Do Markets Remember Accounting Changes? An Examination of Subsequent Years

Abstract

This paper provides a direct test of investor valuation of accounting changes in the years subsequent to the change. We focus on changes in the estimate of the useful life of long-lived assets because these revisions have ongoing effects on reported income that do not reverse in the near term. Descriptive evidence comparing useful life increasers to useful life decreasers and to control firms indicates that useful life increases represent earnings management while decreases do not. We regress returns on earnings components to assess how investors value these changes in the year they occur and the subsequent two years. Results suggest that investors (i) discount the positive effect of useful life increases in the year of the revision, (ii) value the negative effects of useful life decreases in the year of the revision, and (iii) do not forget about the earnings effect of these accounting changes when valuing the firm in subsequent years. These results are consistent with the efficient market hypothesis.