Faculty of Social Sciences
Institute of Economic Studies
MASTER THESIS
Manipulation of Basel Risk Weights: Revising the Czech
Banking Sector
Author: Bc. Tereza Nováčková Supervisor: Petra Andrlíková MSc.
Academic Year: 2013/2014
1. Hereby I declare that I have compiled this master thesis independently, using only the listed literature and sources.
2. I declare that the thesis has not been used for obtaining another title.
3. I agree on making this thesis accessible for study and research purposes.
Prague, May 16, 2014
Signature
The author is grateful especially to her supervisor, Petra Andrlíková MSc. for guiding me through the topic and providing me with valuable comments and suggestions. I am also very grateful to my family and friends for their patience and support.
This thesis provides the empirical analysis of the second Basel regulatory frame- work implementation in Czech banks together with the economic performance in- spection of the Czech banking sector. With Basel II, banks face the possibility to implement internal models to calculate capital adequacy related to bank’s risk ex- posure. This possibility opens a discussion of its economic effect, transparency and potential misuse of the internal models. The empirical part of this thesis examines how the profitability and the reported riskiness change with internal models imple- mentation. Furthermore, the role of cost efficiency to bank’s profitability and risk adequacy ratio is evaluated. The panel data analysis of all Czech banks over a period 2006 to 2012 demonstrates that internal models for capital adequacy calculation in- crease bank’s profitability together with a decrease of the reported riskiness measured by risk weighted assets. Moreover, the cost efficiency has proven to be a significant indicator of both profitability and capital adequacy ratio.
JEL Classification G21, G28, G32
Keywords financial risk and risk management, regulation, banks
Author’s e-mail novackova.tereza@volny.cz Supervisor’s e-mail andrlikova@gmail.com
Abstrakt
Tato práce poskytuje empirickou analýzu českých bank po zavedení druhého basile- jského regulatorního rámce, spolu s analýzou ekonomické výkonnosti českého bankovního sektoru. V rámci Baselu II jsou banky vystaveny možnosti volby zavedení interních modelů pro výpočet kapitálové přiměřenosti. Tato možnost otevírá diskusi týkající se ekonomických dopadů, transparentnosti a možného zneužití zavedení interních mod- elů. V empirické části této práce jsou zkoumány dopady tohoto zavedení na ziskovost banky a na její vykazované rizikové ukazatele. Dále je v této práci zkoumána role nákladové efektivity na profitabilitu a kapitálovou přiměřenost bank. Analýza pan- elových dat český bank v období let 2006 až 2012 ukazuje, že zavedení interních modelů zvyšuje ziskovost bank a snižuje rizikově vážená aktiva. Mimo to, nákladová
Klasifikace JEL G21, G28, G32
Klíčová slova finanční riziko a řízení rizik, regulace, banky E-mail autora novackova.tereza@volny.cz
E-mail vedoucího práce andrlikova@gmail.com
v
List of Tables viii
Acronyms ix
1. Introduction 1
2. Banking regulation 4
2.1. Capital regulation . . . 4
2.2. Credit risk definition . . . 5
2.3. History of the Basel regulatory framework . . . 6
2.3.1. Credit risk regulation . . . 6
2.3.2. Basel I . . . 6
2.3.2.1. Basel I pillars . . . 7
2.3.2.2. Main aspects of Basel I . . . 7
2.3.3. Basel II . . . 8
2.3.3.1. First pillar . . . 8
2.3.3.2. Second pillar . . . 13
2.3.3.3. Third pillar . . . 14
2.3.4. Basel III – more of the same but better? . . . 15
2.4. Problems with the risk weights . . . 16
2.5. Basel standards in the Czech Republic . . . 18
3. Empirical Analysis 20 3.1. Literature review . . . 20
3.1.1. Basel II implementation . . . 21
3.1.2. Bank economic performance . . . 24
3.2. Hypotheses . . . 27
3.3. Data . . . 28
3.4. Methodology . . . 30
vi
3.5. Model 1 . . . 32
3.5.1. Dependent variable . . . 33
3.5.2. Independent variables . . . 34
3.5.3. Model 1 testing . . . 39
3.5.3.1. Pooled OLS . . . 39
3.5.3.2. Fixed Effects . . . 40
3.5.3.3. Random Effects . . . 42
3.5.4. Results . . . 42
3.6. Model 2 . . . 44
3.6.1. Dependent variable . . . 45
3.6.2. Testing . . . 47
3.6.2.1. Pooled OLS . . . 47
3.6.2.2. Fixed Effects . . . 48
3.6.2.3. Random Effects . . . 49
3.6.3. Results . . . 50
3.7. Discussion . . . 51
4. Conclusion 54
References 57
A. Content of Enclosed CD 61
vii
2.1. Basel II - Pillar 1 options . . . 9
2.2. Banks distribution of capital adequacy to credit risk exposure calcu- lation . . . 19
3.1. List of Czech banks . . . 29
3.2. List of Czech branches of foreign banks . . . 29
3.3. Statistical description of ROA . . . 34
3.4. IRB implementation of Czech banks . . . 35
3.5. Statistical description of CIR . . . 36
3.6. Statistical description of BS . . . 37
3.7. Statistical description of DE . . . 37
3.8. Independent variable summarization . . . 38
3.9. Model 1, FE method comparison . . . 41
3.10. Estimated coefficients of Model 1 - overview . . . 43
3.11. Estimated coefficients of Model 1 . . . 44
3.12. Statistical description of CAD . . . 46
3.13. CAD summarization . . . 46
3.14. Pairwise correlation matrix . . . 47
3.15. FE method comparison of Model 2 . . . 49
3.16. Estimated coefficients of Model 2, overview . . . 50
3.17. Estimated coefficients of Model 2, detailed . . . 51
viii
AIRB Advanced Internal Ratings-Based Approach FIRB Foundation Internal Ratings-Based
BCBS Basel Committee on Banking Supervision
CAD Capital Adequacy
CCF Credit Conversion Factor
CF Cash flow
CIR Cost to income ratio
CNB Czech National Bank
DE Debt to equity ratio
EAD Exposure at Default
FE Fixed Effects
FIRB Foundation Internal Ratings-Based FSA Financial Service Authority
GLS Generalized Least Squares
IFRS International Financial Reporting Standards IID Independent and identically distributed
IRB Internal Rate Based
LGD Loss Given Default
LSDV Least Squares Dummy Variables
M Maturity
OLS Ordinary Least Squares
PD Probability of Default
RE Random Effects
ROA Return of Assets
RWA Risk-Weighted Assets
SA Standardized Approach
US GAAP Generally Accepted Accounting Principles
ix
Author: Bc. Tereza Nováčková Supervisor: Petra Andrlíková MSc.
Defense Planned: June 2014
Proposed Topic:
Manipulation of Basel Risk Weights Topic Characteristics:
Regulation of the banking sector, actual topic, which is very important nowadays from the ex post view on the recent financial crisis. In Europe, most of the banks have adopted Basel II Accord, which give them opportunity to use internal rating (IRB) approaches instead of standardized one.
In this Thesis, potential manipulation of Basel risk weights is scrutinized with focus on the European countries, especially the Czech Republic. The situation in Czech banking sector seemed to be more stable than those in the rest of the Europe in crisis times, on the other way the question about the manipulation in compounding risk weights under Basel II approach in Czech banks is still open. Not only from the reason that Czech banking sector is integrated within the European one, but also the most of the Czech banks are owned by foreign institutions belonging to less stable financial environment, and thus the issue of potential manipulation in the Czech banking environment should be examined.
Firstly, considering the main European banks, this Thesis tries to show how timing of IRB approach implementation in the banks influenced their cost effectiveness. Then the Blum’s (2008) hypothesis about reported riskiness declining upon IRB adoption in case of the Czech Republic will be examined. Finally, I try to find the evidence about insufficiency of risk-based indicators as the predictor of bank distress.
Hypotheses:
1. Timing of IRB approach implementation has significant impact on the cost effectiveness of the bank.
2. Average reported riskiness (RWA/TA) in Czech banks declined upon IRB adoption.
3. Risk-free ratios are better predictors of bank distress than those including risk weights.
Methodology:
In the first part, the relationship between timing of advanced IRB approach and cost effectiveness of the European bank is examined. To solve this problem, the probabilities models will be used. Timingof adoption IRB approaches implementation will figure as one of the explanatory variables for cost effectiveness of the European banks.
In the second part, cross-sectional panel regression analysis will be applied in order to explain movement of risk-adjusted ratios after adoption IRB approaches by Czech banks. In this model, risk-weighted asset to total asset ratio as dependent variable will be explained by the year of IRB approach adoption, and by other different control variables for bank specifications. Moreover, the causes of potential change in reported riskiness will be examined.
The last but not least section of the Thesis relates to the previous part. Specific statistical tests should show inappropriateness of risk-weighted indicators as distress predictors. To be more specific, Czech bank will be compared to those of foreign which reported similar levels and trends in riskiness, but which end up with financial difficulties.
All data will be taken from banks' annual reports and from data reported by national banks.
3. Literature review
4. IRB implementation and cost effectiveness 5. Blum's hypothesis for the Czech Republic 6. RWA as insufficient indicator
7. Empirical evidence 8. Conclusion
Core Bibliography:
1. Mariathasan, M., & Merrouche, O. (2012). The Manipulation of Basel Risk-Weights.
Evidence from 2007-10. Department of Economics Discussion Paper Series, (621).
2. Le Leslé, V., & Avramova, S. (2012). Revisiting risk-weighted assets. International monetary fund (IMF).
3. Resti, A., & Sironi, A. (2007). The risk-weights in the New Basel Capital Accord: Lessons from bond spreads based on a simple structural model. Journal of Financial
Intermediation, 16(1), 64-90.
4. Claessens, S., Dell’Ariccia, G., Igan, D., & Laeven, L. (2010). Cross‐country experiences and policy implications from the global financial crisis. Economic Policy, 25(62), 267-293.
5. Demirgüç-Kunt, A., Detragiache, E., & Merrouche, O. (2010). Bank capital: lessons from the financial crisis. IMF Working Papers, 1-35.
6. Haldane, A. G. (2011). Capital discipline. Speech at the American EconoBenink, H., Daníelsson, J., & Jónsson, Á. (2008). On the role of regulatory banking capital. Financial markets, institutions & instruments, 17(1), 85-96.mic Association, 9.
7. Koziol, C., & Lawrenz, J. (2009). What makes a bank risky? Insights from the optimal capital structure of banks. Journal of Banking & Finance, 33(5), 861-873.
1. Blundell-Wignall, A., & Atkinson, P. (2009). Origins of the financial crisis and requirements for reform. Journal of Asian Economics, 20(5), 536-548.
Author Supervisor
Introduction
The history of bank regulation and deregulation is countercyclical to the economic cycle. Times of regulation and deregulation go afterwards to times of economic crisis and subsequent economic prosperity (Balthazar, 2006). The financial crisis starting with bankruptcy of Lehman Brothers’ bank in September 2008 instigated the Basel Committee on Banking Supervision to revise its regulation for the second time.
Lehman Brothers’ bankruptcy has started universal financial instability sequencing in series of economic failures and financial distress over the cross-national banking sector. Banking sector in the Czech Republic is considered to be stable compared to the rest of the Europe, without any bank’s bankruptcy in times of the crisis. With the implementation of second Basel Accord in 2007, banks were exposed to the possibil- ity to design their own models for calculation of capital adequacy related to bank’s risk exposure. However, before implementation of the third regulatory Accord pre- pared by the Basel Committee on Banking Supervision, practical consequences of the second Basel Accord still remain to be assessed.
The objective of this thesis is to provide the empirical evidence of the second Basel regulatory approach implementation in Czech banks together with the economic per- formance inspection of the Czech banking sector. In this thesis, four different hy- potheses are tested; two hypotheses relate to the profitability of the Czech banks, and other two inspect the reported riskiness of the Czech banks.
First of all, the relationship of the implementation of the Internal Ratings-Based models and banks profitability is examined. We assume that the bank’s profitability tends to be higher with advanced methods of capital adequacy calculation. A bank, as a profit-maximizing subject, is supposed to be rewarded by higher profit after investment in employment of Internal Ratings-Based models. Our findings are rel-
evant mainly to the banks and financial institutions intending to implement Internal Ratings-Based models.
Second hypothesis inspects the correlation between bank’s efficiency and its prof- itability. The profitability of a bank is represented by the operational efficiency cov- ered by the cost to income ratio. The correlation is assumed to be positive according to the empirical researches.1 The negative relationship can be interpreted as inap- propriateness of cost to income as a proxy for bank overall efficiency. The suitability of the cost to income ratio as the indicator of the bank efficiency can provide com- parability of the efficiency among particular banks and other financial institutions.
The following hypothesis analyzes the impact of Internal Ratings-Based models implementation on the Capital adequacy ratio representing riskiness of a bank. Re- ported riskiness is claimed to decrease and the capital adequacy ratio to increase with advanced techniques of capital adequacy calculation. This statement is sup- ported by empirical researches performed by Blum and Mariathasan and Merrouche, who detected potential misbehaviour in the cross-country comparison (Blum, 2008), (Mariathasan and Merrouche, 2013). The rejection of the hypothesis of the relation- ship between reported capital adequacy and Internal Ratings-Based models imple- mentation implies the possibility that the banks incline to the riskier assets’ portfolio using advanced models for capital calculation. The result can be interesting from the supervisor’s point of view, since the regulation might have the opposite consequences instead of the original intentions.
The last but not least hypothesis examines the relationship between the cost effi- ciency and Internal Ratings-Based models implementation. If we approve the pos- itive relationship, we can conclude that the operational efficiency of a bank can be generalized to the overall efficiency, or at least efficiency of the risk-exposure treat- ment. The result is relevant both to the investors and to the client of the bank, who can make a picture of their invested capital management.
The structure of this thesis is as follows: Chapter 2 introduces the theoretical back- ground of the banking regulation focusing on the credit risk regulation. At first, credit risk is defined, followed by the description of the history of the Basel Accords. At the end, the problems with the risk weights are stated continued by the implemen- tation of the Basel standards in the Czech Republic description. In Chapter 3, the review of the related literature is summarized and the hypotheses are described in detail. Henceforward testing of the hypotheses is performed on the panel data of
1e.g. research of Christian, Moffitt and Suberly or Mathuva (Christian et al., 2008), (Mathuva, 2009)
the Czech banks’ relevant variables. The Chapter is ended by the discussion of the results. Chapter 4 concludes.
Banking regulation
2.1. Capital regulation
In the theoretical world of perfect competition, where all of its assumptions are ful- filled, there is no need of banking capital regulation. According to Modigliani and Miller propositions, the value of a company, a bank in our case, is not dependent on its capital structure (Santos, 2001, based on Modigliani and Miller, 1958). However, in the real world, a firm faces market imperfections, such as transaction costs, taxes, financial distress costs or information asymmetry (Santos, 2001).
Moreover, financial institutions deal with other concerns. One of the most com- mon, and also most probable, is the so called bank run. “As long as the bank keeps enough reserves to cover the withdrawals of the depositors who actually need their money, which is much less than the total amount of the deposits, the system can func- tion smoothly and efficiently” (Rochet, 2009). The run on a bank can be determining factor of bank to survive financial distress or not (Rochet, 2009). Another problem which the bank faces arises from its role of market liquidity provider. The impor- tance of this task is trying to be secured through compulsory deposit insurance, or the position of the national central bank as lender of the last resort (Bhattacharia, 1998).
To realize the importance of the banking sector, the capital regulation and subse- quent supervision of the banks are the primary objectives of the policy interventions in the economy (Milne and Whalley, 2001). The Basel Committee on Banking Su- pervision holds the position of the international policy maker for nearly 26 years from the first proposal of the first global regulatory supervision (BCBS, 1988). The development of the policy approaches is described in the following chapters.
2.2. Credit risk definition
At the very beginning of this section, definitions connected with the stated problems are introduced; mostly definitions of credit risk and other related risks. The first definition represents a wide point of view on the credit risk theory. Duffie and Sin- gleton are dealing with pricing, measurement, and managing the credit risk. Credit risk can be described as “... the risk of changes in value associated with unexpected changes in credit quality” (Duffie and Singleton, 2012). This definition seems to be very wide, and the linkage to the banking regulation is not straightforward. More evident connection to the topic of this work is provided by the following definitions.
Different perspective represents definition by Bielecki and Rutkowski in their pub- lication, which describes credit risk modelling, valuation, and hedging. From their point of view, credit risk can be understood as generalization of default risk, which the authors characterize as “... a possibility that the counterparty in a financial con- tract will not fulfil a contractual commitment to meet her/his obligation stated in the contract. ” (Bielecki and Rutkowski, 2002). The credit risk is then defined as
“risk, associated with any kind of credit-linked events, such as: changes in the credit quality, (including downgrades or upgrades in credit ratings), variations of credit spreads and the default event.” (Bielecki and Rutkowski, 2002). These definitions are presented since the term of default is closely related to the credit risk. Default and probability of default afterthought create a crucial part of the second Basel Ac- cord and it will be more discussed in the following section.
The last, but not least description of the termcredit riskis provided by the Czech National Bank (CNB) and the definition is thus adjusted for banks’ needs. The term credit risk defined by CNB can be comprehended as “Risk of the bank from losses which result from the failure of a counterparty to meet its obligations in accordance with the terms of a contract under which the bank has become a creditor of the coun- terparty” (CNB, 2013). The reason why this definition is important is that the thesis focuses on the credit risk in the Czech Republic and the implemented regulation by the CNB.
2.3. History of the Basel regulatory framework
2.3.1. Credit risk regulation
Before starting the description of banking regulation under the Basel Committee on Banking Supervision (BCBS), it is important to integrate the following section into the structure of this Thesis. In the previous part, the credit risk that banks are facing was introduced, the next part continues with origin and evolution of banking regula- tory framework by the BCBS.
Regulatory standards known as Basel I, II, and III have been published by Basel Committee on Banking Supervision (BCBS, 1988, 2004 and 2011). Basel Commit- tee itself originated from the Committee on Banking Regulation and Supervisory Practices founded in October 1974 as the answer on the collapse of the Bretton- Woods system and its consequences on the foreign exchange exposures. The original Committee was created by the governors of the central banks and supervisory author- ities from G10 countries and the Netherlands (BCBS, 1988), whose “...aim was and is to enhance financial stability by improving supervisory knowhow and the quality of banking supervision worldwide.” (BCBS, 2013)
The Bretton-Woods system was officially wound up in 1973. In the same year, the European Commission issued a new Directive that was the first true step in the deregulation of the European banking sector. From that moment, there was decided to apply “national treatment” principles, which meant that all banks operating in one country were subordinated to the same rules (even if their headquarters were lo- cated in another European country), which ensured a “level playing field” (Balthazar, 2006).
2.3.2. Basel I
The first important collective action of BCBS came in December 1987, when the first capital adequacy requirement was formulated. It takes a part in the Basel Cap- ital Accord, known as Basel I (BCBS, 2013). All G10 central banks governors and Spain agreed to start proposed standards implementation in July 1988 to stabilize international banking system through cross-border regulation (Blum, 2008).
It is very important to mention that the first Basel Accord considers only credit risk exposure and it was proposed to be implemented in international banks from
developed countries (BCBS, 1988). The first Basel Capital Accord requires 8 % as the minimum capital to risk-weighted assets. This condition should provide the safety capital buffer for banks in case of potential credit risk exposure (Mejstřík, Pečená and Teplý, 2008). Moreover, the BCBS challenges national supervisory authorities to determine additional capital to cover also other risk, such as interest rate risk or investment risk on securities (BCBS, 1988). All involved countries were supposed to implement new capital adequacy requirement till the end of 1992.
2.3.2.1. Basel I pillars
The first Basel Accord is based on four pillars, which are following titled; The Con- stituents of Capital, Risk Weighting, Target Standard Ratio, and Transitional and Implementing Agreements. The first pillar defines two types of capital, Tier 1 Cap- ital and Tier 2 Capital. This type of capital, especially Tier 1, is considered by the Committee as a good indicator of financial wealth, which is easily readable from the financial statements and comparable across countries. The second pillar defines five asset risk classes; 0, 10, 20, 50, and 100 % to provide“international comparisons between banking systems whose structures may differ.” (BCBS, 1988)
The third pillar settles the minimum capital adequacy ratio, i.e. target standard ratio of capital to weighted assets, to be 8 %. Moreover, one half of this 8 % should be covered by Tier 1 Capital. This claim related only to internationally active banks from Member countries and became active at the end of 1992.
The last fourth pillar concerned with transition and implementation of the regula- tory framework. In transition process, obligatory progressive percentages for partic- ular years up to 8 % in 1992 are defined. The whole document is ended up with the announcement that the involved countries in the first Basel Accords should start the implementation of all described arrangements in the national legislation as soon as possible to make legislation compatible with Basel I, national laws and other multi- national agreements (BCBS, 1988).
2.3.2.2. Main aspects of Basel I
It is important to keep in mind that the first Basel Accord was intended to be imple- mented only in G10 countries, i.e. in highly developed countries with banks having great impact to the international banking sector.
As stated above, original Basel I regulation relates only to credit risk events. How- ever, in January 1996, Market Risk Amendment to the Capital Accord was issued to define also regulatory approach to the market risk exposure. This amendment was intended to come into force till the end of 1997 (BCBS, 2013). It is important to mention that in this adjustment, banks were allowed to use internal model for the first time. Banks were accepted to use Value-at-Risk models to calculate capital needed to cover market risk exposure.
2.3.3. Basel II
Two years after the implementation of regulatory requirements relating to market risk the Committee started to work on the Revised Capital Framework which is com- monly known as Basel II (BCBS, 2013). New regulatory framework was published in June 2006 and was applicable six months later, since January 2007 (Mejstřík et al., 2008). Main goals of the new regulatory framework are the following. “Increas- ing the quality and the stability of the international banking system, creation and maintaining a level playing field for internationally active banks, and promotion the adoption of more stringent practices in the risk management field.” (Batlhazar 2006) To achieve these goals, Basel II stands on the three pillars structure. The first pillar defines minimal capital adequacy requirements to risks, which banks are exposed to, than the second pillar describes supervisory review process and the last pillar relates to market discipline of the banks.
2.3.3.1. First pillar
The first pillar extends Basel I requirements on the minimal capital needed to cover risk exposures of a bank. In Basel II, also operational risk was added to credit and market risk. A stance on the market risk capital requirement has not almost changed.
On the other hand in case of credit risk, the updates which came with the second Basel Accord framework can be viewed from two perspectives. From the first side, the attitude to methodology of computation determination of capital requirement went through the biggest changes from the reason that the banks’ possibilities have ex- panded; each bank can choose from the standardized approach on the one side and from internal approach from the other side. On the other hand, in Basel II, capital adequacy requirements calculations are again based on Risk-weighted assets, as in the previous Basel Accord. In this approach banks can chose from three methods how to compute capital requirements for all three risks (BCBS, 2004).
Table 2.1.: Basel II - Pillar 1 options
-Capitalconsumption+ Credit Risk- Credit Risk-
+Complexity-
unstructured exposures securitization Operational risk
Standardized Standardized Basic Indicator
Approach Approach Approach
Internal Ratings-Based - Ratings-Based Standardized
Foundation Approach Approach Approach
Internal Ratings-Based - Internal Assessment Advanced Foundation Advanced - Approach Measurement
Approach Supervisory Formula Approach
source: Balthazar, 2006
This table captures banks’ possibilities how to compute capital needed to cover potential risks banks are exposed to. Balthazar points to the trade-off between cap- ital consumption and complexity of the models. As a bank chooses more complex method for the capital computation, the less capital a bank needs for the same port- folio assets on its balance sheet. A bank undergoes this decision at each risk type determined by the second Basel Accord, i.e. credit risk, operational risk, and market risk (Balthazar, 2006).
2.3.3.1.1 Credit risk
In case of credit risk exposure, a financial institution is obliged to choose from three methods of capital adequacy computing. Each method includes different level of re- quired complexity of internal models. Banks have three available approaches how to determine capital requirement for capital risk exposure; the Standardized Approach (SA), the Foundation Internal Ratings-Based Approach (FIRB), and the Advanced Internal Ratings-Based Approach (AIRB). The Standardized Approach is very sim- ilar to capital adequacy calculation applied in Basel I. In SA, ratings from external agencies are used and there are more risk classes than in Basel I (BCBS, 2004).
With the possibility of choosing a particular option of capital adequacy calcula- tion, a bank faces the trade-off between developing and introducing its own internal models and higher capital requirements. “The more advanced(models)are designed to consume less capital while they impose heavier qualitative and quantitative re- quirements on internal systems and processes” (Balthazar, 2006). This trade-off and possible manipulation within Basel II implementation is discussed in this Thesis.
However, it is important to keep in mind that all methods of calculations have to be
approved by regulatory body, i.e. by their national banks.
Standardized Approach
The first possibility of capital needed to cover credit risk exposure calculation can be understood as widening the computing scheme from previous Basel Accord. The main characteristics of the Standardized Approach can be seen in the extension of credit classes’ number where corresponding ratings are set by external agencies. The approach has the advantage that no internal models have to be developed (BCBS, 2001b).
In the Standardized Approach, the most important matters are determining risk weights and defining the type of the counterparty followed by setting of the counter- party’s rating class (Balthazar, 2006). Subsequently, the corresponding risk weights are assigned to each bank’s asset. The risk weights are mostly 0, 20, 50, 100 and 150
% from the value of the asset. For example, for the corporate counterparty with A+
to A- rating class belongs risk of 50 %.
As the previous Basel Accord, this calculating approach includes both on- and off- balance sheet assets. To evaluate off-balance sheet assets, the Credit Conversion Factor is used as well as in the previous Basel approach. To off-balance sheet items can be classified to risk-weights of 0, 20, 50, and 100 % (BCBS, 2001b).
For its simplicity, transparency, and almost zero implementation costs there was anticipated that the Standardized Approach will be chosen as the possibility of com- puting capital adequacy by many banks (Balthazar, 2006).
Foundation Internal Ratings-Based Approach
On the contrary to the previous approach, the following two ways of calculation are connected by the need of using Internal Ratings-Based models (IRB models). In gen- eral, “Internal Ratings Base methods are based on measures of unexpected losses (UL) and expected losses (EL). The risk-weight functions produce capital require- ments for the UL portion” (BCBS, 2001c). These models are developed by banks themselves and the risk weights are calculated by them instead of using risk weights determined by external agencies. Larger banks have incentives to develop their own risk models, which supposed to be suitable for each of the bank. A bank reduces its capital adequacy requirement this way (BCBS, 2001b).
The first Internal Ratings-Based approach, the Foundation approach allows banks to calculate counterparty’s’ Probability of Default (PD) by the internal model. The Czech National Bank introduces the occurrence of the credit default defined as the moral breaching of the debtor payment. In the connection with the banking regula- tion, we can also meet the term “failure of the debtor”, which can be realized in the moment, when repaying debtor’s obligation in proper way and without delay starts to be highly improbable (CNB, 2013).
As already discussed at the beginning of this chapter, we can define at the PD as “The probability that the counterparty will not meet its financial obligations.”
(Balthazar, 2006). Thus each bank which decides to use Foundation Internal Ratings- Based method for determining the capital adequate to the credit risk exposure can assess counterparty’s probability of default. After PD estimating, the bank calculates each counterparty’s rating and subsequently the capital requirement.
Advanced Internal Ratings-Based Approach
As we continue from the first Basel Accord, we can see more and more sophisticated, complex and complicated way of the capital adequacy assessing. The Advanced In- ternal Ratings-Based models are the most sophisticated ways how to determine cap- ital for credit risk exposure. There are six categories of the exposures the bank is facing; corporate, sovereign exposure, bank exposures, retail exposures, equity ex- posures, purchased receivables exposures (Balthazar, 2006). Moreover at this ap- proach, not only Probability of Default parameter is estimated, but also Loss Given Default (LGD), Exposure at Default (EAD), and Maturity (M) are now assessed by the internal models.
The Czech National Bank defines Loss Given Default as the ratio of the exposure loss during counterparty’s default to the total owed amount (CNB, 2013). With the linkage to the others parameter, the BCBS determines the Loss Given Default the Committee: “LGD must be measured as the loss given default as a percentage of the EAD. Banks eligible for the IRB approach that are unable to meet these additional minimum requirements must utilize the foundation LGD treatment described above.”
(BCBS, 2001c). This ratio relates to the measurement of expected and unexpected losses, and final capital adequacy is derived from it afterwards.
The indicator Exposure at Default can be described as “The expected amount of exposure at the time when counterparty defaults (the expected drawn-down amount
for revolving lines or the off-balance sheet exposure × its CCF)” (Balthazar, 2006).1 The Basel Committee documentation determines that the Exposure at Default can be assessed by “...gross of specific provisions or partial write-offs.” (BCBS, 2001c)
The last, but not least indicator estimated from the internal models is the Maturity.
For each credit risk exposure, the maturity has to be calculated. The calculation of the maturity is described by Balthazar as
Maturity= ∑t∗CF
∑CF (2.1)
where
CF Cash flows (interest and capital) t Time of the cash flow (in years) (Balthazar, 2006)
2.3.3.1.2 Operational risk
Capital requirements for covering operational risk exposure appeared newly in sec- ond Basel Accord. The BCBS defines operational risk as “the risk of direct or indi- rect loss resulting from inadequate or failed internal processes, people and systems or from external events” (BCBS, 2001a). Thus, under operational risk we can under- stand all risks arising from the operations of the bank, such as failure of used software or risk from loss because of natural disasters.
As in previous case of credit risk, a financial institution can choose from three possible methods how to calculate minimal capital requirements which should cover operational risk exposure. The first possibility is Basic Indicator Method, second is the Standard Approach, and the last possibility is the Internal Measurement Ap- proach (BCBS, 2001a). The principal aspect in decision which method to use is the nature of the bank. As well as in the case of credit risk management, calculating methods starts from the simplest and the most transparent one, but they continue with more comprehensive, suitable and thus more sensitive methods of computation, with higher requirements on the regulator supervision (Mejstřík et al., 2008).
In the Basic Indicator Method, level of capital required for covering operational risk is a fixed percentage from the gross income of a particular financial institution
1CCF is a Credit Conversion Factor.
(BCBS, 2001a). The Standard Approach is also very similar. There is main differ- ence in the additional calculation of the percentage of different financial indicator to compound the required capital buffer. Calculation is based on the bank distri- bution to a specific class according to the indicators relating to business profile or business unit of a bank (BCSB, 2001a). If a bank decides to implement the Internal Measurement Approach for calculation of the capital requirement for covering of the operational risk, it is demanded to meet particular qualitative (such as independence of operational risk management or regular audit) and quantitative (reliable internal and external data collection) requirements. All internal approaches have to be also approved by the regulatory supervisor (Mejstřík et al., 2008).
2.3.3.1.3 Market risk
The necessity to cover market risk with additional capital was presented firstly in addition of Basel I. Basel II approach allow banks to choose between two approaches;
between the Standardized Approach and using of internal models (BCSB, 2001a).
To summarize first pillar of Basel II, the bank has to hold capital buffer or reserves in the amount to be able to cover all three types of risk which the bank is exposed to.
The level of reserves or capital requirement is calculated as follows:
Reserves = 0.08∗Risk Weighted Assets +Opertaional Risk Reserves
+Market Risk Reserves (2.2)
(Balin, 2008)
2.3.3.2. Second pillar
The second and the third pillar should perform the support of the first pillar, where the major message of the second Basel Accord about capital requirements for different risks is held. We can see this also from the structure of the document, where pillars II and III take place only on 40 out of 350 pages of Basel II Accord (BCBS, 2004). The second pillar of Basel II structure determines supervisory review process, which in- cludes assessment of risk and capital adequacy of the individual banks and continual
communication with the banks to take control over the internal processes of defining capital requirements (BCBS, 2004). The idea behind this pillar is very closely con- nected to the previous pillar, where the possibility of using internal models requires supervisory monitoring. The regulator can assess required capital additionally if it considers compounded capital to be insufficient (Mejstřík et al., 2008).
The whole second pillar is based on the hypothesis that, ceteris paribus, a bank with higher tendency to risky behaviour should have higher capital requirement than those of capital averse. It should be kept in mind that the minimum capital adequacy ratio is eight percent. Additional capital is required according to the bank risk profile.
This attracts the attention to the central banks as supervisors, because they become more responsible with Basel II framework (Mejstřík et al., 2008).
2.3.3.3. Third pillar
The third pillar’s title is Market discipline. The motivation for the third pillar is the following: “The Committee emphasizes the potential for market discipline to rein- force capital regulation and other supervisory efforts in promoting safety and sound- ness in banks and financial systems.” (BCBS, 2001b) The main attribute of the third pillar is the bank transparency. Each financial institution should provide sufficient information about its capital level to keep the market in the stable and predictable conditions (BCBS, 2001b). The claim on the information should be even more im- portant in case of banks which use internal models to determine capital adequacy (Mejstřík et al., 2008).
The Committee recommends making capital stability indicators such as levels of Tier 1 and Tier 2 capital, capital adequacy ratio including risk-weights, and capital cushions for covering other credit, market, and operational risk information public at least twice a year (BCBS, 2001b). This attitude is intended to be leverage for the stock- and stakeholders, and also for clients. “Basel II hopes to empower sharehold- ers to enforce discipline in the risk-taking and reserve-holding methods of banks, where banks seen to hold too few reserves and take on too much risk are punished by their own shareholders for doing so.” (Balin, 2008)
It is important to mention that the BCBS works on the self-imposed principles. The Committee decisions and statements themselves have no legal force (BCBS, 2013). It can be understood as a meeting place for the suggestions and proposals relating to the banking regulation, then as a collector of the impulses and incentives, and finally, as a proposer of the most appropriate solutions. In general, it depends on the country or
community member itself if it follows the regulatory proposals, and implements it to its legislation. What really matters is the membership in a multinational association.
“In practice, the Accord will be mandatory for all banks and securities firms, even at the national level, in many countries.” (Balthazar, 2006)
2.3.4. Basel III – more of the same but better?
At the end of year 2010, the member states from G20 gave the approval for the Basel III regulation to be implemented in the sovereign legislation. According to King and Tarbert, the new regulatory framework does not continue on the way which has been started by the previous two Accords (King and Tarbert, 2010). On the other hand, the main part of the new regulations is created only by the higher restrictiveness of the old parameters: ”... the essence of these reforms is nicely described; more of the same – and better.” (Haldane, 2011)
The novelty of the third Basel framework comprises of the macroprudential re- quirements. The BCBS established a set of measures on the macroeconomic level, such as countercyclical buffer or a universal leverage ratio, to control the systemic risk within the international financial market (BCBS, 2009). The procyclicality of the Basel II belongs to the main criticism from the previous form of regulation.2
According to the Basel Committee, the Basel III has appeared as the answer to the recent financial crisis. The Committee considers good work of the financial sec- tor as a crucial element for the economic recovery and consequent economic growth (BCBS, 2009). The key component of the third Basel regulation accord is the re- inforcement of the global capital framework. As already mentioned, this objective should be reached by both micro- and macroprudential requirements. The necessity of quality, consistency and transparency improvement of the capital base determina- tion and enhancing risk treatment belong among the microprudential ones. On the other hand, as mentioned above, new macroprudential claims comprise leverage ratio and procyclicality (BCBS, 2010).
To focus on the new regulatory requirement more closely, the microprudential requirements should be strengthen by the simplification of the all Tier capital def- initions, which should also provide better transparency of the capital monitoring.
Another important improvement follows in topic of counterparty credit risk deter- mination. There are plenty of suggestions how to determine credit risk exposure
2The criticism by Heid or Blum and Hellwig is described in literature review more detailed (Heid, 2007), (Blum and Hellwig, 1995).
more accurately, for example changing the determining of the exposure at default to endorse management practises for robust calculation of collateral (BCBS, 2009).
Focusing on the macroprudential requirements, the leverage ratio should accom- pany the risk-weighted assets ratio to cover both risk, one arising from internal be- haviour of the bank, and the second determining risk within global banking sector.
The motivation for another ratio is the excessive leverage of banks on- and off-balance sheet articles. The leverage ratio should be also calculated with respect to structure of the banks portfolio, for example newly certain off-balance sheet items will be added in the leverage ratio calculation with 100% requirement of capital cover (BCBS, 2009).
The procyclicality of the regulatory standards should be reduced by implemen- tation of the requirements to calculate particular parameters for determining capital adequacy using long-term data. Another method how to smooth the effect of the eco- nomic cycle is to employ downturn loss-given estimates and to implement additional loss estimates to capital adequacy calculation (BCBS, 2009).
There are plenty of theoretical evaluating researches of potential effects of Basel III, but the appropriateness of the third Basel Accord swill only be truly assessed ex post after the several years after actual implementation.
2.4. Problems with the risk weights
The complete overview of potential problems with RWA determination is provided by the study of Le Leslé and Avramova. The authors present list of problems with RWA together with their consequences and possible solutions. The problems are presented on a qualitative base without exact evaluating of the impacts.
The problems with the risk-weighted assets (RWA) calculation are divided by the authors into three groups with respect to the regulator, the bank and the investor.
From the regulator’s point of view, the main interest can be regarded to incorrect risk assessment of both on- and off-balance sheet items, which relates to the underestima- tion of this risk. This may lead to the exaggeration of the reported regulatory capital, which is considered as a reliable indicator of the resistance to the financial distress, and to the misleading information about the performance of a bank afterwards. An- other concern relates to the capital adequacy reporting. In general, reported value of RWA ratio increases despite of the overall unfavourable economic conditions. This may be the consequence of the optimization, data adjustment, or calculating methods
modification (Le Leslé and Avramova, 2012). This observation is in correspondence with the Blum’s theoretical study and empirical research of Mariathasan and Mer- rouche (Blum’s, 2008), (Mariathasan and Merrouche, 2012). This problem is also empirically examined in this Thesis for the Czech banking sector.
Another disturbance from the regulator’s point of view is prospective pro-cyclicality of the reported RWA. This statement is supported by the empirical evidence of Heid or Blum and Hellwig. The pro-cyclicality can lead to the asset bubble as banks min- imise their leverage in order to decrease their RWA during the recession (Heid, 2007), (Blum and Hellwig, 1995).
A particular bank as a unit of the banking sector also worries about the level of other banks’ riskiness. Banks with the lowest reported RWA can benefit from the competitive advantage of the low level of unemployed capital. This may lead to the destabilization of the financial sector. Globally, there is a potential threat of the asymmetrical approach of the national supervisors. Although the Basel regulation legislation process should ensure equal supervision, in the consequent implementa- tion, there is a space for differences in the interpretation. This fact can result into the limitation of transparency and global comparison of the risk exposure of the banks (Le Leslé and Avramova, 2012).
From the perspective of the investors and markets, the lack of transparency in the RWA reporting is the most significant concern. The limitations in the compa- rability of banks resulting from all of above mentioned are the crucial worry of the investors. Differences in the internal valuation of risk exposure lead to the asym- metry of reporting RWA within the country, and variation in national supervision harms the multinational comparability. Le Leslé and Avramova recommend using more straightforward indicator for within and cross country comparison, the simple leverage ratio.
The authors also provide a review of key factors potentially leading to the differ- ences in reported RWA. We summarize them to following points:
• Regulatory framework – the bank can choose the extent of internal models implementation,
• Supervisory framework – the interpretation of the Basel regulation into legis- lation can differ across countries,
• Accounting framework – determination of RWA using different accounting standards (US GAAP or IFRS) can lead to the various levels of reported RWA,
• Economic cycle – the calculation of RWA does not take into consideration the current state of the economy, thus international comparison of RWA can be misleading,
• Business model – different types of a bank (universal, retail or investment bank) signify various bank structures and various specializations and thus different risk exposure.
To minimize the effect resulting in the different calculation of RWA, the authors suggest a few proposals. On one hand tightening of current regulation can reduce the diversification in the RWA, on the other hand, cancelling the opportunity of Internal Ratings-Based model employment and applying only the Standardized Approach, or even more just simple leverage ratio is the proposed way how to harmonize the RWA reporting and bring the transparency to the financial sector (Le Leslé and Avramova, 2012).
2.5. Basel standards in the Czech Republic
The second Basel regulatory framework was implemented into the Czech legislation through the regulation no. 123/2007. The fourth part of the regulation is occupied by the capital adequacy. Focusing on the credit risk exposure, the regulation dis- tinguishes two main groups of approaches; the first is the basic approach for the calculation, and special approaches are covered in the second group.
The basic approach is the calculation without internal models employment. The value of the credit exposure is described under § 87 and appendix no. 8 and § 104 and adjustment no. 16 determine adjusted value of the credit risk exposure. The specialized approaches for credit risk exposure calculation are standardized approach using internal models, or Foundation IRB approach, and Advanced IRB approach.
As we can see in the following table, the majority of Czech banks use standard- ized approach without internal ratings employment. The ratio changes just about 5 percentage points over the observed period. The indicator of STA shows the ratio of banks which do not employ Internal Ratings-Based models; the IRB value denotes the ratio for both Foundation and Advanced IRB approaches together. The ratios are calculated as at December, 31 for the particular year.
Table 2.2.: Banks distribution of capital adequacy to credit risk exposure calculation
% of banks 2007 2008 2009 2010 2011 2012
Basel I 80 n/a n/a n/a n/a n/a
STA 6.7 71.4 73.3 75 70.6 70.6
IRB 13.3 28.6 26.7 25 29.4 29.4
source: CNB, 2014
Detailed description of the implementation dates can be found in the next chapter, where the empirical analysis begins.
Empirical Analysis
This chapter deals with an empirical evidence of implementation of the second Basel Accord concerning potential manipulation in the Czech banking sector. We examine this problem from two perspectives. The first perspective relates to the economic performance of the banks represented by return of assets (ROA) which has no rela- tionship to riskiness of the bank’s assets portfolio. From the second perspective we inspect the riskiness of the bank’s assets with respect to the Basel II requirements implementation, which will be characterized by the capital adequacy ratio. We also scrutinize the efficiency of the Czech banks and its impact on the examined ratios.
Last, but not least, we implement the leverage ratio as the indicator of bank’s health and look at the relationship of the both risk-including and riskless performance of the bank.
3.1. Literature review
As already mentioned above, the first part of this chapter examines the relationship between the economic performance of a bank and timing of Basel II requirements im- plementation in Czech banks. In the second part of this chapter, we suggest that after implementation of Internal Ratings-Based (IRB) model for capital adequacy calcu- lation, the banks report lower level of risk-weighted assets (RWA) and thus higher capital adequacy ratio (CAD) than before the implementation. Moreover, we assume that implementation of IRB models reduces on average CAD in comparison to banks which did not employed IRB in monitored period. We also examine the impact of efficiency and leverage indicator to the profitability and level of regulatory capital.
From that reason, the literature overview is divided into two parts. In the first part,
the literature related to the Basel II regulatory approach implementation problems and consequences is introduced, where also the studies related to the potential ma- nipulation are stated. The last part of the review belongs to the empirical studies of the bank performance with respect to the indicator used in this empirical section.
3.1.1. Basel II implementation
The studies analyzing effect on the Basel II regulatory rules can be split into few cat- egories. First category divides the studies to ex ante theoretical analysis and ex post empirical ones. In the first group we can class the Blum’s study of relationship of the capital adequacy requirements and potential riskiness of the bank (Blum, 1999), Heid with his analysis of a representative bank (Heid, 2007) or the criticism or Be- sanko and Kanatas considering sufficiency of the bank’s monitoring skills (Besanko and Kanatas, 1993). Among ex post empirical researches belong Antao and Lacerda with their study about Basel II implementation on Portuguese banking sector (Antao and Lacerda, 2011), or Mariathasan and Merrouche who expressed an idea of po- tential misbehaviour with timing of Internal Ratings-Based models implementation (Mariathasan and Merrouche, 2012).
Another way of surveys classification can be defined as the microeconomic and macroeconomic ones. Microeconomic studies focus on the banking sector in partic- ular countries and assess the impact of the regulation to particular banks. Among those researchers belongs Thompson, Berker and Scottet with empirical research of Turkish banking sector (Thompson et al., 2002) or study of Swiss banks prepared by Bichsel and Blum are examples of such studies (Bichsel and Blum, 2004). On the contrary, macroeconomic surveys measure the effect of the new regulatory frame- work on the economy as a whole. As a representative survey we can list Blum and Hellwig. They present the evidence for potential pro-cyclical effect of banking reg- ulation. If the negative shock in the economy appears, strict capital requirements create incentives for banks to be more cautious about loans providing and thus the economical growth depending on production through new projects needed new cap- ital can be slower (Blum and Hellwig, 1995). The same problem is analyzed in more detail by Heid. His works focuses on both individual and aggregated impact of Basel II implementation for banks. In economic downturn, banks are more likely to be facing with bad loans write-offs than in the economic boom. On the model of rep- resentative bank, the author showed that the extra capital held by the bank to the minimum capital requirements is crucial in decision of further lending. Moreover,
considering procyclicality of bank lending, under Basel I regulatory approach, the capital buffer is predicted to rise in time of economic downturn due to reduction of bank lending. On the other hand, quite the opposite is expected under Basel II from the reason that the reduction in lending is not compensated by the rise of the average risk weights (Heid, 2007).
There are also enough representatives of the opinion that the new regulatory frame- work is more likely to harm the banks. One of these work shows that regardless on the original intention, the capital adequacy requirements can increase the riskiness of the bank (Blum, 1999). The evidence for this is based on the intertemporal trade-off between risk reduction and equity increase. Considering capital adequacy require- ments, an additional unit of equity is worth more. There was shown on a single bank decision problem model that the only possibility how to increase the level of equity tomorrow is to behave more risky today. In accordance with it, Besanko and Kanatas belong to the critics of the capital adequacy requirements. In their study, the moni- toring skills of the bank are reduced due to capital requirements in the competitive market. This fact leads to the deterioration of the banks portfolio not only in terms of reported riskiness, but also as a part of bank’s asset portfolio (Besanko and Kanatas, 1993).
The comparison of Basel I and Basel II regulatory frameworks is also provided by Antao and Lacerda who analyzed capital requirements belonging to the non-financial firms’ credit exposure on the Portuguese banking sector. Their conclusion is not un- ambiguous. The capital for corporate exposures is higher or lower under Basel II in accordance to the value of PD and LGD. Additionally for retail exposures the cap- ital requirements are lower under new regulations (Antao and Lacerda, 2010). On the other hand, Turkish regulatory capital was outlined by Thompson et al. who ex- pressed the speculation about the method and particular parts of capital adequacy calculation. The potential problem can be hidden in overall higher risk in the Turk- ish banking system and also in the misevaluation of the loan portfolio, for example immoderate loans to related parties. Authors stated that reported level of regulatory capital should be lowered under international standards. From that reason and also according rating agency Fitch, the Turkish banks regulatory capital should signif- icantly exceed 8% threshold to cover their actual risk exposure (Thompson et al., 2002).
Benink and Wihlborg expressed the suspicion about the space for manipulation un- der Basel II regulation which leads to the opposite effect compared to the intention
of the regulation. The accidental distortion can result from the insufficient expertise of the creditworthiness of the debtor. On the contrary, the intentional deformation can be seen as the consequence from the opportunities to misuse internal information about the debtor, which is not available to the supervisor. As the most vulnerable part of the capital requirements authors consider translation of the internal rating to the probability of default and its appropriate inspection by the supervisor (Benink and Wihlborg, 2002). According to this, Blum also analyzed the capital requirements from the view of potential manipulation. He also identifies the limitation of the su- pervisory bodies both in the sense of recognizing of the intentional misbehaviour in risk assessing and subsequent penalization (Blum, 2008). To detect potential mis- use of Basel risk weight, Mariathasan and Merrouche analyzed the panel of banks in the period of 2007 to 2010 in the aggregate level. The authors found that on average banks implementing Basel II regime incline more likely to the crisis than banks which do not. To be more concrete, the riskless leverage ratio provide more significant pre- diction of the bank distress than risk-weighted asset in time of high probability of crisis. On the contrary, for banks which did not employ the Basel II approach both leverage ratio and RWA predict distress equally, unconditionally on the presence of the crisis (Mariathasan and Merrouche, 2012).
A contributive survey was prepared by the Financial Service Authority (FSA, 2010), which created the hypothetical portfolio sent to thirty sovereigns, banks and corpo- ration afterwards. The firms were obliged to evaluate the parameters determining capital adequacy under IRB approach. In case of banks, the PD of the portfolio was ranged from 0.03 percent to 0.086 percent with average portfolio mean of 0.054 per- cent. The sovereigns had the lowest range among participants, on the other hand, corporations’ PD estimation differed the most. The possible solution of difference in the calculation can be found in the research of Le Leslé and Avramova. The au- thors performed the overview of both theoretical and empirical explanations of dif- ferences in risk-weighted assets calculation. The study was executed on the main Eu- ropean, North-American, and Asia-Pacific banks with suggestion of potential policy improvement. The work compares the RWA as the indicator of the capital adequacy requirement both within and across particular countries. The authors introduce as the main factors resulting in calculation differences the business model of particu- lar bank, its risk profile, and no less important RWA methodology. On the cross- country level, the crucial aspect can be considered the dissimilarity of supervision.
The authors claim for greater transparency to sufficiently detect possible contagion of bounded banking sector. In their opinion, the higher level of transparency can be
achieved for example by using leverage ratio, as a representative of simple ratios (Le Leslé and Avramova, 2012).
Demirguc-Kunt examined the question of which type of capital holding is appro- priate to avoid financial difficulties during the crisis. The problem was analyzed by the panel of the banks from all over the world to assess the impact of the crisis to the capital ratios, both related and unrelated to the risk exposure. As the indicator of the financial crisis the stock market return was taken. The results showed that in the time of financial distress the structure of the capital is important indicator of the fi- nancial health of the bank and thus stock return. Contrary to this, in pre-crisis times, the structure of the capital does not affect stock return. Moreover, during the crisis, stock return is more sensitive to the riskless leveraged ratio than to the risk-adjusted capital ratio (Demirguc-Kunt, 2013).
The idea of examination of the role of auditors and other supervisors on the capital adequacy requirements to prevent the crisis is the baseline of the survey of Morrison and White. On the general equilibrium model covering two types of market failure, adverse selection and moral hazard, the authors showed the importance of regulator’s reputation. If common confidence in the regulator’s ability to assess the loan portfolio and detect troublesome banks through audit or the reputation alternatively is on the high level, the crisis occurrence is less likely (Morrison and White, 2005).
3.1.2. Bank economic performance
The second part of the literature review relates to the economic performance of a bank, focusing on the important determinants and indicator of bank’s quality status.
There is a group of empirical studies focused on the performance of banking sector in particular countries, represented by Ghosh, Nachane, Narain and Sahoo, Mathuva, Berger, Bichsel and Blum, or Tripe. Ghosh et al. analyse Indian public sector banks on the multivariate panel regression model over the period 1997 through 1999. With changes of the capital ratio as the dependent variable and lagged capital ratio and set of dummies as proxy different regulatory forms authors tried find the evidence of appropriateness of the regulatory framework. The main contribution of the paper exposes that Indian regulatory supervisor should support Indian banks to keep higher capital level than officially required because of the differences in banks’ risk profile.
This fact relates to the second finding about Indian banking sector; that Indian public banks are not risk-averse (Ghosh et al., 2003).
The Kenyan banking sector’s profitability development on the eve of the crisis was studied by Mathuva. On the panel data created by all Kenyan banks from 1998 to 2007 the author studied, besides other things, the relationship between the op- erating effectiveness denoted by cost-income ratio and bank economic profitability represented by the return on asset and return on equity. The author found negative correlation of cost-income ratio to both economic indicators (Mathuva, 2009). Sim- ilar research was performed by Tripe, who uses cost-income ratio from the different point of view; he considers the usefulness of this ratio as a significant measure of bank performance for the Australian banking sector from the view of the banking structure differences additionally to the previous standard use as the explanation for the bank profitability. The author points at the transparency of the cost-income ratio showing that cost-income ratio figure as more comparable across boundaries than cost to asset ratios (Tripe, 1998). Contrary to this, Estrela, Park and Peristianiet con- cern the failure of the predictive power of different bank’s indicators. On the panel of the US banks’ data from 1988 to 1993 they found out that the simple ratios have better predictive power than the risk-employing indicators (Estrela et al., 1999).
Bichsel and Blum focused on the relationship between risk behaviour and the level of banks’ capital in Swiss banking sector using capital adequacy requirements. The relationship in the context of regulatory capital is based on the so-called buffer effect, in other words, the higher level of capital bank possesses, the lower is its probability of default. The problems arise from indirect incentives emerging from the minimum required level of regulatory capital. In accordance with (Blum, 2008), the authors consider the additional capital buffer to be very costly for the bank, and thus to com- pensate for this excessive implies higher risk premium investment. The overall effect is ambiguous, thus increase in regulatory capital can be followed by higher probabil- ity of default. The authors examined the panel of Swiss banks in period from 1990 to 2002 to conclude that even though there is a positive relationship between level of regulatory capital and the probability of default, the predictive power of the capital is not significant (Bichsel and Blum, 2004).
Many researches focus on the US banking market. Berger analyzed the relationship between the capital and profitability US banking sector. The panel of over 80,000 ob- servations created by US banks accounting parameters in period of seven years (1983 - 1989) denied theoretical hypothesis of negative relationship between capital ratio represented by the capital-asset ratio, and after-tax return on equity. The hypothesis following the one bank standard model in perfect competition and without informa- tion asymmetry expects trade-off between the level of capital ratio and risk on equity
and therefore expected return on equity. The author’s rejection of the hypotheses con- firmed the both economic and statistic significant and positive relationship between capital and profitability. Data prove the two-ways positive Granger causality; i.e. the increase in profitability arose as a consequence of unexpected capital increase and vice versa (Berger, 1995).
Hutchinson and Cox analyzed the relationship between structure of capital and profitability with respect to changes in regulatory burden for US banks. The au- thors separated the observed period between two sub-periods; first over 1983 to 1989 represents the period of weak regulation, and second period over 1996 to 2002 cor- responding to the period of stricter bank regulation. Despite earlier empirical evi- dence a positive relationship between financial leverage and profitability represented by return on equity was found for the both sub-samples (Hutchinson and Cox, 2006).
Using similar dataset, Christian, Moffitt and Suberly discussed the significant predic- tor of the future bank earnings by performing the fundamental analysis of financial statements variables. They defined 23 variables, which afterward divided into five classes: Size and growth, Loan exposure, Capital adequacy, Asset quality and Earn- ings. The analysis was performed on the dataset of all national commercial banks US banks over the period 1993 to 1998. The fundamental analysis showed that almost all of the variables (19 out of 23) have a more or less significant impact on the future earnings or returns, but these two groups of variables do not cover. From all 19 vari- ables, no one was significant at 1% level as an explanatory variable for both future earnings and returns. Among the most significant variables in current returns belong variables relating to quality of asset portfolio and capital adequacy. On the other hand, the most significant predictor of future earnings creates current year earnings (Christian et al., 2008).
The comparison of European universal bank and conglomerate cost and profit effi- ciency is provided by the study of Vennet. On the sample of over 2,000 European fi- nancial institutions in years 1995 and 1996 the author found out that de-specialization of the institution can lead to the higher efficiency in the banking system. In other words, average levels of operating efficiency are higher for universal banks than for the specialized ones. In accordance with (Christian et al., 2008), the author identified cost-income ratio as a proxy of operational efficiency as the most important indicator of the bank profitability represented by return on asset, return on equity and ratio of net interest margin to total assets (Vennet, 2002).
