St. Petersburg University
Graduate School of Management
Master in Corporate Finance
BANKRUPTCY RISK OVER THE BUSINESS CYCLE:
FACTORS IDENTIFICATION
Master’s Thesis by the 2nd year student
Concentration – Corporate Finance
Anfisa Lisetskaia
Research Advisor:
Anna E. Loukianova, Associate Professor
St. Petersburg
2016
1
ЗАЯВЛЕНИЕ О САМОСТОЯТЕЛЬНОМ ХАРАКТЕРЕ ВЫПОЛНЕНИЯ
ВЫПУСКНОЙ КВАЛИФИКАЦИОННОЙ РАБОТЫ
Я, Лисецкая Анфиса Геннадьевна, студент второго курса магистратуры
направления «Корпоративные финансы», заявляю, что в моей магистерской диссертации
на тему «Риск банкротства на протяжении бизнес-цикла: определение факторов»,
представленной в службу обеспечения программ магистратуры для последующей
передачи в государственную аттестационную комиссию для публичной защиты, не
содержится элементов плагиата.
Все прямые заимствования из печатных и электронных источников, а также из
защищенных ранее выпускных квалификационных работ, кандидатских и докторских
диссертаций имеют соответствующие ссылки.
Мне известно содержание п. 9.7.1 Правил обучения по основным образовательным
программам высшего и среднего профессионального образования в СПбГУ о том, что
«ВКР выполняется индивидуально каждым студентом под руководством назначенного
ему научного руководителя», и п. 51 Устава федерального государственного бюджетного
образовательного учреждения высшего профессионального образования «СанктПетербургский государственный университет» о том, что «студент подлежит отчислению
из Санкт-Петербургского университета за представление курсовой или выпускной
квалификационной работы, выполненной другим лицом (лицами)».
___________________________________________ (Подпись студента)
_______25.05.2016___________________________ (Дата)
STATEMENT ABOUT THE INDEPENDENT CHARACTER
OF THE MASTER THESIS
I, Lisetskaia Anfisa, second year master student, program «Corporate finance», state that
my master thesis on the topic «Bankruptcy risk over the business cycle: factors identification»,
which is presented to the Master Office to be submitted to the Official Defense Committee for
the public defense, does not contain any elements of plagiarism.
All direct borrowings from printed and electronic sources, as well as from master theses,
PhD and doctorate theses which were defended earlier, have appropriate references.
I am aware that according to paragraph 9.7.1. of Guidelines for instruction in major
curriculum programs of higher and secondary professional education at St.Petersburg University
«А master thesis must be completed by each of the degree candidates individually under the
supervision of his or her advisor», and according to paragraph 51 of Charter of the Federal State
Institution of Higher Professional Education Saint-Petersburg State University «a student can be
expelled from St. Petersburg University for submitting of the course or graduation qualification
work developed by other person (persons)».
___________________________________________ (Student's signature)
________25.05.2016__________________________ (Date)
2
АННОТАЦИЯ
Автор
Название магистерской
диссертации
Факультет
Направление
подготовки
Год
Научный руководитель
Описание цели, задач и
основных результатов
Ключевые слова
Лисецкая Анфиса Геннадьевна
Риск банкротства на протяжении бизнес цикла: определение
факторов
Высшая школа менеджмента
Корпоративные финансы
2016
Лукьянова Анна Евгеньевна
Макроэкономическая среда существенно влияет на
результативность деятельности отдельных компаний и,
следовательно, на вероятность банкротства фирм в различных
отраслях экономики. Однако существующие модели оценки и
прогнозирования риска банкротства в основном учитывают
только финансовые показатели деятельности компании, при
этом упуская влияние макроэкономических факторов. Поэтому
представляется интересным проанализировать проблему
оценки риска банкротства с учётом макроэкономических
показателей, динамика которых демонстрирует циклический
характер и отражается в деловых циклах.
Цель магистерской диссертации – определение
факторов, влияющих на риск банкротства компаний на
протяжении делового цикла в российской макроэкономической
среде. Главные задачи исследования: определить факторы и
причины банкротства; проанализировать существующие
модели оценки риска банкротства, основанные на показателях
отчетности и включающие макроэкономические переменные;
разработать регрессионные модели, связывающие риск
банкротства с показателями деловых циклов.
Результаты проведенного исследования показали, что
макроэкономические показатели делового цикла влияют на
риск банкротства компаний, и степень их влияния меняется в
зависимости от фазы делового цикла: более сильное
воздействие макроэкономических показателей на риск
банкротства проявляется на восходящей фазе цикла.
В свою очередь, воздействие финансовых показателей
компании на риск банкротства тоже зависит от фазы делового
цикла.
Финансовые
коэффициенты,
характеризующие
прибыльность компании, сильно связаны с риском банкротства
на протяжении всего делового цикла, в то время как показатели
структуры капитала проявляют более значительное влияние на
риск банкротства в нисходящей фазе цикла.
Риск банкротства, модели прогнозирования банкротства,
деловые циклы, макроэкономические показатели, финансовые
показатели, логистическая модель
3
ABSTRACT
Master Student's Name
Master Thesis Title
Faculty
Main field of study
Year
Academic Advisor's
Name
Description of the goal,
tasks and main results
Key words
Lisetskaia Anfisa
Bankruptcy risk over the business cycle: factors identification
Graduate School of Management
Corporate Finance
2016
Anna E. Loukianova
External macroeconomic environment strongly affects
performance of separate companies and, consequently, influences
probability of going bankrupt for enterprises in different industries.
However, existing bankruptcy risk estimation approaches are mainly
based on financial ratios of an enterprise, omitting the influence of
external macroeconomic factors. Therefore, it is interesting to
analyze the problem of bankruptcy risk estimation in the light of
macroeconomic environment, which is characterized by cyclical
dynamics and reflected in the business cycles.
The goal of the master thesis is determination of factors
influencing bankruptcy risk over the business cycle in Russian
macroeconomic environment. The main objectives are to identify
factors and reasons of bankruptcy; to analyze existing accountingbased models and models with macroeconomic variables for
bankruptcy risk diagnostics; to develop models that associate
bankruptcy risk with business cycle indicators.
Results of the conducted research showed that corporate
bankruptcy risk is affected by business cycle indicators. However,
the power of influence varies depending on the business cycle
phase: macroeconomic indicators demonstrate stronger relation to
bankruptcy risk during the upward phase of the business cycle than
during the downward phase.
In turn, influence of financial indicators on bankruptcy risk
also depends on the business cycle phase. Profitability ratios are
found significant for bankruptcy risk explanation over the whole
business cycle, while financial structure indicators express stronger
relation to bankruptcy risk in the descending phase.
Bankruptcy risk, bankruptcy prediction models, business cycles,
macroeconomic indicators, financial ratios, logit model
4
TABLE OF CONTENTS:
INTRODUCTION ......................................................................................................................... 7
CHAPTER 1. BANKRUPTCY RISK AND APPROACHES FOR ITS ESTIMATION ....... 9
1.1. Bankruptcy as a stage of crisis processes in a company .................................................. 9
1.2. Factors and reasons of bankruptcy ................................................................................. 13
1.3. Bankruptcy risk in the system of financial risks ............................................................ 17
1.4. Approaches for bankruptcy risk estimation: accounting-based models ..................... 21
CHAPTER 2. INFLUENCE OF THE BUSINESS CYCLE ON CORPORATE
BANKRUPTCY RISK ................................................................................................................ 31
2.1. Dynamics of business cycles in Russia: macro and micro levels impact...................... 31
2.2. Bankruptcy of Russian enterprises: dynamics and sectoral structure ........................ 37
2.3. Approaches for bankruptcy risk estimation:
models with macroeconomic variables................................................................................... 41
CHAPTER 3. DETERMINATION OF BANKRUPTCY RISK FACTORS ......................... 48
3.1. Data description ................................................................................................................ 48
3.2. Applied methodology ........................................................................................................ 53
3.3. Descriptive statistics ......................................................................................................... 54
3.4. Regression analysis and empirical results ...................................................................... 58
CONCLUSION ............................................................................................................................ 67
REFERENCES ............................................................................................................................ 69
APPENDICES.............................................................................................................................. 72
Appendix 1. Number of bankrupt and operating companies in the research sample by
manufacturing industry subsectors ........................................................................................ 72
Appendix 2. Correlation matrix of macroeconomic variables............................................. 73
Appendix 3. Descriptive statistics of bankrupt companies over the upward
and downward phases of the business cycle .......................................................................... 74
Appendix 4. Descriptive statistics of operating companies over the upward
and downward phases of the business cycle .......................................................................... 75
Appendix 5. Examples of models with business cycle variables for upward
and downward phases of the business cycle .......................................................................... 76
5
List of Figures:
Figure 1. Phases of crisis processes in an enterprise ........................................................... 10
Figure 2. Bankruptcy risk as a consequence of financial risks .............................................. 17
Figure 3. Elements of enterprise’s solvency ...................................................................... 19
Figure 4. Dynamics of the annual real GDP growth rate in Russia in 1900-2015..................... 32
Figure 5. Russian economic cycles .................................................................................. 33
Figure 6. Growth rate dynamics of the main cyclical indicators of Russian economy
development for years 2000-2015 ................................................................................... 34
Figure 7. Changes in main macro and micro level indicators over the business cycle............... 36
Figure 8. Number of bankrupt enterprises in Russian economy ............................................ 37
Figure 9. Bankruptcy dynamics in the real sector of Russian economy in 2007-2014 .............. 38
Figure 10. Sectoral structure of bankrupt companies in the real sector of Russian economy in
2007-2014 ................................................................................................................... 40
Figure 11. Dynamics of mean values of financial indicators for operating and bankrupt
enterprises during 2007-2014 ......................................................................................... 57
List of Tables:
Table 1. Financial ratios used in foreign and Russian accounting-based bankruptcy models ...... 29
Table 2. Macroeconomic variables used in foreign and Russian
bankruptcy estimation models ....................................................................................................... 46
Table 3. Number of bankrupt and healthy companies in the research sample by year ................. 49
Table 4. P value intervals and characteristics of bankruptcy risk in logistic model...................... 54
Table 5. Descriptive statistics on bankrupt enterprises ................................................................. 55
Table 6. Descriptive statistics on healthy enterprises .................................................................... 55
Table 7. Correlation matrix of financial variables ......................................................................... 56
Table 8. Dating of turning points in the Russian business cycle ................................................... 59
Table 9. Financial ratios with the highest Pseudo R2 on the upward phase
of the business cycle (with coefficients in univariate analysis)..................................................... 60
Table 10. Financial ratios with the highest Pseudo R2 on the downward phase
of the business cycle (with coefficients in univariate analysis)..................................................... 61
Table 11. Coefficients of logistic regression models for bankruptcy risk estimation over
ascending and descending phases of the business cycle................................................................ 62
Table 12. Coefficients of macroeconomic factors influencing corporate bankruptcy risk
over ascending and descending phases of the business cycle ....................................................... 65
6
INTRODUCTION
Thematic justification. External macroeconomic environment strongly affects
performance of separate companies and, consequently, influences probability of going bankrupt
for enterprises in different industries. Economic booms and recessions, in particular, recent
global financial crisis, emphasize the importance of understanding the link between economic
state in the country and corporate bankruptcy probability in order to timely initiate appropriate
measures on company’s level. To mitigate negative external risk factors, companies should
regularly monitor them and adapt to them.
Beginning from 1960s, a wide variety of bankruptcy risk estimation models was
developed (the most known of them were suggested by Altman, Ohlson, Fulmer, and others).
However, these approaches are mainly based on financial ratios of an enterprise, omitting the
influence of external macroeconomic factors. Detailed consideration of bankruptcy risk
components allows assuming that bankruptcy risk is influenced by both financial and
macroeconomic indicators. Development of bankruptcy risk diagnostics models with
macroeconomic variables is a relatively new field of the recent research. Therefore, it is
interesting to analyze the problem of bankruptcy risk estimation in the light of macroeconomic
environment, which is connected with the business cycle. In the current study the relation of
macroeconomic factors to corporate bankruptcy risk is analyzed via the concept of medium term
Juglar business cycles.
The goal of the current study is determination of factors influencing bankruptcy risk over
the business cycle in Russian macroeconomic environment.
To achieve this goal the following research objectives were set:
1) to identify factors and reasons of bankruptcy;
2) to specify the role of bankruptcy risk in the system of financial risks;
3) to analyze existing accounting-based models and models with macroeconomic variables for
bankruptcy risk diagnostics;
4) to describe the emergence of business cycles in the Russian economy;
5) to identify indicators of business cycles in Russian economy;
6) to relate dynamics of bankruptcy in Russia to the Russian economic business cycles;
7) to develop models that associate bankruptcy risk with business cycle indicators.
The object of the research is influence of macroeconomic indicators on corporate
bankruptcy risk.
The subject of the research – business cycle factors that affect corporate bankruptcy risk
in Russian macroeconomic environment.
7
The theoretical foundation of the current thesis consists of studies and ideas of foreign
and Russian researchers. The most significant were papers of Altman E.I., Ohlson J.A.,
Zmiewski M.E., Giordani P., Jacobson T., Haydarshina G.A., Totmyanina K.M., Juglar C.,
Kondratiev N.D. and others.
Thesis structure. The goal and research objectives determined the structure of the
current study.
The first chapter is devoted to the category of bankruptcy risk and its place in the system
of financial risks. Main corporate bankruptcy factors and reasons are highlighted. Analysis of
existing accounting-based models for bankruptcy risk estimation allows selecting a number of
financial ratios, which are potentially essential estimators of bankruptcy risk and significant for
the purposes of the current study.
The second chapter concentrates on business cycles in Russian economy and their
relation to frequency of bankruptcies in the real sector. Business cycle influence on macro and
micro levels is discussed. This part ends with the overview of several foreign and Russian
models with incorporated macroeconomic variables, which form the basis for our further
research.
The third chapter presents the research methodology, data description and research
findings. Logistic regression analysis, which is the core part of the methodology, is based on data
of Russian enterprises from manufacturing industry.
Theoretical implication of the research is an improved understanding of factors
influencing corporate bankruptcy risk. In addition, directions for further research are formulated
in conclusion. Practical implication is mainly concerned with purposes of internal managers of a
company, suggesting a set of indicators affecting bankruptcy risk, which should be monitored in
order to prevent negative consequences and financial losses.
8
CHAPTER 1. BANKRUPTCY RISK AND APPROACHES FOR ITS ESTIMATION
1.1. Bankruptcy as a stage of crisis processes in a company
One of the key objectives of company's management is to prevent company’s transition
into bankruptcy. In order to prevent adverse consequences, company’s system of financial
management requires elaborate mechanism of bankruptcy risk diagnostics, quickly responding to
any changes in financial and economic activity.
In general, bankruptcy occurs when a firm is not able to cover its obligations to creditors,
suppliers, shareholders, employees, etc (Achim and Borlea 2012).
In Russia, according to laws in action (the Federal Law No.127-FZ dated October 26,
2002 “On Insolvency (Bankruptcy)” with amendments) bankruptcy is defined as “recognized by
a court of arbitration, an inability of a debtor to meet in full the claims of creditors relating to
financial liabilities, payments of severance benefits and/or remuneration to employees and/or to
settle the mandatory payments.”
At the same time, the Federal Law “On Insolvency (Bankruptcy)” distinguishes between
definitions of bankruptcy and insolvency. Insolvency means “caused by the lack of funds,
termination by a debtor firm to fulfill financial obligations and compulsory payments”.
According to the Russian legislation, an enterprise is considered to be a bankrupt if the
following two criteria are fulfilled:
• Financial liabilities, mandatory payments and other obligations have not been paid within
three months after their due date;
• Claims in respect of an indebted entity represent, in total, not less than three hundred
thousand Rubles.
Bankruptcy is aimed, on the one hand, to help business to survive, protecting indebted
company from creditors, and, on the other, to defend the interests of creditors, ensuring full or
partial repayment of provided funds (Nikolaeva and Paluvina 2014). Bankruptcy proceedings are
intended to restore company’s solvency and overcome financial distress via restructurisation of
the company. Furthermore, launch of bankruptcy proceedings implies replacement of executives,
who do not perform their duties.
However, bankruptcy also entails some negative consequences, which include partial loss
of creditors’ funds, job cuts, and the possibility of criminal bankruptcy associated with property
redistribution. A large number of bankruptcies in national economy result in higher
unemployment rate and decreasing effective demand. At the same time, the increase in budget
expenditures related to social payments is in evidence, while tax revenues decline.
9
Managerial
crisis
Acute crisis
Bankruptcy
Absolute
insolvency
Prolonged insolvency
Temporary insolvency
Liquidity crisis
Structural crisis
Strategic crisis
Financial performance of a company
Prodromal crisis
Bifurcation
point
Chronic crisis
Financial crisis
Time (t)
Economic and
legal crisis
Figure 1. Phases of crisis processes in an enterprise
Source: Zhdanov (2011).
The inevitable consequence of Russia’s transition to the market economy was a
development of such concepts as “crisis”, “insolvency” and “bankruptcy”. To analyze and
distinguish among such close but essentially different terms, we refer to the crisis theory.
Crisis can be defined as “an unstable time or state of affairs in which a decisive change is
impending – either one with a distinct possibility of a highly undesirable outcome, or one with a
distinct possibility of a highly desirable and extremely positive outcome” (Darling, Seristö and
Gabrielsson 2005, p.347). According to the definition, crisis is not necessarily a bad event, but it
is clearly accompanied by a certain degree of risk and uncertainty.
Crisis is an integral phase of company’s life cycle and can be divided into three distinct
periods: prodromal crisis, acute crisis and chronic crisis. Additionally, the progression of crisis
process can be described as transformation of managerial crisis to financial distress and, finally,
to economic and legal crisis. These phases include seven stages of crisis development: strategic
crisis, structural crisis, liquidity crisis, temporary insolvency, prolonged insolvency, absolute
insolvency, bankruptcy. Suggested classification assumes that bankruptcy appears to be the end
point of the crisis, when the company is totally unable to cover its debts.
Awareness of the main phases of crisis process facilitates recognition of crisis at an
earlier stage. It reduces the development speed and intensity of crises processes, time of their
occurrence, severity of the crisis and its consequences.
10
Prodromal crisis. Crisis processes start with a strategic crisis, which is triggered by
insufficient development of a strategic management system in a company. Expansion of the
strategic crisis leads to a structural crisis. This stage is characterized by decrease of company’s
activity, market share and profits, staff reduction.
Acute crisis begins with a liquidity crisis, which is marked by increasing debts to
creditors and deteriorating liquidity indicators. If no measures are taken at this stage, the
situation may worsen and transform into a temporary insolvency. Temporary insolvency is
caused by the lack of funds due to the fact that receivables are not fully recovered.
The last phase of crisis processes in a company is a chronic crisis. If no administrative
measures to resolve the crisis have been taken, the company enters a period of chronic or
unsurmountable crisis, which is marked by the absence of internal liquid resources in the
company. Prolonged insolvency occurs when a company cannot repay financial liabilities due to
the lack of assets and to restore the solvency, company has no other choice but to attract external
funding. If things get worse, company moves from prolonged insolvency to bifurcation point of
company’s development, after which a firm either overcomes the crisis and continues its activity,
or launches bankruptcy proceedings. There are two possible outcomes of the situation when the
company faces absolute insolvency – either merger, acquisition and restructuring procedures
without arbitration, or filing for bankruptcy (Vorotnikova and Pshipiy 2015).
It is necessary to distinguish between concepts of “bankruptcy” and “insolvency”,
because insolvency is a result of liabilities excess over assets value and takes place without
recognition by a court of arbitration. Court of arbitration only confirms signs of insolvency and
recognizes a firm to be a bankrupt. Therefore, while the term “insolvency” has economic
meaning, “bankruptcy” is both economic and legal category.
It is reasonable to state a causal link between insolvency and bankruptcy. On the one
hand, the bankruptcy cannot be confirmed without the fact of insolvency. On the other hand,
insolvency is the main reason to apply bankruptcy law. It is worth mentioning that insolvency is
not the fact of company’s bankruptcy, but only a prerequisite, a turning point in development of
a firm (Zhdanov 2011).
Bankruptcy has a variety of modifications. The following classification of bankruptcy
types is the most frequently used:
• Real bankruptcy involves actual loss of capital employed. As a consequence, company is
unable to restore solvency and financial system in subsequent periods. The company starts
legal bankruptcy proceedings, because unsurmountable level of capital losses does not allow
the firm to effectively continue its business activities.
11
• Technical bankruptcy occurs as a result of significant delays in collection of receivables.
However, the amount receivable exceeds accounts payable, and company’s assets
significantly exceed its financial obligations. That is why it is possible to avoid legal
bankruptcy through successful crisis management.
• Intentional (deliberate) bankruptcy is deliberately created insolvency of a company, which is
aimed at infliction of economic damage to the company. This type of bankruptcy is a
consequence of poor management, which pursues personal interests or interests of individual
groups. According to the Russian legislation, deliberate bankruptcy is a criminal offence.
• Fictitious bankruptcy occurs when a company gives knowingly false information about its
insolvency in order to mislead creditors and obtain a delay in payments or debt discounts.
Fictitious bankruptcy is also illegal way to terminate business activity.
• Latent bankruptcy takes place when a company intentionally hides the fact of bankruptcy. If
latent bankruptcy incurs tangible damage to creditors, such activity is a subject to legal
prosecution (Nikolaeva and Paluvina 2014).
In Russia the problem of fictitious and intentional bankruptcy identification is of high
importance. Despite the fact that the fictitious and intentional bankruptcy is part of economic
crime, the trend in this area shows a growing tendency. The use of bankruptcy institute for
personal purposes impedes the implementation of main functions of this institute – improvement
of the economy and creation of effective competitive environment.
Today some factors prevent effective control after fraudulent bankruptcy, and one of the
main problems is the absence of clear description of components of fictitious bankruptcy in
legislation. Other difficulties connected with identification of fictitious and intentional
bankruptcy are connected with concealing assets or financial liabilities, concealing information
about property, transfer of property to other owner, property destruction, falsification of
accounting and other registration documents (Abdullaev 2014).
The main learning point for our research is that existence of such types of fraudulent
defaults may distort bankruptcy statistics, which we use for modeling. However, for the purposes
of the current study we assume all enterprises under consideration to file for real bankruptcy.
12
1.2. Factors and reasons of bankruptcy
In many cases difficulties in the macroeconomic environment, accompanied by general
decline in production and rise in cost of capital, trigger crisis process in business units. Economic
crises may lead to mass bankruptcies, because economic agents are closely related to each other.
As market economy is a complex system of interactions among various entities, which are
connected by contractual relationships, financial difficulties of companies may be transferred to
their business partners and take significant scale. Thus, contractual relationships strengthen
interconnection and interdependence of market participants, when insolvency of one of the
parties and its default on obligations causes adverse financial and economic consequences for its
counterparties.
However, economic crisis is not the only period when it is possible to observe
bankruptcies. Some entities may file for bankruptcy during economic expansion, and explanation
may be in increasing competition. Period of economic growth is characterized by favorable
external conditions, which usually facilitate intensive production expansion and fierce
competition. In such a situation, many companies are unable to compete due to an inefficient
development strategy and management.
It should be realized that cyclical development is inherent for the market economy. In
large part because of the bankruptcy process, periods of economic decline and crises facilitate
the renewal of the economy. Bankruptcy is a necessary mechanism to get rid of inefficient
enterprises, clearing markets for other potentially more effective economic entities.
Therefore, market mechanism entails failures of inefficient economic agents.
Competition, cyclical development, market uncertainty and information asymmetry create
conditions, in which sustainable economic development of a company cannot be guaranteed.
The issue of bankruptcy factors and reasons has been widely discussed in literature.
However, usually the difference between the bankruptcy factors and reasons is not clearly stated.
Lvova O.A. distinguishes between bankruptcy factors and reasons, claiming that bankruptcy
factors exist due to changes in external and internal conditions of company’s operating
environment and exert negative impact only when bankruptcy reasons occur (Lvova and
Peganova 2014).
Bankruptcy factor can be defined as a disturbing event or trend, which indicates the
possibility of crisis with subsequent insolvency and bankruptcy of a company. Bankruptcy
factors affect all areas of business activity. In general, bankruptcy of firms results from the
development of crisis processes due to the influence of macroeconomic and microeconomic
factors.
13
The majority of researchers highlight the following macroeconomic factors, influencing
bankruptcy risk:
• crisis state of the real sector;
• reduction of the innovative capacity in the
• high interest rates;
• structural imbalances in the economy;
national economy;
• lack
• high barriers for entering the capital market;
of
funds
available
for
long-term
investments;
• instability of the tax and customs system;
• decline in consumer demand;
• dependence on export of raw materials;
• deterioration of the investment climate;
• high volatility of the exchange rate;
• high inflation rate.
External factors affecting bankruptcy also include political factors:
• weak government support of home producer;
• undeveloped legal and regulatory framework.
Macroeconomic factors influence the whole economic environment, but only some
companies become insolvent and leave the market. The existence of microeconomic factors
explains this situation. Among the most significant internal factors researchers state as follows:
- production of goods and services with low market demand or of noncompetitive quality;
-
existence of substitute goods;
-
lack of strong relationships with customers, inability to develop customer loyalty that ensures
constant income;
-
absence of stable relationships with suppliers that impedes continuous production;
-
low level of corporate culture and social capital;
-
ineffective advertising.
Thus, a combination of different external and internal, macroeconomic and
microeconomic factors influences company’s bankruptcy. We also need to consider that for
Russian companies possible bankruptcy factor is noncompetitiveness with European companies
on a global scale.
In addition to the abovementioned factors, there are some specific trends causing
bankruptcy, which are unique for the Russian market due to the historical heritage.
The transition from the state-controlled to market economy in Russia still has some
impact on the current economic environment. Even if it might seem that Russia has almost
completely implemented market economic structure, a period of a little more than 20 years
historically is not enough for entire economic transformation. For many years Russian emerging
economy showed problems connected with difficulties after privatization, high militarization,
inflexible large enterprises, managers who got used to receive production plans.
14
As operating in market environment is relatively new practice for Russian managers, one
of the possible factors for emergence of bankruptcy risk is the lack of experience of business
operations in the market-driven economy. Consequently, it entails poor management control and
increasing likelihood of insolvency and bankruptcy.
Many authors mention institutional factors as leading determinants of bankruptcy risk in
Russian economy. Such institutional factors include flaws in current bankruptcy legislation,
which facilitate growth of illegal situations involving bankruptcy. In particular, cases of fictitious
and intentional bankruptcy often occur, although these actions are criminally liable. Bankruptcy
institution in Russia is vulnerable to criminal purposes and may be used for property
redistribution and legalization of illegal property appropriation (Nikolaeva and Paluvina 2014).
One more factor influencing business operations is limited access to lending resources
because of imperfections in the Russian credit system. From the year 2005 Russian banking
sector can not satisfy domestic demand for financial resources from the private sector. This
tendency is reflected in the dynamics of domestic credit provided by financial sector to total
domestic credit received by private sector. While developed European economies have this ratio
higher than 1, for Russian situation during the last 10 years this indicator was below 1 (The
World Bank Database). Such statistics demonstrate that national financial sector does not
provide sufficient funds for the corporate sector, and Russian enterprises are forced to raise funds
from abroad. Moreover, due to high domestic interest rates, lending in foreign banks is
significantly cheaper than in Russian financial institutions. However, despite many incentives for
companies to attract additional financial resources through foreign banks, high country risks
make it difficult for national enterprises to obtain loans in foreign markets.
If some bankruptcy factors exist, appearance of bankruptcy reasons may trigger crisis
processes in an enterprise. Thus, bankruptcy reasons are events resulting in rapid emergence of
bankruptcy risk factors.
There are different approaches to the classification of bankruptcy reasons. It is also
necessary to keep in mind that bankruptcy reasons may include certain combination of
bankruptcy reasons, which is unique for each separate company. The combination of internal
problems connected with company’s business activity may include:
1. Operating reasons:
- high degree of depreciation of fixed assets, low level of used technologies;
- ineffective management of cash flows;
- high proportion of work in progress in current assets, which entails capital turnover
slowdown;
- growth of receivables for goods delivered but not paid;
15
- uncontrolled growth of the business, violation of balanced growth rates, leading to unplanned
expenses;
- inefficient use of operating resources and, as a consequence, high cost of goods sold;
2. Managerial reasons:
- undeveloped crisis management program;
- unprofessional management, entailing inaccurate assessment of risks and making wrong
decisions;
- risky and aggressive development program, suggesting a large borrowings;
- lack of effective audit control;
3. Financial reasons:
- investments in fixed assets, while working capital is managed inefficiently;
- high borrowing costs;
- negative financial leverage;
- underestimation of financial risks;
- inefficient budgeting system and financial strategy (Vorotnikova and Pshipiy 2015).
Under the economic crisis that influenced the whole world economy, the reasons of
business bankruptcies became more and more diverse and complex. Numerous studies mention
many other causes of companies’ bankruptcy, among which are:
• Company age. Established companies which have been in business up to five years have
lower risk of bankruptcy than new entrants.
• Sector of activity. Russian and foreign researchers reveal that the probability of failure
depends on the sector of company’s operations. Haydarshina G.A. (2009) justifies that
adjustment of financial ratios to the industry facilitates predictive accuracy.
• Company size. Evidence shows that bankruptcy is more common phenomenon for small
companies than the big ones. Ohslon (1980), Fulmer (1984), Evstropov (2008), Fedorova and
others (2013) incorporate size variable in their models, because it demonstrates high
significance for bankruptcy analysis.
It is unlikely that the appearance of only one cause will necessarily lead to the inevitable
bankruptcy. Usually a combination of various bankruptcy factors and reasons leads to
unfavorable consequences. It is difficult to determine which particular causes are the most
significant for Russian enterprises. But we should take into account that the level of business
activity in Russia is still not sufficiently high, and therefore the primary role play external factors
- political, economic, financial instability (Monea 2014).
16
1.3. Bankruptcy risk in the system of financial risks
Business activity is associated with many risks, which have strong impact on companies
during financial and economic crises in the national economy. Instability of the economic
situation and high market volatility increase financial risks, accompanying business activities of
both large and small enterprises and associated with financial losses. Adverse external factors in
conjunction with ineffective internal risk-management system lead to crisis situations and later to
financial insolvency and bankruptcy of an enterprise (Y. Sitnikova 2012).
“Depending on the specificity of the economic activity performed, the major risks
possible to affect an entity are: the operational risk, the financial risk, the commercial risk, and
bankruptcy risk. From the multitude of risks the most important to be considered is the
bankruptcy risk, which can be caused by the appearance of all the others types of risks” (Monea
2014, pp.150-151).
The bankruptcy risk is a part of internal business risks and expresses the possibility of
failure to meet timely payments. A comprehensive definition of bankruptcy risk was given by
Haydarshina G.A. (2009, p.86): “The risk of bankruptcy is an economic category, which can be
measured quantitatively and which reflects company’s probability of inability to fully satisfy
creditors’ claims, as well as to settle mandatory payments in the course of decision-making under
external environment uncertainty”. The bankruptcy risk of enterprises is closely connected to the
financial risks. Many authors observe that financial risks could lead a company to the loss of
solvency and reduction of financial stability, which in the worst case result in bankruptcy. Thus,
bankruptcy risk can be divided into two components – risk of insolvency and risk of financial
Bankruptcy risk
Credit risk
Interest-rate risk
Inflation risk
Investment risk
Deposit risk
Currency risk
Stock market risk
Tax risk
Risk of financial
instability
Insolvency risk
Figure 2. Bankruptcy risk as a consequence of financial risks
Source: Frolov (2010, p.98).
17
instability. The system of financial risks, which together lead to increase of bankruptcy risk, can
be presented as in the Figure 2.
Financial risk means the risk of adverse financial consequences, loss of income or capital,
as a result of company’s business activity under uncertainty. According to the Figure 2, the
following financial risks may cause insolvency and financial instability of an enterprise:
• Stock market risk characterizes the possibility to lose assets and funds due to unfavorable
change in stock market rate of securities or implementation of margin trading.
• Currency risk. Companies are subject to currency risk if they are engaged in international
business and receive foreign currency revenue, purchase raw materials or equipment in
foreign currency, have funds or investments denominated in foreign currency. This type of
risk leads to income shortfall or increase of planned expenditures.
• Deposit risk is defined as the risk of non repayment/underpayment of deposit or interest on it
during the contract period.
• Investment risk occurs when a company faces financial losses due to decreasing investment
attractiveness of the undertaken project.
• Inflation risk arises from the possibility of deterioration in capital value and expected income
as a result of inflation. In more detail, this risk may be connected with loss of accounts
receivable value due to delayed payments, increase in cost of goods and services due to
increasing energy prices, transportation costs, wages, etc.
• Tax risk reflects the probability of introduction of new taxes and levies, increase in existing
tax rates, change in terms and conditions of tax payments and repeal of tax benefits.
• Interest-rate risk is the risk of adverse changes in both deposit and credit interest rates.
• Credit risk takes place when a company provides commodity (commercial) loan and sales
goods and services on a deferred-payment basis (Frolov 2010).
It may be noted that many financial risks are strongly connected with macroeconomic and
market factors – interest rate, inflation, stock market. Thus, it is fair to assume that consideration
of macroeconomic and market indicators may facilitate bankruptcy risk assessment.
Bankruptcy risk is an integrated risk, because not only the impact of financial risks may
lead to the thread of bankruptcy. It is also necessary to take into account other types of risk such
as strategic, structural, operational, technical, technological, innovative and commercial, which
are able to bring a company to such a catastrophic result (Y. Sitnikova 2012).
Exposure to the bankruptcy risk is closely related to the state of solvency of the company,
“reflecting the possibility that an entity will no longer be able to honor its payment obligations”
(Bogdan 2014, p.20). Analysis of the bankruptcy risk is based on the idea that bankruptcy is a
18
phenomenon, which does not occur suddenly or unexpectedly; bankruptcy is a result of crisis
processes, which develop in the course of time and influence financial indicators. Degradation of
financial situation, reflected in entity’s financial indicators, denotes increasing bankruptcy risk
that threatens the smooth running of a business.
To link the assessment of bankruptcy risk with company’s financial indicators, it is
reasonable to refer to the concept of “solvency” as a combination of statical and dynamical
stability. In this case, financial stability is defined as equilibrium characterized by the company’s
adaptability to changeable external factors. It is implied that in equilibrium externally influenced
parameters of a firm fluctuate insignificantly and have a tendency to return to the original state.
This approach assumes an enterprise is deemed to be “a system that constantly seeks to maintain
a balance between internal capabilities and external forces (i.e. a self-stabilizing system) in order
to keep steady state” (Makarov and Rakhimova 2014, p.37).
For further measurement of bankruptcy risk, we need to identify the most important
elements of solvency assessment. To analyze solvency structure, Makarov A.S. and Rakhimova
O.S. (2014) suggested the following scheme:
Solvency
Financial position
Financial stability
statical financial
resource capacity
dynamical financial
resource capacity
Potential
repayment ability
Current financial
self-sufficiency
Operating
repayment ability
Availability of
liquid assets to
cover current
liabilities
Availability of
stable liabilities to
finance illiquid
assets
Ability to timely
and fully repay the
liabilities via funds
generated by
business operations
Ability to replace
debt capital with
equity
(self-financing)
Operating margin
Operating liquidity
Proportion of assets and
liabilities in terms of
liquidity and maturity
Liquidity structure
of assets
Maturity structure
of liabilities
Prospective
financial
self-sufficiency
Cost of capital
Figure 3. Elements of enterprise’s solvency
Source: Makarov and Rakhimova (2014, p.36).
19
Solvency of a company is associated with financial resource capacity, necessary for
achievement of company’s goals. Financial resources have a form of cash, liquid assets and
expected income at the disposal of the company. From this point of view, financial position, as a
component of solvency, reflects current financial resource capacity, while financial stability is
related to available financial resources over time.
Company’s financial position reflects the level of insolvency risk. Insolvency risk is
determined by decreasing liquidity of current assets, causing an imbalance of positive and
negative cash flows of the company in time terms. To assess insolvency risk, indicators of
liquidity structure of assets and maturity structure of liabilities should be analyzed.
Risk of financial instability refers to dynamical financial state of the company. Financial
instability risk is determined by the imperfection of capital structure (overreliance on borrowed
funds), which causes imbalance of positive and negative cash flows of the enterprise in volume
terms. This type of risk is characterized by the following financial ratios: equity to total assets
ratio, total debt to equity, assets coverage ratio, equity plus long-term debt to total assets ratio,
earnings to interest and principal expenses. For financial instability risk evaluation, it is also
necessary to consider cash flow and profitability ratios such as operating margin (operating
income to net sales) and operating liquidity (operating cash flow to sales). One more indicator of
financial stability, suggested by Gadanecz and Jayaram (2008), is net foreign exchange exposure
to equity: high levels of this ratio may signal difficulties in the corporate sector arising from
negative currency moves.
20
1.4.
Approaches for bankruptcy risk estimation: accounting-based models
Modern research literature suggests a huge number of models and approaches for
bankruptcy risk assessment. Applying different methods, qualitative and quantitative indicators,
many authors have tried to develop bankruptcy prediction models, basing on the smallest
possible number of parameters but with high predictive power. The most common models for
bankruptcy risk analysis are:
• Accounting models, which use separate financial ratios – liquidity ratio, profitability, cash
ratio etc.
• Theoretical models, based on qualitative criteria: gambler’s ruin theory, option-priced theory,
credit risk theories etc.
• Statistical models, which include univariate analysis, multiple discriminant analysis, survival
analysis, logit and probit models. Statistical models are the most frequently used for
bankruptcy risk analysis and include both financial and non-financial variables such as
company size, sector of activity, country risk etc. Using statistical models for corporate
bankruptcy prediction provides calculation of synthetic risk indicator, which characterizes the
financial state of the company (Achim and Borlea 2012).
• Artificial intelligence models, using soft computing techniques (decision trees, neural
networks, rough sets theory, and genetic algorithm). Although these methods are relatively
new, soft computing techniques have already demonstrated high predictability results. Soft
computing models process and interpret data in a variety of capacities, generalize knowledge
and classify object into one of the previously observed categories (Korol 2013).
According to the research conducted by Aziz and Dar, 64 per cent of case studies on
bankruptcy prediction used statistical models, 25 per cent - artificial intelligence methods, and 11
per cent - other types of bankruptcy risk analysis (Aziz and Dar 2006).
The main difference between statistical and artificial intelligence models is based on
characteristics of included variables. Statistical methods require precise, reliable, and accurate
parameters, while artificial intelligence models tolerate inaccurate data, uncertainty, and
approximation (Korol 2013).
Among all the techniques, the most frequently applied are multiple discriminant analysis
(more than 30 per cent of studies), logit model (21 per cent of studies) and neural networks (9
per cent of models). The average overall predictive accuracy of logit and neural network models
(one year before actual bankruptcy) is of 87 per cent, multiple discriminant analysis – 85 per cent
(Aziz and Dar 2006).
21
As part of bankruptcy risk prediction, an important issue is the selection of an optimal set
of financial and economic indicators, which have the best predictive ability. The complexity of
this issue is reflected in the absence of generally accepted methodology and theoretical
approaches for determination of such measures. For this reason, studies devoted to the
bankruptcy issue highlight a wide range of predictive indicators. It is worth mentioning that
financial and economic variables, which have weak dependence on macro factors and are
essential for any company as an economic entity, are supposed to provide more promising
predictive results (Kopelev 2014).
The first step in our research is to review existing risk assessment methods. We compared
the most frequently applied methods of bankruptcy prediction models, starting with classical
approaches, which are used as the background for contemporary research.
Beaver
More comprehensive analysis on corporate bankruptcy prediction has started in the
1960s. In 1966 William H. Beaver raised the question of applicability of accounting data (i.e.,
financial statements) for corporate default prediction. His univariate study proved that the
financial ratios of bankrupt firms generally differ from those of non-bankrupt firms. In his study,
Beaver revealed that one year before failure the non-failed firms continue to grow while the total
assets of the bankrupt firms decline. During the research, the author tested several financial
ratios, including cash flow to total debt, working capital divided by total assets, current ratio,
total debt divided by total assets, net income to total assets. Beaver concluded that the best
predictor among the analyzed indicators was cash-flow to total debt ratio, where cash is
calculated as net income plus depreciation, depletion and amortization. Classification accuracy
of the cash-flow to total debt ratio was in the range from 87 per cent (one year before
bankruptcy) to 78 per cent (five years prior to failure) (Beaver 1966).
Altman Z-score
In 1968, Edward I. Altman, Professor of finance at New York University, continued
Beaver’s research and introduced multivariate discriminant technique for predicting firms’
bankruptcy (MDA). Altman conducted his research, basing on data of 66 (33 bankrupt and 33
non-bankrupt) listed manufacturing corporations. Having tested 22 variables, the author selected
five financial ratios as demonstrating the highest prediction ability of corporate bankruptcy. The
following discriminant function was derived as the result of Altman’s research:
𝑍 = 0.012𝑋1 + 0.014X2 + 0.033X3 + 0.006X4 + 0.999X5
(1)
where X1= Working capital / Total assets
22
X2 = Retained earnings / Total assets
X3 = Earnings before interest and taxes / Total assets
X4 = Market value of equity / Book value of total debt
X5 = Sales / Total assets
All firms having a Z-score of greater than 2.99 clearly fall into the “non-bankrupt” sector,
while those firms having a Z below 1.81 are all bankrupt. The area between 1.81 and 2.99 is
defined as the “zone of ignorance”, because of the susceptibility to error classification.
Altman’s model was highly accurate one year before bankruptcy - 95 per cent of the
analyzed firms were correctly classified. As the lead time increases, the overall effectiveness of
the discriminant model reduces to 72 per cent for two years preceding bankruptcy, 48 per cent as
of three years before the actual event, and 36 per cent as of five years prior to bankruptcy. Thus,
the predictive ability of the discriminant model deteriorates substantially as the prediction time
period is extended. Altman states that after the second year the model becomes unreliable in its
predictive ability. Trying to investigate the possible reasons for this finding, Altman analyzed the
dynamics of five predictive variables and concluded that “the most serious change in the
majority of these ratios occurred between the third and the second years prior to bankruptcy”
(Altman 1968, p. 606).
Springate S-score
It is also worth mentioning about the study of Gordon L.V. Springate conducted in 1978.
Using multiple discriminant analysis, the researcher selected the following four financial ratios
with the highest predictive ability, which he included in his model:
1) Working capital / Total assets
2) Net profit before interest and taxes / Total assets
3) Net profit before taxes / Current liabilities
4) Sales / Total assets
The model showed an accuracy rate of 92.5 per cent (Springate 1978).
Ohlson O-score
In 1980, James A. Ohlson, Professor of accounting at New York University, contributed
to the research with his logistic regression model for corporate bankruptcy prediction. Ohlson
based his model on a larger sample of companies than Altman. The data sample consisted of
listed industrial companies - 2058 individual non-bankrupt and 105 bankrupt enterprises. To
amplify model quality, Ohlson included two dummy variables and company size variable.
Ohlson’s score is given by the equation with nine independent variables:
𝑂 = −1.32 − 0.407𝑋1 + 6.03X2 − 1.43X3 + 0.0757X4 − 2.37X5 −
−1.83X6 + 0.285X7 − 1.72X8 − 0.521X9
(2)
23
where statistically significant variables can be divided into five groups:
1) size:
X1 = log (Total assets / GNP price-level index)
2) financial structure as reflected by a measure of leverage:
X2 = Total liabilities / Total assets
3) measures of current liquidity:
X3 = Working capital / Total assets
X4 = Current liabilities / Current assets
4) performance measures:
X5 = Net income / Total assets
X6 = Funds provided by operations / Total liabilities
𝑁𝐼 −𝑁𝐼
X9 = |𝑁𝐼 𝑡|+|𝑁𝐼𝑡−1 |, where NIt is the net income for the most recent period.
𝑡
𝑡−1
5) dummy variables:
X7 = One if net income was negative for the last two years, zero otherwise
X8 = One if total liabilities exceed total assets, zero otherwise
The probability of bankruptcy can be obtained using logistic transformation:
exp (O−score)
. If the result is larger than 0.5, there is a high probability of default within two
1 + exp(O−score)
years. The prediction accuracy of the model is 95 per cent two years prior failure and 92 per cent
as of three years before the bankruptcy (Ohlson 1980).
We also should note here the studies of Taffler R.J. (1983), Zmijewski M.E. (1984) and
Fulmer J.G. (1984). The explanatory variables that researchers found significant for bankruptcy
risk diagnostics are:
Taffler model: 1) Profit before tax / Current liabilities, 2) Current assets / Total liabilities, 3)
Current liabilities / Total assets, 4) No-credit interval. The last ratio determines how many days
for the company would be able to finance its continuing operations in case that it stops
generating revenue (Taffler 1983).
Zmijewski Score: 1) Net income / Total assets, 2) Total debt / Total assets, 3) Current assets /
Current liabilities (Zmijewski 1984).
Fulmer H-score: 1) Retained Earnings / Total Assets, 2) Sales / Total Assets, 3) EBT / Total
equity, 4) Cash flow from operations / Total debt, 5) Total debt / Total assets, 6) Current
liabilities / Total assets, 7) Log (Tangible assets), 8) Working Capital / Total debt, 9) Log (EBIT
/ Interest expense) (Fulmer, et al. 1984). This model showed considerably accurate results, when
it was tested on Russian manufacturing companies (Fedorova, Gilenko and Dovzhenko 2013).
24
Pang‐Tien, Ching‐Wen and Hui‐Fun
Using logit regression analysis, Pang-Tien et al. (2008) established financial earlyearning models that enable to predict the probability of impeding financial distress. The study
was based on the data of 116 business groups 1 from Taiwan that experienced financial distress
during the years 2002-2007.
Having tested 37 independent variables (28 financial and 9 non-financial), researchers
concluded that “financial ratio variables remain the primary variables for predicting corporate
financial distress”.
However, authors found that the combination of explanatory variables
changes depending on the remaining time before bankruptcy. Financial ratios make good work
with predicting financial problems one and two years prior to bankruptcy. But the longer the
time that remains before occurrence of financial distress, the less explanatory power financial
indicators have. To explain bankruptcy three years before the fact, researchers included
ownership structure and corporate governance variables in the model. As a result, three equations
with the following explanatory variables were developed:
• One year prior to the occurrence of financial distress, the predictor variables include:
1) Debt ratio = Total debt / Total assets
2) Times interest earned = EBIT / Interest expenses
3) Interest expense ratio = Cash from operating activities before interest and tax / Interest
expenses.
• Two years prior to bankruptcy:
1) Debt ratio = Total debt / Total assets
2) Operating expense ratio = Operating expenses / Net sales
3) Net income ratio = Net profit after tax / Net sales
4) Retention ratio = Earnings after distribution / Net profit after tax.
• Three years prior to the occurrence of financial distress:
1) Cash flow ratio = Net cash flow from operating activities / Current liabilities
Ownership structure variables:
2) Establishment of independent directors and supervisors = One if the company has no
independent directors or supervisors, zero otherwise
3) Pledge ratio for shares held by directors and supervisors = Shares pledged by all directors
and supervisors / Shares held by directors and supervisors.
Generally speaking, the logarithm regression model has significant predictive accuracy
above 90 percent over all three time frames (Pang‐Tien, Ching‐Wen and Hui‐Fun 2008).
1
The term “business group” refers to a group that assembles independent firms under common management and
financial control.
25
As mentioned many Russian authors, foreign forecasting models of default do not
demonstrate satisfactory accuracy when applied to Russian companies’ data and need to be
adjusted for national conditions. While some authors attempted modification of Western and
American models, others are of the opinion that peculiarities of Russian economic conditions do
not allow to use foreign models and require the development of national models with different
set of explanatory variables. The main reasons of low accuracy of foreign models in Russian
conditions are:
• differences in data used to establish models. Foreign and Russian models are based on
noncomparable normative parameters of the balance sheet structure and performance
indicators of enterprises.
• various macroeconomic conditions. Due to diverse economic development level, coefficients
of bankruptcy risk assessment models, designed for enterprises in countries with developed
market economies, as a rule, do not apply to countries with transition economies.
• multicollinearity of factors that causes distortion of coefficients estimates;
• specificity of industries is not taken into account. Most foreign models are originally
developed as “universal” for businesses of all industry segments. However, optimal values of
the key financial variables vary greatly for different industries (Haydarshina 2009).
The main difficulty with development of Russian bankruptcy models is connected with
short history of bankruptcy institute in Russia and lack of bankruptcy statistics. The most known
methods of bankruptcy risk diagnostics for Russian enterprises were developed in 1998-1999 by
Zaitseva, Davydova and Belikov (R-score), Saifullin and Kadykova. However, existing Russian
models show unsatisfactory prediction accuracy, which is sometimes even lower than that of
foreign analogues (Fedorova, Gilenko and Dovzhenko 2013). This fact explains the necessity to
develop effective methods for bankruptcy risk valuation for Russian enterprises. Further we
consider a couple of Russian models that are notable for significant forecasting ability.
R-score (Irkutsk model)
In 1999 one of the first Russian bankruptcy risk estimation models was developed by
Davydova G. and Belikov Yu. To select appropriate explanatory variables, researchers have
conducted a survey among managers of 80 commercial enterprises in Irkutsk. Basing on the
results of the survey, official regulatory methodology and Altman model, authors applied
discriminant analysis for development of the following equation, which is called R-score:
𝑅 = 8.38𝑋1 + X2 + 0.054X3 + 0.63X4
(3)
where X1= Working capital / Total assets
X2 = Net income / Shareholders’ equity
26
X3 = Revenue / Total assets
X4 = Net income / (Cost of goods sold + Operating expenses)
The rule for bankruptcy risk estimation is stated as follows: the higher the R-score, the
lower bankruptcy risk. If R-score is negative, the probability to become bankrupt is the highest –
90-100%. If R-score is more than 0.42, there is minimum bankruptcy risk (Davydova and
Belikov 1999).
The main drawback of the model is that it is efficient only when time period prior to
bankruptcy is very short – less than 3 quarters. Hence, R-score model have low potential to be
used for bankruptcy risk diagnostics. However, this model demonstrated the highest predictive
accuracy (of 71.8 per cent) among other Russian models, when it was tested on manufacturing
companies’ data (Fedorova, Gilenko and Dovzhenko 2013).
Evstropov
In 2008, Evstropov M.V. was the first Russian researcher who applied logit regression
analysis to national companies’ data for bankruptcy prediction. Russian manufacturing
enterprises were an object of the study. The main drawback of the model is the limited sample,
which consists of only 16 companies.
Evstropov developed two models. The first model is designed to predict default state four
years prior to the actual fact, using the following explanatory variables: 1) Book value of stock
shares / Total debt, 2) Current assets / Total assets, 3) ln(Total assets/GDP price-level index), 4)
Net sales / Average fixed assets. The second model, which forecasts bankruptcy two years before
the event, includes five financial ratios:
1) EBIT / Total assets
2) Net sales / Average accounts receivable
3) Revenue / Long-term debt
4) Annual revenue growth rate
5) Cash and cash equivalents / Current liabilities.
The last model demonstrated a high accuracy rate of 90.5% (Evstropov 2008).
Fedorova, Gilenko and Dovzhenko
One of the recent bankruptcy studies was conducted by Fedorova, Gilenko and
Dovzhenko in 2013. For the research 3056 Russian enterprises from the manufacturing industry
were chosen. Among 134 explanatory variables, the authors selected eight financial ratios and
estimated the following logit-model:
𝐹𝐺𝐷 = 10.3 − 6.2𝑋1 − 5.649X2 − 0.818X3 − 1.08X4 −
−0.638X5 − 1.932X6 − 0.928X7 − 2.249X8
(4)
27
where X1= Cash and liquid assets / Current assets
X2 = Net income / Total liabilities
X3 = lg (Tangible assets)
X4 = Inventory / Current liabilities
X5 = Revenue / Total liabilities
X6 = Noncurrent assets / Total assets
X7 = Gross profit / Cost of goods sold
X8 = Current assets / Total liabilities
The interpretation of FDG index is as follows: if FGD1 > 0, then there is a high
probability of bankruptcy of a company and if FGD1 < 0, the company is recognized financially
stable. The overall forecasting accuracy of the model is 87.14 per cent (Fedorova, Gilenko and
Dovzhenko 2013).
Shirinkina and Valiullina
In 2015 Shirinkina E. and Valiullina L. summarized existing foreign and national models
of risk estimation and highlighted six the most frequently used coefficient, among which are:
1) Return on assets = Net income / Assets
2) Assets turnover = Sales / Total assets
3) Current ratio = Current assets / Current liabilities
4) Return on equity = Net income / Shareholders’ equity
5) Current assets / Total assets
6) Profit margin = Net income / Net Sales
Authors also provide bankruptcy risk model with abovementioned explanatory variables, but
there is no any information about model’s prediction accuracy (Shirinkina and Valiullina 2015).
Financial ratios, which were used in the previously discussed accounting-based
bankruptcy risk models, are summarized and presented in Table 1.
In general, solvency, liquidity, efficiency and profitability ratios constitute the basis for
both Russian and foreign bankruptcy prediction modeling. But also some differences in groups
of variables used can be noticed. In particular, financial structure and cash flow indicators are not
frequently considered by Russian researchers for bankruptcy risk modeling, while these variables
showed high significance in foreign equations. In addition, asset structure and operational
efficiency ratios demonstrate satisfactory predictive ability in national models, while variables of
these groups are rarely included in American and Western models.
To choose appropriate variables for bankruptcy models, which we are going to construct
in the third chapter, two criteria were used. The first criterion was frequency – common use in
28
Table 1
Financial ratios
Const
Total debt / Total assets
Equity / Total debt
Total liabilities / Total assets
Current liabilities / Total assets
Current assets/Current liabilities
Current assets / Total liabilities
Absolute liquidity ratio
Working capital / Total assets
Working capital / Total debt
EBIT /Current liabilities
Net income / Total liabilities
Revenue / Total liabilities
Revenue / Long-term debt
EBIT / Interest expenses
Net sales / Average receivables
Net sales / Average fixed assets
Revenue / Total assets
Gross profit / COGS
Net income /(COGS + Oper. exp.)
Altman
(1968)
Springate
(1978)
Ohlson
(1980)
Taffler
(1983)
Fulmer
(1984)
Zmiewski
(1984)
Pang-Tien
et al. (2008)
R-score
(1999)
Evstropov
(2008)
Fedorova
et al. (2013)
Shirinkina
et al. (2015)
Financial ratios used in foreign and Russian accounting-based bankruptcy models
+
+
+
Financial structure indicators
+
+
+
+
+
+
Solvency and liquidity indicators
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Operational efficiency ratios
+
+
+
+
+
+
+
+
+
+
Profitability ratios
Operating expenses / Net sales
Net income / Revenue
Earnings after distrib./Net income
Retained earnings / Total assets
EBIT / Total assets
EBT / Equity
Net income / Total assets
Net income / Equity
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Cash flow indicators
CFO / Current liabilities
CFO / Total liabilities
CFO / Total debt
CFO / Interest expenses
+
+
+
+
Assets structure indicators
Current assets / Total assets
Noncurrent assets / Total assets
Cash & liq. assets /Current assets
Inventory / Current liabilities
Size variable
Total liabilities > Total assets
Net income < 0
+
+
+
+
Other variables
+
+
+
+
+
+
+
+
Note: CFO – Cash from operating activities, EBIT – Earnings before interest and taxes, COGS – Cost of goods sold
Source: The present study.
29
the literature. The ratios advocated in the literature are perceived by many authors to reflect the
important relationships. The second criterion was that the ratios performed well in one of the
previous studies. This criterion will enable the study to examine the consistency of its findings
with those of the previous studies.
A large number of various financial ratios were considered for modeling bankruptcy risk
in the observed studies. As the result of the previous analysis of bankruptcy risk assessment
techniques, several accounting-based variables are selected to be tested for significance in the
third chapter of the current study.
First of all, we refer to Russian models, because they are supposed to be more accurate
when applied to the data of Russian companies. From the previous Russian research we choose
the following variables:
1) Revenue to total assets;
2) Net income to equity;
3) Share of current assets in total assets;
4) Share of noncurrent assets in total assets;
5) Company size variable.
In foreign literature the most common indicators are:
1) Total debt / Total assets;
7) Equity / Total debt;
2) Working capital / Total assets;
8) Current assets / Total liabilities;
3) Net income / Total assets;
9) EBIT / Current liabilities;
4) Current liabilities / Total assets;
10) EBIT / Interest expenses;
5) Retained earnings / Total assets;
11) Net income / Net sales;
6) Current assets / Current liabilities;
12) EBIT / Total assets.
For convenience these financial ratios are divided into five groups: 1) Financial structure
indicators, 2) Solvency and liquidity indicators, 3) Operational efficiency ratios, 4) Profitability
ratios, 5) Assets structure indicators.
Incorporating the financial structure indicators, we account for economic distress - events
when firms demonstrate balance-sheet-based insolvency if the value of the liabilities exceeds
assets value. The earnings and liquidity ratios provide significant information related to whether
a firm is at risk of financial distress, reflected in a lack of liquid assets to cover debt payments
and current expenditures. Hence, abovementioned financial ratios are potentially important
estimators of corporate bankruptcy risk.
30
CHAPTER 2. INFLUENCE OF THE BUSINESS CYCLE
ON CORPORATE BANKRUPTCY RISK
2.1. Dynamics of business cycles in Russia: macro and micro levels impact
Economists revealed that market economy develops cyclically. Cyclicality of economic
development is reflected in continuous fluctuations of production and business activity – periods
of growth give way to recessions and vice versa. Bankruptcy is usually associated with the
downward phase of a cycle, when economy moves from the boom to crisis. As many researchers
argue that the number of bankruptcies is connected with the state of economy and fluctuations in
macroeconomic variables, we refer to the business cycle theory, which helps to understand
determinative processes in Russian economy, causes for transition from economic expansion to
recession periods and whether cyclical changes in Russian economy facilitate explanation of
bankruptcy frequency.
The comprehensive definition of a business cycle was given by Burns and Mitchell
(1946, p.3) in one of the basic papers devoted to this issue: “Business cycles are a type of
fluctuation found in the aggregate economic activity of nations that organize their work mainly
in business enterprises: a cycle consists of expansions occurring at about the same time in many
economic activities, followed by similarly general recessions, contractions, and revivals which
merge into the expansion phase of the next cycle”.
As a result of numerous attempts to explain mechanisms responsible for economic
dynamics, various concepts justifying the cyclicality of the economy were created. Existing
economic cycles are different in its duration (long-, medium- and short-term) and generating
factors. The most common is the typology of the world economy business cycles by the
periodicity:
• the Kondratiev long technological waves of 50 to 60 years. Kondratiev analyzed interest rates
and prices, having noticed that the ascending part of an economic cycle is associated with low
interest rates and rising prices, while downward phase assumes high interest rates and lower
prices. He connected the existence of long wave with changes in capital investments and
technological innovations.
• the Kuznets building cycles of 18 to 22 years. Kuznets explained these waves by demographic
processes, which caused changes in construction intensity or infrastructural investments.
• the Juglar cycles lasting for 7-11 years and associated with fluctuations of investments in
fixed assets.
31
• the Kitchin inventory cycles lasting from 3 to 5 years. These cycles exist due to surplus or
scarcity of goods in warehouse, which appear because of delays in getting business
information and time needed for decision-making (Kuzmenko 2012).
Approximately each half of a century Kitchin, Juglar, Kuznets and Kondratiev economic
cycles simultaneously enter into the downward phase, causing a resonance effect. This period is
usually marked by severe economic and financial crises. Such a situation occurred in the 1870s,
1920s, 1970s and in 2007-2008 (Aivazov 2013).We can assume that these periods are
characterized by the higher number of bankruptcies in the national economy. Due to the lack of
data, we can analyse only the period during and after the last global economic and financial
crisis. This analysis is provided in the next section of this chapter.
Business cycles (or economic cycles) reflect the fluctuations of activity in an economy
and are usually measured by the growth rate of gross domestic product (GDP). Many researchers
proved that GDP growth rates of developed economies and world economy in total demonstrate
cyclical fluctuations, which support the existence of business cycles (Tsirel 2012).
To understand how business cycles emerge in the Russian economy, we analyze the
dynamics of national real GDP growth rate for the XX-XXI centuries (Figure 4). It can be
noticed that this indicator demonstrates cyclical trend, but its direction is not clearly associated
with Kondratiev waves in the world economy. For example, during the descending phase of the
Kondratiev wave in 1914-1946, the average annual rate of global GDP growth fell from 2.57 per
cent to 1.5 per cent, while in Russia this period is marked by the highest rate of economic
development for the whole XXth century (from 1923 to 1940 the average GDP growth rate
20
15
GDP growth rate
10
5
0
-5
-10
-15
1901
1905
1909
1913
1917
1921
1925
1929
1933
1937
1941
1945
1949
1953
1957
1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
2009
2013
-20
Figure 4. Dynamics of the annual real GDP growth rate in Russia in 1900-2015
Source: Simchera (2007); Federal State Statistics Service.
32
Juglar
cycles
Kitchin
cycles
2008
2012
2016
Kuznets
cycles
Year
2020
Kondratiev
cycles
Figure 5. Russian economic cycles
Source: Tyapkina, Mongush and Akimova (2014, p.12).
was 14.25 per cent). Thus, till 1990s cycles of the Russian economy are mainly connected with
change of rulers and methods of economy management. Only after establishment of the market
economy, some relation of economic fluctuations in Russia with the Kondratiev cycle is noticed.
Ascending trend in Russian GDP growth rate from 1993 to 2007 coincides with the upward
phase of Kondratiev cycle in the global economy.
Figure 5 exhibits projected dynamics of business cycles in the Russian economy. Now
Russian economy is in the ascending phase of Juglar investment cycle, moving to the turning
point. Kitchin short-term inventory cycle is in the downward state, but Kuznets cycle should
have positive impact in the current upward phase. As for Kondratiev waves, Russian economy is
in the transition period between descending and ascending stages of long cycles.
Except GDP growth rate, other macroeconomic variables may be considered as indicators
of economic cycles. For example, dynamics of stock market capitalization to GDP and volume
of financial assets to GDP. In USA dynamics of these two indicators is closely related to
ascending and descending Kondratiev waves. In turn, fluctuations on the Russian stock market,
which was relatively recently involved in the world’s financial processes, correspond to external
shocks.
In general, Russian economy is vulnerable to external fluctuations in the world market
conditions. For Russia the world's economic cycles are external factors, and Kondratiev waves
arise endogenously in the national economy. Russian economists conclude that Russian
economic development is exposed to the global Kondratiev waves and corresponds to their
dynamics. However, national internal economic processes support mid-term and short-term
33
business cycles. Detailed analysis of the recent Kitchin and Juglar cycles in Russian economy is
provided by Tyapkina M.F. and others (2014).
For our further analysis we concentrate on Juglar cycles, because this type of business
cycles describes development of macroeconomic environment in the medium term, and that is
why easily observable in the dynamics of macroeconomic indicators.
In the literature on business cycles several macroeconomic variables are highlighted as
business cycle indicators, among which are profits, investments, unemployment rate, money,
credit and interest rates (Zarnowitz 1997).
The National Bureau of Economic Research, American research organization that
monitors economic cycles, suggests the following macroeconomic indicators as determinants of
business cycles:
1) GDP;
2) unemployment rate;
3) industrial production growth rate;
4) consumer price index;
5) investments in fixed assets.
Figure 6 shows the dynamics of abovementioned economic indicators in Russia. It can be
noticed that all indicators, except the unemployment rate, move in the same direction.
Unemployment rate changes in the opposite direction, because it goes up in recession periods.
Gross domestic product
Consumer price index
Industrial production
Fixed assets investments
Unemployment rate
30
25
20
Growth rate
15
10
5
0
-5
-10
-15
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Figure 6. Growth rate dynamics of the main cyclical indicators of
Russian economy development for years 2000-2015
Source: Federal State Statistics Service.
34
GDP growth rate, industrial production, consumer price index growth rate and investments in
fixed assets demonstrate rather simultaneous fluctuations. However, the most significant drop
during the global financial crisis was noticed in the growth rate of fixed assets investments
(Tyapkina, Mongush and Akimova 2014).
There is a distinct medium-term cycle, starting from 2002 and ending in 2009. This is the
evidence of Juglar cycle, driven by fluctuations in fixed assets investments and lasting for 7-8
years. In year 2009 the next medium-term cycle began, and it lasts till now.
The cyclical character of economic development demonstrates different impact on
various
industries.
Economic
decline
dramatically
influences
industries,
producing
manufacturing equipment and durable goods (cars, furniture, and household electronics). The
reason is that in periods of economic crises people tend to postpone purchases of durable goods
in order to save money and spend them on current needs. In this case decrease of demand for
expensive products causes drop in production and employment in the relevant industries.
Each stage of economic cycle significantly influences industries’ state, companies’ level
of output and profitability.
Early-cycle phase. This stage is associated with the economic recovery and lower prices
on factors of production. Production and employment, having reached minimum value during
slump, start to revive. Interest rates go down, creating favorable environment for production
expansion and investments in new enterprises, technologies and equipments. When relatively
cheap credit is readily available, many enterprises tend to increase their personnel, equipment,
inventories (Zarnowitz 1997).
Mid-cycle phase. Usually the longest stage of the business cycle. Production growth rate
is moderate, getting slower than that in the previous phase. New enterprises develop,
unemployment reduces, wages and volume of fixed assets investments increase. Due to fast
expansion of production and demand for credit, interest rates rise. On the firm level, this phase is
marked by growing inventories, sales and profit. Because of high demand, business has many
opportunities to profit and, thus, default probability is low.
Late-cycle phase. The distinctive attribute of this stage is above-trend inflation rate and
restrictive monetary policy. Shrinking credit availability limits investments in fixed assets. On
the firm level, companies’ profit margins and sales growth deteriorate. Economy gradually slips
into recession.
Recession phase. Crisis is accompanied by reduction in economic activity, declining
profits and increasing costs. To launch investments and economic recovery, more favorable
credit conditions are created. During slump, it is more difficult to keep business profitable and
that is why filing for bankruptcy is more likely.
35
Changes on the macroeconomic level
GDP, Industrial production
Unemployment
CPI
Credit availability
Interest rates
EARLY
MID
LATE
RECESSION
+
Economic growth
Sales
Profit
Inventory
Credit
Changes on the company level
Figure 7. Changes in main macro and micro level indicators over the business cycle
Source: based on Emsbo-Mattingly, Hofschire and Betro (2014).
The information about macroeconomic and company indicators dynamics over successive
business cycle phases is summarized in Figure 7.
Among macroeconomic indicators, consumer price index and prime interest rates are
lagging indicators, which reach maximum and minimum points behind business cycle trend
(Loznev 2006).
Thus, there exists relationship between business cycles and corporate performance. On
the company level, three cycles correspond to macroeconomic fluctuations - corporate profit
cycle, credit cycle and inventory cycle (Emsbo-Mattingly, Hofschire and Betro 2014).
36
2.2. Bankruptcy of Russian enterprises: dynamics and sectoral structure
The next question we need to address for our research purpose is whether the number of
bankrupt firms in Russian economy moves counter cyclically: decreases in periods of economic
prosperity and moves up during recession.
First of all, it is necessary to analyze statistic data on bankruptcy dynamics in Russian
economy for the whole period of bankruptcy institute existence (Figure 8).
It can be noticed that peaks in 2002 and 2006 are not associated with macroeconomic
fluctuations. In general, the graph does not show any correlation of bankrupt firms’ number in
the whole Russian economy with Juglar cycle, discussed in the previous section of this chapter.
Now it is curious to understand the reasons for such a discrepancy between theoretical and
practical perspectives. Some authors suggest that the changes in the number of bankruptcies are
rather explained by institutional reasons, than macroeconomic variability (Selevich 2013).
To explain dramatic rise in corporate defaults in 2002 and 2006, we need to briefly
review the development of bankruptcy legislation in Russia. The First Bankruptcy Law was put
into effect in 1992 - the Federal Law No.3929-1-FZ “On Insolvency (Bankruptcy)”. While this
law was in action (1992-1998), default growth rate was very low. In 1998 the Second version of
the law was introduced, changing criteria of bankrupt firms. It loosened barriers to file for
bankruptcy that led to growing number of defaults in 1998-2002. In practice, this period is
characterized by active property redistribution through bankruptcy institute. At that time
bankruptcy institute allowed to quickly, cheaply and reliably change the owner, providing the
legality of process and legitimacy of new owner’s rights. For this reason bankruptcy statistics of
90000
80000
Number of bankrupts
70000
60000
50000
40000
30000
20000
10000
0
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
Figure 8. Number of bankrupt enterprises in Russian economy
Source: Supreme Arbitration Court of the Russian Federation.
37
years 1998-2002 is significantly distorted. Furthermore, in 2002 the adoption of new Bankruptcy
Law, aiming to impede property redistribution schemes, was expected. That is why all
bankruptcy processes were intensified in order to accomplish bankruptcy procedure according to
the less strict Second Law.
For the sharp increase of bankrupt companies in 2006, Selevich O.S. (2013) suggests the
following explanation.
Until 2004 bankruptcy procedures were conducted by the Federal
Bankruptcy Service, which had no interest and money to bankrupt absent debtors. Appointed to
the authorized body for bankruptcy in 2005, the Ministry for Taxes and Levies (currently the
Federal Taxation Service) was more interested in absent debtors’ bankruptcy, because it had an
opportunity to get uncollected taxes. Moreover, additional budget funds were allocated for
bankruptcy procedures (in 2006 — 964 mln. Rub, in 2007 — 2.5 bln. Rub). Active work of the
Federal Taxation Service in 2006 resulted in a high number of firms declared bankrupts. In the
next years bankruptcy procedure became more expensive, and the number of absent bankrupt
companies decreased.
Therefore, the period from 1992 to 2006 is a developing stage of bankruptcy institute in
Russia, characterized by institutional changes. For the reasons mentioned above, bankruptcy
statistic data for this period cannot be reliable and does not show real number of companies,
filing for bankruptcy because of poor financial state. Only starting from 2007 figures probably
reflect the real bankruptcy level, because institutional reasons are no more dominant.
Excluding the most distracting values from the trend (2002 and 2006), we get smoother
figures and conclude that the current level of 14-20 thousand bankruptcies per year (or about
0.5% of the total number of registered enterprises) is normal for the economy (Selevich 2013).
1200
Number of bankrupts
1000
800
600
400
200
0
2007
2008
2009
2010
2011
2012
2013
2014
Figure 9. Bankruptcy dynamics in the real sector of Russian economy in 2007-2014
Source: Mogilat (2015, p.160).
38
As aggregate bankruptcy dynamics does not demonstrate obvious link with
macroeconomic fluctuations, further we consider only the real sector of Russian economy
(Figure 9).
For our analysis under the real sector of economy we assume the following industries:
industrial sector (includes extractive industry, electrical energy industry and manufacturing
industry), fishing industry and agricultural sector.
Figure 9 shows the number of enterprises in the real sector of Russian economy, for
which supervision (the first stage of bankruptcy proceedings) or simplified bankruptcy procedure
is opened.
Statistic data of Russian enterprises show increasing number of companies filing for
bankruptcy. This is a direct consequence of worsening economic state, slowdown in economic
growth in Russia and high volatility on the global markets (Matrosova 2015).
The period from 2007 to 2014 is very heterogeneous, considering the number of
bankruptcies in Russian real sector. A sharp rise in the number of bankrupt companies after the
global financial crisis is observable.The whole period from 2007 can be divided into three stages:
1) pre-crisis period (2007-2008): due to favorable economic conditions the number of
bankruptcies did not exceed 250 enterprises annually;
2) crisis “splash”: sharp increase of number of bankruptcies - for 2009-2010 almost in 4.5 times
in comparison with the pre-crisis period;
3) slight decrease in 2011 and the subsequent steady growth, significantly accelerated in 2014
(the average number of bankruptcies in 2011-2014 was about 800 enterprises annually). The
number of bankruptcies in 2014 even exceeded the peak level of 2010 (a consequence of
crisis 2008-2009).
Noticeable trend for the period under consideration is that the number of companies,
conducting simplified bankruptcy procedure, is growing. The main characteristic of simplified
bankruptcy procedure is that it omits some essential stages of ordinary bankruptcy proceedings “supervision”, “financial sanation”, and “external management”- and moves strait to the
“receivership” stage. That means that there is no possibility to restore the solvency for the
company under simplified bankruptcy procedure. The tendency of increasing number of
companies, conducting simplified bankruptcy procedure, is potentially driven by two factors:
• systematically growing number of companies regarded as “hopeless bankrupts” (in fact, these
enterprises has already ceased their business activity);
• increasing number of legal entities, which started the liquidation procedure but, due to the
insufficient assets to fulfill existing commitments, switched to bankruptcy proceedings
(Mogilat 2015).
39
Manufacturing industry
Extractive industry
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
12%
15%
20%
49%
2007
8%
14%
Electrical energy industry
Fishing industry
6%
11%
6%
14%
14%
10%
18%
56%
2008
68%
68%
2009
2010
Agricultural sector
5%
5%
3%
6%
21%
17%
24%
18%
14%
17%
14%
14%
59%
60%
59%
61%
2011
2012
2013
2014
Figure 10. Sectoral structure of bankrupt companies
in the real sector of Russian economy in 2007-2014
Source: Matrosova (2015, p.88).
These tendencies indicate unstable state of economy and sensitivity of real sector to
macroeconomic environment.
Now we consider the sectoral structure of the Russian real sector (Figure 10). As shows
statistics, for years 2007-2014 sectoral structure of bankrupt enterprises in Russian real sector
is
rather
stable, although the crisis of 2008-2009 had some impact. Bankruptcies in
manufacturing industry prevail - their share is 60% on average from the whole number of
bankrupt companies in the real sector. It can be noticed that the share of manufacturing
enterprises bankruptcies increased from 49% in 2007 to 68% in 2010. At the same time decline
in share of defaults in such industries as electrical energy, extractive and fishing sectors is
observable.
As manufacturing industry dominates in the number of bankruptcies, we consider it in
more detail. The average bankruptcy rate in manufacturing industry was 1.6% in 2009-2010,
1.3% in 2011-2012 and 1.5% in 2013-2014. Despite the fact that generally manufacturing
industry has relatively low bankruptcy level, in the context of certain subsectors the situation is
more alarming. Manufacturing subsectors with the highest intensity of bankruptcies include food
industry (2.4%), wood-processing industry (2.3%), production of other non-metallic mineral
products (1.7%) and machinery manufacturing (1.3%) (Mogilat 2015). In addition, these four
subsectors account for more than 60% of the total number of bankruptcies of manufacturing
enterprises (Matrosova 2015).
Main factors, which impede production growth in manufacturing sector and influence
bankruptcy rate, include macroeconomic factors such as lack of demand in the domestic market
40
and economic uncertainty (Federal State Statistics Service). Other factors are high tax rates, lack
of financial resources and high loan interest rates. Thus, manufacturing industry is very
vulnerable to macroeconomic situation. That is why in the third chapter for our further analysis
we focus on companies from this economic sector.
2.3. Approaches for bankruptcy risk estimation:
models with macroeconomic variables
As shows statistics, it is very likely that not only internal factors, reflected in financial
ratios, but also external factors influence the bankruptcy probability. Apart from the management
problems and other firm specific issues that would cause a loss in its profitability, changes in
market and economic conditions (such as changes in interest rates, GDP growth, exchange rate,
unemployment rates, and industry specific shocks, etc) may affect the overall profitability of the
enterprise.
Business cycles have great impact on the profitability of individual firms. Therefore, they
influence the risk profile of a given company or industry. The incorporation of variables that
capture changes in macroeconomic environment is important, because such variables add a
dynamic component that adjusts probability of insolvency in relation to changing
macroeconomic conditions.
There exists small but growing number of studies investigating the importance of
business cycle variables for corporate default probability. However, due to the last global
economic crisis in 2008-2009 this field of research is currently of high interest both in the world
and in Russia.
The relationship between macroeconomic factors and the probability of default on an
industrial level was analyzed by Qu Y. (2008). He showed that in European countries changes in
macro factors such as industrial production, interest rate spread and exchange rate influence the
probability of default on the industry level. However, the impact of different macroeconomic
variables varies. In general, Qu Y. concluded that exchange rate demonstrates higher importance
for the level of default rate than other macroeconomic factors.
If talking about the exchange rate, its effect on the probability of default depends highly
on industries. Indeed, when the exchange rate goes up, importing becomes more expensive,
exporting becomes easier, and then fewer competitors in an international arena will result in a
decrease of default of national companies.
41
One more important finding is that sensitivity to macroeconomic fluctuations varies with
the quality of the company. The better the company is, the less its probability of default will vary
with the macro factors changes (Qu 2008).
One of the ways to approach the effect that macro economy has on the default probability
is to analyze it directly from the relationship between business cycle and individual firms. Now
we refer to the recent studies devoted to the relationship between macroeconomic fluctuations
and corporate defaults.
Hernandez Tinoco and Wilson
Using a sample of listed companies in United Kingdom during the period 1980-2011,
Hernandez Tinoco M. and Wilson N. (2013) combined accounting, market-based and macroeconomic data to explain corporate financial distress. The final model included three types of
variables:
1) Accounting ratios: Total funds from operations / Total liabilities, Total liabilities / Total
assets, No credit interval, Interest coverage ratio.
2) Macroeconomic variables: the Retail Price Index (measure of consumer inflation in UK) and
the real short term Treasury bill rate.
3) Market variables: equity price, past stock excess returns, market size of the company
(company’s market capitalization relative to the total market capitalization), market
capitalization to total debt.
The study showed that incorporation of all three types of explanatory variables provides
the highest bankruptcy prediction accuracy of 91.9% (Hernandez Tinoco and Wilson 2013).
Giordani et al.
Modeling default risk, Giordani and others (2014) argue that “financial ratios remain the
important information source” in case of private companies. At the same time, macroeconomic
variables clearly contribute to the improvement of predictive accuracy.
The research was based on the data of all incorporated Swedish businesses over the
period 1991-2008. The following explanatory variables were chosen:
1) Macroeconomic variables: annual gross domestic product (GDP) growth rate, repo rate (a
short-term interest rate set by the Central Bank of Sweden);
2) Accounting variables: Total liabilities to total assets, EBIT to total assets, Cash and liquid
assets to total liabilities;
3) Control variables: firm size, firm age.
The main finding of the research was nonlinear relationship between firm’s bankruptcy
and leverage, liquidity and profitability. For example, threshold effect for the relation between
42
the leverage ratio and default probability was observed. One conclusion of the research is that
debt level demonstrates moderate impact on default risk, when leverage ratio is within 30–60 per
cent region. However, the bankruptcy risk increases fourfold for leverage values of 60–100 per
cent. Other observation is connected with relationship between EBIT to total assets ratio and
bankruptcy risk. The bankruptcy risk decreases until the earnings ratio is less than 15 per cent
and increases thereafter. Firms reporting earnings ratios above 15 per cent are associated with
higher failure risk, and authors find evidence suggesting that this is driven by high cash-flow risk
in combination with limited and costly external financing.
To take into account these nonlinear relationships, researchers applied new technique –
they introduced spline functions into a logistic regression. As they claim, this approach improved
the quality of default forecasting (Giordani, et al. 2014).
Haydarshina
Russian
analog
of
bankruptcy
prediction
logit-model
with
incorporation
of
macroeconomic variable was developed by Haydarshina G.A. (2009). The research sample
consisted of 350 companies from three different industries – trade, agricultural and
manufacturing sectors. The author intentionally based the model on enterprises with various
characteristics – company’s size, annual revenue and operating sector. The model is represented
by the logarithm function with eleven explanatory variables, among which accounting ratios are
still core elements:
1) Macroeconomic variables: refinancing rate of the Central Bank of Russian Federation;
2) Accounting variables: Return on company’s assets, Growth rate of assets, Return on equity,
Growth rate of equity, logarithm of equity value, EBIT / Interest expenses, Current ratio;
3) Binary variables: “age” of the company, credit history, region.
The model allows taking into account the most important aspects of bankruptcy risk
assessment, which include the macroeconomic situation in the country, efficiency, liquidity,
financial stability of the enterprise, as well as level of company’s activity and industry specifics.
These factors characterize company’s business activity from different perspectives that facilitates
comprehensive assessment of bankruptcy risk. According to the author, model’s accuracy in
bankruptcy risk assessment was 85.6 per cent.
However, inclusion of eleven explanatory variables makes the model cumbersome. In
world practice, the optimum number of the indicators is from five to seven (Haydarshina 2009).
Totmyanina
One more example of Russian model with macroeconomic variables was developed by
Totmyanina K. M. (2014) for Russian construction industry.
43
Choosing the macroeconomic variables for the model, Totmyanina states that the
combination of this type of indicators may differ depending on the country and time period under
observation. The author made a list of macroeconomic variables that account for business cycle
phase and are potentially valuable for bankruptcy risk valuation:
• GDP indicators: nominal and real GDP, investments in fixed assets, volumes of export /
import, consumption;
• Foreign exchange market: bi-currency basket, exchange rates of main currencies;
• Money market and banking sector: money supply, volume of loans to nonfinancial sector,
loans to individuals, volume of loans to GDP;
• Price level: Consumer price index, Producer price index, GDP deflator;
• Other indicators: oil price, unemployment rate, capital inflow / outflow, government
expenditures.
The following five indicators were selected as having the strongest relation to default rate
in Russian construction industry:
• Oil price;
• Export of goods and services;
• Import of goods and services;
• Unemployment rate;
• Loans to individuals.
The model with incorporated variable Import to GDP demonstrated the highest
explanatory ability basing on determination coefficient (Totmyanina 2014).
Jacobson, Linde and Roszbach
In a recent paper, Jacobson, Linde, and Roszbach (2013) analyze the impact of
macroeconomic factors on corporate bankruptcy risk.
The authors estimate multiperiod logistic regressions on firm-level default data. The
model is estimated for Swedish businesses from ten industries covering years 1990–2009. In
addition to an extensive set of financial statement variables, four standard macroeconomic
variables are included:
• output gap (i.e., the deviation of GDP from its trend value);
• yearly inflation rate;
• repo interest rate (a short-term nominal interest rate);
• real exchange rate.
The researchers compared indicators from different industries and concluded that the
influence of the macroeconomic factors appears to be more important in industries that are more
44
cyclical. For instance, the output gap is more significant in the construction and in the real estate
sectors in comparison with other industries, while the nominal interest rate is very important for
the financial services and the real estate sectors. Inflation and the real exchange rate in general
demonstrated weaker relation to bankruptcy risk. In turn, depreciating real exchange rate (i.e., a
higher value for the variable) is connected with a significantly lower bankruptcy probability in
the manufacturing sector, which is the most export-oriented industry in Sweden.
The model with microeconomic variables works both for listed and privately held firms.
This is important because privately held businesses typically account for over half of GDP in
developed economies.
Suggested model at the aggregate level is very effective and accurate in explaining the
extreme default frequencies observed during the Swedish banking crisis of the early 1990s as
well as the considerably lower default frequencies in the late 1990s. Thus, macroeconomic
fluctuations play an important role in understanding the absolute level of firm default risk. The
results show that “macroeconomic factors shift the mean of the default risk distribution over
time” and thus are significant determinants of fluctuations in the average level of corporate
default (Jacobson, Lindé and Roszbach 2013).
Nam et al
The sample used in this empirical study consists of 367 companies listed on the Korea
Stock Exchange over a period from 1991 to 2000.
The novelty of the research was in specification of the baseline hazard rate with
macroeconomic variables. To directly estimate the baseline hazard rate with macroeconomic
variables, the authors examined two macroeconomic indices: volatility of foreign exchange rate
and change in interest rates.
From a theoretical perspective, all macroeconomic indices reflecting the market condition
might directly or indirectly affect each firm’s hazard rate. Moreover, the mechanisms affecting
the firm’s hazard rate are considerably diverse because the intensity or time lag of certain
economic shocks differs. Intuitively, the macroeconomic variables, highly correlated with hazard
rate, can be regarded as variable with the highest explanatory power.
Both variables, volatility of foreign exchange rate and change in interest rates, show a
pattern very similar to the change of average unconditional hazard rates for all firms. Because
the two macroeconomic variables that share a similar pattern have a high degree of collinearity, a
serious multicollinearity problem can occur if the model includes both of those variables. The
volatility of foreign exchange rate was chosen as the main explanatory macro-variable
considering the high currency exposure of the Korean economy (Nam, et al. 2008).
45
Table 2
Totmyanina
(2014)
Haydarshina
(2009)
Linde, and
Roszbach (2013)
Giordani,
Jacobson et al.
(2014)
Hernandez
Tinoco, Wilson
(2013)
Macroeconomic variables
Qu Y.
(2008)
Nam et al
(2008)
Macroeconomic variables used in foreign and Russian bankruptcy estimation models
GDP indicators
GDP growth rate
+
Output gap
+
Industrial production index
+
Interest rate indicators
Short-term government bonds rate
+
Change in interest rates
+
Interest rate spread
+
Repo interest rate
+
+
Refinancing rate
+
Inflation indicators
Consumer price index
+
GDP deflator
+
Exchange rate indicators
Exchange rate
+
Real exchange rate
+
Volatility of foreign exchange rate
+
Export/Import indicators
Import of goods and services to GDP
+
Export of goods and services to GDP
+
Other macroeconomic indicators
Oil price
+
Unemployment rate
+
Loans to individuals
+2
Source: The present study.
2
Totmyanina created several models. Each model included only one macroeconomic variable because of the strong
correlation between variables. The best result showed the model with import of goods and services to GDP.
46
From the business cycles theory, for our further research the following cyclical indicators
should be examined for significance in bankruptcy risk estimation:
• GDP and industrial production;
• Consumer price index;
• Interest rates;
• Investments in fixed assets;
• Unemployment rate.
The main indicator of medium term Juglar cycles, which we take as the basis for our
further research, is growth rate of investments in fixed assets. However, other abovementioned
macroeconomic indicators also demonstrate medium term fluctuations associated with Juglar
cycles, and that is why should be tested for significance in corporate bankruptcy estimation.
Then we have compared existing approaches for bankruptcy risk estimation.
Macroeconomic variables, which were used in the previously discussed bankruptcy risk models,
are summarized in Table 2.
Currently there is no general approach for macroeconomic variables selection, and we
can notice that each model incorporates its unique macro indicators.
Foreign models include wider range of macroeconomic variables than Russian models. In
Russian theory and practice the research field of macroeconomic influence on bankruptcy
probability is relatively new and only recently started developing. The reason for this is that
Russian bankruptcy institute is young, and earlier insufficient bankruptcy statistics was available.
Having analyzed existing foreign and Russian bankruptcy prediction models with
macroeconomic variables, it is possible to highlight the most informative and potentially
effective macro indicators for bankruptcy risk analysis.
First of all, the group of interest rate indicators showed significance in both national and
foreign approaches. However, the particular variables taken for prior modeling are various;
among them are real short-term government bonds interest rate, repo and refinancing interest
rates and interest rate spread. All these different types of interest rates will be tested for
significance in the next chapter of the current study.
GDP, inflation and exchange rate indicators till now were omitted in national models,
while incorporated in foreign models. At the same time, one Russian model includes such unique
macro variables as export, import, oil price, unemployment rate and loans to individuals. The
indicators from abovementioned macroeconomic categories will also be checked for importance
in corporate bankruptcy estimation.
47
CHAPTER 3. DETERMINATION OF BANKRUPTCY RISK FACTORS
The general idea of the current research is to analyze the relationship between corporate
bankruptcy risk and business cycle indicators, which reflect macroeconomic fluctuations, in
order to find out how macro factors contribute to explain the probability of bankruptcy on the
firm level. The study focuses on Russian large and medium private companies of manufacturing
industry.
We start with detailed description of data taken for the modeling. Financial statements
information is obtained from Spark-Interfax database. Macroeconomic data was gathered from
several sources - Federal State Statistics Service, the World Bank and Central Bank of Russian
Federation databases.
3.1. Data description
Conducted research is based on financial data of bankrupt and non-bankrupt companies
from Russian manufacturing industry.
First of all, it is necessary to clearly define what we assume under the bankrupt
enterprise. For the purpose of the current research a firm is defined to have bankrupt status if it is
legally declared bankrupt, and the receivership procedure was introduced concerning this
company. Receivership is the last stage of bankruptcy proceedings, meaning that there is no
more chance for financial recovery for the company. All prior bankruptcy stages (“supervision”,
“financial sanation” and “external management”) allow for restoring of company’s solvency by
implementation of special measures under external manager supervision.
Enterprises, which undertook other ways to terminate business activity (merging with
other enterprise or liquidation of business) or restored their solvency, were excluded from the
current research.
The year of bankruptcy is deemed to be the year, in which the company was declared
bankrupt by court of arbitration, and the receivership was introduced concerning this company.
Information about the year of bankruptcy is taken from Spark-Interfax and Federal Register on
Bankruptcy information.
For the bankruptcy prediction modeling balance sheet and income statement data on
bankrupt and operating enterprises was taken. All financial data used in the current analysis are
got from financial statements prepared according to Russian Accounting Standards. The final
data sample consists of 1000 firms – 250 bankrupts and 750 non-bankrupts.
48
Table 3
Number of bankrupt and healthy companies in the research sample by year
Year
Number of bankrupt companies
Number of healthy companies
2007
5
15
2008
24
72
2009
55
165
2010
78
233
2011
43
129
2012
22
66
2013
9
26
2014
14
44
Total
250
750
Source: The present study.
Very few companies provided financial data on and after the year of bankruptcy. That is
why our analysis concentrates on financial ratios one, two and three years prior to bankruptcy.
All companies, selected for the research, operate or operated in Russian manufacturing industry,
which includes the following subsectors: food products including beverages; tobacco products;
textile industry; wearing apparel; leather, leather products and footwear; wood processing,
products from wood and cork; production of pulp and paper; publishing and printing; chemical
industry; production of coke, oil products and nuclear materials; rubber and plastic articles;
production of other non-metallic mineral products; metallurgical production; production of
finished metal products; machinery manufacturing; office facilities and computer machines;
electrical machinery and equipment; manufacture of electronic components, radio, television and
communication equipment; automobiles, trailers and semi-trailers production; production of
crafts, aircrafts, spacecrafts and other vehicles; production of furniture; recycling of secondary
raw materials. Sample structure by subsectors is provided in Appendix 1.
Data includes non-operating companies, which were declared bankrupts during the period
from 2007 to 2014. Number of healthy companies is proportional to bankrupt companies for
each particular year. Additionally, for each company financial data three, two and one year prior
to estimation period was collected. Thus, research sample includes data for the period 20042014.
The next criterion for sample selection was private ownership; government-owned
enterprises were excluded from the sample. Government enterprises are not taken into account,
49
because they may receive additional financial support or benefits, which can disturb final results
of the model.
One more restriction is connected with company’s size – at least 250 employees should
work in the company. This filter excludes micro and small enterprises from the sample, because
they have specific factors influencing their financial risks. Thus, the current research focuses
only on medium and large enterprises.
In addition, we put additional constraint on the period company operates in the market.
Final sample includes only companies existing at least ten years in the market, because newly
established firms have additional risks and demonstrate higher probability of insolvency and
bankruptcy.
Thus, for company’s selection the following criteria were applied:
- company operates or operated in Russian manufacturing industry;
- privately held company;
- average annual number of employees is not less than 250;
- company’s age is at least 10 years. Maximum company’s age in the sample is 23.
Dependent variable
The dependent variable bankrupt is a binary variable. It equals 1 if the company filed for
bankruptcy in a particular year and 0 if not.
Further we describe independent variables that were considered during the research. Two
types of explanatory variables are taken for the current study: financial ratios and
macroeconomic variables. Both financial and macroeconomic variables are taken on the annual
basis.
Independent variables: Financial ratios
Having analyzed existing models for bankruptcy risk estimation, the following groups of
financial ratios were selected in the first chapter of the paper to be tested for significance in
bankruptcy risk prediction:
• Financial structure indicators:
1) Total debt / Total assets;
2) Equity / Total liabilities;
3) Total liabilities / Total assets;
4) Current liabilities / Total assets;
• Solvency and liquidity indicators:
5) Current assets / Current liabilities;
6) Current assets / Total liabilities;
50
7) Working capital / Total assets;
8) EBIT / Current liabilities;
9) EBIT / Interest expenses;
• Operational efficiency ratios:
10) Revenue / Total assets;
• Profitability ratios:
11) Net income / Revenue;
12) Retained earnings / Total assets;
13) EBIT / Total assets;
14) Net income / Total assets;
15) Net income / Equity;
• Assets structure:
16) Share of current assets in total assets;
17) Share of noncurrent assets in total assets;
• Control variables:
18) Company’s size variable = ln (Total assets / GDP deflator);
19) Company’s age.
For further analysis, from asset structure variables we choose Share of current assets in
total assets. Firstly, this indicator is a determinant of Kitchin inventory cycle. Secondly,
variables Share of current assets in total assets and Share of noncurrent assets in total assets are
perfectly correlated as they express the same balance sheet relation and cannot be incorporated
into the model together.
Thus, 18 variables, including 16 financial ratios and two control variables – firm’s size
and age, are selected as potential bankruptcy predictors. Among these financial indicators we
will pick several variables with the highest forecasting ability.
The next issue we encountered with is the absence of Earnings before Interest and Taxes
(EBIT) indicator in financial statements prepared according to Russian Accounting Standards.
Thus, we needed to derive the Russian analogue of EBIT based on International Financial
Reporting Standards.
Calculation of EBIT in Russian practice is based on such items as income tax
reimbursement, extraordinary income/expenses and interest paid/received. Due to the lack of
financial data, we take the following approximation for calculation of EBIT:
EBIT= Earnings before Taxes + Interest paid – Interest received
(5)
51
Many abovementioned models of bankruptcy risk estimation include company size
variable. Various models suggest two main approaches for calculation of company size: through
logarithm of tangible assets or logarithm of total assets. In the current research we calculate size
of a firm as logarithm of total assets adjusted for inflation. Firm size and age are generally
associated with less volatile income and cash flows, and thus lower default probability.
Independent variables: Macroeconomic indicators
The main concern of the current research is to assess contribution of macroeconomic
factors to bankruptcy risk level. For this purpose several macroeconomic indicators will be tested
for explanatory ability. Partly, these indicators selected basing on the previous approaches for
default probability estimation. Other macro indicators are taken from the business cycle theories,
which were discussed in detail in the previous chapter.
Several groups of macroeconomic indicators are deemed to be business cycle
determinants:
• GDP indicators;
• Inflation indicators;
• Interest rate indicators;
• Other macroeconomic indicators.
In the current study we consider several macro variables in each abovementioned group
as potentially significant factors of corporate failure probability:
• Gross Domestic Product indicators:
1) Real GDP growth rate;
2) Industrial production index growth rate;
• Inflation indicators:
3) Producer prices index in manufacturing industry;
4) Consumer price index;
5) GDP deflator;
• Interest rate indicators:
6) Repo interest rate;
7) Refinancing interest rate;
8) Moscow prime offered rate (six month rate);
9) Short-term interest rate on federal loan bonds;
10) Long-term interest rate on federal loan bonds;
11) Interest rate spread;
• Other macroeconomic indicators:
52
12) Investments in fixed assets in manufacturing industry;
13) Oil prices;
14) Unemployment rate;
15) Money supply M2 growth;
16) Effective exchange rate.
All in all, 16 macroeconomic variables are selected for testing in the model.
3.2. Applied methodology
Many Russian researchers agree that logistic regression for bankruptcy prediction has
shown significant efficiency in foreign countries, and it can be assumed that the use of the same
technique on the sample of Russian companies will have high predictive potential.
Logistic regression is given by the formula:
𝑃=
where P is the probability of bankruptcy, and
1
1 + 𝑒 −𝑌
𝑌 = 𝑎0 + 𝑎1 𝑋1+ 𝑎2 𝑋2 + ⋯ 𝑎𝑛 𝑋𝑛
(6)
(7)
This implicates that the probability for a firm to go bankrupt in a certain year is given by
the logistic distribution function which argument is a linear function of a constant, several
financial explanatory variables, control variables of firm characteristics and macroeconomic
indicators.
Logistic regression model has some very important advantages, which support
application of this kind of model for bankruptcy prediction purposes:
• the logit-model assumes nonlinear relationships between factors, that is the case of
bankruptcy factors;
• the logit-model does not require normal distribution of variables. In practice, financial
indicators of insolvent firms are rare normally distributed. According to Shapiro-Wilk test,
none of the financial variables taken for the current analysis is normally distributed;
• the logit model is easily interpreted, because it can take values from 0 to 1 and determines the
nominal value of the probability of bankruptcy;
• there is no “grey areas” as in discriminant analysis models.
According to the suggested methodology, P value intervals are associated with different
bankruptcy risk levels (Table below).
53
Table 4
P value intervals and characteristics of bankruptcy risk in logistic model
P value intervals
Characteristics of bankruptcy risk
0 < P < 0.2
Minimum risk
0.2 < P < 0.4
Low risk
0.4 < P < 0.6
Moderate risk
0.6 < P < 0.8
High risk
0.8 < P < 1
Critical risk level
Source: The present study.
3.3. Descriptive statistics
As our research includes data on two separate types of enterprises – bankrupt and
operating companies – we provide descriptive statistics for these groups separately. Tables 5 and
6 show mean, minimum and maximum values of financial ratios in both bankrupt and healthy
firms one year before estimation. Here significant differences in mean values can be noticed.
Unlike healthy companies, bankrupts demonstrate mostly negative values of solvency,
liquidity and profitability ratios.
As insolvent companies actively accumulate debt, the huge discrepancy is noticed in
leverage ratios. Debt to total assets ratio is on average 2.7 times higher in non-operating
companies than that of healthy companies. Total liabilities related to total assets increase on
average more than three times in case of bankrupt enterprises, while the most significant rise is
observed in short-term liabilities. Mean value of equity to total liabilities ratio is negative for
bankrupts and highly positive in case of operating companies.
Among solvency and liquidity indicators the greatest difference is noticed in interest
coverage ratio (- 39.96 in bankrupt vs. 181.46 in operating companies). On average current
assets are more than twice higher in relation to total and current liabilities in healthy enterprises.
For default companies these indicators drop to 0.797 and 0.567 respectively. Mean value of
EBIT / Current liabilities is negative in case of non-operating companies and positive for healthy
enterprises.
When it comes to operational efficiency ratios, asset turnover ratio (Revenue / Total
assets) is 1.3 times higher for healthy firms than that for bankrupts.
54
Table 5
Descriptive statistics on bankrupt enterprises
Variable
Obs.
Mean
Std. Dev.
Financial structure indicators
Total debt / Total assets
Equity / Total liabilities
Total liabilities / Total assets
Current liabilities / Total assets
250
248
249
250
0.5352
-0.1821
1.6617
1.4663
Min
Max
0
-0.9870
0.1996
0
9.9569
2.0598
8.4153
8.4153
0.0111
0.0111
-7.8519
-4.3163
-793.1000
7.8183
1.7995
1
1.4222
3.5127
2.0131
0
9.7298
7.9305
1.3941
0.6726
0.7403
2.2454
-58.0693
-7.8149
-3.9298
-4.9276
-8.9307
0.0837
0.3447
0.3907
0.1024
8.5685
0.2560
0.0315
1
0.9789
0.4002
1.2976
1.2011
Solvency and liquidity indicators
Current assets / Current liabilities
Current assets / Total liabilities
Working capital / Total assets
EBIT / Current liabilities
EBIT / Interest expenses
249
248
250
249
152
0.7969
0.5666
-0.7931
-0.2333
-39.9581
0.9539
0.3230
1.2999
0.3944
115.3152
Operational efficiency ratios
Revenue / Total assets
250
Net income / Revenue
Retained earnings / Total assets
EBIT / Total assets
Net income / Total assets
Net income / Equity
245
250
250
249
249
Current assets / Total assets
250
1.4683
Profitability ratios
-2.2917
-0.8663
-0.4020
-0.4444
0.4066
Assets structure
0.6972
Source: The present study.
Table 6
Descriptive statistics on healthy enterprises
Variable
Obs.
Mean
Std. Dev.
Financial structure indicators
Total debt / Total assets
Equity / Total liabilities
Total liabilities / Total assets
Current liabilities / Total assets
750
750
750
750
0.1970
2.5746
0.4989
0.3893
Min
Max
0
-0.2293
0.0222
0.0222
1.0434
27.5471
1.2725
1.1541
0.0607153
0.058668
-0.6887
-0.8673
-4.5178
34.3120
34.3120
1.1079
9.8746
8577.041
1.237
0.0928
9.2822
0.0851
0.2902
0.1407
0.1111
0.7127
-0.7903
-0.3839
-0.2682
-0.1584
-1.7610
0.4464
1.0624
1.2454
0.6919
7.8380
0.2035
0.0521
1.1839
0.2319
4.0226
0.2836
0.2491
Solvency and liquidity indicators
Current assets / Current liabilities
Current assets / Total liabilities
Working capital / Total assets
EBIT / Current liabilities
EBIT / Interest expenses
750
750
750
748
527
2.8497
2.2485
0.2439
0.8433
181.4627
3.4804
3.0129
0.2623
1.3766
858.0081
Operational efficiency ratios
Revenue / Total assets
750
Net income / Revenue
Retained earnings / Total assets
EBIT / Total assets
Net income / Total assets
Net income / Equity
750
750
748
750
750
Current assets / Total assets
750
1.9004
Profitability ratios
0.0549
0.3639
0.1467
0.0927
0.2447
Assets structure
0.6329
Source: The present study.
55
Average profitability indicators are negative in case of default companies. The only
exception is Return on equity (Net income to Equity), which demonstrates positive mean value
because mostly both net income and equity have negative values. The largest discrepancy is
noticed in profit margin (Net income / Revenue), which is deeply negative for default companies
(-2.29) and higher than zero for operating enterprises (0.055). Other significant indicator, which
varies between two types of companies, is Retained earnings to total assets. Descriptive statistics
shows that in our research sample Retained earnings of healthy enterprises comprises 36 per cent
of total assets, EBIT – around 15 per cent, Net income – 9 per cent of total assets. All three
indicators are negative in case of bankrupts: - 0.87, - 0.4, and -0.44 respectively.
On average, the share of current assets in total assets is higher for insolvent enterprises
than for operating companies.
When building models, we need to consider correlation between explanatory factors.
Correlation matrix of all financial ratios under consideration is presented below (Table 7).
Table 7
Correlation matrix of financial variables
DebtTA_1
ETL_1
TLTA_1
CLTA_1
CACL_1
CATL_1
WCTA_1
EBTCL_1
EBITInt_ex~1
RevTA_1
NIRev_1
RETA_1
EBITTA_1
NITA_1
NIE_1
CATA_1
EBITInt_ex~1
RevTA_1
NIRev_1
RETA_1
EBITTA_1
NITA_1
NIE_1
CATA_1
DebtTA_1
ETL_1
TLTA_1
CLTA_1
CACL_1
CATL_1
WCTA_1
EBTCL_1
1.0000
-0.2727
0.7178
0.6180
-0.1675
-0.2595
-0.6416
-0.2406
-0.1198
-0.1728
-0.2558
-0.7372
-0.6130
-0.6499
-0.0124
-0.0871
1.0000
-0.4446
-0.3575
0.7517
0.9341
0.3305
0.7021
0.2422
0.0091
0.0718
0.3940
0.2750
0.2661
-0.0329
-0.0741
1.0000
0.9397
-0.3621
-0.3959
-0.9154
-0.4046
-0.1698
-0.1139
-0.2695
-0.9616
-0.7923
-0.7804
0.0889
0.0543
1.0000
-0.3870
-0.3033
-0.9465
-0.3438
-0.1344
-0.0454
-0.2605
-0.9150
-0.7737
-0.7645
0.0666
0.1419
1.0000
0.7835
0.4126
0.6123
0.2110
-0.0183
0.0667
0.3548
0.2481
0.2396
-0.0172
0.0878
1.0000
0.3501
0.6086
0.2454
0.0446
0.0693
0.3756
0.2555
0.2459
-0.0234
0.1507
1.0000
0.3258
0.1443
0.1034
0.2553
0.9143
0.7752
0.7611
-0.0506
0.1846
1.0000
0.2991
0.0779
0.0916
0.3937
0.4620
0.4049
0.0171
-0.0471
EBITIn~1
RevTA_1
NIRev_1
RETA_1 EBITTA_1
NITA_1
NIE_1
CATA_1
1.0000
0.0496
0.0268
0.1538
0.1558
0.1368
0.0186
0.0333
1.0000
0.1254
0.1531
0.1415
0.1019
0.1061
0.1805
1.0000
0.2888
0.2610
0.2618
-0.0045
-0.0108
1.0000
-0.0897
0.0047
1.0000
0.0468
1.0000
1.0000
0.8274
0.8165
-0.0578
0.0167
1.0000
0.9502
-0.0663
0.0205
Source: The present study.
56
Bankrupt companies
3 years before bankruptcy
2 years before bankruptcy
1 year before bankruptcy
2
1,5
1
0,5
0
-0,5
-1
-1,5
-2
-2,5
Total liabilities Current assets Equity / Total
Working
Net income /
/ Total assets
/ Total
liabilities
capital / Total Total assets
liabilities
assets
Retained
earnings /
Total assets
Net income /
Revenue
Operating companies
3 years before estimation
2 years before estimation
1 year before estimation
3
2,5
2
1,5
1
0,5
0
Total liabilitiesCurrent assets / Equity / Total
Working
Net income /
/ Total assets Total liabilities liabilities
capital / Total Total assets
assets
Retained
earnings /
Total assets
Net income /
Revenue
Figure 11. Dynamics of mean values of financial indicators for operating
and bankrupt enterprises during 2007-2014
Source: The present study.
57
Correlation matrix demonstrates that variables within one group (financial structure,
solvency and liquidity, and profitability ratios) are, in general, highly correlated. For modeling
only variables with low and moderate correlation coefficients may be incorporated together in
one equation. Under factors with low and moderate correlation we assume those variables that
have correlation coefficients of more than -0.5 and lower than 0.5. Further each considered
model will be tested for factors correlation.
Correlation matrix of macroeconomic variables can be found in Appendix 2. In general,
many macroeconomic indicators are highly correlated with each other. For example, GDP
growth rate and investments in fixed assets, inflation indicators and interest rates. To avoid
multicollinearity problem, further we include business cycle variables one at a time into logistic
regression model.
Figure 11 exhibits seven the most varying variables between bankrupt and non-bankrupt
enterprises. Financial ratios are taken for one, two and three years before bankruptcy or
estimation period. Dynamics of the financial ratios shows that already three years before failure
indicators of insolvent companies differ significantly from those of healthy enterprises.
Operating firms demonstrate rather stable performance indicators, while financial ratios of
insolvent companies deteriorate rapidly with the course of time. On average, profitability ratios
already three years before bankruptcy have negative values. In case of default companies, total
liabilities tend to grow in relation to total assets, while current assets to total liabilities ratio
decreases with the course of time.
3.4. Regression analysis and empirical results
To associate bankruptcy risk with the business cycle, we compare factors influencing
bankruptcy risk over two distinct business cycle phases: ascending and descending periods. For
this purpose all companies in the research sample are divided into two groups:
• The first group includes enterprises which filed for bankruptcy during the upward phase of the
business cycle. Data on operating firms is also taken for the same period. This group consists
of 590 companies.
• The second group consists of enterprises, which were declared bankrupts during the
downward phase of the cycle. Data on operating firms for downturn period is also included.
410 enterprises are in this group.
58
Table 8
Dating of turning points in the Russian business cycle
Investments in
fixed assets
GDP
Trough
Peak
Trough
Peak
4Q 1998
2Q 2008
3Q 2009
3Q 2012
3Q 1998
2Q 2008
2Q 2009
-
Source: Dubovsky, Kofanov and Sosunov (2015).
The separation of an upward trend from downward period of the business cycle is made
according to the dynamics of Juglar medium term cycle, which is reflected in the growth rate of
investments in fixed assets. In this case GDP is a secondary business indicator; however, its
dynamics closely corresponds to the trend of investments in fixed assets. Troughs and peaks for
both indicators are presented in Table 8. Currently Russian economic is on the descending phase
of the business cycle.
Thus, basing on dynamics of investments in fixed assets and GDP growth rate, we refer
years 2007, 2010, 2011, and 2012 to the upward phase of the business cycle. In turn, 2009, 2013,
and 2014 refer to the downward phase. As the peak of Investments cycle in 2008 took place in
the beginning of the second quarter, the most part of this year was characterized by recession.
That is why we associate year 2008 with descending business cycle phase.
Further we create separate bankruptcy risk models for economic downturn and economic
growth periods. All regression analysis calculations were conducted in Stata 12.1 software.
To understand how the macroeconomic variables contribute to bankruptcy risk, firstly we
need to consider models with accounting variables only.
Models with financial ratios
Upward phase of the business cycle
Univariate analysis showed that 15 variables out of 16 financial ratios under
consideration are statistically significant (at the 1 per cent confidence level) for bankruptcy
prediction one year before the event. An exception is Net income to equity ratio, which is not
significant at this level of confidence. The following ratios showed the highest predictive ability
(basing on Pseudo R2) in univariate analysis (Table 9).
All four ratios, which demonstrated the highest explanatory power for one year before
bankruptcy, are classified as profitability ratios. Thus, we conclude that profitability ratios
demonstrate the best forecasting ability one year prior to corporate bankruptcy (with exception of
Net income to equity). It is also worth mentioning that Total liabilities to total assets ratio and
59
Table 9
Financial ratios with the highest Pseudo R2 on the upward phase
of the business cycle (with coefficients in univariate analysis)
1 year before bankruptcy
Net income to total assets
(-27.137)
Net income to revenue
(-16.9688)
Retained earnings to total
assets
(-9.2983)
EBT to total assets
(-20.4359)
2 years before bankruptcy
3 years before bankruptcy
Retained earnings to total
assets
(-6.9946)
Net income to total assets
(-20.3476)
Total liabilities to total
assets
(6.4994)
EBT to total assets
(-15.447)
Retained earnings to total
assets
(-5.4265)
Total liabilities to total assets
(5.1452)
Equity to Total liabilities
(-2.1733)
Net income to total assets
(-12.36380)
Note: all coefficients are significant at the 1 per cent confidence level
Source: The present study.
interest coverage ratio (EBIT / Interest expenses) are also effective bankruptcy predictors,
coming next after profitability ratios.
Received signs of coefficients are logical and expectable. Profitability ratios are inversely
related to bankruptcy probability: the higher company’s net income to assets and revenue to
assets ratios are, the lower the bankruptcy risk is.
According to conducted univariate analysis, the most important determinants two years
prior to bankruptcy with the highest Pseudo R2 are: Retained earnings to total assets, Net income
to total assets, Total liabilities to total assets, Earnings before taxes to total assets. It can be
noticed that still the best predictors are profitability ratios.
For three years before bankruptcy still Retained earnings to total assets indicator
remained the strongest explanatory variable. However, financial structure indicators (Total
liabilities to total assets and Equity to total liabilities) also show significant relation to
bankruptcy risk. Total liabilities to total assets ratio is positively related to bankruptcy
probability: the higher the ratio is, the more a company is exposed to bankruptcy risk. In turn,
Equity to total liabilities ratio is negatively related to bankruptcy risk: higher share of equity is
associated with stronger financial position of a company.
In general, for the upward phase of the business cycle, profitability ratios demonstrate the
strongest link to bankruptcy risk in all three periods – one, two and three years before company’s
failure. However, financial structure indicators - Total liabilities to total assets and Equity to
Total liabilities – reveal their influence two and three years prior to bankruptcy.
60
Table 10
Financial ratios with the highest Pseudo R2 on the downward phase
of the business cycle (with coefficients in univariate analysis)
1 year before bankruptcy
2 years before bankruptcy
Retained earnings to total
assets
(-11.954)
Total liabilities to total assets
(8.00)
Net income to total assets
(-26.0246)
EBT to total assets
(-22.0216)
Retained earnings to total
assets
(-8.9004)
Total liabilities to total assets
(7.156)
Equity to Total liabilities
(-3.4842)
Net income to total assets
(-18.38)
3 years before bankruptcy
Total liabilities to total assets
(7.4873)
Equity to Total liabilities
(-3.837)
Retained earnings to total assets
(-7.3603)
Current liabilities to total assets
(5.2627)
Note: all coefficients are significant at the 1 per cent confidence level
Source: The present study.
Downward phase of the business cycle
For the downward phase of the business cycle we can notice that financial structure
indicators express stronger relation to bankruptcy risk than in the upward phase. Total liabilities
to total assets ratio is already one and two years before bankruptcy positively related to
bankruptcy risk. Three years prior to bankruptcy this ratio demonstrates the strongest correlation
with bankruptcy risk than other variables. Then comes equity to total liabilities ratio, which also
from financial structure indicators group.
For one and two years before bankruptcy, Retained earnings to total assets ratio shows
the strongest relation to bankruptcy risk. On the downward phase of a business cycle Retained
earnings to total assets and Total liabilities to total assets ratios reveal strong link to
bankruptcy risk during all three periods – one, two and three years before bankruptcy.
In general, according to conducted unvariate analysis, it turns out that first signs of
insolvency appear in financial structure indicators already three years prior to bankruptcy.
During the course of time, profitability ratios start worsening and become dominant explanatory
variables one year before failure. Other groups of financial indicators – solvency, liquidity and
assets structure variables – also deteriorate as bankruptcy event approaches. Their coefficients
are also significant in univariate equations; however, Pseudo R2 is lower than that of
abovementioned dominant explanatory ratios.
Having compared two phases of the business cycle, we also conclude that financial
structure indicators express stronger relation to bankruptcy risk in the downward phase than in
the upward phase.
61
Table 11
Coefficients of logistic regression models for bankruptcy risk estimation
over ascending and descending phases of the business cycle
Bankruptcy risk models
Ascending phase
Descending phase
(1)
(2)
(3)
(4)
(5)
(6)
Const
20.8276
12.9722
10.6960
7.3276
-2.8572
-8.7842
Size
-1.3033
-0.9121
-0.8571
-0.7634
-6.5478
-8.8537
Retained Earnings /
Total assets
Net Income / Total
assets
-21.9516
-19.1325
EBT / Total assets
-21.2576
Net Income / Revenue
-10.3647
EBIT / Interest expense
Revenue / Total assets
-26.2556
-0.4990
-0.6165
-0.9150
One year lagged variables
Current assets / Total
assets
4.6097
7.3271
4.0438
Equity / Total liabilities
3.4023
3.3206
-3.8891
Total liabilities / Total
assets
5.0253
Pseudo R2
0.7628
6.1285
0.7619
0.7562
0.7866
6.3858
0.7443
0.7563
Note: all coefficients are significant at the 1 per cent confidence level
Source: The present study.
In our further multiple factor analysis we take financial indicators with the highest
explanatory power as a basis for logistic equations. Inserting remaining variables one after
another, we select several equations, which better explain changes in bankruptcy risk.
The results of models selection are presented in Table 11. All six models under
consideration are significant (basing on the likelihood ratio chi-square, which has p-value of
0.0000 for all models). For each phase of the business cycle three equations with the highest
Pseudo R2 were chosen. Then two models, which better explain variation in bankruptcy risk over
the business cycle periods (according to Pseudo R2), were picked for the further analysis with
macroeconomic variables.
62
All three regression models, developed for the ascending phase of the business cycle,
include company’s size variable, which is negatively related to bankruptcy probability, while this
variable showed significance only in one equation in case of downward stage.
Variable of company’s age did not demonstrate close relation to bankruptcy risk during
all phases of the business cycle. It can be explained by the fact that the research sample includes
only companies, operating 10 or more years in the market. Basing on the results of regression
analysis, we can conclude that for mature companies age is not important determinant of
bankruptcy risk level.
One year lagged indicators of financial structure and assets structure (Current assets to
total assets) add explanatory power to the models. While assets structure is significantly related
to bankruptcy risk over the whole business cycle, financial structure indicators demonstrate
greater relation to bankruptcy risk during the downward phase.
Upward phase of the business cycle
In the upward phase of the business cycle the following model demonstrated the best
explanatory ability:
𝑁𝑒𝑡 𝑖𝑛𝑐𝑜𝑚𝑒
𝑅𝑒𝑣𝑒𝑛𝑢𝑒
𝑡
𝑌 = 20.8276 − 1.3033 ∗ 𝑆𝑖𝑧𝑒𝑡 − 21.9516 ∗ 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠𝑡 − 0.6165 ∗ 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠
+
+5.0253 ∗
𝑇𝑜𝑡𝑎𝑙 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠𝑡−1
𝑡
𝑡
(8)
𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠𝑡−1
Thus, in the upward phase the most significant ratios come from profitability and
financial structure groups. According to the coefficients, Net income to total assets ratio shows
the strongest negative relation to bankruptcy risk.
Downward phase of the business cycle
Changes in bankruptcy risk during the downward business cycle phase are better
described by the following equation:
𝑁𝑒𝑡 𝑖𝑛𝑐𝑜𝑚𝑒
𝑅𝑒𝑣𝑒𝑛𝑢𝑒
𝑡
𝑌 = 7.3276 − 0.7634 ∗ 𝑆𝑖𝑧𝑒𝑡 − 19.1325 ∗ 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠𝑡 − 0.9150 ∗ 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠
+
+ 6.1285 ∗
𝑇𝑜𝑡𝑎𝑙 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠𝑡−1
𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠𝑡−1
+ 4.0438 ∗
𝑡
𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝑎𝑠𝑠𝑒𝑡𝑠 𝑡−1
𝑡
(9)
𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠𝑡−1
During the downward phase assets structure indicator expresses higher influence on the
corporate bankruptcy risk. It is positively related to the risk: the larger share current assets
comprise of total assets, the higher the default probability is.
As in the upward phase, according to the coefficients, Net income to total assets and
Total liabilities to total assets ratios (lagged one year) demonstrate the strongest relation to the
bankruptcy risk.
63
For both phases of the business cycle profitability indicators show the strongest relation
to bankruptcy risk one year before the bankruptcy. Two years prior to bankruptcy assets structure
indicator (Current assets to total assets) is significant for both business cycle phases. Financial
structure indicators have greater relation to bankruptcy risk during the downward phase, while
company’s size stronger influences default probability during the upward phase.
Models with business cycle variables
The next step in our analysis is incorporation of macroeconomic variables into developed
equations. It will show whether macro indicators add some explanatory value or not.
Upward phase of the business cycle
We add macroeconomic indicators into the basis equation (Formula 8). As it was
discussed earlier, many macroeconomic variables are highly correlated with each other. That is
why we incorporate them into the model one at a time. Significant macroeconomic variables are
presented in Table 12 with corresponding coefficients. It should be mentioned that all models
with incorporated business cycle variables have Pseudo R2 higher than 0.7700, which is better
than that of the initial model with financial indicators only.
Macroeconomic variables, which are closer related to bankruptcy risk on the upward
phase of the business cycle, include Industrial production index, Repo rate, and GDP deflator
growth rate.
The example of the model with business cycle indicator can be found in Appendix 5.
Thus, we conclude that during the ascending phase of the business cycle, macroeconomic
variables contribute to more accurate explanation of changes in corporate bankruptcy risk.
Downward phase of the business cycle
Firstly, we incorporate business cycle variables into the initial equation (Formula 9). The
result is that none of the macroeconomic variables is important for bankruptcy risk estimation.
Only if Current assets to total assets ratio is excluded from the equation, one macro variable Industrial production index – becomes significant at the 3 per cent confidence level.
Then we proceed with the second best model with financial ratios for downward phase of
the cycle (model №6 in Table 11). And again, Current assets to total assets ratio should be
excluded to make macroeconomic variables significant. Macroeconomic variables with
significant coefficients are shown in Table 12. The example of the model with business cycle
variable for the descending phase is presented in Appendix 5. In this case, macroeconomic
variables that add more explanatory power are Industrial production index and Producer price
index.
64
Table 12
Coefficients of macroeconomic factors influencing corporate bankruptcy risk over
ascending and descending phases of the business cycle
Bankruptcy risk factors
Ascending phase
1 year before 2 years before
Descending phase
3 years
before
1 year before
3 years
before
bankruptcy
bankruptcy
Industrial production
index growth rate
0.0488
Non-signif.
-0.0600
Non-signif.
0.2742
Non-signif.
GDP growth rate
9.8313
-7.6683
Non-signif.
Non-signif.
24.5767
31.0019
Producer prices index
0.0997
Non-signif.
-0.0804
Non-signif.
0.0952
Non-signif.
CPI
-0.2682
-0.2059
0.2463
Non-signif.
0.3483
Non-signif.
GDP deflator
0.0967
-0.0701
Non-signif.
0.0998
Ivestments in fixed
assets
0.0556
-0.0460
-0.0411
Non-signif.
Refinancing rate
-1.1077
-0.2737
Non-signif.
0.2391
Non-signif. Non-signif.
Repo rate
-1.2814
-0.3225
Non-signif.
Non-signif.
Non-signif. Non-signif.
Short-term interest rate
-0.2536
-0.3861
Non-signif.
Non-signif.
Non-signif. Non-signif.
Long-term interest rate
-0.9203
Non-signif.
0.4964
Non-signif.
MosPrime Rate
-0.1224
Non-signif.
Non-signif.
Non-signif.
Non-signif. Non-signif.
Lending rate
-0.2247
Non-signif.
0.2938
Non-signif.
Non-signif. Non-signif.
Unemployment rate
-0.8710
0.5969
Non-signif.
1.6606
Non-signif. Non-signif.
M2 growth rate
16.995
13.9107
-4.8418
Non-signif.
5.0667
Non-signif.
Oil prices
0.0532
0.0284
-0.0206
Non-signif.
Non-signif.
-0.0292
Effective exchange rate
0.0829
Non-signif.
Non-signif.
0.3522
bankruptcy
bankruptcy
2 years
before
bankruptcy bankruptcy
Non-signif. Non-signif.
0.0783
-1.0016
Non-signif.
Non-signif.
Non-signif. Non-signif.
Note: all coefficients are significant at the 3 per cent confidence level.
Source: The present study.
However, in the downward phase macroeconomic variables demonstrate weak relation to
corporate bankruptcy risk. In addition, none of the models with business cycle variables showed
better Pseudo R2 than that of the initial model with financial variables only.
Thus, inclusion of macroeconomic variables into the downward phase model does not
add explanatory value. All information, which is necessary for bankruptcy risk estimation, is
already included into financial indicators of an enterprise.
65
It is observable that the influence of the business cycle on the bankruptcy risk is more
noticeable during the ascending phase. The following relations between bankruptcy risk and
macroeconomic indicators can be described regarding to the upward period:
• Investments in fixed assets are positively related to bankruptcy risk in the short-term period
(one year prior to bankruptcy) and negatively in the long-run. This indicator shows
significance for the whole three-year estimated period. Change of the sign for one and two
years before estimation may be explained by the fact that the effect of investments in fixed
assets can be observable in medium and long term. Consequently, investments in fixed assets,
made two and three years prior to estimation period, contribute to decrease of bankruptcy
probability, while the result of investments in short term is not yet apparent.
• Positive effect of GDP and industrial production growth is observable only with the lag. Thus,
GDP growth rate is negatively related to bankruptcy risk two years before estimation, while
for industrial production this period is three years. In short-term both indicators are positively
related to bankruptcy risk.
• Consumer prices index and GDP deflator show negative association with bankruptcy risk two
years prior to estimation period.
• Unemployment rate shows its positive relation to bankruptcy risk during the descending phase
earlier (one year before estimation), than in the upward phase of the business cycle (two years
before bankruptcy).
• All interest rate indicators demonstrate negative relation to bankruptcy risk in short-term.
• Effective exchange rate is positively correlated with the default probability one year prior to
bankruptcy. The explanation of this relationship may be found in the fact that depreciating
exchange rate is associated with more expensive import of raw materials and equipment for
domestic production.
Thus, negative association of main business cycle indicators with corporate bankruptcy
risk means that in the upward phase macroeconomic environment indicators contribute to risk
mitigation.
During the descending phase a few business cycle indicators show relation to bankruptcy
risk. Among them are industrial production index, GDP growth rate, producer and consumer
price indexes, and investments in fixed assets, which are positively related to the bankruptcy risk
two years prior to the period of estimation. In general, during the downward phase of the
business cycle influence of macroeconomic factors create potential conditions for the rise of
bankruptcy risk level.
66
CONCLUSION
Basing on the goal and objectives of the research and conducted analysis, we came to the
following concluding remarks.
Firstly, macroeconomic environment contributes to development of crisis processes in a
company. As showed our analysis, corporate bankruptcy risk is affected by business cycle
indicators. The specific characteristic of macroeconomic indicators is that most of them
(Investments in fixed assets, Industrial production and other) reveal their influence on
bankruptcy risk with the lag of two-three years.
Secondly, macroeconomic indicators demonstrate stronger relation to bankruptcy risk
over the upward phase of the business cycle than over the downward period. During the
ascending phase of the business cycle macroeconomic variables contribute to accuracy of
bankruptcy risk estimation, while in the descending phase these indicators don’t contain
information missing in financial indicators of an enterprise.
Thirdly, depending on the business cycle phase different macroeconomic indicators
express their relation to corporate bankruptcy risk.
• In the downward phase industrial production index, GDP growth rate, inflation (measured by
producer price index and consumer price index), and investments in fixed assets demonstrate
positive association with default risk two years prior to bankruptcy.
• In the upward phase effects of business cycle indicators on corporate bankruptcy risk can be
divided into three groups: long-term, medium-term and short-term effects.
- In the long-run (3 years prior to bankruptcy) the following indicators relate to corporate
bankruptcy risk: investments in fixed assets, industrial production, and oil prices. All of the
indicators demonstrate negative association with default probability.
- Medium-term impact (2 years before bankruptcy) on default probability demonstrate such
macroeconomic factors as refinancing rate, repo rate, inflation (measured by consumer
price index and GDP deflator), investments in fixed assets, and GDP growth rate. The
relation of these indicators to bankruptcy risk is negative.
- Short-term (1 year before bankruptcy) effects are connected with effective exchange rate,
short-term and long-term interest rates on federal loan bonds. Effective exchange rate
reveals positive relation to bankruptcy risk, while interest rates are negatively correlated
with default probability.
Finally, influence of company’s financial indicators on bankruptcy risk also varies
depending on time remaining before failure and the business cycle phase. Among all financial
ratios under consideration, profitability and financial structure indicators reveal the strongest
67
association with bankruptcy risk. Financial structure indicators express closer relation to
bankruptcy risk during the downward period than in the upward phase. In the descending phase
of the cycle first signs of insolvency appear in financial structure indicators already three years
prior to bankruptcy. During the course of time, profitability ratios start worsening and become
dominant explanatory factors one year before failure.
Managerial implications
Research findings are potentially useful for internal managers of enterprises. As cyclical
development is essential for the market economy, and influence of macroeconomic factors
cannot be controlled, managers should adapt company’s policy to external changes.
During the ascending phase of the business cycle negative relation of the main
macroeconomic indicators to bankruptcy risk describes favorable macroeconomic environment
for expansion of business. Moreover, weaker impact of financial structure indicators on
bankruptcy risk means good time for raising debt funds for business development purposes.
In the downward phase business cycle indicators demonstrate positive relation to
bankruptcy risk, which characterizes unfavorable external conditions and increased pressure on
bankruptcy risk. In this business cycle phase managers should take additional measures to keep
business profitable. During this period bankruptcy risk is vulnerable to changes in financial
structure indicators, that is why reliance on equity and restriction of outside borrowings
contribute to stronger financial position.
In addition, expected values of business cycle indicators can be useful for estimation of
macroeconomic influence on bankruptcy risk in the next period. Expected trends of business
cycle indicators help to adjust business development strategy, decide whether it will be
appropriate time for business expansion or not.
Limitations and directions for further research
Firstly, current research is based on data of medium and large Russian companies. Small
firms may be affected by other bankruptcy factors, and this also requires separate research.
Secondly, only companies from manufacturing industry were considered. It is very likely
that industrial differences also take place.
Thirdly, only mature companies were included in research sample, consequently,
conclusions of the current study cannot be applied to young companies. Factors affecting
bankruptcy risk in newly established companies may be a subject for subsequent research.
Finally, annual financial and macroeconomic data was used in the study. Quarterly or
even monthly data may improve results accuracy.
Taking into account these limitations, further research may be conducted basing on data
with higher frequency, data on companies from other industries and newly established firms.
68
REFERENCES
1.
Abdullaev, A.A. "Problems of identifying fraudulent bankruptcy in Russia." Business strategies, no.
2 (2014): 6-9.
2.
Achim, M.V., and S.N. Borlea. "Considerations on business risk of bankruptcy." Review Of
Economic Studies and Research Virgil Madgearu 5, no. 2 (2012): 5-29.
3.
Aivazov, A.E. «The Periodic System of the World Capitalist Development.» Partnership of
Civilizations, № 3 (2013): 253-289.
4.
Altman, Edward I. «Financial ratios, discriminant analysis and the prediction of corporate
bankruptcy.» Journal of Finance 23, № 4 (1968): 589-609.
5.
Aziz, M., и H. Dar. «Predicting corporate bankruptcy - where we stand?» Corporate governance
journal 6, № 1 (2006): 18-33.
6.
Beaver, William H. "Financial Ratios as Predictors of Failure." Journal of Accounting Research, no.
4 (1966): 71–111.
7.
Bogdan, O. « The bankruptcy risk in the construction sector .» Analele Universitatii ‘Eftimie Murgu’
Resita. Fascicola II. Studii Economice, 2014: 20-27.
8.
Burns, Arthur F., and Wesley C. Mitchell. Measuring Business Cycles . New York: National Bureau
of Economic Research, 1946.
9.
Campbell, John Y., Jens Hilscher, и Jan Szilagyi. «In Search of Distress Risk.» Journal of Finance
63, № 6 (2008): 2899–2939.
10. Darling, J., H. Seristö, and M. Gabrielsson. "Anatomy of сrisis management: a case focusing on a
major cross-cultural clash within DaimlerChrysler." The Finnish Journal of Business Economics 5
(2005): 343-360.
11. Davydova, G. V., and A. Yu. Belikov. "Methodology of quantitative estimation of the bankruptcy
risk." Upravlenie Riskom, no. 3 (1999): 13-20.
12. Dubovskiy, D., D. Kofanov, and K. Sosunov. The dating of the Russian business cycle. Moscow:
Russian Presidential Academy of National Economy and Public Administration, 2015.
13. Emsbo-Mattingly, L., D. Hofschire, and M. Betro. The Business Cycle Approach to Equity Sector
Investing. Fidelity Investments (AART), 2014.
14. Evstropov, M.V. "Appraisal of opportunities for bankruptcy prediction in Russian enterprises."
Vestnik OGU, no. 4 (2008): 25-32.
15. Fedorova, E. A., E. V. Gilenko, and S. E. Dovzhenko. "Models for bankruptcy forecasting: Case
study of Russian enterprises ." Studies on Russian Economic Development 24, no. 2 (2013): 159–
164.
16. Fedorova, E.A., и S.E. Dovzhenko. «Developnig the methodology for assessing the financial
stability of organizations in manufacturing industries.» Finance and Credit, № 25 (2015): 2–10.
17. Frolov, V. «Company’s financial risks.» Bulletin of the Moskow State Regional University.
Series:Economy 4 (2010): 95-99.
18. Fulmer, John G. Jr., James E. Moon, Thomas A. Gavin, and J. Michael Erwin. "A bankruptcy
classification model for small firms." The Journal of Commercial Bank Lending 66, no. 11 (1984):
25-37.
69
19. Gadanecz, Blaise, and Kaushik Jayaram. "Measures of Financial Stability - A Review." Irving Fisher
Committee Bulletin 31, 2008.
20. Giordani, P., T. Jacobson, E. von Schedvin, and M. Villani. "Taking the twists into account:
predicting firm bankruptcy risk with splines of financial ratios." Journal of Finance and Quantitative
Analysis 49, no. 4 (2014): 1071-1099.
21. Haydarshina, G.A. "Improvement of bankruptcy risk assessment methods for Russian enterprises in
current economic environment." Property relations in the Russian Federation, no. 8 (2009): 86-95.
22. Hernandez Tinoco, M., and N. Wilson. "Financial distress and bankruptcy prediction among listed
companies using accounting, market and macroeconomic variables." International Review of
Financial Analysis, no. 30 (2013): 394-419.
23. Hol, S. «The influence of the business cycle on bankruptcy probability.» International Transactions
in Operational Research, № 14 (2007): 75–90.
24. Jacobson, T., J. Lindé, и K. Roszbach. «Firm default and aggreagte fluctuations.» Journal of the
European Economic Association, № 11 (2013): 945–972.
25. Kopelev, I.B. «Factors of predicting bankruptcy risk of a company.» Finance and credit 45 (2014):
42-49.
26. Korol, T. «Early warning models against bankruptcy risk for Central European and Latin American
enterprises.» Economic Modelling 31 (2013): 22-30.
27. Kuzmenko, V.P. "The system of economic cycles and global financial crisis." Kondratiev waves:
aspects and prospects, 2012: 156-178.
28. Loznev, T.G. "Management of company value and the cyclical nature of macroeconomic
development." Siberian Financial School Journal, no. 2 (2006): 27-30.
29. Lvova, O., and O. Peganova. "Factors and reasons of companies bankruptcy in present economic
conditions." Public Administration: Electronic Journal, 2014: 65-68.
30. Makarov, A., and O. Rakhimova. "Developing conceptual framework, instruments and methodology
to analyze the financial viability of an organization ." Financial analytics: science and experience
47, no. 233 (2014): 32-41.
31. Matrosova, S.V. "Bankruptcy of Russian companies: dynamics and sectoral structure." Innovative
Science, no. 10-2 (2015): 86-89.
32. Mogilat, A.N. «Bankruptcy of Real Sector Companies in Russia: Main Trends and Financial
"Portrait" of a Typical Bankrupt.» Treatise: Institute of Economic Forecasting, Russian Academy of
Sciences, № 13 (2015): 156-186.
33. Monea, M. «Considerations on entity's risk analysis.» Annals Of The University Of Petrosani
Economics 14, № 2 (2014): 149-156.
34. Nam, C. W., T. S. Kim, N. J. Park, и H. K. Lee. «Bankruptcy prediction using a discrete-time
duration model incorporating temporal and macroeconomic dependencies.» Journal of Forecasting,
№ 27 (2008): 493–506.
35. Nikolaeva, I., and A. Paluvina. "The bankruptcy: causes, features and consequences in Russia."
Izvestia VSTU, no. 17 (144) (2014): 31-39.
36. Ohlson, James A. "Financial ratios and the probabilistic prediction of bankruptcy." Journal of
Accounting Research 18, no. 1 (1980): 109-131.
70
37. Pang‐Tien, L., L. Ching‐Wen, and Y. Hui‐Fun. "Financial early‐warning models on cross‐holding
groups." Journal of Industrial Management & Data Systems 108, no. 8 (2008): 1060 - 1080.
38. Qu, Y. "Macroeconomic factors and probability of default." European Journal of Economics,
Finance and Administrative Sciences, 2008: 192-215.
39. Rankov, Sinisa, and Kotlica Slobodan. "Bankruptcy prediction model used in credit risk
management." Megatrend Review 10, no. 4 (2013): 37-58.
40. Selevich, O. S. "The postcrisis analysis of the Russian bankruptcy institute." Tomsk State
Pedagogical University Bulletin, no. 12 (2013): 9-14.
41. Shirinkina, E. V., and L. A. Valiullina. "Formalization of the model of the enterprise insolvency risk
prediction." Actual Problems of Economics and Law, no. 4 (2015): 169–180.
42. Simchera, V.M. Development of Russian economy over the last 100 years. Historical series, trends
of the century, institutional cycles. Moscow: Economics, 2007.
43. Sitnikova, Ya. V. "Financial risks determination as a basis of bankruptcy prediction." Vestnik
NSUEM, № 1 (2012): 197-206.
44. Springate, Gordon L.V. Predicting the possibility of failure in a Canadian firm. Unpublished M.B.A.
Research Project, Simon Fraser University, 1978.
45. Taffler, R.J. "The assessment of company solvency and performance using a statistical model."
Accounting and Business Research 13, no. 52 (1983): 295-308.
46. Totmyanina, K.M. "Default models for Russian enterprises with taking into account macroeconomic
situation." Journal of corporate finance research 1, no. 29 (2014): 18-30.
47. Tsirel, S.V. "Kondratiev waves in Russia and other countries ." Kondratiev wave: issues and
prospects, 2012: 362-372.
48. Tyapkina, M.F., Y.D. Mongush, and I.O. Akimova. "Investigation of the cyclicity indicators of the
economic development of Russia." Russian Entrepreneurship 24, no. 270 (2014): 4-14.
49. Vorotnikova, I., and I. Pshipiy. "Estimating the probability of bankruptcy of companies: a modern
approach." Financial life 3 (2015): 69-72.
50. Zarnowitz, V. "Business cycles observed and assessed: Why and how they matter." National Bureau
of Economic Research working paper 6230, 1997.
51. Zhdanov, V. "The anti-crisis mechanism of bankruptcy risk diagnostics." Management of Economic
Systems: Scientific Electronic Journal, no. 32 (2011).
52. Zmijewski, Mark E. "Methodological issues related to the estimation of financial distress prediction
models." Journal of Accounting Research 22 (1984): 59-82.
Databases and electronic resources:
53. Central Bank of Russian Federation. 2016. Accessed May 25, 2016. http://www.cbr.ru/.
54. Federal Register of information
http://bankrot.fedresurs.ru/.
on
bankruptcy.
2016.
Accessed
May
25,
2016.
55. Federal State Statistics Service. 2016. Accessed May 25, 2016. http://www.gks.ru/.
56. Spark-Interfax database. 2016. Accessed May 25, 2016. http://www.spark-interfax.ru/promo/.
57. The World Bank Database. 2016. Accessed May 25, 2016. http://databank.worldbank.org/.
71
APPENDICES
Appendix 1. Number of bankrupt and operating companies in the research sample
by manufacturing industry subsectors
Food products including
beverages
Production of other nonmetallic mineral products
Wood processing, products
from wood and cork
Electrical machinery and
equipment
Metallurgical production
Machinery manufacturing
Textile industry
Automobiles, trailers and
semi-trailers production
Production of finished metal
products
Electrical machinery and
equipment
Production of crafts, aircrafts,
spacecrafts and other vehicles
Office facilities and computer
machines
Rubber and plastic articles
Chemical industry
Leather, leather products and
footwear
Manufacture of electronic
components, radio, television
and communication
equipment
Production of furniture
Publishing and printing
Production of finished metal
products
Production of pulp and paper
Recycling of secondary raw
materials.
Other
Total
Number of bankrupt
companies
Number of operating
companies
Total
42
121
163
31
89
120
22
72
94
19
61
80
18
15
16
52
49
56
70
64
72
11
31
42
12
28
40
7
22
29
5
13
18
6
12
18
6
5
17
11
23
16
4
14
18
4
12
16
4
4
16
12
20
16
3
10
13
3
11
14
2
5
7
11
250
36
750
47
1000
Source: The present study.
72
Appendix 2. Correlation matrix of macroeconomic variables
IPI_gr_1 GDP_gr_1
IPI_gr_1
GDP_gr_1
IFA_w_1
IPP_1
CPI_r_1
GPD_defl_1
Ref_1
Repo_1
ST_ir_1
LT_ir_1
spread_1
Ef_exr_1
Unemp_1
Oil_1
MPR__6m_1
M2_wide_1
ST_ir_1
LT_ir_1
spread_1
Ef_exr_1
Unemp_1
Oil_1
MPR__6m_1
M2_wide_1
IFA_w_1
IPP_1
1.0000
0.3835
0.3847
0.8549
0.3376
0.2637
-0.7933
-0.9453
0.2685
0.8106
-0.8309
0.5750
-0.8142
0.5477
CPI_r_1 GPD_de~1
1.0000
0.9337
0.8972
0.5600
0.0772
0.7974
-0.0020
-0.0643
-0.9544
-0.9243
0.6066
0.9383
-0.7061
0.5787
-0.9804
0.5145
1.0000
0.9748
0.3371
0.3749
0.9083
0.3461
0.2832
-0.8761
-0.9765
0.3864
0.9061
-0.8452
0.6509
-0.8592
0.4181
1.0000
-0.1332
0.1810
-0.4959
-0.5595
-0.4125
-0.3127
0.3777
0.4461
0.0851
-0.3107
-0.5757
0.8628
1.0000
0.5122
0.9017
0.8759
0.0135
-0.2903
-0.4125
0.2298
-0.2681
-0.0410
0.1073
0.0738
ST_ir_1
LT_ir_1 spread_1 Ef_exr_1
Unemp_1
1.0000
0.8736
-0.7618
-0.9523
0.6697
-0.6987
0.9614
-0.2926
1.0000
-0.3504
-0.8750
0.8892
-0.6779
0.8760
-0.4150
1.0000
-0.8279
0.6557
-0.1394
1.0000
0.6682
-0.1053
0.4424
-0.6844
0.0108
1.0000
-0.6255
0.5681
-0.8991
0.3096
1.0000
0.5213
0.4809
-0.8155
-0.8278
0.4677
0.8816
-0.6807
0.6295
-0.6987
0.1984
Ref_1
Repo_1
1.0000
0.9957
0.0116
-0.2780
-0.3924
0.1533
-0.4194
0.2647
0.1675
-0.2428
1.0000
0.0524
-0.2152
-0.3874
0.1096
-0.3733
0.2608
0.2237
-0.3284
Oil_1 MPR__6~1 M2_wid~1
1.0000
-0.5875
-0.2567
1.0000
-0.4909
1.0000
Source: The present study.
73
Appendix 3. Descriptive statistics of bankrupt companies over the upward and
downward phases of the business cycle
Bankrupts in the downward phase
Variable
Obs.
Mean
Std. Dev.
Financial structure indicators
Total debt / Total assets
102
0.5719
1.1075
Equity / Total liabilities
101
-0.1928
0.4003
Total liabilities / Total assets
102
1.6559
1.2810
Current liabilities / Total assets
102
1.4578
1.2007
Solvency and liquidity indicators
Current assets / Current liabilities
101
0.8160
0.8848
Current assets / Total liabilities
101
0.5861
0.3331
Working capital / Total assets
102
-0.7301
1.2090
EBIT / Current liabilities
101
-0.1732
0.2418
Profitability ratios
Net income / Revenue
100
-1.6354
6.2261
Retained earnings / Total assets
102
-0.8742
1.3972
EBIT / Total assets
102
-0.3401
0.6666
Net income / Total assets
102
-0.3817
0.7575
Min
Max
0
-0.9870
0.2100
0
9.9569
2.0598
7.6157
7.1142
0.0111
0.0111
-6.1529
-1.3468
7.8183
1.6362
1
0.4118
-55.1672
-7.8149
-3.9297
-4.9276
0.0836
0.1840
0.2067
0.0689
Min
Max
0
-0.9720
0.1996
0.1031
7.6958
2.0094
8.4153
8.4153
0.0234
0.0139
-7.8518
-4.3163
7.8108
1.7995
0.7759
1.4222
-58.0692
-7.6193
-3.5436
-4.5436
0.0756
0.3447
0.3906
0.1024
Bankrupts in the upward phase
Variable
Obs.
Mean
Std. Dev.
Financial structure indicators
Total debt / Total assets
148
0.5099
0.8825
Equity / Total liabilities
147
-0.1747
0.4013
Total liabilities / Total assets
147
1.6656
1.3132
Current liabilities / Total assets
148
1.4721
1.2053
Solvency and liquidity indicators
Current assets / Current liabilities
148
0.7839
1.0011
Current assets / Total liabilities
147
0.5532
0.3163
Working capital / Total assets
148
-0.8365
1.3613
EBIT / Current liabilities
148
-0.2742
0.4674
Profitability ratios
Net income / Revenue
145
-2.7442
8.9113
Retained earnings / Total assets
148
-0.8607
1.3966
EBIT / Total assets
148
-0.4447
0.6756
Net income / Total assets
147
-0.4878
0.7274
Source: The present study.
74
Appendix 4. Descriptive statistics of operating companies over the upward and
downward phases of the business cycle
Operating companies in the downward phase
Variable
Obs.
Mean
Std. Dev.
Financial structure indicators
Total debt / Total assets
307
0.1603
0.2107
Equity / Total liabilities
307
2.9472
4.1649
Total liabilities / Total assets
307
0.4611
0.2842
Current liabilities / Total assets
307
0.3647
0.2501
Solvency and liquidity indicators
Current assets / Current liabilities
307
3.0407
3.4869
Current assets / Total liabilities
307
2.4310
2.9972
Working capital / Total assets
307
0.2646
.2633
EBIT / Current liabilities
307
1.1324
1.6295
Profitability ratios
Net income / Revenue
307
0.0719
0.0851
Retained earnings / Total assets
307
0.3963
0.2871
EBIT / Total assets
307
0.1815
0.1653
Net income / Total assets
307
0.1193
0.1247
Min
Max
0
-0.2293
0.0350
0.0315
1.0191
27.5419
1.1927
1.1541
0.0607
0.0587
-0.5995
-0.7968
26.4138
25.7665
1.1079
8.5586
-0.1657
-0.2576
-0.1215
-0.1494
0.4464
1.0624
1.2454
0.6919
Min
Max
0
-0.2141
0.0221
0.0221
1.0433
27.5471
1.2725
1.0502
0.2163
0.1632
-0.6887
-0.8673
34.3120
34.3120
0.9163
9.8746
-0.7902
-0.3839
-0.2682
-0.1584
0.3893
0.9640
0.6818
0.5444
Operating companies in the upward phase
Variable
Obs.
Mean
Std. Dev.
Financial structure indicators
Total debt / Total assets
443
0.2225
0.2425
Equity / Total liabilities
443
2.3164
3.9049
Total liabilities / Total assets
443
0.5252
0.2804
Current liabilities / Total assets
443
0.4063
0.2472
Solvency and liquidity indicators
Current assets/ Current liabilities
443
2.7172
3.4736
Current assets / Total liabilities
443
2.1219
3.0207
Working capital / Total assets
443
0.2297
0.2608
EBIT / Current liabilities
441
0.6420
1.1276
Profitability ratios
Net income / Revenue
443
0.0431
0.0832
Retained earnings / Total assets
443
0.3415
0.2905
EBIT / Total assets
441
0.1224
0.1148
Net income / Total assets
443
0.0742
0.0965
Source: The present study.
75
Appendix 5. Examples of models with business cycle variables for upward and
downward phases of the business cycle
Model for the upward phase
Logistic regression
Number of obs
LR chi2(5)
Prob > chi2
Pseudo R2
Log likelihood = -74.664585
BN
Coef.
Size_1
NITA_1
RevTA_1
TLTA_2
IPI_1
_cons
-1.369937
-24.06177
-.6202429
4.992677
.0488143
17.39372
Std. Err.
.257908
3.87319
.2087679
1.106372
.0181321
5.056612
z
-5.31
-6.21
-2.97
4.51
2.69
3.44
P>|z|
0.000
0.000
0.003
0.000
0.007
0.001
=
=
=
=
590
513.13
0.0000
0.7746
[95% Conf. Interval]
-1.875428
-31.65308
-1.02942
2.824228
.013276
7.482945
-.864447
-16.47045
-.2110654
7.161126
.0843526
27.3045
Note: NITA_1 – Net income to total assets one year before estimation; RevTA_1 – Revenue to total assets one
year before estimation; TLTA_2 – Total liabilities to total assets two years before estimation; IPI_1 – Industrial
production index one year before estimation.
Model for the downward phase
Number of obs
LR chi2(3)
Prob > chi2
Pseudo R2
Logistic regression
Log likelihood = -56.988266
BN
Coef.
TLTA_2
EBTTA_1
IPI_2
_cons
7.074195
-19.77442
.2742164
-36.78433
Std. Err.
1.321749
3.260697
.1163035
12.91014
z
5.35
-6.06
2.36
-2.85
P>|z|
0.000
0.000
0.018
0.004
=
=
=
=
409
345.46
0.0000
0.7519
[95% Conf. Interval]
4.483615
-26.16527
.0462656
-62.08774
9.664775
-13.38357
.5021671
-11.48093
Note: TLTA_2 – Total liabilities to total assets two years before estimation; EBTTA_1 – Earnings before taxes to
total assets one year before estimation; IPI_2 – Industrial production index two years before estimation.
Source: The present study.
76
Отзывы:
Авторизуйтесь, чтобы оставить отзыв