Confidence is complicated. Ted talks try to teach you how to get it; parents want it for their children; for those who don’t have it, it seems almost impossible to acquire. Entrepreneurs, social-media influencers and job applicants will project certainty even when internally racked with self-doubt – which is perhaps no surprise, given that virtually every talk on confidence is about how to acquire more of it, not less.
And yet too much confidence can have serious consequences. It contributed to the 2008 financial crisis and the Post Office scandal. In my years working as a behavioural scientist, I’ve measured the confidence levels of thousands of civil servants, and created case studies of how overconfident decision-making in government leads to failed policies. I have also seen how it can affect senior decision-making at the highest levels – at the Covid inquiry Matt Hancock, the health secretary in 2020, was described as responding to the crisis with “nuclear levels” of overconfidence.
Most of us are thankfully not tasked with responding to a pandemic, but none of us are immune to the effects of too much swagger. I bet you can think of someone in your work or personal life whose confidence is not commensurate with good outcomes. If so, they are not alone. At the Behavioural Insights Team, where I work, a new study shows that overconfidence is widespread among the UK public.
We gave a test to 2,000 adults, asking them 30 general knowledge questions, and how sure they were in their answers. We found eight in 10 were overconfident – by which we mean more sure their answers would be correct than turned out to be the case. Strikingly, 84% of baby boomers (people aged 60 to 78) were overconfident compared with 75% of gen Z (those aged 18 to 27). This 10-percentage-point gap was twice as large as the distance between how overconfident those who hold a university degree were (79% overconfident) compared with those whose level of qualifications was GCSE or lower (84%).
Not knowing things is fine – the key is knowing what you don’t know. Gen Z were pretty confident (82%) they knew the real name of famous YouTuber MrBeast – but that was in line with reality: 86% correctly said Jimmy Donaldson. This close correspondence between correctness and confidence is called being “well calibrated”. In contrast, boomers were 61% confident but only 36% correct – classic overconfidence. With age, of course, comes experience and for most of us that’s a net benefit. But it also seems that most of us age into greater overconfidence.
Confidence plays a trick on all of us. We see the certainty of high achievers such as the Olympic pistol shooter Kim Yeji (who said she had “nothing to improve” when asked how she was preparing for the recent Olympics) and we want to emulate it. But the confidence of Olympians is justified by the fact that they really are among the best in the world. Olympians also benefit from regular, direct feedback about their performance, which gives them an accurate sense of how good they are relative to others. Most of life is not like that.
If overconfidence is overrated, what is the alternative? Instead of pursuing an ever-increasing sense of ungrounded self-assurance, aim to be well calibrated. Know what you know, and what you don’t know.
If you are bringing up children, celebrate their achievements, but emphasise that you are proud of their growth – rather than implying they’ve reached a terminal destination. “You are doing great, and will be better in the future” is the balance to strike. In the workplace, encourage and value the input of less outspoken colleagues. The underconfident are an essential counterweight to the overconfident majority, and helpfully remind the rest of us to double-check our assumptions instead of blindly charging ahead. Underconfidence is also a better starting place than overconfidence – if you’re underconfident, the world can eventually teach you that your judgments are pretty good, and that you should back yourself more. The trouble with overconfidence is that it makes people feel bulletproof – so certain they are right they don’t notice when they have made mistakes, so they don’t change their behaviour.
Interestingly, in my experience running calibration workshops with dozens of organisations, I have noticed that the more underconfident groups tend to have a higher number of younger women. And yet in many organisations, you find that leadership positions become increasingly male-dominated – no wonder overconfidence is such a pervasive problem.
Focusing on calibration means, for most people, reining in a natural tendency towards overconfidence. Forget the advice to stick your chest out and blag your way through. Overconfidence is a fool’s gold that looks like the real thing, which is why so many of us are convinced by it. But in reality it isn’t something to aspire to – in public life and in our personal lives, it tends to deliver disappointing and occasionally catastrophic results.
Think of calibration as basic cognitive hygiene – much as you expect to get check-ups for your teeth or eyes, you might consider regular tests to monitor the accuracy of your judgments. After all, like your gums and vision, it tends to get worse, not better, over time.