The impact of technological change on skills

automation skills eu

Terence Hogarth is Senior Advisor at Fondazione Giacomo Brodolini and has over 30 years’ experience of carrying out research on vocational education and training in the UK and the EU. 

Terence is this month’s guest blogger:

Reflecting upon the employment impacts of rapid technological change in the guise of artificial intelligence, robotics, Industry 4.0 and so forth, various authors in the recently published collection of essays “Economy employment and skills: European, regional and global perspectives in an age of uncertainty” draw attention to the issue of skills.

The threat to jobs

Without doubt Industry 4.0 and the like pose a threat to jobs, especially those whose features can be automated.  It is also apparent that the range of jobs susceptible to substitution by robots has increased and continues to do so at pace.

This has led some commentators to the conclusion that great swathes of jobs will be lost in the future, but more measured analysis in the chapter (link below) by Suta and her colleagues at Cambridge Econometrics suggests that the impact on jobs, whilst substantial, may be more modest than predicted in some of the more commonly cited pessimistic forecasts.

While this is indubitably good news, it is maybe less so for those in the labour market with less highly developed skills because it will be this group who are likely to bear the brunt of job losses resulting from technological change.

Although technological change may make some jobs and skills obsolete, it is likely to create new jobs too.  This has been self-evident throughout history.

From a policy perspective preparing people to take these new jobs poses a number of challenges related to:

These are not mutually exclusive.  If people are to be equipped with the skills that will match them to the jobs of tomorrow then, in the first instance, there is a need to know what these skills will be.

Identifying emerging skills needs is far from straightforward

Conventional methodologies for identifying emerging skill needs – such as those employed in the European Skills and Jobs Survey and in Cedefop’s projections of future skills demand – provide robust, detailed evidence.  But the analysis that leads to the production of robust results takes time.

With advances in computing power it has become possible to complement these traditional methodologies with new approaches that search for data linking technologies, jobs and skills from myriad data sources (for instance, recruitment websites, patent applications, academic papers, etc.) and then categorises them.

The chapters by Fantoni and his colleagues from the University of Pisa (link below) in the aforementioned collection of essays give examples of these new approaches.  Typically they provide an almost real-time insight into emerging skill needs, though they are less able to reveal scale or the extent to which those skill needs are being met.

The point here is that no one approach will be able to provide all the information required, rather there is a need for plurality of approaches that will allow a detailed, realistic, and nuanced view of future skill needs.

Identifying the characteristics of emerging or future employment and skill demand is only the first step.  This information then needs to be used to guide the skills acquisition not just those making their initial entry into the labour market but also those whose jobs might be at risk from robots.

Meeting the needs of this latter group can be notoriously difficult, especially so if the aim is to provide them with the skills that will sustain them in employment before their current job comes under threat.

The case for continuing professional development

If people are increasingly expected to spend longer in the labour market – given that some countries have raised their retirement rates – then it stands to reason that the skills people acquire in their initial education will be unlikely to carry them through to retirement.

It is well known that highly skilled and qualified people are more likely to participate in continuing training.  But this group’s employment is perhaps less susceptible to technological change.  The pressing policy issue, from an occupational and skills perspective, is how individuals can reinvent themselves.  And, given that technological change can sometimes affect an entire industry with the impact typically geographically concentrated, there is often the need for a local as well as a national response.  So one is often looking at how a local area or region can reinvent itself through a combination of industrial and skills policy.

Diagnosing and suitably responding to the changing patterns of skill demand resulting from technological is important.  A failure to do so, for the individual, can lead to social exclusion.  And in aggregate it can have damaging impacts at the national, regional, and local levels.  For instance, the extent to which people are employed in hi-tech sectors and access to training varies by Member State (see Figure 1).  So there is the ever present danger that technological change, in itself, has the capacity to exacerbate both inter- and intra-country differences with respect to who is able to access the most secure jobs.

The tools we have can help us plan effectively for the future

By being able to effectively anticipate skills change, using the variety of tools now available, and ensuring that the provision of skills supply is responsive to emerging labour market demands, there is every hope that the benefits to be derived from Industry 4.0, robots, and AI will be maximised.

But this is dependent upon skills anticipation using the full range of tools now available – skills forecasts, surveys of individuals and employers, big data analysis – to diagnose new and emerging skill needs and ensuring skill systems at the national, regional, and local level are responsive to the messages those tools convey.

Future Employment and Automation, Suta et al.

Defining industry 4.0 professional archetypes: a data-driven approach , Fantoni et al.

No Comments

Add your own comment

Your email address will not be published. Required fields are shown with a *

Allowed HTML tags are <a href="">, <p>, <br>, <i>, <em>, <strong>, <b> and <strike>.