By Guillermo Montt.
In a previous post we wrote about the potential of machines to transform and replace jobs. The potential for automation is limited when it comes to social skills, which is why social skills are increasingly rewarded in the labour market. Technological change is shaping the future of work through, in part, a skill-biased effect on employment.
But what about other drivers of the future of work?
Demographic change, changing economic centres, a shift in management practices and changes in product demand among others, will also shape the types of tasks carried out in each job and, consequently, the skills needed and available in the workplace.
Globalisation – the increasing integration of production across borders – is perhaps one of the strongest drivers behind our economies’ trajectories. What is the relationship between globalisation and skills and work?
In a recent edition of Economic Policy Sascha Becker and Marc-Andreas Muendler follow German occupations and industry sectors for three decades to understand the interplay between engagement in international trade and the impact on tasks performed on the job. Their study is unique in that – in contrast to most of the literature in the subject – they are able to identify trends within industries and occupations and that they are able to follow these trends for three decades, from 1979 to 2006.
Globalisation has had two phases. Globalisation 1.0 refers to trade of final products. Globalisation 2.0, which we are experiencing today, refers to the trade of intermediate products. In this second phase, production lines cross borders and, as a result, production stages – and jobs – are offshored. The types of jobs that are offshorable are not necessarily low-, medium- or high-skilled jobs – they are typically jobs that do not require personal contact or geographic proximity, they are also jobs that can be codified into specific steps. Jobs that rely more on tacit knowledge are less likely to be offshored.
There are several interesting findings in Becker and Muendler’s analysis. First, they note that Globalisation 2.0 was already well-established in Germany in 1979, with many production processes already integrated across borders, but has increased steadily since then. Second, in following the activities and performance requirements for workers since the 1970s, they find that the number of tasks associated with each job has increased: there has been an increase in multi-tasking. If, for example, in 1979 the average worker performed 1.67 activities in his or her job (e.g. produce goods, repair, prepare food, transport, measure, sell, programme a computer, apply legal knowledge, train, nurse, advertise, oversee, control machinery or processes, etc.), by 2006 the average worker performs more than seven of the 15 activities they measure. Becker and Muendler argue that this process of multi-tasking is a response to offshoring. German jobs have specialised in non-offshorable tasks.
Importantly, this process towards multi-tasking is seen mostly within occupations and sectors: it is not driven by changes in the composition of occupations or industries but rather by changes in what is expected for each job. Since 1979, all tasks grew in importance relative to the most offshorable (manufacture, produce goods) and this process was already evident in 1986 relative to 1979. Consistently, imports for intermediate use grew most in sectors that have a greater concentration of what are commonly considered offshorable jobs. Globalisation in Germany has changed what tasks are required of each job and what skills are needed to carry these tasks out. Note that Becker and Muendler control for other drivers of change by adding sector and year fixed effects.
Labour market institutions – at least those that are sector-specific and vary within Germany – explain little of the variation in changing tasks (though, predictably, the authors observe faster offshoring in sectors with less tight labour markets and lower unionisation rates).
What does this mean for skills? For countries that offshore some of their work, it means that the jobs specialise in non-offshorable tasks that tend to require interpersonal and multitasking skills and that rely on more tacit knowledge. But this conclusion only applies to countries (possibly industries or specific firms) on the “sending end” of offshoring, those that are already high up on the global value chain or those that are climbing their way up. Codified tasks are still performed somewhere in the world. The point for skills policies in a specific economy is to understand where firms and industries stand in the global value chain and what their specific needs are.
The weak findings in terms of labour market institutions do not necessarily mean that labour market institutions are irrelevant in explaining the penetration of globalisation and its impact on tasks. It could very well be that the variability in labour market institutions is relatively low across sectors of the German economy, and that these effects may emerge more clearly if the country sample is broadened and/or lengthened to include more recent data taking into account the implementation of the statutory minimum wage in Germany or the rise in the use of “mini-jobs”.