Financial Distress in Indian Aviation Industry: Investigation Using Bankruptcy Prediction Models

The Indian aviation sector over the recent years has shown a significant growth prospects on various parameters like passenger traffic, freight traffic, aircraft movements and number of airports, among others. But ironically the financial performance of most of the air carrier individually is not at all impressive. Every five years one airline in India is being grounded, latest being Jet Airways which had to suspend its operations in April 2019 due to severe financial crunch. The primary aim of this paper is to assess the current financial health of different Indian airline companies. This study uses four different models i.e. Altman Modified Z” Score Model, Pilarski Model, Fuzzy Logic Model and Kroeze Model to test the existence of financial distress and simultaneously aims to assess the applicability of these models on the Indian aviation sector. The models have been applied on four airline companies. Models indicate the existence of severe financial distress in three out of four chosen airlines and also indicate their suitability to be applied to the sector. The study contributes to the existing literature on Indian aviation by attempting to indicate the suitability of studied models for indicating financial distress which can lead to potential bankruptcy.


Introduction
Air transportation has become a significant industry over time and it plays a key role in global tourism and supply chain functions. This industry generates substantial employment and contributes majorly to global economic growth (Fung, Law & Ng, 2006). However, it is also susceptible to several intrinsic and extraneous risks. These risks include economic boom and bust cycles, volatility in oil prices and exchange rates, infrastructure challenges, protectionism, wars and political upheavals, among others. This industry is also vulnerable to various other events such as weather conditions, terrorist attack as well as natural disasters 1 . All these issues lead to significant fluctuations in the profitability of air carrier or airline, the words used interchangeably in the paper.
In any free market economy, the competition will increase between the parties involved. The hallmark of this economic system is that some firms will inevitably fail. The more efficient firms will succeed, and the poorly managed will fail, allowing others to take their place. Thus, cost efficiency is of great importance (Lukic, 2014). According to economic theory, an existing airline will be able to succeed as long as it is operated and managed efficiently; if it fails, a more efficient airline will replace it. Hence, it is extremely important to have an early warning signal of financial distress in such a scenario. Warren Buffett once referred the aviation industry as the death trap for investors 2 .
The Indian aviation sector which is highly competitive in nature has shown considerable growth prospects especially in the domestic segment in the past few years. India is the third largest domestic civil aviation market in the world and is expected to move to become third largest air passenger market by 2024 3 . As the airline industry in India operates in a tight competition, the ability of different airline companies to increase prices is restricted to a great extent. More often prices are even slashed to attract the customers. But for maximizing profit, revenue maximization is also as necessary as minimizing cost (Jayaraman & Srinivasan, 2014). Since profitability is considered to be critical for growth and survival of any industry (Tyagi & Nauriyal, 2016), the companies belonging to the aviation sector have to pursue new strategies to maintain profit margins. In such a dynamic scenario, there is a very high propensity for financial distress for these companies which eventually lead to financial bankruptcies.
Several instances of financial distress have been observed among airline companies in India within the last decade. For example, Kingfisher airlines had to stop its operation in 2012 due to their inability to pay for its liabilities. Spice Jet cancelled more than 2000 flights at the end of 2014 due to huge accumulated losses. It was bailed out only by additional funding from promoters. Similarly, in 2016, Air Asia had to put its expansion plan on hold due to a severe cash crunch. Air Pegasus became bankrupt in 2017. In the same year, Air Carnival and Air Costa had to close down their operation. In recent times, in 2019, Jet Airways operations are completely suspended due to financial crunch. This sequence of financial distress and bankruptcies of several air carriers in India has led to an interest to study the probability of others becoming bankrupt in the near future. This prediction of bankruptcy is considered to be very significant in providing valuable insight to various stakeholders including customers.
According to Altman (1983), financial distress can be predicted before its occurrence (maybe even up to three years) using financial ratios with appropriate statistical analysis. If it can be accurately calculated to predict failures well in time, there is always a possibility to take corrective measures to reverse the phenomenon. In this backdrop, the primary aim of this paper is to investigate the current financial health of different Indian airline companies. An in-depth analysis of the same will be done using different models available including both generic and aviation industryspecific. The study also wishes to assess the applicability of these models in the Indian scenario. The study contributes to the existing literature on Indian aviation by attempting to indicate the suitability of studied models for indicating financial distress which can lead to potential bankruptcy.
The organization of the paper is as follows. A snapshot of the Indian aviation industry has been provided in Section 2. Next, in Section 3, a brief review of the literature is undertaken. The research methodology used in the paper is discussed in Section 4. The research findings and discussions are presented in Section 5, followed by certain concluding remarks in Section 6.

Indian Aviation Industry: A Snapshot
The aviation industry in India has observed a rising splurge post 2012. India became the fastest growing aviation market followed by China. According to the Director General of Civil Aviation (DGCA) 4 , there was a growth of 18.1 per cent in overall passenger traffic (domestic plus international) leading to a total of 147.1 million passengers in 2017-18 5 as compared to the previous year (Table 1). In the same period, the growth rate in domestic traffic was 18.9 per cent totalling to 123.3 million travellers and in international traffic, it was 14.4 per cent totalling to 23.8 million travellers. It can be noted that the number of passengers flown within and outside the country has increased by more than double in the past seven years. DGCA predicts that this number of air passengers would increase by almost 3.3 times in the next 20 years making it approximately 500 million passenger journeys per annum 6 . The year 2012-13 showed a negative growth both in terms of domestic as well as international passengers probably due to grounding of Kingfisher airlines during that time and relatively higher fare. The growth in air passenger traffic was only 3.7 per cent in 2019, showing a steep fall from the growth rate of 18-19 per cent of the previous year. This is due to various reasons, prima facie being grounding of Jet Airways and a subdued economy. However, there is a very significant growth potential of this sector in India. There are several factors that may contribute to this growth: (a) decrease in the cost of air travel; (b) growth in the population of the middle-income group; and (c) inadequate capacity of other travel alternatives like, railways 7 .
Another important aspect in the airline industry is the significant use of leased aircraft. Data from Center for Asia Pacific Aviation (CAPA) fleet database exhibits that out of 652 commercially operated aircraft (including that in storage) in India as on October 17, 2018, approximately 81 per cent are leased 8 (Figure 1). This is much higher as compared to 53 per cent of leased aircraft share globally and approximately 52 per cent in Asia. Indigo with 144 leased aircraft and Jet Airways with 106 leased aircraft together accounts for 47 per cent of the total leased aircraft in India followed by Air India (87 aircraft) and Spice Jet (58 aircraft).

Figure 1. Proportion of Leased Aircrafts from Different Air Carriers
Source: CAPA Database (2018)

Market Share of Different Airlines
In India, Low Cost Carriers 9 (LCC) viz. Indigo, Spice Jet and Go Air, have more than 60 per cent shares in domestic operations compared to Full Service Airlines 10 (FSA) viz. Air India, Jet Airways and Vistara.
Generally, tickets of LCC are priced lesser as compared to full-service airlines, however, the increasing competition within the industry has led to a significant reduction in gap in ticket prices in an endeavour to retain and increase the market share. The market share of different airlines until January 2019 is depicted in Figure  2. It is evident that Indigo leads the domestic industry with 42.5 per cent share followed by Spice Jet at 13.3 per cent. In fact, the market share of Indigo is more than the sum total of the market share of airlines in the next three positions.
It may be noted that Jet Airways has suspended its operations from April 2019 due to severe financial crunch 11 . Post grounding of Jet Airways, DGCA data of December 2019 12 showed an increase in market shares of different airlines i.e. Indigo (47.1 per cent), Spice Jet (14.9 per cent), Air India (12.7 per cent), Go Air (10.6 per cent), Air Asia (6.2 per cent) and Vistara (5.2 per cent).

Figure 2. Market Share of Different Airline Companies Operating within India
Source: DGCA (accessed on January 31, 2020)

Review of Literature
The success of any bankruptcy prediction model lies in its ability to forecast as to which company would possibly become bankrupt that too few years before they file for bankruptcy. The initial attempt in this direction was made by Beaver (1966). He used cash flows for prediction using a univariate model. He identified six financial ratios as the ratios having discriminating power out of 29 ratios which he examined between 79 bankrupt and 79 non-bankrupt firms. These six ratios analysed profitability (net income plus depreciation and amortization/total liabilities, net income/total assets), long term solvency (total debt/total asset) and short-term liquidity (net working capital/total assets, current assets/ current liabilities, and cash, short-term investments, accounts receivable/operating expenses excluding depreciation and amortization). As per Beaver's analysis, the profitability ratio, net income plus depreciation and amortization/total liabilities best predicted the bankruptcy potential. Altman (1968) developed the first bankruptcy model by using multiple ratios based on Multiple Discriminant Analysis (MDA) Model which more popularly is known as Z score model. It was generic in nature attempting to explain the potential bankruptcy amongst manufacturing companies that were publicly traded. The Z score model was later replaced with ZETA model by Altman et al. (1977). Again, in 1993, the five variable model of Altman Z-Score model was revised to a four variable model popularly known as Altman Z" score model. The model is believed to be more effective for non-manufacturing firms. Hanson (2003)  respectively. Altman and Gritta (1984) used Altman ZETA model for assessing US aviation sector.
Studies on bankruptcy gained significant attention after the works of Beaver (1966) and Altman (1968). Various attempts were made from time to time by various authors to validate, to improve the existing models and to come up with new models. Ohlson (1980) used logistic regression for predicting bankruptcy. Zavgreen (1985) developed a logit model that could predict bankruptcy up to five years. Mixed logit method was used by Jones and Hensher (2004). Gepp and Kumar (2008) combined discriminant and logit analysis to create a bankruptcy prediction model.
Two discriminant analyses were conducted in 1980s specifically for aviation sector by Gritta (1982) and Altman and Gritta (1984). Over a while it was felt that industry specific models can probably give superior or more accurate results as compared to generic models. AIRSCORE model was developed by Chow et al. (1991) using airline data. Pilarski and Dinh (1999) developed a model known as P-Score for air transportation. Davalos, Gritta and Chow (1999) constructed a Neural Network Model for major US airlines and Gritta et al. (2000) constructed a neural network model for predicting financial distress in small carriers. Though these models predicted bankruptcy quite accurately for samples until one year but no major breakthrough was observed in their prediction capability over MDA or logistic regression (Gudmundsson, 2002).
The first study on the application of prediction models in the case of businesses in countries outside the United States was done by Altman and Gritta (1984). This study covered more than ten countries but included only one type of statistical model. Various studies have been carried out in the past to find out the best prediction model. But, most of the studies done were conducted on developed economies.
In the Indian context, most of the researches have focussed on applying Z score and modified Z" score to select companies belonging to various sectors including aviation and other industries. Barki and Halageri, (2014) applied Altman Z score model on select companies in the Indian textile sector to analyse their financial strength.
Panigrahi (2019), tried to assess the existence of financial distress in the Indian pharmaceutical sector by applying Altman Z Score Model to select companies belonging to this sector. Kumar and Anand (2013) used both Altman Z score and modified Z" score to assess the financial health of Kingfisher Airlines using data from 2005-06 to 2011-12. The results confirmed the poor financial health of the airline. Another study on selected companies in the Indian aviation sector using original and other variants of Z score model also confirmed that Kingfisher airline was on the verge of bankruptcy (Vasantha et al., 2013). In another study calculating Altman Z score for bankruptcy prediction for the airline industry in India, it was concluded that the overall aviation sector in India is in financial distress excluding only Indigo which is in the safe zone (Kulkarni, 2018).
While existing researches have used Z score model and modified Z Score Model in context of Indian Aviation Sector (Pandey & Rathore, 2013;Safiuddin, 2017;Kulkarni, 2018;), none of the previous studies to the best of our understanding have considered Altman Modified Z Score, P-Score model, Fuzzy Logic and Krueze models all together for a combined analysis. The present study is an extension of previously carried researches in the aviation industry. The main purpose of this investigation is to assess the suitability of major bankruptcy prediction models by applying them to key airline companies in India. The objectives of this study are: (a) to analyse the financial situation of Indian airlines companies using various models (generic and industry specific); (b) to examine the existence of financial distress in the Indian airline industry; and, (c) to assess the applicability of different models in the Indian scenario.

Research Methodology
This study included four leading airlines of India viz. Indigo, Jet Airways, Air India and Spice Jet. Purposive sampling is used based on the market share and availability of data for the air carriers. The audited financial statements provided in the annual reports of these companies are the main source of financial data. The financial ratios which serve as input for various bankruptcy models are calculated from these statements. The data considered for this study is from 2015 to 2018. Only four years of data are considered as existing literature states that financial ratios can predict bankruptcy quite accurately within this time period (Kroeze, 2005).
In this paper four different models viz. Altman Z" Score Model, P-Score Model, Fuzzy Logic Model and Kroeze Model are used to study possibilities of the bankruptcy of Indian airlines companies using financial data. Of these four models, Z" score is a generic model and the remaining three models are specifically designed using airline specific data. The selected models were chosen on the basis of extensive literature review and their predictive abilities as specified in the literature. Also, another criterion was that model should be non-proprietary so that intercept terms are publicly available. A brief about these four models are as follows:

Altman Model (Z" Score)
The original bankruptcy forecasting model of Altman popularly known as Z score has been successfully used by Gritta (1982) to predict the bankruptcies of both Braniff and Continental airlines, several years in advance before their actual filing of bankruptcies. The model is represented as: where, X1 = liquidity ratio calculated as net working capital to total assets; X2 = profitability ratio measured as retained earnings to total assets; X3 = profitability ratio measured as operating profit to total assets; X4 = leverage ratio measured as the market value of equity to book value of debt; X5 = turnover ratio measured as operating revenues to total assets. Here, X1 to X5 are different ratios measuring different aspects of financial strength. This model was developed by Altman (1968) using multiple discriminant regression.
However, because of the increased use of operating lease in aviation sector, the reliability of this model has reduced over time (Gritta et al., 1995). There is a viewpoint that because of the significant use of operating lease, X5 ratio can lead to distorted results (Gritta et al., 1995). Altman (1983) himself suggested the usage of a modified model the Z" for service firms.
Based on the value of Z", the companies are categorized into three groups (Table 2). Increased Probability that insolvency and bankruptcy will not happen in the near future.
The modified Z" model has been used in this paper because of significant use of lease in the Indian aviation sector. One of the limitations of Altman Z" score model is that it is generic in nature and not specifically designed for the aviation industry.

The Pilarski or P-Score Model
This is a logit model estimating the probability of bankruptcy developed by Pilarski and Dinh (1999). This model was primarily used for assessing the financial condition of major US air carriers and is believed to be giving superior results as compared to other models (Goodfriend et al., 2005). Popularly called as P-Score Model, this model gives P value which is the Probability of Bankruptcy. Equation 3 and 4 shows twostep calculation of Probability of Bankruptcy P.
Here 'e' is a mathematical constant and its value is equal to approximately 2.718.
Higher the P-value more is the probability of bankruptcy and vice-versa. Combining equation (3) and equation (4), it is clear that the P-Score is a measure of five different characteristics of the air carriers rather than just profitability. The P-Score model is considered to be appropriate for this study because it is specific to air carriers with a prediction rate of approx. 85.1 per cent. This model has also been used by the US Department of Transportation for tracking the financial strength of their air carriers (Gritta et al., 2006).

Fuzzy Logic Model
Another approach used by several researchers for predicting air carrier insolvency is fuzzy logic. Silva et al. (2005) applied a multivariate technique known as Hybrid Financial Statement Analysis (HFSAT) and tested the financial situation of a few American and Brazilian airlines. HFSAT is the combination of a discriminant analysis multi-variable model and the application of fuzzy logic to the firm's financial data. Following is the model used:
This model uses only three variables and applies Multiple Discriminant Analysis. Positive value of Ya indicates a non-bankruptcy and the negative value of Ya indicates a situation of bankruptcy. Variable X2 defined as retained earnings/total assets is considered to be the most important predictor of bankruptcy as negative retained earnings are a sign of financial distress. Kroeze used this model to successfully predict the situation of Air Canada, US Airways and Hawaiian four years before their actual bankruptcies and that of Trans World Airlines (TWA) and American Trans Air (ATA), three and two years respectively before their actual filing of bankruptcies.

Research Findings and Discussion
This section is being divided into two parts for bankruptcy prediction and subsequent discussions. In the first part, all the four models discussed above have been analysed for four selected air carriers. The second part attempts to relate the results derived from the application of different models on select air carriers with the prevailing situation to assess the applicability of these models in Indian context.  The values of equity/debt ratio (X4) are worth mentioning in this aspect. This ratio shows the relative proportion of debt and equity in the total capital structure. Higher is the proportion of debt, riskier is the firm. Except for Indigo, almost all the companies are showing negative equity/debt ratio and this is because of the 13 Negative equity is a situation where liabilities exceed assets. negative equity. This negative equity is due to the accumulated losses over the years and then excessive debt to cover these losses. On the contrary, in the case of Indigo, the Z" score has been positive in all four years, although with varying degrees. This indicates that Indigo is clearly in a low degree of financial stress.

The Pilarski or P-Score Model
As already stated, P is the probability of going bankrupt. Hence, greater the P-value, the higher is the financial distress and more is the probability bankrupt.
It is clearly evident from Table 4   This is a very significant observation as this particular carrier has been grounded since mid-April 2019 because of financial issues. Government sponsored Air India is also showing more than 70 per cent probability of becoming bankrupt. This airline is surviving only due to financial assistance from the government.

Fuzzy Logic Model
From the Fuzzy Logic Z Scores portrayed in Table 5, it is noticeable that none of the airlines are in a healthy zone with Z < 1.82. Jet Airways and Air India are in the zone of severe financial distress as good as being insolvent with Z score greater than 2.73 for all the four years since 2015. Very high total debt coupled with negative values of retained earnings and equity is responsible for this plight of both these air carriers.
Indigo is the only airline that has gradually moved into a low risk category as per the classification pattern of this model. This is because of the increase in its X1 ratio over the period of study. Spice Jet is another airline that has moved into medium risk category shown by consistent improvement in its Z score. However, the negative value of equity and retained earnings remain a significant cause of concern for Spice Jet.    Table 7 shows the consolidated result of all the four models used in this paper. Indigo is found to be the only stable and consistent air carrier in the Indian aviation industry among the four major airlines. The data presented in the balance sheet also confirms the same.  16 . It has revived splendidly over the last few years from almost a near-death like situation in 2014, when the airline was almost about to shut down its operations due to severe cash crunch 17 . The probable reasons for this positive development may be tightening of cost and expansion to new routes which generated better revenues. However, Spice Jet is yet to be considered out of danger as per the scores of various models in Table 7.

Comparison of All Four Models
The operations of Jet Airways are being suspended since the middle of April 2019 due to mounting financial burdens and this poor situation of Jet airways is illustrated appropriately by all the models used in this paper. This company is struggling with a debt of approx. INR 11,261 crore 18 . The airline has not been able to pay to any of the stakeholders including, banks, creditors, vendors, lessors and most regrettably, salary to its employees.  To sum up, the result of all the models depicts a relatively healthy low risk situation for Indigo and gradually improving condition for Spice Jet. In case of Jet Airways and Air India, all the models captured their severely bad financial position over the period of study.

Concluding Remarks
The primary objective of this paper was to analyse the financial situation of four major Indian airline companies. Except for Indigo, all the other chosen airlines (viz. Spice Jet, Jet Airways and Air India) have been under financial distress as specified by all the four chosen models i.e. Altman Z" Score, P-Score, Fuzzy Logic and Kroeze Model. For Indigo, Altman Z" Score, P-Score and Kroeze Model have shown healthy stable financial position. Fuzzy logic has put Indigo also in medium to low risk category based on its financial situation.
The models have highlighted the existence of severe financial distress in the Indian aviation sector. The results are in line with the previous studies conducted (Tyagi and Dutta, 2017;Kulkarni, 2018) which have also indicated the existence of financial distress in the overall aviation sector in India. The study also aimed to assess the suitability of different models in the Indian context. These models despite drawing data from financial statements for calculating various ratios give different weightage to various ratios and employ different statistical techniques.
The scores from various models have been good indicators and successfully captured the prevailing financial conditions of different air carriers in India. All the four models have depicted the existence of extreme financial stress for Jet Airways leading to a very high probability of bankruptcy well in advance right from the year 2015. Jet Airways operations are suspended since April 2019. From the study it can be safely concluded that all of these models can be applied in Indian Aviation Sector for predicting the possibility of bankruptcy.
However, it should be noted that the existence of financial distress does not always lead to a situation of bankruptcy. It only indicates a likelihood of future failure which can also be reversed by taking appropriate corrective measures and strategies either by the company itself or by the government. This particular analysis of the Indian airline companies is also limited to the extent of availability of accurate financial data. The result of this study shows the potential for further research in this sector. Further studies can focus on identifying various variables leading to the financial distress and in identifying variables critical in improving the financial performance of airlines on a case to case basis. This kind of analysis for any industry can be useful for different stakeholders like stockholders, bankers, customers, lessors, and other creditors, among others. The management of different companies and the government can also take appropriate policy measures well in advance to improve the financial performance on a case to case basis.