Filtros de cartéis baseados em dinâmicas de preço: uma aplicação ao varejo de combustíveis do Brasil

Abstract

This work aims to identify and implement methodologies capable of filtering anticompetitive behavior by analyzing the dynamics of firms’ prices in the Brazil’s retail gas market. The objective was to evaluate methods capable to filter markets with the greatest collusion potential, analyzing pricing patterns similar to ones used by the sector’s cartels. The main reason for this study is to contribute in the existing cartel evidences to the SBDC (Brazilian System of Competition Defense), which still lacks depth studies on cartel detection instruments. To implement the proposed objectives, it analyzed the cartel cases judged by CADE along with the characterization of the market. The pricing patterns associated with collusive market were compared with simulated competition data, through the construction of indicators estimated by time series methods to compose own statistics of each competitive nature. The time series models chosen were based on the tests of: cointegration, price asymmetry, variance, structural breaks and changes of regimes. Finally, it evaluated the indicators on their ability to differentiate competitive from collusive behavior in Brazil’s retail gas market, also it was built an indicator to differentiate these behaviors. The overall results showed that are relevant dynamic price for filtering cartel being analyzes of variance over time , variance between firms and pricing regime changes, the most significant to infer the possibility of collusion.