In the 20 years between 1995 and 2015, Spain’s industry lost 15% of its jobs while the rest of the economy’s sectors gained 37% more employment. But this phenomenon was not limited to Spain: deindustrialisation is having a widespread effect on advanced economies in general and in some, such as the US, it has been taking place for more than half a century.

Deindustrialisation is characterised by a combination of three factors: a reduction in industry employment, both in absolute and relative terms (in Spain the share of industrial employment went from 19.0% in 1995 to 12.7% in 2015); to a lesser extent a decrease in the sector’s contribution of value-added to the total gross domestic product (from 19.5% to 16.4% for the same period in Spain) and, in contrast, an increase in the value-added contributed by the sector in real terms (+33% in the same 20 years and +56% per worker).

There are many different reasons for deindustrialisation but the fundamental cause is the fact that industry benefits more from technological progress than the other sectors as a whole. The automation or robotisation of processes boosts productivity so that much more can be made with less, which ends up displacing part of industry’s employment to other sectors. In addition to this technological effect are other factors that also improve productivity, such as outsourcing certain services (for example, accounting) and offshoring part of the production to other countries, possible thanks to globalisation and the development of the emerging economies. With regard to demand, it has also been shown that, as economies develop, demand for services tends to grow more quickly than the demand for goods.

Although the origins of deindustrialisation lie in unmistakably positive factors such as improvements in productivity, it is also true that this phenomenon is often perceived with mistrust and unease. And for good reason. On the one hand because industry’s loss of relative weight is also due, in part, to a number of obstacles that could be minimised with the right economic policies. And, on the other, because deindustrialisation does not benefit everyone (a worker who has lost their job is not consoled by the fact that the economy as a whole is growing).

Some propose deceptively simple recipes to minimise job losses and promote reindustrialisation, such as barriers to imports and direct subsidies for the sector. This is Donald Trump’s approach, for example, but economic history is full of resounding failures by such policies, which have ended up encouraging corruption and inefficiency rather than sustainable development. Such an approach could also lead to an increase in protectionist measures at a global level and a decline in the world economy, as happed in the Great Depression.

Classic industrial policies, through which governments attempt to directly promote the development of certain industries, do not enjoy a good track record either. When it comes down to it, predicting which industries are more likely to expand successfully is no easy task.

A strategy that is not so simple but more effective involves taking measures in several areas: compensating those losing out to technological innovation and globalisation, for instance with training programmes and help in finding another job; having an educational system that is in line with industry’s new requirements, such as modern professional training; promoting infrastructures that make complete sense from a cost-benefit point of view, such as the Mediterranean corridor, and redesigning those regulations that act as a brake on companies growing in size, which tend to hit the industrial sector particularly hard. It is no longer a question of positive discrimination for a sector; the aim should be, at the very least, not to penalise it.

Enric Fernández

Chief Economist

31 October 2016