 Okay, so hello everyone. Thank you for being here. So my name is Estefania Alban. I'm from University of La República in Uruguay And I'm going to present the paper entitled why has technological change not close the gender wage? I want to thank also UN wider and the International Economic Association for the opportunity of presenting this So driving by technological advances labor market in high-income countries have experienced relevant structural changes One of the most documented consequences is a relocation of employment from routine work to non-routine tasks these two these Chops in which technology complements work a phenomenon that has been called routine bias technological change So this has led to a growing in the territory investigating the effects of technological change on inequality And extend to which labor markets are increasingly polarized However, we still know little about the effects of technological change on gender inequality Although these structural changes are in principle gender neutral They may result in gender bias impacts if men and women have different employment distribution At risk occupations that were differently affected by technological change And therefore this might have had relevant implications for the dynamics of the gender wage disparities So this paper there's a question of how have changes in wages and employment structure At risk occupations associated with the technological progress affected the gender wage cap So with the advancement of technology the occupations that love the most in terms of employment and wage returns Are those traditional male-industrial occupations which have a high content of routine tasks? So by looking at the employment structure recent work suggests that women have relatively Benefit from technological change and that they are more likely to be employed at cognitive high wage occupations Despite this the comrades in the gender wage cap staniated in the last decades in most of the developed countries has shown by Extensive literature put it in question the extent to which the impact of technological change on Occupational wages has been favorable to female workers So this paper comes to reconcile some way these two pieces of evidence by Investigating why is it that the large wage gains that are observed in those occupations in which women are increasingly employed in Detriment of traditional male-industrial occupations have not led to a further reduction in the gender wage cap So to answer this question this paper uses administrative panel data for Germany And investigate the effect of technological change on wage trajectories for men and women across occupations That were differently affected by technological change The German labor markets offers an interesting case to study this question first because it has one of the highest and most persistent gender gaps among developed countries and Also having a large industrial sector the effect of technological change on the employment structure has been Remarkable as shown by previous work So what I do I first estimate changes over time in the occupation specific wage premiums These are defined as the component of the workers potential wage that is common to all workers in a given occupation in a year after accounting for the effect of industrial selection of workers in After accounting for the effect of occupation specific return to individuals skills and for that I use the panel data approach of Cortez Which I correct for the industrial selections of workers into occupations. I consider five for all occupation groups distinguishing between tasks that Can be automated and tasks that require analyzing or interacting with others and where often the knowledge it complements work And then I analyze how those changes in the wage premiums for male and female workers Affect the gender wage gap distinguishing between a composition or sorting channel That is the gender differences in the employment distributions and the gender differences that take place within occupations so Well, this paper contributes to extensive literature on the causes of the gender inequalities in the labor market and Although the women have been upgrading their occupations the occupational structure is still a relevant factor in explaining the gender gaps and In particular there's a literature that focus on the effect of technological change On gender gaps and it does it by looking on the employment structure that is Investigating to what extent women shops are more or less subject to automation compared to those of males For example negate and petrongolo and Serena and co-authors find that the relocation of labors from good producing sectors To service industry favorite women by creating jobs that are less physically demanding and more intense in interactive skills Also by looking at the task content of shops some recent literature find that there was a rise in the use of interpersonal tasks in The US and the women have a comparative advantage in those increasingly value skills My paper is particularly related with this black and speed so in a which also focus on West Germany And they find that women have women's increases in non-routine analytical and interactive tasks within industry and occupation cells which they interpret as a positive effect of technological change on a women's job In this paper by looking not only at the employment, but mostly at the wage dynamics across occupations I contribute to this literature finding that although the Occupational structure has favored women a wage premiums for main workers grew more rapidly than those of females within the analytical and interactive non-routine occupation and this explains why although women have been less exposed to the Automation of work and increase their employment in these highly value skills Occupations and this did not lead to a further reduction in the gender wage gap So my paper uses the administrative social security records for Germany This is a sample of integrated labor market biographies Which is a 2% random sample of all mandatory notifications made by employers to social security agencies My sample is composed of male and female workers aged between 25 and 55 years old in West Germany for the period 1975 to 2010 Because there is no clear information on ours work. I restrict my main analysis to full-time workers So that wages are comparable although I do some robustness also with the part-time and Then I combine this data with the information on tasks coming from a representative labor forced cross-section Each covering around 30,000 Individuals and from this information what I use is the workers of reports on the task contents of their work The tasks are the activities that workers have to perform normally during their jobs I consider a classification of five dimension based on the task content of occupations based on the the one that is developed by speed trainer and Basically, there are five categories. The first is the analytical non-routine Which includes activities such as research and analyzing the second one interactive non-routine activities that require interacting with other people such as Managing or teaching these are high skill high pay occupations. Then we have cognitive routine activities such as calculating bookkeeping Manual routine. These are typical industrial occupations. These are like medium-grade occupations And finally we have the manual routine, which are basically non-skill service and repairing So I use this information on tasks from the qualification and career survey and then aggregated at the 3d sheet occupational level to classify my occupations in the administrative data So following previous literature I interpret changes along these tasks groups as being related to routine bias technological change So here are some descriptive so my sample here. The only thing I want to highlight is that So So that men are mostly employed in the manual routine Group while women are mostly employed in the interactive non-routine and then if we go to the high skill Jobs and men are more represented in the analytical routine while women in the interactive So here are the changes in employment shares by occupation groups and What we can see is that in line with the routine bias technological hypothesis There was a strong decrease in the manual routine groups for both men and women of around 15 percent For men this decline was explained was compensated by increases in both extreme of the wage distribution manual routine and analytical non-routine and for women workers There was a strong increase in the interactive non-routine group So to estimate the occupational wage premiums. I follow the panel data approach of Cortes 2016 assuming that productivity is locked linear in skills the potential wage in occupation J For an individual eye of a scale level set Can be defined as a wage premium occupation component that is common to all workers in occupation J at the empty Component that interacts the individual skills and the return occupation specific return to those skills In principle, I will assume that these skills are time invariance. So we can Call this second component gamma, which is an occupation spell fix effect Which varies for an individual across occupations, but it stands contents whenever the individual stays in the same occupation So I'm pretty golly the server wage depends in which occupation the individual is selected So there's a dummy that takes value one if the individual selects it into one of these five world occupation groups And the regression also includes year fix effects and controls the place of work at the federal state Experience where the individual is chairman or a foreigner They identify an assumption is that selection into occupations only depends on the occupation wage premiums and Indi-U.s workers ability that is the individuals will try to select in the Occupation which they can have the higher returns given these two factors so So Cortes developed this Empirical approach to analyze the effect of technological change for a sample of many workers in the US But as I want to analyze the gender effects of technological change I allow occupation wage premium to differ by gender so I do this by introducing an interaction term with with the occupation fix effects and Adami for female Occupation time fix effects and another me for female. So I estimate these details with our Which can be interpreted as the wage premium for male and female workers Because of the inclusion of the occupation spell fix effect the occupation time fix effects are identifying only from variation Over time within occupation spells. So this should be interpreted as a lot of difference That is they identify the changes over time in the occupational wage premiums relative to the Manual routine, which is the omitted category in the in the regression the manual routine for male So then I want to explain how these changes in the wage premiums contribute to the gender wage cap. I Consider the employment distribution across these five occupation groups and The two complementary channels which can explain how the changes in the occupation specific wage premiums affect the gender wage cap The first one is the composition or sorting across occupation That takes place if women are less likely to be employed at those occupations that most increase the gender wage premiums The the the wage premiums then the second one is the within occupation differences that take place If women obtain a smaller increase in those occupation premiums than men for the same occupation group So I follow the approach of card and routers who did something very similar to the composer's Film premiums, so I adapt this to the composite changes in the occupation premiums Using this wajaka blinders style the composition. So the first term is the within occupation which considers the differences in the Estimated coefficients considering either the fate the male or female distribution and then The gender the sort the second term is the sorting across occupations which considers the the differences in the distribution taking either the female or male coefficients So here the results this is considering The estimating for a sample of male and female or together So no gender differences in the coefficients and we can see that in line with the robustness by with the technology and spectral Technology so we can see that there's a strong decrease in the manual routine occupations and increase in the cognitive especially in the analytical non-routine and in the interactive non-routine so the manual routine is the The omitted categories. So all of them all of these are with respect to the manual non-routine And here are the results when estimating gender specific occupational premiums So what we can see again this are all with respect to Manual routine for men and what we can see is that there is Much strong dispersion in the changes in occupation premiums for male There is a strong increase in the in the cognitive Groups the analytical and routine and interacting non-routine while for women. There's some string Increase in the analytical and routine, which is much lower compared to to that of men and These two occupations which are like the ones that employ most of the women Evolved very similar to the manual routine for men, which is the omitted So then I go to see how these changes in the in the wage premiums affect the gaps considering the the employment distribution and What we can see is that if we take the mix by gender occupation premiums This will have been like favored to women but when we consider the gender specific occupational wage premiums the Variation for a females is negative while for male is positive 40% of variance so if Men and women Receive the same variation in the occupation premiums Then the gender wage cap will have decreased by 40% instead of the 20% to 21% that we observe then the composing the wage premiums gap between the two channels we can see that The fact that occupational wage premiums for men grow more rapidly than female premiums with a certain occupation groups Is the main explanatory factor of the average gender differences over time and the sorting across Occupations in fact contributes to decrease the the wage gaps So I think I'm going to skip the cohort analysis and the robustness checks I'm going to conclude so this paper uses administrative panel data for Germany to Investigate the effect of technological change on the dynamics of the gender wage Differentials the effect of gender differences in certain at risk of patients has mostly Benefit women Contributing to narrow the gender wage cap However, what I find is that wage gains for male workers within cognitive occupations Groom more rapidly than those of females and this effect is in stronger for young cohort of workers. So We should not expect that this effect is the Decreased like in near future So one possible lesson from the paper is that it can be misleading looking only at the show automation exposure Although lower exposition to the automation of work and occupational Grading women still face certain constraint that did not allow them to benefit from the increase of overall wage returns in the upper part of their skill distribution so Policies aim at reducing gender gaps are still very relevant and we should not expect that the Structural change by itself is going to decrease these differences So yeah, we can finish here and how some more time for question