Размещено на http://www.allbest.ru/

Размещено на http://www.allbest.ru/

INSTITUTION OF HIGHER PROFESSIONAL EDUCATION

NATIONAL RESEARCH UNIVERSITY

"HIGHER SCHOOL OF ECONOMICS"

International College of Economics and Finance

Master Thesis

Corporate governance and performance during a global crisis

in the field of 38.04.01 «Economics»

educational program «Financial Economics»

Anna Klykova

Moscow, 2018

Table of contents

• Introduction
• Chapter 1. Literature review and main hypotheses
• 1.1 Effect of leverage on company's performance in crisis
• 1.2 Corporate governance effect on company's performance in crisis
• 1.3 Institutions effect on company's performance in crisis
• 1.4 Hypotheses
• Chapter 2. Data description
• Chapter 3. Methodology
• Chapter 4. Results
• 4.1 Main Results
• 4.2.1 Results for ROA
• 4.2.2 Results for Profit Volatility
• 4.2.3 Results for Stock Return
• 4.2 Discussion
• Chapter 5. Conclusion
• References
• Appendices

Introduction

In the last decade, corporate governance has undergone major changes. If previously the super-profits of investors and the unlimited growth of markets made it possible to smooth out the effect of poor corporate governance and weak institutional support of investors rights, at the time of the global crisis the situation changed radically. In the face of limited demand and the precariousness of the financial system, ways of servicing the debt burden in conditions of limited access to financial resources along with corporate governance determined the sustainability of the organizations (Jones A., 2012).

Corporate governance is not only the organization of the activities of the board of directors, but the system of relations with shareholders and the observance of their rights. It should be noted that it is an integrated approach to building mechanisms by which business is carried out. This concept is based on the development of strategic planning, risk management, ensuring compliance of company activities with legislative requirements that would allow it to pass crises with less loss.

Thanks to the Global crisis, the companies decided to advance their management by introducing effective corporate governance, which not only attracts investments and opens the doors to the capital markets, but also improves business transparency, enhances investor loyalty in building long-term relationships (Yakovlev A., Danilov Y., Simachev Y., 2010).

It is well known that the main task of the company's management is to maximize the wealth of its shareholders. This can be achieved through a competently built investment strategy and financial policy. However, despite the general tendencies to build an effective business, management decisions on the leverage level and corporate governance methods are greatly influenced by the institutions of the country in which the company operates.

Surely, leverage level can change incentives of managers. In crisis leverage typically grows, because in case of financial instability, company may face lack of financial resources and therefore, increase its' debt burden, which, in turn, can change managers behavior depending on institutional environment. Depending on the level of corporate governance and the strengths of institutions of investors' protection and overall legal climate managers can positively respond to the shock by strengthening discipline, changing the structure of capital, investing in profitable projects. Nevertheless, there can be negative effect - opportunistic behavior of managers in the form of assets withdrawal. Institutes of the country level may prevent managers from private benefit extraction. In economies with strong institutes, managers are limited in their actions, which makes it possible to keep them from dishonest actions to the detriment of the company and its shareholders. Poor legal environment probably induce managers to undertake not optimal decisions, which harm the business.

Consequently, the problem of finding the optimal ratio of own and borrowed capital is one of the most important one in the development of company's financial strategy. The Global crisis has suppressed the business of the majority of international companies, restricting access to capital markets, making the debt load colossal.

In this paper, we are aim at exploring the effect of leverage, institutes and overall corporate governance on company's performance during the crisis. Firstly, we explored theoretical relationship among leverage, institutional indicators and corporate governance and its independent and joint influence on company's performance. Secondly, the model, which quantitatively estimates the effect of the interest, was constructed. Thirdly, analysis of the results to understand how corporate governance affects the way company responses on crisis and increase in debt were performed.

We suppose that corporate governance changed significantly under the pressure of crisis reality (Martin Conyon William Q. Judge Michael Useem, 2011). Under the influence of institutional and macroeconomic factors, managers began to change their behavior. Major changes may affect restructuring of investment policy and capital structure. The management decisions during the crisis directly depended on the pre-crisis level of the corporate governance and the development of the institutions in the country, which can affect both positively and negatively the company's performance.

From the theory of corporate finance, we know that the effect of leverage can have either positive or negative effect on the company's results. In the first case, leverage plays a disciplining role, allowing to reduce excess expenses and realize only investment-profitable projects. In the second case, the growth of the leverage facilitates the withdrawal of assets, which can lead to the undermining of the business or bankruptcy. We assume that the direction of the leverage effect depends on the maturity of corporate governance.

When it comes to effect of country-level institutions of investor protection, it should be noted that in weak regimes the agency problem becomes more acute, which requires more effective corporate governance to ensure guarantees to investors (Durnev, Kim, 2003). Surely, we see interesting interconnection between institutions, corporate governance and leverage. The joint and individual influence of these factors on the company's performance during the crisis is to be assessed in this paper. Since the Global crisis radically changed the vision of business, it is most interesting to consider this period.

Our research hypotheses are based on in-depth review of the scientific literature. There are numbers of papers studying individual effect of country-level institutions, leverage and corporate governance on company's key performance indicators. Nevertheless, literature lacks research on the joint influence of institutions and leverage on the company's performance during the global crisis.

The relevance of this work lies in studying the influence of the level of leverage, country-level institutions and corporate governance on the performance of the company during the Global crisis 2007-2008. As far as we know, previous researchers did not conduct research on establishing connections between leverage and institutions on the global scale, which distinguishes our work from the rest.

The empirical analysis are based on a large amount of data obtained by means of a Bloomberg and DataStream. We are interested in non-financial companies of developed countries in the period from 2007 to 2009. Financial companies are excluded from the analysis due to its specific capital structure requirements.

The structure of the paper is as follows: literature review and main hypotheses are presented in Chapter 1; Data description is reflected in Chapter 2; Methodology is described in Chapter 3; Results are shown in Chapter 4.

We found that corporate governance affected positively company's performance in all cases. We also saw that even in case of good corporate governance, the effect of leverage was negative, and the higher the score for corporate governance, the more negative the effect of leverage. Therefore, in crisis times in case of high leverage, corporate governance did not help to overcome possible negative incentives of managers. Moreover, we saw that in poor legal environment, there was more positive influence of corporate governance on company's performance. However, we got this result only for ROA and obtained mostly insignificant results.

Chapter 1. Literature review and main hypotheses

The main goal of the work is to identify the impact of leverage and country-level institutions along with corporate governance on company's decisions during the crisis. Literature is rich in studying these factors separately, but in the scientific world, there is no study of joint influence of leverage and institutions.

A large pool of literature can be divided into two types of influence on the company's performance. Firstly, there were studies on the dependence of corporate governance and the effect of leverage. Secondly, there were studies on the dependence of corporate governance and country-level institutions.

1.1 Effect of leverage on company's performance in crisis

Investigations of the influence of the capital structure on the effectiveness of the company's activities are vastly extensive and explore both financial and non-financial companies. In accordance with the purposes of our study, we turn to the authors who investigated companies in the non-financial sector in the period of exogenous shocks.

One of the most significant works is an empirical study by Ofek (1993). In his work, the author tried to find a relationship between capital structure and company's activities in response to short-term shocks. The author examined 358 firms in the period from 1983 to 1987, which showed average or above average results before the crisis period, and in the moment of shock their figures dropped sharply. The author showed that a high level of financial leverage contributed to increasing the probability of taking operational actions, such as asset restructuring, staff reduction at the time of the crisis. The key result of the author was the finding of a positive relationship between the level of leverage in the pre-crisis period and the company's financial sensitivity to the crisis phenomena. The author emphasized that the higher the leverage, the companies were more inclined to operational and financial changes. For example, the reduction of dividends, the withdrawal of assets and even bankruptcy. Another interesting conclusion was the finding that the more capital was concentrated in the hands of management, the less likely they would undertake operational deeds, especially if they do not generate cash flow. These findings were consistent the results of other authors, such as Stulz (1990), who found that management would likely undertake investment decisions if it generated low cash flows.

In their work Opler и Titman (1994) showed that a large share of borrowed funds in the company's capital structure adversely affected firm's results in normal time. Moreover, the authors emphasized that in the most concentrated industries large financial leverage negatively affected the company's operational and financial effectiveness.

Consistently with Ofek (1993) Kang and Shivdasani (1997) investigated what factors affected companies during the period of overall decline. The authors concentrated on Japanese market in the period of 1986 - 1990. The sample included 92 firms, which were restructuring during the period of interest. The authors set out to understand how much the American firm's reaction to the crisis was different from the reaction of the Japanese ones. The main difference lied in the fact that Japanese firms were less inclined to cardinal staff cuts and asset restructuring than their American counterparts. From this work, we can draw an important conclusion, made by the authors based on the research; the differences in the structure of corporate governance differ significantly across developed countries depending on their historical and cultural changes, as well as the legal environment. For example, in Japan, where corporate governance differs from other countries, restructuring and smaller staff cuts correlate with improving the company's performance.

Another sound work is research made by Berger and Patti (2006). The authors tested the agency cost hypothesis, which implied that higher leverage level reduces the costs of outside equity and thus increases the value of the company by disciplining managers. Nevertheless, there was also the reverse side of the coin, when leverage became high and increased the costs of operational and financial transformation, outside sources of equity became more attractive, which led to the fact that agency costs of outside debt could exceed costs of outside equity. Therefore, an increase in leverage could result in increase of total agency costs. The authors set the goal to study both the fundamental hypothesis of corporate finance and the reverse causality case. The results of the research spoke in favor of a fundamental hypothesis, that is, a higher leverage is connected with an increase in efficiency of profits. However, it should be noted that the authors investigated the banking industry of the United States, but this conclusion is useful for constructing hypotheses in our research.

Denis and Shome (2005) in their research investigated factors that affect the restructuring of assets and company's performance in response to declining economic conditions. Due to the multivariate analysis, the authors found that the implementation of optimization reforms in companies negatively correlated with operational efficiency and positively with the level of leverage and diversification. After the restructuring companies became more focused and had smaller amount of debt, which had a positive impact on their efficiency. Based on this, we can conclude that asset restructuring would have a positive impact on the company's performance during distresses only along with mature corporate governance, because it would prevent managers from assets stripping in their own needs.

Based on the presented survey of studies, we can conclude that there exists an individual leverage effect on the company's performance indicators during the crisis.

1.2 Corporate governance effect on company's performance in crisis

A large number of literature devoted to corporate governance concerns the period of the Asian crisis of 1997-1998.

One of the soundest investigations is the research work by Johnson (2000). The author studied the reasons for the stock market decline and instability of exchange rates, which was called “the Asian crisis”. The financial crisis closed access to capital markets and contributed to a general decline in business activity, which could be explained either by internal factors or by the macroeconomic policy of the state. In his paper, the author proved that corporate governance measures, and in particular the level of protection of minority shareholders' interests, had a greater impact on the market than macroeconomic factors. The author's argumentation is quite logical, because if expropriation by managers increased, while expecting a low rate of return, then with unfavorable shock, it would increase even more, which would lead to imbalance between inflows and outflows from a country in favor of the latter, which would in turn entail negative macroeconomic consequences. In countries with a low level of corporate governance, the crisis more seriously affected managers' decisions to expropriation, which led to a decrease in the company's assets. Conducting a cross-cultural multifactor analysis, the author concluded that the overall level of corporate governance was the main factor that influenced the economy during the crisis.

Mitton (2002) investigated 398 firm from the East Asia and found that dynamics of stocks positively correlates with the level of transparency of company's activities as well as with high level of outside ownership concentration. Along with La Porta, Lopez-de-Silanes, Shleifer and Vishny (1997, 1998, 1999, 2000) the author emphasized that corporate governance was the crucial factor in developing of firm value and overall financial market in the crisis. The author studied not only cross-cultural dependencies, but also tried to understand why performance of companies differs in one country. The author stated that each country needed to build strong institutional foundation before opening the market to foreign cash flows. He concluded that firms with weak corporate governance, namely, weak legal protection of minorities, lost more during the crisis.

Stocks return is one of the key indicators of the company's efficiency. Lemmon and Lins (2003) expanded the study of Mitton (2001), by investigating 800 firms. The authors investigated the influence of the ownership structure on the shareholder returns in the East Asia. Companies in which controlling shareholders had more voting rights, but differentiated their control rights and rights on cash flows were more affected during the Asian crisis. Author suggested that the main reason was that managers themselves decided in which projects to invest in order to maximize their own profitability to the detriment of the interests of shareholders. The authors also concluded that the ownership structure could help to determine incentives of insiders to expropriate minority shareholders during the crisis.

Bae et al. (2012) investigated the same relationships. The authors found that companies with weak corporate governance experienced huge drop in stock prices during the crisis, but recovered faster in post-crisis period. As it was said in previous works, during the crisis, the probability of expropriation of minority shareholders by the controlling ones increased, since panic of reduced profitability began. The authors tested a model that revealed that companies in which there was a large disparity between voting rights and cash flow rights were more exposed to the negative impact of the crisis. That is, if there was a lack of trust between management and shareholders, the likelihood of a decline in the company's value increased, since managers were less motivated to effectively manage the company.

Baek et al. (2004) explored Korean market in the research. They aimed at establishing the relationship between corporate governance and stock returns. They found that availability of alternative sources of financing, disclosure quality and high concentration of foreign investors allowed companies to experience a smaller drop in share prices. The largest drop in shares was found in companies owned and operated by families.

Erkens et al. (2012) examined the relationship between corporate governance of financial institutions and their performance during the crisis of 2007-2008. The sample consisted of 296 financial firms from 30 countries. The authors found that firms with more independent board of directors and higher institutional ownership suffered more from stocks return declines. They explained it by the suggestion that higher institutional ownerships encouraged managers to take more risk prior the crisis, which resulted in huge shareholders losses during the crisis. Additionally, firms with independent boards of directors tended to raise more equity capital during the crisis, which led to wealth transfer to debtholders. These results are not consistent with ones obtained by authors discussed above. It postulated that good corporate governance in terms of risk-taking and financing policies negatively influenced on financial firms during the crisis.

We found the work made by Gupta et al. (2013) very interesting and useful for the present research. The authors studied huge sample of more than 4000 companies from 22 developed countries. They found that in the crisis good governed firms did nor outperform poorly governed ones. The authors made very interesting conclusion that actually stock markets were less efficient in incorporate internal corporate governance indicators into stock prices. Moreover, the authors stated that differences in country level institutional environment did not help to explain the lack in influence of corporate governance on stock process during the Global Crisis.

Despite the fact that most articles are devoted to the study of the Asian market, the literature analysis allows us to conclude that there is a strong connection between the level of corporate governance and the performance of the company in crisis.

1.3 Country-level institutions effect on company's performance in crisis

It is not enough for a company to simply demonstrate high level of corporate governance, since if the country's legal system is weak, internal mechanisms will not always be able to protect investors, which creates problems for capital inflows.

Klapper and Love (2003) investigated the relationship between firm-level corporate governance and country-level investors' protection. The authors postulated that corporate governance was more important in countries with poor legal investors' protection and weak judicial efficiency. Cross-country differences implied that legal environment influenced on dividend payout ratios, access to capital markets, ownership structure etc. Hence, firms, which operated in poor legal regimes, tried to establish good firm-level corporate governance to counterbalance the weaknesses of the investors' protection environment.

Among the literature touching on this topic, the most interesting for us is the work by Enikolopov et al. (2014). The authors seek to show that country-level and corporate-level corporate governance institutions were complements in their effect on the performance of a company during the crisis. The authors chose the effectiveness of the government order and bureaucratic quality, and the rule of law as indicators for legal environment. The authors considered the period of Global crisis of 2007-2009. Testing the model resulted in postulating that in crisis firm-level transparency was more important in countries with strong legal protection, while in weak regimes transparency played almost no role. Moreover, in countries with stronger legal protection less transparent firms lost significantly more than their transparent counterparts did. In pre-crisis times, good corporate governance was more important in weak regimes, because of the need to attract investors' capital. Previous works by Durnev and Kim (2005), Klapper and Love (2004), Bruno and Claessens (2010) who showed that corporate governance and country-level institutions were substitutes in normal times, Enikolopov et al. (2014) contributed in literature by showing that they actually became complements in terms of the effect on firm's value during the crisis.

Durnev and Kim (2005) made one more sound investigation. They documented that firm's choice of corporate governance mechanisms and disclosure practices positively depended on demand for external financing, growth opportunities and concentration of cash flow rights. Such relationships were stronger in weaker legal regimes. It was quite logical, because firms aiming at attracting capital were eager to establish high-level corporate governance in poor legal environment to ensure investors' returns. Moreover, in weak investor protection regimes, ownership concentration became an effective way to resolve agency problem between controlling and minority shareholders. The authors also found that firms' value and the strength of legal regimes were positively related. But this relation became insignificant when individual-firm scores were added. This could be explained by the fact that even in poor legal environment a firm could enjoy high valuation if it adopted high-level corporate governance and disclosure practices.

One more sound investigation was made by Dan Yanga (2017). The authors showed that effectiveness of corporate governance procedures could be more or less effective depending on the environment. For example, they argued that in underdeveloped institutional environment high ownership concentration would play positive and important role than for dispersed systems, because in such situations agency problems were less significant and corporate governance mechanism worked as planned.

Thus, a review of the literature allows us to conclude that the level of protection of investors in the country and the general legal situation affect the effectiveness of the company. The results of previous researchers also suggest that institutional factors and corporate governance can be considered as substitutes, and as complements.

1.4 Hypotheses

Based on the review of the literature, we concluded that in the case of poor corporate governance, managers might seek to redistribute assets for their own personal purposes, which led debt overhang and the abandonment of prospective projects, which in turn reduced the company's value.

A large level of leverage can smooth weak corporate governance, disciplining managers. That is, if the company has low standards for protecting shareholders' rights, it needs more diverse instruments to mitigate agency conflicts; in this case, the financial leverage plays a positive role.

Certainly, in the crisis, the situation changes radically and the direction of the leverage effect can depend on corporate governance.

Our hypotheses are as follows:

We assume that a high-level corporate governance positively affects the company's performance indicators in a crisis.

Hypothesis 1. Corporate governance has a positive effect on the company's performance in a crisis.

As it was said above, we assume that good corporate governance mitigates the shortcomings of debt in times of crisis.

Hypothesis 2. In case of good corporate governance, leverage has a positive or less negative impact on the company's performance in the crisis, in case of poor corporate governance it has negative influence.

As we have different views on the influence of institutions on company's performance, here we assume that investors demand more efficient firm-level governance in case of poor outside legal protection.

Hypothesis 3. Corporate governance affects the efficiency of the company in poor legal protection more than in well-developed investor protection systems during the crisis.

It is interesting to explore how leverage affects companies' performance according to the level of investors protection and overall juridical environment in a country.

Hypothesis 4. In case of good institutional environment in a country, leverage has a positive or less negative impact on the company's performance in the crisis, in case of poor environment it has negative influence.

This research is unique in the way that it aims to study the influence of three variables (leverage, corporate governance and institutions) both individually and in combinations on the company's performance during the Global crisis. Literature lacks research on the joint influence of institutions and leverage on the company's performance during the Global crisis.

Chapter 2. Data description

In this paper, dataset are represented as a sample of international public companies of developed markets traded on exchanges. Financial companies and banks were not considered, because such companies had a specific capital structure that reflects the mandatory standards (reserves). Moreover, for such companies, money is not a source of funding, but a resource.

Data were collected for 2007-2009 period. For collecting data for variables of interest DataStream/Eikon, Bloomberg, World Bank and S&P were used.

In the analysis, we used data for developed countries, which were chosen by several indices, which are presented in the Appendix B. We used several global indices to identify developed countries. They are based on several criteria such as economy and capital markets development. The main criteria was that the country must be high income, but it also had openness to foreign ownership, ease of capital movement, and efficiency of market institutions. We also saw what countries were the members of OCED to be sure of its advanced economics. We exclude from the analysis Iceland since it has two negative ratings and Cyprus due to relatively small firm sample. For collecting data, Thomson Reuters was mainly used. After several manipulations, we collected data for 24 developed countries and 751 company. The main reason for excluding companies from the sample was the absence of corporate governance indicators. The distribution among countries are presented in the Picture 1. Obviously, there were mostly US and UK companies, since they suggest the full information about corporate governance and financial indicators we needed for the analysis.



Picture 1 Model sample description

corporate governance global crisis

The corporate governance indices and financial information were collected from Thompson Reuter's database. The institutional variables were collected from the Databank of the World Bank.

In the model we considered the following independent variables:

· Corporate governance

· Leverage

· Institutions

Corporate governance were measured by the Indices The definitions for corporate governance indices were taken from Thomson Reuters Eikon database, which were constructed by Thomson Reuters using self-reported information of companies:

· Aggregate corporate governance Index (ESG Score) measures the systems and processes of the company that ensure that members of the board of directors and top management act in the interests of its shareholders. It reflects the company's ability to use advanced management methods, direct and monitor the rights and responsibilities of management by creating positive incentives, and through “checks and balances” create a long-term value of equity;

· Management Score measures the commitment and effectiveness of the company's management to the principles of corporate governance. It reflects the company's ability to have an effective board, creating independent board committees with different tasks and responsibilities;

· Shareholders Score reflects the company's ability to guarantee an efficient exchange of ideas and an independent decision-making process through an experienced, diverse and independent board as well as effectiveness towards equal treatment of shareholders and the use of anti-takeover devices.

The value of the indices varied from 0 to 100, where 100 meant that the company maximally satisfied the notions of good corporate governance.

Leverage was measured by the market value (the ratio of the total debt to the market value of assets).

Institutions were measured by several indexes proposed by the World Bank The definitions for institutional variables were taken from the World Bank website: http://info.worldbank.org/governance/wgi:

· Government Effectiveness Index aims to assess the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies;

· Rule of Law Index assesses the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement and property rights.

· Control of Corruption Index includes indicators that measure the perception of corruption in society, the degree of use of state power for mercenary purposes, the existence of corruption at a high political level, the impact of corruption on economic development.;

· Regulatory Quality Index shows perceptions of the business and society about abilities of the government to implement effective and sound policies and regulations that help in promoting private sector development.

· Voice and Accountability Index measures to what extent citizens are able to select the government, the degree of freedom of association, freedom of mass media;

· Political Stability No Violence Index measures the likelihood of terrorism and political instability.

The described indices varied from -2.5 (weak) to 2.5 (strong) government performance.

As control variables we chose firm size (ln(revenue)), industry (binary variable indicating whether the company belongs to the industrials) and pre-crisis profitability of the company (EBIT to the value of all assets ratio, used as control for ROA and profit volatility).

2.1 Descriptive statistic

This Section presents the distribution of the variables of interest.

As we used variables of interest of two subsequent years, we looked at both. The descriptive statistic for ROA is presented in Table 1.

Table 1 Descriptive statistic for ROA

	ROA 2008	ROA 2009
N	751	751
Mean	0.077	0.061
Std Deviation	0.063	0.059
Skewness	0.171	0.178
Kurtosis	5.50	3.012
MIN	-0.342	-0.201
MAX	0.340	0.334

We got that mean ROA in 2008 was 7.7% with maximum of 33.9% and minimum of -34.2%. For ROA in 2009 we had mean of 6.1%, maximum of 33.4%, minimum of -20.1%. It was notable that skewness for both years was about 0.17, so the distribution was right skewed. In finance, it is commonly interpreted as receiving few extreme gains, and bearing small losses. The kurtosis was about five, which meant that the distribution had fat tails. Importantly, such deviations from normal distribution are common for financial data.

For profit, we obtained the following results (Table 2).

Table 2 Descriptive statistic for Operating profit

	Operating Profit 2008	Operating Profit 2009
N	751	751
Mean	2 338 098 399	1 843 400 000
Std Deviation	5 684 138 571	3 798 618 103
Skewness	-6.971	-4.892
Kurtosis	20.620	18.291
MIN	-2 519 900 000	-2 753 128 216
MAX	81 013 000 000	35 409 000 000

It had the mean of $2bn, maximum of $81bn and minimum of -$2bn. The distribution was negatively skewed (skewness=-6 for 2008 and -5 for 2009) and had extremely large kurtosis of nearly 20. Such results were compatible with the period of observation when companies faced extremely large volatility of profits. From the data, it was also notable that there were companies, which could make huge profits on the declining markets.

For stock returns, we obtained the following results (Table 3). Stock return was calculated as ratio of the price of the bottom of the crisis to the price of the start of the crisis minus one. The dates used for each country are presented in the Appendix B.

Table 3 Descriptive statistic for Stock return

	Stock Return
N	751
Mean	-0.464
Std Deviation	0.232
Skewness	0.313
Kurtosis	-0.121
MIN	-0.963
MAX	0.378

The mean was -46%, the minimum was -96% and the maximum was 38%. It was obvious that in the crisis period there were mostly negative stock returns due to huge decline on the markets. The standard deviation was about 23%. The skewness and kurtosis were equal 0.31 and -0.12 respectively, which were close to zero or normal distribution.

For market leverage we had the following results for pre-crisis year (Table 4).

Table 4 Descriptive statistic for Market leverage

	Market Leverage 2007
N	751
Mean	0.428
Std Deviation	0.25
Skewness	9.791
Kurtosis	26.251
MIN	0.000
MAX	2.929

The mean was 43%, the minimum was 0% and the maximum was 293%. The standard deviation was about 25%. The skewness and kurtosis were equal 9.7 and 26 respectively. Such results were compatible with the period of observation when companies faced extremely large volatility in leverage.

Chapter 3. Methodology

For hypothesis testing, a standard regression model were used, where the dependent variable were considered in the crisis period and independent ones in the pre-crisis period.

The regression equation was as follows:

Company performance_i,t = в₀ +в₁*Leverage_i,t-1+в₂*Corporate governance_i,t-1+в₃*Institution variable_i,t-1 +в₄*Leverage_i,t-1*Institution variable_i,t-1+ в₅*Leverage_i,t-1*Corporate Governance_i,t-1 + в₆*Institution variable_i,t-1*Corporate Governance_i,t-1 + в₇*Firm Size_i,t-1 +в₈*Industrial dummy_i+?_i

Depended variable - company performance was expressed as four indicators: Profit volatility, Return on assets (ROA) as an accounting measures of company's performance, stock return as market-based measure.

1) Profit volatility was used instead of ordinary profit because in crisis risk shifting (from shareholders to debtholders) can take place, so we can observe asset substitution. We measure it as a relative standard deviation (or coefficient of variation) of operating profit. Precisely, we took quarterly data, and compute the ratio of standard deviation to the mean. Such measure is commonly used in finance (Reed G., 2002).

2) ROA was used, because it reflects how managements uses assets to generate earnings. Additionally, we can observe the effectiveness of assets utilization in crisis.

3) Stock return was used to see how the market responses to changes in companies behavior during the crisis. Stock return was calculated as ratio of the price of the bottom of the crisis to the price of the start of the crisis minus one. The dates used for each country are presented in the Appendix B.

Regressors were conditionally divided into two groups: independent and control variables. The data for these variables were taken for the pre-crisis year (2007) in order to reflect the influence of the factors on company's performance in the crisis.

For independent variables, we used corporate government and institutional indices and leverage described in the Chapter 2.

For control variables, we used ratio of EBIT to total assets, logarithm of revenue as firm's size measure and dummy for industrial companies.

Chapter 4. Results

In the following Chapter main results are proposed. The quantitative part of the thesis was conducted in the SAS and R studio software.

4.1 Main Results

In the research, we run several regressions specifications to obtain the most robust results. The main focus was on the interaction term between leverage and corporate governance, and leverage and institutions. Additionally, we run regressions on interactions between corporate governance and institutions. We chose different combinations of factors and implemented model selection procedure. Overall, the results were controversial, since we did not obtain absolutely robust results in each regression. The significant specifications for each dependent variable are proposed in the Appendices C, D, E. For model selection, we used R-squared and F-value statistics. Regressions were tested with 10% and 5% significance level. Below we discuss the main results for each dependent variable.

4.2.1 Results for ROA

For each regression with ROA dependent variable as independent ones, we used market-based leverage, EBIT divided by assets, logarithm of Revenue and dummy variable as an indicator of Industrial sector. As for other factors, we used multiple combinations of variables describing corporate governance and level of country institutional environment. We obtained approximately the same results for ROA in 2008 and 2009, therefore, we present results only for 2008. We obtained relatively small R-squared for the models, therefore the model had a lot of residual variance, but some coefficients were significant.

Concerning value of leverage, the results were the same for all other variables combinations. To compute aggregate effect of leverage; we used average values of corporate governance indicators and institutional variables. The market value of leverage had negative effect on the ROA of the company in both years taking into account interaction with corporate governance and institutes. The leverage was significant despite of including or excluding the interaction terms with corporate governance or institutions. Following the obtained results we could conclude that market value of leverage was significant factor (at 5% significance level) in describing ROA during the crisis.

From the economic point of view, we saw that even in case of one standard deviation from the mean leverage, the coefficient was still statistically significant at 5% level. Precisely, if mean leverage increased for =0.25, ROA would decrease by 0.025 (for the first specification in Appendix C).

Turning to corporate governance indicators, we saw that generally they had positive, but small individual influence on ROA. Importantly, corporate governance indicators were statistically significant at 5% level only when interaction with leverage was included in the model.

Concerning institutional variables, we obtained controversial results. We found that Voice and Accountability, Political Stability No Violence, Regulatory Quality and Control of Corruption were insignificant in the most of the regressions specifications. We got only Government Effectiveness and Rule of law as significant factors. This could be explained by the fact that these indicators mostly influence on the business. Government Effectiveness in most specifications had slightly negative influence on the ROA, whereas Rule of law had positive influence. Interestingly, the indicator of Corruption level was insignificant, despite the intuitive influence on the financial indicators of the company. Overall, there were no strong evidence that all institutional variables were significant.

The main interest of the study was to explore the combined influence of factors on the dependent variable. Mostly in all cases, we got that interaction between corporate governance indicator and market value of leverage had negative influence on ROA. This meant that despite good level of corporate governance, the leverage had negative influence. We can conclude that leverage affected the performance of the company more during the crisis. Other important result was the interactions between institutional variables and leverage had positive influence, but were mostly statistically insignificant at 10% level. Concerning interaction of institutions and corporate governance, we also obtained mostly insignificant results except from combined influence of Rule of Law and ESG Score at 5% significant level, Rule of Law and Shareholders Score at 10% significant level. The significant factors had positive influence on the ROA.

We obtained results for Hypothesis 2, 3, 4. We investigated that in case of good corporate governance the aggregate leverage effect on ROA were negative and statistically significant at 5% level. Therefore, we can conclude that even in case of good corporate governance, leverage had negative effect on ROA during the crisis. We also found that in case of poor institutional environment, there were more influence of corporate governance on ROA, but results were insignificant at 10% level. We did not receive significant results for Hypothesis 4.

4.2.2 Results for Profit Volatility

As one of the dependent variables, we chose profit volatility to detect asset substitution pattern in the crisis period. For regressions, we used the same structure as for ROA. Results can be found in Appendix D. It was important that the data we used proposed that mostly companies had huge losses in profits during the crisis, so we considered profit volatility as a negative pattern of company's performance.

To compute aggregate effect of leverage we used average values of corporate governance indicators and institutional variables. The results for leverage were the same for all combinations of variables - had positive effect on profit volatility. It could be explained by the fact that the more leverage a firm had in the crisis year the more volatile it in further years. Such result could be treated as existence of asset substitution problem in crisis years. Aggregate effect of leverage was statistically significant at 5% level.

Turning to corporate governance indicators, given the mean value of leverage, they had mostly negative influence on profit volatility that meant that the higher a firm was evaluated for corporate governance the less volatile it were in crisis time. It can be explained by the fact that corporate governance prevented management from the actions to the detriment of the shareholders and strong corporate governance stood for more stable financial situation for a firm. Precisely, ESG Score and Management Score were statistically significant at 5% level. Shareholders Score was mostly significant and had positive influence on profit volatility.

Concerning institutional variables, we obtained controversial results. Given the mean value of leverage, we found that Voice and Accountability, Political Stability No Violence, Regulatory Quality and Control of Corruption were insignificant in the most of the regressions specifications. Government Effectiveness and Rule of Law were significant at 10% level and had negative influence on the profit volatility. Generally, the stronger institutional environment in a country of operation, the less volatile profit of a firm.

For combination of ESG Score and leverage we obtained significant positive estimates. For interaction between Shareholders and Management Scores we obtained negative estimates. It could be explained that combined positive effect of corporate governance and leverage increased profit volatility. The interactions of institutional indicators and leverage were mostly insignificant and we could not make any grounded conclusions.

We also obtained result for Hypothesis 2, 3, 4. We investigated that in case of good corporate governance the leverage effect on profit volatility was negative and statistically significant at 5% level. We also found that in case of poor institutional environment, there were more influence of corporate governance on profit volatility, but results were insignificant at 10% level. We did not receive significant results for Hypothesis 4. Such results perfectly corresponded to the results for ROA as dependent variable. Nevertheless, R squared for all models were less than 30%.

4.2.3 Results for Stock Return

For stock return, we used approximately the same structure for regressions, but did not include EBIT to total assets ratio as a control variable.

For stock returns, we again obtained statistically significant aggregate results for market value of leverage at 5% level. It had negative influence on the dependent variable, which, intuitively, due to the fact that stock price dropped stronger for those firms with higher level of leverage resulting in decrease in returns.

Given the mean value of leverage, all corporate governance indicators were significant at 10% level only in case of interaction term with leverage inclusion. This meant that solely corporate governance level had no influence on stock returns.

Concerning institutional variables, only Government Effectiveness and Rule of Law were significant at 10% level and had positive effect on stock returns.

Interaction terms of corporate governance and institutional variables were statistically insignificant. R squared for all models were less than 30%.

Overall, we obtained economically significant results for leverage influence on company's performance during the crisis. Namely, one standard deviation from the mean kept the leverage significant at 5% level. Precisely, if mean leverage increased for =0.25, Stock return would decrease by 0.0225 (for the first specification in Appendix E).

4.2 Discussion

In the research we tried to detect whether leverage, corporate governance and country-level institutions influence company's performance during the Global crisis.

From the regression analysis we got that leverage had negative influence on ROA and stock returns. Considering corporate governance influence, we obtained that it had positive influence on ROA and stock returns only in case of interaction term inclusion. The result for stock returns was confirmed by the results of Gupta (2013) meaning that corporate governance effectiveness was not incorporated into stock prices. For profit volatility, we obtained opposite results with aggregate positive influence of leverage and aggregate negative effect of corporate governance.

We found that almost in all cases leverage and corporate governance were substitutes for company's performance. This meant that huge corporate governance worsen the negative effect of the leverage. Therefore, the higher leverage, the smaller the effect of corporate governance. We explained it as following: in crisis times in case of high leverage, corporate governance did not help to overcome possible negative incentives of managers. We also got that interactions of institutions and corporate governance were mostly insignificant and we could not make any grounded conclusion about combined influence of the factors.

Considering sole effect of institutional variables we got that only two out of five indicators were significant - Government Effectiveness and Rule of Law.

Overall, the obtained results allowed us to make conclusions for stated hypotheses. Firstly, corporate governance affected positively company's performance in all cases. Therefore, the Hypothesis 1 was confirmed. Secondly, we saw that even in case of good corporate governance, the effect of leverage was negative, and the higher the score for corporate governance, the more negative the effect of leverage. Consequently, Hypothesis 2 was rejected. Thirdly, we saw that in poor legal environment, there was more positive influence of corporate governance on company's performance. However, we got this result only for ROA and obtained mostly insignificant results, therefore, the Hypothesis 3 was rejected. Fourthly, we did not obtain significant results for Hypothesis 4.

It is important to mention that our results could not correspond to literature we discussed in Chapter 1 due to the fact, that researchers used different indicators for institutional factors and for corporate governance.

...