Pod Insights Real Vision: Finance & Investing How to Profit from the Coming Commodity Supercycle ft. Hari Krishnan About this episode π₯ Visit https://realvision.com/token2049 for 15% off tickets, only with the code 'REALVISION'.π₯ ππ’ππ‘ Real Vision for FREE https://rvtv.io/3Y4t5Pw. Hari Krishnan, author and hedge-fund manager, joins Ash Bennington to explore the intricate world of virtual warehousing for commodities. They tackle the significant challenges in the natural gas and power markets, with Hari breaking down the technical aspects and offering a deep analysis of the potential investment impacts.π’ This episode is brought to you by TOKEN2049 Singapore. Join over 20,000 attendees for the world's largest crypto event: TOKEN2049 Singapore from 18 to 19 September. Balaji Srinivasan, Solana's Anatoly, Arthur Hayes, and over 250 others will hit the stage, as TOKEN2049 takes over the iconic Marina Bay Sands in Singapore. With over 500 side events during TOKEN2049 Week, Singapore will transform into a crypto hub from 16 to 22 September, capped off by AFTER 2049, and the Formula 1 Grand Prix race weekend. Everyone will be there β this is the one event you can't miss this year. π₯ Visit https://realvision.com/token2049 for 15% off tickets, only with the code 'REALVISION'. About Real Visionβ’:We arm you with the knowledge, the tools, and the network to succeed in your financial journey. Elevate your brand with Real Vision. Connect with us at [email protected] to explore advertising possibilities.π₯ Get ππ₯ππ πππππ¦π¦ to Real Vision https://rvtv.io/3Y4t5Pw Connect with Real Visionβ’ Online:Twitter: https://rvtv.io/twitterInstagram: https://rvtv.io/instagramFacebook: https://rvtv.io/facebookLinkedin: https://rvtv.io/linkedinDisclaimer: https://media.realvision.com/wp/20231004185303/Disclaimer-1.pdf Learn more about your ad choices. Visit podcastchoices.com/adchoices
Transcript 10,167 words Β· 4 speakers
0:00 Speaker 00 Join over 20,000 attendees for the world's largest crypto event, Token 2049 Singapore, on the 18th to the 19th of September. Balaji Srinivasan, Solanas Anatoli, Arthur Hayes, and over 250 others will hit the stage as Token 2049 takes over the iconic Marina Bay Sands in Singapore. With over 500 side events during Token 2049 Week, Singapore will transform into a crypto hub from the 16th to the 22nd of September, capped off by After 2049 and the Formula One Grand Prix race weekend. Everyone will be there. This is the one event you can't miss this year.
0:40 Speaker 00 Visit realvision.com/Token2049 for 15% off tickets only with the code REALVISION. Link in description.
0:58 Speaker 01 Welcome to Real Vision. I'm Ash Bennington. Harry, my old friend, welcome back.
1:04 Speaker 02 It's a pleasure. Thank you, Ash.
1:06 Speaker 01 It's always a pleasure to have you here. I really enjoy these conversations and I'm really excited. I know you have a new idea to discuss with us, something I've been looking forward to talking about on this show for some time now. But Harry, I guess we should give full disclosure. You and I have known each other for some time. We're friends and we've written a book together.
1:24 Speaker 02 All of the above, yes. We are friends. We've known each other for a long time. And you were kind enough to help me write Market Travers. So, yes.
1:35 Speaker 01 Well, you were kind enough to let me piggyback on the process for your idea that you come up with, because you really are one of the most original inventive thinkers. You really do think about things from first causes. One of the reasons why I always enjoy having you on this show to talk about your ideas. Harry, let's set this one up. By the way, I should say also, full disclosure, you and I have talked about this. I've seen the slide deck, but we haven't rehearsed this, which is always fun because we get to discover this together. The audience gets to discover the idea along with me. Big picture, Harry. Frame it up for us. What's the new idea?
2:10 Speaker 02 Well, I'd call it virtual commodity warehousing. Which means, um. If you believe that commodities are in the beginning phases of a super cycle. And yet you don't have access to the physical, you don't have all the logistics that the major companies have in the grain space and the energy space. In metals and so on. How can you gain exposure using listed futures and options? In such a way that you don't pay prohibitive carry costs? And it's what, there are a bunch of stylized facts and I'll rattle off a few.
2:48 Speaker 01 Let me jump in there real quick, just because that's a lot for people to digest, just to get a sense of the picture. Uh, the point that you made, uh, you just made there is such an important one, this idea of without incurring excessive carry costs, that really is the trick, uh, and the magic behind most of what you do is not just finding a way to express the idea, but expressing the idea in a way that isn't overwhelming in terms of cost, so that you don't lose, bleed the, the, uh, potential gains that you have on the trade in the future.
3:18 Speaker 02 Yep, that's correct. So, uh, if I were a farmer and I grew corn. Uh, I'm going to make up some numbers here. It might cost me thirty to forty cents per bushel to store the corn in the farm. And it might cost me about ten cents to take it to the local elevator via a truck. That includes fuel, um, labor costs and so on. However, if corn is at three dollars on average, historically, let's say since the eighties. Uh, it would cost me about one dollar to buy and roll the front month five times a year. And so my break, even if I wanted to buy corn when it's cheap. Let's say at three dollars a bushel is four dollars or over thirty percent above the price.
4:04 Speaker 02 So even though it seems attractive, given that many commodities, especially in agricultural markets are cheap now.
4:12 Speaker 02 It's it can be very expensive to express a long term view by buying and rolling futures. And that's the problem that I've tried to solve.
4:21 Speaker 01 Yeah, very well said. And by the way, implication is that it's actually cheaper to store the physical corn than to pay the financial carry.
4:29 Speaker 02 Yes, absolutely. And you know, farmers have a real option. They can keep the corn in the bin. Or they can sell it on and they have advisors and they have their own experience, which leads them to try and take profits along the way. So as prices go up, they can either sell their inventory or go into the futures markets and sell forward in order to convert their more speculative position in physical to a more carry based.
4:57 Speaker 01 Well, that's financial innovation for you. It costs you more to store a court on paper than it does in reality to actually physically take it to an elevator.
5:05 Speaker 02 Can do, yeah, yeah.
5:08 Speaker 01 Let me ask you one more question before we unpack the thesis, because when you started that, when you made your thesis statement, you began with the word if you said, if you believe that commodities are at the beginning of a super cycle. Is that in fact a belief that you do have yourself?
5:26 Speaker 02 I do, but I never have been in the business of bringing rolling, bringing out a crystal ball and saying X is going to happen in the next month or even in the next year because there are plenty of people who do that for a living. And my job is more to say, if you believe something and you want to express it as efficiently as possible, maybe I have a solution. Now, there are a lot of good reasons to be bullish. I wouldn't say bullish in the short term, but in the longer term, and I'll give you both the arguments. The long term argument has to do with geopolitics, for one thing. Deglobalization and increased emphasis on food security and energy security.
6:06 Speaker 02 The potential dangers of remilitarization globally, which require lots of metals and energies. The green revolution, which requires rewiring large swathes of the electric grid, which takes a lot of copper and so on. Electric cars, again, which require a lot of copper, all sorts of things that require
6:29 Speaker 02 the sorts of things that really exist in the physical world, not shares, not derivatives, but real physical things. And with all the demographic changes with people. Rising to the middle class in Asia, notably China and India, their food habits are changing. So that is putting pressure on many commodities that have been depressed for quite some time.
6:54 Speaker 02 We can throw in the
6:58 Speaker 02 large amounts of liquidity that have flooded the global financial system that often find themselves in resulting in inflationary forces. Now, that may have been suppressed for a long time, but since 2022, things have changed a bit. And, you know, that money, which does lead to headline inflation, often trickles back into the commodities markets. And I can go into that too, if you like.
7:24 Speaker 01 So the back of the envelope, lots of reasons to be bullish here, though you don't take a formal position on that yourself. But the important point, the takeaway for our listeners and our viewers to understand is that if you are bullish, you've constructed this solution as a novel mechanism to express that view in markets.
7:43 Speaker 02 Yeah, all true. And, you know, I never thought I'd come on sounding like Warren Buffett, but I will. Because in commodities, I've found that the biggest margin of safety is when you buy something cheap. So if I buy corn at $3 a bushel instead of $6, I would argue that my margin of safety is far greater. And that's true statistically, as well as from a common sense approach. So statistically, it's true, because if you look at the skewness or the tendency for big surprises, most of the surprises are to the upside when something is cheap, whether it's oats, corn, natural gas, you can go down the list.
8:22 Speaker 02 Whereas when the price rises, as we saw notably in the cocoa markets recently, the risk becomes more two sided. You have big up moves and big down moves. So if cocoa is trading at a 10 year historical low or close to that, all the risk is to the upside, or most of it. Whereas if cocoa is rallied 2, 3, 400%, now if you get into the market, you're facing a lot of downside risk as well as upside. So my view is that if you want to get into the commodity markets, especially certain markets now which are depressed, the risk reward is hugely in your favor if and only if you can manage the carrying costs.
9:06 Speaker 02 Well, that makes perfect sense.
9:09 Speaker 02 I hope so. I mean, uh, my goal in this is not to necessarily ride stuff that is already expensive to the moon. Although I'm not looking to take profits quickly, it's more to get in early where the risk is in my favor or in the client's favor. And to, uh, just manage the, control the timing risk in the position so that even if we're a bit early, the client won't bleed too much along the way. And there are various ways to do that, and I can explain the high level picture and how that's been extended in the past, uh, few months as well in some new research. But the high level picture is that
9:48 Speaker 02 people who really engage in commodities, forget about speculators for a moment, need to hedge on both sides. And I'll start with oil. Oil is the easiest case study. Producers are worried about declining prices. And users are worried about sharply rising prices. So producers tend to sell futures or O T C contracts on the way up because they're in the business of controlling their revenues
10:18 Speaker 02 rather than trying to make lottery ticket type payouts on production. So really what they want to do is sell futures on the way up or sell equivalent swap or forward contracts to control the volatility of their revenue streams. And users, notably airlines, uh, are not in the same position. They're not worried about a $1 rise in the price of crude, namely because they're not buying crude, but because their business involves other factors as well. What they're worried about is a very sharp increase. And if you look at the airlines industry, it's typically, or in recent history, has been in bad financial shape. So what does that mean?
11:00 Speaker 02 Airlines are unlikely to take directs high directional risk when they hedge against upside moves. So they're usually advised to buy things like call options, out of the money call options. Those are insurance contracts that pay out if the price goes up a lot. Or if the price becomes significantly more volatile. Now, what does that mean? It means that there's an asymmetry. There's a force that producers are exerting on price that pushes it down most of the time by taking profits on the way up. And there's an asymmetry in the sense that the end user is more likely to be a buyer of insurance rather than a seller.
11:43 Speaker 02 Which raises the cost of out of the money insurance, probably beyond fair value. So the trick in this game of managing carry is to maintain a long position. Sell those expensive out of the money options and then buy other options around it. It still maintain a long payout profile. While paying for the costs of carrying the position in the futures. Now, in oil, you're actually in the lovely position where the forward curve is in backwardation. Twenty twenty five oil is cheaper than spot W E T I. Twenty twenty six oil is cheaper still. So you actually get paid to hold such a position.
12:26 Speaker 02 But my main point was that there's an asymmetry in the way users of a commodity behave. And manage their financial risk and producers behave. Producers have one to one exposure to the markets they're involved in. Uh, end users, I'll give you another example because it amuses me. Cereal manufacturers, if I buy a box of cereal that has a lot of corn in it, maybe 10 cents of the cost is actually corn and related grains or sugar. The rest is really uh, marketing, manufacturing, all of the other things that go into it. So you have much less direct exposure.
13:08 Speaker 02 But you do have exposure to very severe price rises, which creates that asymmetry. That's what I'm trying to mind. That that difference in behavior between end users and producers of a given commodity.
13:24 Speaker 02 That asymmetry is implicitly what pays the carry cost for keeping a long position. And there's a lot of technical
13:33 Speaker 02 stuff to it that I won't go into in great detail, but I'm always open to questions and comments in the chat.
13:42 Speaker 01 So with that said, shall we dive into the deck?
13:45 Speaker 02 We can. I wanted to go a little off piece, just to keep you honest here.
13:52 Speaker 02 I know, I know it can scare the both of us, but
13:56 Speaker 02 I want to give a little bit of credit to a good friend of mine and three students he had who did a wonderful piece of work in the commodity space that I think is highly illuminating. And I have a little story, little fictional story about it. And please stop me as I go through it, but the professor's name is Stefan Storm. He is a math professor at Worcester Polytechnic Institute in Massachusetts. He and three very smart students did a little exercise.
14:26 Speaker 02 And here was my exercise. I had this hypothetical situation. I know there are lots of doomsdayers out there, so I will indulge them for a moment. Imagine that civilization died out. And all that was left was a series of futures prices.
14:44 Speaker 02 No names. Just prices going up and down. No time scale either. All you saw was this, a big stack. Now, could a machine reasonably distinguish between corn and natural gas? Or
15:02 Speaker 02 copper and wheat or something like that? In other words, is there some ground truth? In the same way that if you gave a machine a stack of images of cats, the machines are pretty good at identifying cats, even if they're in the shade, bigger or smaller, with a different angle or so on. Could they do the same thing with time series? It turns out that they could. And using some fancy methods that I will talk about in another lecture somewhere else.
15:31 Speaker 02 They were able with ninety degree plus percent accuracy to distinguish between storable and non-storable commodity prices purely based on the prices, no labels. And so there is some characteristic movement in different markets. And that lends itself to the storage strategy that I apply across various markets. So metals are highly storable, gold particularly, grains less so, oil even less so, and things like natural gas and power or electricity are non-storable. The beauty of this is that we've developed strategies that cover even the non-storable case, which is a lot to unpack.
16:13 Speaker 02 But I'm happy to go into that too, as you probably know.
16:18 Speaker 03 Hey, everyone. We're going to take a quick pause and hear a word from our partners. We'll be right back.
16:26 Speaker 01 So as soon as your friends research, represent or demonstrate a continuum along a sort of storability axis so that you can see different price behavior depending upon its degree of storability correlates with the factors that they've discovered.
16:42 Speaker 02 Absolutely. And you actually see that in the options markets. That's the cool thing. There isn't just a pie in the sky or a fanciful exercise. In other words, the story that I gave you about oil, which is a storable commodity, where the producer has a bunch of real options to either keep oil in the tanker or in the ground or take it out.
17:06 Speaker 02 For non-storable commodities, that isn't true. So there's a very high premium paid for hedging in those markets, far beyond what you see in the copper markets, which are relatively tame, or the gold markets. And that requires a different set of warehousing strategies to satisfy. And again, I can go into that in more detail, but I just wanted to give the audience the idea that these markets are not all one, but the same techniques apply, given the level of storability of the underlying asset.
17:42 Speaker 02 Where metals are the most durable followed by, as I said, grains and salts followed by energies that require a huge amount of volume, natural gas which requires more volume still and effectively cannot be shipped in its gas form. And electricity where the current has to be flowing at the point where it's required. And you can almost, you know, as someone who's interested in commodities in the audience, you can almost back this out intuitively. It's great that we did it scientifically, but from the idea that if you can't store something, then the amount of supply that's, the risk of a major price shock is far higher than if you can store it.
18:25 Speaker 01 Right.
18:26 Speaker 02 Because you can always bring that extra inventory to market. You cannot store it. A slight demand imbalance can cause a huge movement of price. And you see that in prices. Natural gas is much wilder than copper is or than aluminium is or something like that. Electricity is wilder still. And being able to give clients exposure to all of those things using
18:52 Speaker 02 option strategies plus an awareness of positioning risk, I think is a powerful concept.
19:00 Speaker 01 Really interesting stuff, Harry.
19:02 Speaker 02 I hope so, yeah, yeah.
19:05 Speaker 02 I'm happy to rabbit on about this endlessly, but maybe we could go to a few slides in the deck just to give, to bring things down to earth a bit.
19:13 Speaker 01 Yeah, I think it's always helpful for people to be able to visually see it as well. How long did it take you to begin to apply your friend's research to, you know, actually building a strategy? Obviously, a friend who's a math professor is doing this in a very abstract way. How did you begin to take that information and apply it to an actual strategy where you could put trades on?
19:34 Speaker 02 Well, I had a strategy that, I have a strategy that works cross-sectionally very well across most markets. So it works for corn, beans, soybeans, wheat, copper, silver, gold, WTI oil and Brent crude. But it failed, that same strategy, which has similar parameterization, similar options, strikes and maturities, similar quantities at futures, didn't work at all for natural gas.
20:05 Speaker 02 Dutch gas, German power, French power and so on. And so I knew something that was dynamically or qualitatively different.
20:14 Speaker 02 To use an overly fancy phrase, the econophysics or the nature of price movements in non-storable commodities look different. So I wanted to gain a good understanding of why those structures failed.
20:29 Speaker 02 And how to address it. And the real reason that these structures fail is because when a commodity is non-storable, you have to pay a huge premium
20:41 Speaker 02 to buy it in the forward market when the commodity is cheap. So for example, natural gas typically trades at a huge premium in deferred months, especially winter months to the spot price when gas is cheap. It's very much like the VIX, which is how I kind of vectored into this. The VIX, when things are calm, tends to trade at a big premium in deferred months, further out months, to the spot VIX. That's because it's very hard to replicate, nearly impossible for the average investor. So since there is an imbalance between people who want to hedge using the VIX, the world and their dog is along the S and P 500, their dogs are along the S and P 500. So there's a premium on hedging.
21:27 Speaker 02 And the thing isn't storable, so it's hard for it to be supplied in a physical format, which in the case of the VIX means by buying and owning every out of the money strike on the S and P in the front two months. That tends to create a huge premium, which is why some of the strategies which require buying at the money insurance to sell out of the money insurance don't work. The at the money insurance is just too expensive when the physical asset isn't storable. I know that sounds like a lot, but it actually all comes out in the wash when you just look at the steepness of the forward curve for natural gas or power when supply is abundance in the short term.
22:15 Speaker 01 Harry, one of the reasons why you and I work so well together is because you're the genius who comes up with these ideas and I'm the guy who's never afraid to ask the dumb questions. So let me just ask you this, what a simpler way of saying this be to say this, the less storable a physical commodity is, the more reliant market participants are on financial mechanisms of replicating price moves to either hedge the position or speculate in it, or I guess hedge the position might be the better way to say it. So in other words, if you're reliant for something in your supply chain on a physical commodity, if the commodity is storable, you're able to lock in price action on a forward
22:57 Speaker 01 basis or a future basis at a lower premium because there's physical supply that can come onto the market. If it's non-storable, then the only mechanism that you have of mitigating your exposure risk is essentially through synthetic financial derivatives.
23:12 Speaker 02 Correct, yeah. And the fact that there's no guarantee it can be supplied in sufficient quantity in the future leads to it being very expensive when purchased in the forward markets. That's a crucial thing. Something like copper, I know that we may have a shortage of copper in five years or even two years, but copper not only is storable, but it's also recyclable. So scrap can be put back into a super hot furnace, melted down, and then copper can be reused and very little copper goes to waste. So the supply is not only physically storable, it's also recyclable. Natural gas, once you've used it, it's gone.
23:55 Speaker 02 Electricity, once the current flows, it's flowed. You can't get the electrons back. So it's a very different market where the premium is far higher and where the virtual warehousing skill, skill set is much more severe. Took me a lot longer to make significant progress in that space. Now, you mentioned the idea of a theoretical mathematician doing this. Now, I'd like to give him and his three students, Tora, Kathleen, and Adam, a huge heads up. But in addition to that, I want to say that he's been acutely aware of a lot of the applications. And I was very aware of what he could do.
24:38 Speaker 02 At least in principle. And so I kind of had an idea in my back pocket that I threw their way without giving all the reasons why I wanted to have done. And I think that works very well. Um, it's something I plan to continue to. So.
24:53 Speaker 01 By the way, I should say you are a very humble guy, but you are also a recovering mathematician by training. Um, some very impressive degrees in in mathematics that you.
25:04 Speaker 02 I guess, I guess, yeah, it's a long time ago. Um.
25:11 Speaker 02 So, uh, yeah, that's, that's what I wanted to say about that. But if the slides are up, we can go through a little bit more.
25:16 Speaker 01 Slides are up.
25:18 Speaker 02 And, uh, I could talk about that. So the risk is hugely in your favor. So you can see long periods where the price does nothing, always followed by a big up spike. The up spike sometimes reverses. Uh, the timing is unknown, but the risk is hugely in your favor if you buy commodities. Cheap and cheap for me means nothing more than at a low price relative to history and not trending down too rapidly. Um, you don't want to catch falling knife in these markets. So that's the first chart. We can go to the second one.
25:55 Speaker 02 Which kind of compares that to the futures market. Uh, forget about the VIX for now, because this is a growth for uh, corn. If you look at buying and rolling the futures when corn is cheap, which is the blue line. You can see that even in the tail of a forgotten asset case, you lose money. Why do you lose money? Because when a commodity is cheap in the spot market, uh, the forward curve tends to pay to be punitive. So you have to pay a lot to buy and roll the position. And as I mentioned, it's like the VIX. Uh, a flatlining market, even though it's asymmetrically in your favorite in the spot price setup actually works against you in the forward markets. And that's the problem that we try to solve.
26:40 Speaker 02 Because if you don't time the for the front month futures well, and the commodity is cheap, you just pay too much to break even. So you lose the power of the best.
26:51 Speaker 01 And that's really, that's really the problem right there.
26:54 Speaker 02 Yeah, that's it in a nutshell. Buying stuff that's cheap is great, but the forward market will make you pay for that, for that privilege.
27:02 Speaker 01 Right.
27:02 Speaker 02 How do you get around that? So again, just to repeat, let's attempt to buzz through things a little quickly. The green line is the spot market, spot price for corn. The blue line is the P and L, the profit and loss from buying and rolling front month futures in corn.
27:23 Speaker 01 So, so just, just looking at this chart, what you see during the, the, the period, it, it's that bleed that you see there from, say, let's say, December 12.
27:32 Speaker 02 14 to 20, something like that, 2014 to 20, 2015 to 2020. Yep.
27:40 Speaker 02 Yep.
27:43 Speaker 02 Which is a kind of a flat period for the
27:47 Speaker 02 physical asset, but a bad period for the futures.
27:51 Speaker 01 Flat period for the physical asset, a bad period for the futures.
27:56 Speaker 01 And that's the bleed.
27:58 Speaker 02 That's the bleed. That's the cost of buying high and selling low again and again and again, five times a year, which corresponds to the number of liquid contracts available.
28:12 Speaker 02 I did an estimate that the average level of contango is around 20 cents, three dollars a bushel, which costs you about a buck to do five times a year. Now, some of the ETFs or ETPs are getting around that. I think Tukrium has some small exchange traded products that don't just buy and roll the front month. They do three different months on some weighted, they have some weighted average of three different months, roughly equally weighted, weighted. And then they roll them, which cuts your roll cost a bit, but it doesn't solve the entire problem.
28:46 Speaker 01 Hey, can you explain, Harry, real quick, the differences here between when these markets are in contango and in backwardation? And what that looks like in terms of, in terms of this bleed off that you see in roll yield?
28:59 Speaker 02 Yeah, if corn is at three bucks a bushel in the physical market, at some elevator, let's say in Des Moines, I don't know my towns too well.
29:11 Speaker 02 Then, and we're sitting in August, it may be that November corn is at
29:20 Speaker 02 Three dollars and twenty five cents a bush. Um, I know there's seasonal effects and so on, so don't quote me on this. Which means that uh if I buy in November at three twenty five, I'm going to be selling it as October approaches or into October. And buying the next month, which might be Jan or something in twenty twenty five. And um. That that's going to cost me about twenty five cents to do or twenty cents to do because the three twenty five is going to roll down toward three, assuming that a bush or corn costs the same in the physical market. But the premium will still be retained
30:03 Speaker 02 in the January contract. So I keep buying high, which means buying January and selling low, which means selling November as it gets close to expiration. So I keep locking in that loss in an effort to maintain a long position.
30:20 Speaker 01 And that's the the price of the of the of the forward contango as it rolls down the spot.
30:27 Speaker 02 Yes. So in order to maintain a long position, you need to actually trade periodically. So as you get close to the expiration, you have to sell out of the front month, which is going away. You don't want to take delivery and then reestablish the position um further out.
30:47 Speaker 01 Yeah. And that's just the definition of how you rule a contract.
30:50 Speaker 02 Yep. That's correct.
30:54 Speaker 01 And in the case of backwardation, where you see it rising to spot.
30:59 Speaker 02 Well, that's a glorious setup if the backwardation is sustainable.
31:05 Speaker 02 So in markets like natural gas, you'll you you may have a price spike because there's a supply shortage that's instantaneous.
31:14 Speaker 02 And so the front month skyrockets and in principle, you're paid to hold a deferred month and sell the front.
31:23 Speaker 02 The problem with that is that backwardation is not that sustainable a state for markets such as natural gas or the VIX, which I keep referring back to. So you're not able to keep monetizing that. Heating oil used to be a market that was consistently in backwardation. Crude oil currently is a market that's in backwardation, as I pointed out earlier, which is actually nice because structural backwardation is a setup where you don't need to be great at carrying the position. You just buy the deferred futures and you get paid to wait on the assumption that you think that, in this case, the price of oil is unlikely to go down substantially in the near future.
32:06 Speaker 02 So your breakeven is actually in your favor.
32:10 Speaker 03 Hey, everyone. We're going to take another quick break and hear a word from our partners, and then we'll be right back.
32:18 Speaker 01 Assuming that the price of spot does not collapse dramatically, essentially you have positive rolling on the position in backwardation.
32:26 Speaker 02 Yeah, so if every commodity forward market were in backwardation, you wouldn't need me to come on the show and say, oh, I've come up with this bag of tricks for warehousing cheap commodities. You just go and buy deferred months and it'd be done. But sadly, WTI is a pretty rare case.
32:44 Speaker 02 So, yeah.
32:46 Speaker 01 Talking of your bag of tricks, we've got a slide we can pull up. I believe it's the next slide called "Our Approach," which talks about precisely that bag of tricks.
32:57 Speaker 02 Yeah, it does. I mean, the first bullet point is a little bit, isn't something that's as central to the argument as what I previously thought. But what I wanted to do initially was to set a benchmark for what the Lewis-Dreyfes's, the Cargills, the Chevrons of the world do, because they can do arbitrages left and right, which result in profits that go to their trading desks, hopefully, that sadly are never passed on to the equity owner, because the equity shareholder has more of a bet on some hybrid of the S and P and a corporatized version of what goes on on the trading desk. Then the producer, then these big multinational companies that have the logistics
33:40 Speaker 02 to move physical supply around as well as trade financial contracts. But having said all of that, um, my point is that I thought a good benchmark would be to say, how much does it cost for the real producer to exercise the storage option, which is wait or sell now? Versus what I could achieve in the futures markets. And I think at least in the base case, ignoring some of the fancier techniques that they have in terms of logistics, we can do quite well.
34:15 Speaker 02 Purely financial. Yes, purely financial and purely listed. Again, this is not a sales pitch, but I think there are lots of interesting ideas um baked in this this concept. So, yes, um, well, the way that the the math worked, or at least the engineering worked, was we tried to do what I did in the second leg down, which was to test various combinations of option structures, standard option structures uh in the markets that we trade, find consistent patterns, and then try to explain why those patterns were there, why there were structural misprices.
34:54 Speaker 02 And a lot of progress was made in in that direction uh uh probably a year ago to three months ago. And I can describe a little bit of that in the next few charts. Uh, the cautionary notes are worth mentioning, which is that I'm not going out and telling you that you should go out, even if you have this warehousing technique and buy every depressed commodity. There are other factors involved as well. And in our own strategies, we do use a lot of curated research. Um, and I've been quite reliant on my colleague Mark Roberts to help me in that journey. Uh, so I'd like to give him a big shout out, but um
35:37 Speaker 02 uh, the idea is to buy depressed commodities, to use my warehousing techniques to hold on to them without worrying about carry costs. And then to hope for that upside surprise potential that's embedded in cheap commodities, especially given the state of the world today.
35:58 Speaker 01 I think when we go down to Chart 7, this is just a couple of bullet points here on current market coverage. I guess this is essentially where you're applying the techniques right now.
36:09 Speaker 02 Yes, it's been expanded since this chart was created, but that's a good baseline list. We've added gold and silver. We've added oats. We've added WTI and Brent crude oil, and we are in the process of adding German and French power. So
36:30 Speaker 02 and again, these other markets, these less storable markets are are attackable now based on the increased level of knowledge we have about the best option structures to trade. But the point I do want to make is that this is not a heavily optimized process. As you go to the left of the storability axis toward the most storable commodities, the structures tend to be surprisingly uniform. The same sorts of mispricings exist across markets.
37:03 Speaker 01 All right, and that gets us to the next slide, which is a walk through on some of the bullet points were fiscal corn, storage, and hauling estimates. This is about the fiscal commodity, how easy it is to store, and how easy it is for individuals and firms to take delivery once it is stored.
37:20 Speaker 02 Yeah, this is a bit of self-education. It's not entirely
37:27 Speaker 02 crucial to understand these costs in order to
37:32 Speaker 02 come to grips with the virtual warehousing concept. But for me, it was a lot of learning because I have no experience in farming whatsoever. And I was rightly ridiculed in a previous podcast for wearing a John Deere cap in the podcast, which I found amusing, but I didn't wish to take anyone off. But even basic things like realizing that grain is stored in metal bins instead of silos and things like that were things I didn't know. But the crucial point here is that if you do these back of the envelope
38:05 Speaker 02 cost analyses, what you'll find is that it's actually a little cheaper to store corn.
38:12 Speaker 02 Especially in percentage terms, but also in absolute terms when it's cheap. In absolute terms, let me say, when it's cheap than when it's expensive. And that's in the physical markets. That's because your insurance costs go down. Your implied loan costs go down. And depreciation, things like shrinkage or damage to the grain. Those sorts of estimates are actually expressed as a percentage of price instead of in pure dollar terms. So it costs less to store in dollar terms, three dollar corn than six dollar bushel corn. But the crucial point is that the reverse is true in the forward markets, in the futures markets, unless you have means of dealing with the carry.
38:57 Speaker 02 So you can look at the nice pictures. But the main point is that the physical operator has an edge over the financial virtual operator. The guy who runs a virtual silver mine or copper mine or virtual farm. Because they actually pay less in absolute terms when the physical is cheap. Instead of more, which is what I pay when I buy and roll the futures.
39:25 Speaker 01 So interesting, and it must be interesting. I mean, I guess you could sort of, to analogize from foreign policy, this idea of the billiard ball more model, where the internal price dynamics don't matter. It is interesting to actually walk through and try and understand the internal price dynamics of the physical market and how that disconnects from, as you would say, the virtual market, the financial market.
39:48 Speaker 02 Absolutely, yes, that's correct.
39:52 Speaker 01 It sure would be fun to get on a plane with you and go out and meet some farmers and ask them, what do guys on Wall Street totally misunderstand about the nature of this market? But you get great answers, I'm sure.
40:05 Speaker 02 We do, and a lot of people would be a little bit, and it's within their rights, a little antagonistic to the idea of a virtual miner, a virtual farmer and so on. Because they're dealing with real stuff. I know that they have government support and so on and so forth. But they have real revenues to be concerned about. Which I think is a large reason why they behave in the way they do, which is to take profits perhaps a little bit quickly. Because they have to, you know, they have fixed costs that they have to pay. So they have to sell into the forward markets if the price goes up. And so their goal isn't to capture the peaks of revenue, but to smooth out the revenue stream so they can keep running their business appropriately.
40:45 Speaker 01 Right. Their goal is never to get blown out on the downside. Right. I mean, that's really, it's lived to fight another day and not literally lose the farm.
40:52 Speaker 02 Which is why um they tend to dampen most price moves, but the ones that get away can really rip. Because no one has hedged against them. The end users have hedged too low and too little. Uh, and you know, the feedback loops and the contagion effects that occur based on the interaction between speculators, namely trend followers, and people who have to bail out of the market for whatever reason, namely end users who need to hedge, can cause the occasional outsized positive surprises that seem to to defy all reason for a physical asset.
41:32 Speaker 01 Right.
41:35 Speaker 01 Okay, next slide is expected roll costs. This slide actually has a chart on it. Walk us through what we're seeing here.
41:42 Speaker 02 Yeah, we kind of covered this where I said that it costs about twenty to twenty cents. Uh, how the average contagion is about twenty cents between the second and front bump.
41:54 Speaker 02 For three dollar a bushel corn. If you look at the x-axis, three hundred cents is three dollar bushel corn. If I plug that into the regression equation, forget the math for a moment, but it's about, but don't get bust by the math. It would give you something like five percent of three dollars, which is fifteen cents, plus thirty five cents, which is twenty cents, is the average contango for three dollar bushel corn over the period from two thousand to twenty twenty three.
42:25 Speaker 01 All right, Harry, explain the axes here. So you got price on the x. What's on the y? This is.
42:30 Speaker 02 Uh, the differential between the second month futures and front month futures in cents. So when you roll, you're going to have to roll by selling the front, by buying the back. So if it's a twenty cent differential, you're paying twenty cents to keep your long position going when corn is cheap.
42:52 Speaker 01 Wait, if the if the Y is in cents, that means as a percentage, that's going to decline dramatically as the X axis increases, right? But it's flat.
43:03 Speaker 02 Well, it's noisy, but you can see that there's a big, there's a large clump of dots at 600 cents.
43:13 Speaker 01 Right.
43:13 Speaker 02 Below the below the Y axis. That gives you that downward sloping dependence. It's much less stable as the price goes up. And there are a bunch of reasons for that one, a seasonality, but also just the noisiness of price moves, which which where you lose a little bit. But the crucial point is on the left side, where you see a more solid clump with less noise. You're paying a lot.
43:43 Speaker 02 You might, go ahead.
43:45 Speaker 01 No, I was just gonna say, if you if you if you drop out the outliers, the the clump at the top slides slips down only slightly. That's that's unusual, isn't it? Explain what we're seeing there.
43:56 Speaker 02 I'm not prepared for this question, but what I will say by in some politically fashioned way, I'll say that if you think in percentage terms, even if that line were flat, you're paying a lot more to buy and roll for a fixed dollar investment, potential investment at three bucks a bushel than at six. I agree that there's a lot of noise there.
44:21 Speaker 02 Yes.
44:22 Speaker 01 All right. And the next slide is about fewer about the futures term structure, seasonality.
44:28 Speaker 02 Yeah, there were a couple of things that we found out when we studied futures using kind of an interest rate focused approach. So in rates, everyone, I'm sure people do this in the ag markets too, but it was news to us. In the rates markets, people talk about slope, sorry, level slope and convex.
44:48 Speaker 02 So the level is just where the 10-year note yield is. Slope is maybe the differential between the 10-year and the three-month yield, or the 10-year and the two-year yield, whatever your fancy benchmark is. And convexity is something like the short end plus the long end minus two times something in the middle. So it's how humpy is the yield curve. In other words, how much does the slope vary at the short end versus further out? And that tells you a lot about interest rate markets. So in the same way, here you get a remarkably clean chart.
45:31 Speaker 02 If you graph the, I think what I did was the first month plus the fifth month, minus two times the third month over time. So the fifth month is basically the next year, one 12 months ahead. And what you see is a huge amount of seasonal dependence in the futures curve.
45:56 Speaker 02 And the seasonal dependence is largely based on the timing of the harvest.
46:03 Speaker 02 So
46:07 Speaker 02 how should I put this in, in, in plain English?
46:12 Speaker 02 If I am planting corn in the US, I guess the harvest is around November. Correct me if I'm wrong. Uh, so there is quite a lot of uncertainty about whether the harvest will be a good one or a bad one until maybe June, July.
46:31 Speaker 02 Early June to July, because soil conditions aren't really known. Weather is especially an unknown factor. How hot will it be? How dry will it be? And so a lot of the premium in the harvest month drains out as more is known about the harvest. One could argue that most harvests are good ones. But there's the occasional surprise that causes the November contract, or the December contracts, let me say, which is usually indicative of a new crop to go. Skyrocket higher. So you're basically playing a risk premium game that is embedded in the forward curve as the months go by, which shows that there's a surprising amount
47:10 Speaker 02 of structure in the ag markets that can be mined. And then maybe the next chart, if it goes into, what I think it does, if you go down two charts.
47:22 Speaker 01 So just before we flip, what we're looking at right now, because this is really interesting, corn futures convexity calendar day, you're seeing that flip above the zero line between that's like March and August.
47:34 Speaker 02 Yeah, that's right. That's right. So convexity is basically saying, how does the slope change from the first month to the third month, vis a vis the third month to the fifth month, which is an implied statement about how much premium is baked into the third month relative to the front or relative to the next year. And so my comment about futures holding its value until more is known about the crop is what drives this dynamic.
48:08 Speaker 01 And then you have, it looks like, I guess you could say convexity of the convexity as it slopes up from March to July.
48:16 Speaker 02 Yep.
48:19 Speaker 01 All right. That's too many derivatives for me, Harry. Next slide.
48:22 Speaker 02 Yeah, let's, let's skip this slide because it's quite complex.
48:26 Speaker 01 Okay.
48:28 Speaker 02 Although I guess what I could say is that there, there are more dynamics that show, that are required to understand what the most efficient place to buy the futures might be at. And let me try and explain that. So if I'm sitting and looking over at US corn, and I know that the market is more complex than that because there's corn, significant amount of corn is now grown in Brazil that's cheaply transportable to places like China, whether it's as feed or just as a raw material for various uses. But assuming that there were only Chicago, there were only US corn, sorry.
49:09 Speaker 02 The thing to keep in mind is that, uh, supply is likely to be highest just after the harvest. And it's likely to be lower in the following summer. Because there's no new corn that is being harvested. All the supply is sitting in a bin or in an elevator somewhere. And so the supply is more constrained some months before harvest, especially before the harvest is known. And so what you get is slower decay in the old harvest, which is the July contract for the following year, or for the year following a given year's harvest, than in the new contract where supply is expected to be more abundant. Why do I say this and why should the listener care about this?
49:53 Speaker 02 Uh, well, because that already gives you a mechanism for reducing your cost to carry. If you always were to focus on the July contract in the following year, instead of the new harvest contract, you're likely to pay lower carry costs because you're taking into account the idea that supply will be more constrained
50:17 Speaker 02 with an unknown future harvest in the July old crop contract. So that's not something that we necessarily do in an idiosyncratic way. It's built into the overall modeling, but it's a way to give the listener or the viewer an indication as to how you can cut the costs of carrying positions when you have a bullish view in a given market.
50:43 Speaker 02 Uh, that's the real takeaway. That's not the way that we actually model things per se. But it's a way to understand why there may be various devices you can use to cut the costs of holding on to a position or virtually warehousing, uh, a corn farm or corn production operation, uh, by selecting futures in a superior way to just buying it from a market relevant.
51:13 Speaker 01 Okay, so on the next slide, we talk a little bit about structural imbalances and futures. Walk us through, this is the difference between the longs and the shorts. Walk us through what we're looking at here and why it's significant in the thesis.
51:25 Speaker 02 Now, as others have pointed out, the categories in the commitment of traders reports are a little bit fuzzy. And some producer hedging that occurs, so some farmer hedging isn't listed in the hedging category in the commercial hedging category.
51:45 Speaker 02 So there are various, let me take a quick step back and say that the CFTC, Commodity Futures Trading Corporation, produces a commitment of traders report periodically, I think it's once a week.
51:58 Speaker 01 Yeah.
51:59 Speaker 02 Where they categorize the open interest according to real commercial hedgers, whether they're producers or end users, speculators, swap dealers, and I think there's one other category,
52:15 Speaker 02 which somehow I slipped my mind, but let's just focus on producers for now. If you view the producer or the commercial hedging category, literally what you see in the chart is
52:27 Speaker 02 two stylized, what you see are two stylized facts. One is that short open interest among commercial hedgers in the futures markets almost invariably outweighs long open interest, which suggests that the hedging is taking place by producers who are directly exposed one to one on a revenue basis or an earnings basis to the price of corn.
52:55 Speaker 02 The other thing you'll see is that short interest tends to increase as prices go up. Prices are in the yellow line.
53:03 Speaker 02 What that means is that producers or farmers are doing what they are instructed to do or what they see most fit to do in terms of controlling the volatility of their revenue stream, which is that they sell futures all the way up. They take profits. They try and achieve a better average price using the forward markets as a tool than they would if they just dumped everything at harvest or played a waiting game where they were permaballs on the crop that they just took out of the ground until there was an oversupply in their bin and the next harvest was imminent. So that's a way of saying that if you believe that this is a good characterization of what producers are doing, they tend to hedge
53:48 Speaker 02 all the way up. Now, if that is true, which I think it is, what that should do is it should take most upward price trends and attenuate because there's going to be selling pressure from the real commercial hedging community producers on the way up. And so my belief there is that that results in more modest sized rallies in the underlying, at least in the corn markets. This tends to be true for other grains too, but let's focus on corn for a moment. That does not say anything about the ripe tail because what you will see as well is that in a few of the peaks, such as in early 08, and again,
54:31 Speaker 02 I'm not a corn historian, you can see that hedging activity peaked before the price in yellow peaks in its maximum spike. So what that means is that hedging is done in a disciplined way until the producers basically hedge the entire crop. And maybe speculators are in or maybe some exogenous event has occurred and the price can keep going higher. So this does not say anything about the decreased frequency of ripe tail events. What it suggests is that moderate moves tend to be a little less big. Then they would otherwise have been based on the somewhat systematic or predictable profit-taking nature of producers. So that's the takeaway there, and that leads to various option strategies.
55:16 Speaker 02 It doesn't say don't buy the right tail. In fact, I do buy the right tail in all of these strategies. But I am trying to make money on the lottery ticket to the upside that the end user doesn't seem to care about. And the producer is too busy running his or her business to be concerned about.
55:33 Speaker 02 But it also suggests that a lot of upside moves will be capped.
55:39 Speaker 01 Until they break through and become right tail events.
55:42 Speaker 02 Exactly. So you want to be long the right tail. Sort of the back of the envelope statement is you want to be long the right tail. You want to be long sort of insurance that isn't too far out of the money. And moderate outsized moves, you don't want to be long because that's where the cereal companies or the other end users of corn,
56:06 Speaker 02 ethanol producers and so on, will be hedging. They don't care about a five cent move in corn because they can pass that cost on to the buyer of pack of corn, corn nuggets or whatever they're called, corn flakes.
56:23 Speaker 01 So it suppresses, it suppresses volatility at the center of the distribution, but not on the right tail.
56:29 Speaker 02 And various physicists have already found that in various markets, the distributions are not Gaussian. They tend to have more mass in the middle and fat tails. So the middle, the belly of the distribution is compressed, but the outliers tend to, to be larger than expected.
56:52 Speaker 02 And you see that at least in a stylized fashion from this chart.
56:59 Speaker 01 All right, I think we're getting a little bit tight on time here. And I know that there's a lot more to come. I want you to just give you the opportunity to sum up and maybe hit a couple of key slides before we run out of time. What's most important for people who are listening to this thesis to understand right now?
57:14 Speaker 02 Okay, if you're bullish on commodities, long term, you have a series of options. One is to buy commodity companies in the equity market.
57:29 Speaker 02 Another is to buy and roll futures. A third option, which is basically untenable for most, is to get directly involved in these markets. What I'm offering here is a way to participate in a long-term bull market, buying commodities that are cheap while minimizing carry costs. So you don't have to time your entry points very well. So I think what I'm trying to do is to fill a gap in the market where investors who want to be long commodities in a pure way.
58:02 Speaker 02 I know that there are a lot of arguments to be long commodity companies in the equity market based on the limited level of ownership and so on. But in my view, you're not getting a pure commodities best. So if you want to be bullish, long term, don't want to pay the carry costs, especially for cheap commodities. There are a variety of techniques based on the way that real commercial players hedge that allow you to subsidize the cost of warehousing that position. That's what I call the virtual silvermine or the virtual farm and so on. It's a way to do things in a clean way, perhaps without all of the alpha
58:44 Speaker 02 of running a real infrastructure-heavy business, or certainly without that alpha. But with clean exposure, low carry costs, upside participation, and fairly low downside risk, given that you're getting into positions when they're cheap.
59:06 Speaker 02 That's the virtual warehouse.
59:09 Speaker 01 That's what this is all about. Let me ask you this. For folks who have been watching this and attempting to understand it, understand what it means for them. Do you have a view? You mentioned, obviously, there are these very large commodity producing companies here in the United States and abroad. Do you have a view on that? And you mentioned this idea that it doesn't perfectly mirror the underlying price of commodities. I guess that's because you have factors like, you know, operational factors as well as management factors and those sorts of things. But do you think about that in the equation? Or are you more focused on the pure price action of the derivatives?
59:52 Speaker 02 I don't focus. For the purposes of this discussion, I'm not saying anything about the corporations that are heavily involved in these businesses. And it's partly for the reason that I mentioned, which is that, you know,
1:00:07 Speaker 02 any company in the S and P 500 is a
1:00:11 Speaker 02 is being driven to some degree by the index, even though the index is composed of the components. The index tends to drive the components as well, in some kind of backward, sort of tail wagging the dog fashion.
1:00:26 Speaker 02 So I just believe that even though there is a lot of space for commodity companies to increase
1:00:33 Speaker 02 as a percentage of the S and P or whatever index you tend to look at,
1:00:38 Speaker 02 they're still at the mercy of what the index does and what the global carry trade does and what the global risk on trade does. Things like grains, although other markets are different, are remarkably uncorrelated to that sort of thing.
1:00:54 Speaker 02 Other markets do have a higher correlation, such as copper or oil. And they're involved in this kind of very complicated feedback loop where even with inflation, oil might be a small component of the CPI, maybe 5%, but it explains quite a bit of variability in CPI, disproportionate amounts. And in addition, if people are worried about inflation and inflation goes up.
1:01:20 Speaker 02 That can feed back into the price of commodities as well as people make adjustments as to what the input costs for production will be. And so even those people who are worried about inflation as kind of an abstraction should be concerned about the price action of commodities at the time.
1:01:42 Speaker 01 Very interesting. Harry, as we come to the conclusion of this conversation, final thoughts, key takeaways that you'd like to leave our listeners and our viewers with?
1:01:50 Speaker 02 Well, think about commodities. I know that there have been lots of arguments over the years that they don't go up at the same rates as as things like the US equities do. But I think it is a very good time to be thinking about commodities, to be investing in the ones that are cheap and perhaps to be hanging out and waiting for those who are bearish on the risk asset setup or a material sell-off in risky assets, which will trigger, really, I don't want to oversell it, but bargain basement type prices for economic growth sensitive commodities. And that's exactly what we're looking for. We're looking to be around and active at the time when these things are really depressed,
1:02:34 Speaker 02 you know, the more economically sensitive ones are, so that we can participate with low carrying costs. And the engineering part is the part that I can talk about in great detail. And I feel it's a very fruitful area to for investors to think about going forward, especially if they believe equities are expensive and they believe that inflation will be relatively sticky and that the cleanest expression of inflation is those inputs that actually drive the rising costs of goods and services. So I think I'll leave it at that.
1:03:12 Speaker 01 And certainly, Harry, to put a cap on it, there are many who believe precisely that. We've talked to many here on Real Vision. Great conversation. Say again?
1:03:21 Speaker 02 I'm sure I'm sure you have. That's why I'm letting them give the speech and I can just. Hang out for the ride. So yeah.
1:03:29 Speaker 01 And speak to the engineering, which is really where the genius of this type of thinking is. Harris, thanks so much for joining us.
1:03:37 Speaker 02 It's always a pleasure. Thank you, sir.
1:03:39 Speaker 01 Thanks for watching. Thanks for listening. Have a great afternoon, everybody.
1:03:42 Speaker 00 Join over 20,000 attendees for the world's largest crypto event, Token 2049 Singapore on the 18th to the 19th of September. Balaji Srinivasan, Solanas Anatoli, Arthur Hayes, and over 250 others will hit the stage as Token 2049 takes over the iconic Marina Bay Sands in Singapore. With over 500 side events during Token 2049 Week, Singapore will transform into a crypto hub from the 16th to the 22nd of September, capped off by After 2049 and the Formula One Grand Prix race weekend. Everyone will be there. This is the one event you can't
1:04:22 Speaker 00 miss this year. Visit realvision.com/Token2049 for 15% off tickets only with the code REALVISION. Link in description.
1:04:33 Speaker 01 We hope you enjoyed this episode. At Real Vision, we arm you with expert knowledge, time-efficient tools, and a powerful network to help you succeed on your financial journey. Get a taste of financial freedom with our free offer at realvision.com/free. That's realvision.com/free.
Previous Crypto Underperforms: What's Going On? | REKT Vision Next Global Macro: Has the Fed Abandoned Its Inflation Target? ft. Andreas Steno & Darius Dale Page updated October 18, 2026 Β· transcript and summary are machine-generated