Dan Villalon, Global Co-Head of Portfolio Solutions, AQR Capital Management

[00:00:01] Hello, this is Dean Curnutt and welcome to the Alpha Exchange, where we explore topics in financial markets associated with managing risk, generating return, and the deployment of capital in the alternative investment industry. Today's world of ETFs and mutual funds increasingly feature new flavors, a popular one of which is derived from embedding optionality.

[00:00:28] There are plenty of ways in which one might contemplate risk managing and shaping the distribution of equity returns using options. Common strategies like overriding create income but limit upside. Others, like the zero-cost collar, create both upside and downside guardrails on returns. These strategies can be backtested. Because they also exist in the market with more than $200 billion in AUM, the performance of the funds can be evaluated as well.

[00:00:56] With this in mind, it was a pleasure to welcome Dan Villalon, Global Co-Head of Portfolio Solutions at AQR, back to the Alpha Exchange. Dan walks us through the findings from his research, published in a two-part series on the AQR website. In these notes, Dan dissects the drawdowns and returns across these funds. The findings are rather striking.

[00:01:20] Across a wide sample of buffered funds and option-based strategies, very few delivered both higher returns and smaller drawdowns. In fact, most underperformed their beta-adjusted benchmarks on both fronts, meaning they not only lagged in returns, but also failed to meaningfully protect against losses in periods like the COVID crash, the 2022 inflation-driven drawdown, and the volatility of early 2025.

[00:01:48] Even strategies designed explicitly for downside protection often fell short when it mattered most. I'm a big believer in option strategies and in the value of the S&P options market as a vehicle for risk transfer. These results were a surprise to me. Dan outlines three key drivers, their persistent cost of buying options, the structural frictions involved in implementation, and the surprisingly high management fees for such rules-based products.

[00:02:18] Dan also introduces a more behavioral theory, what he calls the placebo effect. The idea that investors feel safer simply because they're told they're protected, even when the data shows otherwise. I hope you enjoy this episode of the Alpha Exchange, my conversation with Dan Villalon. Dan, it's great to welcome you back to the podcast today. That's great to be here, Dean. Thanks for having me back. Yeah, excited to have this conversation.

[00:02:46] You and Cliff, but really you, authored a couple of papers. One of them is clearly titled by Cliff, but it's on this emerging industry of risk-managed equity, where there is an option overlay embedded in typically, I would say, a long S&P base exposure. A lot of these strategies are called buffer strategies.

[00:03:14] So your first one, which was out in March, is called rebuffed, a closer look at option risk management strategies. And then the second one, which I think has got to have a cliff title, is called buffer madness. But it's a review that you did. You dove in and you got your hands on some data, which AQR is known to do.

[00:03:39] And you took a very close look at the performance of this overall category. You've got some conclusions that, you know, don't, I don't think jump off the page as advocating for these. And so that's really what's going to be the basis of our conversation. We're going to talk about your original work and then your follow-up.

[00:04:05] We'll talk about a lot of the feedback, which I think has been a big part of what's landed in your inbox on the back of some of the work. So why don't we set the table? So why don't you start with perhaps even starting with what got you interested in starting to look at the risk management segment of the ETF and kind of mutual fund industry? So why don't we start there?

[00:04:34] What got you interested in doing this work? I will disclaim off the bat, this research was not my idea. I have a partner here at AQR. His name is Brad Jones. And he stopped in my office, I guess a couple months ago now, and said, hey, Dan, would you mind taking a look at this segment of the industry?

[00:04:55] Something that he had been hearing out in the field more and more about sort of these call it options enhanced or options altered ways of harvesting the equity risk premium. And he said, I'd love to know kind of what's the AQR take on it. And so that was the genesis of it. And back then, I didn't really have a narrow focus on any particular type of fund. You know, there's these buffered things. There are these sort of like more income oriented things.

[00:05:23] So I just said, well, let's just look at all of them. And so the first analysis kind of looked at it. I know this is sort of a nerdier podcast, so I'll use nerdier words. We looked at a fairly heterogeneous collection of strategies. Three, in fact, defined outcome. I always forget the next name here. Derivative income. And the last was a subset of the equity hedge category in Morningstar called options trading.

[00:05:53] And those kind of three together seemed to me like they would represent most of what was going on, most of the opportunity set for investors who were looking to allocate to equities, but to kind of tweak, to kind of shift the expected distribution of returns they would receive while being invested. So that was the genesis of the first paper.

[00:06:17] I have to admit, I didn't think that it would have gotten much of a response. And so I kept it short. You know, I didn't think people were actually going to respond to it. And I was dead wrong. Right. Yeah. Well, listen, there's a lot of vested interests in these products. I think this is a very interesting conversation because, you know, I was just on LinkedIn last week and there was a post basically,

[00:06:46] I think it was almost like celebrating 20 odd years of a paper from one of the French banks. And the French banks are very known for their incredibly complex derivatives expertise. Right. I mean, you know, if you thought a variant swap was complicated, what these folks are doing on the vol and correlation front, it's super complicated. And the only thing harder than understanding Greeks is Greeks with a French accent.

[00:07:15] I'm with you. Totally comprehensible. Right. Well, when we talk about gamma and vega, they've got cross gamma. They've got things like Vanna and Volga, all these different cross Greeks. But, you know, one of the points around that structured product industry, which really has been a European and Asian, I would say, centric industry, has been complexity, which really brings a lot of cost as well.

[00:07:45] These are very illiquid. They're optically cheaper or optically a better deal than they really are oftentimes. And it's sort of complexity for complexity sake. And I think what's interesting, and I would just sort of juxtapose versus the area that we're going to have a conversation on is a lot of these are embedding options on the S&P 500. And, you know, maybe along with the U.S. government bond market, it's about as liquid as they come.

[00:08:15] The vol surface for S&P options is very well governed by lots and lots of people. It's very efficient. And so I think that's an interesting sort of backdrop is we're dealing with a market where the options are very, very liquid. And at least at face value, the frictions to get in and out of these things are on the very low side.

[00:08:39] Again, just sort of juxtaposing that versus the industry of highly structured products that, you know, often is seen in Europe and Asia. So I thought that was interesting. Tell me a little bit about the Morningstar database. So is that the source of all of the funds that you looked at? It's from the Morningstar database. And is that some combo of mutual funds and ETFs? What's in there?

[00:09:02] Great. Yes. So the Morningstar database we used as what we thought of, what we have access to as being the most comprehensive set of sort of liquid solutions that investors might have who are contemplating these types of products. You're completely right. It's mostly ETFs and mutual funds. Of the two, it's really dominated by ETFs, particularly when it comes to buffered funds.

[00:09:32] The vast majority there are in ETF form. And to your earlier point, it's usually against extremely liquid markets, the so-called reference assets to these funds. By and large, tends to be equities. And among, if you look at all of them, the number one is S&P 500. And what percentage of that? Do you know what percentage of the cap? What is the capital base in the sample size?

[00:09:59] Yeah. So for defined income funds, so these would be ones, you can think of these as covered call. So you're long the underlying, the reference asset, and you sell some out of the money calls. That is actually the largest category, at least as of last month. That's about $126 billion. The next category, defined outcome. So these are really where the buffered funds, in fact, buffered funds, I think, account for over 98% of the AUM in this category.

[00:10:29] And we can get into the construction there. It's a little bit more complicated than a covered call. They're about $64 billion as of the end of last month. And then the third, this sort of subset of equity hedged called options trading. That one's about $51 billion. And that one, you can kind of think of as like a hybrid of those first two. This is one where, for the most part, you are, on average, long some sort of bet on equities.

[00:10:57] But it involves a little bit more timing, I would say. It's a little bit more opportunistic in terms of when to go long vol or short, when to buy or sell a putter or a call. That one tends to have an average equity exposure of somewhere between sort of a 0.4 and a 0.6 beta, but fairly time-barying. And one of the questions I had, just as I was sort of thinking through this subset of funds in these three categories,

[00:11:26] is the degree to which, and you sort of touched on it there, the degree to which they are incredibly systematic, perfectly tied to a mandate with no wiggle room, versus there's some flexibility, whether it's where they strike their options, how they roll them, the extent of the timing of either increasing or decreasing that beta. Maybe that's a function of judgment.

[00:11:56] Maybe it's a function of some rule where if the option goes, in the money by a certain amount, I have a rolling scheme. Did you get a sense as to how mechanistic or how much flexibility they had? Yeah, these are highly mechanistic, highly systematic. I suspect that is a selling point of these, that this is sort of a, you know, no discretion. This is you sign up for a plan and you get that plan.

[00:12:27] Let's take defined outcome, the so-called buffered funds as an example. 90% by AUM and the vast majority just in terms of number of funds make one trade a year. They own annual options, you know, on the third Wednesday of a given month, they buy. 12 months later, they kind of switch those out and bring in a new one.

[00:12:54] So it's, I mean, it's a pretty mechanistic strategy. And we didn't really find much evidence when we kind of dug into these of ones involving much, call it discretion. You know, you think of some like more tail risk hedge fund types where you would be a little bit more tactical or opportunistic in terms of when do you go long volatility. But these strategies we looked at, for the most part, are highly rules-based. And I suspect, like I said earlier, I think that might be part of the appeal.

[00:13:24] It really makes it kind of simpler to explain to the potential investor exactly what they're signing up for. And I have to assume that's one of the reasons they've been able to garner so much in terms of assets under management and are predicted. They're expected to continue to grow. Right. And so one of the things that you found early on as you started to gather the data was that there had been a significant amount of growth.

[00:13:52] Why don't you walk through just a little bit of the, you know, the backdrop for the study itself in terms of the timeframe over which you looked at the measurement? What were you looking for? Why don't you walk through that? Great. I'll start with kind of blog post number one, because blog post number two, I responded to some of the feedback and we got a little fancier. But let's start with kind of the first one as our baseline.

[00:14:20] So for this, I wanted to have sort of an update that I might be able to come to some reasonable conclusions. So I, you know, admittedly, I just picked a date. I picked January 1, 2020. I was looking for about five years of data to analyze. So I looked at, okay, let's take a look at all of these categories of funds since then. And for simplicity, I'm only going to consider two things.

[00:14:49] Two reasons someone might allocate, someone might invest in these things. One is the expectation of returns. And the second would be something about mitigating their risk. So that's it. So if you read that piece, it's the exhibits are just two by twos. It asks, did the strategy deliver superior cumulative returns? And I'll get into a little bit more of that. Yes or no.

[00:15:15] And then did the strategy have milder drawdowns? Yes or no. And so I just sort of said, okay, how many of the funds end up in each one of these quadrants? Now, we mentioned this briefly a couple of minutes ago. These strategies are a heterogeneous bunch. That means a few things in terms of how much risk, how much exposure to the equity risk premium are they taking?

[00:15:44] So in the first study, we said, well, let's take a look over these five years. What was their average sensitivity to equity markets? Or in, yes, throw out a Greek term, what was their beta to equities? And we use that as the benchmark. And so when we asked about, are the returns superior? Is the risk superior? Superior to what? We said, well, it's superior to a beta matched index. So if- Is that by fund?

[00:16:13] Each individual fund is a beta? Thank you for that clarifying question. In retrospect, I could have made that clear in the original post. So if there was a fund, let's say that had a buffer or a put protection that covered all the downside, you would expect that to have a little bit less beta than one that maybe had much more exposure to the downside. So for each of those two, we would compare it to sort of the right beta, you know, a lower beta benchmark for the first and a higher for the second.

[00:16:42] So for each one of these hundred or so funds that we looked at, we figured out what was that individual fund's exposure to equities? And that informed the benchmark, you know, how much equity risk or how much equity allocation should we use? And then we just put the rest in cash. That was kind of our benchmark. That was what we were asking. Is it better than just being equities plus cash in terms of returns or in terms of risk?

[00:17:07] And so, of course, what you published originally and then followed up in Buffer Madness is you've got a two by two matrix. So the first part of that is on return. And the second part of that is on drawdown. Exactly right. And what we found was a very small minority of these funds looked good on both metrics.

[00:17:36] So kind of superior average returns or cumulative returns to this cash stock benchmark and milder drawdowns. The majority looked worse on both counts. And I think that was something that may have catalyzed some of the feedback we got. Yeah. I mean, listen, we're both from the University of Chicago and, you know, so much of our thinking

[00:18:05] is going to be from a no free lunch starting point. Right. And so a worse, worse, you know, almost sounds too bad to be true instead of too good to be true. Right. And so, you know, some of the questions that I had as I sort of thought through this and I've done my share of back tests, but you and AQR have done, you know, many, many more than I have.

[00:18:27] And so the thought process behind how you set up the, what you're trying to do is an apples to apples as best you can. And there are so many considerations that go into that. And I just was curious as you set this up and, you know, first maybe thought about on your risk, the risk dial part of the matrix, you have drawdown. There's other ways to measure risk.

[00:18:53] You know, for me, I always look at the realized vol of the series, right? If I'm might not improve on my drawdown, but if I've lowered my realized vol sufficiently, in some ways that's my sharp ratio input. I'm just curious as you sort of, as you set the tests up, you know, how you thought about the approach was this, this came to mind and this is how we're going to do it.

[00:19:19] Or were there other maybe competing ways to try to create the apples to apples comparison? Yeah. We looked at it in a bunch of ways and ended up reporting what we thought would be the simplest statistic. You know, this original piece was destined to be a mere blog post. So the kind of thinking was short and sweet. We have a limited amount of people's attention.

[00:19:47] So what's the easiest way to communicate that these things have been a disappointment? Well, if you're investing in something, you can disappoint either in terms of returns or risk. So what's the easiest way to show that? And we came up with the two by two. You are exactly right in that, you know, reality is always richer than what a two by two can describe. So since then, we have looked at other things. We've looked at volatility.

[00:20:14] We've looked at a sharp ratio, for instance, because that gives us a better picture of kind of both parts. Not surprising, as you would expect from a two by two, where the majority of observations end up in that sort of worst quadrant. As you would expect, the average sharp ratio of these strategies versus their reference asset is also inferior. And so that's how you end up in the bottom quadrant on average is some combination of worse risk and worse return.

[00:20:44] And indeed, we do find that when we look at our sharp ratios as well. Your starting point is a fascinating one in terms of date, right? Because not but two months after January, we've got really one of the most remarkable drawdowns in market history, along with an 80 VIX. And so these options, if I'm just sort of thinking through the mechanics of these, if I'm buying,

[00:21:09] you know, these options in February, I'm paying probably 12 to 15 implied volatility. And those are going to 80 implied vol. But I'm probably buying them then too, right? And I think that's some part of your work. I think some previous work that you had done around hedging fast and slow was that, you know, some part of, again, if we're mechanistic about it, we're saying, look, we're just buying these options.

[00:21:38] And in some ways, it's irrespective of the cost of them, right? So we know when COVID hit, the market experienced such a violent drawdown that these options became incredibly expensive. And so if you're, you know, truly systematic, you're paying that 70 to 80 vol along the way as well. Yeah, I think that's right. And in terms of the period over which we looked, since 2020, it's been a pretty,

[00:22:05] and I'm going to use this term very academically, it's been a pretty rich period to look at. And when I say rich, I just mean it's been full of interesting observations. You know, the worst thing for an analysis is when the backdrop is just the same thing, because you don't actually learn anything about how this thing could behave. So one of the benefits of looking since 2020 is, yeah, you had the COVID drawdown,

[00:22:30] you had 2022s, and then you had all of the craziness at the start of Q2 of this year. And for me, that makes the analysis more robust and probably more worthy of people taking seriously than if we had just done it over a sort of placid period. You know, for instance, by starting right before COVID, and like I said earlier, these strategies,

[00:22:56] most of them buy an annual combination of options, and then just sit on it for 12 months. So they kind of got that protection at a pretty good time, you know, before they needed it. You know, it really is interesting. And I thought it was, well, we'll kind of circle back to this, because analyzing the sub periods, to me, you know, whenever I do a back test, I always try to think, okay, what might I have imposed on this in terms of some bias?

[00:23:25] If I'm rolling options, did it just so happen that my rolling strategy was lucky or unlucky? You know, and so we'll try it a bunch of different ways. Or as you said, am I going through an especially sanguine period or the opposite? And so as we've just talked about the, you know, call it starting point of your analysis in 2025. And then, of course, through something like April, that was just, you know, incredibly chaotic.

[00:23:55] And then the same thing, I'm sorry, in 2020. And then the same thing happened in 2025, which we'll circle back to. I would love to learn more about the different sleeves, these three categories. And if there are any insights that you've been able to glean in looking at differentiation. And so I'll just sort of set the table by saying, when we talk about buffered, I think we typically mean a collar, right? I'm buying, put, selling, call, defined.

[00:24:24] I've got my guardrails up. So I've got literal protection there. And I'm trying to fund it by selling off some of the upside versus just overriding where I'm taking in some income, but boy, if the market just collapses, I, you know, that income's just not going to really do much for me. So those are different. And I just would be curious, of the three categories, any thoughts on what you learned from the differentiation on the risk and the risk profiles of them?

[00:24:54] Economically, just on first principles, I would have expected strategies that are net short volatility, or in other words, collecting the volatility risk premium. So covered call strategies, as an example, I would have expected them to look better than they ultimately did. And for the same reason, I would have expected strategies that were usually long volatility.

[00:25:23] So ones that, you know, we're paying a lot for puts to underperform. And, you know, we found that they underperformed by even more than we had expected. But I would have expected some symmetry there, depending on sort of which side funds were on. We found less than I would have expected. So both categories looked a bit worse in our analysis than I would have guessed ex ante. That's kind of like, one thing.

[00:25:54] In terms of the derivative income funds, there I would have expected worse peak to trough drawdowns versus sort of the beta matched benchmark. And, you know, indeed, we did find that for the buffered funds, I would have expected them to look good in peak to trough periods. And we didn't find that.

[00:26:20] So in the COVID drawdown, we found 90% of buffered funds underperformed for peak to trough, their beta matched benchmark. In 2022, that number was 69%. And then in 2025, the number was 83%. And the reason for mentioning all three of these is each of these drawdowns was pretty different.

[00:26:46] You know, you had pandemic, you had inflation oriented, you had kind of trade policy. It really kind of helps give some sense of robustness to our findings. The other thing that I should mention is buffered strategies. You know, most of them use options that have an annual ten or an annual holding period. But they also have different series.

[00:27:13] So there's a January series, a February series, a March series, etc. That's really useful for someone looking to study these things, because it means the path dependency, which month you happened to write these options, doesn't really matter that much. It turns out that these strategies are fairly evenly split by AUM across each of these monthly series. And so you're able to kind of like diversify out some of that noise about like, you know, what happened?

[00:27:40] The one that, you know, bought protection right before the COVID drawdown may have gotten lucky just in terms of when it decided to rebalance. But by looking at all of these, we're really able to kind of diversify out that rebalancing luck. There is one more kind of clarifying point I'd like to make about buffers. And that is they are a little bit fancier than a collar. So a collar, you have exposure to the underlying, you sell a call and you buy a put.

[00:28:10] Buffers do one more thing than that. In addition to buying a put generally at the money, they sell an out of the money put. And so they do have downside participation after a certain percentage of losses. And so in some way, I like to think of sort of the evolution of customizing return outcomes as, you know, in the beginning, you have the equity risk premium.

[00:28:37] The equity risk premium is hard for investors to hold. So they want to do something about it. They buy a put. They realize that doing that is expensive. You know, it's called the equity risk premium because you get compensated for bearing risk. You know, if you hedge out that risk, you know, goodbye to a good chunk of that compensation. I think a lot of folks said, okay, well, what can I do about that? How can I improve my expected returns? Ah, let's sell a call.

[00:29:04] Well, and that's your collar and kind of variants like a cashless collar. Buffers, I think, are the next step in this sort of evolution. I think folks have realized that collars aren't free either. There's no free lunch there. They tend to have lower risk adjusted returns than the underlying asset whose returns you're trying to convert to something more appealing. And so what buffers do is say, okay, well, you know what? Let's sell an out of the money put as well.

[00:29:29] That way we'll kind of like lower our cost of the, you know, the explicit cost protection. You also end up weakening your downside protection. And I think that's really the insight of how can it be that these strategies, these supposedly downside oriented strategies end up giving you worse peak to trough drawdowns than just simply de-risking your equity allocation from the beginning. Yeah, you know, we talk about the equity risk premium.

[00:29:58] Then, of course, there's the vol risk premium. The old saying goes, sell the straddle and go to lunch. You know, that vol clears the market at a level that is higher than it ultimately materializes. You know, I always go back to Fama and think, okay, well, you know, who knows? Maybe there's a 50% one-day drawdown in the S&P waiting in the wings. We just don't know about it yet, right? Yeah.

[00:30:25] So we just, you know, we're trading our experience with the distribution, but at least empirically, vol, implied vol has cleared the market higher than the observations we know of so far. And then, you know, adding to that, there is in some ways, I'll just call it the skew risk premium, which is, you know, there's not just strategies in selling vol, but in also selling the skew.

[00:30:49] So whereas the collar is buying, put, selling call, there is a cottage industry of, you know, quanti folks that are doing the opposite. You know, they are selling, put, buying call with the idea that the realized skewness of the S&P empirically has been lower than that, which is implied by the skew, right?

[00:31:16] So whenever you do that collar trade, you are paying, I don't know, it could be 25 implied vol for the put and selling the call for 17. Maybe it's 20, something like that, but there's a spread. And I think that's, you know, one part of the collar where people might say, okay, that's baked in to the S&P vol surface. You know, it's the global asset of wealth and it's the thing that everybody kind of wants to ensure.

[00:31:44] And there's a lot of demand for those puts and maybe that's why that vol surface clears the way it does. Um, that's kind of one, you know, one thesis on maybe why there's that underperformance. And it leads to a, I think a great observation, which is if you find a strategy involving, you know, options that tends to, uh, underperform, take a look at what the opposite of that does.

[00:32:11] In theory, uh, if you can, if you can pursue it frictionlessly in terms of, uh, trading costs all that, uh, maybe that's, maybe that's a better strategy. Going back to our periods. And I think you nailed it. You've got your first, as your analysis starts, it's the beginning of 2020, big, big drawdown, massive vol 2022, you know, a 20% S&P drawdown, but a quieter one covered a lot of ground,

[00:32:39] but, you know, VIX, I think was 35 was a high for the year, nothing even close to 2020. And then this last one in 2025, you know, a short lived risk event, but a doozy as well. Tell us about the, I want to say you went from January through April of 2025. So very fresh that analysis, what did you find there? So this kind of gets to our second piece.

[00:33:06] So our first piece we put out, um, uh, got a ton of feedback and, um, which I was very thankful for, I mean, this is really the only way in which we as an industry get better is we publish, we get feedback and we, you know, reanalyze based on it. So what I did was like, I collected all of the comments I got and I, I kind of grouped them.

[00:33:31] And the second piece we wrote was responding to the six most common bits of feedback. Some related to the horizon over which we did our original analysis. People had an issue with going back to January, 2020. Um, and so we said, all right, let's look longer and let's look shorter. I think there was this belief that doing this analysis from 2020 until today, even though

[00:34:00] there were plenty of crazy events out there, equity markets did pretty well over that period. And I think there was a belief that, okay, well, it's not fair. It's not fair to compare say buffered ETFs versus equities in a period where equities have done really well. Um, I totally agree with that. I mean, the first thing we did to control for that was we beta adjusted everything. We said, okay, well, how much equity risk are these funds actually exposed to in the first place? That will be the benchmark, not a hundred percent equities.

[00:34:30] We wanted to make it, you know, as fair as possible, but still, uh, we wanted to respond to, uh, the comments of, nah, you know what? Maybe there's something funky about the period in which you looked. So we look longer, we look shorter in terms of looking shorter. We said, okay, let's just look year to date, 2025. So January until at the time, April, like you said, um, high volatility, high uncertainty, high macro instability, and the S and P just getting crushed in the first week.

[00:35:00] Of April. And we found basically the same results. I mean, if anything, even worse results for the defined outcome or so-called, uh, buffered segment of, of, of the industry. And I think it gets to, you know, you, you, you, you invoked Chicago, um, you know, options markets are decently efficient, particularly when the underlying, the reference asset is

[00:35:25] something like the S and P just because you are in a high ball or low ball environment. Doesn't mean that options are somehow intrinsically mispriced. You know, you, you wouldn't be surprised that if something that doesn't look great in previous crashes, um, also doesn't look great in a new crash. You know, this is really the markets are able to price themselves pretty well. And I think for that reason, you know, we don't find, um, a crash, the latest crash coming

[00:35:53] to the rescue for, for these strategies. Yeah. You know, again, I just go back to this timeframe, this last five years. And, um, one of the, I think more interesting aspects of the COVID crash was the equity derivatives fallout was really, I would say more substantial than most any other market. You know, if you were long a distressed credit, of course that thing in a mark to market basis

[00:36:19] got really dislocated, but you, all you'd had to do was, was hold on. I'm not saying that was easy, but if you held on, it was, the price was repaired and it was repaired relatively quickly. Equity derivative strategies don't work that way. You're crystallizing losses, right? If you sell volatility or you're short something that's got negative convexity, you can't just hold it. It's a mark to market loss that gets crystallized.

[00:36:46] And a lot of the, those losses were very public. Like AIMCO, uh, Alberta investment management company lost several billion dollars. Um, Allianz, you know, some of that was fraud, but they lost billions of dollars. This hedge fund Malachite shut down. Um, and so one of my observations, and, you know, I look at this market really closely was that the capital base in the market really got, it fell off.

[00:37:14] There was just not nearly as much capital supporting, um, the risk taking capacity for the market was compromised. And that really was into 2020 and then through 2021, it started to come back in 2022. And so again, as I just try to think through the potential uniqueness of the, you know, the sample 2021 by at least the counts of one large investment bank, I won't say which, but they had done some

[00:37:43] backtesting, they said it was one of the worst years they could find to basically hedge, you know, think about maybe 2009 is a terrible year because vol is super high and you're still buying puts at very, you know, high prices and the market's going up. And that was kind of 2021 where the vol risk premium, which is always a thing was especially high in 2021. So that's just another thing I was thinking, you know, perhaps that's some part of it.

[00:38:13] I know you've got a, you know, there's a consistent story here, um, as well, but, uh, I just wanted to share that, you know, the fallout from COVID really made its way into pricing for a year and a half afterwards. I'm fully with you. And that that's always the risk in any empirical analysis is that if there was some element of the, or some part of your history that was idiosyncratic and dominated the average, then you got to be careful.

[00:38:41] And one of the nice things with this is we've been able to slice and dice the history by high rates versus low rates, high volatility versus low volatility, different crashes to see, you know, is the behavior is, are the conclusions we're coming to driven by just one thing? And maybe these strategies work great in some specific environment. And if that environment is identifiable in advance, then maybe there's a good reason for these strategies.

[00:39:07] We just found consistent results to what we published, no matter how we, no matter how we looked at it. And if you were to think about the, the source of the underperformance, and maybe this is asking the same question, just a different way. But when I think about an overrider, right? So you're just long the index, you don't have the put, but you're gaining a little bit of income, but you are subject to the opportunity loss due a market that just charges through your strike.

[00:39:37] Do you have a sense as to, you know, when, when you sort of think about the underperformance, is it that the market's just going up more than it's supposed to? Are the, the hedges, you know, in the context of a buffered strategy, are they just not struck close enough to the money? Just curious if you have any thoughts on, on just distinguishing around the sources of that underperformance? It's a great question.

[00:40:06] If we were to do an attribution for, why? It can't be that markets went up or down or even sideways during this period, because we've accounted for that by doing the beta adjustment. One thing I know that you have, like I said earlier, sort of a nerdier than average listening audience. So I will go into one more test that we did.

[00:40:33] Our original analysis in some ways, I'm going to use an air quote here, cheated in the sense that it estimated the beta of each one of those hundred or so funds using the entire sample. So you can imagine that maybe one of these funds started with a high beta and then just sort of lowered their beta over time. Let's just imagine monotonically over those five years, each month, it just took a little bit less equity risk.

[00:41:03] And if equities did sort of the inverse, they did better and better over those 10 years, then that would be totally unfair for our analysis of measuring kind of that look-ahead bias would have made them look worse because of the methodology and equity performance. What we did in our second piece was an out-of-sample version of beta. So there's no look-ahead bias. We estimate the beta looking only at returns that have been realized. And then say, that's going to be our next period's beta.

[00:41:33] And we find the same thing. So it can't be because of market behavior. That can't be the reason these things underperformed. It has to be something about the funds themselves. There are, I think, three candidates to explain the underperformance in our analysis. I don't know how to allocate across the three, but the three, the first is, you mentioned it before, buying options is expensive.

[00:42:02] This is, you know, if you buy that put, you are going to be on the wrong side of the volatility risk premium in terms of expected returns. The second is that there are frictions involved in buying these options. You can't get them for free. And then the third is, these funds also cost something. For a fund that makes one trade a year, that is well-defined, they are surprisingly expensive.

[00:42:30] It's some combination of these three of these. And the two add up over time because options expire. And that's really why I think that's the only explanation I can come up with for why these things are underperforming to the extent and to the consistency that they have. Yeah. And again, I keep, I come back to this idea of the vol surface, right? It's the way in which the market prices options, both across time and then across strike.

[00:42:57] And if we think about this quote skew across strike, you know, Black Shoals will tell us there's no such thing as the skew, right? That would be a flat line in their world. Normally distributed and the market just doesn't buy it and it shouldn't, right? It's a, it's a, it makes sense for the skew to be priced the way it is because of 1987, right? You know, we have just bigger down moves than we have up moves, escalator up, elevator down.

[00:43:26] That's the equity market as they, as they say. And so that surface, I think, and again, I studied this. That's what I, you know, what I focus on. It wiggles around. You know, there are times when that skew is especially steep and there are other times and, you know, 2022 is a super interesting one because we had that tightening cycle coming. Everybody was sort of saying, okay, the fed's got a lot to do.

[00:43:54] And most of the like prime broker data would say people were really under allocated to equities that year. So the market went down a lot and people kind of saw it coming and they didn't have a lot of allocation. And so what happened to this vol skew was it flattened a tremendous amount. In other words, the difference between the call vol and the put vol got really small.

[00:44:20] And so it almost made, it did, it made the economics of the collar strategy a lot better, actually. And which was, you know, it's a rare thing to see it like that. So, you know, again, I come back to some part of the underperformance here as a function of the shape of this skew. If it was always shaped the way it was for that time period, which was kind of a six month time period in 2022, I feel like the economics would be better because the collar would be

[00:44:50] a better deal. You know, the market would be providing a better deal. In a black shoals world, you're not paying up for the collar at all. Yeah. Right. So that's a super interesting one. This might be one of the differences between how someone like you might pursue a strategy like this versus how the industry writ large does. The industry writ large, fully rules based. Every year I buy, I wait a year, rebalance.

[00:45:16] Whereas you would be, you know, looking more at the underlined macro fundamentals and pricing and then making the call based on the opportunity set. These are set it and forget it. And I think that that's probably a reason that a lot of folks who trade options for a living are befuddled for why we found the results that we do. Well, I think your point at the start of the conversation is very right, which is, I think,

[00:45:44] to make this a product, you've got to provide some systematic rules based approach. I think that makes sense rather than, you know, having someone wake up and decide whether they think the options are rich or cheap. You've got a chart on your second piece, Buffer Madness, which is looking at the peak to trough drawdown improvement for the buffered category. And you're looking at 16 different periods.

[00:46:14] Walk us through that. I mean, that's a staggering chart. So this was part of our response. The responses I collected from the first piece and one that I totally agree with was you guys put together very different categories into a single analysis. That's not going to tell you much about, you know, any of the underlying. And so what we did in the second piece, we said, good point.

[00:46:43] You know, we're hoping to keep that first one short, but given there's interest, let's do a longer paper. And so one of the exhibits was we said, okay, let's take a look at each of these categories separately. So the covered call guys, you know, the derivative income, the buffered guys, so the defined outcome, and then those options trading equity hedge strategies. And we did that sort of two by two analysis for each of them. So, you know, better returns and better drawdowns.

[00:47:11] When we split it apart, buffered funds didn't look quite as bad as they did in the original analysis in the following sense. Fewer of them landed in the worst quartile when we split out the universe than when it was all together. A good number, 38% of them, made it into the top left quartile, which is the better drawdown, but worse returns.

[00:47:41] And that was kind of an interesting thing. And this is sort of your intuition that you mentioned earlier, Dean, is you thought that, you know, if markets were efficient, we wouldn't expect so many in the worst, worse. You would expect more, you know, one for twos, a trade-off kind of a thing. And so we found more move into a trade-off where you had better downside protection, but worse average returns, worse cumulative returns. And so we just wanted to drill down into that.

[00:48:09] So those 38% that kind of, you know, graduated to that one for two quadrant, we said, let's take a look at those 16 funds and get a sense of the magnitude. What was the trade-off? What was the trade-off in terms of improved drawdown versus cumulative loss in returns? And that's the exhibit that you're referring to. So those 16 funds, we just organized them. We laid them out by least drawdown protection to most drawdown protection.

[00:48:35] So the first, oh boy, the first 10, they protected the peak to trough drawdown by less than 1%. So we are talking, you know, you have to take it out to a couple of decimal points to say, yeah, these guys helped you on the drawdown. But what you found was more striking is that on average, you lost 7% just being in these things. So even though you've made, you know, 50 basis points in peak to trough drawdown improvement

[00:49:00] for that one bad drawdown, you paid for it pretty dearly over the longer term. And that's what we wanted to do there. We want to take a look at, you know, and ask, you know, for investors, was that trade-off worth it? I haven't met one who said yes yet. And is this daily data? What's the periodicity? Yeah. So what we did in the original analysis, we had five years. So we said, let's just look at monthly data. That's how most kind of academic studies are done.

[00:49:26] For this particular exhibit and the ones around it that look at a shorter period, you know, January to March or January to May, you don't want to use monthly data because you get four observations. So there we broke it down into weekly data just so we could have some more, you know, some more data points in case maybe something fancy, amazing happened intramonth. We want to capture that. We also look at daily data as well. In the defined outcome bucket and the buffered bucket, that bucket,

[00:49:57] is it always the put spread? Is it sometimes just a put or is it always a put spread? The vast majority of the time, it is the put spread. You are exposed to downside after you've kind of reached that buffer threshold. I think something like 4.7% of these funds by AUM have a 100% protected downside.

[00:50:24] In other words, you can think of as collars. So the market, I think that investors know that protecting your downside fully is extremely expensive. And I think that's why you see so few assets in those particular forms of these funds. And, you know, 95 plus percent in ones where, you know, after you've lost a certain amount in the S&P, then you are participating one for one with the drawdown. Yeah.

[00:50:55] And, you know, one of the ways I like to think about the put spread collar, right? So the long put spread short call is you've bought a slightly out of the money put. Maybe it's 3%, 5% of the money. And you've basically funded that with a strangle. You've sold maybe a 20% out of the money put. And now we've got to find the call that, you know, kind of solves for the whole thing. If you just bought the put and didn't sell the downside to form the put spread,

[00:51:23] the call strike would be so close to the money. It would be very difficult. And I think that's kind of a big part of it. And, you know, as I sort of think through our drawdowns, we've got, of course, COVID, which I want to say was 30 odd percent at one point, very fast. Yeah. And obviously snap back to all the intervention from the government. This last one got to 19% in April.

[00:51:50] And so these put spreads are roughly 15% wide. Could be a 95, 80, you know, put spread. If you had just the put and you went through those COVID lows down 30 odd percent, I got to hope that would have protected you at least somewhat. But because you had the put spread, your protection, yeah, it got you something, but you were well through it as you commented, right? You're only getting a slice of the protection.

[00:52:19] I think one of the misperceptions or misconceptions with these strategies, you know, you can draw sort of the payoff diagram. That only holds for the one specific observation from the start of when you buy that option to when it expires, if it's a European option. It doesn't tell you about anything in between.

[00:52:45] And I think that's part of the reason, like, how can it be that these things have breached the, you know, that sort of protected area or done even worse than the downside participation area? It's because unless you happen to look only at inception, go to sleep for a year and take a look exactly a year later, you are going to realize returns very different than what you may have thought you had signed up for.

[00:53:12] One of the, I'll call it innovations, that the buffered fund industry has seen over the past few years is so-called laddered funds. The idea for these is, okay, well, you're going to have some path dependency if you're only buying and rebalancing in January or in March or in June. So instead, let's come up with, think of it as a fund of funds that allocates one twelfth

[00:53:39] to each one of these monthly kind of series. And that should kind of give you a little bit more consistent protection. I think there, the fiction of returns looking at anything like a simple payoff diagram evaporates very quickly. But that is something that we have seen more investors go to because I think in their mind, more consistent protection is a good thing.

[00:54:06] Our hope in publishing stuff like this is for people to see like, ah, it's really promised as opposed to realized protection. It might seem good in theory, but in practice, it doesn't really hold up. You had mentioned this before, just in terms of these funds are running a business and they've got lots of fixed costs associated with that business. And so they're clearly charging an assets under management fee.

[00:54:32] Did you get a sense as to how significant a driver of the underperformance that might be? That's a great question. I think the median one of these funds, now just thinking about the defined outcome category, I think the median price is something like 79 basis points per year. That one exhibit that we were looking at that sort of showed the cumulative return difference, that was far greater than 79 basis points.

[00:55:02] So it explains something and maybe for funds that are on the margin, the difference is the management fees. But I think it is only part of what's going on. Another way to see that it's only part of what's going on is you can build these strategies using indexes. So the theoretically free version of it, ones without transactions, costs, and management fees. And there you find underperformance too.

[00:55:31] One of the things you do in your second piece is you list the areas of feedback, some of which I would think are a little bit more relevant than others. You'd said one you pushed back on was really this is coming from the capital that I'm putting towards this strategy is really a fixed income allocation. And you take some time to really push back on that.

[00:55:57] I thought that was a thoughtful way of pushing back. I'd love for you to talk a little bit about that. Yeah. And of all of the feedback that we got from the first piece, this was the one that was most surprising, or this was the one that taught me the most. We did our original analysis evaluating these strategies against equities. The reference asset for all of these is equities.

[00:56:26] So to us, it made sense to ask, okay, well, are you better off investing in this versus equities? Specifically, just having less inequities. What I didn't appreciate and only did with the second was, no, Dan, that's not how people are going into this. They're actually getting out of bonds, or maybe they're sitting on cash. And this is their way of kind of inching their way back into the market.

[00:56:52] The accusation was the analysis was flawed because the benchmark was wrong. And so to your point, what we did in that second blog was we had to be very clear. There is a difference between adding value relative to a benchmark and the opportunity cost of what you were funding from. You could allocate to a buffered fund from anything. You could sell real estate. You could sell crypto. You could sell cash. You could sell bonds.

[00:57:23] In some sense, what you sell doesn't really tell you if the buffered fund was worth the fees. You could have picked anything here and made the comparison. What you want to know, you'd ask two questions. Would you want to kind of sell out of starting asset for something in stocks? And then are these better than that thing in stocks?

[00:57:49] It's really two questions that people need to disentangle when evaluating, like, should I get out of bonds to go into this? The first question is, should I get out of bonds to have exposure to the equity risk premium? And then if the answer is yes, then it's, okay, are these better than the equity risk premium? It's really a two-stage. The second tells you whether or not the benchmark is right. The first tells you whether or not it's kind of the opportunity cost, whether you're making the right choice from an asset allocation standpoint.

[00:58:17] And as you received lots and lots of feedback, what are some of the areas of feedback that you got that you really did think were, you know, areas for consideration? As I said, you know, these fact tests, you could set two really, really bright folks in a room with the same data sets and ask them to try to solve the same problem. And they're probably going to come up with different answers because they're going to make some simplifying assumptions

[00:58:46] in the construction of their analysis and their back tests, which could lead to different outcomes, right? We see that all the time in the literature. What are some of the areas of feedback that you found, you know, thoughtful and got you thinking? All of the feedback, the vast majority of the feedback we got was very easy to respond to or reply to with more analysis, more detailed analysis, more thoughtful analysis, more data, less data, looking at specific windows.

[00:59:16] And those are pretty easy to do. And robustness too, you know, looking out of sample versus in sample for beta, using daily and weekly, as opposed to only monthly returns. I think the two hardest to address, one was this whole notion of funding cost. And it wasn't hard to answer for why, you know, bonds are not the right benchmark for these. But I think what made it harder was the realization that,

[00:59:45] or the implication, that some of these strategies are being marketed to investors as a bond substitute. Yes, they are. They are not bonds. They have nothing to do with bonds. They are not a bond substitute. Or folks selling bonds for something that seems safe, they're just getting more equity risk with a lot of bells and whistles in the form of options. So that was something that really opened my eyes to what's going on in the industry. The second bit of feedback, which I have to admit,

[01:00:11] I was unable to answer with numbers, was that's not the point of these things. The point of these things is that they help investors sleep better at night. There's no test. There's no analysis I can do to show that that's not true. And so what I did was I came up with a theory for how can it be

[01:00:40] that investors are able to sleep better at night for something that has shown no evidence at being better in terms of returns, nor risk. And that was the placebo effect. I think people sleep better at night with these because they are told that these will help them sleep better at night, as opposed to because these things actually help you in terms of risk mitigation. That was the trickiest feedback to come up with.

[01:01:08] I hope my response wasn't too snarky in saying that this is really a placebo effect. The placebo effect, for those of you who either don't watch hospital dramas or went to medical school, so the placebo is sort of classically the sugar pill. It is a treatment given as a control that has no therapeutic benefits, but just to see, you say, hey, this is the thing and see if the patient gets better. The placebo effect is a phenomenon

[01:01:37] in which the patient claims, and in some cases has been shown to get better, even when given a placebo. And I think that that is a reasonable explanation for the popularity of these funds. They are told that it will help them, even though empirically it has not. Maybe that's what keeps people from panic selling. Panic selling is always bad. Well, I can't say always bad. Generally bad.

[01:02:05] But offering a placebo in order to mitigate that risk, doesn't seem like a worthwhile for investors. I think one of the developments in, especially on the ETF front, where increasingly these funds have some embedded derivative is that there are now single stocks with ETFs with embedded derivatives.

[01:02:32] NVIDIA has got an income fund. I think Tesla might have one. And so those are super interesting also because very easy to test as well. And they're not hundreds of billions, but they're meaningful enough. Maybe that's your next thing to test. From a purely academic standpoint, having a broader cross-section to look at is great. I would think,

[01:03:00] maybe we can catch up in five years when I get to do the analysis on individual securities and single stocks. But to the extent hedging out sort of undiversifiable risk, like the S&P 500 is costly. You might expect the cost of hedging out a pretty diversifiable risk, like an idiosyncratic stock risk, might be a little bit cheaper. If that's the case, then maybe there's less of a headwind in that particular sense

[01:03:30] for single securities. Now there's all sorts of other problems with having a large portion of your wealth in a single security. But that might be for another day. I will mention one more thing, and that is a bit of feedback that we have not yet addressed, but we are in the process of. So this is sort of a sneak preview. Someone said, you know, blogs are great, but let's see an actual academic study. So we are working on one now

[01:03:59] and are hoping to submit it to a journal next month. So hopefully, you know, stay tuned. You will get to see more of the cross-section in the analysis. We'll be making a handful of changes and a bunch more exhibits to really kind of dive deep into what we have found in this part of the industry. Excellent. Well, that's a great way to cap off a really interesting conversation. Dan, I appreciate you taking the time

[01:04:29] to come back on the podcast and discuss this work. It was great to catch up. It's been a lot of fun, Dean. Thanks so much. You've been listening to The Alpha Exchange. If you've enjoyed the show, please do tell a friend. And before we leave, I wanted to invite you to drop us some feedback. As we aim to utilize these conversations to contribute to the investment community's understanding of risk, your input is valuable and provides direction on where we should focus. Please email us at feedback

[01:04:57] at alphaexchangepodcast.com. Thanks again and catch you next time.