In value relevance literature, there are two basic types of valuation model that have extensively used by prior studies. Price model testing how firm`s market value relate to accounting earnings and equity book value. As seen in equation 1, based on Ohlson (1995), the model expresses the firm value as a function of its earnings and equity book value. The other type of value relevance valuation model is return model, which describes relation between stock returns and accounting earnings. The value relevance of accounting information studies using returns-earnings association is motivated by the seminal work of Ball and Brown (1968). Then, Easton and Harris (1991) popularized a specific version of annual return model including both earnings levels and earnings changes. The return model used in this study, based on Easton and Harris (1991), is provided in equation 2.
Chen et al. (2001) discuss two advantages of price models compare to return models. When stock markets anticipate any components of accounting earnings and incorporate the anticipation in the beginning stock price, i.e. prices leading earnings, return models will bias earnings coefficients towards zero. On the other hand, Kothari and Zimmerman (1995) argue that price models yield unbiased earnings coefficients because stock prices reflect the cumulative effect of earnings information. Return models only assess the value relevance of accounting earnings, whereas price models based on Ohlson (1995) show how a firm`s market value is related both accounting earnings and equity book value. In addition, the use of Ohlson model will expand the scope of assessing value relevance into both income statement and balance sheet.
However, Kothari and Zimmerman (1995) argue that return models have less serious econometric problems of heteroscedasticity than price models. In fact, we can find large number of value relevance studies use both price and return models. Following previous studies and based on Kothari and Zimmerman suggestion, this study uses both price and return models in assessing value relevance of accounting information.
P_it=α_0+α_1 E_it+α_2 EBV_it+ε_it 1)
P_it = stock price of firm i (at three months after end of year t),
E_it = Earnings per share for firm i during period t,
〖EBV〗_(it ) = Equity book value per share for firm i at the end of period t.
R_it=α_0+α_1 E_it/P_(it-1) +α_2 ((E_it-E_(it-1)))/P_(it-1) +ε_it (2)
R_it = Stock return firm i for year t (annual return from month -9 to +3),
E_it = Earnings per share of firm i for year t,
E_it-E_(it-1) = Change in annual earnings per share,
P_(it-1) = The stock price at the beginning of nine months prior to fiscal year end.
This study estimates price model and return model described above to examine the hypothesis, and the results of the two models are expected to complement each other. To...