Operating Asymmetries and Propensity Score Matching in Discretionary Accrual Models

• In Event: Earnings Management and Discretionary Accrual Proxies

Abstract

Earnings management research often uses discretionary accruals from Jones-type expectation models. These standard models assume a linear relation between sales changes and accruals. However, we predict and find that sales changes have a non-linear asymmetric effect on accruals
through managers’ operating decisions. By forcing a linear specification on this non-linear effect, the modified Jones model overestimates discretionary accruals for moderate sales changes and underestimates them for extreme sales changes. This causes excessive type-I error in tests of positive (negative) discretionary accruals for subsamples with moderate (extreme) sales growth. We generalize the performance matching approach of Kothari, Leone, and Wasley (2005) and use propensity score matching on multiple accrual determinants to address the bias caused by non-linear effects of sales changes and other variables. This modification improves type-I errors relative to standard matching on ROA, successfully mitigating the bias, and changes inferences about some of the major findings in the literature.

Authors

• Rajiv D Banker, Temple University

• Dmitri Byzalov, Temple University

• Shunlan Fang, Kent State University - Kent

• Byunghoon Jin, Temple University