Yesterday I was speaking at the 13th Annual OECD LEED Forum in Prague on ‘Resilient and Diversified Local Economies’, or in more old fashioned academic terms: structural economic change and productivity.

It was good to be in the country of Vaclav Havel and Joseph Schumpeter (yes, he was born in Triesch – Moravia, current day Czech Republic). Both great thinkers and proponents of open societies. Today I largely stood on the shoulders of Joseph Schumpeter, the inventor of the concept of “creative destruction”: productive innovations making old firms and industries redundant.

One of the questions today was whether regions grow faster if they are more specialized or more diversified. In academic research you can find almost as much evidence pro as contra both options. So, well, it depends…, is the unsatisfactory answer. In general for the economy to evolve and prosper you need experiments and successful experiments to be scaled up and diffused. A diversified economy is generally better at enabling experiments, but for scaling up you need to develop a critical mass, sometimes leading to a specialized economy. So a specialized economy may be more the outcome of successful development than the cause. The latter phenomenon is good for growth, but along the lifecycle of an industry, it’s not good to be stuck with a dying industry in your (specialized) region. Check Detroit! So best is a diversified region to experiment, but with possibilities to specialize in emerging and growing industries, and step out of low productivity industries. Here’s where the related variety discussion enters: if you want to step into new industries, they better be related to your existing competence bases. Regions tend to diversify in new activities related to their existing activities from which they draw and combine local competences, unrelated diversification is the exception. Computer hardware production is unlikely to emerge in a region that specializes in growing bananas, while software production is likely to emerge in regions with computer hardware production capabilities, and in a more radical way, Fintech is more likely to emerge in regions with capabilities in the IT industry and banking than regions without these capabilities.

Another question was how can an entrepreneurial ecosystem approach supports new growth sectors. And, what is an example of a region where an ecosystem has been developed to support industrial diversification? An entrepreneurial ecosystem approach takes into account both agency (entrepreneurs, firms) and structures (system) to stimulate economic development. A key element of economic development is structural change: the creation of new industries and the destruction of old industries. For this process of creative destruction to take place, policy needs to stimulate the entrepreneurs and businesses that have the potential to create new sectors. Targeted industrial policy has many problems. An effective option may be to create the best context for productive entrepreneurship: a flourishing entrepreneurial ecosystem. Examples around the globe are places like Boulder-Colorado, Cambridgeshire-UK, and in my home country: Amsterdam and Eindhoven. These places have not had explicit ecosystem strategies to start with, these largely emerged with a mixture of public and private involvement. They were certainly not a top down exercise, although places like Singapore and the United Arab Emirates might provide exceptions here.