Bankruptcy Modelling of Indian Public Sector Banks: Evidence from Neural Trace

Bankruptcy Modelling of Indian Public Sector Banks: Evidence from Neural Trace

Bikramaditya Ghosh
Copyright: © 2017 |Pages: 14
DOI: 10.4018/IJABE.2017040104
(Individual Articles)
No Current Special Offers


The paper estimates earnings per share (EPS) of top three Indian public sector banks on the basis of Ohlson O score, Zmijewski score and Graham Number, for a period of 12 years (2004-2015), with the help of the generalized method of moments (GMM), along with the use of an artificial neural network (ANN) algorithm. The time period has been carefully selected so that it could capture crash and consolidation phase, along with unprecedented bull rally too. It has been found that the fitment of ANN based model is accurate. Thus, using this model, their future EPS during distress could be predicted with a higher degree of precision. The authors believe this to illustrate a clear trace of the availability heuristic, timid choice, bold forecast and herding, as bulk deals by institutional investors decide the feat of a stock even on the futuristic possibility of bankruptcy.
Article Preview

1. Introduction

Bankruptcy is a puzzle which corporations try to avoid, analysts love to measure and find difficult to predict in general. The case of Indian Public Sector banks expands the puzzle due to a set of current developments. Firstly, the Indian central government has decided to infuse 22915 Crores in the economy for recapitalisation of 13 public sector banks. Among these beneficiaries, the State Bank of India (SBI) will receive a staggering amount of 7575 Crore, followed closely by Punjab National Bank (PNB) with 2816 Crore. The potential danger of such measures lays in aspects like the poor asset quality and weak capitalisation, proven in other historical occasions. Only in the latest Union budget, the central government has allocated 25000 Crore under this capital infusion scheme termed as “Indradhanush” (Bhowmik, 2015). Whenever the state-run banks are facing sharp rise in non-preforming assets and nearing possible bankruptcy, the centre comes out as a saviour to infuse cash. However, the justification remains vague and calls for further investigation. A noteworthy observation is that earnings per share of SBI have experienced an avalanche breakdown from March 2012 (Quarter ending) with 174.46 to March 2016 (Quarter ending) with 12.82. Whether technical or behavioural parameters are behind this fall is yet unclear. Both Bank of Baroda (BOB) and Punjab National Bank (PNB) faced similar fates, without any apparent plausible reasons. Thus, we postulate that an accurate prediction of EPS holds the key for understanding such situations. This study attempt to provide a tailored explanation by using various bankruptcy and valuation variables in a unique fashion.

The context is set by different approaches indicating a clear behavioural effect in the bankruptcy puzzle (Nickles, 2006). Numerous behavioural scientists observed that investors tend to simplify their decisions with sophisticated heuristics (Haug & Taleb, 2011), but they are still prone to errors. Between the idea that a cognitive bias is stubborn in nature (Burke, 2006) and the link of these behavioural traces to the dynamic gradient of financial literacy, there is an important gap to be explored. According to a recent comparative research (Cole, Sampson, & Zia, 2009) it has been found that Indian financial literacy is substantially lower than Indonesia. This is complemented by the fact that the penetration level of banks and insurance companies is substantially lower as well. As an operational answer, Otuteye et al. (2015) developed a heuristic named OS-heuristic, including factors like profitability, financial stability, susceptibility to bankruptcy and margin of safety, based on the rationale before. Moreover, Coelho et al. (2010) discusses about the psyche of retail investors and the identified similarities to gambling profiles. They considered the price impact of this trader behaviour and document the instant and constant churning. Such securities lead, on an average, to a negative realized absolute return of in and around -28% during the 12-month post-announcement period. This is an outcome which is found to be inconsistent with traditional asset pricing models. These innovative findings open up a new question on the presence of gambler’s fallacy, strengthening the links between behaviourally informed traces and bankruptcy.

Complete Article List

Search this Journal:
Volume 12: 1 Issue (2023)
Volume 11: 1 Issue (2022)
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing