By Lixing Sun

Just a few weeks ago, apparently out of frustration when Donald Trump edged closer in polls, Hillary Clinton bashed Trump supporters by calling half of them “deplorables.” She quickly regretted it for the ensuing PR blunder. In reality, she was certainly wrong. Even two of my friends, who can’t be more removed from her portrayal also told me that they were leaning for Trump. You may think they are denizens of the dark red states in the Deep South. Wrong. They live in the bluest of the blue: New York and Maryland. When asked why, “you see,” they confided, “one is untrustworthy and the other is crazy.”

Colorful characterization aside, my friends’ attitude reflects most Americans’. In fact, this is a bizarre election in many decades when the unfavorable rates of both candidates exceed 50%. Even so, why do so many voters still choose Trump over Clinton? A key to the answer lies in an evolved human mentality: avoiding loss by gambling.

This mentality was first systematically studied by pioneers in behavioral economics such as Richard Thaler, Amos Tversky, and Daniel Kahneman in the 1970s. They found that people tend to gamble when facing a sure loss. Use an example from Kahneman’s popular book, Thinking, Fast and Slow: when people are given a choice between 1) a sure loss of $500 and 2) 50% chance to lose $1,000, the majority will prefer option 2. Such tendency to gamble can go out of the ballpark of rationality as shown in a recent study by researchers from University College London: when facing a sure loss of £20, 62% of people would rather go for a 50% chance to lose £50.

Such gambling mentality when facing a sure loss has a deep root in evolution. If you 1) show and give capuchin monkeys one slice of apple, or 2) show them two slices of apple and then take one slice back, guess which option the monkeys will take? Economist Keith Chen and his collaborators at Yale University show that the monkeys would swoon for option 1 (nearly 80% of the time) only because option 2 stands for a sure loss. Apparently, plants also gamble when facing a sure loss. Efrat Dener at Ben Gurion University in Israel tricked some famished pea plants to allocate roots between two pots with the same low level of nutrition. The only difference lies in that the supply is constant for one pot but erratic for the other. He found that the innocuous pea plants, rather than withstand a constant strain of nutrients (a sure loss), place bets (by growing more roots) on the pot where nutrition comes fitfully.

No doubt, loss aversion by gambling is crucial in economic decisions. But how can it be retooled for use in political voting?

For the current presidential election, there is one big difference between the two candidates. A political veteran for over three decades, Clinton embodies predictability and continuation for many of the current policies. A political novice, Trump, on the contrary, is seen as the opposite of the status quo. So for voters who are dissatisfied by the current political, economic, and social situations, a Clinton presidency stands for a sure loss and a Trump presidency represents a gamble—a chance to win, if things go as Trump blueprints. Hence, Trump, even being more disliked, can gain a better part of the support from these voters for no other reason than our evolved penchant to gamble when facing a sure loss. Now we know why there are so many “deplorables” out there to back Trump.

Although the gambling mentality for loss aversion has a deep evolutionary history, how reliable is it in today’s world? Can we still let it reign freely in our economic, social, and political choices and hope we will somehow end up unscathed?

Unfortunately, the answer is most likely no. The reason lies in that in today’s complex world, we are often unable to gauge the scale of risk with our Stone-age instincts and intuitions.[i] So, betting without knowing the magnitude of risk can be dangerous. We all know that gamblers lose big when they try to recoup a small loss by taking more risk. Even so, such gambler’s fallacy is still prevalent in investment. It can make many of us hold onto a losing position for too long. When every small loss in our asset (known as sunk-cost in economics) becomes a reason to keep it, we may end up with losing our shirt. Think about the spectacular fiasco of the supersonic Concorde project. How could it be possible to happen if the British and French governments cut the joint venture loose early on when it had already shown obvious signs of failure? To hedge against a devastating loss, it’s often better to write off a small cost and move onto a greener pasture elsewhere rather than, as the Chinese would say, “stick to the same tree until you hang yourself.” In all these cases, the gambling mentality can fail us in a spectacular manner as folk wisdom warns us (remember the children’s rhyme: “For want of a nail the shoe was lost.…”).

Now, what is the risk in electing the much less predictable Trump? Although we can’t rely on our intuitions and instincts in this case, we can assess it by observation and analysis.[ii] Trump hinges his economic policies on slashing taxes, especially corporate taxes, which few mainstream economists think they will work but adding a national debt of over $5 trillion in the next 10 years. Trump’s economic policies even turn off leaders of large corporations, the presumed beneficiaries: none of the CEOs of the Fortune 100 companies has donated to the Trump campaign. (Goldman Sachs even bans its bankers from opening their wallets for Trump.)

How likely will these policies help the hardworking rural folks and non-college educated, blue-collar workers in the bottom?

Trump’s business practice is another matter of concern. Among many others, he has been involved in 4 bankruptcies (he declared a gigantic $916 million loss in his 1995 tax return) and is still mired in several lawsuits including one against his Trump University. Even his responsibility as a citizen is in question when he equates his failure to pay taxes as being “smart.” (He might have paid no income taxes at all for nearly two decades after 1995.)

Can we hand over the Oval Office to such a person with zero experience in public service and questionable sense of basic fairness?

Trump’s bossy temperament, erratic antics, and reckless claims may be suited for a private company with a steep hierarchy. But can we trust this person as a leader in a democratic nation? This is not to mention his anti-woman, anti-immigrant, and anti-Muslim rhetoric. There are few signs that he will be able to, using his own campaign slogan, “Make America Great Again!” but all signs that he will divide America. These are some major risks for those of us who don’t like either candidate to consider regardless of our political views and affiliations.

The bottom line is that we don’t need to take any risk to gamble when the magnitude of the risk can be assessed. This worked spectacularly well for a friend of mine in London when the Brexit was up for a referendum last July. The polls were close between “Leave” and “Stay.” But just one day before the vote, he swapped all of his savings from the pound to the dollar. “Even though the odds were 3:1 for Stay,” he reasoned to me, “it was not worth the risk of gambling.” On the second day, when the result was out, all the major banks in the world were in an enormous panic, as the sterling dropped more than 10%. My friend could be literally laughing to the bank. The moral is that when we are facing a sure loss and opt to gamble, our evolved instincts often fail us. In this sense, my hat is off for many (especially conservatives such as George H. W. Bush) for exercising their sound judgement and civic responsibility to resist the pull of partisanship when facing a weighty choice for the future of our nation.

2016 October 7

[i] Here is an example from risk analyst Gerd Gigerenzer’s 2003 book, Calculated Risks, to test how good you are in sensing risk in today’s world: “The probability that a woman of age 40 has breast cancer is about 1 percent. If she has breast cancer, the probability that she tests positive on a screening mammogram is 90 percent. If she does not have breast cancer, the probability that she nevertheless tests positive is 9 percent. What are the chances that a woman who tests positive actually has breast cancer?” (Answer: 10 percent.)

[ii] In psychology, perceptions with instincts or thinking with intuitions or inveterate habits is known as System I (fast) Thinking. In contrast, conscious reasoning and analysis are known as System II (slow) Thinking.