I work in the investment business, where risk-taking is an occupational necessity. There isn't anyone successful at managing a mutual or hedge fund who avoids risk; we just need to face it carefully. Traditionally the industry encourages a solo approach to evaluating risk; at Fidelity Investments, where I worked for over two decades, each fund is assigned to one person who makes all the buying and selling decisions. When results are strong, the manager basks in the glow, prestige, and compensation attached to outperformance. When performance suffers, a situation experienced by anyone who has managed a fund for over ten years, you feel like an impostor and reach for the Pepto Bismal.
Nine years ago, when I co-founded an investment firm, we chose a different approach to decision-making, characterized by group risk taking. Two to four of us, working on each type of investment involved - stocks, alternative asset classes, venture or private holdings -- must all agree on the merits of the transaction.
Why did we decide on this, rather than the traditional "czar" structure?
As a fund manager, I was very aware that the pain of losing money was markedly worse than the satisfaction of gaining an equal amount. I didn't realize that noted academics Amos Tversky and Daniel Kahneman were studying this exact phenomenon, which they officially named "loss aversion."
I felt that I would be perfectly happy, in our new enterprise, to forego the personal credit for good decisions if I could also share blame for the mistakes. I anticipated, but without any evidence, that shared pain would be less severe than solitary suffering. My partners agreed that the group approach made sense.
While I acknowledged that, perhaps selfishly, I wanted to avoid the disproportionate emotional response to loss, this was not the only reason we preferred a group risk-taking approach. We had decided to create a concentrated portfolio of 30-35 names, wide-ranging enough to capture at least 90% of the diversification of the S&P 500 but of a manageable size so that three of us could carefully analyze and follow all of our holdings, an impossibility with a large number of stock positions. Each of us would need to convince others of the merits of her or his selections, backing up conviction with strong research. We would only buy or sell a position if everyone agreed. Our combined effort offers "less opportunity to be sloppy" (as one of my partners puts it) and forces an extremely thorough analysis and close monitoring of our holdings. The bar to convince your colleagues is higher than for just yourself.
Studies have now shown that loss aversion leads individuals to make irrational decisions, but that groups often make better ones. One study found that, in various lottery scenarios, groups of three or more were significantly more risk averse than individuals, particularly when the probability of winning was 40% or less. Once people joined a group and discussed their options, they consistently moved toward a risk-neutral position. Groups did accept slightly more risk in the highest probability bets.
How is group risk taking working for us? Because our performance has been good (knock on wood) since our inception, I have to like the process. Through the financial meltdown in 2008, I still felt sick staring at my screens, but I appreciated that we bonded over how to manage through the market implosion rather than point fingers at each other (although I confess, that still happened occasionally - we're only human).
What are the potential downsides?
Groups may play it safe, with each individual choosing to pitch ideas most likely to win broad approval. While we all insist otherwise, there may be a subconscious bias against spending the time building a case for a concept that could be rejected as too risky.
Group efforts could also result in horse trading, where I might support one partner's idea in the hope that he'll back me the next time I pitch a name. In over 9 years, I have not been part of any side dealing with any of my colleagues, and I suspect that this has not happened at all.
Another concern is that group decision-making may simply take longer. I don't sense any lengthening of the days taken between first hearing the idea and pulling the trigger, but I know we devote more time collectively to the analysis. (Research has also found that groups don't necessarily take longer to make decisions.)
Of course, no system of group risk taking could succeed without positive group dynamics, mutual respect for each other's abilities, and a collaborative spirit - but that's true of all teamwork, not just investing.
This post first appeared on the Harvard Business Review blog.