Perfectly Confident Leadership

Confidence and Performance

Research does indeed suggest that confident people are more likely to be elevated to positions of status and leadership. We trust them in part because we rely on the presumed positive correlation between confidence and performance. We behave as if we assume that their confidence portends future success.

There is good reason to believe in the power of confidence. All around us, we see the correlation between confidence and performance. Confident athletes are more likely to win. Confident candidates are more likely get elected. ⁴ And more confident businesspeople are more likely to succeed. But we should interpret these correlations with caution, remembering the dictum that correlation is not causation.

Just because two things are correlated does not mean one causes the other. Companies with large corporate headquarters find themselves defendants in more lawsuits than do small firms, but those lawsuits are not the cause of their size. A correlation can result from some third variable that affects both of them. In the case of confidence, it is entirely possible—and likely, as I have found—that a confounding variable, such as talent, strength, or ability, contributes to both confidence and to success.

Controlled experiments are necessary for testing causal relationships. Experiments eliminate the causal ambiguity that constrains correlational studies by directly varying the variable being tested while leaving other things constant. An experiment randomly assigns individuals to experimental conditions, in which the conditions differ only with respect to the independent variable—the variable the experiment is designed to test. Random assignment renders it unlikely that the conditions differ systematically, other than in the experimental treatment.

Research has tested the causal effect of confidence on performance. ⁵ Researchers began by asking people when they thought confidence improved performance most. Among the things they said was they thought confidence boosted performance on tests like math tests. So researchers put this hypothesis to the test.

The experiment had three conditions. The first two conditions manipulated the confidence of participants. Participants began by taking a pre-test and answering some survey questions. The high-confidence condition induced confidence by telling participants that their pre-test performance predicted they would perform well on the math test. Those in the low-confidence condition expected to perform poorly. No surprise, then, that those in the high-confidence condition had higher expectations than did those in the low-confidence condition.

Participants assigned to a third “observer” condition learned about the two experimental conditions and the manipulation. They learned that people in the two main conditions had been randomly assigned to get the encouraging or discouraging message. Researchers were careful to verify that they understood the random assignment, and that participants in the two conditions did not differ with regard to underlying math skill. Observers were to bet on how the test-takers would perform on the math test. The observers were ready to bet on the optimists.

However, contrary to their naïve expectations, there was no difference in performance between those induced to have high vs. low confidence. That is, in the absence of actual difference in ability, attempts to manipulate confidence did not have a significant effect on performance, even though our participants expected it to.

Because a math test might not have given confidence its best chance to shine, the researchers conducted additional experiments using different tasks. These included trivia quizzes, perceptual tasks, tests of athletic performance, physical endurance, and even boring visual tasks. Despite these efforts, evidence failed to support the claim that greater confidence causes stronger performance. And yet, observers expected confidence to heighten performance on each task. The results suggested that, at least in the situations we studied, people tend to exaggerate the performance benefits of optimistic expectations.

Confident Leaders

I cannot claim that confidence never contributes to performance. Indeed, there must be some times when it does, such as when greater confidence persuades you to enter a competition that you win. Confidence can also contribute to your success when it earns you credibility with others. Research shows how confident displays can boost others’ faith in you as a leader. ⁶ It leads others to think that you know the path forward to and can help the group be successful.

Confidence is also intimately bound to the exercise of leadership. Even if confidence does not actually influence your performance, it can increase others’ faith in you as a leader. In times of crisis like the global pandemic in which we find ourselves, the yearning for confident leaders is particularly poignant. We want people who can guide us out of our predicament. The problem is that would-be leaders can fake confidence, displaying more bravado than ability. Demagogues and charlatans may attempt to fulfill our yearning by pretending they know what the future holds.

Honest leaders know that the future is uncertain. The future is best thought of as a distribution of possible outcomes, some more likely than others. Mapping that uncertainty allows you to place smart bets that maximize expected value or minimize the expected loss. Listen for leaders’ acknowledgment of this uncertainty as a bellwether of trustworthiness. All of us should want—and ultimately need—leaders with the courage to confront uncertainty and use it to make wise decisions.

Despite the enormous benefits of acknowledging uncertainty, it is easy to find advice that counsels otherwise. Business consultant Rebecca Knight advised that leadership in a pandemic requires you to “force yourself to think positively.” ⁷ It is worth parsing what that means because there are different potential interpretations. The dangerous interpretation is encouragement to lie to ourselves and to others about the chances of success.

When Elizabeth Holmes told Theranos investors she could conduct hundreds of blood tests using only a drop of blood, her confident claims did not magically overcome the scientific and technological barriers to the success of her company’s technology. Her idealistically confident fervor persisted even in the face of setbacks. “You’ll get knocked down over and over and over again, and you get back up,” Holmes advised other young entrepreneurs. ⁸ “I’ve been knocked down a lot, and it became really clear that this was what I wanted to do, and I would start this company over 10,000 times if I had to.”

Holmes’s confidence did not lead to success at Theranos. In fact, her relentless optimism may even have undermined it. By silencing the pessimists and punishing those who had the courage to criticize Theranos’s products, Holmes made it more difficult for the company to come clean and confess its limitations.

Those whose lives depend on preparing for calamity are particularly aware of the benefits of pessimism. They obsess about all the ways in which things can go wrong and try to avoid becoming too overconfident about risk. Consider those who engage in risky activities like rock-climbing, skydiving, or big-wave surfing. Survival means understanding safety margins and not allowing themselves to become overconfident. In the words of big-wave surfer Brett Lickle, “As soon as you think, I’ve got this place wired. I’m the man! you’re about thirty minutes away from being pinned on the bottom for the beating of your life.” ⁹

Those in high-reliability industries such as air travel or nuclear power must similarly anticipate potential failures, errors, and accidents. We all owe a deep debt of gratitude to the careful and cautious planners who institute safeguards that improve the safety of air travel. Their contemplation of future accidents, errors, and failures is precisely what keeps air travel so safe.

Even if your overconfidence does not lead directly to loss of life, complacency puts you at risk. Every successful company is vulnerable of being overtaken by fleeter rivals. Business history is littered with the carcasses of wealthy, powerful companies that were overconfident of their continued dominance. Just think of Kodak and its failure to make the move to digital photography, or of Blockbuster Video, undone by online streaming video services. He was thinking of these sorts of risks when Andy Grove, former CEO of Intel, wrote, “Only the paranoid survive.” ¹⁰

If blind optimism puts us at risk, the let us then consider a second interpretation of the admonition to think positively. That is the encouragement to act by reminding people of their options, even in the face of adversity. Even in the darkest hour, it could be worse. There is still a path forward. Pull yourself together. We almost always have choices about how to act, how to invest, and how to respond to even the most difficult situations.

“We shall fight,” Winston Churchill told the British people in the depths of the Second World War. “We shall fight on the seas and oceans, we shall fight with growing confidence and growing strength in the air, we shall defend our island, whatever the cost may be.” Was Churchill lying to the British people about the perils ahead? No, Churchill was also forthright about the grim challenges they faced: “We have before us an ordeal of the most grievous kind. We have before us many, many long months of struggle and of suffering.” Nevertheless, action is preferable to despair.

Fake It till You Make It

A number of motivational speakers have advised would-be leaders to “fake it till you make it.” ¹¹ That is, imitate confidence, competence, and an optimistic mindset, so as to help realize those qualities. Stand up and fight for what you believe in, even when you are not sure you will be victorious. Or, in Shakespeare’s words, “Assume a virtue if you have it not.”

This is useful advice if motivating yourself to act is, itself, the heart of the achievement. If you would like to be more extraverted, friendly, or generous, then act that way. Actions can bring about the desired end state. Even if it can feel unnatural at the beginning, practice can increase your comfort with a new behavioral routine and even develop a new habit. ¹²

The display of confidence can be a winning strategy, but only if, in fact, you have what it takes. Bill Gates tells the story of his youthful excitement about one of the world’s first commercial computers, the Altair. ¹³ Gates and Paul Allen, who would go on to found Microsoft together, told Altair’s makers that they had a programming language that could run on the new machine. The Altair people expressed interest in seeing what the two young men had to offer.

Gates and Allen got to work writing the program they imagined, despite the fact that they did not actually have an Altair on which to test their code. When Allen flew to Albuquerque two months later, the program got its first test. Their code actually ran. Allen recounted, “A wave of relief surged over me. I almost couldn’t believe it had worked the first time.” ¹⁴ The people at Altair were impressed enough that they offered Gates and Allen jobs. In this case, Gates’s confidence was well placed because he was able to deliver.

Later in Microsoft’s history, the company’s record of delivering on its pledges was less consistent. The company’s overconfidence frustrated customers, competitors, and regulators. A Microsoft engineer coined the term “vaporware” in 1982 to describe the company’s habit of announcing new products that might never actually arrive. Tesla’s leadership has made similarly early announcements of plans for new products, vehicles, and features. This practice can discourage competitors from developing rival products, and so the U.S. Department of Justice worries that it can violate antitrust law. ¹⁵ When you cannot actually deliver on your confident promises, you reveal your overconfidence.

Tesla CEO Elon Musk got himself into trouble with another sort of overconfident announcement. On August 7, 2018, Musk posted on Twitter, “Am considering taking Tesla private at $420. Funding secured.” The tweet sent Tesla stock into gyrations and led to charges from the Securities and Exchange Commission that he had committed securities fraud. Musk ultimately settled with the SEC by paying a $40 million fine and agreed to step down as the chair of Tesla’s Board of Directors. ¹⁶

There are obviously other risks associating with “faking it” by promising more than one can deliver. Con men routinely behave as if they are qualified, capable experts in order to fool others. Frank Abagnale’s supreme confidence allowed him to get away with an astonishing series of con jobs. However, when Abagnale posed as a doctor, his confidence nearly proved fatal for a child handed to him for treatment. ¹⁷ As a leader, you must have a realistic sense of what you can fake and what you cannot. Abagnale frequently posed as an airline pilot, but he had the wisdom to avoid attempting to fly a plane.

Here, it is worth highlighting the crucial distinction between confidence and overconfidence. Confidence can obviously contribute to success, especially when backed by actual ability, as the story of Gates and Allen illustrates. But when confidence outstrips reality, it crosses the line into overconfidence. This is a dangerous territory in which you are liable to undertake tasks, enter competitions, or announce products that will end in failure. You may be sued by the U.S. Justice Department or by unhappy investors. The challenge, obviously, is to calibrate your assessments so that they are accurate. You want to be as confident as reality can justify, but not more. This challenge is particularly acute when you find yourself in a confidence arms race.

The Confidence Arms Race

A colleague of mine found himself in a conundrum when called to testify before the U.S. Congress on the subject of drug legalization. This UC Berkeley psychologist had studied drug abuse, drug enforcement, and the potential consequences of legalizing different classes of controlled substances. His research suggested that legalization of marijuana was unlikely to lead to substantial increases in the number of users, the number who became addicted to other drugs, or the number of fatalities due to overdoses. Being a good scientist, he attempted to present a balanced interpretation of his findings and admitted their limitations. Since marijuana was still illegal in the United States, he did not yet have evidence for what would happen if the nation legalized it. But his research did have evidence from other nations that had legalized marijuana, which offered useful evidence.

He was followed by heartfelt testimony from a mother who had lost her son to drugs. Sandra Bennett’s opinions were not informed so much by evidence as by anecdote, opinion, and dogma. She offered a simpler and more confident narrative: she railed against the idea that recreational drugs might be legalized and insisted that legalization would be a colossal mistake. There was a profound threat, she argued, to our children’s health. Legalization of marijuana would constitute an embrace of the drug culture, and was only the start of a larger effort to legalize all psychoactive and addictive drugs.

Speaking for the hundreds and thousands of parents who have lost children to drugs, I cannot understand how this country can listen for even one moment to those who advocate making illicit drugs easier to come by, particularly when most of the leaders of this effort are admitted drug users. Unfortunately, credibility is given to those disingenuous scofflaws when they are invited to the podium, paraded on TV, glorified on PBS, and asked to serve as consultants to government agencies and the media, who then parrot this misleading, deceitful and dangerous propaganda. ¹⁸

My colleague’s presentation was balanced and cautious, presenting both sides of the evidence. But it was Sandra Bennett’s passionate arguments that won the day, and Congress took no action on decriminalization. Bennett was like the one-armed economist that President Harry Truman reportedly asked for. Truman was tired of economic advisors who hedged their advice by following up any recommendation with a cautionary, “. . . on the other hand.” Truman, like most of us, appreciated someone who offered clear, confident, unambiguous advice.

Those who exude confidence, like Sandra Bennett, are naturally more persuasive. In my research, my colleagues and I have found that greater confidence makes people more credible. ¹⁹ Would-be leaders know that they stand to benefit from displaying confidence. Michael Lewis noted this dynamic in his profile of Barack Obama. Lewis wrote about the fact that the President must make difficult decisions with complex and uncertain outcomes. At the same time, Lewis noted the value in displaying confidence about those decisions: “After you have made your decision, you need to feign total certainty about it. People being led do not want to think probabilistically.” ²⁰ Leaders’ confidence inspires faith in their abilities and attracts followers.

Anticipating the Confidence Arms Race

Voters, investors, venture capitalists, and the rest of us routinely rely on confidence as a signal of potential performance. So long as the correlation between confidence and performance is positive, this might be justifiable. But my research also shows that confidence is not a reliable signal of competence. ²¹ Highly confident individuals are more likely to rise to positions of status or authority within groups, as one might expect, but they rise even when their confidence is not justified by their ability. In one study, we asked groups of Berkeley undergraduates to work together. Those who thought they were better at the task tended to talk first and talk more often, and other members of the group rated them as more influential. ²² This makes perfect sense, since the groups sought capable leaders. They did not have good ways to assess ability, but they did have clear signals of confidence. The only problem was that the most confident people in the group were not necessarily the most capable.

Too often, we rely on superficial displays of confidence rather than real experience, a proven track record, or a credible plan. Entrepreneurs who understand the persuasive power of confidence may try to bamboozle with braggadocio. And the more entrepreneurs escalate their displays of confidence, the less useful it becomes as an actual signal of future potential. It was surprisingly difficult for people in our experiment to tell who was really capable and who just had a big mouth. As a result, by elevating the most confident members, groups tend to select the more overconfident people.

Up until this point, I have highlighted the ways in which the dangers of overconfidence outweigh its benefits in decision-making. But the calculus is more complicated when it comes to questions of persuasion and leadership. The setting involves different incentives. When the goal is persuading others, confidence provides what is called social proof, which can give other people faith in one’s abilities or one’s argument.

The candidate that is most persuasively confident is more likely to win the day, but overconfidence comes with risks. One way to think of this dilemma is by considering the analogy to reverse auctions. In a reverse auction, sellers compete with one another to offer better terms to a potential buyer. One example is contractors bidding on a building contract.

Contractors know that they can increase their chances of winning the contract by bidding low, thereby expressing confidence in their ability to accomplish the project more efficiently. Confident bidders do, in fact, increase their chances of winning. But being overconfident and under-bidding the actual cost of construction could make you a victim of the so-called winner’s curse. Winning can be a curse if your winning bid cannot cover your costs. Unless you have good reason to believe that your costs are actually lower than those of your rivals, the fact that you have priced your bid lower than theirs suggests you may have bid too low. You should expect to lose money if you have agreed to complete the work for less money than it will cost you to do so.

In formulating strategy, it is worth remembering that if you get elected, funded, or selected, then you are likely to be accountable to deliver on your confident claim. If exaggerated confidence helped you edge out your rivals, then you are at increased risk of disappointing those who have placed their faith in you. Even if you can reconcile yourself with the ethics of such hypocrisy, you may nevertheless worry about potential complaints, lawsuits, or reputational damage. The confident assurances offered by Elizabeth Holmes did not assure the success of her company or protect her from prosecution for fraud.

The Lure of Confidence

Those who must choose between contenders for leadership face the other side of this “reverse auction.” The confidence arms race between contenders represents competing bids in an auction to earn buyers, followers, investors, supporters, or voters. Political candidates hope their confidence will persuade us to vote for them. Job applicants hope that their confidence will earn them the job and help them negotiate better salaries. Entrepreneurs hope that confident pitches will bring them investments from venture capitalists. Contractors hope that their confidently optimistic bids will earn them business.

When choosing which projects to support, whom to believe, or what to invest in, we should choose the person who is most capable and who speaks the truth. We should seek revealing tests of others’ claims. For instance, before selecting a contractor to renovate the house or to develop a new software system, we should examine their track record of performance. How often did they deliver on prior projects and contracts?

Many aspiring leaders may seek to display confidence that eludes falsification. ²³ They may promise phenomenal success without committing to a specific sales figure. They talk a big game without offering specifics or guarantees. Knowing this, wise investors and buyers will demand leaders who are clearer and more explicit in their claims, assurances, and promises. By seeking clarity, we can help raise the standards of what we expect from our leaders and reward honesty and well-calibrated confidence.

If you want to avoid being taken in by a confidence man, ²⁴ remember the perils of the winner’s curse. When we choose the more confident contender, we may be rewarding escalation in the confidence arms race and putting ourselves at risk of selecting the most overconfident who is most likely to disappoint us. How can you avoid falling victim to the winner’s curse? Resist persuasion by “cheap talk” forms of confidence such as assertive body language and vague optimistic claims. Instead, seek out specific claims and testable guarantees.

Bridgewater Associates, by some measures the world’s most successful hedge fund, seeks to document every employee’s credibility with a tracking system that awards them “dots” for making wise and well-calibrated forecasts that turn out to be right. This system is key to the company’s goal of being an “idea meritocracy” in which the best ideas rise to the top, regardless of who proposes them. ²⁵

Anyone who has hired a contractor for a construction or remodeling job has seen the dysfunctional dynamics of the confidence arms race at work. By selecting the bidder who promises to complete the work fastest and for the lowest price, you risk rewarding the overconfident. When my family collected bids to remodel the kitchen in our home, prices varied dramatically. The lowest bid came from a contractor who did not have enough of a record of accomplishment to give us faith in them. Instead, we preferred the contractor who had impressed us with their experience and long list of satisfied customers. When I asked how long our remodeling job would take, the contractor claimed they could completely remodel the kitchen in three weeks. Knowing how common it is for remodeling projects to go past deadline, I was wary their claim.

So I offered what negotiators call a contingent contract. ²⁶ I wrote their three-week forecast into the contract, along with a promise that, if they completed the work to my satisfaction in three weeks, I would pay a bonus. One of the benefits of contingent contracts is that they can create incentives for performance. The result was that the contractor completed the kitchen remodel ahead of schedule and I happily paid the bonus.

A contingent contract like this can offer confident contenders the chance to bet on their claims. If they believe what they say, they should be eager to take the bet. In employment contracts, performance-based bonuses can serve this function. Sometimes, for instance, entrepreneurs ask for generous funding, justified by confident claims of future performance. Investors can counter with contingent contracts whose value depends on that performance. There are many opportunities to structure incentives to reward good confidence calibration.

A Way Forward

Wise leaders have a way out of the dilemma between expressing confidence and telling the truth. Evidence suggests leaders can honestly confess the uncertainties inherent in the future without undermining their own credibility. ²⁷ The way to do this is to report the uncertainties with confidence. That is not a contradiction.

“I’m confident there is a 60% chance we will win the game” expresses assurance that a full analysis of all the relevant evidence suggests a 60% probability, but also that uncertainty remains. ²⁸ If you are worried about confessing uncertainty, imagine how much more you jeopardize your credibility by feigning certainty. You can acknowledge that there are many ways to fail, even as you pursue the actions most likely to achieve success. Jeff Bezos cautioned early investors in Amazon that there was a 70% chance they would lose their investment. ²⁹

Gather the best evidence you can, assess the expected values of different courses of action, and place your bets accordingly. Do not delude yourself about your chances of success. In 2016, Bezos said the following:

Outsized returns often come from betting against conventional wisdom, and conventional wisdom is usually right. Given a 10% chance of a 100 times payoff, you should take that bet every time. But you’re still going to be wrong nine times out of 10. ³⁰

That is why Bezos boasts that Amazon is “the best place in the world to fail.” ³¹ Investing in Amazon itself has turned out to be a good bet, despite the high risk of failure, because of its substantial up side. You can be confident that you have made the choice with the highest expected value, even while you accept the risk of failure.

The trust placed in you as a leader gives you a special kind of responsibility to provide just the right amount of confidence. There is a path forward through tough times. In this pandemic depression, that path will be long and painful. But we have choices to make, and honestly confronting them will help you and those who depend on you to choose the best course.