Li-Wen Hong , Shin-Kun Peng (p)

We adopt Montagna and Noccos (2013) unionization framework and introduce search frictions into homogeneous sector to study how the unionsbargaining power affects the unemployment rate before and after trade. When the unions bargaining power increases, the minimum production threshold required to stay in the market decreases. It allows the entry of relatively less productively firms. The selection soften effect may be generated through the reduction in the toughness of competition via different labor demand elasticity of firms. Firms with higher productivity pay higher wages to labors. That is, raising unions bargaining power do harm to the firms with higher productivity, and then it may leave the market space for firms with lower productivity. The number of entrant is decreasing and the number of surviving firms depends on the magnitude of unionsbargaining power. In autarky, we find out that when the unionsbargaining power is low enough, and increase in the bargaining power would cause the fall of aggregate unemployment rate. The effect of increasing in surviving firms outweighs the shrinking in average labor demand of firms. When the bargaining power is in the median level, the number of firms is still increasing but the average labor demand per firm decreases significantly. The later effect dominates the former; therefore, thus the total number of unemployed labors is increasing. When the bargaining power is high enough, then the unemployment rate increases in the bargaining power. After international trade is allowed,firms now could choose to serve domestic and foreign market or only producing for domestic market. Labor unions negotiate wages with distinct profit centers within a firm, that is, domestic and exporting departments. Labors in each department may earn different wages because of the variable price elasticity faced by different departments. We highlight the case that unions have asymmetric bargaining powers in two countries. Increasing in unions bargaining power of one country always lowers the total number of employed labors in exporting sector, while the number of employed labors in domestic sector depends on the trade-o¤ between the magnitude of trade openness and bargaining power, so does the unemployment rate.

Merja Kauhanen (p)

Rapid technological change including robotisation, digitalization and artificial intelligence is transforming labour markets and the structure of employment. As regards labour markets this development has also raised concern about its impact on employment and technological unemployment. The estimates given by different experts and evaluations differ as to how many jobs are threatened by this development in the future (see Brynjolfson and Mcafee 2014; Frey and Osborne 2013 & 2016; Autor 2015). There is relatively much earlier research on the impact of ICT on employment and its structure, but still much less research on the impact of robots on employment.

The aim of this paper is to provide new empirical evidence on the impact of an increased robot usage on employment in seven European countries between 1993-2015 and thus contribute to the developing literature on the labour market impacts of robotisation.

The analysis utilises robot data on industrial robots (and later on data on service robots) combined with employment and other relevant data from EUKLEMS. The seven European countries that are (so far) used in the analysis are Germany, France, UK, Italy, Spain, Finland and Sweden. These are countries for which there is data available for all the years between 1993-2015. We analyse the employment effects of robotisation for the whole period but also separately for the subperiods 1993-2007 and 2008-2015 of which the latter also includes the financial crisis years of 2008-2009 and the recovery period after that. We utilise OLS and IV regressions in the empirical analyses.

The preliminary results suggest that there is a negative impact of industrial robots on hours during the period 1993-2015 for the seven countries used in the estimation when controlling for the country fixed effects and initial wages and capital-labour ratio. The impact is negative and statistically significant but very small. The corresponding OLS estimates by the skill level suggest that the impact of increased robotisation has a statistically significant negative impact only for the low-skilled workers' working hours.

Begoña Cueto, Matías Mayor (p), Patricia Suárez

In the definition of economic resilience, entrepreneurial activity can play a key role (Martin, 2012; Martin & Sunley, 2015). Therefore, the analysis of determinants of entrepreneurial resilience can help to understand why some regions can retain a higher level of entrepreneurial activity than others. At the same time, recent studies (Fotopoulos & Storey, 2016; Fritsch & Wyrwich, 2014) point out that entrepreneurship rates have risen in the last years, however, the regional differences remain constant over the time. On one hand, this finding implies the existence of a persistence spatial heterogeneity, and on the other hand, the reduced impact of entrepreneurial spatial spillovers. The aim of this research is to analyse the determinants of entrepreneurial resilience of the Spanish regions focusing on this spatial context. First, we discuss how the expected regional behaviour should be computed, second, we analyse the contribution of the local related and unrelated variety and industrial specialization together with the impact of these characteristics in the neighbouring regions using a spatial econometric model. Third, we consider in which measure a lower improvement in terms of accessibility is acting as a dead weight reducing the recovery capacity.

Nikolaos Terzidis (p), Steven Brakman , Raquel Ortega-Argiles

This meta-analysis systematizes the empirical evidence on the labor market impacts from various types of factor-biased technical change and participation in global value chains. Our evidence draws upon 1162 empirical estimates from 71 scientific articles published between 1987 and 2017. Despite the vast heterogeneity in the estimated outcomes and the weak evidence of publication bias in the relevant literature, our three-step empirical process documents how the encountered features of heterogeneity alter the likelihood for outcomes of specific sign and significance. In particular, factor-biased technical change is found to hurt industrialized labor markets more in terms of employment rather than wages, while the negative effects concentrate on low-skilled workers. In contrast, engaging in global value chains benefits skilled employees in developed countries and trade in services is associated with positive labor market outcomes compared to trade in manufacturing.