Our modern, technologically-advanced, societies require very specialized knowledge and expertise to function.   In such societies, it benefits individuals to specialize. Despite the beliefs of management consultants and the old Bell System, it is not true that everyone can do anything.

In the 1950s and 1960s, for instance, the British Government successfully promoted the development of a highly-specialized cadre of nuclear energy physicists and engineers, able to design, build and operate nuclear power stations.  Once that  technology became mature, however, Government policy shifted and it was thought that country could purchase nuclear power technology “off the shelf”; the country had no need for the skills involved (it was argued) and thus de-skilled.  The French government took a different view, with the consequence today that young French nuclear engineers are in high demand in Britain.

Just as it benefits individuals to specialize, so too with cities and regions.  If there are many companies in the same industry near to one another, recruitment of specialized, skilled staff is easier, exchanges of ideas and business occurs more often, and collaborative partnerships and common campaigns are facilitated.   This is why, for example, the world’s leading commercial insurance companies operate near to one another in Trinity Square, London, and have done for centuries.  This is why Stamford, CT, is a similar centre for insurance companies.  This is why, despite the so-called abolition of distance by the Internet, the key US companies in telemedicine all operate within a few blocks of one another in Manhattan.   Michael Porter’s work on regional industrial clusters has been rightly compelling in explaining the causes and consequences of these phenomena.

But what of countries?  Most national borders are historical or geographic artefacts, contingent accidents of history that could well be otherwise.   So, prima facie, what is true of regions should also be true of countries:  it should benefit countries to specialize.  Since David Ricardo’s theory of comparative advantage in 1817, economists have believed that countries gain from specialization in the production of goods for which they have relative advantage (despite the theory’s flaws).    Why then do many people think it necessary for countries to NOT specialize, to have strong services sectors AND strong manufacturing industries AND a strong agricultural base?   Many Marxists seem to think this – that all countries should have large manufacturing sectors – and none I have questioned has ever been able to give me a good justification as to why.   (Perhaps believing that a proletarian revolution is a necessary stage of every country’s history leads one to believe that an industrial working class is also necessary, and hence a large manufacturing sector.)
Since the Great Global Recession of 2007-?, conventional public policy wisdom in Britain has been that the country’s economy needs “rebalancing” to reduce the role and proportion of financial and professional services, and increase the role of manufacturing.   But why?  Surely, most jobs in services are better paid, have better working conditions, and are generally more intellectually and emotionally challenging,  than the repetitive, dirty, noisy, foul-smelling, physically-demanding jobs of factories.  Of course, modern factories are often clean, quiet, and air-conditioned, because robots, unlike people and trades unions, refuse to work in any other conditions.

Now, according to The Economist,  the British Government is planning to throw money at industrial sector strategy again – “picking winners” is the term of art.  Not only did this fail last time Britain did it (in the 1960s and 1970s), but even MITI – the once all-powerful Japanese Ministry of International Trade and Industry – failed at it.   Japanese attempts to enter the avionics industry were a bust, for example, despite MITI’s great desire, focus, power, and resources.

I can see a valuable role for government in overcoming problems of collective action – for example, when the actors lack knowledge of each other’s capabilities, beliefs or intentions, or when there are network effects or externalities associated to actions, or when it is in everyone’s interest to do something, but in no one’s interest to be the first to do that something. In these cases, government can can bring relevant actors or stakeholders together; it can convene; it can co-ordinate; it can develop common visons for the future; it can suggest, request, cajole, morally suade, and even harry participants to act for the collective good against their own self-interest.  But none of these government actions or policies requires the government to choose winning companies or perhaps even winning sectors or regions. And none requires vast sums of money.