Essays on SMEs insolvency risk

Gupta, Jairaj

October 2014

Thesis or dissertation

© 2014 Jairaj Gupta. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright holder.

In light of the new Basel Capital Accord, Small and medium size enterprises (SMEs) play a fundamental role in the economic performance of major economies. Several lending communities proposed to treat SMEs as retail clients to optimize capital requirements and profitability. In this context, it is becoming critically important to have a detailed understanding of its risk behavior for appropriate pricing of credit risk. Thus, this thesis presents four essays on SMEs insolvency risk starting from chapter 3 through chapter 6 that investigates different dimensions of their default risk. My first essay makes distinction among SMEs that report operating cash flow and those which do not while modeling their default risk. However, I do not report any significant improvement in model’s classification performance when operating cash flow information is made available. Similarly, my second essay considers domestic and international SMEs separately while modelling their default risk and report almost identical classifications performance of the models’ developed for both the groups. The third essay compares the default risk attributes of micro, small and medium-sized firms respectively with SMEs. Test results suggest significant difference in the default risk attributes of only micro firms and SMEs. On a different line, my fourth essay deals with the methodological issues that have been witnessed recently in the bankruptcy literature that use hazard models for making bankruptcy predictions. This essay highlights the critical issues and provides appropriate guidance for the correct use of hazard models in making bankruptcy predictions. Here, I also propose a default definition for SMEs which considers both legal bankruptcy laws and firms’ financial health while defining the default event. Empirical results show that my default definition performs significantly better than its respective counterparts in identifying distressed firms with superior goodness of fit measures across all econometric specifications. Detailed abstract of respective essays are as follows.

Evidence pertaining to SMEs financing strongly motivates me to believe that firms which are unable to generate sufficient operating cash flow (OCF) are more susceptible to bankruptcy. However, the role of OCF in bankruptcy of SMEs lacks empirical validation. Thus, my first essay (chapter 3) investigates the role of operating cash flow information as predictors in assessing the creditworthiness of SMEs. One-year distress prediction model developed using significant financial information of United Kingdom SMEs over a period of 2000 to 2009 confirm that the presence of operating cash flow information does not improve the prediction accuracy of the distress prediction model.

My second essay (chapter 4) considers domestic and international small and medium-sized enterprises (SMEs) of the United Kingdom separately while modelling their default risk. To establish the empirical validation, separate one-year default prediction models are developed using dynamic logistic regression technique that encapsulates significant financial information over an analysis period of 2000 to 2009. Almost an identical set of explanatory variables affect the default probability of domestic and international SMEs, which contradicts the need for separate default risk models. However, the lower predictive accuracy measures of the model developed for international SMEs motivate me to compare the weights of regression coefficients of the models developed for domestic and international firms. Test results confirm that four out of the nine common predictors display significant statistical differences in their weights. However, these differences do not contribute to the discriminatory performance of the default prediction models, given that I report very little difference in each model’s classification performance.

A huge diversity exists within the broad category of Small and medium size enterprises (SMEs). They differ widely in their capital structure, firm size, access to external finance, management style, numbers of employees etc. Thus, my third essay (chapter 5) contributes to the literature by acknowledging this diversity while modeling credit risk for them, using a relatively large UK database, covering the analysis period between 2000 and 2009. My analysis partially employs the definition provided by the European Union to distinguish between ‘micro’, ‘small’, and ‘medium’ sized firms. I use both financial and non-financial information to predict firms’ failure hazard. I estimate separate hazard models for each sub-category of SMEs, and compare their performance with a SMEs hazard model including all the three sub-categories. I test my hypotheses using discrete-time duration-dependent hazard rate modelling techniques, which controls for both macro-economic conditions and survival time. My test results strongly highlight the differences in the credit risk attributes of ‘micro’ firms and SMEs, while it does not support the need to consider ‘small’ and ‘medium’ firms’ category separately while modelling credit risk for them, as almost the same sets of explanatory variables affect the failure hazard of SMEs, ‘small’ and ‘medium’ firms.

My fourth essay (chapter 6) considers all serious and neglected concerns while developing discrete and continuous time duration dependent hazard models for predicting failure of US SMEs. I compare theoretical and classification performance aspects of three popular hazard models, namely discrete hazard models with logit and clog-log links and the extended Cox model. I report that discrete hazard models are superior to extended Cox models in making default predictions. I also propose a default definition for SMEs which considers both legal bankruptcy laws and firms’ financial health while defining the default event. My empirical results show that my default definition performs significantly better than the default definitions which are only based on legal consequence or firms’ financial health in identifying distressed firms. In addition, my default definition also shows superior goodness of fit measures across all econometric specifications.

Business School, The University of Hull
Gregoriou, A. (Andros), 1977-; Healy, Jerome
Sponsor (Organisation)
University of Hull
Qualification level
Qualification name
2 MB
QR Code