your average entrepreneur
Brayden
Typical stories of entrepreneurs in American culture center on the individual’s creativity, personal vision, risk-taking, and other personal characteristics. We laud the entrepreneur as someone very special who knows how to “think outside the box.” In reality, entrepreneurs are not much different than your average middle-class, suburban next-door-neighbor. In fact, your neighbor may very well be an entrepreneur. Scott Shane makes this point in his new book, The Illusions of Entrepreneurship. The book brings varied data sources together to correct common misconceptions about entrepreneurship. Shane’s book contains a smorgasbord of data and facts about entrepreneurs, many of which are quite surprising.
Rather than summarize the entire book in one post, I thought I’d talk about a common myth that Shane debunks – the myth of the genius entrepreneur. Entrepreneurs, we believe, have unique qualities that qualify them to rise above the fray and come up with brilliant ideas that innovate the way we do business. Sergey Brin and Bill Gates are role models for this type of entrepreneur. They couldn’t help but be entrepreneurs; it almost seems destined.
Shane finds that most entrepreneurs are very average, and if they’re not average, they’re unique in unexpected ways. Here is the typical entrepreneur (pg. 41):
- “He is a white man in his forties.
- He is married with a working spouse.
- He attended college but might not have graduated.
- He was born in the United States and has lived here his whole life.
- He has spent much of his life in the town where he started his business.
- He is just trying to make a living, not trying to build a high-growth business.
- He worked previously in the industry in which he started his company, something like construction or insurance or retail.
- He has no special psychological characteristics.”
With regards to the latter, it’s difficult to identify any psychological characteristic that is a good predictor of being an entrepreneur. In fact, our best predictors are demographic characteristics. Being a white man greatly increases your odds of starting your own business. Being unemployed for a significant period of time improves your chances of being an entrepreneur. Oddly enough, “people who dealt drugs as teenagers are between 11 and 21 percent more likely than other people to start their own businesses in adulthood,” a fact which probably ties into the causal role that past experience plays in shaping future entrepreneurs. As Fabio notes, structural network characteristics make you more likely to be an entrepreneur. All of these things have greater predictive power than attitudinal or psychological characteristics.
One of the main messages of Shane’s book is that it’s time we stop thinking about entrepreneurs as somehow being more special or more psychologically qualified than others. The truth is that entrepreneurs are pretty average.
I have not yet read Shane’s book, but I think it’s also important to stress the role of the intra-preneur. I would argue that intrapreneurs, individuals who practice entrepreneurship within a larger corporation rather than start their own – further hit on the idea that entrepreneurial characteristics are more common than it may seem. Perhaps instead it is their ability to lead well that makes the most notable entrepreneurs stand out the way they do.
michael@oddjobia.com
February 19, 2008 at 7:18 am
I have also not read the book, but I think it’s important to point out that it’s not average entrepreneurs who create value (in fact most entrepreneurs fail in the two years following the creation of their firm, if I remember well), but the 1% that get the most successful. The Pareto distribution is extremely skewed, and you might find young black innovators or other more original profiles among these exceptionnal creators. (this is just an uninformed guess, does he have data about this?)
Markss
February 19, 2008 at 10:13 am
Averages are informative (as is the book, I am sure), but can be misleading (well, of course depending on what one is trying to explain). If we took samples of the Milky Way, we’d end up with an average temperature of absolute zero and no life. As suggested by the comment above, the skewed outcomes and extremes of entrepreneurship are interesting.
tf
February 19, 2008 at 3:13 pm
I know entreprenuership studies can devolve into definitional arguments, but does Shane use the definition of ent. as someone who starts a firm? If so, then I am not surprised that the average ent looks like the average person given that there have been millions of firms that have started. I think I am siding with Teppo on this one.
fabiorojas
February 19, 2008 at 3:37 pm
“The Banality of Entrepreneurship?”
fabiorojas
February 19, 2008 at 4:05 pm
Well, of course, there are good reasons to study the outliers too! If you’re primarily interested in studying innovation and change, then your best bet would be to look at the most novel entrepreneurial organizations, i.e., rare events. But if you’re interested in understanding what most people’s entrepreneurial experience is like, then you should seek a more representative sample.
The problem that Shane notes is that our common conceptions of entrepreneurs are shaped by those rare events and not by typical experience. Thus, people who start a new business have a very unrealistic view of what sorts of problems/challenges they will face or what kinds of outcomes to expect. We haven’t done our homework in studying the most basic forms of entrepreneurship – the family business, the local startups, etc. The other reason we should be interested in the average entrepreneur is because these businesses constitute the vast majority of for-profit businesses in the world. As Howard Aldrich has argued, our views on organizations are slanted because our research focuses narrowly on the rare successes, i.e., Fortune 500 firms, Silicon Valley success stories. Our knowledge of organizations is plagued with sample selection bias. Knowing more about the businesses that constitute the largest share of the organizational pie can only improve our theories of survival, growth, etc.
I suppose that the case(s) you choose to study ultimately comes down to what theoretical questions motivate you. If you’re interested in understanding processes of radical, transformative change, studying average entrepreneurs may not be the best place to start. But even then, our knowledge of average entrepreneurs could be helpful. How do you know what’s different about the rare events unless you’ve adequately studied the most common events first?
brayden
February 19, 2008 at 4:15 pm
European research indicates that the single most important defining factor is whether one’s father was an entrepreneur. It’s almost hereditary. Awkward that this factor is not mentioned in the above list. This doesn’t play a role among american entrepreneurs — or did the researchers forget to control for this factor?
Olivier
February 19, 2008 at 5:03 pm
I didn’t list all of the factors Shane mentions in the chapter about entrepreneur characteristics. But I’m pretty sure that he didn’t include parental employment status as a predictor.
brayden
February 19, 2008 at 5:05 pm
Maybe the issue is that firm starting is so common that it’s uninteresting by itself, except that it selects for people who really want to be their own boss, have supportive networks, and can accept risk (e.g., the prior drug use correlate). A guy who starts a hot dog cart and Bill Gates would be justifiably considered “entrepreneur.” Maybe firm starting is really a “low level” concept and that ent. studies needs to considered more “advanced” concepts like “firm starting in a new industry,” or “starting hybrid firms.”
I’ve always thought the most interesting side of ent. is when people do something that the average person would consider unprofitable, or to build a firm around an innovation that no one foresaw. That’s the difference between the guy who starts a restaurant chain and the guy who starts Apple computers. The first guy is simply taking a risk in a fairly established market, while the second guy is really trying to invent a new market. I bet there are some interesting difference between these two types of firm starters. Or does Shane show that’s an incorrect belief?
fabiorojas
February 19, 2008 at 5:49 pm
I’m only halfway through the book. Maybe that will be the topic of my next post….
brayden
February 19, 2008 at 6:22 pm
Fabio: The Shane book, as Brayden suggests and based at least on the table of contents, has some of that intuition (differences in types). Then the Aldrich and Ruef book — Organizations Evolving (remember Howard shared those figures at the comparative conference, I thought they were from this book, the updated edition) — I believe also has the numbers for just how unlikely the extreme events (Apple, IPOs, venture funding, etc) are; though granted, not every entrepreneur has those extreme goals from the outset (so, the stats throw every mom-and-pop, which is most of what entrepreneurship is, into the mix).
tf
February 19, 2008 at 6:52 pm
Teppo – That’s actually where Shane and Howard might have some disagreement. Howard treats all organizations as if their natural lifecycle is to become big (or else they fail). Shane shows that most entrepreneurs have no intention of expanding beyond a small business.
brayden
February 19, 2008 at 6:57 pm
Right, that indeed was the flaw in the VSR intuition suggested by Aldrich in those figures — the actual numbers would be interesting to see (though, probably still staggering and easily conducive to explaining origins via randomness — I remember all the dog-and-pony shows I evaluated when I worked for a VC firm, lots and lots of variation with some potential ‘hits’ – of course hard to judge a priori).
tf
February 19, 2008 at 6:59 pm
This discussion makes me want to toss out the word “entrepreneur” and its connotations and replace it with less romantic words like “firm creator.” Boring? Yes, but once we say that, we see Shane’s results in a less shocking light and Aldrich’s arguments seem to miss the point.
If the capitalist process requires a huge number of firms to yield a few winners, then the Shane approach is really only the beginning. It’s like saying that the study of athletes is boring because lot of normal people try out athletics on a small scale. That’s certainly true and worthy of study, but the real motivation for studying athletes is to study excellence, not the average. So while Aldrich is correct, it actually makes the study of outliers even more important. If any joe-schmoe can start a firm with modest goals, and many do, then what is it that distinguishes the folks who make the markets?
Last rehash: Perhaps what I am saying is that most markets are relatively fixed entities and most firm creators are competing for space within an economic domain by exploiting some niche. That’s the normal economic process and that’s what Shane studies.
What really interests me are the people who wreck markets, Schumpeter style. Aldrich says I shouldn’t care because they are rare. But I think that’s an important component of economic growth because some firm is going to be pioneer in a market that increases our total welfare. We should definitely care about that!
fabiorojas
February 19, 2008 at 9:36 pm
Agreed. There is lots of research coming out in that vein (creating markets). The new journal Strategic Entrepreneurship Journal for example is publishing this type of research. And, all the Austrian stuff — that our O&M friends know lots and lots about — of course is also closely related.
tf
February 19, 2008 at 11:09 pm
But there is something missing in this discussion. In research on artists (or scientists or any cultural producer) careers, we know that you can start with a normal distribution of talent, and end up with a scale-free (with a “badly behaved second moment” [sorry I've hanging around physicists lately]) distribution of outcomes (rewards). We also know that the only plausible micro-level mechanism that can produce such a transformation is some sort of cumulative advantage process. So it is a foolhardy venture to guess backwards from the distribution of rewards (or influence or market-breaking prowess) to the initial distribution of talents without taking into account the Mertonian Mathew-effect, since it is precisely the interesting social process that makes that distribution not correspond to the original, and produces the illusion that the actual distribution of talents is somehow skewed, with lots of untalented people and a few gifted ones. But if the original distribution of talents is indeed normal, studying the “average” as Shane does is useful. Studying the post-facto successes is misleading on the other hand.
Omar
February 19, 2008 at 11:51 pm
Thanks Omar. You said that better than I did.
But I should also point out that Shane isn’t just studying the mean or the median entrepreneurial outcome. He’s also examining the modal outcome. Another reason to collect more data on these average organizations, aside from illuminating the theoretical concerns that motivate us, is so that our research has more practical use for real entrepreneurs. I’m just a sociologist (so what do I know?) but it seems like if you’re going to teach business students about entrepreneurship, you should know something about the everyday experience that the majority of entrepreneurs face. You can’t explain to inexperienced students what to expect as they begin the process of a business start-up if you only study the most novel businesses. It will likely be completely foreign to their probable experience in the real world.
brayden
February 20, 2008 at 12:02 am
That (representativeness, ability to draw unbiased inferences, etc.) is also the justification of that Hannan, Freeman, Carroll et al give for studying entire organizational populations. Coincidentally, they are not very popular in strategy classes.
Omar
February 20, 2008 at 12:21 am
Students may not want to hear that there is a high likelihood of failure, but they should still have to hear it!
brayden
February 20, 2008 at 12:23 am
Very interesting discussion on Shane and entrepreneurship. Some people called for an update on accumulated findings related to entrepreneurship: In addition to the Aldrich and Ruef book, there is a recent review series called “Foundation and Trends in Entrepreneurship” edited by David Audretsch and Zoltan Acs that provides much of the generalized “stylized fact” on entrepreneurship. I can recommend in particular Simon Parker’s issue on The Economics of Entrepreneurship and Kim and Aldrich “Social Capital and Entrepreneurship.”
In regards to Fabios call for distinguishing “the innovative start-up” and “the average start-up”: There is much suggested in the literature that these two should be distinguish but up until recently, lack of reliable data did not allow researchers to test for systematic distinctions between the two. Through random-digit dialing, the Panel Study of Entrepreneurial Dynamics (PSED) project was able to sample fairly large random sample of “nascent” firms – i.e. firms in the start-up stage. Various publications related to this project highlights the heterogeneity of firms sampled, one study in particular (Samuelsson & Davidsson, Small Business Economics 2008) found systematic differences between start-ups based on “innovative” vs. “replicative” business opportunities. Teppo, I think “the actual numbers” indicate that less than 10% of all start-ups are based on some idea that is (fairly) unique while the rest is a variant on something pre-existing.
In regards to ability: There is little robust evidence to data that some latent “entrepreneurial” trait exist (if so it would indicate that the World’s entrepreneurship courses would be futile, perhaps to the sadness of the 500- or so recently endowned chairs in Entrepreneurship…). However, some studies indicate that entrepreneurs have on average higher IQ than employees (For example De Wit & van Winden, Small Business Economics 1989). For measures of unobserved ability, see the recent book by Dutch economist Mirjam van Praag. Some recent projects investigates genetic or hereditary preferences for entrepreneurship (see e.g. one of the last chapters in the Aldrich and Ruef book, or Nicolaou and colleagues in Management Science). This line of work suggest that genes might increase the probability of engaging in entrepreneurship by, for example, increasing the probability of seeking specific environmental stimuli that increase the likelihood of engaging in entrepreneurship, increasing the probability of selecting into environments that are favorable to entrepreneurship, or by psychological differences such as extraversion and internal locus of control that predispose people to engage in entrepreneurship.
In regards to the sociological aspects of parental heritage: There is a lot of evidence that parental heritage is important for entrepreneurship. I think I remember some of this evidence is also included in the Shane book (see e.g. Carroll & Mosakowski ASQ1987 or the recent review by Kim and Aldrich in Foundation and Trends in Entrepreneurship on how parental self-employment influences children’s propensity to become self-employed. Interestingly enough, I remember Aldrich saying something that these evidence were well known by sociologists already in the 1950s (small business ownership showing one of the strongest probabilities of occupational heritages from fathers to sons) but are not as apparent in the most recent U.S. data (e.g. PSED 2 or Kauffman firm study). An explanation would be that for today’s young, entrepreneurial role models are readily available in popular media, school, or even on TV shows, decreasing the importance of parental role models.
So, if less than 10% of start-ups are based on a fairly unique idea, the role of ability is not yet known, and role models and social network is important, what then motivates “the average entrepreneur” described by Shane?
Some research indicate that entrepreneurs are quite aware of the average low returns to entrepreneurship but that they have individual preferences for risk – they are “skewness-lowers” (Asterbro, Economic Journal 2003) or they exhibit systematic overconfidence (Shepherd and colleagues, Management Science 2006). The Management Science issue also includes a paper by Shane and colleagues on funding that is related to Brayden & Teppos discussion on motivation/self-selection both on the behalf of entrepreneurs and on VCs.
One large part of the explanation seems to be that (the average) entrepreneur is actually _not_ primarily profit oriented – self-employed people reported much higher job satisfaction on various measures in many studies – perhaps the most comprehensive of them being the study of German, Swizz U.S. and U.K. workplaces by Bruno Frey & Mattias Benz that recently appeared in Economica).
So – Entrepreneurship makes people happy? Or, the average start-up is life-style motivated but only the extraordinary start-up is growth/ profit oriented enough to pass through Aldrich’s VSR model.
Karl
February 25, 2008 at 3:19 am
Thanks, Karl, for summarizing all that research for us.
tf
February 25, 2008 at 3:23 am
Yeah, thanks for the great summary Karl!
brayden
February 25, 2008 at 5:42 am
Yeah, seems I have nothing else to do. Did I mention my papers are increasingly made up of literature reviews? :)
Karl
February 26, 2008 at 5:12 am
[...] my earlier post about Scott Shane’s book, The Illusions of Entrepreneurship, I reported Shane’s [...]
your average entrepreneur fails « orgtheory.net
March 4, 2008 at 5:03 pm
Thanks for the good suggestions. I have not read the book yet but it is on my list now. I am not a traditional entrepreneur, but I have been selling my jewelry at flea markets believe it or not and doing very well. If you are considering it as a way to get started, I highly recommend a DVD that showed me how to do it successfully. It talks about making $1,000 per day at flea markets and I have been very successful since I started using these techqniques. Take a look at http://www.salesandmanagementsolutions.com/lp_flea.htm to see if it might help you.
Joseph Murphy
July 16, 2008 at 12:54 pm