Summary
In this paper, experimental, computer simulation methods are used to demonstrate how a depreciation-type adjustment influences the distributional form of accounting earnings. The results confirm conjectures that earnings distributions generally, with or without depreciation adjustments, tend towards a normal form as a function of increasing 'activity' levels. They also indicate that depreciation is likely to accelerate the transition towards a normal form as activity levels increase and to transform a non-normal form to one that is significantly closer to the normal at relatively low activity levels. The impact of the fixed asset 'impairment' rules is also investigated. The results reported in the paper have implications for standard-setting, risk analysis and inference using accounting earnings and related numbers, including ratios based upon earnings.
See the full content of this document
Extract
The Impact of Depreciation-Type Adjustments On the Distribution of Accounting Earnings
(ProQuest-CSA LLC: ... denotes formulae omitted.)
1. IntroductionThis paper reports the results of a series of computational experiments investigating the statistical properties of accounting numbers. It uses Statistical Activity Cost Theory (SACT) to model the impact of the depreciation adjustment on the form of the probability distribution of accounting earnings. It extends previous work based upon the same theory, which shows how the depreciation adjustment may reduce the variance of accounting earnings (e.g. Lane and Willett, 1997).The paper is motivated by both theoretical and practical considerations and an interest in the statistical properties of accounting numbers. Concern has been diverted away from measurement aspects of accounting theory in recent years by a focus upon disclosure and user-related aspects of accounting. However, there remain important unanswered questions, relating to the purpose and usefulness of depreciation-type adjustments, that the approach taken in this paper addresses. This has the potential to inform the manner in which accounting information could be better derived, disclosed and used.The results of the experiments reported in this paper demonstrate hitherto unknown properties of depreciation-type adjustments. By depreciation-type adjustments is meant accounting entries that are added to or subtracted from earnings, which utilise information about the expected time duration and lifetime costs of activities to supplement the information provided by accrued transaction (invoice) costs. Such adjustments are shown to affect the shape as well as the dispersion of the statistical distribution of accounting earnings. Specifically, depreciation-type adjustments that use reasonably accurate estimates of lifetimes and expected costs improve, more or less dramatically, the approximation of the distribution of undepreciated earnings to the normal distribution. Moreover, as the analysis of the impairment provisions shows, ignoring new information or incorporating changes in depreciation-type calculations in the absence of objective data reduces or negates this effect.The theoretical framework and modelling approach demonstrated here could be a useful addition to the technical armoury supporting policy choices of standard-setters. The assessment of the 'quality' of earnings is a key concern. To what extent may a depreciation-type adjustment enhance the quality of earnings? It is difficult to see how such a question can be adequately answered without a systematic and rigorous method for evaluating the relative merits of alternative earnings calculations according to some criteria. The findings also have implications for the analysis of risk in financial reporting and cognate areas such as management accounting and auditing. Additionally, they have significance for empirical work using regression analysis containing earnings or functions of earnings variables, including financial ratios.The next section of the paper briefly reviews prior research relevant to the issue at hand. Section 3 then provides the motivational background to the paper. Section 4 discusses the modelling framework and this is followed by more detail in Section 5 outlining the method used and the variables examined in the computer experiments. After Section 6 describing the approach to analysing the samples, the results of the experiments are presented in two parts. Section 7 reports the results on the limiting form of the earnings distribution. Section 8 reports the results of experiments concerning the distributional form that earnings, whether inclusive of a depreciation adjustment or not, takes at relatively low activity levels. Section 9 extends the results to assess the effect on earnings of depreciation pr...See the full content of this document
Sponsored links
