To further validate that the observed higher discretionary accrual sin the early years of CEOs’ service represent earnings overstatement, we examine whether these accruals reverse.
We use a sub sample corresponding to CEOs who stay in office for at least nine years.
Given that discretionary accruals to manage earnings can take multiple years to reverse, this sub sample should provide for a long enough CEO tenure for observing if and when discretionary accruals reverse after early years of CEOs’ service.