CEO departure and discretionary accounting choices

Dechow and Sloan [1991] investigate the hypothesis that CEOs, during their final years of office (the "horizon" years), manage discretionary expenditures to improve short-term earnings performance. Using a sample of 261 firm-years, this study extends the Dechow and Sloan model by including additional control variables. It also examines whether the discretionary components of earnings (discretionary accruals, discretionary revenue, and capital expenditures) provide departing CEOs a monetary incentive (bonuses) to manipulate these income factors. The general results of this study do not support the hypothesis that departing CEOs have a greater monetary incentive than incumbent CEOs to manage discretionary earnings to maximize their bonus schemes. A possible reason this hypothesis is not supported may be due to the fact that previous research has treated incumbent and departing CEOs as separate, homogeneous samples a treatment that the extant income-smoothing and CEO turnover research suggests may be flawed. Income smoothing literature provides evidence that some incumbent CEOs manipulate earnings to a predetermined target to avoid a "ratcheting" of expectations while CEO turnover research suggests that the "relay" process mitigates some departing CEOs' manipulations of earnings. Since agency theory predicts that management of accounting earnings will vary between groups of incumbent and departing CEOs, as well as within these two groups, the present study partitions the sample on the median change in operating cash flows for departing CEOs. This study finds evidence that departing CEOs in the above-median partition do increase income-enhancing discretionary accruals in their final year with the firm, and they have a significant economic incentive to do so. However, there is apparently no economic incentive for departing CEOs with an above-median change in operating cash flows to reduce discretionary revenue or capital expenditures.