Reputation counts for much in money management--nowhere more so than in the hedge fund industry. Everyone understands this, although exactly how reputation influences investing strategies beyond the anecdotal evidence is poorly understood.
What is known includes the fact that all investing strategies move in and out of favor. Whether it's plain-vanilla value and growth strategies, or merger arbitrage and global macro, leaders and laggards are forever switching places. That's life in the money game. Explaining why that's so, and in a quantitatively/economically satisfying way is something else.
Enter a Yale finance professor and his student, co-authors of a new working paper that proposes a model of so-called reputation concerns among fund managers ("Nickels versus Black Swans: Reputation, Trading Strategies and Asset Prices," by Steven Malliaris and Hongjun Yan, assistant professor of finance, Yale School of Management). The paper unveils an econometric explanation for how the interplay between manager status and investor perception is an agent of change in the shifting fortunes of hedge fund strategies.
Why does this matter? Because a deeper understanding of why capital flows in--and out--of investment strategies, and how those flows affect performance is critical for clear-eyed analysis of managers and their funds.
The paper starts by noting that hedge funds can be grouped into two basic types: nickel strategies and black swan strategies. The former is a reference to picking up nickels in front of a steam roller--a risky strategy, but one that tends to generate modest returns most of the time. But the smooth performance comes with the risk of a blowup. The classic example is the leveraged arbitrage strategy practiced by the infamous Long-Term Capital Management, which earned steady and seemingly reliable gains, right up until it imploded in 1997.
The flip side is the black swan strategy, a term popularized by author Nassim Taleb. In this case, small but recurring losses are typical until an extreme market event that almost no one expects delivers a huge payoff. Buying out-of-the-money puts, for instance, is a black swan strategy that will probably incur small losses until and if the market crumbles.
Malliaris and Yan's research finds that hedge fund managers favor nickel strategies, which hold roughly half of the assets among the major strategy categories. The true black swan strategies, by contrast, have a tiny minority of assets.
Looking deeper into the distinction between nickel and black swan strategies, and how manager reputation fits in is a productive exercise for rethinking conventional asset pricing models. Once upon a time, individual investors were the prevailing force in setting asset prices. That was true in the 1960s, when the capital asset pricing model was invented. Today, institutional investors and money management shops are the dominant players, and Malliaris and Yan incorporate the change into their model.
The rise of professional investors has many implications for finance, Prof. Yan advises. One example is refining our understanding of markets, investor behavior and the theory of asset pricing, as he explains in a recent interview.
How does your new research fit in with asset pricing theory?
The traditional approach is discounting cash flows and considering the risks. A series of asset pricing models have been created over the years, but empirically the models don't work very well. That gives my profession a huge motivation for thinking about moving beyond the basic models.
One general line of research I'm focused on is introducing extra institutional details for pricing assets. Typically, when creating a model for explaining how the world works, you throw away what appear to be less important details. But it's time to bring in more institutional details [into pricing models]. Institutions now own a large share of assets, and hedge funds are playing a bigger role, too. If you think about how hedge fund managers behave, reputation is a key issue, which is related to their ability to keep clients and raise capital.
How does your model work?
In our study, we're looking at a particular behavior, which is that managers tend to choose strategies that generate pretty decent returns most of the time but once in a while suffer a huge crash. If you look at most of the popular hedge fund strategies, they tend to have negative skewness in their return distributions [an indication of frequent small gains and rare but large losses]. The question is, why choose that kind of strategy? There are two hypotheses: One, the managers have no clue that there's a possibility for a crash with the strategy. I don't think that's the case. But if they knowingly choose the strategy, why do they choose it? In search answers, we created a model (see Table below left) which measures the skewness of hedge fund returns.
What does skewness in returns tell us?
Skewness measures the bias in a return distribution. Hedge funds, for instance, tend to have bigger losses and moderate gains [as indicated by negative skewness in the table]. Imagine a strategy with pretty decent returns most of the time, but once in a while the strategy crashes. Now imagine a manager building a phenomenal track record over, say five or 10 years with such a strategy. That's a huge benefit for the manager, because he can attract more clients with the record. That opportunity gives managers a huge incentive to use the strategy.
In the paper, you call that a nickel strategy, which is really a label for a number of different hedge fund strategies with a common theme of negative skewness. What's the alternative strategy?
That's what we call the black swan strategy [positive skewness], and it's pretty rough because every month there's usually a small loss and only once in a while--once every five to 10 years, for instance--will you hit a home run with return. Even if a home run compensates for all the losses, in the meantime the clients will probably fire you by taking their money out of the fund. That's a huge problem for hedge fund managers.
Overall, we're saying that hedge fund managers care a lot about their reputation, and so they favor nickel strategies. That's true even if they understand that there's likely to be a bubble risk with nickel strategies; they're still reluctant to use the other strategy--the black swan strategy.
The reputation factor, if you will, is specific to professional money managers?
Right. If you're managing your personal money, there's no reputation issue. You're not worried about clients taking money away. That gives you a stronger position to deal with a crisis.
Your research also makes a case for two states of market equilibrium rather than the traditional view of just one.
That's what academically is called multiple equilibria. As an example, if people suspect that their bank is insolvent, they would rush to withdraw their deposits, and this makes the bank insolvent even if the bank is otherwise healthy. On the other hand, let's say people are not worried about their bank. The bank can actually invest in long-term projects and make enough money to pay back its depositors. There's another theme in the paper related to what academically is called multiple equilibria. In the model we find that a hedge fund strategy can be very popular; it can also be very unpopular. We're trying to make a point that when big investors care so much about reputation, the equilibrium situation is very fragile. When there's a shock, there may be a shift from one equilibrium state to another, with very large price movements and capital relocation, even when there is no news about the fundamentals. For example, if investors believe a strategy is popular among talented managers, they'll be more tolerant of the managers using the strategy. Even if those managers generate a loss, investors won't be so suspicious about a manager's abilities. That's one equilibrium state. On the other hand, if investors think a strategy is unpopular among talented managers, investors will be more intolerant of a loss in the strategy. That's the other equilibrium state. So, depending on investors' perception, they can be quite tolerant or intolerant of a strategy.
In early 2004, for example, the convertible bond arbitrage strategy was extremely popular, and so investors were tolerant of losses. After 2004, and a bunch of very bad quarters, there seemed to be a shift to the other equilibrium, in which investors were intolerant of convertible bond arb strategies. In that case, the managers were scared to implement the strategy because a small loss could cost them their jobs.
Your paper also explains that expected return varies for investing strategies as markets shift from one equilibrium state to the other.
One example: The theoretical implied expected return in convertible bond arbitrage, after 2004, was rising because fewer managers were implementing the strategy. In other words, the strategy became more attractive after a run of poor returns. Why didn't capital flow back to the strategy? Traditional finance theories have a hard time explaining this. But in the context of reputation concern, this is a natural phenomenon. The reason is that investors are intolerant of the strategy in this new equilibrium, and so managers face bigger reputation risks.
So the idea of one equilibrium state that applies at all times for investing strategies is in fact more dynamic, according to your model?
Exactly. Prices may move without fundamental reasons. The reasons they move are because the hedge fund managers are moving capital around. Because of reputation concerns, managers may have to move a large amount of capital very, very quickly, and that can have a large impact on asset pricing.
Meanwhile, the more unpopular a strategy, the higher its expected return, and vice versa?
Exactly. That raises a question: When convertible bond arb, for instance, becomes so unpopular and expected returns are so high, why don't people put capital back into the strategy? One reason is that when investors are intolerant of the strategy, managers face bigger risks. They know that if they invest in such a strategy, a small loss may cost their job, and so they're very cautious.There's a bigger career risk at certain times, even if expected returns are high.
What's the basic message in your research for thinking about investing generally?
When investing in hedge funds, one should consider the possibility of a crash, which can so easily be ignored. Also, one should not completely rely on the track record. An extraordinary 10-year record does not mean that there won't be a crash next year. Taking this into account, one may want to reconsider how much to allocate to the fund. From the policy point of view, one should think about the systematic risk that might be generated by the hedge fund industry. An effective policy design probably should take the reputation problem seriously.
James Picerno (firstname.lastname@example.org) is senior writer at Wealth Manager.