The COVID-19 pandemic is a reminder that life is risky and the future is uncertain. Suddenly and simultaneously, we were all facing the risk of losing our health, lives, freedom of movement, income, and savings.

Communication is a critical part of reducing anxiety and turning uncertainty into risk. Communicating risk is hard, however, because doing so relies on abstract concepts, probability, and statistics that not everyone comprehends. Storytelling can be an effective, alternative way to convey more complex concepts.

In “Learning about Risk Management—Insights from Unconventional Risk-Takers,” a CFA Institute Research Foundation Brief, Allison Schrager, senior fellow at the Manhattan Institute and co-founder of risk advisory firm LifeCycle Finance Partners, LLC, shows how stories improve risk management practice and communication. “Risk models are in many ways a parable, an abstraction,” Schrager says, that offer insight into an important lesson or relationship. We can use these allegories to explain risk and the value of risk management.

Schrager’s research shares knowledge from unconventional risk-takers. While their own fields have little in common with the financial industry, the players all face the same problem: how to balance risk and reward. The following is a summary of the parables Schrager presents in her research.

Hollywood, the land of broken dreams
Schrager starts with a cautionary story from Hollywood. In the movie industry, it’s impossible to predict which movies will be blockbusters and which ones will flop. The only way to manage risk is to make a lot of movies: while most won’t make money, a few big hits will let studios cover the costs of the others.

A good risk estimate, Schrager says, requires data that can reveal past lessons—which, in turn, prove to be relevant in the future—and predicts certain past outcomes are more likely than others. The nature of moviemaking, however, means that its data cannot be used to generate a good risk estimate. Box office revenue distribution is skewed, with many data points in the tail, because most movies lose money or barely break even. This extreme skew makes it difficult to get reliable risk estimates.

Schrager’s cautionary tale is the story of Ryan Kavanaugh. Kavanaugh charmed Hollywood with a Monte Carlo simulation that promised to make the unpredictable predictable. He first started off with a venture capital fund in the 1990s but his firm failed after the dotcom bust in 2000. Subsequently, he co-founded Relativity Media in 2004 and marketed himself as a math whiz who could provide the predictability that Hollywood craved. Kavanaugh’s experience as a venture capitalist served him well in convincing people in Hollywood, because a venture capital firm’s investment strategies are similar to those of movie studios.

His timing was perfect, because movie studios at the time were looking for new sources of financing while hedge funds were looking to invest in high-yield assets as interest rates fell. The studios had been dependent on a German tax shelter that gave investors and studios financial incentives to invest in movies but, after Angela Merkel took office in the mid-2000s, her government discontinued the shelter.

Kavanaugh claimed his model could generate a reliable estimate of risk even when data was skewed. He selected certain movie characteristics (actors, directors, genres, budgets, release data, and ratings), then analyzed data for the same characteristics from previous movies to predict a winner. The model produced a range of potential profits based on how these characteristics previously performed. The movies that he picked then delivered 13 to 18 percent returns to his investors in 2005 and 2006.

Kavanaugh’s downfall, Schrager says, was greed. He began investing in movies himself after Elliot Management, a $21-billion hedge fund, bought 49.5 percent of his firm. Soon after, his magic model stopped working and he was selecting box office failures instead. Elliot Management pulled out in 2010, and Relativity Media was bankrupt by 2016.

Schrager’s story about Kavanaugh’s is a warning as to why investors should never be seduced by the power of their own model or—more importantly—ever fall for someone else’s model. Perfect risk estimates do not exist, and past data is less useful when changes happen quickly, she points out. Skill and judgement are required to know which data are the most relevant and how to make reliable estimates from tail cases. More data estimation techniques, such as machine learning, may lead to more reliable risk estimates. However, when tastes and technology both change quickly, even big data cannot produce perfect estimates.

Overcoming behavioural bias
People get better at making good risk decisions over time or as they gain more experience from facing the same risk problems regularly, says Schrager. Risk management suggests that we do not have to be slaves to our emotions and can deal with risk in a logical way.

Schrager’s second case study is of Phil Hellmuth, a world champion poker player who has overcome the “break-even effect,” as coined by behavioural economists Richard H. Thaler and Eric J. Johnson. Thaler and Johnson have argued that people bet more when they are losing so they can get back to the break-even point. Statistically, however, the odds to win or lose are the same, no matter what happened in the earlier hands, making this approach not a good way to take risk. Hellmuth credits his own success to not falling into this trap.

Schrager reports three strategies that Hellmuth uses to keep his emotions in check:

1. Don’t have too much of your own money on the table.
   Schrager says that Hellmuth goes into large tournaments “staked,” where he has outside investors put up money for him to play and they receive a share of his winnings in return. His own personal stake never exceeds US$10,000,  which enables him to win big without losing too much of his own money, and keeps him from feeling too desperate when he is down because only a small fraction of his own money is at stake.
2. Eliminate extreme downside risk.
   Hellmuth makes side deals with his opponents to split the prize money and still offer the winner some extra upside. Having a guaranteed pay whether he wins or loses, in addition to being staked, helps him stay focused. An everyday example of this would be buying insurance: insurance offers us peace of mind because the potential loss is smaller.
3. Remind yourself “this is just one hand out of many” (broad framing).
   Schrager described how Hellmuth’s broad framing is a way of playing the long game—frame an individually risky decision as part of a larger gamble, for example. Broad framing help individuals to think clearly and avoid overreacting to temporary loss.

Wisdom from big wave surfers
Once we have defined, measured, and understood risk, Schrager says, we can find the best strategy to manage it. There are two ways of managing risk: hedging and insurance. Hedging reduces the upside in exchange for less downside, while insurance allows you to keep the upside and pay someone else to reduce the downside. Both strategies, however, come with complications. It is hard to strike the right balance with hedging because investors don’t want to give up too much upside, and so may end up hedging too little risk. Insurance, however, can create a moral hazard, or false sense of security, that enables people to take on more risk than they should.

Schrager’s third case study is of big wave surfers. Big wave surfers seek out waves that are 20 to 80 feet high, often in remote locations, and must balance the thrill of riding a big wave with the risks (which include sharks, rocks, and cold water).

The common hedging strategy they use is in picking the right wave to surf, as waves tend to travel in sets, she says. If the waves are a part of a five-wave set, a surfer can hedge by taking the fourth wave. That way, after the surfer finishes or wipes out, they are not held underwater by the next big waves in the set. A surfer who is tempted to take on extra risk by riding an earlier wave in the set may be risking their life.

Jet skis serve like insurance for big wave surfing. The jet skis can cut through rough waters to bring injured surfers to shore quickly for medical attention, providing protection if things go wrong while still allowing surfers the unlimited upside of tackling the big waves. Like any insurance, however, jet skis may also result in moral hazard as less experienced and less skilled surfers rely on the jet skis to save them, and may then take bigger risks than they should.

Schrager says that taking on these big risks poses costs to others: resources are diverted from helping others in need, the lives of the rescuers are put at risk, and additional expense is incurred if the coast guard has to be called in. In finance, excessive risk-taking transfers risk to institutions that are unprepared to bear the risk, and sometimes requires a government bailout.

Each innovation—jet skis, inflatable vests—makes surfing safer and expands the boundaries of what surfers can do; however, those innovations also encourage more risk. The key to better risk-taking, Schrager says, lies in better education on how to use the tools property and understanding their limits.

The stories above show that successful risk management, relies on the following factors, regardless of industry. Schrager identifies them as:

1. the ability to identify, define and measure risks;
2. the ability to overcome behavioral bias; and
3. the knowledge on how to use the risk models, risk reduction strategies, technology to manage the risks, and understanding their limits.

Concluding thoughts
Schrager concludes that communication is critical for successfully conveying the value of risk management, which is often underestimated in good times and obvious only in hindsight in bad times. Risk management and storytelling are complementary. Risk models in many ways are parables, Schrager says: “an abstraction to assist us in understanding a complex, ever-changing world so we can make decisions,” and appreciate the value of risk management even in good times.

But risk management is always incomplete because risk models are imperfect, Schrager cautions, as models cannot account for everything that could happen and are not meant to. Technology and innovation cannot eliminate risk, either. However, risk management is still useful to help us understand typical challenges and enable us to think through what could happen. Even when we’re faced with uncertainty, risk management can be adapted and re-optimized to lessen the new, previously unimaginable downside risks.