Putting faces to the jobs at risk of automation

By Stefano Scarpetta (OECD),

Read the full article here.

Screen Shot 2018-05-02 at 10.53.42.png

The tasks that AI and robots cannot do are shrinking rapidly

New technologies such as robotics are artificial intelligence are rapidly changing peoples jobs and lives. They have the potential to free up workers to do more productive, less routine tasks and to provide consumers with access to more and better products and services. However, technology will likely change many of the existing jobs, requiring workers and companies to adjust. Some jobs may become entirely redundant although the extent of automation will likely depend on policy, institutions and social preferences.

There has been a considerable public debate about the extent of job destruction and whether automation and digitalization are leading to mass technological unemployment, in which many jobs will be done by computers and robots. This is fuelled by the perception that technological change is faster paced and broader based than in the past, making many more jobs automatable than previously thought. While only 15 years ago computers performed poorly in non-routine cognitive and manual tasks (Autor et al, 2003), cutting-edge technologies now open the possibility for tasks as diverse as medical diagnosis, insurance brokerage and driving to be automated.

While only one in seven jobs may be lost to automation, many others will change significantly

Screen Shot 2018-05-02 at 11.02.00.png

Given the current state of knowledge about the tasks that cannot be automated (the so-called engineering bottlenecks, see the box on the right), new research by OECD suggests that 14% of all jobs across the 32 countries analysed have a high risk of automation (Nedelkoska and Quintini, 2018). A further 32% of jobs may experience significant changes to how they are carried out (chart 1).

Chart 1. Large shares of jobs are at risk of automation or significant change.

Chart 1. Large shares of jobs are at risk of automation or significant change.

Whether workers lose their jobs as a result of automation of have to adapt to new tasks and job content, the challenge for adult learning systems is considerable. At present, about 40% of workers participate in job-related training, on average across OECD countries, but participation often amounts to just a few hours per year. In addition, there are substantial differences across countries and socio-demographic groups. For instance, data from the OECD Survey of Adult Skills shows that only 16%of workers in Turkey and Greece participated in job-related training in the year precedenting the survey, compared to almost 60% in Denmark and New Zealand (Chart 2). At the same time, in all of the countries surveyed, the low-skilled received much less training than their more skilled counterparts: only 17% participated in training on average in a 12-month window. 

The training challenge is amplified by the fact that the risk of automation not distributed equally among workers, making existing differences in access to training even more problematic.

Low-skilled people and youth are among those most at risk

Despite recurrent arguments that automation may start to adversely affect selected highly-skilled occupations, this prediction is not supported by the Frey and Osborne (2017) framework of engineering bottlenecks. If anything, AI puts more low-skilled jobs at risk than previous waves of technological progress, whereby technology replaced primarily middle-skilled jobs creating labour market polarisation.

With the exception of some low-skilled jobs that, for instance, involve caring for and assisting others, the risk of automation declines as educational attainment and skill level rise. Occupations with the highest risk of automation are predominantly low-skilled, such as food preparation assistance, cleaners and helpers, and labourers in mining, construction, manufacturing, and transport. Occupations with the lowest risk of automation all require professional training and/or tertiary education and include teaching, management, and health professionals.

Looking at the risk of automation by age group, young people are most at risk, followed by older workers. Youth often combine studying with work in elementary occupations such as unskilled laboureres, ceaners and in food prparation. In addition, even hihgly-educated youth tend to enter the labour market in junior and routine roles which are more prone to automation before progressing to posittions that make better use of their cognitive and social skills. On the policy side, these developments suggest that new ways must be found to help youth obtain work eperience while studying. In addition, the potential disappearance of entry positions would have implications for career progression in some profressionak areas such as law or accounting.

Chart 2. Low-skilled adults receive little training in most countries

Chart 2. Low-skilled adults receive little training in most countries

Automation may also be putting downward pressure on wages and working hours

Automation is not a far-flung reality but is already reflected in employment outcomes, such as unemployment, hours worked and earnings. Today, workers in occupations with a high risk of automation display much higher unemployment rates than those with low risk. Further, workers in the most automatable jobs work approximately eight hours less per week than workers in the least automatable professions.

Turning to wages, workers in jobs with a high risk of automation have lower hourly earnings in most countries, On average, a 10% higher risk of automation corresponds to a 4.3% decrease in hourly earnings. This suggests that those at risk of automation are often already in a vulnerable position in the labour market, which may be exacerbated as automation progresses. 

There are reasons why the future may not be jobless

While the prospect that every second job in the OECD may be impacted by automation is daunting, it is only a technical possibility. The actual implementation of full or partial job automation will depend on a range of other factors such as technology penetration and adoption, the cost of human labour relative to the new technologies and the social preferences for automating certain tasks. Take-up of new technologies lags behind their technical feasibility and some innovations are never widely implemented,. For example, even if care services for the elderly could be automated, people may continue to attach a value to these tasks being performed by humans (Artz et al, 2016).

Moreover, even if technology makes certain jobs redundant, it also creates new jobs and complements existing ones (Autor, 2015, OECD, 2017). Historical examples include the introduction of automatic telling machines (ATMs), which performed routine tasks handled by human bank tellers but freed up their time for more productive tasks such as non-routine marketing and individualized customer services, ultimately leading to an increase in employment numbers (Bessen, 2015). Similarly, jobs are likely to be created for the development, implementation, maintenance, and use of new technology.

Key policy priorities

The unequal distribution of the risk of automation raises the stakes involved in policies to prepare workers for changing job requirements. Education systems will need to adapt to the changes brought about by automation and teach children and youth the skills that allow them to take full advantage of the current wave of technology adoption, This includes skills such as cognitive and social intelligence but also extends to the skills needed to work effectively in a digital context, both as specialists and users of digital technologies.

For those already in the workforce and whose jobs are being affected by technology, adult learning is a crucial policy instrument. Unfortunately, the likelihood of participation in any type of training, on-the-job and outside the job, is found to be significantly lower among workers in jobs at risk of being automated. Workers at the highest risk of automation are more than three times less likely to have participated in on-the-job training, over a 12month period, than workers in non-automatable jobs. These differences also apply to training duration: individuals in jobs with the highest risk of automation spend 29 hours less in job-related training annually than those in non-automatable jobs, ceteris paribus. These findings point to the importance of training providded outside the workplace, particularly for workers whose jobs have the highest risk of being automated.

Effective and well-targeted adult learning opportunities for re-skilling and up-skilling workers are not all that is needed. For the smaller group whose jobs may disappear entirely, requalification must be accompanied by reinforced helo from labour market and social policies. Newly created jobs mare require very different skills from those that are being destroyed and may be located in a different region. These displaced workers would need adequate social protection, including income support, and tailored re-employment assistance.

Much of the variation in the risk of automation between industries and occupations reflects the way jobs are organised. Jobs with smaller proportion of routine tasks or a more-balanced mix of routine and non-routine tasks have a reduced risk of automation. Hence is it is crucial to promote good practice in work organsation within firms and across industries.

Source: http://www.oecd.org/els/emp/future-of-work...