Posted by on

In my last post, I presented a ranking of entrepreneurship in Canada’s largest cities based on the share of the labour force who are self-employed, and then discussed some reasons why that ranking doesn’t capture what policymakers often mean when they talk about entrepreneurship. Basically, the idea was that being self-employed doesn’t tell us anything about a person’s intentions for their business: do they want to innovate and build a large business, stay small (or never employ anyone at all) and live a more relaxed lifestyle, or just pay the bills until they can find a job?

In terms of economic growth we want to focus on the first type of self-employed person, those who want to take risks, grow big, and employ a lot of other people in the process. And since we don’t have any data on the intentions of the self-employed, to capture this group of entrepreneurs we could focus on results and measure whether a city’s entrepreneurs have been successful at building large businesses.

Two Swedish researchers named Magnus Henrekson and Tino Sanandaji did just that, ranking countries on entrepreneurship based on their number of billionaire entrepreneurs. They generously shared their data on Canadian billionaires with me, and I’ve created a similar ranking for Canadian cities. Unfortunately there’s a problem: it’s no more reliable than the self-employment ranking. In fact, it may even be worse. I’ll still show a ranking based on the billionaire data at the end of this post, because that’s what I committed to doing, but please read the caveats and take it with a grain of salt.

First, I’ll describe the data sent to me by Henrekson and Sanandaji and how I adapted it to Canadian cities:

  • The source of the data is Forbes Magazine’s The World’s Billionaires lists from 1996-2014. Henrekson and Sanandaji went through each individual or family in the lists excluded those who inherited wealth or did not start a business (for example early employees of successful businesses). An individual is included in the dataset if they appear in the Forbes list in any year from 1996-2014, even if they later drop off the list due to death or loss of wealth. Finally, billionaires are allocated to countries based on citizenship.
  • I then allocated each billionaire to a city based on where their business was founded. I decided against allocating by city of birth or residence as in the original dataset, because it’s probably more likely for a Canadian entrepreneur to have moved within Canada before or after building their business, as opposed to moving between countries. As a result, three billionaire entrepreneurs with Canadian citizenship but who founded their business outside of Canada were excluded from the final dataset.
  • After allocating billionaires based on founding city, the resulting dataset has 33 billionaires representing 31 unique businesses.

Here’s what the resulting ranking of entrepreneurship, using billionaire entrepreneurs per 100,000 population, looks like:


Just like in the self-employment measure, Edmonton sits at around the middle of the ranking as the 9th most entrepreneurial major city in Canada. One change from the self-employment measure that makes intuitive sense is that Victoria (number one in the self-employment ranking) has dropped into the bottom group of cities, while technology hub Kitchener-Cambridge-Waterloo (near last in terms of self-employment) is now the most entrepreneurial city in Canada.

But also like the self-employment measure, using billionaire entrepreneurs to measure entrepreneurship for cities comes with a whole lot of problems and caveats. Here are a few:

  • The biggest problem is that the sample size is just too small. There are only 33 billionaires in our dataset, and more than half of Canada’s largest cities have no billionaires allocated to them. With such a small number of billionaires to work with it becomes difficult to say whether a given city has a high billionaires per 100,000 population because it’s especially entrepreneurial, or just due to random chance.
  • The billionaire cut-off is too high. When we’re measuring entrepreneurship on the scale of cities, only counting individuals who built businesses that made them billionaires is too high of a bar. What about an entrepreneur whose business earned them hundreds of millions of dollars? They might not be significant when comparing countries, but would certainly be among the most successful entrepreneurs in Canada and should be included in a cross-city comparison. Unfortunately we don’t have any data that tracks entrepreneurs at this level of success.
  • The measure doesn’t account for multiple founders of a single business. Kitchener-Cambridge-Waterloo has three billionaires in the dataset, all whom are associated with a single business, Blackberry Limited. The billionaire entrepreneur measure would treat it the same as a hypothetical city with the same population, but three billionaires who all founded separate businesses. Is this the correct way to measure entrepreneurship? I would argue that the hypothetical city is more entrepreneurial.

So does the billionaire measure of entrepreneurship improve on the self-employment measure? They both have strengths and weaknesses, but overall I think that the self-employment measure is a more reliable indicator of entrepreneurship on the scale of cities. But really, the upshot here is that there is no easy answer when it comes to measuring entrepreneurship. Like with many questions about the economies of Canadian cities, the data we have available just isn’t sufficient to capture what local policymakers really want to target.

Governments have made it clear that fostering entrepreneurship is an important policy goal, and have put resources behind all sorts of initiatives with that goal in mind. That’s great, but let’s not forget that to measure our progress towards that goal, we also need to devote resources to finding new and more accurate ways of measuring entrepreneurship.

Thanks to Carol Wong for her help researching this post.


Join the conversation