Avaliação de ofertas públicas iniciais no Brasil segundo as abordagens de Múltiplos e de Lucros Residuais

Abstract

The main objective of this study was to verify which of the Comparable Method and the Residual Income Model has the best ability to evaluate Initial Public Offerings in Brazil between 2010 and 2018. The Comparable Method was applied based on the Price-to- Earnings (P / E), Price-to-Book (P / B), Price-to-Sales (P / S), Price-to-EBITDA and an evaluation given by the four multiples together, called the Joint Evaluation. The Residual Income Model was adopted according to two assumptions regarding the behavior of Residual Profits after the IPO. The first case considered that the companies would present the same Residual Income observed in the year prior to the IPO in the three years after the event, resulting in the model called RIM – RE (Residual Income Model – Residual Earnings). The second case considered that the companies would present the same ROE observed in the year prior to the event in the years following the event and gave rise to two models. The first one considered the existence of Residual Profits for the sample in a period of three years after the IPO, generating the RIM - ROE (Residual Income Model - Returno on Equity) model. The second considered that the third year Residual Income obtained with the ROE of the year prior to the IPO would be perpetual, resulting in the model RIM - ROE (P) (Residual Income Model - Return on Equity (Perpetuity). The models were compared with the offer price and with the closing price of the first trading day in the market, due to the underpricing phenomenon in the universe of IPOs. It was verified that the Joint Evaluation as well as the multiple P/EBITDA were presented as the most accurate models, with Average Absolute Percent Error of 49%. The model RIM - ROE (P) presented the best adjustment of the model, in the amount of 28,41%, however, it presented a Mean Aboslute Percentual Error of 62%. We also found that valuation models are more accurate when companies present positive accounting information as a basis for the valuation. There is not enough evidence to state that the models are more accurate pricing for companies with more than 10 years at the time of the IPO, as indicated by Kim and Ritter (1999), since only one model among all used presented better pricing for these companies.