come up with that would apply to wheat? Instead, I read stuff. I read the same providers again and again. I try and pick their brains either in person or

[00:54:50] just by reading their material. And then I look for patterns that I can test. And so my job isn't really to be an idea originator when it comes to macro, but rather a sieve where I can look for patterns in the way people think and then try and implement them in my own trading. And having something like Claude with a database, and I'll give a shout out to my friend Felix at Alasso, can be very powerful in terms of very rapid testing and prototyping of models. The other thing it's really good

[00:55:19] for is customization. So you can run a risk analysis on a client's existing portfolio and then do a grid search or some kind of multi-dimensional search for the best option structures relative to what they already have. And so it may be the case that selling the ratio spread is attractive for one client but not for another. So it's not only the intrinsic properties of the hedge, it's also the relevance of the hedge for the client and being able to view everything in an integrated

[00:55:48] platform is very powerful. So for me it's a really powerful backtesting engine, a really powerful way to figure out what the existing research is in a given area, to summarize research papers. This is something that used to take me ages. It would take me a good three or four days to read a hard research paper and see if I could use it. And that's lost time unless you know that it's going to be of some value. So I don't find it that useful for coming up with

[00:56:17] original ideas but better for processing stuff that's already out in the world and seeing if it's relevant to what I do. I would say those are all things I do from the option standpoint. At SCT we actually use machine learning methods directly. I'll leave a teaser on that front. This is something that the guys have been doing. They try to use images for prediction. So they believe that the eyeball, the human eyeball, tends to see patterns in data

[00:56:47] that lead to trades that cause follow-through in one direction or another. And that by directly analyzing the images, you can glean things that you couldn't from looking at tables. Now that may seem a huge leap of faith, but they've gotten some very interesting results doing that and I am part of that, although indirectly part of that process. So yes. Well, as we've talked a lot about commodities and some of the more off-the-run commodities that are just less and less adjacent to the typical

[00:57:16] financial markets that we spend all our time on, things like S&P and rates and credit spreads, I wanted to ask you just lastly about prediction markets and wondering if those are just pure probabilities, Arrow, Dubrow, securities in some ways, they are making their way onto the Bloomberg terminal now and in some ways, of course, a lot of these markets are incredibly thin, but if I were to think about just truly independent

[00:57:45] events, there's a lot of them up there that are really independent. Some of them are quirky and, of course, very subject to manipulation. Is that an area that you're spending any time on? As of now, no. I'm very interested in them as you probably are, but the granularity is not that attractive to me yet. I would need enough liquidity in these markets to even consider running them for clients. All right, we covered a good amount of ground here. We focused on traditional vol

[00:58:14] complexes like the S&P and then got into some of your observations and strategies on out-of-the-money skew on the corn side. I really appreciate you taking the time to be a guest. This was a different one than a lot of the things we talk about typically around monetary policy and things like basic hedging and equity markets. I think this one's a seriously interesting conversation. Thank you for having me. I enjoyed it greatly. Thank you. You've been listening to The Alpha

[00:58:44] Exchange. If you've enjoyed the show, please do tell a friend. Before we leave, I wanted to invite you to drop us some feedback. As we aim to utilize these conversations to contribute to the investment community's understanding of risk, your input is valuable and provides direction on where we should focus. Please email us at feedback at alpha exchange podcast dot com. Thanks again and catch you next time.