Several authors suggest that the managerial labour

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: xecutive directors. We empirically test the assertion that earnings management is negatively associated with average non-executive director’s tenure. Several authors suggest that the managerial labour market for outside directorships rewards effective outside directors with additional positions as directors, but disciplines outside directors who have a record of poor monitoring performance (Fama and Jensen 1983; Milgrom and Roberts 1992). Empirical evidence indicates that non-executive directors of financially-distressed firms lose outside directorships after leaving the board of the troubled firm (Gilson 1990) and that non-executive directors of firms charged with accounting and disclosure violations by the SEC are more likely than others to lose their other directorships (Gerety and Lehn 1997). Hence, non-executive directors have an incentive to monitor effectively because the fact that they are directors of well-run firms signals their competence to the managerial labour market. For that reason, the number of directorships a board member holds is a signal of his or her competence. Also, additional directorships allow the director to acquire governance competencies and to gain knowledge of best practices for boards of directors. These results suggest that additional directorships may be associated with monitoring effectiveness. Thus, we expect a negative relationship between the number of outside directorships of non-executive directors and the level of earnings management.1 4. Research design The objective of this study is to determine whether good governance practices reduce the level of earnings management as measured by the discretionary accruals. Our sample is 14 drawn from the population of US firms that appear on Compustat in 1996. In order to increase the power of our tests, we choose those that have the largest amount of discretionary accruals (both negative and positive) and those that have almost no discretionary accruals. Discretionary accruals estimation We use both income increasing and decreasing discretionary accruals because discretionary accruals can be used either to conceal poor performance or to save current earnings for possible use in the future (DeFond and Park 1997). Moreover, the reversing nature of accruals make it possible that a firm that had large income increasing accruals in 1995 has to reverse them in 1996, the year we are observing it. Our sample is based on the complete set of firms on Compustat with a December 31 1996 year-end and complete accruals data for 1996. We exclude firms from regulated (SIC 4000 to 4900), financial (SIC 6000 to 6900), and government (SIC 9900) sectors because their special accounting practices make the estimation of their discretionary accruals difficult. The discretionary component of the total accruals is estimated with the modified Jones (1991) cross-sectional model (Defond and Jiambalvo 1994; Francis, Maydew and Sparks 1998; and Becker, DeFond, Jiambalvo, and Subramanyam 1998). This requires the estimation of a cross-sectional regression for each industry (two-digit SIC codes), so we eliminate firms from industries with less than ten firms. These requirements leave 3,947 observations for the calculation of discretionary accruals. Discretionary accruals (DAC) for each firm i in industry j are defined as the residual from the regression of total accruals (the difference between Cash from Operations and Net Income) on two factors that explain non-discretionary accruals, the increase in revenue and the level of fixed assets subject to depreciation. [( ) ( ) ( ˆ ˆ ˆ DACijt = TAC ijt A ijt −1 − α j 1 A ijt −1 + β1 j ∆RE ijt A ijt −1 + β 2 j PPE ijt A ijt −1 where: 15 )] (1) DACijt = Discretionary accruals for firm i in industry j in year t; TACijt = Total accruals for firm i in industry j in year t; A ijt −1 = Total assets for firm i in industry j at the end of year t -1 (Assets-Total in Compustat); ∆RE ijt = Change in net sales for firm i in industry j between year t-1 and t. (Sales (Net) in Compustat); PPEijt = Gross property, plant and equipment for firm i in year t (Property, Plant and Equipment (gross-Total)). ˆˆˆ Where α j , β1 j , β 2 j are the industry-specific estimated coefficients from the following crosssectional regression. ( ) ( ) ( ) TACijt A ijt −1 = α j 1 A ijt −1 + β1 j ∆RE ijt A ijt −1 + β 2 j PPE ijt A ijt −1 + e ijt Consistent with DeFond and Park (1997) and Subramanyam (1996) we drop 438 firms with income and cash flows from operation in excess of the top and bottom 2% of all observations and the 58 firms with a large Cook distance to eliminate the effect of outliers.2 The sample used to estimate equation (2) separately for each of the 39 industries that meet our requirements contains 3,451 firms. Discretionary accruals are then computed for each firm from equation (1). To obtain a sample composed of firms with high levels of DAC and firms with almost no DAC, we rank the 3,451 firms by the size of their DAC. Since we want to include both positive and negative discretionary accruals, of the firms for w...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online