ICT specialist occupations and other ICT task-intensive occupations made a positive contribution to employment growth in almost all countries between 2011 and 2017, including in countries where employment fell overall. In Luxembourg, where employment increased by 21% over this period, ICT specialists accounted for one-in-ten new jobs, and a further three jobs were in other ICT task-intensive occupations. In the United States, employment grew by around 10%; one third of these additional jobs were in ICT task-intensive occupations.
ICT specialists are most likely to work in Information industries, whereas ICT task-intensive occupations are pervasive in a variety of sectors. In the countries presented, approximately a quarter to a half of employees in the information industries are ICT specialists. Other ICT task-intensive occupations make up a relatively small share of ICT industries’ employment in most countries. These nevertheless represent a majority of ICT-related employment in other industries, which employ around four people in other ICT task-intensive occupations for every one ICT specialist, on average.
The ways in which digital technologies are changing jobs, and the implications arising therefrom, is a key concern for workers, employers and governments. Identifying the tasks that are most likely to be substituted by technology – those involving basic exchange of information, buying and selling, and simple manual dexterity – and the workers performing them helps to shed light on what the future of work may look like. The OECD Survey of Adult Skills (PIAAC) dataset provides a detailed breakdown of the tasks workers perform on the job. Each worker can thus be assigned a probability of being impacted by digital technologies, and by automation, in particular. According to Nedelkoska and Quintini (2018), 14% of jobs across all countries in the sample have a high (over 70%) likelihood of being automated, while another 32% have a 50% to 70% probability of facing significant change. Workers in these jobs perform several automatable tasks, alongside tasks that are not currently automatable. Meanwhile, the estimates also suggest that about a quarter of jobs have a less than 30% chance of automation. Overall, these estimates indicate that automation could affect a wide range of jobs, though the nature and extent of these impacts will vary greatly across occupations, industries and countries.
The estimates also highlight significant differences across countries, with automation highly likely to affect between 6% and 33% of all jobs. Similarly, the share of jobs estimated to have a significant likelihood of change varies between 23% and 43% of all jobs. These, however, are not necessarily the jobs displaying the lowest skill requirements. Marcolin et al. (2018) show that the relationship between skill and routine intensity is negative but not very strong, and insignificant for jobs which display medium routine intensity.
