Zero-one decisions in crisis

Crises are unexpected, rare events that pose an existential threat to an individual, an organisation, or a society. They present novel problems and are associated with uncertainty, emotions, and time pressure. The Covid-19 pandemic, for instance, left societies coping with a host of new medical, social, and economic challenges. But while each crisis might seem unpredictable and new, human behaviour in crisis is not unpredictable. We tend to exhibit recurring behavioural tendencies when our mental resources are stretched and when we face threats that may be existential. These conditions can trigger what I call zero-one categorical decisions. This has implications for resource allocation decisions, evaluation of others, risk taking, and learning in crises.

Crises present a shock to a system, putting it in a state of sudden loss. This may be an individual who is facing a medical crisis, a company with a faulty product facing massive returns and lawsuits, or a town facing a nuclear disaster. Crises often trigger an instinctive motivation to “fix things”, make them return to how they used to be. However, how do we balance cost-benefit trade-offs and manage our limited resources?

Emotion and crises

In my crisis leadership class (with Professor Niro Sivanathan), we analyse a critical decision facing Yoshida, the late plant manager of the Fukushima Daiichi nuclear power plant. Yoshida believes there is a chance of nuclear meltdown that could kill those in the plant and living nearby. He can inject seawater into the plant, but this would make the plant obsolete. Should he inject the sea water? One group of students in my class learned that injecting the seawater would reduce the chances of nuclear fission from 2/1000 to 1/1000. The other group of students learn that injecting the seawater would reduce the chances from 1/0000 to zero – thereby eliminating the risk. The latter group of students were significantly more likely to recommend injecting the seawater and scrapping the plant, even though the reduction in risk is the same in both cases. This result captures the intuition, proposed by researchers such as Kahneman and Tversky, that eliminating a risk is more impactful than reducing the risk by the same amount. Later research showed that this is especially the case for outcomes that are more emotionally laden and for people who are in an emotional hot rather than cold state*. We tend to react to risk in a categorical zero-one fashion, especially during crises, keen on eliminating risks, but much less sensitive to reducing risks. 

My colleague Jonathan Berman and his co-author showed something related with respect to causing harm. In one study, people were willing to sacrifice efficient use of resources to avoid causing any harm to the environment when this was possible. However, when causing harm was unavoidable, and only a reduction in harm was possible, people were more interested in efficient use of resources. As a last illustration, people react strongly to images of a single victim. Recall the image of five-year-old Aylan Kurdi, a migrant whose lifeless body washed to the coast of Turkey. People are willing to sacrifice significant amounts of resources to help individual victims, identified and vividly described, perhaps feeling that at least that one problem is “solved”. They often react with less compassion to help alleviate human losses in larger scale tragedies.

These three interrelated phenomena capture zero-one approach to judgments and decisions prevalent especially during crises. Feeling like we are fully, categorically solving a problem is alluring. But crises often involve multiple problems, several fires that need to be controlled, even if not completely put out. These natural human tendencies are costly because they result in potentially inefficient use of resources for the sake of the satisfaction associated with elimination of risk and harm, even if greater benefit could be achieved with the same resources to reduce risk and harm. This is especially critical when resources are stretched, as with crises. 

Judging by outcome

Consider now how we evaluate other people’s decisions in and after crises. Decision researchers have long documented an outcome bias, a tendency to evaluate the quality of a decision not by the process that led to that decision, but instead based on whether its outcome good or bad. Outcome bias leads us to punish some people for bad luck, and reward others for good luck. In recent research, we found some evidence that outcome bias may be more pronounced in crisis situations. In face of an existential threat, short term thinking may kick in and outcome may be seen as all that counts, undermining processes, principles, and values which we would cherish at normal times. Who cares, a boss of a company might exclaim, about processes and principles, if we no longer exist as a company? Evaluating by outcome is also easier and quicker, attributes we may appreciate especially during crises. We also want to find someone to blame, or reward, in or after a crisis. Thus, we end with heroes who somehow foresaw and predicted what was to happen, people such as Richard Clark (September 9-11), Nuriel Roubini and Michael Burry (2007-8 financial crisis), Bill Gates (Covid 19), and Yitzchak Brick (Israel Oct 7th). We also seek to identify villains on whom we are keen placing responsibility.

Such outcome based zero-one approach to evaluating others may be reassuring from a psychological perspective, the feeling that some people could predict the crisis and could have averted it, if only they were listened to by our failed leaders. The alternative, that some crises are simply unpredictable and cannot be avoided, is psychologically too disturbing, even if it carries greater truth. However, such greater prevalence of evaluation of others by outcome comes at a cost. Our research shows that when people believe they would be judged by outcome, they avoid taking risk.

They fear that any potential losses associated with taking a risk would be weighted heavily and result in being blamed, even if these were calculated and warranted risks. Ignoring the process of decisions and weighing only results can thus curb risk taking – something which ironically is most needed under times of crises. The tendency to judge by outcome and place undue credit or responsibility on people can also detract attention from processes we may need to develop to cope with uncertainty and with complex novel situations. 

Recurring features of crises make certain behavioural tendencies more likely and pronounced. I gave here a few examples of phenomena that broadly fit under the umbrella of zero-one categorical decisions. While crises may be unpredictable, the predictability of certain biases in a crisis, and an understanding of what their psychological drivers, offer potential for increased behavioural resilience and preparedness.

David Faro is Associate Professor of Marketing at London Business School