RESUMO Este trabalho se propôs a identificar os fatores que podem explicar as atribuições dos ratings, com especial atenção ao impacto do uso de derivativos. A lacuna explorada por esta pesquisa reside no ineditismo em se analisar a percepção, por parte das agências de ratings, dos reflexos causados pelas informações relacionadas ao uso de derivativos pelas companhias brasileiras de capital aberto. Além disso, este estudo transfere os achados anteriores que se debruçaram sobre os analistas de ações para as agências de ratings, reforçando a discussão sobre a complexidade dos derivativos no processo de avaliação do risco de crédito. A contemporaneidade deste tema de pesquisa se dá pela recente adoção da norma International Financial Reporting Standards 9 (Comitê de Pronunciamentos Contábeis - Pronunciamento Técnico n. 48), que entrou em vigor em janeiro de 2018. A partir dessa normatização, a principal novidade apresentada neste artigo foi a verificação do reflexo do uso de derivativos utilizados pelas empresas com a finalidade de hedge nos seus ratings de créditos, ajudando a suprir uma lacuna empírica na literatura da área. Os resultados encontrados contestam a teoria de que o uso de derivativos para hedge é visto positivamente pelos investidores. No entanto, apesar de nenhum impacto significativamente estatístico ter sido encontrado nos ratings das empresas que utilizam derivativos, observou-se que as empresas que usam derivativos e têm os maiores valores nocionais foram as que receberam as melhores notas da agência Moody’s. Com isso, ampliamos o debate sobre a complexidade das informações atreladas ao uso dos derivativos. Foram analisados 2.090 ratings atribuídos a companhias não financeiras com ações negociadas na Brasil, Bolsa, Balcão [B]³ entre 2010 e 2016, por meio de análise dos dados em painel, conferindo maior robustez às análises e aos achados. Contrariando a hipótese central desta pesquisa, os resultados aqui apresentados mostram que, no Brasil, as empresas que se utilizam de instrumentos financeiros derivativos para hedge não recebem as melhores notas de classificação de crédito por parte das agências de ratings. Uma das principais contribuições deste estudo centra-se nos indícios de que as agências Standard & Poor's e Moody’s não foram capazes de incorporar, de maneira consistente, as informações relacionadas ao uso dos derivativos, ampliando a discussão sobre a complexidade desses instrumentos financeiros.

ABSTRACT The purpose of this study is to identify the factors that may explain the attribution of credit ratings to firms, focusing especially on the impact of derivatives. The gap explored by this research lies in the novelty of analyzing how rating agencies perceive the effects caused by information related to derivatives use by Brazilian publicly-traded companies. In addition, this study shifts the previous findings from stock analysts to rating agencies, reinforcing the discussion about the complexity of derivatives in the credit risk assessment process. This research topic is currently of interest due to the adoption of International Financial Reporting Standard (IFRS) 9 (Accounting Pronouncements Committee - CPC - 48), which came into effect in January of 2018. Based on these rules, the main novelty presented in this article was its verification of the effect of the derivatives used by companies in order to hedge their credit ratings, thus helping to fill the empirical gap that exists in the literature from the area. The results found challenge the theory that the use of hedge derivatives is viewed positively by investors. However, although no significant statistical impact was found on the ratings of companies that use derivatives, it was observed that the companies that use derivatives and have the highest notional values were those that received the best ratings from Moody’s. With this we broadened the debate about the complexity of the information linked to derivatives use. In the study, 2,090 ratings attributed to non-financial companies with stocks traded on the Brasil, Bolsa, Balcão [B]³ exchange were examined between 2010 and 2016 by using panel data analysis, which lends robustness to the analysis and findings. Contrary to the central hypothesis of this research, the results presented here show that, in Brazil, companies that use derivative financial instruments for hedging do not receive the best credit ratings from rating agencies. One of the main contributions of this study is the evidence that Standard & Poor’s and Moody’s were unable to consistently incorporate information related to derivatives use, thus broadening the discussion about the complexity of these financial instruments.

RESUMO O presente estudo investiga se as empresas de capital aberto brasileiras gerenciam os tributos diferidos em direção a uma estratégia de big bath. O "big bath” earnings management é uma estratégia na qual as empresas reconhecem elevados valores de perdas não recorrentes. O estudo obtém resultados originais que suportam a hipótese do big bath com base no CPC 32/IAS 12. As despesas de tributos diferidos representam uma forma conveniente de reduzir os resultados, por conta de sua subjetividade e sua tempestividade. O estudo propõe uma estratégia específica de investigação para a análise do excesso de tributos diferidos, que é baseada na elevada homogeneidade de ambas as normas contábeis e a legislação tributária nas empresas de capital aberto brasileiras. Essa análise provê novas evidências de ajustes de big bath que não foram encontradas pela literatura contábil atual. O estudo analisa 226 empresas de capital aberto brasileiras no período de 2011-2015 e constrói um modelo linear para a estimação do excesso de tributos diferidos, o qual é baseado na independência condicional entre tratamento-efeito decorrente das exigências constantes na norma contábil CPC 32/IAS 12. A análise segue a abordagem convencional dos mínimos quadrados com variáveis de controle como a análise principal. O estudo aplica a abordagem de dois estágios para o controle de possível viés de variável omitida. Foram encontrados indícios de que empresas de capital aberto brasileiras gerenciam tributos diferidos em direção a uma estratégia de big bath. Os resultados indicam que as empresas com prejuízos divulgam um excesso significativo de despesas de tributos diferidos, e esse excesso é positivamente relacionado ao crescimento dos prejuízos.

ABSTRACT This study investigates whether Brazilian loss-making firms manage deferred income tax as a form of big bath strategy. "Big bath” is a strategy in which a firm manages earnings by intentionally recording large non-recurring losses. We found original evidence supporting the hypothesis of big bath through the managing of deferred taxes under CPC 32/IAS 12. Deferred tax expenses can be used as a tool for reducing earnings because of the subjectivity and timing involved. To analyze the excess of deferred taxes, we propose a particular research strategy that is based on the increased homogeneity of accounting standards and tax regulation in Brazilian listed firms. This analysis provides new evidence of big bath adjustments that was never described before in the literature. We analyze 226 Brazilian listed firms for the 2011-2015 period. We designed a linear model to estimate deferred tax excess that is based on the conditional independence between treatment and effect under accounting standard CPC32/IAS 12. For our baseline analysis, we used least squares with controlling covariates. We also used two-stage least squares to control for omitted variables bias. This paper finds evidence that Brazilian firms can manage deferred income tax as a form of big bath. Results indicate that loss-making firms disclose significantly higher excesses of net deferred tax expenses, and that these excesses increase with losses.

Abstract Market analysts, by means of issuance of their opinions (earnings per share, target prices, and recommendations), have been the object of several earlier studies. Previous literature identifies analysts as information intermediaries between companies and investors, and identifies their role in reducing the information asymmetry. This study investigates whether the estimated standard deviations of price-target shares issued by capital market analysts are informative, and ascertains whether it is possible to identify smaller errors in analysts' forecasts from the verification of the consensus among them. The study is carried out in Latin American countries, and relies on a database of 23,367 estimates of target-price shares during the period from October 2010 to January 2017. It also takes into account the number of analysts who issued the estimates, the company market value, and the government effectiveness between the countries. The results indicate that the greater the consensus (smaller standard deviation), the smaller the forecast errors. Thus, the standard deviation of target-price estimates presents an informative tool to investors about forecast accuracy. Another important result shows that the greater the government effectiveness, the greater the forecast accuracy of target-price estimates issued by the analysts.

The stock analysts have a relevant role in the capital market, since, directly or indirectly, they contribute to the paper pricing and to the composition of the investment portfolio. The purpose of this study is to verify if it is possible to obtain extraordinary returns, above those offered by a market portfolio, with the monitoring of the stock recommendations issued by Brazilian capital market analysts, one of the most important in Latin America. Based on a wide range of consensual recommendations concerning the period from 2000 to 2010, and with the monitoring of the historical series of paper returns covered by the analyses, the performance of two portfolios were compared, one formed by stocks that received favorable and the other one formed by stocks that received unfavorable analyst recommendations. The results showed bias in recommendations, since there is, systematically, a greater number of favorable against unfavorable recommendations. The results mainly showed that the analysts were unable to identify the stocks that actually offered greater returns within the period considered.

Los analistas de acciones desempeñan un papel importante en los mercados de capitales, de modo que, directa o indirectamente, contribuyen a la fijación de precios de las acciones y a la composición de portafolios. El propósito de este estudio fue verificar si es posible obtener rendimientos extraordinarios, por encima de los ofrecidos por una cartera de mercado, con el acompañamiento de las recomendaciones de acciones realizadas por los analistas en el mercado de capitales brasileño. Basado en una gama de recomendaciones de consenso para el período comprendido entre los años 2000 y 2010, más el seguimiento de series de tiempo de los rendimientos de los títulos que abarca el análisis, fueron comparados la actuación de dos carteras, una formada por acciones que recibieron recomendaciones favorables y otra formada por las acciones que recibieron recomendaciones desfavorables de los analistas. Los resultados mostraron la existencia de la parcialidad en las recomendaciones, ya que hay, sistemáticamente, un mayor número de recomendaciones favorables en comparación con las recomendaciones desfavorables. Y se mostraron, principalmente, que los analistas no fueron capaces de identificar las acciones que ofrecieron una mayor rentabilidad en el período estudiado.