Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Vincent C. Brenner


There is little extant accounting research on the turnaround phenomenon, when a financially distressed firm regains its financial health rather than merging with another firm or filing for protection under the Bankruptcy Act. The goal of this research was to determine whether financial and other publicly available information could be used to predict whether or not a distressed firm would turn around. To achieve that goal, a sample of distressed firms was identified based on their initial rather than final financial condition. A measure external to the firm--Standard and Poor's earnings and dividend ranking for common stock--was used to identify the financial condition of the sample and assign a value for the dependent variable (turnaround or continued distress). A model of recovery from bankruptcy developed by Casey, McGee and Stickney (1986) was used to model recovery from financial distress. Six logistic regression models (all significant) were used, consisting of deflated (by the nonresidential fixed investment implicit price deflator) and undeflated static and change proxies for the independent variables. All four independent variables (Size of the Firm, Free Assets Percentage, Prospective Earnings and Management Ownership Concentration) were significant in the deflated static model. The predictive accuracy of the model was assessed with the jackknife procedure. The model correctly classified sixty-five percent of the total sample, seventy-one percent of the turnaround firms and sixty-three percent of those firms that continued in distress.