Discretionary accounting changes in relation to income smoothing, firm characteristics and bankruptcy prediction models

Abstract

The purpose of this study was to extend the body of research pertaining to: (1) the use of discretionary accounting changes by management; (2) possible characteristics that are useful in isolating those firms most likely to initiate discretionary accounting changes; (3) the impact of discretionary accounting changes on bankruptcy prediction models. Seven hypotheses were used to investigate these areas: (H(,o)-1) Bankrupt firms do not make on the average significantly greater numbers of discretionary accounting changes than nonbankrupt firms. (H(,o)-2) The magnitude of the discretionary accounting changes made by the two groups is not significantly greater. (H(,o)-3) Firms that report discretionary accounting changes have the same median size as measured by total assets as the nonchanging firms. (H(,o)-4) Firms that report discretionary accounting changes have the same median size as measured by total revenues as the nonchanging firms. (H(,o)-5) The use of discretionary accounting changes by firms (changers or nonchangers) is independent of the major industry classification of the firm. (H(,o)-6) The types of discretionary accounting changes that are reported are independent of the group (bankrupt or nonbankrupt) membership. (H(,o)-7) The discriminating ability of the discriminant function excluding DAC as a potential discriminating variable is not significantly different from the discriminating ability of the discriminant function including DAC as a potential discriminating variable. ...