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
Decoding the Dynamics of Hedge Fund Arbitrage: Unveiling Merger, Distressed, and Event-Driven Strategies (S1 E17)
Unlock the secrets of hedge fund arbitrage strategies with me, Edward Finley, as we navigate through the thrilling intersections of finance and probability. Prepare to have your perceptions challenged as we dissect the complex mechanics behind the high-octane world of merger arbitrage, distressed arbitrage, and event-driven arbitrage. Through this episode, you're bound to gain a deeper understanding of how hedge fund managers play the odds in a market where knowledge of financial intricacies, legal frameworks, and regulatory landscapes can make or break multi-million-dollar deals.
Venture further into the labyrinth of hedge fund tactics as we scrutinize the enigmatic performance of distressed arbitrage strategies and their deceptive volatility. We'll dissect this high-risk investment game, revealing how it thrives on the brink of bankruptcy, drawing in astute investors with the allure of high returns amidst the smoke and mirrors of market fluctuations. With an analytical lens, we'll break down the statistics that lay bare the true nature of these strategies, from their negative skew to the high kurtosis that defines their risk profile.
Finally, we cast a spotlight on the allure of multi-strategy funds as we probe the delicate balance they strike in pursuit of portfolio diversification and efficiency. The episode culminates with a critical analysis of their performance, peeling back the layers to determine whether these funds truly deliver on their promise of attractive returns for the savvy investor. Join me for this riveting journey through the risk-laden terrain of hedge funds, and arm yourself with the insights needed to navigate these complex investment strategies.
Episode Notes: https://1drv.ms/p/s!AqjfuX3WVgp8uwOODPqoWCYdwAFz
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 sometime professor at the University of Virginia and a veteran Wall Street investor, and you're listening to Not Another Investment podcast.
Speaker 1: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 really making our way through hedge fund strategies now. They're complex, as promised, but, I think, also really interesting, serve very interesting purposes both in financial markets as well as in portfolios, but also are the kind of thing that when we read the papers and we talk to people about hedge funds, we know so very little about it, and I hope that in these episodes, you're able to come with some appreciation for both the complexity and the good uses they might be put to. Last time, we started our evaluation of the hedge fund arbitrage strategy with those that primarily arbitrage fixed income securities, namely convertible arbitrage and fixed income arbitrage. Today, we'll continue the arbitrage strategies, but focusing now on those that primarily arbitrage equity securities. They merger arbitrage, distressed arbitrage and event driven arbitrage, and event driven is sometimes called special situations. We'll finish off this episode by considering a very popular hedge fund strategy among individual investors, namely multi strategy funds. These are designed as a portfolio of hedge funds that avoid the highs and lows of market movements, while it's promised delivering attractive returns, and we'll take a look to see whether we think that really lives up to its promise. As a reminder, arbitrage strategies generally are those in which managers trade in securities where the pricing of risk is especially challenged, sometimes, as we saw, for behavioral reasons, sometimes due to their complexity and sometimes due to uncertainty. But in all events, managers trading the assets that are especially difficult to price risk hope to earn returns without taking much market risk. And so we turn now to the equity, so-called equity strategies.
Speaker 1:First up, merger ARB. So what does a merger arbitrage manager do? Well, in general, it might be better to call this strategy risk arbitrage, because its outcome depends on correctly forecasting the success of a proposed merger. The manager here takes a position on the spread between the bid and the market price of a target company in a merger. So let's unpack that for a second the bid and the market price. Well, the target company is the one that is being targeted for acquisition, usually by another company, and they will have made an offer at a certain price, and that price will be different than the current market price. And so what this type of manager does is he or she takes a position on the spread, on the difference between the bid and the market price of a target company in a merger, and they'll do this by a mixture of financial analysis, company intelligence and legal and regulatory advice. And so I point this out here, because this is kind of the first time that we really considered a hedge fund strategy using something other than pure financial logic.
Speaker 1:Here it's as important maybe some would argue more important to understand the legal and regulatory framework around a proposed merger as it is to understand the fundamentals of the bid and the way the market is trading. The greater the risk of failure of the merger, the greater the spread will be. In other words, if a company is currently trading at 100X per share and someone comes in and offers to buy that company for 110X per share, the share price will trade at a difference. It will be less than the offer price by an amount. That represents how risky it is that people think that that merger will go through. After all, they're offering 110X per share. If that's a 10% premium, that sounds pretty great. But if you don't think the merger is going to happen either, because the company won't get its financing, or because regulators won't permit it to go through, or because they can't agree to some legal terms. The market is trying to discount that uncertainty, so the stock may actually trade down. By how much it goes down will be dependent on how much risk the market perceives in a successful deal. The manager, as a result, is going to want to enter into a transaction where the spread between the bid and the current market price is wide enough so that when the deal goes through, that's what the manager is counting on. They think they know better than the market. When the deal goes through, they will earn that spread.
Speaker 1:Historically, managers in this space entered into transactions only after a deal was announced, and that's because the speculation and the risk of being accused of insider trading were considered too great to take a chance in trades like this. And increasingly competitive pressure to earn returns has led managers in this space to enter the trade soon before it's announced. They tried to get a little bit ahead of the trade. Managers in this strategy tend to increase their positions slowly over time. They don't put on a position all at once, nor take it off. It's sort of a build-up and a build-down and, depending on the spread.
Speaker 1:It usually means that the manager will be long the target company Remember that's the one that's trading at a discount because the market's unsure if the deal will go through. So long the target because you think it's going to go up, and short the acquirer. And short the acquirer because in general the acquirer's stock tends to go down when a merger is completed. This price goes up at first. People are very positive and optimistic about this possible acquisition, but maybe a little too optimistic, and it tends to go down. So short the acquirer. But it's a lot more complicated than that. That might be the basic trade, but at times, during the course of the manager's exposure to this target company, the trade might actually be the opposite.
Speaker 1:Trade Markets may, in the interim, move in a direction that causes the manager to short the target and long the acquirer. The trade is highly asymmetric and remember that just means it's not like a normal distribution where we have a bell curve. It's highly asymmetric. There's a high probability of a small gain, that is, the transaction occurs and the target stock trades up to the bid price, but there's a low probability of a large loss. The transaction fails and you're long this company that is now trading well below what you paid for it. As of Q3 2023, we saw merger arbitrage strategies account for a reasonably large percentage of the market. It occurs at about $91 billion, which is about 2% or 3% of the entire hedge fund universe. So what are the trade dynamics when a manager is engaged in this type of strategy? Well, in those transactions which are cash only, that is, the acquiring company is going to pay all cash for the target company the manager is going to be long the target and short some amount of systematic risk. So that's very different trade than what I described a moment ago.
Speaker 1:In a stock transaction, whether it's all stock or partially stock, the manager will be long the target and short the acquire and, given the asymmetric returns, there's usually going to be a fair bit of leverage, either explicit leverage through borrowing or just the regular ordinary use of leverage through the use of options. Managers in this space tend to focus on large cap companies. It needs to be sufficient liquidity to support the shorts and there also needs to be a market for options. And there's always going to be some directional equity risk in these trades, since bull markets make deals more likely to happen and bear markets make deals less likely to happen. So, if you're trading on the likely outcome of a successful merger and the market overall takes a nosedive, that's going to make it less likely the deal will go through, whereas, conversely, if the market is going through a great uptick, that makes the deal more likely to happen. So there's always some directional equity risk here.
Speaker 1:Well, as we've done in the past, what do we expect to see based on these trade dynamics and based on the general view? Well, we expect to see that this is going to be a strategy that's non-normal. This is going to be a strategy which has either some skew or some fat tails or both, and that's because, as we said, these are highly asymmetric trades. You're either right or you're not right, and if you're right, you've got a lot of leverage and that's going to help you make a decent return. But that's a pretty large probability of a small gain. But you're also could be wrong and there's a small probability of a large loss, all of which will be amplified by leverage. We also would expect to see that the strategy is going to earn most, if not some great part of their return from those non-normal risks and, as I mentioned just a moment ago, we would expect that in general, there to be some directional exposure to equity risk.
Speaker 1:So let's see what the data tells us. This again is from 1997 through 2023. And we start with the regression results. And the regression results tell us that our traditional risk factors equity risk, interest rate risk, credit risk and high yield credit risk our traditional factors explain only about a third of the strategy's returns. There are still drivers that are not captured in our factors. We have an excess return a return not explained by those risks of about 4% out of the total of nearly 6%. There's statistically significant coefficients for each of the factors, but economically they're not quite as significant as one might expect. So, for example, the equity risk factor has a correlation coefficient of 10%, and remember that just means if equities increase by 1%, the strategy's returns will increase by 10 basis points 1 tenth of 1%. So that's very little movement for a 1% equity movement. The equity correlation coefficient is 10%, the interest rate is 8%, investment grade is 12% and high yield is 12%. So our traditional risk factors don't really give us much to go on. We are certainly earning a lot of our return here from other sorts of factors.
Speaker 1:Let's turn our attention to the summary statistics to see if it gives us any further insights. The average annual return here was a little shy of 6%, with volatility of 4.6%. Now that gives us a very strong information ratio of 1.29. That's about double the information ratio of owning equities. So that's awfully impressive. But if we look at the correlation to US equities and the beta to US equities, we see some evidence of this non-normality. We see that correlation is about 57%. It's certainly directional, but the beta is only 15%. That means something is artificially depressing the volatility of the strategy relative to the volatility of equities.
Speaker 1:We also see a skew of negative 2. That's quite large. That's quite large. And we see a kurtosis of 16. Again, that's about eight times larger than equity risk. So lots more tails and skewed to the negative return bucket.
Speaker 1:And if we look at the histogram we see that most of the monthly returns are centered around, say, 4 tenths of 1% to 1.4% per month. But a significant number, more than half as much, is negative 0.6% to positive 0.4%. We also see the tails. So that's the skew part. We also see the tails. The worst month in the strategy was somewhere between negative 8.6 and negative 9.6. That's the worst month. The two best months were between 4.4 and 5.4%. So again we have these tail events, but they're far deeper. That's both evidence of some skew and also evidence that we have outliers In our up-down volatility ratio. What we find is that in general, there's negative skew because more of the volatility of the strategy is to the downside than it is to the upside. All of which is to say that we've got lots of evidence to tell us that volatility does not fully account for the risk in this strategy. So when we look at an adjusted information ratio of 1.29, we should take it with a grain of salt. That volatility number is probably artificially depressed.
Speaker 1:When we turn our attention to state-dependent returns, we find a similar story. There's higher correlation with equity returns in the worst months than there is in the best months. So there's 63% correlation in the worst months, 41% correlation in the best months. That suggests to us a sort of nonlinear exposure. Moreover, it's got that telltale smiley face shape, which is to say that what we're looking at here is convexity, that is, as returns go up, the amount by which the returns of the strategy go up accelerate, and when the returns start going down, the amount by which they go down likewise starts to accelerate. The participation rate in the strategy also shows similar evidence of convexity Only 6% participation in the worst months of equity returns, 20% participation in the best month of equity returns and it shows how managers are able to limit much of the market losses that the market experiences from being experienced in the strategy.
Speaker 1:There doesn't seem to be any asymmetry to the sensitivity of returns to interest rates. We're at 9% in the worst months and negative 8% in the best months. It's pretty much suggesting that there might be some directional exposure to interest rates, and that shouldn't be too surprising because a lot of companies are going to be financing the acquisition of a target company and therefore, if interest rates go up, that will mean that it becomes more expensive, more difficult, to accomplish the transaction. The participation rate in the 10-year return is pretty much a similar story to the participation rate in equity returns. It's very, very low in the worst months of the 10-year returns. It's likewise very low in the best month of the 10-year returns and where its highest is in the middle quintile, and so it suggests that maybe what we're looking at here is some convexity to interest rate risk. But when we consider it in the aggregate we think that it looks rather independent. Correlations at every quintile are very near zero. So it's, on balance, I think, some sensitivity to interest rate risk, but not a ton.
Speaker 1:Let's turn our attention, then, to distressed arbitrage strategies. So what are they? Well, in some sense or another it's kind of the mirror image of merger arbitrage. Merger arbitrage suggests that a company is buying another company, and usually that suggests a lot of health Health in the acquirer, a healthy business in the target. Distressed, however, is where managers invest in securities of companies that are in or about to enter or exit bankruptcy. They're in real trouble. To give you some sense, the bonds of these kinds of companies are usually rated below triple C, so they're really in some meaningful distress. But these managers are investing because they have the expectation that the market is over-estimating the risk of failure or, to put it another way, they think the company will recover.
Speaker 1:So managers in this kind of strategy can invest in a wide array of securities. They can buy the equities of the firm, they can buy bonds, they could buy trade claims, they could buy receivables. But typically, if a manager is trading in a firm's credits whether that's bonds or trade claims or receivables the manager is doing that with the expectation of being able to influence the outcome in a bankruptcy or restructuring, and so that's really akin to being an equity owner right, because if you are a debt holder, you typically don't have those kinds of rights in terms of what the company is doing, but in bankruptcy you do have those rights. And so for investors in this strategy, while they might be investing in the firm's credits, that is really a way of investing in the control of the firm, which we would typically think of as equity risk.
Speaker 1:In addition to the uncertainty of recovery, there's a great deal of illiquidity to this strategy. The firm's equities may be delisted, there may be legal constraints on the securities and how they might trade, there may not be analyst coverage of these securities if they haven't yet been delisted, and likewise, there are going to be legal and other metrics that you have to look at in order to assess, to evaluate, the likelihood of this company coming out successfully. You know, typical valuation methods for equities is something called the discounted cash flow model. It's basically a way of saying well, how much cash flow do I think this company will deliver over time and therefore, this is how much I would be willing to pay for that cash flow. Those valuation methods all assume that a company is a going concern. Those valuation methods go out the window when the option is that the firm may go bankrupt.
Speaker 1:In addition, there are going to be certain contract rights of other claimants that might have an effect on your ability to control the way that this company exits its distress. There's going to be a really long process. Normal trades, like shorts, become very, very difficult in a strategy like this where the process may take upwards of a year or more. So it's important to see how, when we say generically arbitrage strategies, arbitrage, mispricings of risk because of behavioral reasons or the difficulty in valuing the risk, et cetera, you can see that all in play here. But again, like merger arbitrage, we have non-finance specialists who are involved in making this decision. There are going to be bankruptcy lawyers, there are going to be securities lawyers, contract lawyers, all of whom are going to have something important to say about the things that will affect the likely outcome in the company's distress, all of which will be crucially important to the manager's trade.
Speaker 1:There are three ways generally that a manager can execute on a distressed arbitrage strategy. The first is through control. Managers acquire a sufficient amount of the company's equity to control the company or they acquire a sufficient amount of debt in order to participate directly in the management of the company during bankruptcy. But in either case, all because it means the manager will have influence on how long the recovery takes and how to allocate and maximize the resources to benefit the security owners. It's as simple as that. These are not altruists, these are not white knights looking to save a company. There are investors who say I think this company will recover, I think it can be a going concern, I think I can make a profit based on how it's trading today. But, crucially, I need to be in control in order to make my strategy work. The second possible way to execute on this strategy, little different than control, through the trading of preferred stock. If the manager acquires a sufficient amount of preferred stock, then they can play an active role in the reorganization. They just won't have any control over the management of the company. So they can affect the timing of the reorganization, because preferred shareholders are given some preference in bankruptcy, but they won't be able to affect management. So, depending on the kind of recovery the manager expects, means the manager might tilt toward prefers or might tilt toward control or might tilt toward the third possible way to execute on this strategy, which is debt. Here the managers play no role in the reorganization or the management of the company. They're simply buying the debt in order to earn some profit on its recovery. So three basic ways that managers can execute here.
Speaker 1:The strategy is, by its nature, directional, that is to say there's a long bias. When you're investing in the firm that is in distress, it's very, very difficult to hedge any firm specific risk. That's the risk you're taking. You can try to hedge some systematic risk, but in general, as we've talked about many times before, even firm specific risk is tuned to is affected by the general market risk. In addition, the economic cycle where we are in an economic cycle, plays a large role in the outcomes of distressed investing. If we're in a growth cycle, it makes it more likely the company will recover faster, but there's not a lot of good distressed paper out there to buy. That is to say, the market is generally abolient and the economy is growing. Sure, this company is in distress, but maybe it's not trading at such a big discount, whereas if the economy is in a contraction cycle, it's much more likely that the company will take longer to recover or maybe not recover at all. There's going to be a lot more good distressed paper to buy. For smart investors who are thinking about how to own a piece of this recovery, the amount that you can make will be a lot larger. Distressed investing is really a very, very small part of the overall hedge fund world. It was about $16 billion in assets under management by the third quarter of 2023. That's less than 1% of all hedge fund strategies.
Speaker 1:Well, let's evaluate, then, what we expect to see here. So I think one thing we would expect to see is, again, a lot of non-normality. We would expect to see some really significant negative skew because of the asymmetry and results, right. So, mirror image of merger arbitrage, you have the prospect, you know, sort of say, a low probability, a high probability prospect of a fairly modest return, but a low probability prospect of a total loss if the company doesn't exit bankruptcy. So we would expect there to be a good negative skew there. We would also expect there to be pretty serious tails, pretty serious outliers, and again, that's because it's kind of binary, isn't it? Either you got it right or you didn't. There's no kind of middle ground here. Moreover, because these are companies that are in distress, we probably would expect to see a pretty substantial amount of illiquidity. Illiquidity, certainly in the securities, but also illiquidity, because while the company is in its distress and before it exits, it's unlikely that its value will change much. It will just seem to stick where it's at until the outcome is known ["The Outcome"]. And so let's take a look and see if that plays out in the data.
Speaker 1:So first, again with the regressions, we see that the traditional factors explain a fair bit of the variation in returns of the strategy, accounting for about 40% of the strategy's returns. But to put that another way, that means 60% of the strategy's returns are the product of some other risks. Statistically significant unexplained return of 4% out of the total for the period of 7%. So about 60% is unexplained by the factors. Moreover, all of the factors are statistically significant, but only investment grade credit and high yield credit have any economic significance. Credits correlation coefficient is 30% and high yield is 33%. Still, it's not very big, but it's meaningful. If we turn our attention to the summary statistics, we get some further color about the non-normal risks, the risks not represented by these traditional factors, and the role they play. First I should just mention average return was 7% for this period.
Speaker 1:Volatility of 9%, which delivered a decent information ratio of 0.8. That's about 50% larger than equities generally earn. We see directionality the correlation with equity returns is 61%, but the beta is only 25%. So again, positive directionality, but we know that there's something driving volatility artificially low. Likewise, 10-year treasury correlation and 10-year treasury beta are suggesting that there's an inverse relationship, that that is to say that there's some positive relationship with rates. Remember that the returns on bonds will be the opposite sign of changes in rates. So if rates are going up, bond returns will be going down. And so here we see that the negative correlation suggests that when rates are going up, bond prices are going down. That means that the strategy is positive, it's doing better than it was. There's a pretty significant asymmetric distribution here. The skew is about negative 1.5. And there's another really big tail, a little shy, of seven, and that suggests again that volatility and beta are not good measures of the risks that we have here.
Speaker 1:A look at the Historogram can really tell you the story. It's very, very clear that there's negative skew and most of that negative skew is the result of some large numbers of months one, two, three, five, six, 11 months out of the total number of months when returns in the month were between 3% down and 11% down, whereas on the positive side there were eight months in which returns were up 4% in the month or as much as 7% in the month. So you see, the size of the positive months is lower 7% as opposed to 11% on the downside. And the number of them eight on the upside and 11 on the downside tells us a pretty reasonable story there about non-normality. We can also see the non-normality in the up-down volatility ratio it's 0.6. That tells us that the strategy is much more sensitive to down markets than to up markets and we expect it as much.
Speaker 1:Right when we said that the economic cycle plays a pretty important role here. On the likelihood of a company exiting, exiting successfully. We can also see really strong evidence of illiquidity here. The auto correlation and remember that's just the degree to which the following month's return is likely to be the same or different than the prior month's return as a kind of rough measure of illiquidity. The auto correlation here is about 41%. That's very significant. Equity markets generally have auto correlations below 10%. Indeed, merger arbitrage, which we thought should be a little illiquid, had auto correlation of 11%. So very strong evidence of illiquidity in the strategy, for which you would expect that managers are gonna earn a return. We could confirm the non-normality when we look at the equity state dependent returns, where the correlation in the worst quintile was about 70% and the correlation in the best quintile is around 10%. This is the classic U-shape of the correlation.
Speaker 1:So it's telling us that it's convex, it's non-stationary, it moves as markets are changing, it becomes higher and lower and in particular, it's when markets are in their worst months that this strategy tends to feel that much more potently. There's also some evidence that the strategy's correlation to high yield shows sensitivity to the same down markets. Now, why high yield? Usually we compare its sensitivity to tenure treasury, but here, because we're talking about distress, I felt like a more interesting comparison. And also when we looked at the regression. The regression suggested that high yield is maybe more meaningful a risk than the tenure. To see how that looks. And we see modest, a modest convex relationship to high yield, not nearly as much as the relationship to equity risk, but certainly there doesn't seem to be a whole lot of convexity there because the participation rate really stays pretty stable throughout, with one quintile's exception.
Speaker 1:The next step is we'll talk a little bit about event driven or, as I say, sometimes called special situations. What's going on here? What's the strategy? Well, here managers invest in securities of companies that are engaged in some kind of corporate transaction, some sort of capital structure change. It could be a security issuance or share repurchase, it could be a sale of assets, it could be a division spin-off, it could be an acquisition that's going to result in a new division of the company or some other catalyst that will affect the balance sheet of the company. Now you might be thinking, as I described, that, well, wouldn't merger arbitrage be that kind of event? But merger arbitrage is rather special. It's not really about the change in the capital structure of a company, as much as it is about the combination of two companies and the attendant risks involved. Here we're not focused on that part. We're focused on just things happening to a particular company's balance sheet.
Speaker 1:These managers will focus on both announced transactions as well as situations which pre or post-date their trade, and no formal announcement is expected. That is to say, they just follow certain companies where they make judgments about changes in their capital structure that might be relevant. The managers will employ an investment process that focuses broadly on a wide spectrum of corporate life cycle events, and so, as I mentioned, we'll focus on all of those things that could happen in the life cycle of a company's balance sheet where it will materially impact the firm's capital structure. And they identify those situations how, by following companies and doing fundamental research. These are not quantitative managers who are looking at historical prices or historical trades and then making a judgment about the mispricing or the pricing of some risk.
Speaker 1:The managers here can invest in a wide array of securities Certainly they can, but primarily they invest in equities. And event driven is pretty significant. As of Q3 2023, we're about $242 billion in assets under management and that's about 14% of all hedge fund strategies. To put that into perspective, that's the second largest of all the arbitrage strategies. Fixed income arbitrage is the largest, the second at 973 billion, and event driven is the next largest at 242 billion.
Speaker 1:So what might we expect to see in an event-driven strategy? Well, it's kind of hard to say. I mean intuition tells me that we're mostly investing in equities. We're mostly focused on changes to the capital structure. Who will benefit or be hurt by changes in the capital structure? Answer equity owners. It's unlikely that bond owners are going to be affected by that, and so these managers trading on anticipated corporate events in the life of a company's balance sheet are almost certainly going to be focused on equity risk.
Speaker 1:However, we know that the kinds of transactions that they might be focused on may be those of companies whose shares are less liquid, or those kinds of transactions where it's more complicated to assess the risk or to compute what the risk will really result in in terms of the change of the company, and so we probably would expect to see again negative skew and larger tails. These are going to be rather binary judgments. I think this change to the balance sheet is going to be really good for the company. I'm going to be long. I'm right or I'm wrong. It's impossible to say more than that and so that would suggest that there are going to be some pretty large tails and that skew will be to the left. We also would imagine this going to be a fair bit of illiquidity, because, again, these are companies with that managers trading that are probably not those companies with the largest amount of float. So, general directionality to equity risk, but beta probably not a very good measure. Information ratio not a very good measure because volatility is not a terribly good measure. Why? Because of these non-normal asymmetric risks that lead to the depressed value for volatility. All right, let's test out what we expect and see what the data tells us. This is a different time period than all the rest. The HFRI index for special situations is newer than the other indices, and so here our data is only from 2008 to 2023. Typically, we talk about 1997 to 2023. So that's a meaningful difference, but I think we can still take away some useful insights.
Speaker 1:First, let's start, as we always do, with the regression results. Here, for the first time, we see that the traditional factors explain nearly all the strategies' returns. So there is an intercept. There is an unexplained return of about 80 basis points a year, but it's not statistically significant. That is to say, we can't predict that that is any different than zero, and so we have to say it's not predictable in that respect. So nearly all of the returns, if not all of the returns, can be explained by the traditional risk factors. The traditional risk factors that are the most statistically significant. No surprises equities, investment grade credit and high yield Interest rate really plays no role in terms of the returns for the strategy and, in addition, those three risks are all about equally economically significant. So the Wilshire 5000 coefficient is 26%, investment grade credit 33%, high yield 25%. All in the same neighborhood and all basically moving in the same direction at the same size.
Speaker 1:We can turn our attention to the summary statistics. First. I'll just point out we had on average a 5% return for the strategy with 10% volatility, and that's an information ratio of about 0.5%. That's about the same as equity markets. So it's not great but it's not bad. But we see the evidence of non-normality pretty quickly. The correlation to US large cap equities is about 82%, yet the beta is 43%. So highly directional to equity risk, highly, highly directional. But the volatility is really being understated here. In addition, we have skew of negative 1.3, and we have kurtosis of 7. Again, large negative skew, large kurtosis, large tails. And again just a glimpse at the histogram and we see that the skew and kurtosis represent themselves ably in some of the worst monthly returns. There was one month in the period where returns were negative 14%. The very best month's returns were between 7.8 and 8.8. So all very much suggestive that this is non-normal and that volatility is not a terribly good measure. Likewise, ill liquidity seems present, not quite as peaking as it was in distressed, but here 23% auto correlation, which is really quite serious and suggesting that these are trades where ill liquidity plays some role and for which managers expect to be compensated.
Speaker 1:We can look at the state dependent returns to get an even more interesting sense for any non-normality here and what we see is not terribly surprising. We see that the equity state dependent returns we have much higher correlation in the worst month of equity returns 70% and equally high, or nearly as high, correlation in the best month. That classic smile shape for the correlation chart tells us we have some real convexity. But what's interesting is that the participation rate is rather stable. The participation rate doesn't move terribly much because while we do see the convexity in correlations, we see that the participation rate increases only slightly If we compare it to high yield. And again, why high yield? The answer is because in the case of these strategies, these are typically companies that are smaller, more highly levered and their debt is very likely to be high yield debt. So rather than comparing it to the tenure also, the regressions told us that investment grade and high yield were more meaningful we see a similar story. We see that the correlation is highest in the worst months for high yield, at 80%, and they are at their lowest in the best months. So this suggests not so much convexity as just inverse relationship, a pretty stable inverse relationship, which would be one way to understand or to think about how the manager is able to produce the returns that she produces with the amount of market risk it seems that she takes. But, as I say, there doesn't need to seem to be much convexity here, and the participation rate, with a couple of exceptions, really hovers at around 30 or 40%.
Speaker 1:So, last but not least, we'll finish up our discussion of hedge funds by talking about multi-strategy funds. So what are multi-strategy funds? Well, essentially, these are strategies that design a diversified portfolio of hedge fund strategies. So imagine that some investment professional puts together a portfolio of all the strategies that we've talked about for the last couple of sessions, and why would they do such a thing? Well, their objective would be to significantly lower risk measured as volatility, and it would do so by not putting all of our eggs in one basket, right? So not investing only with the one manager. It would also do so by not investing in only one strategy and, as we saw, these strategies bear very different relationships to equity risk and from each other, and so putting them together may, in a sense, create a more efficient portfolio than trying to pick off these strategies one at a time.
Speaker 1:The manager of a multi-strategy fund has discretion in choosing which strategies to invest in, how much to invest in each. They can allocate them to lots of managers within one strategy, or with lots of managers in multiple strategies. It's really kind of up to them. Those can be in-house, that is to say, the manager uses their own internal managers for each strategy, or it can be outside the firm, that is, it could be third party managers that are put together in this kind of a strategy. The minimum investment in a multi-strategy fund is usually lower than the minimum investment in any individual hedge fund, and so here this has the advantage of providing diversification among managers and styles with a lot less capital than investing in separate managers, making it especially popular for individual high net worth investors who might not have they might not be ultra high net worth, and so they might have trouble allocating to individual hedge fund strategies. Overall, as of Q3 2023, multi-strategy funds accounted for about $689 billion in assets under management, which is quite significant. That's about 14% of all hedge fund strategies, so it's way up there in the scheme of things.
Speaker 1:Well, what should we expect? What should we expect from a portfolio of these different hedge fund strategies? Well, I will tell you that my intuition was that it should produce a more efficient portfolio. But the data really tells us a very different story, and partly that's because, if you recall all of our discussion about hedge funds, by their nature they are dynamic in their risk allocations. It's different than saying equities has this set of characteristics, bonds has this set of characteristics, and if I mix them together in a portfolio, I will get this outcome, because those two risks are not really very dynamic. They tend to move systematically. But hedge fund strategies are all very dynamic and there is very little way to know at any given moment in time whether the strategies are moving together or not together, whether the strategies are owning the same risks or different risks. That is to say, it's hard to know whether it is diversified or not. But when we look at the data, we might expect to see maybe the worst of all worlds, maybe not really getting the better part of any of the strategies, but only accumulating the worst parts of all. What do I mean the worst parts of all, I mean all the non-normal risks, I mean that that is what you might wind up owning, but it's unclear intuitively. So I'm going to tell you the data in a minute, but I don't want you to think I'm sort of guessing by knowing the answer. I'm saying intuitively. I would have been at crossroads here to say what the real look might be.
Speaker 1:Well, first let's start with the regressions. We find that the factors traditional factors explain a fair bit of the variation in returns of the strategy. So the traditional factors account for about 64% of the strategies' returns. That's not all. So obviously they're still about a third of the returns are not accounted for by the factors. So there's some non-normality there. And the factors that have statistically significant correlations are all, but none of them is particularly economically significant. So they all have very strong statistical significance but not much economic significance. Equity risk is 18%, interest rate risk 11%. Investment grade credit 20%. High yield credit 12%. These are all rather small, all contributive and all meaningfully contributive but not terribly determinative. So traditional factors explain a fair bit, but it's a bit all over the map what you get when we turn our attention to the summary statistics. We sort of see a little about why.
Speaker 1:So first, just to put it out there, the annual return was about 4.5%, with volatility of about 7%, delivering an information ratio of 0.6. That's about the same as equity returns. So if volatility is a reasonably good measure of risk, then yeah, this type of strategy does rather well in the scheme of things. But it really probably doesn't. The skew is a meaningful negative 0.8. So that's a moderately negative skew. The kurtosis is still quite high at nearly a 5%. Remember, traditional equity markets usually have kurtosis of around 2% and there's significant illiquidity in the strategy. The auto correlation is about 29%. Up-down volatility ratio likewise is less than 1. So that tells us that more sensitivity to downside risk than to upside risk. And so all of that signals to us that volatility is probably a pretty poor measure of risk. That's confirmed when we look at the equity correlation, which is high directionally 68% equity correlation compared to a negative 14% 10-year correlation. So there's a fair bit of directional equity risk here. So, like most of the equity strategies we saw. But the beta is only 24%, so signaling that our measure of risk, volatility, is not terribly, terribly good.
Speaker 1:The state dependent returns further spells out that story. The correlations here don't show much sign of convexity, but they certainly show signs of non-normality. So we've got. When we look at the state dependent returns on equities, we see that the highest correlation is in the worst quintile and the lowest correlation is in the best quintile, and so, as a consequence, we note that the participation rates how much does the strategy earn as a percentage of what that market did at that moment is fairly restricted on the downside, but also equally restricted on the upside. So we see a lot of the sort of notion of the equity-long short risk where it's highly collared but we don't see any of that convexity for which we might earn returns. And then we can likewise look at the correlation and the participation to the tenure, and here we see that the correlation just hovers around zero. It's essentially independent of interest rate risk. The participation rate changes, but it changes in kind of uninformative ways, suggesting to me that we don't see any more of that convexity risk that we might have seen in some strategies.
Speaker 1:And so, on balance, I think that the second thing I predicted is what really is evident in the data. I think that you earn fairly modest returns under 5%. You can't look at the volatility as telling you whether that's good or bad, because the volatility is rather ineffective. We've got pretty significant tail risk, pretty significant negative skew, very large directional exposure to equity risk, and that directional exposure to equity risk tends to be the most peaking during down months. And so, again, when you put it together, the multi-strategy approach, while it might sound like it would be a very, very good way to invest in hedge funds, seems to me a little bit fraught with owning most of the non-normal risks but not really quite enough of the traditional risks to make it the strategy payoff.
Speaker 1:Well, I'll stop there. That concludes our discussion of hedge funds. I hope that you found this interesting and enlightening, but, like everything on this podcast, it's 10 miles wide and about an inch deep, so no one's going to be an expert on hedge funds, but I hope you've got a better insight into what they're all about. When we come back next time, we're going to move on to our last asset class, and that is real assets. Thanks again for listening and I look forward to talking to you next time.
.jpeg)