Tech Perspectives

Nature Risk: The Velociraptor Next Door

What’s the difference between the risks nature imposes on humans and the risks humans impose on nature? What are some examples, and what steps can risk managers take to identify, disclose and mitigate these threats?

Thursday, March 28, 2024

By Aaron Brown


Just as many risk managers are struggling with newly-mandated climate risk disclosure, nature risk reporting has appeared on the horizon. Forward-thinking managers therefore at least need to think about how to approach disclosure and management of nature risk.

Those seeking advice on the best strategy can look to a clear nature risk statement recently issued by Norway’s sovereign wealth fund:

“Estimates suggest that over half of global GDP is moderately or highly dependent on nature. However, global biodiversity is disappearing at an unprecedented rate, and most of the ecosystem services on which business and society depend are in decline. As a globally diversified investor, we have an interest in understanding how the companies we own depend and impact on nature.”

a2r1W000000x3uDQAQ_Aaron-BrownAaron Brown

While this only affects companies and managers hoping to get capital from Norway’s fund, the principles are supported by many environmentalists with influence over politicians, investors and regulators.

Identifying, disclosing and managing nature risk are now clearly on risk managers’ radar. But where can they find guidance that will help them on this journey?

Truth in Fiction

Michael Crichton’s 1990 novel, Jurassic Park, is an excellent and entertaining place to start. In addition to being a spectacularly successful popular novelist, producer and director, Crichton was among the most prescient and incisive scientists thinking about the risks of technology and nature — as well as the dangers of the very human trait of hubris. The theme of the Jurassic Park novel, diluted in the movie version, is the illusion of control among researchers and businesspeople exploiting technological advances.

In the book, the reason for bringing in outside experts to Isla Nublar is the escape of a small species of dinosaur to Costa Rica, where one injures a small girl. (In the movie version, it is an accident in which a caged velociraptor kills a worker.) The builders of Jurassic Park believe they can control their creations with biosafety protocols in the research lab, a remote island location and electric fences.

More safeguards came in the form of creating only females, so they could not reproduce, and depriving the dinosaurs of the ability to produce lysine, so they could not survive without supplements given to them in the park. Paleontologists Ellie Sattler and Alan Grant, along with chaotician Ian Malcolm, immediately realize how inadequate this approach is. In Malcolm’s (really Crichton’s) words, “Life will find a way.”

A key difference between movie and book is the movie begins with a normal, well-understood risk — animals in zoos killing staff. The disaster depends on the confluence of an enormous storm and a rogue employee, also common events. The conclusion is that the park depended too much on automation, a standard response to disasters.

In the book, the disaster occurs before the beginning. Dinosaurs have escaped the island and are thriving, reproducing and hurting people on the mainland. The storm and malfeasance cost a few lives on the island, but the big problem is the engineered creatures can evolve and spread to threaten everyone and every ecosystem. The mistake of Jurassic Park wasn’t overreliance on automation but failure to understand life.

Real-Life Nature Risk

Moving from fiction to fact, it seems plausible that the decision to do gain-of-function research on coronaviruses led to seven million global deaths — with over 100,000 additional deaths per year projected for the indefinite future. Of course, we don’t know this for sure, and we also don’t know how much benefit there was from the gain-of-function research that did not lead to a global pandemic.

Much of the public discussion on the COVID-19 pandemic focuses on who or what to blame — e.g., poor adherence to biosafety protocols, reckless individual researchers, and officials with varying protocols in different countries. However, this may be like blaming the storm and a criminal employee for Jurassic Park. If life will find a way, creating a virus with pandemic potential is fundamentally unwise, regardless of precautions taken.

Most risk that humans impose upon nature is like this — different from the risks that nature imposes on humans (fire, earthquake, pandemic, etc.) or humans impose on each other (crime, war, financial crises, etc.). Serious thinking about nature risk disclosure and management must take this into account.

Risks nature imposes on humans are well modeled by standard probability theory derived from casino games. Unfortunately, this is the only theory of probability many people learn in school.

Risks from hurricanes, earthquakes, falling trees and other attacks by nature can be studied by actuaries, with probability and loss distributions estimated using historical data and theory. But applying standard probability theory in inappropriate domains is what Nassim Taleb labeled the “ludic fallacy.”

When dealing with risks humans impose on each other, an entirely different mathematical approach should be used: game theory. Game theory models uncertainty as choices by rational beings, rather than something created by nature.

We don’t estimate the risk of embezzlement, for example, by hiring actuaries to estimate what fraction of people are dishonest. We think about the incentives created by our organizational culture and how they influence temptation to steal, the chance of getting caught, the rewards for honesty and the penalties for dishonesty.

Although the mathematics of natural and game theory probabilities are identical in formalism, the meaning is entirely different — often opposite. For example, in standard ludic decision theory, more information and more freedom of action can only increase utility. In game theory, on the other hand, information and freedom are often bad.

Take, for instance, the famous Prisoners’ Dilemma paradox. Under that scenario, if both partners have the ability to reduce their sentences by informing on the other, both will do so, and both will spend more time in prison than if they did not have the freedom to inform. In arms races, meanwhile, both sides waste resources and risk war when they would be better off if neither side were able to build offensive and defensive devices.

A risk manager applying tools designed for natural disasters or financial markets to criminal trials or wars would fail. Neither standard probability theory nor game theory can help much to analyze the risk if our next-door neighbor breeds velociraptors — or for unprecedented experiments with nature in general.

Rather, when assessing the risks humans impose on nature, practitioners should weigh, for example, what I call innovative nature tampering — e.g., the risks of genetically-modified crops, biological warfare research, and genetic medical therapies. Even nature modification in the form of non-innovative experiments, like CO2 emissions, plastics and deforestation, should be considered.

Human-Driven Nature Risks: Focus on Fitness

What sort of mathematics is appropriate for risks humans impose on nature? Rather than focusing on the probabilities of outcomes, we should think of their fitness. We know nature can bring about events of nearly infinitesimal probability via randomness and selection. It’s not the most likely scenarios that happen, but the most fit ones.

The mismanagement of nature risk at Yellowstone National Park, which Crichton explained in detail in a speech he gave in 2005, is a great example. Speaking of the goal of Yellowstone, President Theodore Roosevelt once said: “Our people should see to it that this rich heritage is preserved for their children and their children’s children forever, with its majestic beauty all unmarred.”

The problem is nature is constantly changing – it cannot be preserved, and there is no such thing as “unmarred.” Yesterday’s forest is today’s wildfire and tomorrow’s scrubland. Yesterday’s mighty river is today’s canyon and tomorrow’s lake.

Crichton recounted that when Yellowstone’s managers noticed that elk were declining, back in the 1890s, they began feeding them to preserve the population. This approach, however, was flawed from the start, because the managers were thinking in mechanical terms, not natural terms.

The burgeoning elk population overgrazed and killed off the aspen and willows, which in turn killed off many antelope and deer. Without aspens, moreover, the beaver disappeared, taking their dams with them, which destroyed the water management in the park.

Subsequently, when even the elk population began to decline, the Park Service starting killing off the predators — wolves, cougars and coyotes. But that only made things worse.

Yellowstone managers repeated this same type of mistake often. For example, until the 1980s, all fires at the park were suppressed; consequently, deadwood eventually built up, leading to the catastrophic 1988 fires that nearly destroyed the park.

Earlier in Yellowstone’s history, bears were encouraged to interact with humans, until they harmed so many people that they were moved out of the park and became endangered. Moreover, after wolves were reintroduced in 1990s, they promptly began using the park as a base to attack domestic animals outside their limits.

The Dorner Experiment, and Robinson Crusoe vs. Swiss Family Robinson

Another example of the risks humans impose on nature can be found in a famous experiment by Dietrich Dorner. He built a computer simulation of an African village called “Tanaland,” with immediate issues like disease and a shortage of water.

Participants in the experiment were able to manage the village in various ways, such as introducing vaccinations to reduce disease or drilling wells to get more water. All but one participant managed to destroy the village, killing everyone and leaving a wasteland. Immediate problems were attacked without considering future consequences — reducing disease, for example, led to population growing faster than the food supply, while drilling wells exhausted the groundwater.

It’s important to note that the game was not designed to be impossible, or even difficult. The village was modeled with considerable accuracy.

All information was available to players to understand interactions and future consequences. One player managed to bring about a stable village at a considerably higher population and standard of living than the original — but many such solutions were possible.

A final example of the impact of humans on nature can be found by comparing two famous works of fiction: the 1719 novel Robinson Crusoe, by Daniel Defoe, and the 1812 novel inspired by that work, Johann Wyss’ The Swiss Family Robinson. Crusoe works with nature on his island; the Swiss Robinson’s try to convert theirs to a Swiss village.

Ironically, Defoe wrote his book to entertain, while Wyss wrote to teach natural history. Defoe based his work on the real-life adventure of Alexander Selkirk, while Wyss relied entirely on the theories of Jean-Jacques Rousseau.

One other crucial difference is that in the century between the two books, humans had gotten further away from reliance on nature. Crusoe describes a realistic and practical way to convert a deserted tropical island into a comfortable home for a single human. Wyss’ Swiss family, on the other hand, would have died quickly from the ecological disaster they created.

Lessons Learned from Fictional and Real Disasters

When analyzing the risks humans impose on nature, we can learn a lot from outcomes. Just think about the ramifications of the fictional and real-life events we have discussed:

  • Jurassic Park. If you recreate dinosaurs and release them to the open air, you will never prevent them from spreading to all suitable environments and evolving. It seems likely they will be invasive in at least some ecosystems, with the enormous advantage of being immune to all modern microbes and having no natural predators. Dinosaurs dominated the Earth for 180 million years, so they would be expected to displace many existing species. Only mad scientists need apply.
  • Yellowstone. Try to gather data, or at least estimate, the historical range of variation among animal and plant populations, fires, water and other natural elements. Do your best to minimize the impact of human visitors, pollution and other stresses. As long as the park seems to be orbiting an acceptable stable equilibrium, leave it alone. If the park threatens to drift out of equilibrium, or to an unacceptable equilibrium, nudge it back slowly and humbly, watching closely for feedback — with allowance for lags and with consideration (to the extent possible) all downstream consequences of actions.
  • Tanaland. Begin with a goal — a sustainable population at a sustainable level of consumption. Check the requirements for water, fertilizer, healthcare and other inputs to support this level, and make sure that all people, plants and animals have what they need. Then move toward this goal, slowly and humbly, watching closely for feedback with allowance for lags.

  • Crusoe. Unlike Yellowstone or Tanaland, a lone human in a real environment lacks the capacity to imagine a practical future state. Instead, he must learn to live with what nature provides, and gradually experiment with things to make life better — e.g., domesticating goats, making raisins, and using thorn bushes for defensive walls. By keeping each experiment small and not relying on success, he can prosper without excessive risk.

Parting Thoughts

Reporting on and managing nature risk is an exciting emerging field for risk management that will require a different perspective from existing practice. Forcing it into patterns developed for risk from nature or humans, however, will lead to disaster.

Any risk manager tasked with serious real decisions regarding nature risk should begin with a careful study of how risk has been mismanaged in the past, both in reality and in fiction.


Aaron Brown worked on Wall Street since the early 1980s as a trader, portfolio manager, head of mortgage securities and risk manager for several global financial institutions. Most recently he served for 10 years as chief risk officer of the large hedge fund AQR Capital Management. He was named the 2011 GARP Risk Manager of the Year. His books on risk management include The Poker Face of Wall Street, Red-Blooded Risk, Financial Risk Management for Dummies and A World of Chance (with Reuven and Gabriel Brenner). He currently teaches finance and mathematics as an adjunct and writes columns for Bloomberg.


BylawsCode of ConductPrivacy NoticeTerms of Use © 2024 Global Association of Risk Professionals