Date of Award

Fall 2002

Document Type


Degree Name

Doctor of Business Administration (DBA)


Economics and Finance

First Advisor

Rohan Christie-David


The purpose of this study is to develop a model that predicts failure and estimates the time of survival of dotcoms using a number of financial and non-financial factors. This model can be used as a warning tool for stockholders, creditors, and consumers to protect themselves from such failures.

I employ the Cox (1972) Proportional Hazards Model in a cross-sectional and time-varying context using financial data over the 1998–2001 period. Results from a cross-sectional analysis reveal that the coefficient estimates for variables CFTL and NSTA are consistently negative and highly significant. This suggests that higher sales and cash flows lower the potential of failure.

The results also show that NITA is negative and significant at the 10% level, suggesting that higher revenues improve the survivability of a firm. Moreover, TLTA and WCTA show no significant effect on failure. On the other hand, the coefficient estimate on TA is positive and highly significant, suggesting that larger firms have higher odds of failure. This could be the result of an unsustainable growth rate among dotcoms. The excessive and rapid need for external sources of funds may raise the concerns of creditors about the financial position of the company and can lead to higher cost of funds and closer monitoring.

The results from event-time data show qualitatively similar findings. However, the coefficient estimate for TA becomes negative. On the other hand, the event-time model does not show much significance in the overall effect of the regressors.

The time-dependent analysis, however, shows a few differences in results, in that; sales have no significant effect on the potential of failure. In contrast, the coefficient estimate on NITA becomes negative, and highly significant.

Results also reveal that stock returns add little to the predictive capability of these models. Moreover, matching companies by size to account for the size effect do not significantly alter the results. Finally, findings from industry-specific models, namely, retail, service and manufacturing, are not conclusive.