How macroeconomic variables affect admission and dismissal in the Brazilian electro-electronic sector: A VAR-based model and cluster analysis

Abstract Brazilian industries in the electro-electronic sector impact other areas of the economy, as they contribute to the generation of employment and income. The objective of this research is to analyze the short-term dynamic behavior vector autoregressive (VAR) models of admission and dismissal, using cluster analysis as a pretreatment to classify variable endogeneity and observe the macroeconomic variables that most impact unemployment. Cluster analysis will be used to determine if the macroeconomic variables related to the Brazilian electro-electronic sector can be classified and selected based on their exogeneity. The VAR model will be used to show the contemporary relationship among the variables and, through the use of the impulse response function and the variance decomposition, help explain their dynamics. The shock imputed by the macroeconomic variables causes employment rate instability that takes two months to return to normal. There is a significant relationship (p ≤ 0.01) between the macroeconomic variables and the number of people admitted and dismissed in the sector, indicating that abrupt changes in the macroeconomic variables will result in fluctuations in the number of employees in the Brazilian electro-electronic industry. The cluster analysis was able to classify the variables as endogenous or exogenous variables, but it was not able to order them based on exogeneity.