Small Business Overconfidence

The August Journal of Economic Psychology says that across 18 countries "subjective, and often biased, perceptions have a crucial impact on new business creation." For example, the strongest predictor of who starts a new business is "whether the person believes herself to have the sufficient skills, knowledge and ability to start a business."  But for those who do start a business, the higher such confidence, the lower their "approximate survival chances." Furthermore, "some countries exhibit relatively high rates of start-up activity because their inhabitants are more (over)confident than in other countries."

GD Star Rating
Tagged as: ,
Trackback URL:
  • michael vassar

    Did the Journal also affirm that the sky is blue?
    How could these things *Not* be the case?

  • Douglas Knight

    michael vassar,
    it is not surprising that within a country entrepreneurs are more confident about their entrepreneurial skills. The last claim is surprising, if for no other reason than that it seems to contradict the widely reported claim that cross-national variation entrepreneurship is controlled by variation in the social stigma of failure.

  • g

    Social stigma of failure and overconfidence might be quite strongly negatively correlated. The more you hear, and the more negative it is, about failed businesses, the less likely you are to think blithely “yeah, sure, I could do that”. (This is mere guesswork; I have no actual evidence.)

  • Barkley Rosser

    What is perhaps surprising, and way underreported is how vastly overconfident most entrepreneurs are and how many of them fail. I think we do not like to publicize this, as we need entrepreneurs to keep the economy growing. Publicizing too much how poorly so many of them do might not be a good thing for the economy. All the better to let them wallow in their self-confident ignorance.

  • g

    If, as claimed, over-over-confidence makes entrepreneurs more likely to fail, then what we want is a precisely calibrated level of overconfidence…

  • Robin Hanson

    Barkley, that comes awful close to “exploitation.”

  • j

    Personal knowledge of entrepreneurs is a prime factor. You see what they are doing and soon you arrive to the conclusion that I could do it better than this idiot. Imitation is also a factor.

  • The Decision Strategist

    This is kind of funny in a way. You need a minimum level of confidence to start a business, but anything more than that decreases your chance of success.

    My biggest question is more about what they are measuring versus what they are concluding. It makes perfect sense that arrogance and inflexibility would be negatively correlated with success. Confidence and adaptability, on the other hand, are probably positively correlated.

    Maybe I’m splitting hairs, but to me a willingness to listen to other points of view and flexibility is crucial to success.

  • ungoverned

    Any more than a certain amount of confidence probably indicates a lower level of self awareness, or an inability to apply negative feedback to ones own behaviors.

  • Bob

    I agree with ungoverned, and suspect that this ties back to the Planning Bias discussion – it seems likely that detailed plans become increasingly inaccurate the more overconfident the planner is in their ability (and the less apt they are to apply general experience).

    I should think further on Robin’s “exploitation” point but it seems to assume that entrepreneurs earn below-market returns on their effort (in fairness, there is some empirical evidence that this is the case, at least for economic returns). My observation is that many people fail to see the potential associated with risk/change because they cannot easily imagine the many potential positive outcomes that can arise from the change, that is, the probability of the planned path to success may be small, but starting the process will often reveal other desirable paths yet unseen.

  • Michael Sullivan

    If you believe that being an entrepreneur is such a bad deal on the whole, I’d be interested to know how you refute Paul Graham’s arguments to the contrary at least for young people just out of college or grad school. Basically that if you can start with a small amount within a few years of graduation, you have relatively little to lose: worst case is that you took a very interesting job that paid you much less than other opportunities and ended up broke but young and with a fairly interesting resume: IOW, no worse off than your compatriots who overspent like mad or went into the peace corps or whatever.

    People who are middle aged and on track to retire comfortably choosing to invest their life-savings in a speculative venture would fit Barkley’s characterization better. That said, I think the main reason entrepreneurs do so poorly on the whole is not because it is intrinsically risky or poorly paid endeavor, but because so many of the people who try it are so spectacularly ill prepared.

    There are no barriers to entry for the foolhardy, only for the informed. If everyone who wanted to be a research scientist got hired out of to learn on the job and see what they came up with, most of them would fail spectacularly. 90% of research scientists do nicely, because hardly anybody ever gets paid to do that job until they’ve already proven they can do it in grad school.

    Not so entrepreneurs. The only way to find out that you stink is to roll the dice and take your chances.

  • Barkley Rosser

    There are clearly other factors involved as well. One of them is the pleasure of being one’s own boss, even though it is also well known that entrepreneurs, especially the successful ones, put in a huge amount of hours. Looking at that even further lowers the expected purely monetary return, worsening the potential exploitation. However, there are an awful lot of people out there with bad bosses who really relate to that old Johnny Paycheck song, “Take this job and shove it.” One may not be making much money, but if one can keep one step ahead of the creditors, does not need to pay for kids in college, and really enjoys what one is doing, it may be worth it, although again, the evidence suggests that there is a lot of overconfidence and poor information out there.

  • Robin Hanson

    Michael, I don’t recall Paul Graham citing any statistics on expected payoffs.

  • michael vassar

    Michael Sullivan: Great point about entrepreneurs vs. research scientists. I think that you are basically correct, but I have long said the same thing about parents vs. doctors. Judith Rich Harris shows that within cultural norms parenting seems to have no statistically detectable effect on life outcomes, but even if it is fairly easy to have large atypical positive effects on your kids we would expect to see it very rarely because parents, like entrepreneurs, have no barriers to entry (pun sincerely not intended). If even doctors have difficulty conveying statistically significant health benefits in aggregate despite their (fairly poorly chosen) barriers to entry, and despite strong validation on particular medical techniques, I would expect much worse from parents.