Elasticidade PIB e desemprego no estado de São Paulo: uma aplicação do modelo vetorial de correção de erro (VEC)

This study analyzed the relationship between number of unemployed workers in the metropolitan region of São Paulo and the GDP of the state of São Paulo for the period from January 2003 to December 2017. Data from the Fundação Sistema Estadual de Análise de Dados (Seade) were used. Method X-12 was used to deseasonalize the series. For the determination of the integration order of each series, the Dickey-Fuller Aumentado (ADF) unit root test was used. In order to verify if the variables converge to the equilibrium in the long term, the Johansen cointegration test was used. From the vector error correction model (VEC), it was possible to estimate the long-run elasticity between GDP and the number of unemployed. The result shows that for every 1% change in GDP, unemployment is reduced by 3.49% in the long term, which was estimated to be equal to one year. The result of the Impulse Response Function shows that after a shock in the GDP, unemployment declines for twenty months and then stabilizes, indicating a level of full employment in the metropolitan region of São Paulo.