CONCENTRATED OWNERSHIP AND PREDICTION OF FINANCIAL INSTITUTION FAILURES

Download This Article

Piruna Polsiri

https://doi.org/10.22495/cocv8i4p5

Abstract

In an emerging economy where ownership concentration is common and legal protection of outside investors is weak, financial and economic factors that are widely documented might not have been sufficient in constructing sound models to predict financial institution failures. Using the data of financial institutions listed in the Thai stock exchange during the 1997 East Asian financial crisis, this study showed that to develop sound prediction models that are robust across time to failure models, ownership variables should be incorporated in the models. Specifically, in the logit models that include both financial and ownership variables, 85.45%, 85.41%, and 91.49% of financial institutions were correctly classified in the models using the data of one, two, and three years prior to failure, respectively. It was also find that the presence of family as the largest shareholder increases the probability that a financial institution was closed. This evidence supports the expropriation effects of controlling families. Finally, the results suggested evidence of a “too-big-to-fail” policy in the closure procedures of Thai financial institutions during the East Asian financial crisis.

Keywords: Ownership Structure, Failure Prediction Model, Early Warning Systems, East Asian Crisis

How to cite this paper: Polsiri, P. (2011). Concentrated ownership and prediction of financial institution failures. Corporate Ownership & Control, 8(4), 84-95. https://doi.org/10.22495/cocv8i4p5