

臺大管理論叢
第
26
卷第
3
期
85
Working Capital Management and Estimations of Discretionary
Accruals
1. Purpose
Topics regarding earnings quality and earnings management have been the focus of
much academic research. Because managers can manipulate discretionary accruals at low
costs without being easily detected, they tend to employ accruals for earnings management
to meet earnings target. However, no existing method can be used to accurately measure
discretionary accruals. In other words, the creditability of empirical outcomes relies on
whether the estimation model used can effectively estimate discretionary accruals. The Jones
(1991) model is among the methods used in academic research to estimate abnormal
accruals, which serves as a proxy for discretionary accruals. Other available models include
the modified Jones model reported by Dechow et al. (1995) and performance adjusted (or
matched) model reported by Kothari et al. (2005). These models extend or revise the Jones
model and mainly rely on revenue growth and fixed assets to estimate normal accruals.
Jones-type models are developed by regressing total accruals on changes in revenue,
which are proxies for changes in the operating environment. However, companies adjust
their working capital in response to changes in the operating environment, which may not be
concurrently observed through changes in revenue. In other words, working capital
management would affect accruals. Stickney and Brown (1999) argued that revenue fails to
instantly reflect the effect of working capital management on the operations of companies. In
addition, Collins et al. (2011) contended that revenue growth fails to completely capture the
effect of working capital management on accruals. Overall, accrual models estimated
through changes in revenue cannot distinguish accruals resulting from working capital
management as nondiscretionary accruals and create biased estimates (Ball and Shivakumar,
2008).
Furthermore, biased estimates of discretionary accruals could be related to the direction
of working capital management. A company at the growth stage is attributable to the
expansion of investment in fixed assets and working capital, which results in a greater
amount of accruals. However, when a company is facing a recession, little investment in
working capital is required, and the company may even reduce its working capital.
Chi-Hua Li
, Assistant Professor, Department of Accounting, Fu Jen Catholic University
Yan-Jie Yang
, Associate Professor, College of Management, Yuan Ze University
Kuo-Chih Cheng
, Professor, Department of Accounting, National Changhua University of Education