EVALUATION OF SHORT AND LONG TERM EFFECTS ON BRAZILIAN COCOA PRODUCTION
Price, Exports, Production, Cocoa, VECM Model.
This work aimed to verify the dynamics of how price and export have influenced the Brazilian cocoa production. For this, the following methods were used: augmented Dickey-Fuller unit root test (ADF), Johansen Cointegration test, Vector Autoregressive Model (VAR), Vector Error Correction Model (VECM), Breusch-Godfrey test , Portmanteau test for serial autocorrelation, ARCH effect and Jarque-Bera test for normality of residuals. The data used in the work were obtained from the Institute of Applied Economic Research (IPEA), ComexStat and the Brazilian Institute of Geography and Statistics (IBGE), for the period from 1994 to 2020. Regarding the results, it was observed that the variables Cocoa Quantity Produced (QPC), Cocoa Exports (EXP) and Average Cocoa Price (PMC) have a long-term relationship with at least one cointegration vector. Short-term variables do not significantly influence cocoa production, but when considering a 10% significance level, there is the influence of the production of 4 years ago and the price of the previous year. The adjustment coefficient revealed that cocoa market imbalances are adjusted at a rate of 40.7% per year. Serial correlation and autocorrelation were not detected in the residuals, which have a normal distribution and there is no arch effect.