Will automation destroy jobs?

MT at 50: Carl Benedikt Frey, co-director of the Oxford Martin Programme on Technology and Employment, gives his expert view on the effects of automation on jobs - and the news isn't all bad.

by Ian Wylie
Last Updated: 15 Jun 2016

Carl Benedikt Frey of the Oxford Martin School at the University of Oxford has become (with colleague Michael Osborne) the leading authority on the automation of jobs. Their methodology focuses on how reliant tasks are on human aptitudes such as social perceptiveness, persuasion, originality and manual dexterity. We asked him about the impact automation will have in the future.

What will determine the pace of change?

Adoption – the technology needed to automate 47% of jobs is already there. The pace of adoption depends on the relative cost of technology and labour. In China, for example, we're seeing rising wages, which means the relative cost of technology is becoming cheaper – so the pace may become even faster.

Regulation is another determinant – no matter how good the Google Car is, unless insurance frameworks are adopted to allow for driverless cars, the pace of adoption will be slow.

I think it unlikely that robots will enter the domain of complex social interactions or jobs that require creativity or perception, and manipulation of irregular objects.

What about the UK?

The UK has a strong base in financial services which are very automatable, so in terms of remaining competitive, the adoption of technology in UK financial services will be essential to driving productivity. However, some countries are being particularly pro-active. China's new five-year plan explicitly embraces automation as a strategy for remaining competitive in manufacturing.

Will jobs lost through automation be replaced?

It seems from our research that digital companies such as Facebook and Google do not create as many jobs as the technology companies of the past – the GMs, Fords and Dells. But the fact that there are not as many people working directly in technology does not necessarily mean we will see a decline in the general number of jobs. 

Our research suggests that once a technology job is created, it leads to the creation of five new jobs in the local economy. And in some new research that we hope to publish soon, we’ve seen that in developing economies, the multiplier effect from new technologies is even higher.




Find this article useful?

Get more great articles like this in your inbox every lunchtime

Upcoming Events