 Welcome to the ECB podcast bringing you insights into the world of economics and central banking. My name is Michael Steen and today's topic is digitalisation and jobs in the Euro area. The digital economy is transforming how we work, where we work and when we work. Some even worry it might change whether we work. In recent months the coronavirus pandemic has sped up this transformation for many employees as they work from home. But the digital economy goes further than that. It affects output, investment, productivity, prices and labour markets. So we decided to spend the next two episodes looking at this topic. How does the increased use of digital technologies change the way we work? Is the historical fear of automation taking away our jobs justified? Or do our tasks and skills simply evolve to meet the new conditions? Before you turn to our first guest, I'd just like to say we're back in the building, sort of. We've recorded our last few podcasts from home and in this one we're doing it in a socially distanced way in a meeting room at the top of the ECB. Our first guest today is Robert Anderton, an economist at the European Central Bank. Robert, thanks a lot for taking the time and welcome to the podcast. Thank you for having me. So you recently co-authored this paper on digitalisation and its impact on the euro area economy. And we wanted to discuss some of your findings with you and your co-authors. But the first question is let's maybe briefly define digitalisation. What do we actually mean by that? Yes, it covers also many decades, starting off with mainframe computers that eventually lead to personal computers. And then we move on to nowadays things like artificial intelligence. So it's a very broad term covering all these aspects of computerisation, automation, robotisation. The paper we're talking about today is called Virtually Everywhere Digitalisation and the Euro Area and EU Economies. And we'll obviously link to that in the show notes. Now there are 19 different economies in the euro area and there are 27 in the European Union. What did you mean by Virtually Everywhere? Is that in a sense we're not quite there yet? The answer is it depends. So our paper considers a number of measures of the degree of digitalisation because no measure is perfect and because there are many different aspects of digitalisation. The first is a composite indicator computed by the European Commission, which is the Digital Economy and Society Index. And what this index shows is that some dimensions of digitalisation are now at roughly similar levels across the EU, for example broadband connectivity, while others such as the use of the internet can vary significantly across European countries. And according to this measure, the most digital economies in the EU are Finland, Sweden, the Netherlands and Denmark. When we looked at the size of digital economy sub-sectors, including manufacturing of digital products, as well as IT services, telecommunications and other related activities, it is clear that the digital economy is larger in the US than in most of the euro area and EU countries, with the digital economy in the euro area about two thirds the size of the US. The third measure I'd like to mention is robots because robot adoption is another aspect of digitalisation. The EU and the euro area are on par with the US, but both are far behind Japan. However, some EU countries are also at the frontier in terms of robotisation. For example, Germany has as many robots as Japan. Digitalisation is indeed virtually everywhere and transforming our economies, but to differ in degrees across the euro area and EU countries and across continents. Okay, so surprisingly, Germany on the par with Japan on the number of robots, but the big picture there is lagging behind against the US, and then these quite significant differences between euro area and EU countries individually. How do we explain these differences? Well, a lot has to do with country characteristics. Some of those characteristics are at least to a large extent given, such as the size of an economy or its structure. Other characteristics may be heavily influenced by economic policies and so-called framework conditions, including institutions and governance. A country with a large manufacturing base may have had more opportunities to apply automation and robotisation to production processes and or be involved in the manufacture of digital products. Economic policy also plays a role. Where the framework conditions are favourable to the adoption of new technologies, those technologies are likely to diffuse faster and be more widespread across the economy. What we found is that the countries with high quality institutions and governance tended to have higher levels of digitalisation. Conversely, progress may be held back if the returns to investing cannot be fully realised because of weak institutions or governance. So one way this has a major effect on the economy is labour markets. I think your research shows that whether a job is lost or not depends on the tasks and skills it involves and of course the industry. Also the way a country is set up is policies, institutions and governance play a role. So the question is what's the situation in the euro area? How do the differences between different countries affect their labour markets? Yeah, the effects of digitalisation that occur through the labour market are indeed very important and for two main reasons. One is that the labour market implications of digitalisation may be important for each and every one of us. Another is that over history technological advances have entailed fears of job losses. The evidence suggests that digitalisation raises productivity and economic activity but may not always and not everywhere increase employment. Whether a job exists or not depends on the tasks and the skills it involves. The sector or industry in which it resides and also how the country is set up in terms of policies, institutions and governance. The effect on employment depends on whether technology is a complement or a substitute to labour and this is not always clear cut. Consider the role of robots for example. Where the exposure to robots is high some jobs become redundant. I mean those jobs that consist of routine tasks that can be taken over by robots. At the same time other jobs are created to perform tasks like putting in place the robots, monitoring and maintaining them. Secondly the results of a special ECB survey of large firms provide some insights into the impacts of digitalisation on European labour markets. Digitalisation was seen by those firms as replacing some low and medium skilled jobs but not high skilled jobs. Digitalisation was regarded as increasing the ratio of high skilled to low skilled workers with firms emphasising the need for retraining and the reassignment of workers to new tasks supported by digital technologies. I'd say it's important to realise that digital technology is going to have further impacts on labour markets and also that the fear of job losses without new job opportunities arising is unfounded. What role should labour market policies play here? There are a number of policy implications regarding income distribution, competition, innovation and education which may require efforts at both the EU and national level. As a starting point it's important to provide general access to the new technologies and the corresponding education and training along with help for business models supporting digital skills, tasks and jobs as this will be particularly important to digital inclusion. It may also be necessary to adapt regulations of labour, product and financial market regulations in order to fully reap the potential gains from digital technologies while keeping an eye on the more vulnerable groups in the labour market. Digitalisation may entail more market concentration among firms and it is desirable to maintain equal opportunities and incentives for innovation for all firms while supporting those in the labour market particularly affected by the transition to a digital economy. Digitalisation offers many opportunities but structural and framework conditions may need to be adapted for the Euro area and EU countries to fully reap the potential gains from digitalisation while maintaining inclusiveness in the labour market. Okay Robert, thank you very much. Thank you. Our next guest is Valerie Jarvis who also works at the ECB as an economist. She's done a lot of work on labour markets over the past few years. Valerie, welcome to the ECB podcast. Thank you. Now let's dig a little deeper into how digitalisation affects job creation and start with a question that automatically comes to mind when speaking about technological progress. The historical fear of automation leading to massive job losses. In your paper you mentioned the Luddite movement in Northern England in the early 1800s. The Luddites were English textile workers and they destroyed textile machinery out of fear that the machines would replace them. A rebellion that lasted some five years. I guess the question is do modern day Luddites have a point? Well I mean structural change is always part of the dynamic economic process and of course technological change often induces technological anxiety and insecurity but I doubt that the Luddites foresaw that the industrial revolution proceeded in retrospect an era of remarkable improvements in standards of living and incomes. Now in my view the digital revolution also has the potential to bring about large improvements in productivity and living standards. Of course there can be disruption to labour markets in cases where digitalisation occurs very rapidly, where workers lose their jobs and find it difficult to get back into employment even for prolonged periods. And this is of course of importance to policy makers and to citizens alike. But digitalisation also generates new jobs and tasks. So in our paper we find that European countries with higher shares of digitally intensive employment are typically those with higher aggregate employment growth. If we take a look at employment growth we've had several decades now of this technological progress automating lots of human tasks. The other obvious question is how can we actually generate enough jobs for people the fear of there being jobs disappearing? Well going back to my last point I said that European countries with the high shares of digitally related employment are typically those that are associated also with higher rates of employment growth. One of the main findings and consistent findings of the paper is really the considerable degree of cross-country variation heterogeneity across a range of digitalisation metrics in Europe. And we find the same degree of differentiation also when we look at the sort of shares of digitally dependent or digitally intensive employment. Now there's no standard definition of digitally dependent employment. It's difficult to measure because we don't want to just count the number of IT specialists in each country and we don't just want to look at the IT producing sector. We should also take account of those working in sort of digitally intensive occupations and tasks because actually for every one IT specialist in Europe there are roughly three to four additional workers working in digitally dependent work. So you're taking quite a broad view there of what a digital job is. Absolutely. And what we find is that there are really quite stark differences in the proportions of digitally dependent employment across Europe. So at one end of the spectrum, Luxembourg, with a share of digitally intensive employment of around, well, in excess of 20%, surpasses even the shares that we see in the United States. Now there are other countries up there, Sweden, the UK, Lithuania, Estonia, the Netherlands. The EU and Euro area averages are around 11%. But at the bottom end of the spectrum, these shares are only around 7%. And again, I want to reiterate that it's the countries at the top of the spectrum which have typically been those with the highest rates of employment growth and lower unemployment. Okay, so that's really interesting. So countries with more digital jobs tend to have faster employment growth. And that's a finding that you came out with from the paper. I mean, what else do you draw from that? Yeah, I think it's really important to understand the mechanisms of digital job creation here. And if we look at European countries which typically score well across a range of digitalisation metrics really, whether we're looking at the shares of IT specialists or digitally dependent employment, whether we're looking at the relative sizes of the digital economy across countries or even metrics of the proportions of firms making use of really state-of-the-art high-level cloud computing technologies. We see that three European countries, Sweden, Estonia and the UK, are typically and consistently towards the top of all of these lists. Now we could ask what these three countries have in common. Clearly all three of them have larger IT and telecommunications sectors than most European countries, but what's behind that? And what has helped support so much employment growth in these sectors? Sweden and the UK both have large audio-visual and broadcasting sectors. Perhaps for different reasons in the UK this reflects their important global position in broadcasting and advertising. While in Sweden this is more likely the results of its strong presence in the development of consumer-oriented media apps, so Skype and Spotify. The UK's strong performance in fintech, business consultancy and e-commerce is also likely to add to its demand for IT specialists. And in Estonia strong efforts to create public services, so e-government as it's known, has probably been a big driver. But really there's no single mechanism. So what you're saying is they've all got very different routes there. Yeah, they're taking different paths to developing this stronger share of digitally dependent employment. And that should be encouraging to other European countries to know that there are different ways to get there and for those that are hoping to increase their own digital sectors. I think the real point from a policy perspective is to remember that those countries with stronger digital employment growth tend to be the ones that have stronger total employment growth and over the past 20 years have also been those with the lowest unemployment rates. Right, so no one should be scared of digitalisation, they should be embracing it really. We should understand that we have opportunities. Okay, and just to circle back very quickly, you mentioned earlier Luxembourg with a very high ratio of digital to non-digital jobs. Just to fill us in, what's their success there? I mean obviously smaller country, but... Well, I think again, like the UK, it has a very large fintech sector that tends to also accompany higher rates of digitally dependent employment. Okay, so financial services and technology again. Valerie, thank you very much for your time today. I'm very happy that you invited us, thank you. This brings us to the end of this episode. So we've talked about the digital revolution and how it's transforming Euro area countries in different ways. We took a close look at labour markets and cleared up the myth that more technological progress automatically leads to less jobs. In fact, it looks like the contrary might be the case. Countries that are more digitalised tend to be associated with lower unemployment rates. Stay tuned for the next episode of the podcast in which we'll focus on how digitalisation can open the door to new jobs with new perspectives. We'd love to hear your feedback and thoughts for future episodes via social media. You can use direct messages and comments. You've been listening to the ECB podcast with Michael Steen. If you like what you've heard, please subscribe and leave us a review. Until next time, thanks for listening.