Too much economic development policy is based on a naive analogy to portfolio theory

Cities looking to strengthen their economies should concentrate on building upon and  extending current specializations

One of the most widely agreed upon bits of folk wisdom in economic development is the idea of “economic diversification.” The notion is that your local economy is like a stock portfolio, and if you are invested in just a few industries, then you are more vulnerable to dislocation if those industries decline. Like a prudent investor, if you were more diversified, you’d be less likely to suffer a decline if any particular industry experienced a downturn.

This bit of folk wisdom got a boost a few years back from economist Ricardo Hausmann the author of the Atlas of Economic Complexity, which assembled a clever statistical portrait of how simple or sophisticated a nation’s economy is based on its mix of industries. Hausmann argues that diversification is the key to higher incomes for cities and nations:

Larger cities are more diversified than smaller cities. Among cities with similar populations – say, Salvador and Curitiba in Brazil, or Guadalajara and Monterrey in Mexico – more diversified cities are richer than less diversified cities. They tend to grow faster and become even more diversified, not only because they have a larger internal market, but also because they are more diversified in terms of what they can sell to other cities and countries. What is true at the level of cities is even more applicable at the level of states and countries. The Netherlands, Chile, and Cameroon have a similar population size, but the Netherlands is twice as rich as Chile, which is 10 times richer than Cameroon. Looking at their exports shows that the Netherlands is three times more diversified than Chile, which is three times more diversified than Cameroon.

On its face that seems like a plausible claim, but upon closer inspection, there are five good reasons to question its accuracy and utility as policy advice.

First, Hausman and others note that highly diversified cities (and countries) tend to be richer.  But this is typically because they are bigger and better educated, not simply because they are more diverse.  A New York can achieve scale in more different industries than Des Moines, so its more diverse.  Also, as we know, educational attainment is critical–something these analyses often leave out.  And the cross-country evidence here is misleading:  Chile and Cameroon are poorer than the Netherlands, not because they are less diversified, but because they have much lower levels of education, and are still primarily dependent on resource-based industries.

Second, economies tend to be more diverse if they are richer.  Greater income and wealth create a greater demand for services, which naturally leads to more diversity  The third world village suffices with rudimentary generalist care; large wealthy nations have a diverse assortment of medical professionals.  But the cause of the diversification is the wealth; and not so much the other way around.

Third, its doubtful that this is of much practical application to local economic development policy.  The admonition to diversify your economy is based on a naive analogy to portfolio theory.  If one could change one’s economic base as easily (and cheaply) as one buys and sell stocks, it might make sense.  But its not clear how a city could easily change its mix of industries. Taken literally, the argument to diversify says that it would be a good thing if your biggest industries got smaller (that would make you more diverse).  But would Seattle really be better off if Amazon, Microsoft or Boeing was half the size it is today?

Fourth, the key lesson of clusters is that firms draw competitive business advantage from having other similar and related firms nearby.  By attracting talent, developing specialized suppliers, and promoting intense competition and benefiting from specialized knowledge spilling over, you get stronger, better firms, and a healthier economy.  Specializations are seldom static: one specialization often provides the knowledge base for new specializations: The process of economic development is often about related diversification:  being good in one technology at one time sets the stage to be good at generating the next technology at the next time.  The important thing is this isn’t random:  its path-dependent.

Fifth, the really pernicious thing about simple-minded diversification thinking is that it leads to the fad-of-the-month club style of economic development.  Yes!  We can be the next big biotech hub and that will diversify our economy!  And it leads people to neglect or ignore, or simply take for granted their existing strengths–which are likely to be much more plausible sources of future economic growth. Are you more likely to succeed in a field you know something about, or in a field in which you know nothing?

While there may be advantages to having a more diverse economy and avoiding excessive reliance on one or a few sectors, its far from clear that a city has much leverage to change its portfolio of industries. And pursuing diversity for diversity’s sake may sacrifice opportunities to build on existing strengths.