Every penny counts: comprehensive income value relevance from valuation, informational and forecasting perspectives

Abstract

The present dissertation is organized in three studies and aims to investigate the value relevance of Comprehensive Income (CI) and Other Comprehensive Income (OCI) and OCI components from valuation, informational and forecasting perspectives compared to P&L (Profit or Loss) after the mandatory adoption of the International Financial Reporting Standards (IFRS) by Brazilian listed companies. Since FASB (Financial Accounting Standards Board, the issuing body of the United States accounting standards) and the IASB (International Accounting Standards Board, the issuing body of IFRS) required the presentation of both P&L and OCI in the Financial Statements, based on the clean surplus theory, this topic drew attention of academics and practioners around the world. Clean surplus supporters argue that the financial statements prepared in accordance with this paradigm provide more useful information to stakeholders than its alternative (dirty surplus approach). The value relevance research of accounting data aims to examine the correlation between accounting information and the information used by investors to make economic decisions. The information can be considered relevant when it influences the decision making of users of financial statements by helping them predict future cash flows and/or confirm previous assumptions. Therefore, based on clean surplus and value relevance theories, the studies provides empirical evidence of clean surplus income measure (CI) value relevance compared to dirty surplus income measure (P&L) by examining the relative and incremental association of P&L, CI, OCI and OCI components with (i) share prices and (ii) share returns. Additionally, the forecasting ability of P&L, CI, OCI and OCI components to predict future OCF (Operating Cash Flows) and future P&L is empirically examined. The present study adopts existing well-established research models on value relevance and forecasting ability of P&L, CI, OCI and OCI components and develops additional statistical models based on those theories and concepts. This dissertation analyzes hand-collected data for a sample of Brazilian companies listed on BMF&BOVESPA and comprises the period from 2010 to 2015. The results indicate that P&L is more value relevant than CI, even though CI provide value relevant information. However, the CI coefficient is lower than P&L coefficient. Although OCI does not provide incremental value relevant information, OCI components add incremental value relevant information compared to standalone P&L, especially adjustments in fair value of available-for-sale financial instruments, gains and losses from translating the financial statements of a foreign operation and adjustments in fair value of cash flow hedging instruments. The present dissertation provided additional insight to the ongoing discussion on value relevance of P&L, CI, OCI and OCI components in the research community, as well as on the standard setter level and contributed to fill, even partially, the lack of research on the issue in the Brazil.