Accuracy comparison of accounting-based bankruptcy prediction models of Springate (1978), Ohlson (1980) and Altman (2000) to US manufacturing companies 1990-2018

Abstract

The importance of corporate bankruptcy has risen to ever more prominence since the recent financial crisis. The field of bankruptcy prediction has become even more popular among academics and is considered to be an industry itself. It is estimated that at least 40,000 people are dealing with corporate distress. The estimation of bankruptcy is intriguing but still a bankruptcy prediction model with high accuracy rate remains a challenge since the models are based on certain industries and tend to be sample specific.

Recently, market-based bankruptcy prediction models have gained popularity among researchers in the field, however, there is no supportive evidence of their superiority compared to accounting-based models that aim to predict financial distress using financial accounting data. Although new complex models e.g. neural network techniques have emerged to the literature, the accounting-based techniques are still most popular in the research.

This paper attempts to add finding to the literature by comparing three accounting-based bankruptcy prediction models of Altman (2000), Ohlson (1980) and Springate (1978) in order to present comprehensive computational comparison of methodologies to fulfil the strategic information needs of investors and other stakeholders. The aim is to statistically show that the models differ in accuracy rates in US manufacturing industry. The sample of the study consists of thirty-three bankrupt and 414 non-bankrupt United States manufacturing companies that were listed in either NYSE, Nasdaq or American Stock Exchange in 1990-2018.

The result of one, two and three years prior to bankruptcy indicates that the three accounting-based bankruptcy prediction models of Altman (2000), Ohlson (1980) and Springate (1978) have different predicting power to bankruptcy in US manufacturing companies. Furthermore, the results show that the Altman´s (2000) model performs better than the models of Ohlson (1980) and Springate (1978) when predicting bankruptcy in US manufacturing companies but the differences in the accuracy rates are all not statistically significant.