ANALYSIS MODELS OF THE BANKRUPTCY RISK IN ROMANIA S ENERGY SECTOR

Transcription

1 ANALYSIS MODELS OF THE BANKRUPTCY RISK IN ROMANIA S ENERGY SECTOR MIRON VASILE CRISTIAN IOACHIM PHD. STUDENT 1 DECEMBRIE 1918 UNIVERSITY OF ALBA IULIA cristi_mir89@yahoo.com VOICULESCU ALINA PHD. STUDENT 1 DECEMBRIE 1918 UNIVERSITY OF ALBA IULIA POPA (JELER) IOANA PHD. STUDENT 1 DECEMBRIE 1918 UNIVERSITY OF ALBA IULIA Abstract The risk, as a concept found in the economic sphere of business represents an analysis area often approached by researchers from the finance and accounting field. Although it is often seen, along with cost-effectiveness and value, as a fundamental element of finance (Stancu, I., 2007), the risk has often facets that make it useful also in analyzing other sides of the economic sphere of business, such as financial position and economic performance of it. From these meanings, we believe that the most suitable for this purpose is the one through which it is analyzed the ability of an entity to avoid bankruptcy. The present study has as main objectives the presentation of bankruptcy risk of an entity from a theoretical point of view and the analysis (from an empirical and comparative point of view) through the scoring method of the implementation of various models for analyzing the risk of bankruptcy (Altman, Conan-holder, Taffler, Robertson) in Romania s energy sector, in order to issue an opinion regarding the optimal method for analyzing the bankruptcy risk in the energy sector. The results show that there are significant differences regarding the analysis of the bankruptcy risk through the appliction of different models, proposing the Conan-Holder model as the most appropriate for this sector. Key words: bankruptcy risk, risk, Altman, Conan-Holder, Taffler, Robertson. JEL classification: M40, M41 1. INTRODUCTION In our opinion, the financial position of an economic entity is nothing but a concept which is characterized by financial balance and risk. Among the forms that risk can take in this context, we believe that the risk of bankruptcy is the form that characterizes best the way of expressing an economic entity's financial position. Economic and financial performance is a concept whose expression depends on many factors, including the risks assumed by the entity. We also consider that there are forms of risk, such as the risk of bankruptcy, which express in inverse ratio even the performance of an entity. In other words, we can assume that an entity that is underperforming has a higher bankruptcy risk, and vice versa, an entity that is registering a high performance has a low bankruptcy risk. Considering these aspects, we can assume that the risk of bankruptcy can even be regarded as a mathematical measurement indicator (inverse ratio) of the continuity of an entity. This approach can be seen as a response to the issues raised by Miron, V., C., I., [9], who draws attention to the lack of some mathematical indicators for assessing the compliance with the continuity principle, principle existing only declaratively in the annual financial statements. 2. LITERATURE REVIEW The financial difficulties that the economic entities are facing are realities encountered in any economy, regardless of its development degree. To have an image of the future evolution of an entity, the analysis of the bankruptcy risk is one of the most common methods. 249

2 Bankruptcy risk analysis is just an empirical approach of the concept of bankruptcy. In a broader context, the analysis of the concept of bankruptcy must be achieved even through correlations between the number of bankruptcies and business cycles, demographic movements of economic entities, efficiency of the judiciary system as regards the conduct and duration of the bankruptcy proceedings, the behavior of companies which have gone bankrupt, etc. [5]. The importance of the analysis of bankruptcy risk is a high one, especially for stakeholders who have claims to a particular entity, such as banking societies, various creditors, investors, employees, suppliers, etc. We subscribe to the idea presented by Petrisor, M., B., and Lupu, D., [13] according to which the losses that would be suffered by these groups of stakeholders in the case of a bankruptcy or insolvency would be much more higher than the direct costs with the insolvency or bankruptcy (expenses with various charges, accounting, audit, legal assistance, etc.). Solvency, as a concept of assesing financial balance widely known, that describes the state of a company able to meet payment obligations when due, is an economic entity's ability to overcome the risk of bankruptcy [12]. The specialized literature devoted four methods used to analyze the risk of bankruptcy [1]: Comparison method or analysis of indicators in dynamic; are taken into account indicators presented in the annual financial statements which are compared from year to year, or comparing indicators for companies in difficulty; Methods based on rates: they have the advantage of allowing extensive comparative analysis in time and space; Financial flows method, highlighting the impact on cash, aspect which has a decisive influence in the case of bankruptcies and insolvencies; Scoreing method: it appeared in response to the classical methods of analysis and it has the advantage that it takes into consideration a fairly large range of financial and nonfinancial rates. In our opinion, the scoring method (regardless of the chosen model) brings the highest contribution in the assessment of the bankruptcy risk, because it is based on a number of indicators that are able to cover all the sensitive areas of an entity: working capital, assets, profits, liabilities, turnover, etc. The principle of applying the scoring method is based on a mathematical function, designed so it assess with approximation if the entity is heading towards bankruptcy, or it will register high economic performance in the following periods that come after the analyzed ones [15]. However, there are bibliographical sources [7] that consider that these analyzes can be useful even for longer periods of time. In our opinion this is possible only where it is considered a dynamic analysis of the indicators which express the mathematical function, for a period of minimum 5 years. Among the established models of the scoring method we can recalled models like Altman, Conan-Holder, Taffler, Robertson, The model of the Central Bank of France, IN99, IN05 etc. Recognized companies that are dealing with risk analysis often use more complex models of risk analysis and rating, such as Logit model (used by JP Morgan rating agency) or KMW model (used by Moody's). Theory and practice from the field of risk analysis often recognize the supremacy of Altman model, but there are specialized studies showing that in some circumstances, this model can be surpassed by models like Logit [6]. Due to the continuous changes undergone by the economic environment in which companies activate, over time these patterns of risk analysis have been continuously updated, and the reality in which we live makes us believe that the update trend will increase in the future. These models are internationally recognized, but in the majority of the economies different models of domestic analysis of bankruptcy risk have been created, according to the vision of various researchers in the field, but there are also own models used primarily by banks in accordance with their own procedures. Moreover, the study drawn up by Machek, O., [7] on a sample of companies in the Czech Republic, shows that models geared largely towards the creditors have the highest rates of correct predictability regarding the risk of bankruptcy (the only model with rates of over 50% accurate predictability is IN05 model, which, according to the source, is geared towards creditors and shareholders). Regarding the development of models applicable only at the country level, we believe that this is somehow not applicable, given the trend towards globalization that we are witnessing. Cîrciumaru, D. [4] states that in Romania, there is often difficult to develop such models due to different forms of manifestation of bankruptcy in Romania compared to other states. The source says that in Romania there is a large number of companies that are in financial difficulty but against which it has not yet been initiated the bankruptcy procedure, but also companies that present financial reports containing information possibly manipulated. However, we believe that such claims, without solid evidence, are not worthy of consideration. In our opinion, the difficulty of developing such models for the reasons mentioned above is non-existent. However, we subribe to the idea presented by Mosionek Schweda, M., [11] which states that the results of a single model is not sufficient to characterize the state of a company. In our opinion, even within the same methods of analyzing bankruptcy risk, it is required that the analysis is made through several specific models, and the obtained results should be regarded as complementary, so that the image of the entity is as comprehensive as possible. The empirical study was conducted by analyzing data from the annual financial statements issued by entities that activate in the energy sector and are listed on the Bucharest Stock Exchange, namely Romgaz (SNG), Petrom (SNP), Electrica (EL), Transgaz (TGN), Transelectrica (TEL), Nuclearelectrica (SNN), Conpet (COTE), Rompetrol Rafinare 250

3 (RRC), Oil Terminal (OIL), Petrolexportimport (PEI) and Dafora (DAFR). The observation was made for the individual financial statements (except Electrica, for which we have used data from the consolidated financial statements) of the mentioned companies for the years 2012, 2013, RESULTS AND DISCUSSION The expression of the bankruptcy risk has been achieved through the application of recognized method of scoring such as Altman, Conan-Holder, Taffler and Robertson. 1. The Altman model This model is one of the most used models for analyzing bankruptcy risk, often being considered a benchmark in this area of analysis. Given the fact that the model includes an indicator obtained based on market capitalization, this model is only applicable for listed companies. However, over time, this model has often been adapted even for non-listed entities in order to meet their needs for information of interested parties (banks, suppliers, etc.). Developed in 1968 as a score (function) in which are weighted 5 indicators, the model was continuously improved, having today the following form (1): =3,3 ' +0, ,6 - +1,2 0 +1,4 2 1 The variables considered for this model have the following form: ' = / + = / - = / 0 = / 2 = / We believe that a better representation of the model in relation to its purpose (to provide a picture of the risk of bankruptcy), required some adjustments regarding the calculation of variables, as follows: Instead of EBIT, we will use the Profit before tax (gross profit), because in our opinion the costs and revenues of interest (which are not taken into account in calculating EBIT, but included in the calculation of profit before tax) are essential elements that can influence the bankruptcy of an entity; The market capitalization of a company is often dependent on its external elements (national and international economic context, the vision and subjectivism of investors who choose to buy or sell shares based on their perception, etc.), which is why we use the equity instead of the market capitalization. We believe that the equity gives a more accurate value of a company's than the market capitalization; Interpretation of results involves framing the analyzed companies in one of the three risk categories proposed in the model, as follows: Z <1.8 - imminent risk; 1.81 <Z < gray area; Z> safe area. The results of this model for the sample under consideration are shown in Fig. 1: 251

4 12,00 10,00 8,00 6,00 4,00 2,00 0,00-2,00 1,95 2,67 1,73 7,59 6,61 7,28 2,74 3,33 3,00 Altman Model ,91 1,55 1,35 2,35 3,47 3,47 1,18-0,66 0,92 6,65 6,08 9,46 Fig. 1: Applying the Altman model on the analyzed sample Source: author's own processings 4,74 8,05 10,42 3,27 2,09 1,82 2,84 2,11 2,48 0,46 3,81 5,87 2,39 2,62 2,23 SNP SNG TGN TEL RRC DAFR COTE OIL EL SNN PEI Media The data presented in the Figure no. 1 one can notice that we are dealing with entities such as DAFR that frames in imminent risk zone in all 3 years analyzed. In the same risk category were also classified TEL (for 2012 and 2013) and SNP (2012), but for these companies the situation has been remedied. The fact that in case of DAFR was declared insolvency procedures in 2015, confirms that Altman is a useful model in the energy sector. Also, we consider a worrying situation for PEI, which has a continuous decrease during the analyzed period, with categorizing imminent risk of bankruptcy in On the other hand, the model shows no difficult situation for RRC. However, the Rompetrol Group (in whose composition enter and RRC) asked its insolvency in 2015, but the request was rejected. The mean recorded in the analyzed sample shows that we are dealing with a risk of bankruptcy situated in the gray area. This is due to a very good level of the score recorded by companies such as COTE, OIL, or SNG, but also to a high wheigjt that have companies like SNP, SNG and TGN in the calculation of averages. 2. Conan-Holder model As form, this model is somewhat similar to Altman model, but the difference is that it takes into account variables that characterize much internal activity of the entity, without dealing with items such as the exchange value of the shares. This makes the model applicable both for listed companies and for unlisted. However, given the fact that external influences on the entity are excluded, insisting on characterizing relations like liquidity / chargeability, we believe that this model is applicable to companies that do not exceed 1,000 employees. The form of Conan-Holder model is the following (2): =16 ' (2) The variables considered for this model have the following form [8]: ' =( + h )/ + = / - = / 0 = / 2 = / We believe that a better representation of the model in relation to its purpose (to provide a view of the risk of bankruptcy), an adjustment is necessary as regards the calculation of the last variable; Thus, we consider appropriate to replace gross operating surplus in the last variable operating result. Our option is supported by the fact that elements of significant operating activity, such as expenses and income with adjustments for impairment or provisions related to operating activities and other expenses and income related to operating activities are excluded from the calculation of gross operating surplus, they being only included in the calculation of operating result. Interpretation of results obtained by this model score is presented in Table no. 1: Table no. 1 Interpretation of results for Conan-Holder model Score Bankruptcy Entity s situation 252

5 Negative >80% 0-1,5 75%-80% Unfavorable situation 1,5-4 70%-75% 4-8,5 50%-70% 8,5-9 35% Uncertain situation % % Favorable situation % Z>16 <15% Excellent situation Source: accesed at ,00 100,00 0,00-100,00-200,00-300,00-400,00-500,00-600,00-700,00 29,77 38,15 30,41 The results of this model for the sample under consideration are shown in Fig. 2: 137,26 101,43 126,32 Modelul Conan-Holder ,17 57,38 49,10 43,65 39,34 32,85 0,80 9,59 2,23-1,00-25,35 17,09 111,63 94,37 48,80 29,28 30,45 23,63 55,40 38,99 36,35 62,73 21,07 6,03-611,54 37,82 56,05 Fig. 2: Applying the Conan-Holder model on the analyzed sample Source: author's own processings 35,51 34,90 28,72 SNP SNG TGN TEL RRC DAFR COTE OIL EL SNN PEI Media Analysis of the Figure no. 2 highlights two issues. On the one hand we can see that there are similarities with the conclusions reached by the application of Altman model, similarities as: Situations where healthy for SNG, TGN, COTE, OIL, etc.; DAFR particularly worrying and alarming even for PEI (high risk of bankruptcy in analysis of 2014 after previous years the company has shown a great situation). On the other hand we note that there are also different conclusions from those reached by applying Altman model, such as: Entities such as SNP, SNN or TEL, and the annual average are seen as totally free of bankruptcy risk, but if they were analyzed by Altman model they were on gray area; RRC is classified as high-risk ok bankruptcy, while, in the Altman model was framed in gray zone, or area without risk of bankruptcy. (3): 3. Taffler model The algorithm of application of this model is based on a 4 variables that are weighted according to the tool below =0,53 ' +0,13 + 0,18-0,16 0 (3) The variables considered for this model have the following form [2]: ' = / + = / - = / 0 = / 253

6 Interpretation of results involves framing the analyzed companies in one of the two risk categories proposed in the model, as follows: Z <0.2 - high risk of bankruptcy; 0.2 <Z - low risk of bankruptcy. The results of this model for the sample under consideration are shown in Fig. 3: 2,50 2,00 1,50 1,00 0,50 0,00-0,50 1,89 1,74 0,18 0,60 0,34 1,30 1,08 0,74 0,59 0,16 0,09-0,04 Taffler Model ,56-0,34-0,35-0,23-0,47-0,16 0,97 0,73 1,25 0,06 0,07 0,04 0,36 0,19 0,11 0,23 0,15 0,03 0,02-0,46-0,69 0,15 0,26 0,16 SNP SNG TGN TEL RRC DAFR COTE OIL EL SNN PEI Media ,00 Fig. 3: Applying the Taffler model on the analyzed sample Source: author's own processings Analysis of the Figure no. 3 confirms the findings from Altman model for companies like DAFR, SNG, COTE, etc. However, as in the case of the Conan-Holder, we remark the unfavorable situation existed for RRC, but there are also other companies that are classified as high risk of bankruptcy: TEL, SNP (2014), SNN (2012 and 2013), OIL and the annual average (for 2012 and 2014). Very interesting towards the first two approaches is the case of PEI. Although all models discussed in 2014 indicated a high risk of bankruptcy, Taffler model shows the existing situation as in the two preceding years. Moreover, if in the case of the Altman and Conan-Holder could see a worsening situation from year to year, the Taffler model show the opposite: the gradual improvement of the situation of the company (even if it fails to leave the area imminent risk). However, we believe that this model of analysis of bankruptcy risk labels too easy companies in the high risk of bankruptcy (6 out of 11 companies considered with high bankruptcy risk). In our opinion, the model could be improved, not necessarily by reconsidering or weighting the variables which are taken into account when calculating the score, but by defining intervals in which a company may be assigned a risk category or another. 4. Robertson model Professor Robertson identified elements that produce changes in the company's financial health such as market stability, low profits, decrease working capital, loan growth, etc., proposing a function of 5 variables in the following form: (4) =3 ' ,6-0,3 0 +0,6 2 (4) The variables considered for this model have the following form: ' =( )/ + = / - =( )/ 0 =( )/ 2 =( )/ In our study, we consider more appropriate using the indicator Cash and cash equivalents at the numerator of the last resort. Interpretation of the results of this function is slightly different from the models mentioned above: for this model is made an analysis of the dynamic evolution function, from one period to another. If there is depreciation by 40% or more, we believe that the entity faces a worrying risk of bankruptcy, and if over two consecutive periods is found is lower by 40% or more, we believe that the entity presents a high risk of bankruptcy. 254

7 The results of this model for the sample under consideration are shown in Fig. 4: Robertson Model ,00 0,00-50,00-100,00-4,48-2,88-2,32-8,32-8,59-9,19 7,83-2,22-4,75-2,20-2,86-2,09 1,50 1,41 1,49-0,69-4,28-3,20-7,65-6,57 13,11-11,90-13,14-11,78-0,01-2,09-1,79-13,59-14,92-15,15-230,62 2,20 2,53-2,28-2,38-2,04 SNP SNG TGN TEL RRC DAFR COTE OIL EL SNN PEI Media ,00-200,00-250,00 Fig. 4: Applying the Robertson model on the analyzed sample Source: author's own processings Analysis of bankruptcy risk by this model, especially in this way (dynamic), must be done, in our view, complementary to one of the previous models as the independent application of this model does not provide enough information (eg. DAFR ). As shown in Figure no. 4 entities that have a high risk of bankruptcy are SNP, COTE or PEI (alarming situation). Thus, by combining information obtained by this model with those obtained by one of the previous models, possibly also with those obtained from the analysis of financial balance or financial flows (or other issues), we consider that informational power on the risk of bankruptcy (or even on the global risk of the entity) is much higher. CONCLUSIONS Scoring method has proved to be a useful tool in assessing the risk of bankruptcy in the energy sector. This approach, however, should not ignore the analysis carried out by other methods, such as analysis of financial flows and financial balance (and possibly other matters), all of these contribute to determining an entity's overall risk. Given that bankruptcies will continue to exist, this being the starting point of analysis related to the risk of bankruptcy, we believe that such studies are always topical. We appreciate assumption confirmed by Posada, M., G., and Mora-Sanguinetti, J., S., [14] in their study that the introduction of high costs for bankruptcy procedures would lead the entities to mortgage loans thus reducing the number of failures, but we believe that this approach is not in all cases the best solution. Being caused primarily by too high level of debt, bankruptcy cannot always be avoided by additional debt. In our opinion, it must be prevented through the regular bankruptcy risk analysis as those presented in this study (or other analysis), and take appropriate action at the right time. A personal comparison between models of analysis of bankruptcy risk analyzed in this study leads us to conclude that, in our opinion, Altman model is most useful if desired both the analysis of bankruptcy risk and the analysis inversely related to concepts involving no risk of bankruptcy, such as going venture. Although Altman model for analyzing the risk of bankruptcy is highly standardized and rigid, literature presents studies that show how easy it the classification can be changed in a risk category or another, depending on how they may see the concept of working capital [10]. However, after the empirical analysis, we conclude that Conan-Holder model is the most useful model for analyzing the risk of bankruptcy in the energy sector, especially due to the results obtained from the analysis of DAFR and RRC. We believe that Taffler analysis model of bankruptcy risk labels too easy companies in the high risk of bankruptcy (6 out of 11 companies considered with high bankruptcy risk in 2014). In our opinion, the model could be improved, not 255

8 necessarily by reconsidering or weighting variables taken into account when calculating the score, but by defining intervals in which a company may be assigned to a risk category or another. Regarding the Robertson model, our conclusion is that it is rather useful when it is applied together with other analytical models to confirm or refute the findings obtained by other models. Moreover, in a study published by Cîrciumaru, D., et. al., [4] it is shown that analysis of bankruptcy risk by scoring method, applying different models may lead to different results, proposing to correlate these results with other financial aspects, such as turnover, rates of return, rate debt, financial balance, etc. The overall conclusion reached by this study is that regardless of the model under consideration, the Romanian energy sector (in the sample analyzed, but the findings can be extended to the entire sector) is generally characterized by a medium risk of bankruptcy, meeting both companies with a sensitive situation (some which have already entered into insolvency proceedings, others on which there is a high probability that they will get in this situation), and solid companies, over which there is no risk of bankruptcy in the near future. REFERENCES [1] Balcaen, S., Ooghe, H., (2004), 35 Years of studys on Business Failure: An Overview of the Classic Statistical Methodologies and their Related Problems, Vlerick Leuven Gent Working Paper Series 2004/15; [2] Bordeianu, D., D., et. al., (2011), Analysis models of the bankruptcy risk, Economy Transdisciplinarity Cognition, George Bacovia University, Vol. XIV, Issue 1, pp ; [3] Cîrciumaru, D., (2011), The Score Models For Analyzing The Bankruptcy Risk. Some Specific Features For The Case Of Romania, Article provided by University of Craiova, Faculty of Economics and Business Administration in its journal Revista Tinerior Economisti(The Young Economists Journal), Volume 1, Issue 16, pp ; [4] Cîrciumaru, D., et. al., (2015), Return Versus Risk. Evidence From Romania, Article provided by University of Craiova, Faculty of Economics and Business Administration in its journal Annals of Computational Economics, Volume 1, Issue 43, pp. 7-28; [5] Hautcoeur, P., C., și Levratto, N., (2006), Does the law alone explain the rise in bankruptcies in XIXth century France?, PSE Working Papers n <halshs >; [6] Lis, B., et. al., (2011), The Proposition Value Of Corporate Ratings - A Reliability Testing Of Corporate Ratings By Applying Roc And Cap Techniques, Article provided by Lucian Blaga University of Sibiu, Faculty of Economic Sciences in its journal Studies in Business and Economics, Volume 6, Issue 2, pp ; [7] Machek, O., (2014), Long-term Predictive Ability of Bankruptcy Models in the Czech Republic: Evidence from , Article provided by University of Economics, Prague in its journal Central European Business Review, Issue 2, pp ; [8] Mandru, L., et. al., (2010), The Diagnosis of Bankruptcy Risk Using Score Function, articol available at accesat la data de ; [9] Miron, V., C., I., (2015), Financial Position And Its Relevance To Stakeholders, Article provided by Constantin Brancusi University, Faculty of Economics in its journal Constatin Brancusi University of Targu Jiu Annals - Economy Series, Volume 2, Issue April), pp ; [10] Miroslava, D., (2014), Which Altman Model Do We Actually Use?, Article provided by University of South Bohemia in Ceske Budejovice in its journal Acta Universitatis Bohemiae Meridionales, Volume 17, Issue 2, pp ; [11] Mosionek-Schweda, M., (2014), The Use Of Discriminant Analysis To Predict The Bankruptcy Of Companies Listed On The Newconnect Market, Article provided by Uniwersytet Mikolaja Kopernika in its journal Equilibrium. Quarterly Journal of Economics and Economic Policy, Volume 9, Issue 3, pp ; [12] Munteanu, G., (2010), Metode de analiză a riscului de faliment, Revista Română de Statistică, nr. 12, pp ; [13] Petrișor, M., B., and Lupu, D., (2013), The Forecast Of Bankruptcy Risk Using Altman Model, Article provided by Stefan cel Mare University of Suceava, Romania, Faculty of Economics and Public Administration in its journal The USV Annals of Economics and Public Administration, Volume 13, Issue 2, pp ; [14] Posada, M., G., and Mora-Sanguinetti, J., S., (2014), Are there alternatives to bankruptcy? A study of small business distress in Spain, rticle provided by Spanish Economic Association in its journal SERIEs, Volume 5, Issue 2, pp ; [15] Răscolean, I., et. al., (2012), Predictive Analysis Software for Modeling The Altman Z-Score Financial Distress Status of Companies, Article provided by University of Petrosani, Romania in its journal Annals of the University of Petrosani Economics, Volume 12, Issue 3, pp ; [16] Stancu, I., Bălu, F., (2007), Utilizarea metodologiei VaR pentru măsurarea şi prevenirea riscului valutar, Revista ECTAP, articol disponibil la accesat în data de ; [17] accessed at ; 256