The Effects of Earnings Variables on Stock Returns among Public Companies in Norway: A Multiple Regression Analysis

The usefulness of accounting information to stock investors is measured through value relevance studies. Value relevance studies typically apply multiple regression analysis to investigate the ability of accounting variables to explain stock returns. This paper examines the statistical association between stock returns and income statement information. All analyses are conducted on a sample of Norwegian listed companies, based on accounting and stock market data provided by the Oslo Stock Exchange. The study starts out with a standard value relevance analysis where stock returns are regressed on aggregate accounting earnings. More complex analyses are then introduced; the earnings numbers are split into underlying components and the regression coefficients are allowed to be dependent on the sign of earnings. The more advanced analyses illustrate that traditional value relevance studies generally underestimate the usefulness of financial reporting. This paper suggests that the value relevance as measured by the explanatory power of regression analysis doubles if both the sign and the disaggregation effect are incorporated into the analysis. 1. Introduction One objective of financial reporting is to assist stock investors in predicting cash flow and estimating company value (Holthausen & Watts, 2001). Accounting information can be termed value relevant if it contains the variables used in a valuation model or assists in predicting those variables (Francis & Schipper, 1999). Thus, value relevance can be seen as a measure of accounting usefulness from the perspective of stock investors (Beisland, 2009). In practice, value relevance is typically measured as the statistical association between accounting information and stock prices or returns (Francis & Schipper, 1999). Lev (1 989) assesses the usefulness of accounting earnings by evaluating a large number of studies on the relationship between stock returns and accounting earnings. He finds that most studies report a remarkably low statistical association between stock returns and current earnings. More recent studies also find that earnings information often is able to explain less than 10% of the variation in stock return (Ali & Lee-Seok, 2000; Ball, Kothari, & Robin, 2000; Brimble & Hodgson, 2007) and some studies even report that the value relevance of earnings has been decreasing over time (S. Brown, Kin, & Lys, 1999; Francis & Schipper, 1999; Kim & Kross, 2005). However, several papers suggest that the seemingly low value relevance is a matter of mis-specification of statistical models. For instance, a number of empirical analyses have shown that the informational content of earnings is sign dependent (Basu, 1997; Hayn, 1995; Joos & Plesko, 2005). The return-earnings association is not constant across earnings levels and prior research (Francis, Schipper, & Vincent, 2003; Hayn, 1995) suggests that the return-earnings association improves if sign dependant earnings response coefficients are applied in the empirical analyses. Empirical studies have also documented that the value relevance of earnings may increase substantially as earnings are disaggregated into components (Barth, Beaver, Hand, & Landsman, 2005; Barth, Cram, & Nelson, 2001; Carnes, 2006; Ohlson & Penman, 1 992). This conclusion holds both as earnings are split into underlying line items (Carnes, 2006; Ohlson & Penman, 1992) and as earnings are split into cash flow and accrual items (Barth, et al., 2005; Rayburn, 1986). None of the above listed value relevance studies incorporate both the sign effect and the disaggregation effect simultaneously. In fact, most value relevance studies tend to ignore both effects. This study adds to existing research by showing that the sign effect and the disaggregation effect are incremental to each other, and none of the can be disregarded. Prior research which has failed to recognize that earnings components generally do not add up in valuation (Pope, 2005) has typically underestimated the usefulness of accounting information for stock investors. …

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.