Content area
Full Text
Introduction
During recent years, the issue of earnings value relevance has received great attention from both investors and managers. Managers are interested in sustaining an attractive earnings growth, since their position and compensation are closely related to the earnings figures reported in the firm's financial statements. Analysts, on the other hand, are interested in providing useful and reliable information for investors in order to achieve profitable investing decisions. By the term "value relevance", we mean the ability of financial statements to summarize valuable information that affect stock price movements and assist investors to assess the value of the firm.
The "true and fair view" principle has been legally embedded in general accepted accounting principles (GAAP), in the majority of the developed countries worldwide. However, the reality sometimes is quite different. Many studies in the past have documented the existence of "creative accounting" through managers' efforts to manipulate earnings figures for convenience ([31] Naser, 1993; [36] Shah, 1998; [28] McNichols and Wilson, 1998). Consequently, the existence of earnings manipulation by the market participants could result in stock market returns that deviate from their originally correct values, resulting in a misleading picture for the market and the returns-earnings relation. Thus, an effective measure of earnings relevance could prove very helpful in assessing the power of future stock price movements and the overall earnings-return relation.
Many studies in the past have documented that earnings figures and earnings changes are associated with positive abnormal returns ([27] Latane and Jones, 1979; [17] Foster et al. , 1984; [7] Bernard and Thomas, 1989). However, this relation is expected to alter depending on the validity of earnings change. To state it differently, the above relation may be true if we observe a genuine improvement in profitability, or false if manages have applied aggressive accounting to manipulate the earnings figures. Thus, the inclusion of an earnings quality measure in the returns-earnings models could yield more convincing results for the aforementioned relation.
Researchers have developed many methodologies in order to determine how genuine the financial statements are. One group of studies has attempted to build models to predict possible management fraud. For instance, [6] Beasley (1996) and [21] Hansen et al. (1996) used logit regression analysis and generalized qualitative response models, respectively. [20]...