Research Policy, Volume 53, Issue 3, 2024, 104928
International Journal of Data Science and Analytics
Word embeddings have proven extremely useful across many NLP applications in recent years. Several key linguistic tasks, such as machine translation and transfer learning, require comparing distributed representations of words belonging to different vector spaces within or among different domains and languages to be aligned, known as embedding alignment. To this end, several existing methods exploit words that are supposed to have the same meaning in the two corpora, called seed lexicon or anchors, as reference points to map one embedding into the other.
Research Policy, Volume 53, Issue 2, 2024, 104908
This paper provides empirical evidence on the role of technology in affecting the relationship between the participation of EU countries and industries in Global Value Chains (GVCs) and their employment structure over the period 2000–2014. The empirical analysis is based on country-sector level data for 21 EU countries on employment, trade in value added, patents and investments in intangible assets, and focusses on backward linkages within GVCs.
Research Policy,Volume 53, Issue 2, 2024,
Establishment closures have lasting negative consequences for the workers displaced from their jobs. We study how these consequences vary with the amount of skill mismatch that workers experience after job displacement. Developing new measures of occupational skill redundancy and skill shortage, we analyze the work histories of individuals in Germany between 1975 and 2010. We estimate difference-in-differences models, using a sample of displaced workers who are matched to statistically similar non-displaced workers.
CESifo Working Paper No. 10955
This paper measures the exposure of industries and occupations to 40 digital technologies that emerged over the past decade and estimates their impact on European employment. Using a novel approach that leverages sentence transformers, we calculate exposure scores based on the semantic similarity between patents and ISCO-08/NACE Rev.2 classifications to construct an open–access database, ‘TechXposure’. By combining our data with a shift–share approach, we instrument the regional exposure to emerging digital technologies to estimate their employment impact across European regions.
Pillars Report
The rise of robots has raised a controversial discussion about their disruptive impacts on domestic labor markets. Robot adoption, which is largely concentrated in a few high-income countries, might also affect labor markets of trade partners through global value chain linkages. This effect could be even more pronounced in developing countries due to the predominance of routine tasks and labor-intensive activities.
Pillars Report
The increasing size and importance of digital platforms in various fields of economic activity has raised expectations that virtually everything will be organised via platforms in the future. But will the digital transformation really go that far, or are there limits to it? To answer this question, we have collected and assessed evidence and theoretical arguments about the scope of the gig economy.
CESifo Working Paper No. 10876, Munich, 2023
We study the effects of technological change on immigration flows as well as the labor market outcomes of migrants versus natives. We analyse and compare the effects of two different automation technologies: Industrial robots and artificial intelligence. We exploit data provided by the Industrial Federation of Robotics as well as online job vacancy data on Germany, a highly automated economy and the main destination for migrants in Europe.
CESifo Working Paper No. 10865, Munich, 2023
This paper investigates the effect of robotization in high-income countries on firm-level North-South trade along the value chain. Using a novel combination of data sources including firm-level export data, input-output linkages, and robot adoption, we show contrasting implications for Southern firms. Increased exposure to robot adoption in the destination country of exports reduces firm-level exports in case of robot adoption in the same industry. However, the opposite holds when accounting for input-output linkages and trade along the value chain.
Review of International Economics 2023; 1–30
We study how technology adoption and changes in global value chain (GVC) integration jointly affect labor shares and business function specialization in a sample of 14 manufacturing industries in 14 European countries in 1999–2011. Increases in upstream, forward GVC integration directly reduce labor shares, mostly through reductions in fabrication, but also via other business functions. We do not find any direct effects of robot adoption; robotization affects labor only indirectly, by increasing upstream, forward GVC integration. In this sense robotization is “upstream-biased”.
World Bank Research Observer
In this paper we analyze the evidence of job polarization—the relative decline of mid-wage jobs—in developing and emerging economies. We carry out an extensive literature review, revealing that job polarization in these countries is only incipient compared to advanced economies. We then examine the possible moderating aspects explaining this lack of job polarization. We distinguish three groups of explanations: Limited technology adoption; structural change; and changes in the global value chains.
Pillars Report
This paper investigates the impact of job training on workers’ susceptibility to automation. Using rich individual-level data from the Programme for the International Assessment of Adult Competencies (PIAAC) across 37 industrialized countries, we construct a unique individual-level measure of automation risk based on the tasks performed at work. We uncover substantial variation in automation risk within detailed occupations,which would have been overlooked by previous occupation-level automation measures.