Capacidade de geração de emprego e flexibilidade do mercado de trabalho no Brasil (2002-2008) : uma análise de insumo-produto integrada com indicadores setoriais de flexibilidade salarial
Resumo
Resumo: Este estudo se propõe a usar duas análises estruturais distintas para o mercado de trabalho brasileiro na última década: o modelo insumo-produto de Leontief-Miyazawa e o modelo econométrico Curva de Salário. O primeiro permite decompor setorialmente o emprego total gerado da economia em direto-indireto e induzido por meio de Matrizes Insumo-Produto (MIP) de 2002 e 2008, para, em seguida, comparar as principais mudanças entre os dois períodos. Pretende-se ainda confrontar o emprego total estimado e o emprego efetivamente observado em 2008, extraindo um "viés" setorial do modelo e, assim, verificar em uma meta-análise se existe alguma relação entre esse viés e uma das hipóteses que fragilizam o modelo, a rigidez dos salários. Para isso, é necessário obter estimativas estruturais do grau de flexibilidade salarial por setor de atividade por meio da Curva de Salário de Blanchflower e Oswald (1994a), com base nas PNADs de 2002 a 2009. Em geral, os resultados mostram que: 1) apesar das expressivas alterações no padrão de consumo e renda das famílias, não houve grandes mudanças estruturais na composição do tipo de emprego gerado; 2) o grau de flexibilidade salarial para o Brasil é menor do que em décadas anteriores; 3) apesar de algumas limitações metodológicas e na base de dados, a curva de salário se mostrou útil em proporcionar estimativas setoriais do grau de flexibilidade salarial; 4) esta variável demonstrou ter associação com o viés percentual de previsão do emprego no modelo insumo-produto, indicando ser relevante sua inclusão na análise. Abstract: This study proposes to use two distinct structural analysis for the Brazilian labor market in the last decade: the Leontief-Miyazawa input-output model and the Wage Curve econometric model. The first one allows sectorally decompose the total employment generated in the economy in direct-indirect and induced by input-output matrices during the years of 2002 and 2008, and then compare the major changes between the two periods. It also aims to confront the total employment estimated and the employment actually observed in 2008, extracting a sector "bias" of the model, and then, verify through a meta-analysis if there is any relationship between this bias and one of the hypothesis that weakens the model, the wage rigidity. For this it is necessary to obtain structural estimates of the rate of wage flexibility by activity sector through the Wage Curve Blanchflower and Oswald (1994a), based on PNAD 2002 to 2009. In general, the results show that: 1) despite significant changes in household consumption and income pattern, there were no major structural changes in the composition of employment generated type, 2) the rate of wage flexibility in Brazil is lower than in previous decades, 3) despite some methodological limitations and in the database, the wage curve proved to be useful in providing sectorial estimates of the wage flexibility rate, 4) this variable have demonstrated association with the percentage bias in employment forecast input-output model, indicating that their inclusion is relevant in the analysis.
Collections
- Teses & Dissertações [9430]