#45 - David Dredge: Risk Management (Part 2)

33:44But in everybody's own investment path, I guarantee they get their one percentile outcome. They must, right? So going back to the race car analogy where there's a windy road and the common strategy is drive only at 60 % speed, which as you, I think, very thoughtfully pointed out, when you do that, it's inefficient because you're going too slow during the good times and you're going too fast during the curves. And so you're likely to crash and underperform over time. So what about this other way of thinking about it, where you have a super diversified portfolio? So a 60-40 portfolio is not that well diversified.

34:27Basically, all your risk is in the stock market, which is a very volatile market. What if you, instead of that road, you travel on a road that maybe isn't as curvy because you have a much more diversified portfolio. Maybe you're invested across various markets. You're invested across managers who have a track record of being low correlated to traditional markets, even during periods of stress. Maybe you have other components of your portfolio, maybe private investments where there's less efficiency, more alpha potential, and you put all that together. And one way to conceptualize that is the road you're traveling on is less windy than the road where the portfolio is less diversified.

35:06So on a road like that, is there less need for good breaks or do you think about it differently? Because even in that portfolio, while you've done, let's argue, you've done a better job of diversifying than a less well diversified portfolio, that portfolio is still relying on assumptions about correlation, assumptions about historical correlation, historical behavior between asset classes or how managers have behaved in other past times. If there's nobody in there that is explicitly asymmetrically negatively correlated, you're still going to run into the same problem. You may not crash, but you're going to still have this concave performance dynamic, this volatility drag on your portfolio.

35:51You're still going to benefit in that portfolio by adding explicit asymmetric negatively correlating protection and replacing some of your other diversifiers that are sharp ratio maximizers where they're not capturing upside and not providing nearly as much downside protection. So the benefit of that asymmetry to allow you to participate to just eke ever more capital into your goal scores, your accelerators is going to make a difference because that's going to still better capture the upside down side magnitude of the largest divergences. And that's going to improve your compounding path. So, yes, the more effectively you diversify, the less risk you have.

36:49But even then, that diversification to the extent it's relying on historical correlations or assumptions about historical behavior, you still need something that cuts the tail. and you need to fund that by taking money away from what I would call defensive midfielders, right? Guys that play really well in the middle of the field, but not so well in the goal scoring boxes and get more goal scorers on and a better goalkeeper on. It's still going to make a difference. So last time we talked about your approach of seeking cheaper insurance by focusing on the second or third order impact from periods of crisis.

37:28So this introduces a concept we didn't talk about last time, which is basis risk with indirect hedges, as opposed to paying more for direct hedges. Would you talk about how you think about that risk? Yeah. So a lot of people will be concerned because people, investors, investment managers, wealth managers, tend to measure their risk in a certain given way. So they might, as I think I may have said last time, they might measure their risk as, oh, we think our risk is S &P beta. Now, even though they're running a global multi-asset portfolio, they've simplified their risk measure down to S &P beta.

38:09And I will tell them all the time, well, your risk is correlation. The whole assumption you're making to measure your S &P beta is an assumption around the historical correlation of all the components of your portfolio to the S &P. And it's that assumption that's going to be the problem, not whether the S &P goes up and down. As you were describing, a well-diversified portfolio. Because the portfolio becomes less diversified when you need the diversification the most. Correct, correct. The assumption you're making to say that this, you know, whatever,$10 billion portfolio has 5 billion of S &P beta.

38:46Guess what? It's going to have more than that exactly when it matters because the correlation you were relying on to reduce the gross risk of 10 billion down to your measure of debt risk is an assumption about correlation. So we tell everybody your risk is correlation. And if you went and looked at the performance of most any market in any significant drawdown, that drawdown aligned with some sort of correlation event the world over, as we were discussing earlier. It's when that correlation changes that undercapitalized risks get forced to be liquidated. And that's literally what creates the skewed return dynamic of markets.

39:29And as we discussed last time, that's heavily driven by the use of leverage. And in a sense, diversification is a leverage tool, right? If I say, well, I've got enough capital to run 5 billion net risk, but I can create that 5 billion net risk through diversification and run 10 billion gross. That's effectively the measure of your risk is de facto leverage. And when it proves wrong and correlation overshoots your assumptions, now you've got to be liquidating. And it's the fact that that's happening throughout the market that causes it. So many people are concerned because they want to match the risk they measure as opposed to match the risk they have.

40:11And so that basis risk you're talking about is they might come and look at us and say, well, our risk is S &P beta or emerging markets or whatever they want to define as the risk. We need to edge that. And we're saying, well, your big risk, your systemic risk is correlation. Now, there's every logic to go and hedge idiosyncratic or what we might call systematic risk. And you might want to buy puts on your actual S &P risk or buy CDS protection on your emerging market bond portfolio. But your systemisk risk, where something happens that you think is unrelated to you and it spreads throughout the whole forest, is best hedged by the things that are most exposed to the spreading of the fire.

40:54And you do that by finding the most attractive price because that price, as you said earlier, inversely relates to the amount of risk. and you will find that over time things seemingly unrelated to the event had much bigger payoffs in the correlated shock and trying to get people to understand because people you know the biggest and biggest of institutions don't want to measure their correlation risk because their whole premise of their sharp world modern portfolio theory portfolio construction methodology is an assumption around stable correlation. And so if you start measuring correlation risk, people's eyes can get opened that, yeah.

41:42So again, you can imagine 60-40 portfolios that made assumptions about the historical correlation benefit of those bonds. Their mismeasurement of that risk in 2022 meant that their losses were much larger than their risk measures that said they could be, which inevitably contributed to the joint bond and equity sell-off as people needed to de-risk because they were taking more risk than their risk methodology claimed. And that nature of that risk was a change in the correlation that they assumed would stay constant in their risk methodology. And so if the true risk is that your main assumption that correlations are stable through time diverges and correlations go up right when you need that diversification.

42:34Then in terms of portfolio construction, you can go out and buy hedges or insurance against that risk. And you can find it cheaply because when it's the cheapest is probably when it's the most valuable. Correct. What we do, and it's hard to do, is all day, every day, all we do is track volatility supply, work with our counterparties, trying to find attractively priced, efficient, ever longer duration convexity. Any market, any asset class anywhere. And it's very difficult for somebody to do that. It's not a full-time job. A, the complexity of some of the products in the markets make it difficult.

43:18The inversion of the credit relationship with the counterparties where you're doing it in bilateral OTC markets require a credit construct that most people don't have in place. The efficiency of doing it on easy to access exchange traded markets, we discussed last time, makes VIX generally egregiously expensive and so somewhat ineffective. Now, having said that, there's a number of my friends and competitors that are in the tail risk or long volatility business that use S &P and VIX because it's a very efficient market, but they need to find ways to cope with the cost inefficiency in that. So they tend to be much more active trading.

43:58They tend to use algorithms. They might use RV strategies. It's just egregiously expensive. My advice, to the extent it's possible for anyone who's listening, is find any and every reputable tail risk long ball guy you can find and start talking to them, start building a relationship with them. Because the beauty is there's no constraints in how many goalkeepers you can have on your team. And just like investing, just like you diversify your investments, exactly to your previous question about basis risk, the spread of that fire is very path dependent. And there's going to be events and occurrences where one style of tail risk strategy might, not mine, definitely be more effective than another style.

44:49But in another occurrence, the other style will be more effective. So pairing tail risk or long vault type managers together that go about it in different ways is a very, very good strategy. And obviously the bigger institutions all do that, which is why we all know each other, because we end up working together on stuff. And while we're out around the world doing things multi-asset class and OTC markets, we work all the time with people that are doing things in S &P and VIX or guys that are specifically doing things in Europe, et cetera. And in interest rates exclusively. But just the conversation, getting out and talking to your clients, talking to you so that they know you're talking to me and other people that do this so that everybody's starting to get an idea of how to construct a portfolio that benefits from the divergences from the expectation, that benefits from better positive correlation at good times and better negative correlation at bad times.

45:53Not an assumption around steady correlation that gives you the exact opposite, where you've got too little correlation at good times and too much correlation at bad times, which I'm sure you've seeing over your careers? I recall the first question I asked you in the last episode is this idea that the vast majority of investors, at least in my experience, focus on returns. Returns are things you can see, yet so little attention is paid to risk. Risk is harder to see. It's harder to understand. The concepts are more difficult to grasp. And what I'm hearing you say is try to think about that more.

46:29And the more you think about it, the more you recognize the impact it has on your portfolio and ways to mitigate it, the better off you are over time. Yeah. And again, history is a terrible measure because history only tells you what did happen. So I tell you, I always use the simple example, two guys climbing up a cliff face. One guy has a rope, the other guy doesn't. They both in a perfect sunny day, dry rocks, no wind. They both climb up the same time and get to the top at the same time. And if you measured it, did they have the same risk when the financial markets they'd say they did have the same risk but the guy with the rope had far less risk the whole time and the next time they're climbing and it starts raining or the wind blows the guy with the rope's got an enormous benefit because the other guy is now like i'm not climbing today back to my shipping example the guy with insurance is shipping every day no matter how bad the weather is and so he's getting the highest fees because the guys who said, I don't need insurance, I'll just hire a good economist to forecast the weather for me.

47:37He's sitting in the dock half the time because the guy said, well, I think the weather's going to be bad. And so this understanding of what could have been and the opportunity it gives you by having that rope to climb every day, no matter what the weather conditions are, not fall off the thing because of a surprise gust of wind. and in theory, maybe climb faster because you've got a rope than the guy who needs to be a little bit more cautious because he doesn't got one. And to simply measure them on the one day when they both climbed the same speed and said, well, same risk is wrong. Yeah. And it's hard to see that the example that you just described is easy to see because we know how much safer it is to climb with a rope than without a rope.

48:20But in the markets, it's not as obvious. And so you could have two strategies that travel through time and they have the same return and one took half the risk, you don't see the risk unless something bad happened. And I think a simple way to conceptualize that is I always go back to the roulette wheel. Let's say there's a hundred spaces and you have two wheels and one of them has 90 winning spaces and the other one has 10 winning spaces and 90 losing spaces. And you have full visibility and you only get one spin Which one do you want to spin? You want to spin the one that has 90 % odds of success.

48:58Yet you could actually spin those and you only get one turn. And the one with the 10 % odds of winning wins. And the one with the 90 % odds of winning, let's say, loses. You could look at that and say the 10 % one was the better wheel. Yeah. Right? And obviously it was not. And that comes back to numerator, right? So easy to measure the numerator. That's what happened. But the denominator, people didn't know the difference between the wheels. They don't know. that's right say well that one so that i'd rather go play that wheel which is nonsense and it's the same thing and this is why you know volatility is a risk measure sharp ratio is terrible terrible risk measure because it's exactly what's exposing me to that far better risk measure is downside volatility or drawdown negative beta what was your beta in negative markets something that shows that lets your eyeball start to see is there superior convexity in that strategy versus this strategy.

49:54And very few people take the time to look through and look at that because as you say, we talked about last time, they forgot that you need to equalize the denominator before you compare the numerators. And by race cards, at the end of the race, we all know the guy with the best brakes won. How do you attribute the value of the brakes? You don't, you just say how fast did he drove? But the reason he did that is because he had the brakes. And it seems like a lot of this misunderstanding can persist for several reasons. So this is kind of summarizing a lot of what we talked about. Number one, you have a limited data set.

50:25And most people don't realize that it's limited because they see a lot of data. So that's issue number one. Number two, there's an effort to try to make a science out of this. And it's not scientific. Not only do you have limited data, but there's a lot of variables that are just difficult to predict. Number three, you have this notion of alternate histories, all the things that could have easily happened, but didn't. and you only see what actually happened, not all the things that could have easily happened that did not happen. And so you put all that together and you could see how we could continue on the path of this conventional view of risk and how to manage portfolios, even though it's obviously faulty.

51:06Yeah, I mean, a great, I make fun of his, you know, having read some of my stuff. Sometimes I poke a little bit of fun at central bankers and both Hugh Pill, the chief economist at the Bank of England and his boss, the governor, Andrew Bailey, have both made comments going back to the period of transitory inflation or high inflation and said, well, you know, our models just didn't work. And it turns out our linear models, when we moved away from our target and it became nonlinear, our models didn't work. They're not changing the models. They're just saying, well, when it moves, our model doesn't work.

51:45And that's a great example of sharp work, right? That distribution for the longest time looks normal. But in reality, the whole time it was a power law distribution. And you'll only find out when you get that data point, right? You'll only find out, oh shit, he didn't have breaks, right? When you get the data point. And so, you know, when you're interviewing managers, when you're talking to your, the guy who manages your pension, your wealth or whatever, these are the questions you need to be asking him you need to be asking explicitly about the convexity of his portfolio what is he doing to put positive convexity to the portfolio is he relying on historical correlations is he relying on things that are low correlation because they were low correlation during all the good periods only to find out i mean i'm sure you've heard it a million times the well yes we didn't keep up with the market but look how low our risk was And so we were being prudent.

52:44And so, yes, our returns are bad in a good part, underperforming a good market. But our sharp ratio is good because our prudent, our conservatism is showing through. And then when the market, when they underperform the downside, they say, well, yeah, but that was an event that we couldn't have seen coming. All the while, they didn't see either side coming. As we talked about last time, they didn't see 2022 being bad at that correlation breaking down. And they didn't see 2023 and tech stock and NASDAQ going to the moon. Both times they missed it. And both times they had bad convexity in their portfolio because they were relying on faux linear, normal distribution, simple assumptions around steady correlation and ensemble averages.

53:34And also in practice, as a practitioner, I'm looking at strategies and managers and return streams all day for a couple of decades. And one thing that is very obvious is there's a massive survivor bias, meaning we automatically exclude all the things that had a big drawdown in the past. A lot of them go out of business. And so what's left are the ones where the risk hasn't surfaced yet. So our analysis is all about not how have you done in the past, but what's the risk that's going to get you in the future? And can I understand that? And do you understand it? And how are you positioned for it?

54:13And it kind of goes back to your convexity thought. And exactly what you said last time, the danger that's always the case. You think about what people, again, traditionally invest in, including the portfolio as diversifiers, what in my soccer analogy I refer to as defensive midfielders. When you go out and hire an absolute return hedge fund, you're probably hiring him as a diversifier against your otherwise equity beta growth assets, real estate, private equity, public equity. So you're hiring him to play defense, but you're incentivizing him to score goals. You've put him on a goal scoring metric, but you've hired him to play defense.

54:59He probably comes in and for the first few games, he's playing a defensive midfielder. But over time, as his performance suffers, he starts playing forward a little more, playing forward a little more. Your example that you gave last time, as he gets further and further down the straightaway, he decides it's OK to drive a little faster. It's OK to drive a little faster. And so just as he gets the most leaning towards the offensive end of the pitch, despite being hired to be a defensive midfielder, is when the ball gets booted the other way and he's not playing defense. And so you see this all the time.

55:33You see the underperformance during the long bull market of supposed low correlating defensive absolute return strategies whose beta inches ever higher and higher, their correlation to the market inches higher and higher and is inevitably at its highest just as the market turns. and then they end up not being and the ones who don't survive as you say drop out and the ones that get evaluated you know is the guy you know the guy that was driving 60 of speed crash but the guy was driving 55 of speed survived but he's still got the same problem you know the next unforeseeable not yet occurred curve is going to get him and so you know the questions that people need to be asking is about convexity, about tail risk, about protection, not relying on historical correlations, not making assumptions about stable correlations.

56:29The risk in your book is the correlation assumption. It's the fact that that correlation will consistently underperform in good markets, underperform in bad markets. It's not stable. One notion you've alluded to a couple of times is this idea of the Fed put. In other words, when there's a massive curve that is going to cause a lot of crashes, the Fed will step in to save the day and they'll provide stimulus and they'll do whatever it takes. So what some believe, and please tell me if this is accurate or not, but what some believe is that I already get free breaks because the Fed is there in case something really bad happens that causes systemic risk.

57:17How do you think about that? Yeah. And this is exactly what happened. The Fed regulates everybody into sharp world practices, in particular the banks. And then when they're all crashing, they've got to come in and save the day. Otherwise, the system goes away. But each time they do this, they're building more fire risk for the future. They create the moral hazard. They create the bad risk management. People say, well, I don't need to worry about it. I don't need to worry about fires because the Fed's always there to put them out. then the next fire starts and spreads a different direction than the one that they protect decided to protect you from and you get caught in it plenty of people got caught in oh wait plenty of people got caught in 2020 plenty of people got caught in 2022 yes the fed comes in and saves the system the system itself has its own sort of natural resilience you know strong capital hands come in and soak up cheap assets that have been forced out of levered weak capital hands.

58:13The system's naturally resilient. But certainly the Fed jumps in and helps when the banking system's falling over and risks a societal pulldown. But each time they do that, it makes the system more risky through time. Yellowstone Park, the more you put fires out, the more risk you get. So the more you have to put fires out. So you think about what the Fed's doing or central banks are doing. They're always fixing the problem that was created by the problem they fixed. when they were fixing the problem that was created by the problem they fixed and it builds and builds and builds and that just creates this you know inevitable self-organized criticality sand pile thing the sand pile builds up and it collapses and it builds up and it collapses it's the history of life in the world it's the punctuated equilibrium of evolution that is not the way darwin drew it up of a nice stable evolution evolution does very little and then jumps and then does very little and then jumps, right?

59:10It all comes back to complex systems. Self-organized criticality. Everybody should go read one of the best books you'll ever read as an investor is a book by a physicist by the name of Perbach, P-E-R-B-A-K, called How Nature Works. It's about how nature works. It's self-organized criticality, Sort of that specific part of chaos theory in physics that talks about sand piles and avalanches and forests and evolution and volcanoes and all the stuff that has this power law, self-similarity, self-organized risk that then disseminates to find restored stable equilibriums. And that's exactly how markets work.

59:56One of the ideas you mentioned last time is the Fed's primary goal is financial stability. So obviously their mandate is reasonable growth, reasonable inflation, but the primary objective is to maintain financial stability. So if every time there is a crash on the road, they come in and they provide stimulus to try to allow the system to survive, how do you reconcile that with the notion that you just described, which is every time you do that, the system actually becomes more fragile. And at some point, you could have a major accident that is not easily recoverable from. Yeah. 2008 is a decent example.

1:00:40The Asian crisis is a better example where you had countries that really went through a cleanse, 25 % GDP contractions, years of 25 % unemployment, 80%, 90 % currency devaluation. That's a scale unlike anything you saw in GFC. But the GFC, pretty severe. And investors' compounding path took it a major hit in that process. with good breaks, you cut off a significant amount of that major hit and get to participate in the recoveries much more aggressively. Same with 2020, same with 2022. So the fact that the Fed at a certain point, hopefully, or central banks, governments are going to come in and bail it out, hopefully, doesn't negate the fact that you had a 50 % drawdown.

1:01:31And if you can cut off that 50 % drawdown back to my S &P 480 month path, 10 worst months contribute 40 % of that 40 year compounding path. You cut off those 10 worst months, you will structurally change your terminal outcomes massively. And you'll get to participate if you're doing it right, you're participating more in the 10 best months. So you've just driven the contribution of how you've constructed your portfolio to the parts that contribute the most. And so there's no point ignoring the drawdown because you say, well, the Fed won't let the world come to it in. I'll just write it out. Why not take advantage of it?

1:02:13Why not benefit from it? It's the consistently, the most attractively priced opportunity out there because the fact that that happened probably is because it's priced like it couldn't happen. And it happened because the reason, again, you had that extreme left tail, left skew tail of end is because there was a whole bunch of leverage build up around the belief in the moral hazard that they wouldn't let that happen. I guarantee you, because I've talked to a lot of them, none of the central bankers wanted to let all the banks go. None of them wanted to have to bail out every major bank in the world.

1:02:49They didn't do it on purpose. None of the Asian central banks wanted their currencies to devalue 80%. They were swearing up and down, they'll do everything they can do, have to do to prevent it. Yet it happened. And so I don't know what the next peak and bubble and correction and redistribution is going to be, but I do know it will be. You know, the world's not going to somehow magically for the first time in all of history, you know, it barred on a straight line, seven and a half percent annual gain of investments for the next hundred years. It ain't happening. The variance will matter a lot more than the mean.

1:03:33And you construct your portfolio, optimize to the variance, optimize to the geometric compounding, and it will tell you, own the convexity. All right. If you optimize your portfolio to annual returns, one slice of time, short-term probabilistic, it'll say, ah, sell convexity. But then you're destroying value in the divergences. Something that you just said strikes me as really interesting, and that in 2008, 2009, the market fell 50%, yet the Fed put was still obviously in force. During COVID, the market fell a third in five weeks, yet the Fed put was still there. And so you've had these major drawdowns, and you mentioned other countries, major drawdowns where the central banks stepped in and did whatever it took to allow the recovery.

1:04:23It's also possible that they can't protect you during the worst of times. So you've had these massive drawdowns, even with that protection, with that hedge. We shouldn't assume that that's always the case. The next crisis, they may not have the credibility or the tools or the ability to engineer a rebound. So that's another reason to try to protect against that tail. Yeah, absolutely. But again, the main reason to protect against that tail is so you can participate more in the good times. The reason to have better brakes isn't because you're afraid of crashing. It's because you want to drive faster.

1:05:03So many people miss this perspective. I'm not telling people to be more cautious. People are already too cautious because they're driving around without brakes. right there as we said last time they're just suckers for the bearish siren song that is just constantly pervading the airwaves and they're always cautious and their their financial fiduciary is cautious and he's being prudent he's optimizing to a sharp ratio where he's forever foregoing upside to reduce downside it's the wrong mindset it's the wrong way to do it and so you know the the point isn't that i think the world's coming to hand or there's going to be some phenomenal crash, there will be crashes.

1:05:43I think everybody pretty much accepts that. The point is, in between the crashes, there's enormous opportunities. Coming out of the sharp curve is the best time to go fast. But everybody's backward-looking risk methodology, exactly as we discussed last time, is telling you after that really scary curve, we better go slowly. And then eventually you said, geez, if we could drive it on the straightaway for three miles now, let's start going faster. So you end up always with the, you get it wrong. You're buying high, selling low. And that's exactly the behavior, again, that creates this dynamic. Too much risk builds up because it now looks good so I can add leverage to it.

1:06:26I can expand my gross in my portfolio, a multi-asset portfolio, benefiting from what's been really good correlation for the last bit of this straightaway. I can gross it up relative to my risk appetite, only to find out I did that exactly the wrong time. And now I've got to gross it down because all of a sudden correlation changed on me. We hit a curve I hadn't foreseen. And that's why markets behave the way markets do. An interesting way to look at what you just described is if you're the driver in that car, it's very understandable that after a sharp turn, you're going to be more cautious and drive slower.

1:07:02and after driving for an extended period on a straightaway, you're going to feel more optimistic that the future will look like that and you'll drive faster. Yet, if you're sitting in the tower 100 feet up and you look down and you're looking at the big picture, you could see what a mistake it is to behave that way. But it's understandable for the person doing the driving. Especially if he's driving without brakes. Right. Right. But if he's driving with brakes and he understands that, he's hitting the accelerator as soon as he comes out the other side of that curve he knows it right he knows i got brakes i'm ready to go he's constantly looking for the opportunity to go go go knowing that he's got brakes if he's got to back it off right whereas the other guy's trying to adjust his speed just by pushing on the gas or releasing the gas pushing on the gas or releasing the gas well inevitably he's going to screw it up He can't react fast enough.

1:08:00And it's that, again, this is the guy driving that way is exposing himself to the volatility drag, this compounding volatility drag where negative compounds pull you further back than the positive compounds recover you versus the guy who's got the convexity who's benefiting from the divergences. He's immediately picking up speed in the good parts and immediately decelerating in the bad parts, not the other way around. And that will change your compounding path just unbelievably and magically. I said to you last time, it's very hard to find the point where you made the right call. You just kind of always are right.

1:08:40You're always benefiting. So you've explained the concept of insurance as it relates to investments as not being zero sum. And it's because you have to add in the fact that the buyer of the insurance can take more risk. Now, outside of that, if we just look at the insurance itself, does the insurance itself have a negative expected return over the long run? No, absolutely not. Right. Again, it depends on price. It depends on unknown future outcomes. But the insurance, again, it's like deposits in a bank. Deposits are a negative carry, but paired with loans, it's a very profitable business. It's non-recourse leverage against higher priced recourse leverage in a bank.

1:09:27And they keep all the difference. And if they screw up their lending, they don't have to pay back the deposit or somebody else will. So, again, buying insurance is the same concept. Now, you don't buy insurance. You wouldn't. Again, it comes down to this difference. Who's on a ergodic view? The insurer who insures every ship in the sea versus the guy who's in the non-ergotic path. He has one ship and he's in the shipping business. what behooves him to buy insurance and make more money delivering every day regardless of the weather you wouldn't bet against the insurance company you wouldn't say well i'm going to take the other side of the insurance company's view as a bet that i think there's going to be more than five percent ships that sink this year now people do that in the markets that's a a trading activity traders might make that bet because they say, oh, everybody's forecasting good weather this week, and I think it's going to be bad.

1:10:23And a few of these ships are going to go out and sink more than the cost of the insurance. So I'll take a bet that more ships sink because I have some different view of the weather. That's a trading activity. That's not an investment activity. That's not a shipping activity. And that's zero sum. So there's a winner and a loser. Yeah. In a bet, there's a winner and a loser. In an insurance activity, there's two winners. In a deposit, in, you know, de minimis, but the depositor makes money. But so does the bank, right? Now, sometimes the bank's relation, because the bank's in the business of intermediation.

1:10:58So his relationship is between the non-recourse leverage of the deposit and the risk he's taking in lending that out. And he doesn't care in any given year, which side the profit comes from. So in most years, the profit looks like it's coming from the loan. In some years, 2022, it might look like it's coming from the deposit because interest rates went up so much. So he made the profits from the liability side. But he doesn't care which side it comes from because what he cares about is the relationship. Same as my shipping guy. He doesn't care which side he makes money on, as long as when he's shipping, he makes more money than the cost of the insurance.

1:11:37And when his boat ships the insurance makes all the money and covers the cost he also bought you know business interference insurance right and so it's the relationship between the two in this insurance activity which is very different to a betting or gambling or trading activity different activities and remember a guy selling puts and a guy buying puts are taking you know in a bet in a trading are taking different views of the market. But a guy buying insurance, the guy selling insurance and the guy buying insurance are taking the same view of the market. Just one providing non-recourse leverage against that view and one taking the non-recourse leverage to take that risk.

1:12:18And I suppose a very simple way to understand why both sides win is you just have to look and see there's a massive insurance industry that has been around for decades, generations. There's life insurance. There's health insurance. There's all sorts of insurance and a property casualty insurance. And those, you know, that both sides do well in that. And it's been around forever. There's a shipping industry, right? There's a banking industry. There's an investment industry. Trust me, all of this vault selling that's getting done in the world, the people taking the other side of it aren't going out of business.

1:12:57Somehow, some way, right? banks aren't going out of business taking deposits despite paying for them because they're using that as insurance non-recourse leverage to go out and take other risks that's the way the financial markets work and i'm very confident that end of the day the guys using this supply of optionality of volatility to manage better their risk are doing just fine doing really really fine relative to the guy who's selling it. Dave, you've been super generous with your time, not just on this episode, but the previous episode. And I know it's late at night in Singapore where you're located.

1:13:39So I have one final question to ask you. So let's for a second talk about implementation. So how would investors implement your strategy at a high level? Is it as simple as they have a, whatever the portfolio looks like, it's a diversified portfolio and they can add a tail risk fund or strategy that acts like insurance where its return behavior may be small, low positive or low negative returns for a while, and then you get a big positive return when you need it most. Is that conceptually how it works? Yeah. So the concept is, so one, you would never look at your insurance on a standalone basis.

1:14:20So looking at a tail risk strategy on a standalone basis is meaningless. Like I said, it's like looking at your insurance and say, well, it costs this much every year. And there was no fire. Yeah, there was no fire. There was no, you know, without even asking how much insurance did I own? I just look at the cost. It's meaningless. You've got it. So the whole point of this is to free up capital. So most portfolios have too much capital tied up in faux defensive strategies and assumptions around correlation. Obviously, most prevalent is bots. So the whole world has for decades invested in 60-40 some sort of assumption around equity bond correlation, which has been a disaster.

1:15:05The bonds have been an enormous, enormous opportunity cost to that portfolio. You would have been way better owning explicit protection and putting more of that money into equities. You would have had something that protected much better in the bad times. And you would have had a lot more money participating in the good times. So the whole point is to free capital that is in inefficient, ineffective diversifiers and get more money participating because you can more efficiently protect with something that's far more explicit. And that combination, even when bonds were good, outperforms. right and the other thing one of my other pet peeves and you know i'm sure you've thought about this before too is the one thing that nobody measures is opportunity cost right nobody measures what i missed out on everybody says well you know my bonds didn't really cost me any money you had 40 percent bonds that have you know flatlined for 13 years now in a real basis or down in a market where equities have tripled the opportunity cost of that everybody ignores because nobody wants to admit to it nobody wants to go and tell the trustee or the investors group yeah we screwed up we should have made this much money because we could have put brakes on the car and put more money at work so the whole point i'm not trying to reduce people's exposure to growth assets to unbounded potential upside i'm trying to help them increase it i'm trying to reduce their exposure to bad diversifiers bonds absolute return sharp ratio optimized low-volve strategies that neither participate nor protect your portfolio everything should be judged does it participate does it protect you start pairing those together negatively correlated stuff and there's this magic thing uh you can tell your your viewers can go and google parando's paradox or shannon's demon there's this magic thing where you pair things together that negatively correlate you can shannon's demon shows you even if they both have a negative expected return and they negatively correlate and you rebalance them it's profitable right And that's a little bit of this magic dust of the convexity in the thing.

1:17:38And so that's really where you're focused. Now, obviously, people need to make decisions about pools of liquidity. If you're an endowment, you've got money that you've got to pay out. If you're a pension and you've got mismatches and inflows and outflows, et cetera, there's a liquidity requirement in things. But in my opinion, that's not part of your investment portfolio. the part of your investment portfolio that's targeted at terminal compounded capital for somebody's retirement, for the next generation of the family, for the university 25, 35, 45 years down the road, needs to be focused on what participates and what protects.

1:18:14And that's where you really need to, and you're not going to do it overnight. You're not going to take a traditional modern portfolio theory, sharp world portfolio and revolutionize it overnight. But every chance you get, you can be reallocating what I call debt capital that's in these ineffective diversifiers and putting some of it into explicit protection and some of it into explicit participation, whether that's equities or real estate or high yield credit, stuff that participates, stuff that participates, knowing that during the good periods of correlation and participation, that stuff does great.

1:18:52And in the fact periods, it's the correlation risk that your hedge needs to kick in and provide the deceleration for you. I think you alluded to it just now, and it's really important because you're buying this hedge. So let's say it's a fund that provides that hedge. And if you look at it on a standalone basis, you may be disappointed, but you have to think of it in an aggregate basis. You have to look at the whole portfolio and look at it through time. And in reality, you're buying this insurance, you're buying this hedge and hoping it does poorly, right? Because that means the rest of your portfolio is doing great.

1:19:29And so if it actually does poorly, you should look at it and say, oh, that was a waste. You want it to do poorly. You want it to do poorly, right? You got to recognize it's allowing you to take extra risk. It's allowing you to drive more confidently, more aggressively, right? That's the whole point. The whole point of brakes is so you could drive faster. You don't put brakes on your car. You don't take your 60, 40 car, your 60 % car and add brakes and still drive at 60%. No logic to that, right? The whole point is to let you be more confident, more aggressive in how you pursue the participating risk.

1:20:07That's the point. Nobody would evaluate JP Morgan separately between the cost of their deposits and the return of their assets. Nobody's going to kick out all the depositors and say that's a shitty business that loses us money almost every year. It's mindless. And exactly as you said, this is exactly how we operate. We hope our investors never need us, right? We hope that their compounding assets and participating assets just grow and grow and grow and grow forever. And we'll spend our days all day, every day, trying to increase and increase and increase the potential asymmetry and insurance that we have if they ever need us.

1:20:46And we high five ourselves every day, month, year that they have a great year. All we focus on is their returns. It's not our job to focus on our returns. There's no point us focusing on returns. Everything we're buying, we're buying because it's priced like nobody thinks it could ever happen. We're not buying it because we think it's going to happen. We're not making a bet, to your example of a zero-sum bet in insurance. We're buying insurance so that they can take risk. We're focused on the relationship between the insurance and the shipping business. We're not focused on a bet that we're trying to make with somebody who's providing the insurance.

1:21:25And that's the mindset that a good tail risk manager has to have. If a tail risk manager is trying to optimize his performance, optimize his compensation, he's lost the plot. That shouldn't be his job. You don't hire a goalkeeper and then incentivize him with goal scoring metrics. Well, Dave, this has been a fascinating two-part discussion. I appreciate you sharing all your insights. I've already heard great feedback about episode one. I assume the same for episode two. So thank you so much. Thank you, Alex. It's a lot of fun and really, again, just like last time, so well prepared. And it's fun to have this sort of come back again, elaborate on that.

1:22:02So I appreciate you making the time for it. Thanks for listening. We hope you enjoyed this episode. Please visit our website at insightfulinvestor.org to access past shows and learn more about our podcast. If you have questions, feel free to email us at info at insightfulinvestor.org. And if you enjoyed the discussion, please subscribe to this podcast to ensure you don't miss future episodes. And don't forget to forward today's conversation to others you think would enjoy listening. This podcast is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice.

1:22:40All opinions expressed by podcast participants are solely their own opinions and do not necessarily reflect the opinions of Evoke Advisors, their affiliates, or companies featured. Due to industry regulations, participants on this podcast are instructed not to make specific trade recommendations, nor reference past or potential profits. And listeners are reminded that securities trading, commodity trading, and alternative investments are complex and carry a risk of substantial losses. As such, they are not suitable for all investors.

1:23:36Thank you. or their services. Similarly, the inclusion of a guest on the podcast does not imply that Evoke Advisors or the host endorses the guest or any company with which they may be affiliated or employed. Evoke has neither paid nor received compensation from guests for their participation.