The effects of ex-CEO’s board service and CEO change type on the association between CEO changes and goodwill impairments: Evidence from U.S. companies during 2002-2009

Abstract

The purpose of this paper is to examine whether there is an association between CEO changes and goodwill impairments in U.S. companies during 2002-2009, and how ex-CEO’s board service and CEO change type affect this association. In 2001, the FASB introduced a new goodwill accounting standard SFAS 142, which allows goodwill to be tested for impairment instead of periodic amortization. To determine whether goodwill is impaired, managers estimate the fair value of goodwill, which is highly subjective in nature. Many researchers have raised concerns about this increased managerial discretion with regard to goodwill accounting.

The highly subjective nature of goodwill impairment tests makes goodwill impairments an ideal earnings management tool for managers. Many prior studies document that new CEOs tend to record initial “earnings baths”, especially after CEO resignations. It is hypothesized that new CEOs use goodwill impairments in their baths, and that this is driven by CEO resignations. CEOs might resist the write-off of personally-acquired goodwill because the write-off could harm their reputation. It is further hypothesized that ex-CEO’s presence on the board prevents goodwill impairments by the new CEO.

The topic of this paper is important because of goodwill’s increased importance in financial statements, substantial managerial discretion afforded by SFAS 142, and managerial incentives related to CEO changes. There is also little prior research on CEO’s board service.

Data used in this study is obtained from Compustat database. This study focuses on years 2002-2009 because SFAS 142 was introduced in 2001, and data when CEOs have left their companies is stored until 2009 on Compustat. The final sample consists of 30,625 firm-year observations. Hypotheses of this paper are studied using both logistic and linear regression. Decisions of whether to record a goodwill impairment in a given year or not are of primary interest.

The main findings of this paper are the following. First, it is shown that CEO changes and goodwill impairments are positively and significantly associated, implying that new CEOs use goodwill impairments in their earnings baths. Second, it is shown that ex-CEO’s board service does not affect the positive association between CEO changes and goodwill impairments, despite the incentives for preventing impairment. Third, it is shown that the positive association between CEO changes and goodwill impairments is driven by CEO resignations and not by CEO retirements. Finally, it is shown that goodwill impairments after CEO changes are also larger.