Job automation risks vary widely across different regions within countries

World Bank/Chhor Sokunthea
A woman irons fabric at a garments factory at the Sihanoukville Special Economic Zone, Phnom Penh, Cambodia.

This article is brought to you in association with OECD.

The risk of job automation is much higher in some regions than others within countries, meaning governments will need to address any widening of job inequality between one area and another in the coming years, according to a new OECD report.

Job Creation and Local Economic Development 2018: Preparing for the Future of Work finds that the geographic variation in job automation risk is strikingly high in the 21 countries for which data is available. The share of jobs at high risk nears 40% in some regions (e.g. West Slovakia) and is as low as 4% in others (e.g. the region around the Norwegian capital Oslo).

Previous OECD analysis has estimated that around 14% of jobs across the OECD area as a whole are at risk of automation, while another 32% are likely to see significant changes.

Within countries, the share of jobs at high automation risk varies the most in Spain, with 12 percentage points of difference between the highest and lowest risk regions. It is also high in the Slovak Republic, the Czech Republic and France. The variation, reflecting in part the fact that the sectors and jobs most susceptible to automation are not spread evenly across countries, is the least in Canada, with just 1 percentage point between the highest and lowest risk regions. Austria and Italy also show much smaller disparities than the average.

“Technological innovation such as automation can drive productivity growth, generate new jobs and contribute to better living standards. But we must guard against any increase in regional divides in job quality and employment,” said OECD Secretary-General Angel Gurría. “Our focus should be on improving skills and firm efficiency across all regions.”

Automation aside, Job Creation and Local Economic Development 2018 finds striking differences in access to quality employment across different regions in OECD countries. Regional disparities have increased in terms of the number and quality of new jobs created, unemployment and educational attainment. More than half of all regions saw their working age population decline from 2010 to 2016. Cities and towns continue to attract young educated workers at the expense of rural areas.

The share of people in temporary and part-time work also varies substantially between regions in the same country. In countries like France, Belgium, Hungary, Italy, Spain or Greece the gap between regions exceeds 10 percentage points. In the French region of Auvergne, for example, the share of non-standard work was 33.6% of total employment in 2016 while in the Ile-de-France region that contains Paris the share was just 21.7%.

Encouragingly, the report finds that 60% of regions in the 21 countries studied have created more jobs at a low risk of automation since 2011 than the number of jobs they have lost in high automation-risk sectors. Regions with a lower share of jobs at risk of automation tend to be highly urbanised with highly educated workers and a strong tradeable services sector.

The report calls for greater efforts to enhance skills in the workforce, particularly in rural areas, and to improve efficiency in firms at regional and city level so that they can reduce the share of workers doing the kind of routine tasks that run a higher risk of automation.

Both central governments and local authorities will need to juggle the need to enable automation to boost productivity with the need to manage the job losses that it could entail, particularly in regions that already have low productivity growth and high unemployment.