Unexpected Changes in Quarterly Financial-Statement Line Items and Their Relationship to Stock Prices

ABSTRACT This study examines the value-relevance of six quarterly financial-statement line items (accounts receivable, inventory, current liabilities, gross margin, SGA expense, and depreciation expense) and finds the size of deviations from one-step-ahead predicted values of the six items is associated with abnormal stock returns. With the exception of inventory, results are consistent with the theory that transitory changes in line items introduce greater noise into the earnings number. Consistent with Jiambalvo, Noreen and Shevlin (1997), inventory changes are viewed as positive leading indicators of firm value. Both unsophisticated models (random walk and random walk with drift) and more complex models (Box-Jenkins ARIMA and vector autoregressive models) are developed for the line items. Overall findings are generally insensitive to the degree of sophistication of the expectation model employed. INTRODUCTION Fundamental information analysis, as defined by Lev and Thiagarajan (1993), aims at identifying financial variables that are useful in security valuation. In this paper, evidence is provided that unexpected changes in quarterly financial-statement line items affect the value of the underlying firm's stock. While there is considerable empirical evidence with respect to the relationship between both quarterly earnings numbers and annual financial-statement line items and security returns,1 the current study extends this research, investigating the nature and extent of this relationship through examining quarterly values of selected financial-statement line items. The line items chosen have been shown by past research to be closely related to earnings. Both income-statement and balance-sheet line items are examined, since each provide information useful in describing and predicting firm earnings or security returns (Lipe 1986, Bernard and Noel 1991, Ohlson and Penman 1992, Stober 1993, Cheng 1998). The line items are accounts receivable, inventory, current liabilities, gross margin, SGA expense, and depreciation expense. Four timeseries models are employed as expectation models: account-specific Box-Jenkins ARIMA models, a random walk, a seasonal random walk, and a vector autoregressive model. Once time-series models for the selected line items are developed, the value-relevance of unexpected changes in these items is examined. I examine the link between line items and stock prices, rather than the link between unexpected values of line items and unexpected earnings, due to the nature of the relationships among these various pieces of information. Unexpected values of line items may not have a transparent effect upon earnings, since the potential change in earnings may be offset through other accruals in an attempt to manage earnings. Indeed, some of the line items examined in this study have been manipulated by firms in order to misstate revenue.2 However, in an efficient and rational security market, the price of a firm's stock should reflect not just current earnings, but earnings expectations based on all available information. It logically follows that current stock prices reflect information about future earnings - such as information contained in unexpected values of financial-statement line items - before that information is reflected in current earnings. This study adds to the body of work that examines the relationship between outputs of the financial-reporting process and the market value of the firm. One reason line items are expected to be value-relevant is because they can provide information that is useful in ascertaining whether changes in earnings are transitory or permanent, a distinction that has been shown to be important in determining firm value.3 Evidence is provided that there is information in the transitory elements of financial statement line items that is consistent with the information set determining security prices. A longwindow event study is performed to determine whether the market recognizes a distinction between permanent and transitory changes in quarterly earnings and in line items, as determined by different values of the Earnings Response Coefficient (ERC). …