Not Another Investment Podcast
Understand investing beyond the headlines with Edward Finley, sometime Professor of Finance at the University of Virginia and veteran Wall Street investor.
Not Another Investment Podcast
Demystifying Hedge Funds: Introduction and Dynamic Market Strategies (S1 E14)
Unlock the enigma of hedge funds with me, Edward Finley, as we explore these complicated (and controversial) investment strategies. We will see that they are not really an asset class but instead are a universe of diverse trading strategies, from directional plays to arbitrage opportunities. In this episode, we dissect the perception and reality of these complex instruments, developing a taxonomy of hedge fund categories based on their myriad risks. We navigate the murky waters of data biases that can obscure the true performance of hedge fund strategies, and we lay the groundwork for a deeper comprehension of the pivotal alternative risks employed by hedge fund managers.
We will look at the universe of securities that a manager trades in, the types of trades she engages, as well as the leverage and illiquidity in order to categorize hedge funds by their risks. We will learn that common risks, like equity risk and interest rate risk, play a relatively small roll in hedge fund performance. Instead, hedge funds deliver a mix of alternative risks that will help us categorize them into three main categories: Dynamic Market Strategies, Absolute Return Strategies, and Arbitrage Strategies.
We will start our exploration of hedge funds with the largest category, Dynamic Market Strategies, including equity long-short and macro.
The largest hedge fund strategy, equity long-short, is an intricate web of systematic risk and nonlinear relationships to broader market returns that are the result of how these managers trade. Through examples, we clarify the often-misunderstood mechanisms of gross and net exposure, leverage, and beta-adjusted market exposure, providing a masterclass on how equity long-short funds balance the scales of risk and reward.
Finally, our analysis of Dynamic Market Strategies is not complete without a foray into the performance and intricacies of macro trading strategies. We unearth insights into the performance macro strategies by revisiting George Soros's legendary bet against the British pound, offering a thrilling case study of macroeconomic speculation and its potential for astonishing profits. This tactic serves as a testament to the high-octane intellectual analysis that underpins successful macro trading but also the huge uncertainties in making those bets.
You will leave this episode enriched with a profound understanding of the complex machinations of these two hedge fund strategies. Future episodes will explore the strategies that comprise the other two categories.
Episode Slides: https://1drv.ms/p/s!AqjfuX3WVgp8ukFS322alTE4tvCt
Thanks for listening! Please be sure to review the podcast or send your comments to me by email at info@not-another-investment-podcast.com. And tell your friends!
Hi, I'm Edward Finley, a Sum Time Professor at the University of Virginia and a Veteran Wall Street Investor, and you're listening to Not Another Investment Podcast. Here we explore topics and markets and investing that every educated person should understand to be a good citizen. Welcome to the podcast. I'm Edward Finley. Well, we're back to asset classes this week and we're going to tackle a big one and that is hedge funds. Now, hedge funds are tricky because they put to work some really interesting concepts and make them kind of a little complex to understand, but they're also tricky because the hedge fund world is very opaque, and that's always been true and I think always will be true. We're going to, in this episode, understand hedge funds broadly and introduce a taxonomy of hedge funds, and so, instead of a fancy word, we're going to introduce a set of labels that will help us think about hedge funds in the terms that they have common risks, in the same way that we have equities they're all different kinds of equities, but they all share a set of common risks. We have bonds they're all different kinds of bonds, but they share some common risks we can think about concretely. It's going to be the same for hedge funds. We're going to build up a taxonomy around it to help us understand that. We'll talk a little bit about hedge funds at a very high level and then we'll go into the first strategy component in our taxonomy, namely dynamic market risk. We'll talk about the two hedge fund strategies in that taxonomy, namely directional, long, short and global macro. Then in future episodes we'll continue with the taxonomy, looking at the different strategies in each. When we look at the strategies in each of these taxonomies, what we're going to do is build our thinking up from the ground. We're going to start by thinking about what kinds of trades in general does a manager in that type of strategy deploy. When we think about those kinds of trades, we're going to then form an intuition about the kinds of risks we expect to see in those trades. Then, after we've built up that intuition about the kinds of risks we expect to see in a strategy based on the trades, we're going to look at the empirical data to see if it supports the theory that we've come up with or not Without further ado.
Speaker 1:Hedge funds First. I think I'm going to maybe surprise you, but hedge funds isn't an asset class. We shouldn't think of it as an asset class Hedge funds literally, if you take it literally, is just a kind of investment vehicle. It describes a kind of investment vehicle in which managers have freedom to do certain things that they're not able to do in other investment vehicles. It's a way of pooling investor assets. There's different fee structures. There's really no limit to the kinds of strategies they can follow. When we say hedge funds, there's no way we can do an evaluation of the risks, because there isn't any one type of risk that we're talking about in hedge funds.
Speaker 1:If I were to give a proper definition of what a hedge fund is, if it's not an asset class, what is it? I would say the one common theme across all hedge funds is that it's a way in which investors can pool their money and access a manager who trades in securities, both traditional securities like stocks and bonds, but also more exotic securities like derivatives, using long positions and short positions in order to gain a mixed exposure to some of the common risks. We've already talked about stock returns and bond returns, but also some other risks that get compensated but that aren't stock and bond returns. I'll repeat it just to make sure you're with me Hedge funds are not an asset class because they don't have a common set of risks. That's what we're going to build up in our taxonomy. Instead, what are hedge funds? But just a pooled investment vehicle that allow investors to access a manager who trades in securities both traditional securities as well as derivatives using long positions and short positions, in order to gain a mixed exposure to common risks like equity risk and interest rate risk, as well as alternative risks.
Speaker 1:How big is the hedge fund world? When we look at headlines, it seems like it's pretty big, and we hear people throwing around numbers in the billions. In the US, as of the second quarter of 2023, a little more than $5 trillion was invested across all hedge fund strategies. While that might seem large, it's relatively small compared to say, how much capital is invested in the US bond market $49 trillion or in the US equity market $45.5 trillion. Hedge funds themselves are really quite a small part of the capital market. What kinds of risk exposures do we have in hedge funds?
Speaker 1:Before we can get to that answer, we first have to talk about some of the data that we have available and its limitations. Any data that we might use to try to identify the risk exposures in hedge funds is going to be susceptible to at least three pretty big problems. Problem number one is something called backfill bias. Most of the vendors that keep data and from which you can get access data on hedge fund returns allow funds to choose when they begin reporting their returns. If you're a fund manager, you're only going to start reporting your returns in the year that your returns are good. This is why we call it backfill bias.
Speaker 1:Second problem is something called extinction bias. A fund that does very poorly and therefore delists or closes is usually no longer reported. Most of the database vendors will exclude those funds from the data. That in turn means that you don't really measure the aggregate return of all of the funds in that strategy, because if the real losers have dropped out, you're artificially inflating the return numbers. Third big problem is something called missing return bias Very successful firms. It's sort of the mirror image of the backfill bias, if you think about it, but very successful firms don't report all their earlier data. You might have a strategy that's doing great, but when you look back the first three or four years the returns were crappy and you tell yourself oh well, we were just getting started and there were all sorts of very weird, idiosyncratic reasons, but when you report your returns. You just don't report those earlier years. You only start reporting the later years.
Speaker 1:Recent academic research has exposed the real difficulties of using traditional risk metrics to evaluate hedge funds. It's really much more complicated to evaluate hedge funds than it is to evaluate equity returns or bond returns, or the returns of an equity manager or the returns of a bond manager. That has mostly to do with the limitations of the data that I just described, but it also has to do with the fact that hedge funds have exposure to risks other than equity risk or bond risk by virtue of how they trade. In order to really assess the risks that are present in a hedge fund strategy, we've kind of got to follow a multi-step analysis. We first have to identify the universe of securities in which the manager trades in order to understand whether she has exposure to what traditional risks Thank you. If she trades only in bonds, then that manager has some exposure to traditional bond risk, but likely little or no exposure to traditional equity risk. Second, we've got to look next at the trading style of the manager in order to identify whether there are going to be some nonlinear risks that are present.
Speaker 1:What do I mean nonlinear risks? Well, think back to the episode in which we talked about how securities trade, when I'm short or long, or when I invest on margin, or when I invest using derivatives. We talked about how the return function is asymmetric. That is, if I say sell options, my maximum gain is going to be the premium I earn, but my maximum loss might be the full price. Likewise, when I'm short a strategy, then I'm going to necessarily have very large upside gains, but I also face very large downside gains because it's a levered strategy. In both cases there's asymmetry and nonlinearity in the returns that I earn.
Speaker 1:Next, we have to look at the trading strategy in order to just get a sense for what kinds of nonnormal risks the strategy is going to be exposed to. Finally, we use a combination of linear regression, which we talked about earlier, and that's just, with all of the dots, what's the best fitted line? But because of these nonlinear risks, a linear regression alone is not going to help us, because in many cases the strategy may just be nonlinear. We'll also want to look at traditional risk benchmarks and we'll want to look at what are called state dependent returns. State dependent returns just means not looking at things year by year, but looking at things in different contexts. How did the strategy do in the worst months of equity returns? How did the strategy do in the best months of equity returns? By looking at those state dependent behaviors, we can get a better sense for the risks that the manager owns.
Speaker 1:Okay, well, when you do all of that, when you conduct an analysis of hedge funds, first by identifying the universe of securities, second by identifying the trading style and what nonlinear risks are present, and then, third, using a combination of regression and traditional benchmarks and state dependent analysis, the result is that hedge fund strategy strategies can be broadly characterized by their exposure to four different risks. Four different risks. The first one is stuff that we've already talked about traditional, systematic risks. Those are the sensitivity of a strategy to changes in equity returns or changes in interest rates, and it can be even more minute. It can be that the strategy has exposure to the size risk factor or it can have exposure to value in the equity space. These traditional, systematic risks are the ones that we're thinking about. The second kind of risk a hedge fund strategy can have exposure to are nonlinear risks. This is the byproduct of how trades are constructed in each strategy, and it may include one or more of the effects of volatility options, something called convexity and binary outcomes.
Speaker 1:Just in a nutshell, what are we talking about? Volatility tends to make the measures of beta, that is, how much equity risk do I have? How much bond risk do I have? Volatility tends to make beta measures an inaccurate way to evaluate traditional risks. For example, in hedge funds there are so-called momentum and convergence trades. Well, those trades have very high auto correlation. Again, recall back to our discussion of probability theory and statistics. What a high auto correlation in returns means is that if one month the return is up, then the next month it's more likely to be up. That's a high auto correlation. A high auto correlation means that volatility does a very bad job of measuring. The dispersion of return does a bad job of measuring risk. Other kinds of trades are like trades in which you sell volatility or you sell a put on volatility. You act like an insurer against volatility. That tends to result in negative skew and higher tail risk. Remember, if we have negative skew and higher tail risk, it means a normal distribution does not describe the strategy. If a normal distribution doesn't describe the strategy, then volatility is going to do a bad job of measuring risk.
Speaker 1:First, nonlinear risk that we care about is that the nature of the trades might render volatility less useful than it was when we were thinking about stocks and bonds. Second is that there is going to be option risk. Typically in a hedge fund strategy, those are those asymmetric payoffs that I just described. The asymmetric payoffs of options make both volatility and correlation very misleading, since it doesn't account for the different result when the underlying security is up or down. That asymmetry will render volatility and correlation very misleading. Third is something called convexity. Convexity is complex, but for our purposes it's easy to understand it as just relating to the way that the options traded create asymmetry. Some securities or pairs of trades and securities will themselves create changing returns over time. That is, it's nonlinear. It sort of bends, gets higher faster and then gets higher more slowly. This is called convexity. That again has to do with the nature of the trades, but convexity means that volatility and correlation are again very misleading in terms of understanding risks.
Speaker 1:Then last are binary outcomes. Some hedge fund strategies have binary outcomes. You're right or you're wrong. When that's the case, you're going to have very large tails, very large right tails. When you're right, you do a great job, and very large left tails. When you're wrong, you're really wrong. Of course, these large tails will mean that volatility is again a very misleading measure of risk.
Speaker 1:To recap that, hedge fund strategies might have some exposure to traditional, systematic risks, and we'd want to know what that is. They also might have exposure to some nonlinear risks, and those will be the result of either the way in which those trades make volatility a bad measure of risk. Those are sort of momentum trades, convergence trades, selling volatility. It might have something to do with option trading, where it's asymmetric and therefore we see that both volatility and correlation are misleading. It may do with convexity, because the return, the payoffs, don't follow a straight line. They change rate, binary outcomes, large tails, all of which tend to make volatility and often correlation poor measures of risk. Risk one traditional, systematic risk. Risk two nonlinear risks. Risk three leverage.
Speaker 1:Now, leverage in hedge fund strategies can merely be the result of their short positions. Remember our discussion of shorts. You are borrowing a stock, selling it, waiting for the price to go down, buying it at the lower market price and returning the stock to the lender, but of course that's leverage. You've borrowed stock and so leverage, as we know, will increase returns in both directions. But leverage can also be the result of derivatives. It can be the result of trading in higher beta securities, but all of them have in common the fact that leverage will increase tail risk, but unlike borrowing in the conventional sense, unlike literally borrowing money and investing it, unlike margins, eh, short sales and derivatives and beta will all likewise have asymmetric payoffs, which means that the leverage will have the result of being much larger tail risks. But those tail risks might be skewed to the bad side, not the good side. And lastly, for its illiquidity, there are gonna be some hedge fund strategies where the trades are themselves illiquid or because the trades find no market for unwinding during a time of distress. In either case, either because the securities themselves are illiquid or because the securities become very illiquid during times of distress, those are also gonna create much larger tail risk, making a very important contribution to hedge fund returns.
Speaker 1:So what can hedge fund returns consist of? Systematic, traditional, systematic risks, non-linear risks, leverage and illiquidity. All of those are risks that you find in hedge funds, that you wouldn't find in stocks and bonds. ["hedge Funds"]. ["hedge Funds"]. Okay, well, if we understand the universe of risks in that way, it means that we can really then look at the mix of those risks and categorize hedge funds broadly into three styles.
Speaker 1:One style I fashion dynamic market strategies. So, generally speaking, these strategies are gonna have some exposure to traditional equity risk. They're gonna have some exposure to directional views of the general economy or directional views of specific securities, and these strategies are gonna seek out securities whose prices aren't consistent with the view and that provides them the opportunity to earn higher returns per unit of risk. All right. So dynamic market strategies, generally speaking, take some directional view on the general economy or specific securities and they trade them in a way that gives rise to an expectation of earning higher returns for a given level of volatility risk, meaning they would expect their strategies to have lower beta or lower volatility. The two strategies that I would put in this category are equity, long short, or sometimes called equity hedge strategies, and macro strategies.
Speaker 1:The second style category I fashion as absolute return strategies. What's going on here? Well, generally these strategies seek to identify anomalies in market pricing and then they trade on those anomalies, allowing them to earn consistent positive returns regardless of what the current economic or market conditions are, and that should allow them to take very little exposure to traditional systematic risks like equity risk or interest rate risk. Okay, so again, absolute return strategies generally are identifying anomalies in market pricing that allow them to trade those anomalies and earn consistent returns regardless of the economic or market conditions. The two strategies that we'll talk about there in a future episode are equity, market neutral strategies and managed futures strategies. And the third category is what I fashion arbitrage strategies. What are they?
Speaker 1:Generally, these strategies are operating in markets or with securities where the price of risk is very challenging to compute. Sometimes that's for behavioral reasons, sometimes it's because of complexity in the marketplace, but sometimes it's just due to plain old uncertainty. So these strategies are operating in markets or with securities where the price of risk is really challenging to measure correctly, either for behavioral reasons, complexity, uncertainty or some combination, which leads to the opportunity of owning only non-directional risks and earning the positive returns. Non-directional is just another way of saying non-systematic. So you just don't really own any systematic risk, you own non-directional risk, non-systematic risk, and you earn positive returns. We'll talk about in a future episode five different strategies that fit that category.
Speaker 1:Description fixed income arbitrage, convertible arbitrage, merger arbitrage, distressed and event driven. So with that in hand we've got our basic understanding of hedge funds broad brush. We understand that it is going to be a mix of traditional systematic risks. It's going to be some non-linear risks, it's going to be some leverage and some illiquidity, and we've got our three categories. We're gonna now focus our attention just on the dynamic market strategies and sort of unpack that idea. So, as I said earlier, generally speaking, dynamic market strategies take some directional view on the general economy or on a specific security and then they try to find securities whose prices are not consistent with that view and this provides them with the opportunity, they think, to earn higher return for a given level of risk.
Speaker 1:There are two strategies that I think fall into that general description equity, directional, long short and macro. If you have access to the slides, I've put a slide in there to show you the breakdown of the assets managed in each of these two strategies Equity Long Short and Macro. But let me just sort of give you a quick overview. Generally speaking, equity Long Short is distinguished from traditional investing because the managers use shorts and they use leverage. They don't manage their strategies typically to a benchmark. Instead they aim to earn absolute returns. Their shorts can be a source of return, that is, they've chosen stocks that they think are going to go down in price and they do, in fact, go down in price. So the source of returns can be from shorts, or the shorts can be not a source of positive return but a way in which the manager reduces the amount of traditional systematic risks that she owns. As of the second quarter of 2023, all of the Equity Long Short strategies accounted for a little more than a third of all hedge fund strategies, so it's really quite an important part of the market and it's the lion's share of dynamic market strategies. It's about $2 trillion out of a total of $2.1 trillion, so primarily Equity Long Short in this space.
Speaker 1:What do the trades look like? Remember, we're going to build from the ground up. Well, again, these are very broad generalizations and so it's not necessarily the case that every manager does all of these things, some of these things, and may do versions of these things. But just to give you a flavor, the trades well, the most common and typical trade is what's called a straight long short. What does that mean? It means that a manager will have a certain style. There'll be a value investor, a concentration investor, a momentum investor, a specialty market investor, you name it. They're going to have a style and they're going to use that style, that insight, to select longs stocks they think are going to go up in price and shorts stocks they think are going to go down in price, but they choose them independent from each other.
Speaker 1:The size of the trades and the betas of the securities that they trade will, in the aggregate, determine how much traditional systematic risk they want to own, and, generally, managers will have some view of how much systematic risk they want to own. And here, of course, when we say systematic risk, we're really just talking about equity risk. That amount of equity risk that the manager owns, though, will time vary. It's why this category I call dynamic market risk, because the managers are not just choosing a certain level of market risk, they're also increasing and decreasing that level of market risk based on their views of the market. The returns in these trades will depend on the beta-adjusted market exposure. So how much equity risk is there on leverage? So how much leverage is embedded in their strategy and skill?
Speaker 1:A second kind of trade that you might see in equity long short is called share class arbitrage. It's just like straight longs and shorts, except the manager is choosing longs and shorts in different share classes of the same company, and that's because they perceive a mispricing in the shares of the different share classes in that security. Third is another very common strategy, and these are called pair trades. Now, pair trades are going to sound similar to straight long shorts, but they're a little different. In the case of pair trades, a manager will select securities that they think are highly correlated with each other. They're either in the same industry or the same sector, or they historically co-move very positively together, and then they're going to determine the price of each of those securities at any given moment in line with historical norms. And so they're going to then belong some stocks that they think are underpriced the historical one relative to the other stock and they're going to short the stock that they think is overpriced, again relative to the other stock, not relative to the whole market or some fundamental valuation. They're going to size the trades they're going to make, eat the longs and the shorts of a given size, consistent with how much overall equity market risk they want to own. And in the end, the returns on these pair trades are more a function of the relative strength of the securities one to the other than anything more related to the market overall.
Speaker 1:Let me give you an example of what I mean by that. So let's say we have a manager who has a fund of 1000X in assets under management and that manager strategy is simple. It's one trade that manager is 1200X long Walmart, and 1000X short Albertsons. And, depending on where you live, these are very similar types of stores. They're not the same but they're similar. Walmart is a big box store, it's national, they've got a large internet presence and they serve a particular price point in the economy. Albertsons is more grocery related. They are regional, not entirely national. They have a smaller internet presence but again, they operate in a very similar type of space. Okay, so we've got 1200X long Walmart, 1000x short Albertson.
Speaker 1:Well, we need to know a little bit more information to understand how this manager's risks are worked out. So, first of all, we could ask what the manager's gross exposure is. Well, the gross exposure is really just adding up the longs and shorts and comparing it to the size of the fund. And so if I add up the two exposures, 1200x and 1000X, the total is 2200X and the funds management size, asset size is 1000X. So that's 2.2 times gross exposure. Well, what that means is is that this strategy is highly levered. It's 2.2 times levered.
Speaker 1:We can also look at the net exposure to understand how much equity risk we might expect the manager to have. Well, the net exposure here is 20%. That's just the 1200X long minus the 1000 short, leaves 200 long. And then I take 200 long and I divide it by the overall long bias of one and I get a 20% exposure to equity risk. But of course that tells me how much equity risk in these stocks I have, but it doesn't tell me how much systematic equity risk I have, because each stock will have a different beta to the market. So Walmart's beta last time I checked was 0.53. And remember that just means the extent to which volatility and co-movement will describe the sensitivity of a stock's price to the overall market. So Walmart's equity beta was about 0.5 and Albertson's beta was 0.24. Well, that means I can beta adjust that exposure, that risk exposure. I can beta adjust it by saying well, I have 1200X of 0.53 beta minus one. That's the other trade 1000X at 0.24 beta. And if I multiply those out and subtract them then I find that the beta adjusted risk is 39.6%. So not 20%, practically double.
Speaker 1:And so when I look at a fund like that, and I think of it from those perspective. That's how we want to unpack the idea of how an equity long short manager is going to be setting up trades where there's some directional view in markets, but they're doing pair trades here with, say, in my case, walmart and Albertson's. Another thing that a manager like this might do is they might sell puts and calls, and the reason they might do that is because their target price for each of those stocks may take some time to achieve. That's particularly true if the manager's style is a value style. It may take a little while for the market price to get where they think it's going to go. But the manager might not be able to sustain the strategy of being short, matching short, for as long as it might take for the prices to get to where she thinks they're going to go. And so the manager might want to catalyze the trade artificially by selling covered puts for the short end of the trade and by selling covered calls on the long end of the train.
Speaker 1:So let's take an example, a different example. Let's say that our manager is long Microsoft, where the trade, the price is currently at $309 a share and their target is $315. Their short Intel, which currently is $55 a share, but they think it's going to be $45 a share. What they might do is they might sell a six month call at their target price on Microsoft and collect a $3 premium. And if the price goes above $315 by maturity then the manager covers and mitigates the loss of the higher price with the premium right, because if that was their short and if the price goes up, then they're going to have a problem with that. If the price is less than 315 by maturity, then the manager is going to cover and mitigate losses down to a price of 312. 312 is just the difference of the price point minus the premium. The manager might also sell a six month put at $45 their target price for Intel, and collect a premium of nearly $10. So if the price of Intel, which they're short, goes down below $45, the manager will cover at their target price and then they'll mitigate any loss on the option with the lower price premium. Likewise, if the price of their short goes above $45 by maturity, the manager is going to cover the trade and mitigate the losses up to that price of $55. Why? Well, because the losses up to $55 are covered by the premium that they sold. If that didn't make perfect sense to you, don't worry about it.
Speaker 1:The real takeaway here is that when executing a direction along short strategy, there are a lot of moving parts. There's the part of the strategy that involves the security itself and its degree of equity risk. There's the size of each trade. There's leverage that's component built into it. And then there's also the nature of selling, puts and calls in order to catalyze trades in a certain time horizon. But we put it all together.
Speaker 1:What kinds of risks would we expect to see in equity long short strategies? Well, in general, we would expect there to be some positive exposure to equity systematic risk, and that's because the nature of these trades is that they tend to be they tend to be what's called long bias. The equity long short manager has a long bias. They have some exposure to systematic risk in equities. We also would expect, therefore, the strategy to have some correlation with equity risk. If it has some exposure to equity risk, we would imagine it will have some correlation with equity risk. The leverage can be determined by the ratio of the strategy's gross exposure to its AUM we talked about that a minute ago where we can figure out how much leverage there is, and that's going to be an important part of the risk. And then their net exposure doesn't tell us how much systematic risk there is, because it doesn't account for the betas of each position. For that we've got to calculate the beta-adjusted net, as I described earlier. In addition, the manager might be selling puts and calls in order to crystallize the strategy in a certain timeframe, and if that's right, then we would expect there to be some option-like payoffs if they're using puts and calls to catalyze trades. So we would expect equity long short strategies to have some equity risk exposure, some correlation to equities, but to earn returns for other risks, and those other risks will be the product of leverage, the product of net exposure to beta and maybe the product of nonlinear risks, like the option-like payoffs.
Speaker 1:Okay, let's look at the data. Does that make sense? And here I'm going to be using the data collected by the HFRI Institute, that's the Hedge Fund Research Institute, and they publish very broad database on returns for different strategies. The database here is no better than any other and in fact suffers from the same problems I indicated at the beginning of the episode. So what's the data tell us?
Speaker 1:Well, let's just start with the linear regression. Let's just start by seeing the extent to which we think common risk factors find exposure in the strategy. What common risk factors am I regressing? I'm looking at equity risk, tenure, treasury risk, credit risk and high yield risk. Well, it turns out that those risks do a pretty good job of explaining the strategy's returns. It explains about 70% of the strategy's returns, but it doesn't explain 100%. The strategy still earns returns for risks other than those systematic risks. Namely, it earns around 2.3% a year beyond the returns for those risks.
Speaker 1:All of the traditional systematic risks we talked about find some exposure in the average equity long-short manager. The most significant of all is, not surprisingly, equity risk. There are statistically significant exposures to the other risks as well. If we turn our attention, then, to the summary statistics, what we see is that the average returns on an equity long-short strategy were 7.59%, which is lower than what equities earned during that period, which was 9.2%. This is our same 1997 to 2023.
Speaker 1:But there's significantly lower volatility. The average equity long-short manager had volatility of 10.8% and the equity markets had volatility of 16.5%. It's also the case that the average equity long-short manager had an especially high auto correlation, and remember that's going to tell us that there's either some illiquidity or some nonlinear exposure, like momentum, in the strategy. The adjusted information ratio is modestly better than equities it's 0.7 compared to equities 0.56. The correlation of the average equity long-short manager to US equity returns was 83% and the beta was 48%, and the difference between the correlation and the beta is really just telling us that, while there is a large exposure to equity risk, the strategy has much lower volatility. But, as you'll recall, because if we look further down the statistics, because the skew is negative 0.27 and because the kurtosis is 2.5 and because the auto correlation is 18%, it means that volatility doesn't tell us the whole risk picture. And so to compare the average equity long-short manager's beta or adjusted information ratio is really going to be very misleading. So instead, what we can look at is we can look at the state-dependent returns.
Speaker 1:What's the relationship between returns in the average equity long-short strategy to the average returns in equities during different periods? And I break the periods down into quintiles, into sort of five sets, with the worst monthly equity returns being the first quintile and the best monthly equity returns being the fifth quintile. And what we find is that we see a nonlinear relationship to equity returns. It's true that the strategies' worst months are in the same category the first quintile as equities' worst months, and their best months are in the same quintile as the equities' best months. But they're very, very different in terms of what those returns were. The strategy was down 2.5% each month on average in the worst months, but equities were down on average 6% a month in the worst months. The strategy was up 3.5% in the best months, but the equities were up 6.5% in the best months, and so you see a kind of collared relationship to equity returns. I don't lose as much when equities are down. I also don't make as much when equities are up.
Speaker 1:We can look a little bit more closely by seeing is the rate of participation the same? And what we find is the rate of participation is not. The rate of participation is only 40% in the worst months and it steadily grows to about 55% in the best months, and that suggests that the exposure to equity risk is convex, it's not linear. The rate at which it's increasing is increasing as well, and we also see that in those quintiles the correlation of the average equity long short manager to equity returns is highest in the worst quintile and highest in the best quintile and is downward from there. So that's sort of a smile shape, if you will, and that is again further evidence of nonlinear exposure in the relationship between equities and hedge funds, and so it tends to match up.
Speaker 1:Equity. Long short strategies are going to earn returns that in general, are going to be a little bit lower than equity returns, but they might hit our bogie. We can't really evaluate it against volatility, because we see that volatility is going to be unreliable for all of the reasons we mentioned before, and we don't just earn returns for the systematic risk, we earn returns for some of these other risks. The strategy will do very much the same thing that equities do, but it will do it in a way that is less extreme, and the less extreme part of that is maybe what investors might find appealing.
Speaker 1:Let's turn our attention, then, to the second strategy in this category, and that is macro. Well, macro is a very different bird. In the case of macro, typically it's a top down strategy. Top down just means the manager looks at the broad economy or the broad market and draws some conclusions about it, and then moves down to look for specific securities to trade that can give expression to that view. They use statistical analysis of the macroeconomic and market conditions in order to form the thesis. They use financial analysis then to choose the timing and the particular security to trade on the thesis. So it's two versions here. Macro as of the second quarter of 2023, was very small $154 billion in macro strategies. That was about 3% of all hedge fund strategies and, as I mentioned at the beginning, a very small part of the dynamic market category.
Speaker 1:What risk exposures do we expect to see? Well, we expect it to be the strategy to be fully directional. There's no attempt to hedge systematic risk. Fully directional, remember, the manager takes a big view of either markets or the economy as a whole and trades those views, and so it will be very directional. It's similar to something we're gonna talk about in a future episode called Managed Futures, except here the directional exposure is present and it's not gonna be present in the other, and it tends to operate the macro manager tends to operate in slightly less liquid markets.
Speaker 1:A macro manager can trade in all asset classes, all financial instruments and both longs and shorts, so there's no single universe of security like with equity. Long short, macro might trade in bonds, they might trade in stocks, they might trade in derivatives and options. It's really unconstrained in a large way. I can't say much about the trades themselves generically, because how broad the mandate might be the manager? What I can tell you is the manager will construct trades very gradually. The managers don't put on trades very quickly and they use lots of different counterparties to trade in order to mask the thesis that they're trading. If the whole market knew the thesis that they were trading, it might price away any advantage that they have, and that means that sometimes if you discover what a manager's trade is at one particular counterparty, it might look like a bear view, but that could really just be profit taking on a bull position at a different counterparty, and so it's very, very difficult just looking at the trades to make any sense of what the manager's risks might be.
Speaker 1:I'm gonna give an example here, because I think that so far my definition, unlike with equity, long short. It sounds kind of ephemeral. So I wanna give it concreteness, and the example I'm gonna give you is a very, very famous example of a trade that made a hedge fund macro manager, george Soros, both very rich and very famous, and that was a trade around the devaluation of the British pound. All right. So here was the thesis, right? Remember? Macro managers start at a very high level and develop a thesis and then they go down to choose the securities to trade. So what's the thesis? Well, this was back in 1990 and the European monetary system, which was the precursor to today's euro. The European monetary system required member states to maintain currencies with a 2.25% fixed rate parity, meaning currency to currency, and not allow more than 4.5% fluctuation against the dollar. Okay, so the system was a political decision that the countries agreed to 2.25% parity across the currencies and 4.5% max fluctuation against the dollar. What else was happening at that moment? Well, at that moment, german unification was happening.
Speaker 1:The wall came down and Germany was coming together and Soros believed that this would lead to strong economic growth. And we know that when there's strong economic growth, that there will be inflationary pressures. And if there are inflationary pressures, the central bank's gonna raise rates, and so we're gonna expect higher rates in Germany. What was happening at the same time? The UK was just finishing up a recession. Well, when you're just finishing up a recession, you have slower growth, you have slower inflation. And therefore, what should the UK do? When you've got slower growth and lower inflation, they should cut rates to boost the economy. But if the UK cut rates, that would have the effect of weakening its currency outside the parameters of the European monetary system. They'd be in breach of the European monetary system. And so Soros looked at this whole situation and said the UK cannot allow its economy to languish without a cut in rates, but it can't defend its currency from devaluation, which will force it to be in violation of the European monetary system. And so Soros predicted the UK would exit the European monetary system.
Speaker 1:Okay, so there's his thesis, right. And so you see, a macro manager's thesis is not like I think Apple is better than Intel. It's a very convoluted, very, very high level interpretation of events. All right, what were his trades? Well, in general, a devaluation of currencies tends to be good for equities, and that's because exports become cheaper and imports become more expensive and you boost local economic growth. Appreciation of currencies tends to be good for bonds, because if currencies are appreciating, central banks will have to lower rates, and if you lower rates, that's gonna increase the price of bonds.
Speaker 1:Soros had access to large credit lines allowing him to borrow money in UK sterling. If the sterling devalued which is what he predicted would happen and the Deutsche Mark and the French Frank appreciated, which is what he predicted, then he would buy sterling and repay his sterling loan. So this was the trade right. The trade was to sort of borrow in sterling and then use the proceeds to buy Deutsche Marks and Franks. And if he was right and sterling devalued, he would and the German Frank and the French excuse me, the German Deutsche Mark and the French Frank appreciated, he would sell those German Deutsche Marks and French Franks for a profit and he would pay off his loan, which is then going to be lower given the depreciation in the sterling. He had a billion dollars to manage in this trade, one billion dollars. He borrowed seven billion of sterling. So that's his short position and he bought six billion of Deutsche Marks and Franks. That's his long position.
Speaker 1:He also remember I said devaluation of currencies is good for equities, appreciation of currencies is good for bonds. He also used a half a billion dollars to buy UK equities and a half a billion dollars to buy French and German bonds. Okay, so his trades again, he has a billion dollars, but he borrowed seven, buys six in Deutsche Marks and Franks, buys a half a million in UK equity and buys a half a billion in French and German bonds. You can see there's a tremendous amount of leverage in his strategy, right, because he's using seven billion borrowing to fund the long side of his equation. He only has a billion dollars in assets. He has exposure to currencies like the French Frank and the Deutsche Mark. He has exposure to French and German bonds, interest rate risk. He has exposure to UK equities. So you see, there's a lot of different risks that are in place. So what happens? Well, he was right. The sterling depreciated by 10%, the German Deutsche Mark and the French Frank appreciated by 7%, uk equities appreciated by 7% and the German and French bonds appreciated by 3%, and so he made a tremendous profit on this trade.
Speaker 1:But what I hope you can take away from this is macro is one of these strategies where it's extremely complicated to make sense of what we would expect the risks to look like, because each manager is going to be very different and the thesis they develop is going to be very different. But what they have in common, it would seem, is that the strategies tend to be fully directional, there's no attempt to hedge systematic risk, and that they operate in less liquid markets and they construct trades in which we would expect to find lots of non-normal risks. All right, so, looking at the data from HFRI, we can look at macro data and, as before, we'll start with the regressions. And so how well do the traditional systematic risks explain the variation in the strategy's returns? And the answer is not very well. Only 20% of the strategy's returns were explained by systematic risks. That's meaning that there are drivers of return, risks that drive return that are not captured by those factors, and to the tune of 4.4% 80% of the strategy's total returns Only equity exposure is statistically significant and reaches the level of some worthy attention, and so it suggests that macro managers in the aggregate probably are using equity exposure more than they're using other types of exposure. But remember, this is an index of all macro managers who have vastly different, highly idiosyncratic trade ideas, and so when we mush it all together, it shouldn't surprise us terribly much that there's little linear relationship between traditional risks and the strategies' returns, and it probably shouldn't surprise us so much that it's mostly equity risk. But that doesn't mean that that's all there ever is. It may just be that different strategies within the index are canceling each other out.
Speaker 1:What about the summary statistics? So the average returns on the strategy were 5.6%, with low volatility 5.8%, giving a much higher information ratio than equities. Information ratio here was 0.96% versus equities 0.56%. The directionality is evident because the strategies' correlation with equities was about 27%. The beta was 9.6%, but, as we've said, the beta is going to be misleading in hedge fund strategies generally, and it means that the strategy earned much higher returns for its equity risk because it had only about a 10% exposure to equity risk, which would have returned you 88 basis points, and the strategy earned 5.6%.
Speaker 1:There's very little evidence of any directional exposure to interest rate risk. The treasury correlation is only 2.5%. There's some evidence, though, of non-normal risks, so there's an asymmetric distribution of the returns. The skew was a positive 0.6%, with reasonably large tails, 1.2%, and so, in combination, volatility and beta are not going to tell us much about the risk exposures of the average macro manager and, as I mentioned, because this is an index of returns, kurtosis probably very significantly understates the tail risks of any given strategy, because you're going to have some who experience very positive tail risks simultaneously with those that are experiencing very negative tail risks. That cancel each other out, and the aggregate tail risks are a little misleading. So, to get a deeper sense, we can look at those state-dependent returns again.
Speaker 1:So how did the strategy do in the worst equity market months versus the best equity market months? Had the strategy do in the worst interest rate months for, say, the 10-year treasury and the best interest rate months? And what you see here is much more typical for the kind of exposure of risks that you would see in a typical macro manager. It's true that its worst monthly returns were in the quintile where equities had the worst monthly returns, its best monthly returns in the quintile where they had their best monthly returns. But the returns to the strategy in its worst months was a minuscule negative 18 basis points per month on average. So almost zero in the worst equity months and one and a half percent average monthly return in the best equity months, when equities earned 6.5% roughly.
Speaker 1:The returns followed equity returns with rising participation. So we saw that, that convexity, that the participation in the first quintile was only 3% but it rose to 22%, suggesting some convexity. It's also the case that the returns for its, when measured against the returns for the 10-year treasury, there's no discernible pattern. Its best months occurred in the fourth quintile of bond returns. Its worst returns occurred in the third quintile of bond returns, and so there's really no discernible pattern with respect to the 10-year, and likewise we do see some evidence of convexity. The correlations and the participation are growing over time, but in the aggregate what we can see is that the way in which a macro manager earned 5.6% returns is not really so much the product of being really smart, because this is just the average manager they can't all be smart but these positive returns are not really the result of exposure to traditional systematic risks, but instead it's some exposure to traditional systematic risks and then returns for exposures to these non-normal or asymmetric risks.
Speaker 1:Well, that's it for the dynamic market strategy category, so we'll leave it there and we'll come back next time to introduce the second category, which is absolute return strategies. Thanks so much for listening and I look forward to seeing you next time. You've been listening to Not Another Investment Podcast, hosted by me, edward Finland. You can find research links and charts at NotAnotherInvestmentPodcastcom, and don't forget to follow us on your favorite platform and leave comments. Thanks for listening.
.jpeg)