Altman Z-Score Model

The original model was created by Edward I. Altman in 1968. Edward I. Altman is the Max L. Heine Professor of Finance at the Stern School of Business, New York University. He is the Director of Research in Credit and Debt Markets at the NYU Salomon Centre for the Study of Financial Institutions. Prior to serving in his present position, Professor Altman chaired the Stern School’s MBA Program for 12 years. Dr. Altman was named to the Max L. Heine endowed professorship at Stern in 1988. Dr. Altman has an international reputation as an expert on corporate bankruptcy, high yield bonds, distressed debt and credit risk analysis. Professor Altman was one of the founders and an Executive Editor of the international publication, the Journal of Banking and Finance. Dr. Altman’s primary areas of research include bankruptcy analysis and prediction, credit and lending policies, risk management and regulation in banking, corporate finance and capital markets (Stern NYU)

The original 1968 model focused on predicting bankruptcy in public manufacturing companies. Altman’s reason for creating this model was to bridge the gap rather than sever the links between traditional ratio analysis and the more rigorous statistical techniques which had become popular among academics at that time. In this study, a set of financial and economic ratios will be investigated in a bankruptcy prediction context wherein a multiple discriminant analysis methodology is employed (Altman, September 1968). The original study/sample model focused on firms with an asset size range from $1million to $25million. Original data for this study was taken from the financial statements one year prior to bankruptcy. Altman selected five variables that do the best overall job together in the prediction of corporate bankruptcy. These five variables are as follows:

Source: Altman, E. (2018). Applications of Distress Prediction Models: What Have We Learned After 50 Years from the Z-Score Models?. International Journal of Financial Studies, 6(3), p.70.

Altman states in his 1968 paper (Altman, September 1968) that one of the limitations of his study was that all the firms examined were all publicly held manufacturing corporations for which comprehensive financial data was obtainable, including market price quotations. He stated that an area for future research would be to extend the analysis to relatively smaller asset-sized firms and unincorporated entities where the incidence of business failure is greater than with larger corporations (Altman, September 1968). Interpreting the Z-Score is as follows:

If above 3 then bankruptcy is not likely, if between 1.8 and 3 then bankruptcy can’t be predicted and if below 1.8 then bankruptcy is likely.

Source: Cengage.com. (2018).

Z-Score Model Advancements

Altman (1983) emphasized that the Z-Score Model is a publicly traded firm model and ad hoc adjustments are not scientifically valid. Altman’s created an updated model for a broad cross-section of industrial sector firms, as well as for firms outside the U.S., the original model was based upon the U.S. (Altman, et al 1995). Non-manufacturing firms, such as retailers and service firms, have very different asset and liabilities structures and income statement relationships with asset levels, e.g. the Sales/Total Assets ratio, which is considerably greater on average for retail companies than manufacturers. It is the 1.0 weighting for the variable (X5) in the original model that causes most retail companies to have a higher Z-Score than manufacturers. Therefore, Altman (1983) advocated a complete re-estimation of the model substituting the book value of equity for the market value in X4.

Altman extracted the following revised Z’-Score Model: Z’ = 0.717?X1 + 0.847?X2 + 3.107?X3 + 0.420?X4 + 0 .998?X5 where X1 = Working capital/Total assets X2 = Retained Earnings/Total assets X3 = Earnings before interest and taxes/Total assets X4 = Book value of equity/Book value of total liabilities X5 = Sales/Total assets Z’ = Overall Index. Due to the lack of a private firm data base Altman did not test the Z’-Score model on a secondary sample.

In case of the modification of the Z-Score Model, the most important changes were:

• The use of firms’ up-dated financial data in order to re estimate coefficients and

• the use of other estimation techniques in order to improve efficacy in comparison to the original model.

The use of Altman’s ratios combined with the MDA modeling techniques have improved the prediction ability. The presentation of new data greatly improved the model performance in the instance of both US and non-US firms.

The results for the updated model are slightly different compared to the original model. When looking at private manufacturing companies a Z-Score above 2.9 predicts bankruptcy is not likely, between 1.23 and 2.9 then bankruptcy can’t be predicted and below 1.23 then bankruptcy is likely. In terms of a private general company a Z-Score above 2.6 predicts bankruptcy is not likely, between 1.1 and 2.6 then bankruptcy can be predicted and below 1.1 predicts bankruptcy is likely. The area that can’t be predicted is known as the grey area.

The ZETA Credit Risk Model

In 1977, Altman, Haldeman and Narayanan (1977) made numerous improvements to the original Z-Score methodology with a second-generation model. The objective of this study was to structure, analyse and test an updated bankruptcy classification model which looks at recent developments with respect to business failures. Modifications have been incorporated into the updated model in the utilisation of discriminant statistical methods. The new model, known as the ZETA Credit Risk Model, looked at a sample of corporations comprising of manufacturers and retailers. The ZETA model for assessing bankruptcy risk of corporations demonstrates improved accuracy in classifying companies failure up to five years preceding to failure.

Studies assessing the effect of numerous elements involved with the application of discriminant analysis to financial problems. Include linear vs. quadratic analysis for the original and holdout samples, introduction of prior probabilities of group membership and costs of error estimates into the classification rule, and comparison of the model’s results with inexperienced bankruptcy classification strategies. The potential applications of the ZETA bankruptcy identification model are in line with previously developed models. These include credit worthiness analysis of firms for financial and non-financial institutions, identification of undesirable investment risk for portfolio managers and individual investors and to aid in more effective internal and external audits of firms with respect to going-concern considerations, among others.

ZETA aimed to remove previous statistical analysis techniques such as MDA, a statistical technique used to classify an observation into one of several prior groupings dependent upon the observations of individual characteristics. MDA was used, early on, to classify high and low price earnings ratio firms. Although it should not be the only means of credit evaluation, it should be paired with further evaluation. For example when looking at loan evaluation important variables such as the purpose of the loan, its maturity, the security of the loan and the deposit status of the applicant is left out. The MDA model focuses on the variables common to business loan evaluation.

When comparing the Altman’s original publication in 1968, studies have shown that there are five reason outlining why an updated Z-Score bankruptcy classification model can progress the statistics.

1. The change in the size, and the financial profile, of business failures. Since the original model the average size of bankrupt firms had dramatically risen with the subsequent greater visibility and greater concern from financial institutions, regulatory agencies and the general public. There is however an exception of Altman’s (1973) railroad study and the commercial bank findings, the majority of previous studies sampled small firms.

2. In order to be accurate, a model should be as relevant as possible to the population in which it will be serving. A new model should be as up to date as possible with respect to the temporal nature of the data.

3. Previous failure model focused either on the extensive classification of manufacturers or on specific industries. Upon applying the appropriate analytical adjustments, retailing companies could be analysed on an equal basis with manufacturers.

4. The data and footnotes to financial statements must be rigorously analysed to include the most recent changes in financial reporting standards and accepted accounting practices. In the studies, a change which was scheduled to be implemented in a very short time was applied. The purpose of these modifications was to make the model not only relevant to past failures, but to the data that will appear in the future.

5. To test and assess several of the then recent advances and still controversial aspects of discriminant analysis.

Altman (1993) reports a 94% correct classification rate for his ZETA model for over 150 US industrial bankruptcies over the 17-year period to 1991, with 20% of his firm population estimated as then having ZETA scores below his cut-off of zero.

Other Prediction Models

Another model that assess bankruptcy classification is the Hazard models, it uses both accounting and market data to assess the risk of a firm going bankrupt. The seminal accounting-based model of Altman (1968) is a widely used benchmark in bankruptcy prediction. Both models can differentiate between failed and non-failed firms. However, there is clear evidence that the z-score model has miscalculations, while hazard models have average default probabilities that are closer to observed default rates. Bauer and Agarwal states that “all models carry significant distress related information” and that the alternative approach of hazards model contains all of the information that is found in Altman’s and the contingent claims-based model. Although, the z-score contains more information about subsequent failure than the contingent claims-based approach. In economic terms, the two hazard models have shown far superior performances in comparison to the Z-Score models claims. In addition, we also find that of the two hazard models, Shumway (2001) model leads to a much higher market share and profit, higher credit quality of the loan portfolio as well as higher return on risk-weighted assets as compared to Campbell et al. (2008) model.

Remedies

The Z-Score was seen as a management tool for recovery. Rather than follow a strategy that had caused a the ratios and the Z-Score to decline, management must alter their outlook. The management must put in place a specifically formulated strategy to avert bankruptcy. Before implementing any decision, the impact on the model must first be reviewed in detail. Inherent in the Z-Score predictor was the message that underutilized assets could be a major contribution to the deterioration of any company’s financial condition. Getting rid of excess assets and utilizing the funds immediately will help the Z-Score increase. The predictor has become an active strategy in avoiding bankruptcy.

Critical Appraisal

After thorough research we have found the opinions of analysts/academics for Altman’s Z model as a bankruptcy prediction model. Bankruptcy models generally provide measures of financial distress and are regularly employed by scholars to examine the financial health of companies (Grice and Dugan, 2001). The technique for failure prediction was first published in 1968 by Altman, Edward, which remains very popular in the literature today (Almamy, Aston, ; Ngwa, 2016). Although many of those in the academic world have different opinions on whether the model is still relevant or if it is outdated. The most well-known arguments are outlined below.

Grice and Ingram (2001) have suggested that although it was developed in 1968, using a small sample of firms from the 1950s and 1960s, Altman’s Z-score model remains a commonly used tool for evaluating the financial health of companies. The advancement in the Z-Score model has helped it to be more effective than it was in 1968. Despite that the age of the model and other attributes, such as its small sample of manufacturing firms and the use of equal group sizes of bankrupt and non-bankrupt firms, it is likely that model is not as effective in classifying firms in more recent studies as it was when it was developed by Altman. Grice and Ingram indicated that those who employ Altman’s Z-score model should re-estimate the model’s coefficients rather than relying on those reported by Altman (1968). Because the coefficients are not stable, significantly better classification results were achieved when the model’s coefficients were re-estimated using 1985–1987 data than when the original coefficients were used. Grice and Ingram considers the Z-score model’s ability to evaluate financial distress conditions other than bankruptcy. If Altman’s model is better suited for predicting bankruptcy than for predicting other financial distress problems, it may be deemed inappropriate for some of the applications for which it is currently being used.

Financial analysts should use other credit risk measurement models like KMV model and to forecast value of firms. As Altman’s Z-score depends on data from financial statements, truthfulness of the model can be threatened by window dressing (Agarwal & Taffler, 2005).

Scott (1981) documented potential search bias in the variable selection technique used by Altman. The lack of a theory of bankruptcy invites the researcher to consider a multitude of variables and then to reduce the original set to the most accurate subset. The resulting subset of variables often proves ineffective when applied to a sample of firms or periods that were not used in order to develop the original models. Scott develops a coherent theory of bankruptcy and, in particular, shows how the empirically determined formulation of the Altman et al. (1977) ZETA model and its constituent variable set fits the postulated theory quite well.

Leon, L. and Miu, P. (2010) also suggest that one might dispute that banks adopting Altman’s approach should recalibrate the equation against the default experiences of their own credit portfolios to achieve the highest goodness-of-fit.

As Scott (1981) mentioned the Z-Score models are commonly censured for their perceived lack of theory. Gambling (1985: 420) strongly agrees with that: ‘… this rather interesting work (z-scores) … provides no theory to explain insolvency. This means it provides no pathology of organizational disease … Indeed, it is as if medical research came up with a conclusion that the cause of dying is death … This profile of ratios is the corporate equivalent of … ”We’d better send for his people, sister”, whether the symptoms arise from cancer of the liver or from gunshot wounds.’

Even though the Z-score model was developed to predict bankruptcy, this event is only one of several indicators (or consequences) of financial distress. As a result, there is uncertainty

whether Altman’s model is specifically useful for identifying firms that are likely to go bankrupt or whether it is a model that is more commonly used for identifying firms experiencing financial distress, nevertheless the model is commonly used for this purpose. While firms that experience financial distress are more likely to declare bankruptcy than other firms, most financially distressed firms do not declare bankruptcy. Gilbert et al. (1990) suggested that financial dimensions that distinguish bankrupt from healthy firms are different from those that separate bankrupt from distressed firms.

Altman original Z-score model has proven to be quite accurate over the last twenty decades and remains an established tool for assessing the health of companies. Not long ago, this evidence was confirmed by Al Zaabi and Obaid Saif (2011) who employed Emerging Markets (EM) Z-score model to predict failure and to measure the financial performance of major Islamic banks in the UAE context. Similar studies to this one have taken place all around the world, proving that the model is still very much used. Al Zaabi and Obaid Saif found that Z-score model is a suitable model to measure Islamic banks’ performance and the factors used in computing Z-score can be considered to give precious instrumental indicators (Almamy, Aston, & Ngwa, 2016). Their findings are consistent with Lugovskaya, 2010, Li et al., 2013, Bhandari and Iyer, 2013, Gutzeit and Yozzo, 2011a, Li and Rahgozar, 2012, Goswami et al., 2014; and Mizan and Hossain (2014).

Reisz and Perlich (2007), assessed the Z-Score as a better measure for short-term bankruptcy prediction than the market-based models and in Das, Hanouna and Sarin, (2009), it was shown to perform comparably to the Merton structural, market-based approach for CDS spread estimation. The question of whether market data is better than accounting data has been raised many times and will continued to be questioned further.

Before the beginning of quantitative analysis of company performance, one would examine early failure predictable models as in the 1930s in understanding the failure of companies.

Several empirical approaches have been employed in bankruptcy prediction models, since the original studies of financial models such as Fitzpatrick (1932), who compared 13 financial ratios of both failed and successful firms. Whereby 19 of each firm status were compared. His results show that, a majority of successful company cases illustrated favourable ratios while the failed firms experienced unfavourable ratios when compared with standard trend ratios. Fitzpatrick concluded that Net Worth to Debt and Net Profit to Net Worth. In addition, he reported that less significance should be placed on current and quick ratios for firms having long-term liabilities.

On the other hand, Merwin as cited in Bellovary et al. (2007) published a study on small manufacturers. His results demonstrate that, when comparing successful and failing firms, the failing firm displays several signs of weakness as early as four to five years before they fail. He identifies Net Working Capital to Total Assets. This is consistent with Altman (1968); Beaver (1966); Ohlson (1980) and Taffler (1982). Statistical methods have been applied in corporate prediction studies of both non-bankrupt and bankrupt firms. Predicting bankruptcy using multivariate discriminant analysis has been commonly employed, for instance Altman (1968); Taffler (1982); Ohlson (1980) with logit analysis. This methodology requires classification of paired sample (failed and non-failed companies) using financial ratios.

The wide level of research on this topic has shown how much two opinions can vary, this is clear when comparing the Z-Score model to the Hazard model.

While there is some evidence that Z?Score models of bankruptcy prediction have been outperformed by competing market?based or hazard models, in other studies, Z?Score models perform very well. (Altman & Iwanicz?Drozdowska , 2017 June). In comparison one opinion states that the hazard models are superior to the alternative models in bankruptcy prediction along all three performance dimensions. The Shumway (2001) model leads to economic outperformance once differential misclassification costs are taken into account.