Search

This paper analyses employment transitions and workers' skills in Brazil using a random sample from the universe of formal labour contracts covering the period from 2003 to 2018. We develop a novel procedure to derive a measure of occupational distance and internationally comparable skill measures from occupations' task descriptions in the country under analysis based on machine learning and natural language processing methods, but without usual ad hoc classifications. Our findings confirm that workers who use non-routine cognitive skills intensively experience the highest employment growth rates and wages. Their labour market exit risk is relatively low, occupational and sectoral changes are least common and, in the case of occupational switching, non-routine cognitive workers tend to find occupations that are higher-paid and closer in terms of their task content. Against the same characteristics, routine and non-routine manual workers are worse off in the labour market. Overall, there have been signs of routine-biased technological change and employment polarization since the 2014 Brazilian economic crisis.

Quais os efeitos futuros das tecnologias de automa??o sobre o mercado de mercado de trabalho brasileiro? Busca-se responder ? quest?o por meio de tr?s ensaios. Primeiro, avalia-se os efeitos das tecnologias de automa??o sobre o setor p?blico, introduzindo-se uma metodologia de estima??o das ocupa??es mais propensas ? automa??o, chamado Bartik Occupational Tasks (BOT). Segundo, realiza-se proje??es de d?ficits no n?mero de servidores por motivos de aposentadoria e investiga-se a automa??o como alternativas para a reposi??o de servidores. Terceiro, analisa-se as tend?ncias futuras para o mercado de trabalho privado no Brasil por meio da discuss?o e aplica??o das novas estimativas de automa??o. What is the future effect of automation technologies on the Brazilian job market? We answer this question with three essays. Firstly, we evaluate the effects of automation technologies on the public sector employment, introducing the Bartik Occupational Task (BOT) method, in order to build estimates on automation susceptibility of each occupation. Secondly, we approximate the deficit of public workers due to retirement, investigating automation as an alternative to replace them. Thirdly, we analyze the future trends for Brazil?s private job market applying the new automation estimates. Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES

A presente disserta??o ? constitu?da de dois ensaios acerca das desigualdades de oportunidades no Brasil. Mensura-se um limite inferior da desigualdade de renda e de oportunidades a n?vel nacional e estadual no Brasil, evidenciando seus componentes eticamente aceit?veis e inaceit?veis. Aplicando a metodologia param?trica de Ferreira e Gignoux (2011) aos dados da PNAD entre 2001 e 2014, o Ensaio I mostra a diminui??o no ?ndice de desigualdade de oportunidades (IOp) de 0,106 para 0,066, uma queda de 37,37% no per?odo. Em compara??o, a disparidade de renda, mensurada pelo ?ndice MLD, caiu de 0,55 para 0,40, uma redu??o de 27,2%. Da queda na desigualdade total, 53% se deve ? diminui??o na parcela injusta da desigualdade, enquanto a parcela justa representa 47% da redu??o. O peso das circunst?ncias da desigualdade de oportunidades manteve-se praticamente constante, com 58% para a informalidade do trabalhador, 27% para as diferen?as de cor e 14% das diferen?as de sexo. Comparando a renda dos grupos por circunst?ncias, mostra-se que a renda m?dia do grupo em maior desvantagem, ?mulheres n?o brancas informais?, ? de apenas 29,5% da renda do grupo em maior vantagem, ?homens brancos formais?. Todavia, os grupos em maior desvantagem foram os que mais se beneficiaram do crescimento da renda. Entre os estados, Santa Catarina oferece o melhor ?ndice de oportunidades, enquanto o Piau? ocupa a ?ltima posi??o. Ficou clara a desvantagem relativa da regi?o Norte e Nordeste, regi?es que apresentam sistematicamente os ?ndices de desigualdade de oportunidades mais elevados. No Ensaio II, analisa-se especificamente o estado do Rio Grande do Sul no ano de 2010 utilizando-se os ?ndices de Desigualdade de Oportunidades Municipais (IOp-M) calculados por Figueiredo et al. (2013). Observa-se como os resultados est?o distribu?dos espacialmente no territ?rio do estado e a sua rela??o com vari?veis produtivas. A an?lise LISA permitiu verificar a presen?a de efeitos de transbordamento do IOp-M de uma cidade para seus vizinhos. A forma??o de clusters espaciais mostra que as desigualdades injustas sofrem depend?ncia espacial. Considerando exclusivamente a estrutura produtiva, o Princ?pio da Compensa??o de Roemer (1998) sugere que a regra alocativa de recursos p?blicos demandaria um direcionamento de recursos para as regi?es com maiores IOp-M?s e mais dependentes do servi?o p?blico e agricultura, enquanto os munic?pios com maior participa??o da ind?stria, com?rcio e servi?os t?m necessidades menores de compensa??o. The present dissertation is composed by two essays on inequality of opportunities in Brazil. It is measured the inferior bound of income inequality and opportunities inequality in Brazil, bringing evidences to its ethical and unethical components. Applying the parametric methodology developed by Ferreira and Gignoux (2011) to PNAD data between 2001 and 2014, the Essay I shows the reduction in the inequality of opportunities index (IOp) from 0,106 to 0,066, a drop of 37,37% in the period. In comparison, inequality of income measured by the MLD index decreased significantly from 0,55 to 0,40, a drop of 27,2%. This fall in total inequality is composed in 53% by a decrease in the unfair share of inequality, meanwhile, the fair share is responsible for 47% of the reduction. The weight of circumstances in inequality of opportunities remains nearly constant, with 58% corresponding to the informality of workers, 27% for differences in color, and 14% for differences in sex. Comparing the income of the circumstances groups, is showed that the mean income of the most disadvantaged group, ?nonwhite informal women?, consists only in 29,5% the mean income of the most advantaged group, ?white formal men?. However, the most disadvantaged groups were the ones that gained more with the growth of income. Amongst the states, Santa Catarina offers the best index of inequality of opportunities, meanwhile Piau? sits at the least position. It is clear the relative disadvantage of states in North and Northeast regions, presenting systematically the larger indexes of inequality of opportunity. The Essay II studies the inequality of opportunity in the state of Rio Grande do Sul in the year of 2010 using the Municipal Inequality of Opportunity Index (IOp-M) constructed by Figueiredo et al. (2013). The spatial distribution of the results is analyzed and its relations with productive variables. The LISA analysis allows to verify the presence of spillovers of IOp-M from a city to its neighbors. The formation of spatial clusters shows that unfair inequalities suffer from spatial dependence. Considering solely the productive structure, Roemer?s (1998) Compensation Principle suggests that the allocative rule of public resources demands a redirection of resources to regions with higher IOp-M`s which are more dependent on public sector and agriculture, meanwhile municipalities with higher industry, commerce, and services participation have lower compensational needs.