Uncertainty is that unwelcome, unsettling feeling we experience when we don’t know what the future may bring. It can range from small scale – like being unsure when dinner guests will arrive while preparing their meal–to something much greater.
And there is no large-scale uncertainty greater than Covid-19–and its delta variant–at the moment, which has laid bare our anxieties, trepidations and lack of caution. In early 2020, most of us suffered from willful blindness; by mid-year we were on knife’s edge, either cowering or throwing caution to the wind; and then we felt exuberant about miracle vaccines. Not everyone experienced this sequence; and some people may want to crawl under a rock again if these nasty variants such as delta multiply.
Why do we handle uncertainty so poorly? Behavioral science dictates that almost all of us desire some control over our lives and uncertainty is an unpleasant reminder that we often can’t. Many events we just cannot control – from the weather, other people’s behaviors and perhaps our own emotions.
The need for control is only one side the story, however, since humans also have a deeply rooted need or desire for variety and surprise. Life would be pretty boring if we could predict or control everything. That’s why we embark on adventures, and explore unknown terrains. Or why Las Vegas exists. But few would consider Covid-19 a welcome diversion or interesting episode, unless perhaps they are research scientists.
To sort out our conflicting emotions, it helps to understand the different kinds of uncertainty, from financial and social, to physical and moral. Our risk tolerances likely differ among these domains. Some people are willing to risk life and limb for fun, like a conservative accountant enjoying the thrill of bungie jumping. Each of us needs to find the right balance by taking less risk in some areas and more elsewhere.
When Covid-19 was thrust upon us, most people countered it by reducing risk taking elsewhere. Once the pandemic subsided in the U.S., we embraced more risk again socially and otherwise. As we continue to rebalance our risk portfolios, try to keep the following pitfalls in mind.
Our Risk Perceptions are Biased
Researchers studying risk perception asked people to rate the risks, from low to high, of a broad cross section of activities, such as skiing, driving a car, living near a nuclear power plant or being exposed to X-rays at the dentist. The researchers then contrasted lay people’s responses with those of experts in those domains. Where the experts focused primarily on the statistical profile of any given risk, lay people were unduly influenced by:
- Whether they could control the risk (yes for cars and no for airplanes),
- how much they knew about the risk (soccer injuries vs. radon exposure),
- whether the risk strikes people in clusters (earthquake vs losing a wallet),
- a risk’s societal image or dread (catching Ebola vs tripping in a shower).
Vivid risks – such as shark attacks or explosions – leave a strong imprint and are typically feared disproportionately to their actual probability of occurring. The car trip to the ocean is usually riskier than swimming in it, even where occasional shark attacks do happen, such as coastal Florida. Another mental factor: how readily we can imagine plausible pathways to a catastrophic outcome. It is much easier to imagine your airplane falling from the sky (due to engine failure or pilot error) than to envision getting stomach cancer or being accidentally electrocuted at home.
Covid-19 is a risk over which we have some control through how we behave, and we learn more about it every day, even though the new variants remain scary. The disease can occur in clusters (like retirement homes) but also strikes randomly. As more knowledge was gained, societal dread declined in the U.S. and Europe although not in all quarters due to lockdown protests, antivaxxers, QAnon and other social media charlatans. Also, the shifting advice of public health officials, reflecting the changing nature of the virus and its prevalence, has heightened some people’s fear of the virus.
Flawed Choices Involving Risk
In addition to distorting the estimates of probabilities and consequences, research shows that people are often inconsistent when making choices involving risk, even if the probabilities and consequences are properly assessed. Here are four major biases:
First, we tend to be risk averse for risks involving gains but less so when it comes to losses. For example, most of us would prefer $100 for sure over flipping a coin to receive either $200 or nothing. But when given the mirror image choice on the loss side, many would opt to flip the coin (risking losing $200 or nothing) rather than accept a sure loss of $100. In general, humans tend to be risk averse for gains but may double down to avoid sure loss, which suggests rather inconsistent risk attitudes.
Second, we tend to factor probabilities into our choices in distorted ways by overweighting low probabilities and under weighting high ones. Consider this inconsistent set of choices: Most of us prefer a guaranteed $100 over an 80% chance of getting $120 and 20% of getting nothing. But when both these choices are reduced equally by a factor of four in likelihood, people will switch preference. They will opt for a 20% chance at $120 over a 25% chance at $100, which is a flip flop of the previous choices. If you like an apple better than a pear, you should also like a small chance of getting an apple over an equally small chance of getting a pear.
Third, people often treat any given risk in isolation from other uncertainties they face and thus fail to adopt a portfolio perspective. Flipping a coin once may seem risky. But flipping it many times will reduce the risk due to the law of large numbers. With many flips, the possible payoffs will start to resemble that famous bell-shaped curve of statistics, known as the Normal or Gaussian distribution. For example, flipping a weighted coin that offers you a 60% chance of winning $100 and 40% chance of losing $100 just one time may be unattractive. But flipping that weighted coin a hundred times likely is attractive since it offers you a $2,000 net gain on average with just a small chance (less than 3 percent) of having lost money after those hundred flips.
Fourth, people tend to prefer known risks over ambiguous ones, even if these cases are statistically identical. Suppose you can draw a ball from an urn containing 50 red balls and 50 white ones versus an urn in which the ratio of red to white balls is unknown to you. Suppose, further, that you get $100 if you draw a red ball blindfolded– which urn would you rather pick one ball out of? Most people favor the first urn even though the second one, with an unknown ratio of red to white, will on average – in the absence of any further information – also offer a 50% of drawing red. Indeed, people will not even prefer the unknown urn if they are allowed to choose which color to bet on! This illustrate our innate aversion to ambiguity, which can be detrimental when choosing between jobs, investments or life experiences.
The upshot is that most people don’t handle risk and uncertainty very well when left to our own devices, as amply demonstrated by the upheavals caused by Covid-19. Unaided decision making in the face of the unknown often results in inconsistent and thus sub-optimal choices. Fear and hope intertwine when we look at risky decisions intuitively and our emotions can easily overshadow more reasoned perspectives.
Fortunately, well-established approaches exist to help you choose rationally under risk. Those include mean-variance analysis or more generally, expected utility theory. Sadly, these theories are still not widely known enough and can be hard to apply at first. But without using some decision aids, biases and error easily creep into our choices and diminish the quality of our life.
Most of us know intuitively that we don’t handle uncertainty well and can fall victim to the numerous quirks and biases that decision researchers have cataloged. Also, most managers realize that the analytic tools that are taught in business school, such as decision trees, utility theory, sensitivity analysis, portfolio models and Monte Carlo simulation, may not quite be up to the task of fully taming uncertainty in real world.
Putting these two realizations together, it is little wonder that we dislike uncertainty. That’s the bad news. The good news is that we can actually do much better by using simple decision tools and time-tested practical pointers, as detailed in books blending behavioral and analytical approaches. Paul Goodwin and George Wright wrote an excellent textbook covering the behavioral and rational sides of decision making, whereas J. Edward Russo and I wrote one for managers about common decision traps and how to avoid them in practice.
Finally, there is always the school of hard knocks, which can offer clear lessons but often at a high price. So, you have to pick your poison by either beefing up on book smarts or grinding it out with street smarts. However, don’t flip a coin, but try both. ,