Why Merger Arb Works

In August of 2020, Paul recessed a beginner boarded a flight in the Middle East for a speaking event in the African country of Burundi. He never made it there. Paul was a human rights hero who saved hundreds of lives during the 1994 Rwandan genocide and later became the subject of a blockbuster movie. That flight set an emotion a stranger than fiction story of an adacious kidnapping, espionage, and a massive global campaign to free Paul from captivity. I'm Robbie Grammer and this is after Hotel Rwanda, a podcast from Foreign Policy. Listen in the slate presents Feed Now. Hey guys, before we begin, a lot of the information in today's episode comes from a 2001 paper, characteristics of a risk and return in risk arbitrage by our colleagues Mark Mitchell and Todd Polvino. And now on to the show. This is AQR's the curious investor, a show that breaks down some of the most important ideas and finance to help us make better investment decisions. I'm Dan Villalan. And I'm Gabe Fagali. A few decades ago, a handful of managers were profiting from a relatively unknown strategy, merger arbitrage. Today, we explore what it is and why it tends to work in the long run. And we're joined by one of the guys who demystified the strategy for the rest of us. My name is Mark Mitchell. I'm a principal CNH partners, which is an affiliate of AQR Capital. In the late 80s, Mark was working at the SEC and he was interested in mergers. These were happening a lot in markets, but no one had really studied them in depth. A merger is simply where two companies combine their operations. Generally, you have an acquire, which is a larger firm than you have the target, which is of course the smaller firm. Mark initially was just curious about how mergers affected the economy. I was trying to understand how, when and why mergers create value. Merger arbitrage at the time was just sort of a sideline research project for me. I did a lot of research on this trying to understand variation in the size of the merger market and what we were able to link it to was industry shocks. Just positive negative shocks industries that cause this change in terms of firm size. Mergers are a way to facilitate that quick movement towards a new optimal firm structure. But the more he dug, the more interesting the world of merger arbitrage became. From his desk at the SEC in the 80s, Mark wanted to figure out what was going on, but it wasn't an easy thing to learn about. Right after I started there, when the crisis of 87 occurred and started talking to merger arbitrage jurors, I was just trying to understand like how do they make money. And there was no data on this at all. There's no books on it. There's no articles on it. Fast forward to the University of Chicago where Mark was now teaching. He got a call from Todd Polvino, a professor at Northwestern University. And Todd reached out to me. It was in the fall of 1997 and said, "Hey, you're doing all this work on merger arbitrage. What do you think about us getting together?" And my response was an instant, yes. Todd had been helping a hedge fund to fund pick managers for their portfolio. In particular, he went out and interviewed merger arbitrage managers and tried to understand their investment process. And so the question I was trying to answer in the same question Todd was trying to answer is, "What is the risk of return to the strategy?" Before we get into the data, let's take a step back and describe how merger arbitrage works. First, cash deals. In a cash merger, you typically have one firm buying another with cash. The acquirer buys the target. Let's say the target stock is currently trading at $15 a share. And there's a merger announced that the target is going to be acquired for $20 a share or months from today. What will happen is when the market opens, that stock price will increase rapidly up close to $20. Let's say it goes up to $19.50. So a merger arbitrage manager would buy shares in that target firm and hope to collect $0.50 the difference between a traded price of $0.50 and the deal price of $0.20. If the deal fails, the price might drop back down to $15 so the merger arbitrage would lose $4.50. But if the deal closes, they make $0.50 or about 2.5% on the initial investment. Now 2.5% doesn't sound like much, but that's not an annualized number. This was just one deal that took four months. Over the next eight months, there could be lots of other deals a merger arbit manager would participate in. Cash deals aren't the only way mergers happen. An acquirer can use their own stock to buy the target instead. Now doing that has some advantages, particularly if it's a merger in the same industry. It's a bit of a hedge that is if the industry goes up, both companies go up with it. You don't have to reprice the deal per se. It's a natural hedge from a risk management point of view. In a stock deal, the merger arbitrage or buys shares of the target and hedges their risk by selling shares of the acquirer. So that's the theory. Now let's get to the data. So you mentioned that there wasn't much data. How did you go about even getting the data to look at? Well, in terms of much data, I would say there was actually, I'd clarify that there was probably no data. Mark and Todd had to actually build a data set from scratch, which meant rolling up their sleeves or rather their research assistant sleeves. And so I then took the effort of hiring a team, a massive team, over a multiyear period of research assistance. And we collected data on every single merger that happened going back to 1962. And this was the 90s. There weren't exactly web scraping algorithms. So this literally took years. At some points, Mark had up to 10 research assistants just pulling and checking and cross checking data. We'd read every single page of the Wall Street Journal index, which would publish every headline in the Wall Street Journal. And so we'd look for articles about mergers. Then once we would find the article, we would actually go to using microfilm or microfish, go to the actual article and pull out the deal terms. Then we'd go to the proxy follies to confirm the deal terms, etc. And Mark wanted to make sure they captured everything. You worry about what if the research assistant fell asleep, you know, page 400. So we'd have two RAs reading each index. Because you had to make certain you collect every deal, especially every failed deal because failed deals drive the results. Then they went even further. They looked at every single daily stock price move that was unusually large compared to history up or down. Because once in a while, you find one that wasn't mentioned in the newspaper index. So this took lots of people, a long time and a lot of money. But eventually, you know, after about four years of this, we had a large data set that we could then set up a pseudo hedge fund, if you will. A merge arbitrage your and get a sense of what the risk and return pay off was to the strategy. You might think new data, the academic community must have been on pins and needles. Well, not exactly. My colleagues didn't think it was the wisest move. Because here was their assumption. Was it? You're not going to find anything. They were skeptical because this wasn't a strategy that seemed to take a lot of market risk. Since arbitrage by definition is about being long one security and short a related one, it means you tend to be pretty market neutral. That's why marks colleagues assumed it wouldn't have high average returns. But Mark and Todd expected a different result. One reason was they'd seen how the strategy struggled when equity markets were doing poorly. If you go through the peers like the crash, etc and suffer all that pain, you would think you're going to end up with more than a risk free rate of return on average. But we had no way of knowing the answer to that until you actually built the data sets. By late 1998, Mark and Todd had their answer. We knew within a year of analyzing the data that we had something which we would view as fairly material and substantial. When a merger is announced, the price of the target firm usually shoots up toward the acquisition price. But it doesn't get all the way there. That remaining amount is called the spread, which varies from deal to deal. You get tight spreads when investors are confident that the merger will happen. You get wide spreads when investors are a little more skeptical. Mark and Todd's research found that spreads were historically pretty accurate. The active merger arbitrage users do a great job in understanding deal risk and setting the spread appropriately. So that was a finding that from academics, everybody would say, "Yeah, that makes sense." From a street perspective, I think the street was probably a little surprised by that finding that is the spread is right after the deal is announced. And again, it's right on average. And so by that, I mean risky deals have wide spreads, safe deals have tight spreads, etc. But Mark and Todd's research also uncovered something a bit more surprising. The other thing we found, in which I think is what our colleagues and academics, academia really liked a lot was we found this non-linear payoff in terms of the risk and return. Most investors are familiar with linear payoffs. For example, let's say you hold a basket of stocks. If the market goes up or down by 10%, you probably expect your portfolio to do the same. There's a linear relationship between market returns and portfolio returns. Because in this case it's one to one, we'd say the market sensitivity or beta is one. Mark found that merger arbitrage had a non-linear profile. Sometimes it's beta is zero, but not always. The most states of the world, a merger arbitrage generates this sort of flat return. And it's invariant with respect to the beta. But during severely declining markets, the fine-as-equity markets, when the SMB 500s down and say 5% a morning given a month, suddenly the merger or beta goes from zero up to say 0.5 or so. The next headline from their paper was probably the biggest. For the first time, someone was able to quantify and explain the reasons behind merger arbitrage's profits. We were also able to document that there's this excess return to the strategy, excess of the risk-free rate and excess of the loading on beta during severely declining markets. And that's what we found very material and important because there we felt like you have this market which is efficient. And yet you can capture this excess risk premium which we call the liquidity premium. Here's how to think about the liquidity premium. Suppose you own shares in a company that's trading at $25. And news comes out that it's going to be acquired for $75. It's open and the price shoots up. Not all the way to $75 but let's say to $73.50. Well, you've just made a lot of money. So what's left on the table? About 2%. If you think about $73.50 versus $75. Are you going to wait it out for next year 2%? No way. You're going to sell. And so we think of that selling pressure that is all your natural holders when it deals announced they no longer want the whole security. They want the exit. That's the whole reason arbitrage exists is that we provide liquidity to those investors. We buy the shares. We hope to collect. I mean, they tripled their money. We're trying to make 2%. So there's a reason for them to exit the security. There's a reason for us to buy the security, right? And at the end of the day what happens is we then transfer our shares to the actual acquire. And a merger is announced that same day a lot of the target companies shares get traded. It's not unusual for the majority of existing shareholders to sell and take their gains that same day. Even if the merger isn't happening for another three or four months. When you're selling of that magnitude, that results in a lot of selling pressure, right? So it's that selling pressure, what we call price pressure, that we as arbitrages, we step in. We provide liquidity, right? To take those shares from those natural holders, hold those shares for the next three or four months and incel those shares to the actual acquire, right? So we're providing liquidity and we charge a price for that. We call that price that we charge a liquidity premium. And this premium is in addition to both the risk free rate and the beta that the strategy sometimes has. It's a different source of expected returns. And that makes it interesting to more than just academics. It makes it interesting to a lot of actual investors. You could even build a pretty good strategy based on this. If you take the basic results from the paper that the spread's efficient, then that result would suggest to invest in every deal and build a well-diversified portfolio. In fact, one of the arps here reached out to us, you know, proposed that we think of in terms of building an index. But there would be a number of things you'd have to consider before building an index based on a strategy like merger, first of all, how do you wait each deal? You could think about it as an equal way to index, for example. And we did that in the paper. You could think about value-weighted index. You know, the issue with a value-weighted index, for example, would be you could have one merger deal, which could be 40% of the market, or 50% of the market, right? So suddenly you have a lot of idiosyncratic risk with respect to your portfolio. If you think about an equal way to portfolio, you know, your median firm is very small. Your median target is very small. You'd invest in investing in a lot of small targets, and you start pushing prices, etc. And in fact, you would be demanding liquidity, if you will. All these implementation details aside, Mark and Todd's paper allowed investors to think about merger arbor in a new way. The basic concept was, if you build a well-diversified portfolio, then you should be able to capture this premium. And in practice, we took it a few steps further that is, we try to think about implementation of this strategy, and then you want to create the most efficient portfolio as possible to capture this liquidity premium. One way to be more efficient is to diversify not just across a large number of mergers, but to also diversify those mergers by risk. And remember, merger deals have binary outcomes. Either they go through or they fail. So traditional measures of risk aren't enough. If it weren't to fail, what would happen to the target stock price and what would happen to the choir stock price? So based on that, you can imagine if we thought if a deal were to fail, and the stock price were to drop a lot upon failure, then we would put a low weight in that deal versus a deal which were to fail, and the stock price were not to drop as much. Unfortunately, estimating the consequences of deal failure isn't easy. It's a difficult process. And it requires a lot of quantitative analysis. We've done a lot of historical research on this, but it also requires a lot of analysis in real time with a lot of qualitative data just to get a goods in what's going on in the industry, et cetera. What's going on with the management team, what's going on with the legal structure, just to have a sense of if the deal fails, what happens to that share price. Mark and Todd's research also looked at the effect of different economic environments on merger arbitrage. Take an expansionary period for the economy. Generally, there are more merger deals in those times, and more merger deals means more opportunity for diversification. That can mean less impact to your portfolio if any one deal fails. But that's not necessarily the ideal time to be a merger arbitrage or. You can also think about during an expansionary period, perhaps risk premiums not as high. And if you have lots of deals invested in, that we all can be very diversified as merger arbitrageurs, right? You may think that spreads are not as attractive in, which is the case. That they're attractive, in the sense of, there's lots of deals, but the spreads aren't necessarily wide. Nero spreads generally imply less risk, but also lower expected returns. Now, in best case scenario for merger arbitrage, giant deals. Think 50 billion dollars or more. Here, spreads might be wider than you'd expect, even during economic expansions. Those mega mergers can be such that we as arbitrageurs are too small to keep that spread efficient. And so those spreads actually may be wider. So that's the best of all worlds where there's lots of deals and lots of mega mergers where spreads are excessively wide. And there's a bit of temporary misprice, and just because as arb's, we can't keep that spread efficient. Spreads might also be temporarily wide when systematic risk is high. And this might provide another time for a merger arbitrage manager to add value. And that world spreads will widen a lot. They reflect that they added risk that deals may be more likely to fail, but there we will like to crank up the exposure in more relatively safe deals because their spreads will have widened as well. Simply because of the arbitrage community taking money off the table from a risk management perspective. Gabe, what do you say we wrapped this episode up? I don't know if I can say arbitrage any more times today. Sounds good. Okay, to me, the big story here is that some hedge fund strategies like merger arb have been figured out. They can be explained in terms of how, why, and when they were. But figured out doesn't mean merger arb is easy. It's an alternative strategy and that means design choices can really matter. So that's it for this week's episode. Join us next time where we'll explore how active managers generate excess returns. This time in the world of fixed income. People look at this and say, "Ooh, it's easy. Fixed income must somehow be easier because every man in his dog can build a portfolio of the beats." What you see in the benchmark. Until then, thanks for tuning in to the curious investor. We'll see you next time, folks. The views expressed in this recording are the personal views of participants as of the date indicated and do not necessarily reflect the views of a QR itself. We can contain in this recording constitutes investment, legal, tax or other advice and it should not be viewed as a current or past recommendation or solicitation of an offer to buy or sell any securities or to adopt any investment strategy. Information in this recording is based on current and historical market conditions, which will fluctuate and may be superseded by subsequent market events or for other reasons. AQR does not assume any duty to update forward-looking statements. The information in this recording has been developed internally and or obtained from sources believed to be reliable. However, no representation or warranty, Expressor implied, is made or given by, or on behalf of, AQR as the accuracy and completeness or fairness of the information contained in this recording. Any liability as a result of this recording, including any direct, indirect, special, or consequential loss or damage, is expressly disglaimed. Copyright 2019 AQR Capital LLC. All rights reserved.