Renewable sources are relevant in a country's energy planning because they are linked to the creation of opportunities for technological, economic, and productive development guided by the principles of sustainability. Thus, the aim of this study was to investigate the relation between electric generation capacity by renewable and non-renewable energies and Brazilian socioeconomic variables. The analysis of the interrelationships between electricity generation capacity and economic growth in Brazil, from April 2009 to March 2017, was carried out by the vector autoregressive and autoregressive distributed lag methodologies. It was verified that the variance of employment is explained by renewable sources: hydroelectric in 7.71%, biomass in 1.99%, wind energy in 3.13%, and solar energy in 10.58%. While, the GDP variance is explained in 3.15% by hydroelectric energy, 0.06% by biomass, 1.70% by wind energy, and 17.38% by solar energy. The export variance is explained by renewable sources: hydroelectric 2.48%, biomass 0.39%, wind energy 2.34%, and solar energy 17.58%. Finally, the variance of the minimum wage is explained by hydroelectric energy in 1.48%, biomass in 5.09%, wind energy in 9.09%, and solar energy in 10.67%. An ARDL (1, 1, 2, 0, 0, 0, 3, 2, 0, 2, 0, 2) model was also adjusted for natural gas, with AIC (13.082) and BIC (13.739), and the ARDL (1, 0, 1, 0, 0, 0, 0, 0, 3, 0, 0, 4) model adjusted for hydroelectric power, with AIC (13.633) and BIC (14.189), considering the variables' order cited above. Through the adjustment of the ARDL model, it was verified that there is a long-term influence of socioeconomic variables on electricity production variables, both renewable and non-renewable ones. The analysis of the impulse response function and the variance decomposition allowed us to verify that the installed capacity for production of electric energy exerts influence on Brazilian socioeconomic variables considered in this study.
Keywords: Economic growth; Granger causality; Non-renewable energies; Renewable energy; Response impulse function; Socioeconomic development; Variance decomposition; Vector autoregressive model.