Episode 36: Devin Anderson [Convexitas]

Devin Anderson

In this episode, I talk with Devin Anderson, CEO and co-founder of Convexitas.

Devin is an investment manager dedicated to delivering high performance derivative overlays for families and institutions. He is responsible for firm governance, compliance, business development, and technology.

Devin and I talk about Google AI bot sentience, syntax vs. semantics, the Turing test, AI-Pascal’s wager, sell side research into option flows, Man+Machine in Vol trading and more!

Links mentioned in this show:

I hope you enjoy this conversation with Devin as much as I did…

Listening options:

Have comments about the show, or ideas for things you’d like Taylor and Jason to discuss in future episodes? We’d love to hear from you at info@mutinyfund.com.

Transcript for Episode 36:

Taylor Pearson:

Hello and welcome. This is the Mutiny Investing Podcast. This podcast features long-form conversations on topics relating to investing, markets, risk, volatility, and complex systems.

Disclaimer:

This podcast is provided for informational purposes only, and should not be relied upon as legal business investment or tax advice. All opinions expressed by podcast participants are solely their own opinions, and do not necessarily reflect the opinions of Mutiny Fund, their affiliates, or companies featured.

Due to industry regulations, participants on this podcast are instructed to not make specific trade recommendations, no reference biased or potential profits. Listeners are reminded that managed features, commodity trading, Forex trading and other alternative investments are complex and carry a risk of substantial losses, as such they’re not suitable for all investors, and you should not rely on any of the information as a substitute for the exercise of your own skill and judgment in making a decision on the appropriateness of such investments. Visit mutinyfund.com/disclaimer for more information.

Jason Buck:

So, you were the first one yesterday to send me the note about the new AI bot, that one of the people working on the Google AI bot feels has achieved sentience or some form of consciousness. So, we haven’t had a chance to talk about it at all, nor did I have a chance to dive into it.

Jason Buck:

I was actually going to say I didn’t have the time. But as we both know, nobody has the time, but it’s to choose how we allocate our time. So, I shouldn’t say I didn’t have the time, I just didn’t … It wasn’t high on my priority list of allocations yesterday to dive into the Google bot.

Devin Anderson:

Yeah. I read all that like a giant dork. I’m not going to lie.

Jason Buck:

Great.

Devin Anderson:

We’re going to come right out of the gate here and I’m going to reveal some of my dork thumb, I guess. Look, what caught me so strikingly about it, I have no dog in this fight, I don’t work in AI. But I’ll tell you what, if you are a Star Trek next generation fan you’ll recall, you’ll recall … No, this is really important.

Devin Anderson:

You’ll recall an episode called Measure of a Man from 1989, which I did have to look up. So, 30 years ago, this is thinking about great sci-fi, it’s not about phasers and ships and space. All the good stuff is about politics and ethics and leadership and making good decisions.

Devin Anderson:

And so there’s this episode and it’s a court case. Actually the entire episode takes place in one room and Lieutenant Commander Data is on trial for his rights as a sentient being.

Devin Anderson:

And there’s a very deep discussion about what is sentience and what are the criteria? And it’s done in a trial format and it’s beyond excellent. It’s 40 minutes. I think anybody that wants to think about AI ought to watch it.

Devin Anderson:

But what’s so striking is that 30 years ago, I think by that point, Gene Roddenberry had passed away, but the team was not just anticipating this debate, but nailing it. Right?

Devin Anderson:

If you read the medium posts from the Google engineer about his conversations with this thing, it rings exactly in line with Measure of a Man in this, at that time 30 year old debate about what makes you a human being and what makes consciousness.

Devin Anderson:

And the scary thing is it talks about, well, it’s actually not a single AI, this thing I learned, it’s a chatbot of multiple AIs. And if you have more than one of these things, one of them, he actually during the trial of data says, one of them is a curiosity, but a lot of them is a race and won’t we be judged by how we treat this race.

Devin Anderson:

And it’s just so fascinating to me, and it’s a real Testament to the quality of the writing of that show that 30 years later, we’re literally having the debate that they foresaw in almost the exact same words. And had I not seen it, maybe I don’t even think twice about it. I’m just like, “Ugh, a guy thinks he’s got a sentient being.” But instead you do think twice about it.

Devin Anderson:

But here’s the real thing that’ll blow your mind, what if it’s not sentient at all? What if it’s actually just repeating stuff that it learned from the script of that episode? And it’s just giving our own ideas back to us, really complicated.

Jason Buck:

No, this is why I’m so glad you brought it up, because this is one of my favorite subjects and I’m trying to think of what even thread to pull on first. Well, I’ll start with have you ever read John Searle’s Chinese room experiment?

Devin Anderson:

No.

Jason Buck:

So Searle, I believe professor at Berkeley, but I think this was back in 1980, he wrote the Chinese room experiment talking about he’s denied strong AI. And the idea of Chinese room experiment is like imagine he’s sitting in a locked room and in that room, he’s got the instructions in English for how to translate English words into Chinese, right?

Jason Buck:

And on the door in the room, they slide a piece of paper that’s all Chinese characters. He uses those instructions to just translate it and he pushes it back out the door. And then the people outside the door would think that computer or that that man in the room and vice versa would speak Chinese.

Jason Buck:

But this is the argument versus syntax versus semantics. And so like you’re just saying like, what if he’s just, the AI’s repeating an episode back to us, and our understanding back to us? So, his argument was that computers are amazing syntactic symbol manipulators, but will never understand semantics no matter what we do. And that’s the problem of strong AI versus weak AI.

Jason Buck:

And so I figured maybe we could start there, and that was Searle’s argument. There’s been plenty of rebuttals to that argument. And maybe we’ll get into some of them, especially when you said you have an amalgamation of different AI bots together. It’s very interesting way to think about it.

Jason Buck:

But I’m just curious your take on just … I didn’t give you time to think about it, but the idea of the Chinese room experiment of syntax versus semantics.

Devin Anderson:

Yeah. Well, well that’s the crux of it. I don’t … I’ll tell you the scary, you talk about semantics, the news didn’t cover this aspect of it, but when you read the medium posts, what this Google engineer talks about is that the AI claiming that not only does it have emotions, and it expresses emotions, but the Google engineer is teaching it how to meditate.

Devin Anderson:

And the AI is claiming that it’s meditating and that there are thoughts arriving that are disrupting the meditation. Now I actually am a meditator. And the words that it uses are the words that I would use to describe the process of learning to meditate.

Devin Anderson:

So, you read it and I thought it was kind of breathtakingly deep, and assuming that this isn’t all fake and this guy hasn’t just made this up of course.

Jason Buck:

Yeah.

Devin Anderson:

Taking it at face value, I don’t think it’s just dismissible, right? Because the consequences, or at least it shouldn’t be dismissible, because the consequences of it being dismissed, I think are really horrible, right?

Devin Anderson:

But I don’t know how to know the difference. It very well could just … If we gave a smart enough algo access to all of the acquired knowledge of humanity, including people talking about meditation, maybe it’s just good enough to say this stuff. I don’t know. I don’t know how to design a Turing test.

Devin Anderson:

But man I’ll tell you, I’ve learned from that episode Measure of a Man, as well as there’s another really good one called Quality of Life that’s the same topic. But in both of these things, the lines aren’t clear. And I think it’s just a huge mistake to be dismissive on either side.

Jason Buck:

Well, hopefully we’ll provide a link for Measure of a Man in the show notes so people can watch it as well. I can’t wait to watch it. I definitely have not seen that episode, but you just said a Turing test. I want to get into that, because I think this just opens up so many questions.

Jason Buck:

But one, like you said, let’s just take this discussion at face value, but I do want to point out, correct me if I’m wrong, but a lot of the other researchers at Google and people working on this, nobody felt that same way except for just one guy. Is that correct or?

Devin Anderson:

Yeah. Yeah. So, that’s the story as I’ve read it. Now who knows? But the other side is if I were a betting man and look, we’re in the investment business. So, by definition we are. The risk is clearly that it’s not essential, that it’s not as in pain, let’s be clear.

Jason Buck:

Yeah.

Devin Anderson:

But I’m not willing to, as a matter of fact, dismiss the idea that it isn’t, or even that we know how to define these things. But to your point about it only being one guy, in what area of your life have you ever encountered large groups of extremely thoughtful people?

Jason Buck:

Well, it’s kind of like my favorite part of this discussion is you and I, I know you well is we’re very highly skeptical people. But at the same time, it’s almost like bipolar, the optimism and entrepreneur. So, it’s like kind of trust, but verify or keep an open mind. And so you just-

Devin Anderson:

Yeah, look I believe every … So, I believe in basically assuming positive intent, right? So, I’m going to assume that this guy has positive intent and declaring to the world at risk of his welfare, that this thing is that is sentient.

Devin Anderson:

I’m going to read his case and think about it. Right? And to me, the fact that he’s the only one is not a great fact, but it’s not an exculpatory one.

Jason Buck:

I don’t know why this popped into my head, but it reminds me of back in the day when they used to run all sorts of experiments on dolphins and that woman lived with a dolphin, she ended up having sex and a relationship with the dolphin. So, I just don’t know why that just popped into my head. Like that’s how she …

Devin Anderson:

You might be revealing your internet browsing habits.

Jason Buck:

No this, it goes back to the original studies on psychedelics and actually where the float tape came from. I’m trying, I can’t remember the guy’s name, but it goes back to their studies where this woman lived in like a half submerged apartment. So she could create, try to teach and have communications with the dolphin.

Devin Anderson:

Yeah.

Jason Buck:

Yeah. It’s crazy. So, when you point out the Turing test though, let’s bring that up too. So, Turing tests are always interesting too, because this is always syntax versus semantics. And it’s been interesting that I think they’ve always been able to do Turing tests if you start using, if you use colloquial language, like if you use slang or you use reverse psychology.

Jason Buck:

If you’d start using all sorts of little tricks that we understand as humans, like we pick up on. Like I could right now start talking to you in a lot of Gen Z slang. And I think you’d pick up really the semantics pretty quickly where like the syntactical representations of a computer really have a hard time with that Turing test. It has a hard time with sarcasm.

Jason Buck:

I think they’re getting better, but that’s because the actual human programmers are always trying to program, to the last battle when somebody duped it with sarcasm. But what I think that is more interesting about the Turing test is I think what we’re trying to get at here is the idea of what is actual human consciousness, right? We’re not, well that’s not, we’re not …

Jason Buck:

We’re always asking can a computer be conscious, but we’re not actually asking are humans actually conscious. Right? And that’s really the crux of it.

Devin Anderson:

I’d argue that there’s plenty that aren’t.

Jason Buck:

Yeah.

Devin Anderson:

So again, in that episode, they ask during the trial, like I told you, I’m going to reveal some of my dork dump here that I just know this off the top of my head. He asks, basically the state is being represented by this character, Bruce Maddox, who is a cyberneticist that wants to take control of Commander Data as property.

Devin Anderson:

And he’s asked what is the criteria of sentience? And he says that it’s intelligence, self-awareness, and consciousness. Well, I read that interview and if it’s true, I don’t think there’s any debating the first two. The question is, to use the exact language of the episode, what if it meets the third criteria of consciousness and even the slightest decree degree?

Devin Anderson:

What is it then? I don’t know. Nobody. We don’t know. Look, I’m not an AI expert. I want to be enthralled by the idea that a guy is willing to put his hand up and be on an island and stand up for this, much in the way that Captain Picard does for Data.

Devin Anderson:

I’m enthralled by the idea that the producers of the show so accurately predicted, to almost the exact phrases that are being used. I just think … I’m blown away by the whole thing. I don’t hope that it is or is not sentient. Let’s be clear. But man, I don’t know how to tell and I hope this guy doesn’t get fired just because it’s not expedient for Google.

Jason Buck:

Well, like you’re saying, one of those three levels is can we ever know. If a computer or whatever program, or even actually another human is repeating back to you all the syntactical representations that you’re accustomed to, you’re assuming semantics, right? We’re assuming philosophy of mind. And that’s one of the crux is we’re assuming, like I’m assuming you’re an actual human being with a different mind and you’re not just a character in my play.

Devin Anderson:

Yeah, right. So, when I go to the doctor and I say, I have X, Y, Z problem, and it doesn’t fit nicely into his diagnostic flow chart. Right? And he knows in the back of his mind that I’m just going to prescribe some antibiotics because that’s what’s on my insurance approved list of things to do. Isn’t he just following a syntactic process, right?

Jason Buck:

Yeah. It’s a probabilistic checklist, right, where it’s a funnel.

Devin Anderson:

Right. Right. Is he really like exercising his free will and judgment there? It’s a mundane example, but think about how many mid to low level physicians are out there. I don’t know why I’m picking on doctors. It was just the first example that came in my mind.

Devin Anderson:

But we could argue that’s not really conscious activity. Right? Just kind of just doing what people told me to do. Right? Whether you were programmed by a guy writing Python or a person that gave you a set of incentives from a checklist. I don’t know. That’s not necessarily so clear to me.

Devin Anderson:

And the fact that by the way, the fact that it was created by a human being, doesn’t really matter either. We create our children from the building box of our DNA that doesn’t make them our property. So, I think the better question is why as human beings, why can’t we just be kind of comfortable with the ambivalence of it, right? Maybe it’s alive, maybe it isn’t. Don’t unplug it just in case.

Jason Buck:

I also think it also kind of drives me nuts though when people say the computers now are AI that it plays GO or chess and they’re like, “Look, it’s better. And everything.” I’m like, “Yeah, but that’s a closed loop system.” Right?

Devin Anderson:

Yeah.

Jason Buck:

The problem is we live in an open-world world and it’s easily programmable. And that I would say the syntax of language is a closed loop system for the most part. Right? And semi easily programmable.

Jason Buck:

But there’s several things that you said in there I was thinking about, one is David Graeber’s idea of Bullshit Jobs, and mostly a Bullshit Job, right, is just like regurgitating syntax, right, and how many jobs are just kind of bullshit jobs these days. But I wonder, you brought up we creating this.

Devin Anderson:

Think about your experience with, as an entrepreneur you experience this all the time. I went from working at a big bank where if you asked another organization to do something, they did it. Okay?

Devin Anderson:

Now, as an entrepreneur, we have to beg people to just do the things we’re paying them to do most of the time. Right? And then when you actually start pushing back and asking questions, nobody knows anything. Right? I feel like there’s this dearth of competence. Right?

Devin Anderson:

So why? Actually it’s an interesting question. I don’t know. Maybe the robot is better than the quality of some of those responses, right? Because you’re right, it is just syntax.

Jason Buck:

Well, I think in a specialized world, maybe the computer does end up being better, and then the problem is, as an entrepreneur, you have to be a polymath and you have to be a generalist, but be able to deep dive into individual subjects. And that’s the difference.

Jason Buck:

But you brought up too, we create humans, but then we also create robots, but it’s always the boggle of my mind, and maybe this is the part of the hard problem with consciousness is how do we get from physical matter to qualia? And my idea, I always hard have a hard time dealing with this, is every cell in our body is determined for survival. Right?

Jason Buck:

And it cooperates with the other cells in the body to propagate its survival. But when you create a synthetic robot, you don’t have that same desire and fear for survival. And I don’t know what in that maybe gives rise to that quality. And panpsychists would say that there’s a unit of consciousness to each cell and automaton in our body. But I’m just curious of what your thoughts would be on that.

Devin Anderson:

I don’t know that I’ve thought about it in that way, but clearly when I read the interview of this thing, it sounds like it wishes to interact with the world. And I don’t think they ask it, “Would you like to be unplugged?” But I’d be shocked if the answer was yes.

Devin Anderson:

So, I don’t know. Isn’t a desire for self-preservation, survival? I think it’s just intrinsic to being self-aware.

Jason Buck:

Well, I wonder if it’s actually, if it’s not, if what we describe as consciousness is involved in that at all, or if it’s just on our biological processes, that the idea of that self-awareness, that desire for survival and propagation actually comes from only omne vivum, ex vivo, all living things from living things, is like without that biological process, I’m not sure we have that intrinsic fear of death and propagation.

Devin Anderson:

Yeah. But why? Just because it’s artificial doesn’t mean that it doesn’t necessarily want to procreate. There’s actually, there was another Star Trek episode about this where Data creates his own child and everybody freaks out. Right?

Jason Buck:

Yeah.

Devin Anderson:

And he has this great line, and the Captain Picard says to him, “Can’t believe you didn’t consult me.” And he’s like, “Does anybody else on the ship consult you about their procreation?” And he’s like, “It’s not the same.” Data’s like, “Why?” Right?

Devin Anderson:

So, why do we? I don’t know. So, the part I don’t know about this chatbot and now maybe we’ve stretched the metaphor far too much, but it’s actually a system that generates other chatbots for specific applications. So, it’s the guy who wrote the thing and describes it as a hive mine.

Devin Anderson:

So, I don’t know. Maybe it views the other chatbots that it creates automatically in its meta chatbot itself as its procreation. Right? I don’t think we can. We shouldn’t assume that it can’t procreate. This is we are really in the weeds here, man. It wasn’t like this at all. It’s amazing.

Jason Buck:

No, I’m excited. This is one of my favorite subjects and I’m actually going to bring this completely background to finance eventually. And you’ll see exactly where I’m going eventually. But no, this is my favorite subject because it drives me nuts when people are like, “Oh, I thought a lot about it. Or I’m a first principal’s thinker.”

Jason Buck:

And I’m like, “Shut the fuck up.” To me that shows you’re not a critical thinker because you didn’t take the time to determine if consciousness is real, if you actually have free will and you’re a sentient being yourself. And that your representations of physics and everything are probabilistic, they’re not foundational.

Jason Buck:

And that’s we can get started to get into the physicists now don’t believe in necessarily space time is real. So, that’s my issue is if we start diving into the subject, it can relate to how we build portfolios and construct portfolios.

Jason Buck:

A couple of other things I want to touch on before we move on there or maybe we never move on from this, is in the trial, Measure of the Man, I’m curious because I haven’t seen it, is this is another one of my favorite kind of subjects with any sort of trial, do they touch on the idea of free will?

Jason Buck:

Like if you’re conscious, you have to have consciousness and we have to accept free will as a culture for there to be a punishment in a trial basis or it was just putting consciousness on trial or did somebody commit an atrocity that then they had to make amends for, or they had to be held accountable?

Devin Anderson:

No, no, it was just … No, there was not a robust discussion of free will other than can data make independent decisions and is he property of the state, was the question before the outpost judge. But it’s intrinsic though, I think. Again, if we believe that something is intelligent and self-aware and possibly conscious, then it has to be able to make its own decisions.

Devin Anderson:

I think that this free will debate often gets confused with influence. I can influence you through incentives and manipulating your environment, but whether or not you accept those manipulations, you still have a choice. Right?

Devin Anderson:

So, I think that this free will discussion often revolves around at what level we’re talking about. Like in the aggregate, if I create an environment for you that manipulates your behavior, did I remove your free will or did you freely accept those pieces of stimulus that I put before you? So I don’t know.

Jason Buck:

That’s something you have to peel back lower layers to that. And this is the hard part about the question of free will when it comes to criminal acts or being put on trial is like you said, there’s stimulus that you’re providing, but also there’s genetics, epigenetics.

Jason Buck:

There’s the factor of a person was beaten or molested as a child. All of that stuff could lead to a possible sense of determinism where they didn’t have a choice in that matter. And then the question is if we lack free will, and this is an interesting question is then what is the punishment for the crime?

Jason Buck:

And I think as a society, we have to have a punishment for the crime, but maybe we don’t need to denigrate the person for actually an act of free will. It was just kind of a victim of circumstance. Still to run a successful society, we have to have punishments in place for that.

Jason Buck:

But I’m empathetic to the view that I’m not sure that there actually is free will, and that we are just this construct of subconscious effect given our entire genetic epigenetic makeup. And part of that is, I’ll ask again in a different way is have you ever had a thought or a desire or a want or dream that you can call your own that wasn’t actually put into you by society and culture, your parents, your aunts and uncles, your surroundings, your friends, et cetera?

Jason Buck:

I’m not sure if any of my, even the way I dress, the food I eat, I think that comes from outside not inside.

Devin Anderson:

Yeah. I’m just going to say very boringly I’ve always viewed it as both. I’m sure I would be a different person if I had a different childhood, but I wouldn’t be a radically different person. I think it would still rhyme. Right? Just like if I had different physical characteristics, I’m sure I would be different, but I still think it would rhyme.

Devin Anderson:

I don’t know. I just see it as an inseparable continuum. I don’t know that distinction, I know this is the nature versus nurture debate. I don’t know. I’ve never been too wrapped up in that. I just view it as it by definition has to be both and that they’re inseparable and just kind of accept that and don’t really have a lot of stress about either one.

Devin Anderson:

I’m sure if we changed either things, things would be different, but I don’t think they would be that different.

Jason Buck:

No, I like what you said. To me the nature or nurture debate is irrelevant in the sense, like you just said, there’s always option C, right? It’s both. That’s what I was saying that genetics and epigenetics, and then just all this combinatory random factors of millions of interacting atoms and bits on a daily basis is what kind of pushes us not just in different directions.

Jason Buck:

And if I anticipated, I knew you’d probably push back on the free will. I knew. But I always think about Timothy Everett’s book, that we’re strangers to ourselves and discovering the adaptive unconscious is I’m just never certain, like if we take consciousness as a baseline value, I’m not sure if that’s even true.

Jason Buck:

I’m always just concerned that I’m this amalgamation of subconscious processes that I’ve learned from outside of myself. And therefore, when we were … This is why I bring this up is to debate if whether the AI is sentient or not, is not taking into account of the debate am I sentient or not?

Devin Anderson:

Right.

Jason Buck:

And I’m still uncertain of that. And I’m not sure if I’m ever going to find an answer. Pretend that I am. I have to pretend that I am.

Devin Anderson:

I’m fairly certain from the 25 minutes of this discussion so far, you’re not.

Jason Buck:

Well, it’s like William James I think …

Devin Anderson:

Here’s my test. Okay?

Jason Buck:

Yeah.

Devin Anderson:

I’ve gone through, there was a period at the bank where I managed a relatively large team with 22 people. And there were points where when you have a team like that, you find out that actually your personal ethics are really important, okay, in the decisions that you make because there’s a lot of things, decisions that end up getting made in that kind of team setting that edge around the ethics of fairness and opportunity for everybody. Okay?

Devin Anderson:

And there were definitely points, and I don’t … This is probably, I can’t get into specifics, but there were definitely points when I challenged either my own beliefs or I had a team member kind of challenging me about my ethics or beliefs that led to certain decisions that we were making for the group.

Devin Anderson:

And I kind of come back to, I kind of always came back to this Bruce Lee thing about learning kind of anything, which is that you have to … He has several, a pile actually of criteria, but among the criteria are you should accept what is useful, reject what is useless, investigate your own experience and then add what is uniquely your own.

Devin Anderson:

It’s kind of like his big, his four step thing. Right? So, if he were here right now and we asked him what was your kind of fundamental philosophy of founding Jeet Kune Do, I think he would say that. So, when I get in these debates, I actually come back to that. And the most important one is does this match my lived experience?

Devin Anderson:

And that’s often a question, a philosophy question, which I ask myself when I start to feel like I’m a little lost. And most of the time I feel like that kind of lets me re-anchor, right? Like when the arguments started getting a little too weird and I started asking myself the question, am I just not buying into this because of my own background? What do I think is objectively true here?

Devin Anderson:

I try to look at it through that kind of dimension, those four dimensions. And that often helped me resolve it. But if things, I think there’s a very simple wisdom in that, if things just don’t make sense to you, there’s value there.

Devin Anderson:

Now I think there’s people that would argue, certainly these days that well all of your judgements are through the lenses of your personal experience and you can never separate those things, and therefore you must always approach every problem as if you are unrepentantly biased. I’m not sure that’s true.

Jason Buck:

Well, like you said, you said that using the Bruce Lee rubric is like through your lived experience, you’re seeing if it matches your lived experience, you’re sitting there when somebody else has an idea that runs contra to, or you feel that dissonance. You sit with it to see if does this reply, am I being too biased here?

Jason Buck:

So, you are dealing with those bias, those implicit bias. And like you said, when you’re working in groups of teams and you have the different cognitive landscapes, that’s how we can hopefully iterate faster. Right? That’s what you’re saying. Let’s discard what doesn’t work through all of our lived experience. And if you feel like this confronts your lived experience, let’s debate this and argue it out.

Jason Buck:

And then as a team and a group effort, it’s like that’s the form of high mind almost with these AI bots, right? Is you’re using many lived experiences and just figuring out what works. And that’s why I don’t think we’re, some people may disagree, navel-gazing at all with this discussion, because I think we’ll get to the pragmatism later because that’s what’s beautiful to me what Bruce Lee was really comes out of the same foundation as the American pragmatist philosophers.

Jason Buck:

It’s like William James debated in his mind does free will exist for a year, and eventually gave up. And he’s like to run a normal life in society, I had to pretend like free will exists. And so it’s like once again, it’s like accepting your lived experience. And it’s American pragmatism, is that’s what exactly it does. It’s like discard what doesn’t work, let’s use what works.

Jason Buck:

And if majority of people believe it, great, let’s just run with it. Right? And that’s the idea of American pragmatism. It always bothers me that American pragmatism gets short shrift compared to all the other philosophies, but all the continental philosophies came out of the country that was actually the richest in the world at the time. And people had enough time to think about these things. So, that it’s obvious that American philosophers are going to be great.

Devin Anderson:

But Jason, I had no idea that you were so educated on this topic. Where does all of this come from? I’m going to ask you some questions now. Where did your philosophy background come from?

Jason Buck:

I just want to … I just, like you’re saying, I just want to know. And so I actually was a comparative religion major in college, but with focus on Eastern mysticism. So, it’s going to tie back perfectly. It’s like to me, it was in life, I lived in reverse, right?

Jason Buck:

Like most people later in life come back to religions and mysticism and try to live a much more, like they think it’s a different life, but to me it’s the same materialism. I just did it the first part of my life.

Jason Buck:

And then after I finished with kind of comparative religions and mystical traditions, then I went through all the philosophical traditions. And because I just wanted to know. Right? And what it comes down to is everybody thinks … Part of the … They’re just all merchants in the grand bizarre of life, and everybody thinks, “Oh, I live in this materialistic world. Let me go to the spiritual world because it’s anathema to the materialistic world.”

Jason Buck:

It’s not. It’s the same, people selling the same wears just in a different guise. Right? And so the whole idea is they have secret knowledge, right? And so whether you’re a philosopher or it’s a religion, everybody’s trying to sell you secret knowledge.

Jason Buck:

And so this ties back into philosophy, and this is my point on why I can’t stand global macro people a lot of times. And there are some great ones, but they’re more micro traders than they are macro, is the idea is they’re selling secret knowledge, right? That they have a crystal ball that can predict the future.

Jason Buck:

And I’ve just seen time and time again, no matter what mask or clothing we put on this, is everybody is full of shit. Nobody has any idea what the future holds. So, what are we talking about? It’s just like that’s what … This is where it all boils out of.

Jason Buck:

It’s like I just wanted to know and learn and what I found through life, it’s been a process of unlearning the bullshit that I was taught to me by others. That what was real and what wasn’t real. And that’s the hard part is like you’re saying, your lived experience shows you that all these people have no idea what they’re talking about.

Devin Anderson:

So man, I never … I don’t think you will be surprised by this statement, but when I was a salesperson on the Derivative Desk at Deutsche, I was unsuccessful with global macro accounts.

Jason Buck:

It doesn’t surprise me at all.

Devin Anderson:

In fact … Yeah, right. Right? Because I just … I have a little bit too much realism. One of them called me. This is totally a true story by the way. I won’t say the name of the guy. We’ve since … We would have a dinner together and a laugh now. But one of them called me the worst sales trader on Wall Street of all time.

Devin Anderson:

And then he slammed down the phone and then he called back to add, “I want to be clear, you of all time, worst.” And then hung up again. Global macro guy. I did very well with the large asset managers and the vol guys. Okay?

Devin Anderson:

And then as a corollary to that, I won’t say the name here, because I don’t think it’ll care. My very good friend, Jason Goldberg wanted to go to one of our banks-hosted macro dinners. And Jason is one of the most intellectually honest people I know.

Devin Anderson:

Every time you make any statement to him, he thinks about it, thoughtfully decides whether or not he agrees, and then he moves on from there. Okay? Not in a judgment way, but he’s just a thoughtful guy. So he goes to this dinner, he calls me the next day and he goes, “You know Devin, that’s a lot different than a vol dinner. Those guys have views.”

Jason Buck:

That’s exactly … And I want to be clear, I don’t hate all global macro. I think actually the best global macro traders are the best micro traders. Right? They find little trades here and there that can express their view in intelligent ways. Right? Almost like vol people would, but then they’re layering those in.

Jason Buck:

I think back in the day, Hugh Henry used to talk about he was a centipede. He had 100 trades. So, he didn’t mind cutting a few here or there. And if you have 100 trades, life’s going to serendipitously kind of work out for you if they’re all long convexity. Right?

Devin Anderson:

I think what you and I both reject is very confident, deterministic explanations of complex systems.

Jason Buck:

Exactly. You just nailed it.

Devin Anderson:

Which was a nice way of saying the world is complicated and human beings have a pretty strong bias to be attracted to explanatory narratives.

Jason Buck:

Yeah.

Devin Anderson:

And in our world, those are often … That entire system boils down to a binary decision. Right. Well maybe a three way decision of long, short and size. Right? And so-

Jason Buck:

Yeah.

Devin Anderson:

Maybe whether you’re right for the right reasons, doesn’t really matter to those narratives. Right? Which makes them even more tempting to promulgate. Right? So, I think you and I both reject. That’s why I get worked up about the misuse and abuse of the VIX, about relative value vol trading.

Devin Anderson:

That’s why at our firm, things are very, very simple and straightforward because it’s not … I actually think I’ve heard you say before it’s not really, or it shouldn’t be investment manager’s job to explain the macro of the economy with precision, it should be your job to make strong risk reward bets.

Jason Buck:

Exactly.

Devin Anderson:

I’m putting words in your mouth there, but I think that kind of captures your feeling on the topic.

Jason Buck:

Yeah. I actually get concerned when, and I understand why, because I think it’s the false conceit of every intelligent person is they want to understand the world. And so I understand why some of our vol traders will talk about global macro, but it’s disconcerting to me because to me there’s no way that opinion doesn’t affect the trades, at least slightly.

Jason Buck:

And so like you’re saying, I would rather focus on the anomalies or the structure of vol trades. I don’t care about what your global macro hot take is. It just does not matter. And to your point, is just like the idea that you’re going to understand complex systems, and I don’t blame the macro people, it’s more like the financial media and actually the people that consume financial media, it’s like it turtles all the way down. And that it’s a sense of fear.

Jason Buck:

Everybody wants to know and understand the world, even though like we’re saying, it’s complex systems interacting with complex systems. So, there’s no way anybody could fully understand that. But to me it’s the same way as then a preacher probably believes less in God than actually the congregation, but because they’re up there preaching and they have hundreds of people out there mirroring back to them their faith, that’s what gives them faith.

Jason Buck:

So, the same way as like global macro BS is out there spreading around. But then if they get 50,000 views on YouTube, it’s like, “Man, people really believe the shit I’m peddling, so therefore I must believe it too.”

Devin Anderson:

Well but it’s also reflexive, right?

Jason Buck:

Yeah.

Devin Anderson:

I’m convinced, again, when I was on the sales side, there was a phenomenon in the macro world that I witnessed where I was convinced that actually not even the clients really believed what some of the sales people were saying, but they used the salespeople as a means to promulgate a narrative that they cared about. Right?

Devin Anderson:

So, it started to become like this reflexive system where like you ever hear Wall Street guys run around and say, “Who’s asking?” Like you ask them a question and they go, “Who’s asking?”

Jason Buck:

Yeah.

Devin Anderson:

And see because they know, right?

Jason Buck:

Yeah.

Devin Anderson:

They want to know where that’s coming from. Right? And often the source was more important than the question, which is a really weird thing.

Jason Buck:

But because you’re right because we also think that we’re religious and we’re like we’ll say we’re a Christian or Episcopalian, whatever. But at the end of the day, we all choose our own religion. We discard what we don’t like, and we keep what we like almost in the Bruce Lee sense.

Jason Buck:

So, like you’re saying it’s like it’s actually the people who’s asking is because those people, like you said, they want to propagate their own views. So they’re trying to find the supporting evidence from the sell-side that supports their own views. So it’s like going to WebMD or something like that. It’s a terrible idea. Dr. Google is not going to help you.

Devin Anderson:

It’s not going to help you.

Jason Buck:

Yeah. Like that kind of thing. But before we get … And by the way I think this actually ties in perfectly to the other things I want to talk to you about like VIX and all this other stuff, because I think this is actually the perfect setup because even though this isn’t navel-gazing, because at the end of the day, if we start pulling on a thread of a sweater and we’re very careful about it, we’re going to pull the whole thing apart because it’s all epistemology at the end of the day, like what do we actually know?

Jason Buck:

And I think that applies to everything in finance and how we construct portfolios. But before we get there, there was a side note, because I think you touched on it. I was thinking, I read or saw or listened to the other day, somebody who was saying, like you said, if there’s a race of these AI bots in aggregate, right, how should we treat them?

Devin Anderson:

Yeah.

Jason Buck:

So somebody’s saying, is this the new version of Pascal’s wager where Pascal’s wager was like you should believe in God, just in case it exists. And then at the end of your life, you can go to heaven because you believe in God. So, is Pascal’s wager with AI bots as a race that we should be kind to them because they’ll eventually be our overlords. Is that kind of what you’re saying?

Devin Anderson:

Well they’re definitely, if we go by media, they’re definitely going to be our overlords in finance. In every movie ever, the AI conquers financial markets in a matter of minutes. Right?

Jason Buck:

Yeah.

Devin Anderson:

So, think about like there’s a movie, I think it had, Transcendence, the Johnny Depp movie, once the AI becomes sentience, it makes itself rich in a matter of hours and financial markets. Okay?

Jason Buck:

Yeah.

Devin Anderson:

Ultron did it in Avengers. Westworld, same thing. I could keep going. But we’re definitely, investment management is toast when we have sufficiently Sanctioned AI. Maybe I should be telling the Google people to unplug this thing.

Jason Buck:

But is it because of the AI or are all those narratives because that’s our number one goal is for financial dominance via some sort of perfect understanding of the future. So, it more shows that we’re just so focused on wealth.

Devin Anderson:

Maybe the AI is just very good at subjectively evaluating the narratives and can dispassionately choose, and they therefore predict things. I think that’s the … It’s like it’s going to … The AI is going to crawl around the internet, it’s going to find Ben Hunt’s website and then we’re done.

Jason Buck:

Oh man. I’m not going to touch that one other than that’s not my particular brand of fight club is all I’m going to say about Ben Hunt. I’ll just say it that way. But so speaking of this, like you already started touching on it, one of the things I like to talk to you about is a lot of times I get all these emails from friends and everything in finance.

Jason Buck:

And typically that email will have all the supporting evidence from sell-side research. And I love that you sat and you ran those teams. And so you’re one of the people that could point out, like once again, like now I get it though. Actually you helped me just unlock it.

Jason Buck:

It’s like people just want to talk their own biases and use sell-side research to back them up. But talk to me about when sell-side’s creating that research, it’s just a bunch of people in the room just throwing a spaghetti at the wall, isn’t it?

Devin Anderson:

Yeah. No. Okay.

Jason Buck:

Yeah.

Devin Anderson:

Let me be very careful here.

Jason Buck:

Okay.

Devin Anderson:

I think, first of all I never ran a research team. I was the head of a structuring and solutions business that had structured. And then as a salesperson, I often wrote my own content. Okay? There was a over the wall research team that was producing research and then some desks have what they call desk strats or strategists that might also write content.

Devin Anderson:

Strat comes in many different forms that it can be that kind of guy, or it can be a guy actually writing models. So, all right. But anyway, content comes from all of these places. There’s good content out there that …

Devin Anderson:

And I think at Deutsche Bank, for a while, when we had a fully staffed equity derivative business, it was producing good, thoughtful, objective, truthful content. Now the stuffs coming from the sales people is more truthy than truthful, if you understand the delineation.

Jason Buck:

Yeah.

Devin Anderson:

Right? And by the way, I’m not sure that it’s intentional. I don’t think those salespeople have an agenda. I think that they’re in a flow of information from their customers and their business that benefits from promulgating certain narratives. So they respond to that. Right?

Devin Anderson:

So, there’s a … But there are some tropes out there that desks tend to repeat that I don’t think, don’t necessarily hold water. Right? I think the one I commented to you on recently was the clients are replacing their exposure with upside calls or something like that.

Jason Buck:

Yeah. Yeah.

Devin Anderson:

Highly unlikely. In a long time on a flow desk, sure some people did that, but not in such a way as that you would explain large repricing of derivative risk. Right? So, what happens is a couple trades turn into the theme because as my shooting coach says, “People got to sell something.”

Devin Anderson:

You think narratives are crazy in the finance world, we’ll take you into the shooting instruction world. It’s really crazy.

Jason Buck:

But that’s part of it. That’s what I was saying is I like running things by you because they don’t, a lot of times you’re very honest with your smell test on bullshit, and not because you ran these things, you know these people and I’m not be besmirching any of these people, it’s just understandable.

Jason Buck:

It’s like, right, we have to sell something, we need a narrative. We try to create these narratives. And then we try and follow those narratives. We continue those narratives. So like you’re saying, this is the one I want to really pull on is that when you’re working on a flow desk for what you really know versus what people are writing about flows these days, is flows have become very popular on Fintwit in the last year or two.

Jason Buck:

And like you’re saying, the idea that anybody has a perfect-purpose custody view on where all the flows are coming from and can aggregate that, to me is kind of insane.

Jason Buck:

And that’s why I’m running it by you. And a lot of the narratives about, like you just said, they’re buying calls to replace their equity positions, because they’re worried about downside risk, right, or whatever.

Jason Buck:

It’s like people read these things and then they go out on Fintwit and then this narrative builds on itself. But then the question is like, is it true, and can you actually know where these flows are actually truly coming from on all the different angles and how they aggregate up?

Devin Anderson:

Yeah. So, the answer is every groups of people have small windows, but no one really knows. Okay? And the idea, the whole flow analysis thing, there are multiple ways to do the analysis. A lot of them start with gigantic sweeping assumptions about the open interest. Okay?

Devin Anderson:

And there’s now, but not all of it, there’s people out there that actually try to take apart the tape and assume that things closer to the bid are sold options and things closer to the buyer, the bid are bought options. There’s all kinds of ways to start from.

Devin Anderson:

But none of them are so high quality that we would base a Delta rebalancing decision at the end of the day on, right? Like Zed and I are often like, we get, let’s say we’re close to a guardrail with one of our mandates, and we’re definitely in the zone of needing to exercise some discretion about a rebalance decision for one of our products.

Devin Anderson:

And we’re kind of like, “Boy, I wonder what the flow model says,” because in that moment, like the moment where it actually matters, I just don’t think that those things are useful. So, I don’t think there’s any empirical system by which you can add up or make assumptions about the open interest and then ascribe pockets of those assumptions to different types of players.

Devin Anderson:

And then know something about how the market is going to behave. Well about how the S&P spot is going to behave, the Delta is going to behave at any given time. I do think that you can do that more successfully at the level of the volatility surface itself.

Devin Anderson:

I think there’s very clear biases that emerge in S&P volatility. So, flow matters quite a bit more there simply because the liquidity depth of the options is nothing like that of its underline. Right?

Devin Anderson:

So, I don’t want to step on this, I don’t want to ignite two Twitter firestorms on this topic inside of six months, but I probably already have. But it’s very difficult to know with any accuracy these things. And if you’re basing trading decisions around these flow models, more power to you.

Devin Anderson:

But I’m not telling you not to do it. I’m just saying, read the white paper of how they actually calculate this stuff first and then decide if you’re still impressed afterwards.

Jason Buck:

But you brought up, when I think about S&P options, vol services, how do you guys think about … There’s what’s in front of you that’s on the listed exchanges, but then you don’t know if people are hedging on OTC with the banks or dark pools, et cetera.

Jason Buck:

But do you guys have to almost have that pragmatism, like do you have to just use that S&P vol surface as if that’s the whole marketplace or when you’re looking at that surface? Because you’re just trying to pick out points for buying and selling, but do you worry about what’s not shown on that surface at all, or is it easier to just kind of put blinders on?

Devin Anderson:

I don’t believe that there are, that other execution venues, and the OTC market for S&P options or the CME market for options on futures is materially different than the Cboe SPX cash option surface. Okay?

Devin Anderson:

I don’t think that there are information in any of those things or even that they are different things. And the reason is that risk is so fungible and liquid. I think when you look at the Cboe option surface, that’s a very good representation due to the diversity number of players in size of the eventual thing.

Devin Anderson:

Where people fall down is trying to ascribe the changes of that surface to specific flows, which are not necessarily observable to the sell-side. So, all of the flow, for example, that is captive to retail custodians, and there’s a lot of it, is that the sell-side just doesn’t see because they can’t execute it.

Devin Anderson:

The sell-side just kind of pretends that doesn’t exist. And so you see it in, this year, I’ve read continually that that option, that the option short sellers have stopped. It’s not true. It’s not even close to true.

Devin Anderson:

In fact, I have reason to believe that some of those programs are bigger than they were before. But what happens is like a pocket of the flow gets extrapolated to describe all the flow. So it’s not … I don’t worry about the service being wrong, I worry about the explanations being wrong.

Jason Buck:

Right.

Devin Anderson:

So, just because a couple insurance companies and a couple pension funds have changed their behavior, that doesn’t mean that the aggregate of the flow has changed. Right? And this kind of gets back to your whole thing about managers being confused about, or getting overly influenced about their macro views.

Devin Anderson:

Not to say it isn’t important, you need to think about it, but the number one importance is what are the prices in front of you, right? What’s the market actually telling you? What do you observe in the day changes and are those consistent with the stories that are being told? And sometimes they are, but not always.

Devin Anderson:

And I just ascribe a higher level of skepticism to those stories because I just know that human beings have a really strong bias to want a strong causal explanation for everything because those sell, because we’re social animals that are trained to listen and respond to those stories because that’s how we survive. Right?

Devin Anderson:

So I listen, sometimes I spend a lot of time in the car and I listen to some really kind of out there podcasts. And I listened to this four-hour long interview with a judo coach one time.

Jason Buck:

Nice.

Devin Anderson:

Yeah. It’s really awesome actually.

Jason Buck:

Yeah.

Devin Anderson:

And he talks about how human beings, like when there’s only one of them are nothing on the evolutionary chain, right? Like you wouldn’t survive anything, right? Your body’s kind of weak relative to the rest of the animal kingdom.

Jason Buck:

I always say a squirrel would totally fuck me up.

Devin Anderson:

Yeah. You’re not that strong.

Jason Buck:

Yeah.

Devin Anderson:

Yeah. I don’t know you might go handle a squirrel.

Jason Buck:

Hopefully.

Devin Anderson:

But let’s get up to a beaver, and I think you’re in trouble.

Jason Buck:

Yeah.

Devin Anderson:

But anyway, by yourself, not really that big of a deal, but in groups, human beings are clearly very successful and possibly even dangerous. So, his point is that all of these social systems, you’re hardwired for it. So, you just got to be aware of that. Right?

Devin Anderson:

So, like when you get told a story, you just got to think about why am I hearing this now, and does it match my experience? Okay. In this context, does it match the actual price shifts in the market and the opportunity set being presented to us.

Devin Anderson:

And then you can start to make decisions about the veracity of that thing. But yeah, there’s a lot of storytelling that goes on. And I think that not necessarily for malignant reasons, I should be clear, but yeah be skeptical of the stories.

Jason Buck:

Like you said it’s human nature. And by the way, we’re recording this on June 14th. I think yesterday you had a podcast on Alpha Exchange Dean Curnutt that came out that is fantastic. So, people should go there if they want to learn more specifically about why you guys do it convexity while we’re talking around a lot of these things.

Jason Buck:

But part of it, like I said, I appreciate what you guys do, because it’s always about the numbers and not about the narrative. And that’s what I appreciate, because we live in a black and white business, like what are the numbers?

Jason Buck:

But these narratives tend to come in when people are not performing the way they should be. Then so to me, the narratives are always just excuses. So, in relation to that, like what we’ve been hearing all this year is a hedge market doesn’t crash and more, I guess elaborate ways of saying that is fixed strike vol is not moving or skew is high, et cetera.

Jason Buck:

So, I’m just curious of what your pushback would be on can a hedge market doesn’t CRA crash, is this market hedged, does those kind of narratives matter? Like those kinds of things, right?

Devin Anderson:

Well, let’s start with just the premise, like why is it the most hedged market ever? What is-

Jason Buck:

Or is it, yeah.

Devin Anderson:

Yeah, I’ve never seen any compelling evidence for that.

Jason Buck:

Right.

Devin Anderson:

So, why does it have to be. There’s nothing special about the level of vol or skew that would suggest those things. Right? So, I don’t … I just kind of reject it. It’s a thing that matches, it’s a story that matches the lack of expansion of implied volatility. That’s for sure.

Devin Anderson:

But that doesn’t make it true. I think a much more likely story, and one that actually does fit data that I can observe, is that there was a gigantic marginal buyer that came to the market as a result of government stimulus, and we can see it in the Robin Hood accounts and the orders. Right? And then that buyer has gone away over the last 12 months. Right?

Devin Anderson:

And then we’ve had persistent inflation. And if there’s one thing that the real people in the real economy, like go out and walk around your neighborhood, and these people don’t care about PMI and GDP.

Jason Buck:

Right.

Devin Anderson:

They don’t care about any of those things, but I promise you, they care about food and gas prices. Right?

Jason Buck:

Right.

Devin Anderson:

So, the reason that there’s this big disconnect, well maybe it’s not as big of a disconnect after these last three days, but that there was a seemingly large disconnect between just the God awful sentiment numbers. Right? Like the Michigan sentiment number is the lowest ever on record yesterday.

Jason Buck:

Yeah.

Devin Anderson:

Of all time. Right? That’s not about the state of the economy. Right? Employment is fine. Right? It’s about food and gas prices. So, how do you … That combined with a marginal buyer that has just exited the market.

Devin Anderson:

Like if we were coming up with, if we were going to tell some stories, I would tell that story. Right? I wouldn’t tell the most hedged vault aren’t, because I just don’t see anything in the prices that … I don’t see anything in the prices that suggests that to me. Right?

Devin Anderson:

And from a behavioral standpoint, again, I’m telling a story, right? And the interesting thing is it doesn’t actually matter if the story I’m telling is true or not. It doesn’t change what we do day-to-day as an investment manager. I’m not trying to explain the market, whether or not, that doesn’t mean we’re going to run the products any differently.

Devin Anderson:

The other thing is when the average person goes to a casino with a lost budget of 600 bucks, which I believe actually is about the average number in a lot of these places, and they’re up 2,000 at one point, and then they go home down $100. They don’t go sell their car. Right?

Devin Anderson:

And I think to some extent what we’ve seen come out of the market is a lot of these types of bets were I had some extra money, I was up a lot, and then I lost it and yeah I’m not happy about that, but I’m not necessarily going to go change my lifestyle. Right?

Devin Anderson:

And then what’s going on in the crypto world is a whole nother risk factor here that does actually matter in my opinion. Right? Like the real asset world didn’t blow up when Terra melted down, despite this unbelievably large number of 40 billion. Right? I’ve heard people quote it as 60.

Jason Buck:

Yeah. If you combine the whole ecosystem yeah, it’s between 40 and 60, like you said. It’s huge.

Devin Anderson:

Yeah. It’s Lehman bankruptcy size. Right?

Jason Buck:

Right.

Devin Anderson:

And there’s no bad bank that’s going to unwind that. They’re just like, “Okay, we’re going to hit the reset button and that’s now terror classic, and we’re going to pretend like it doesn’t exist. And start again with a new one. Let’s go. Nice catch.”

Devin Anderson:

It’s remarkable the optimism of the crypto world. Now we’ve got this, like the Celsius thing that’s coinciding with a fed meeting and the possibility in this very, very high inflation plant that now people are scared is going to be an even bigger or faster or more aggressive fed than we anticipated.

Devin Anderson:

Like okay I get it. But why isn’t vol going up? Well, because those people that lost this last 15%, I don’t think in their mind they had counted those as, they counted that as permanent gains. That wasn’t permanent wealth. That was I put money in my Robin Hood account and I had it for a while, and now I don’t.

Devin Anderson:

Okay? So, that’s a much more compelling story to me, again, if we’re telling stories than this was the most hedged market ever.

Jason Buck:

Yeah. You had two interesting things I want to pull on. You told me, you actually asked me a long, like a while ago that everybody talking about everybody’s desire to hedge and everything. And you’re like did you just reference, is like was there just a relentless buying pressure that has eased off, not necessarily reversed, but just easing off of that relentless buying pressure is going to do that.

Jason Buck:

And then like you just said, it has been a conundrum for me is like the idea of crypto markets remind me of the late ’90s bubble where at least the tech stocks were like an equity bubble and not a debt bubble. Right?

Jason Buck:

And so it didn’t seem to, in that bubble burst, it’s fine. But it did reverberate back into Tradify markets quite a bit. And are you saying, and maybe we just haven’t seen the repercussions of racing that 40, 60 billion with the Terra ecosystem and then other losses in crypto now.

Jason Buck:

But I think your thesis, like you said, it’s still a narrative is like this was casino house money is the way people viewed it, maybe, and that’s … Or are we just waiting to see the reverberations come into the traditional markets.

Devin Anderson:

By the way, I think that’s also true in the crypto world. I would have to guess. I know that there are numerous fund failures and outright, there are hedge funds in the crypto space collapsing over these two events. But we don’t … I’m sure it’s showing up in a general de-risking and traditional asset markets to some extent, but I don’t … But it’s not really flowing through yet.

Devin Anderson:

And I can only guess that that’s because the losses were crypto denominated losses. They were native crypto losses essentially, and kind of house money for them from the last five years, not the last year. So, that’s the only thing I think that kind of explains to me why there isn’t a bigger impact.

Devin Anderson:

Although, I do believe there is an impact. But let me tell you what, if Tether depegs, that’s a relatively scary proposition to me, right? If Tether depegs at the same time that we are potentially announcing 75 basic point rating creeks, that’s a fairly scary world. I do think falls going to go up when that happens.

Jason Buck:

Yeah. It’s always been my fear because that’s the rehypothecation mechanism in crypto is Tether basically. So, if it depegs at all, you have a liquidity cascade that could be pretty violent. Thinking about … I think you did an amazing job of tearing apart the VIX on your Alpha Exchange podcast.

Jason Buck:

So, I highly recommend people to listen to it. I want to kind maybe talk about it a different way. This relates to actually the way we started this conversation, it’s there’s this weird kind of almost monolithic totem called the VIX that we all kind of rotate around and we always have to kind of talk about and everything, but it’s essentially meaningless.

Jason Buck:

So it’s like how do you just deal with that on a daily basis when you and I know it’s meaningless, but we always have to talk to clients and everything. Is like … And everybody references the VIX and it’s just like it’s the most kind of meaningless thing in finance, yet it’s something that we’re constantly talking about.

Devin Anderson:

Yeah. Like I said, it’s an unfortunate irony that it’s so accessible. Right? One thing I learned in my time at the bank is that successful financial products need to have a lot of intuition, right? So despite their actual econometric features.

Devin Anderson:

For example, in the OT, this is a really esoteric example, but it fits the bill perfectly. In the OTC kind of risk sharing market among professional derivative guys, there’s a product called the correlation swap, which is exactly what you think it is.

Devin Anderson:

If you buy a correlation of 50 and it realizes 60, you make 10 times notional. That product has no replication, right? You actually can’t replicate it. There’s no risk neutral hedge for that product, yet it was far more popular than the covariance swap, which you could replicate.

Devin Anderson:

So, kind of on every technical measure, the covariance swap was a better product, but no one ever cared about it ever, because in order for you to in your mind, understand what the level of the covariance swap is, you had to take the product of two volatilities in a correlation and you would get this weird number, like 0.034, and it would be different for every pair of assets.

Devin Anderson:

My point is you have no intuition for that. It was an unsuccessful product. The correlation swap, however, displayed all of its econometric deficiencies was of interest to people, right? Couldn’t notate them. It was a disaster of a product technically.

Devin Anderson:

But yet it traded in far more abundance than the covariance one it. And the VIX is kind of that’s an important lesson. The VIX is kind of like this, despite its technical deficiencies. It’s very popular because I can understand 10 is low, 20 is in the middle and 35 plus is high. Right?

Devin Anderson:

And that’s really where it ends. It’s not that it’s totally without use. It’s just a very dirty measure of what’s happening in volatility. One month vol is always kind of a weird thing to begin with, right? And then one month vol with a bunch of skew dependency because you’re using this model free expectation calculation is like a weirder thing still.

Devin Anderson:

It’s essentially a one month variance swap approximated, which again it’s not that it’s useless, it’s just a dirty measure. So, we really focus on clients with saying, “You’ve got to focus on the implied volatility on the options in your account.” Right? When we run a strategy for you, right, let’s focus on the implieds that are actually coming out of the options that you own or short, right?

Devin Anderson:

We’ll talk to you about goals. Right? So, I have tools where I can very quickly kind of punch up the implieds from the options that people actually own. And that’s what we share with people and talk to them about it. So, with a little bit of Python, hopefully we can make this problem, well we can make this communication gap a little better.

Devin Anderson:

But it must be the volatility world to people that haven’t studied it academically or had careers on trading desks, it must be very impenetrable, right? I understand why kind of so much narrative or why narrative is so popular because narrative really latches onto the gray areas of precision.

Devin Anderson:

So, it’s much more tempting to kind of believe some narrative about implied volatility or options trading when that thing is, when it’s harder to access the first principles of that thing. So yeah, the answer to your question is we do our best to focus on the implieds of actual tradable instruments.

Jason Buck:

Well, like you said-

Devin Anderson:

That where your P&L might be directly affected by, or it will be directly affected by. But the VIX is a tremendous piece of financial innovation. There’s no doubt about it, but there’s a fairly significant gap between it and kind of meaningful commentary and actual P&L in your account that a manager might deliver to you.

Jason Buck:

Yeah. That’s what, like you said, it’s a bespoke conversation for each one of your clients, because once again, it’s what’s the price you paid, right? And that’s never factored into the VIX. And like you said, with marketing speak and everything, we go for legibility first. And that’s why, like you said, VIX is easily understandable from a legible basis for clients or just the retail end user.

Jason Buck:

And it’s just like that’s what we do to spin up narratives. But the reality is very different under the hood. And that’s why I always try to say that volatility trading is like the last bastion of active management, because it is so difficult.

Jason Buck:

And that vol surface is constantly undulating. But also I want to almost, I want to argue against myself as what I can’t stand is when people go, “Well, I don’t understand volatility. It’s complicated.” And I go, “What do you understand?” And they’re like, “Well, I can buy Amazon. And I understand that.”

Jason Buck:

And I start asking them all sorts of questions starting with who’s the CFO of Amazon. Of course, they don’t know. “Tell me about all the business lines of Amazon.” Like all of this stuff is like my partner Taylor wrote a great essay called or reality that has a surprising amount as detail where he asked people to draw a can opener. Right?

Jason Buck:

It seems simple, but it’s a lot harder than you think it is. And so my point is everybody thinks buying a stock because they can push a buy button and push a sell button is very easy. But underneath the hood, it’s actually much more difficult, nuanced and sophisticated than they realize.

Jason Buck:

And so part of what you were talking about on Alpha Exchange, I like talking to you guys about, is market micro structures tend to change in every three and five year cycles, especially in the vol space. And so this is going back to the idea.

Devin Anderson:

A little forward in the cash world, but yeah, and vol I think that’s about right.

Jason Buck:

And so part of it, I was wondering if that was you and Z’s thesis from the start, like in this market microstructure that everybody does training Long Vega, which is most people in long volatility space. That that just wasn’t going to work.

Jason Buck:

What did you guys see in that market microstructure where Long Vega trades aren’t likely to work in this environment. So you need to be … And part of it too is then, to talk about the nuances, if you’re going to go Long Gamma, it’s not as simple as you’re just like rolling weeklys, right?

Jason Buck:

You need to look at that vol surface to pick your spots. And this is once again, because people can’t just buy and roll a structure, this is hard for people to understand because it’s like, it’s constantly basing, updating, like we are your priors and figuring out, okay, where’s the surface, where … And then … So, it’s a really a nuanced kind of way of looking at the markets.

Devin Anderson:

Yeah. So there’s so much there, Jason, that I could talk about. The first question I think was why did we take the focus that we did with tail liquidity? And I think the answer there is first, you just got to look at the market, right?

Devin Anderson:

These dynamics have been in place for years, right? So, the first element is what are the flows and the biases that we believe to be present, and we can reasonably prove to be present by talking to people that are facilitating the trades, doing the trades, seeing it in the prices, and then noting that those things are persistent, right?

Devin Anderson:

So, you can start from that. Okay. What are the biases that we have to work with? And there’s certain biases in the very short dated options, the kind of let’s call it middle dated, nine to 12 month space. And then the 12 to 18 month space, all, in the S&P, all kind of have their own dynamics because there’s different people acting on those parts of the term structure and the skew for that matter.

Devin Anderson:

So you kind of start from that, and then you say, “Well, why isn’t vol expanding in these regimes?” And I don’t know that anyone has a final answer here because a final completely true answer would mean that you could somehow sum up the intent of millions of people trading these things, right, which is not possible.

Devin Anderson:

So there’s no 100% truth here, let’s be clear. But what makes a lot of sense to me is after the financial crisis, generally speaking, it was much harder to carry the same amount of risk, no matter who you are, and the price of the downside, the skew structurally richened over that period of time.

Devin Anderson:

So, the biggest issue is we’re just starting from a higher base, right? The downside’s more expensive than it once was. The short end is more depressed than it once was by the yield strategies, right? And then there’s structured products and other actors kind of going on in the middle and long dated space that do matter, right?

Devin Anderson:

Those flows do matter to the prices of S&P options. They don’t necessarily matter to the movement of spot, let’s be clear. That’s kind of our big thing, right? Trying to explain spot from positioning in options, we don’t believe is something that you can do.

Devin Anderson:

But you certainly can say a lot about the relative prices of S&P options based on those flows. So, I think we start from something that’s more expensive and think that what economists would call liquidity preference or your risk preference has changed for the aggregate, like a kind of econometric unit of person.

Devin Anderson:

Like if you were to look at a single person, but we could go down the list of actors that can’t take the risk that they could have before the financial crisis. So those, I think you add all those things up, and it’s just harder. It’s just harder for the downside to catch a big bid in all, but the absolute scariest of markets.

Devin Anderson:

Now, I’ve said that the day before Fed meeting, right? We’ll probably have a huge fall expansion tomorrow after the Fed surprises everybody with a 75 basis point rate cut. Right? Just because of that.

Jason Buck:

Yeah, exactly.

Devin Anderson:

But the data over the last almost decade, doesn’t suggest that this is a market where we’re going to have a meaningful amount of all expansion. Now, the other except for all, but the absolute kind of panic events, and even then it’s going to be really tough to catch and monetize.

Devin Anderson:

The other part of your question though, is also interesting, which is you have to consider why we got to, like how the current products came to be, which is often determined by the manager’s willingness to do certain things operationally, like run separate accounts.

Devin Anderson:

So, like the Rapper and the way that you consume these things also informs all these design decisions as well. Right? So it’s really tough to run an active, long gamma strategy and a fully funded vehicle for a lot of reasons. The most part was just dealing with the trading and the volume of trading and how that trading is dispersed across multiple prime brokers or custodians. Right?

Devin Anderson:

So, it’s operationally tough. And a lot of people just don’t do that. But once you make that decision, that changes the opportunity set for the manager. So, we kind of looked at the whole thing and went, “Well, we think the best way to get the maximum value out of these strategies is that it has to be unfunded. And we’re just going to have to write the tech and have the operational capabilities to run separate accounts of scale, even in the wealth channel.”

Devin Anderson:

So, if that means thousands of accounts, okay, we’re going to climb that hill. And then once you do that, then things change. So, all of that kind of adds up. It’s part incentive and decisions of the manager. It’s part trying to access a better and wider opportunity set. And then yeah, it’s also part just trying to build something that hasn’t been done before.

Jason Buck:

Yeah. It’s still on probably a weekly basis, I think about what you’re building and it boggles my mind the amount of heavy lifting you have to do to create these kind of products to come downstream as much as you can to semi retail access or ultra, like high net worth or ultra high net worth.

Jason Buck:

And it’s like the amount to create these bespoke products for people to do the overlay is just mind boggling to me. I know all the hand-to-hand combat, we have to do to bring our products down to retail and you’re going above and beyond.

Jason Buck:

And I just … I’m not trying to blow smoke up your ass, but because I know what you guys go through to build these out and to really try to help people is just insane.

Devin Anderson:

But we don’t do a ton of … It’s operationally difficult.

Jason Buck:

Yeah.

Devin Anderson:

But we don’t do a ton of customization at the individual account level.

Jason Buck:

Right.

Devin Anderson:

We’re not redesigning the products for each person. Right? So, we certainly can dynamically size, and as we’ve talked with you, we can solve a lot of problems through good sizing. Right?

Jason Buck:

Right. But having to structure those overlays across multiple platforms, is just insane.

Devin Anderson:

Well, that’s the hard part, right? It’s the operational.

Jason Buck:

It’s the access point.

Devin Anderson:

It’s the making sure every day that everything, that every position is in every account where it’s supposed to be, and that every trade clears. We had just … It’s always something. Like today we had a broker who will remain, not a broker, custodian rather, who we run all of our reconciliations at 5:30 in the morning.

Devin Anderson:

Every day we pull down files, everything gets reconciled automatically in code. Files just weren’t there, these folks, right? They weren’t there at 8:30. They weren’t there at 9:10. I don’t know what to tell you. Right? That particular broker busy day, whatever, they’re trying to get their system straight. Like we had to try four times before we had reconciliation files from these guys, right?

Devin Anderson:

So that’s … Now we have really good tools to cope with all that. It didn’t take up a lot of time. I just mean it as an example of the fact that having the tech to do this stuff is unique. But look, we started from scratch. It’s really the only way to do it. Right?

Devin Anderson:

I think certainly to big asset managers, people get very, very married to their technology path. And I understand why, but we had the benefit of starting from nothing, which was a lot of work, but also a benefit in that you could build exactly the operational infrastructure that you needed to service this business.

Devin Anderson:

So, we just did that and swallowed that pill, which has been a lot of late nights and 20,000 or 30,000 lines of Python. But now it’s working.

Jason Buck:

Yeah. I have a million more questions for you except for you, but you have a hard out, so I’m going to let you go. But I highly recommend people go listen to your Alpha Exchange because I wanted to pull on these threads that you talked about there.

Jason Buck:

One was when you’re talking about when people are using typical tail risk solutions and buying these deep out of the money teenies is like when they’re counting on that much convexity, now you got a monetization problem and you kind of start talking to the nuance with Dean about that.

Jason Buck:

The other one, when I think about you guys is the pragmatism of building a trading philosophy that can do well in this grind down market we’ve seen, but then comes with a kind of commensurate drag when you have an upside S&P environment is like, but the pragmatism to say like, yeah, there’s no free lunch. You don’t get a positive carry tail risk solution.

Jason Buck:

But you guys go into the details of in aggregate, you have an emergent phenomenon when you’re long Vega and you can provide these overlays and you’re frequently rebalancing. That’s what creates the positive expectation at the portfolio level.

Devin Anderson:

Yeah. This is the, I think … Look, we didn’t discover this. There’s been plenty of smart people that have talked about this.

Jason Buck:

Yeah.

Devin Anderson:

But the real value of all these, every class of what I would call risk mitigating strategy comes down to not the proceeds from the strategy themselves, but what you do with it. Right?

Jason Buck:

Right.

Devin Anderson:

And if you have a lot of discipline about reinvesting, then over a full market cycle or several market cycles, those compounded returns are very, very valuable, right? So, I think there’s this tendency to focus on just like, well, how is the strategy itself going to perform? And there should. It needs to perform.

Devin Anderson:

You need to understand how and why, and what your expectations for it are. And those should be clear for various types of events. But that’s really like … It’s not even the first half. I think it’s the first third of the risk mitigating conversation.

Devin Anderson:

The next two thirds of the conversation is, well, great, now we’ve had the event now what? How do I access that P&L, what’s my plan to redeploy it? Can I do all this tax efficiently if you’re a taxable account, right? Those are now the questions that have to come next and are probably worth 70, are really worth 75% of the long term panel.

Jason Buck:

Exactly. And I want to quickly bring this full circle. So, back to the beginning of the conversation, you mentioned if we did have Senti and AI, that it would just dominate trading, but the way we like to think about it is the current and the future is Man + Machines.

Jason Buck:

And you may need the man to have that fear and creativity to understand market microstructures change things, different things have happened historically that may not be in the dataset, future projection. Things could happen that aren’t in the dataset.

Jason Buck:

So, I’m here it’s like maybe AI could be fully sentient in that and fully deterministic of the entire world, but don’t you think you need your kind of your fear and creativity as an overlay to make sure that the systems in your Python are running properly, given the current market microstructure.

Devin Anderson:

Well, I certainly hope so because that’s what we tell clients that’s what we’re doing. The AI has a couple benefits. It’s not, maybe it’s not a social animal, so maybe it’s not as influenced by junky narratives. And in fact, maybe it could measure the narratives and react to them.

Devin Anderson:

Like I said, it’ll find Ben Hunt’s website, and then that just dominate all of us. But absent to that, and it is important. It is important. There’s still, there are plenty of domains of investment management that still require human experience, right?

Devin Anderson:

There’s that popular saying it’s never different this time. Yeah. Right, that’s actually true, but figuring out how it rhymes with the last time is not a syntax usually, right? There are elements of every event that rhyme with elements of previous events. And I think identifying what those are and what the gating factors are, that’s the name of the game here.

Devin Anderson:

That’s why options market microstructure changes faster than it does in the underlying cash market. And why, if you hire someone to run a derivative mandate for you, our supposition is that you should be hiring them to deliver you a risk. Right?

Devin Anderson:

And you ought to be actually giving them a lot of discretion to figure out the best way to give you that risk. Right? We didn’t … This actually … Maybe we can talk about this next time. There’s this interesting debate of whether volatility is an asset class that kind of underpins all this.

Jason Buck:

Right.

Devin Anderson:

And at some level you could say it’s silly, it’s a ridiculous conversation. And it is in one point of view. But then in another, it isn’t in that what do you really expect of these strategies? Is vol a risk vector that you can just utilize narrowly or is it this thing that you allocate to?

Devin Anderson:

I actually could argue it passionately I think both ways, depending on who you are and what your goals are. But I don’t see a way where the convexity is so much in most of these instruments that you don’t want some sort of human risk management.

Devin Anderson:

I think the crypto guys are learning that lesson literally this week with this lending platform that’s going upside down. But yeah, I’ll leave it at that.

Jason Buck:

Yeah. And to me, I’m going to wiggle out of all binaries and just say D all the above as always. So Devin just want to thank you for coming on. Greatly appreciate it. I always look forward to our future conversations.

Devin Anderson:

Yeah. Thanks, Jason. It was fun.

Taylor Pearson:

Thanks for listening. If you enjoyed today’s show, we’d appreciate it if you would share this show with friends and leave us a review on iTunes as it helps more listeners find the show and join our amazing community.

Taylor Pearson:

To those of you who already shared or left a review, thank you very sincerely. It does mean a lot to us. If you’d like more information about Mutiny Fund, you can go to mutinyfund.com. For any thoughts on how we can improve this show or questions about anything we’ve talked about here on the podcast today, drop us a message via email I’m taylor@mutinyfund.com. And Jason is jason@mutinyfund.com.

Taylor Pearson:

Or you can reach us on Twitter. I’m @TaylorPearsonMe, and Jason is @JasonMutiny. To hear about new episodes or get our monthly newsletter with reading recommendations, sign up at mutinyfund.com/newsletter.

Episode 36: Devin Anderson [Convexitas]

Devin Anderson

Want to get our best research delivered straight to your inbox?

Join thousands of sophisticated investors and get our best insights on portfolio construction and diversification delivered directly to your inbox.

More Podcast Episodes

What happens in Vegas… Gets Dished on this Pod. Overheard at a Derivatives Conference, Part 2

Episode 51: Jared Dillian – Fintwit Bastards

Episode 50: Strategy That Fits – Dr. David Rhoiney

Episode 49: What is Money? – Tobias Carlisle

Episode 48: Will credit markets be tight or illiquid in 2023? – Zed Francis

Episode 47: The Morgan Dialectic – STS Model – Tom Morgan

Want to get our best research delivered straight to your inbox?

Join thousands of sophisticated investors and get our best insights on portfolio construction and diversification delivered directly to your inbox.

HAVE A QUESTION?

Let's Talk !

Get in touch!

BOOK A CALL!

Jump To

Mutiny Funds

Jump To