Better Than Average Effect: Why Managers Can't See Their Own Compliance Gaps

Ask almost any driver how they compare to everyone else on the road, and they'll put themselves above average. Safer. Sharper reflexes. Better judgement than the person two cars ahead. Everyone's convinced they're the exception, and the evidence has been piling up since 1981, when a Swedish psychologist called Ola Svenson first put the question to a group of drivers directly: compared with the others in this study, where do you rank for skill and safety?¹ Logic says the answers should cluster around the middle. A median splits any group exactly in half by definition, so at most 50% of a group of drivers can genuinely be safer than the median driver in it. No more. Roughly 80% to 90% of Svenson's drivers put themselves above the average anyway, and a 2023 replication with over 1,200 drivers found exactly the same pattern, across three different ways of asking the question.² It's one of the most reliably repeated findings in the field.

What you might find shocking is that the same bias shows up in managers. A 2022 study, 706 owner-managers surveyed over six months, asked business owners to rate their own regulatory compliance, then separately asked them to rate their belief about their competitors' compliance.³ It wasn't a logistics study, it looked at English accommodation providers, not haulage yards. But sit with that for a second, because I think the finding carries over all the same, and it's worth going deeper into why.

Those owner-managers rated their own compliance at 6.04 out of 7. They rated their competitors' compliance, on the same scale, at 3.60.³ The two scores moved in opposite directions, the more confident someone was in their own compliance, the lower they rated everyone else's.³ That's the driving study again, almost number for number, just with "regulatory compliance" written in where "driving skill" used to be. Managers have the same cognitive bias as drivers. This study found it directly: given a chance to rate themselves against their peers, on a question with no hard external answer attached to it, they came out comfortably, confidently above the pack, and so did almost everyone else answering the same question about themselves.

So here's the real question worth sitting with. If people can't accurately judge something as personal and frequently tested as their own driving, what happens when they're asked to judge something far less tested and far more ambiguous: their own regulatory compliance?

The bias has a name

There is no realistic version of an industry where that self-versus-competitor gap reflects reality. This is a documented, named phenomenon, the better-than-average effect, and it's bigger than this one study.

There's a reason that gap might be self-sealing rather than self-correcting. The same 2022 study raises, without testing it directly, a related possibility worth sitting with: that publicising non-compliance, naming firms that have been caught breaching a regulation, might not correct this bias so much as feed it.³ If an owner-manager already assumes, without real evidence, that competitors are the less compliant ones, a well-publicised enforcement case doesn't contradict that assumption. It confirms it, handing an existing, unfounded belief a concrete example to point to. The study's authors are explicit that this is a speculative concern raised in discussion. It's worth holding as an open question rather than a settled one: does publicising enforcement correct the better-than-average effect, or does it just give it something to point to?

It isn't random which traits trigger it hardest, either. Research on what's sometimes called the hard-easy effect finds the bias is strongest exactly where self-judgment is subjective and gets no immediate outside correction, and weaker or reversed for traits that are rare, difficult, or regularly tested against reality.⁴ Driving skill fits the first category for most people, you rarely get hard, specific feedback on it unless something goes wrong. Regulatory compliance fits it even more precisely. Nobody gets a weekly scorecard on whether they correctly interpreted every clause of every regulation that applies to their operation, unless an audit or an inspection actually happens.

It isn't what you know, it's what you think counts as right

The same 2022 study found something that looks, at first, like a separate issue, but is really the same mechanism from another angle. Respondents' actual knowledge of the relevant regulations had no measurable effect on how compliant they rated themselves. Their perceived knowledge, how knowledgeable they believed themselves to be, had a small effect. Confidence moved the self-assessment. Correctness didn't.³

The study's comments explain why, more plainly than the regression tables can. Respondents said regulation was generally designed to make them "do the right thing," created with "good intentions," and that if they did what they personally believed was "right," meaning just or fair rather than correct according to the regulation's actual text, they would likely comply by default.³ That's not a casual mistake. It has a name in judgment research: attribute substitution. When a question is hard to answer directly, people often substitute an easier, related question and answer that instead, usually without noticing the swap.⁵ "Is this specific action correct according to the regulation" is a hard question, it requires knowing the regulation's exact wording and applying it to a specific situation. "Does this feel fair" is the easy stand-in, available instantly, no lookup required. Cass Sunstein's work on moral heuristics found this same substitution running through legal and moral reasoning generally, not just compliance.⁶ The two questions feel like one question. They aren't.

Why you can't benchmark yourself from inside your own firm

There's a reason this particular blind spot is so hard to self-correct. Smaller firm owner-managers are, by existing research, among the least likely to seek outside help or build wider networks, because they operate in relative isolation and often don't realise how limited their own regulatory picture actually is.⁷ One respondent in the 2022 study put the underlying problem plainly, when asked to assess competitors: "how do you judge what other competitors offer unless you have first-hand knowledge of their buildings and practice?"³ That's an honest admission that the information needed to self-correct the bias was never available in the first place. You can't benchmark against a reference point you can't see.

It would be convenient if the fix were more training, or clearer guidance leaflets, but the same research found owner-managers tend to resist the formal events trade bodies already run to provide exactly that.⁸ Given the knowledge-confidence gap above, that resistance makes a grim kind of sense: if you already believe your compliance is well above average, sitting through a training session is solving a problem you don't think you have.

It's worth pausing on the strongest version of the other side of this argument. Self-assessment isn't always baseless. Operators do get some real signal over time, insurance claims, complaints, the odd inspection or prohibition notice, and none of that is nothing. The 2022 study's own statistical models, for what it's worth, only explain about 10% to 11% of the variation in how people rate their compliance.³ That's a real, measurable effect, but it also means roughly 90% of what actually drives someone's self-assessment is something other than the psychological factors described here. This piece is making a narrower claim than "self-assessment is worthless." It's making the claim that self-assessment is systematically optimistic in a specific, well-documented direction, and that an unverified gut sense of compliance, on its own, isn't a safe substitute for an outside check.

What actually closes the gap

The research behind this piece doesn't end on the diagnosis. Its own authors, looking at what might actually move the needle, found that passive channels, leaflets, webinars, the training sessions owner-managers resist, had limited effect, while they suggested direct, active engagement, someone physically visiting the business and looking at what's actually happening on the ground, was the more promising route.⁸ That's a specific, checkable claim about what works, not a generic call to "raise awareness," and it describes an audit rather than a leaflet.

The reason an outside check works where self-assessment doesn't isn't that the operator is careless or being misled. It's that self-assessment was never going to catch this particular error. The better-than-average effect, the common-sense substitution, the missing reference point, none of them are solved by trying harder to be honest with yourself. They're solved by someone else looking at the actual operation against the actual rule, because that's the one thing self-assessment structurally can't provide.

The takeaway

The evidence here supports a specific, narrow claim: confidence in your own compliance is not, on its own, good evidence of your own compliance, and that gap is wider and more common than it feels from the inside, for reasons that have nothing to do with how careful or honest an operator is. The fix isn't more self-scrutiny. It's an outside measurement, the one thing the better-than-average effect reliably prevents anyone from generating for themselves.

If you'd like an outside, independent read on where your operation actually stands against the rules that apply to it, get in touch.

References

1. Svenson, O. "Are we all less risky and more skillful than our fellow drivers?" Acta Psychologica 47, no. 2 (1981): 143–148 (source of the original finding that roughly 80–90% of surveyed drivers rated themselves as safer and more skilled than the median driver in their comparison group). 2. Koppel, L., Andersson, D., Tinghög, G., Västfjäll, D., and Feldman, G. "We are all less risky and more skillful than our fellow drivers: Successful replication and extension of Svenson (1981)." Meta-Psychology 7 (2023) (preregistered replication, n=1,203, confirming the original effect across three response-scale formats). 3. Betton, M.E., Branston, J.R., and Tomlinson, P.R. "Perceived regulatory compliance in micro-firms: The case of English accommodation providers." Journal of General Management 0, no. 0 (2022): 1–14 (source of the own-firm versus competitor compliance figures, the knowledge/confidence findings, and the unsolicited respondent comments; studies English accommodation micro-firms specifically, not transport, and surveys perceived rather than actual compliance throughout). 4. Kruger, J. "Lake Wobegon be gone! The 'below-average effect' and the egocentric nature of comparative ability judgments." Journal of Personality and Social Psychology 77, no. 2 (1999): 221–232; and Moore, D.A., and Healy, P.J. "The trouble with overconfidence." Psychological Review 115, no. 2 (2008): 502–517 (sources, cited within source 3, for the hard-easy boundary condition on the better-than-average effect). 5. Kahneman, D., and Frederick, S. "Representativeness Revisited: Attribute Substitution in Intuitive Judgment." In Heuristics and Biases: The Psychology of Intuitive Judgment, edited by T. Gilovich, D. Griffin, and D. Kahneman, 49–81. Cambridge: Cambridge University Press, 2002. 6. Sunstein, C.R. "Moral Heuristics." Behavioral and Brain Sciences 28, no. 4 (2005): 531–542 (extends attribute substitution specifically into moral and legal reasoning). 7. Lynch-Wood, G., and Williamson, D. "Understanding SME responses to environmental regulation." Journal of Environmental Planning and Management 57, no. 8 (2014): 1220–1239 (cited within source 3 as evidence that smaller-firm owner-managers are among the least likely to seek outside help or engage in wider networks). 8. Bennett, R.J., and Robson, P.J.A. "Intensity of interaction in supply of business advice and client impact: a comparison of consultancy, business associations and government support initiatives for SMEs." British Journal of Management 10, no. 4 (1999): 351–369 (cited within source 3 as evidence that owner-managers generally resist formal training events trade associations provide; source 3's own authors recommend direct, active engagement over passive channels as the more promising alternative).