Earnings persistence of restating firms: Should all earnings restatements be treated equally

The rapid increase in the volume of restatements in the past decade has led to a large volume of academic research on the characteristics of restatement firms and the negative market reaction to the announcement of a restatement. Using earnings persistence as my proxy for earnings quality, I attempt to determine if restatements are just a short-term fix of the accounting problems in the restated year alone or if there are longer-term effects on the earnings quality of restatement firms after the most recent restated year. My analysis combines a short-term and long-term approach to earnings restatements by directly comparing the persistence of the restated and originally reported earnings from the most recent restated year and looking to see how the earnings persistence changes as a result of the restatement. I also consider the impact of the cause and initiator of the earnings restatement on future earnings persistence in the years after the most recent restated year.;I find that the restated earnings are more persistent than the originally reported earnings which let me conclude that earnings restatements do "fix" the accounting violations specific to the restatement year. I also provide evidence that restatements have a significant impact on earnings persistence beyond the restatement year by showing that, relative to a control sample of similar non-restating firms, there is a larger increase in earnings persistence from the period before to the period after the restatement year for earnings restatement firms. I also find that the earnings persistence in future years of firms that restate core accounts is higher relative to firms that restate non-core accounts. Overall, I believe that earnings restatements do have a positive effect on the earnings quality of restating firms beyond just the restatement year and that the type of account being restated is important in any analysis that involves firms that restate earnings.