1
“The implications of the introduction of Basel II in terms of risk-weighted assets and internationalization exposure for European banks”
University of Groningen
Faculty of Economics and Business MSc Double Degree with Uppsala University
International Financial Management
RICCARDO BERTOZZI S3061892
Supervisor RuG: Dr. E. Karmaziene
2
Abstract
Considering a sample of sixty-two European banks, I investigate the implications of the introduction of Basel II in terms of risk-weighted assets and internationalization exposure. I challenge the Market Risk and Portfolio Diversification theories jointly with regulative factors of probability of default and asset correlation. I find that after the introduction of the Accord, the whole sample increased in total assets, eventuality discarded by previous authors. Furthermore, I found that under a highly financial regulated environment, financial institutions subjected to Basel II increased statistically their level of internationalization. Outcome that is supported by the Portfolio Diversification theory and asset correlation concept.
1. Introduction
The path of internationalization of banks during the last 150 years has been subjected to numerous variations in terms of exposure degree. Historically, financial institutions increased cross-border strategies at the end of the 19th century in order to finance the significant costs related to the
establishment of colonies.
This increasing trend was subjected to a reduction in the period between the World Wars and increased again in 1970s aiming to keep pace with the emerging financial innovation. The peak that the banking system reached in terms of internationalization degree was in the early 21st century
(Focarelli & Pozzolo, 2005).
Due to the strong presence of internationalization strategies within the banking system, this paper investigates the relationship between the level of cross-border exposure and the concept of risk-weighted assets. This study will focus in the context of the implementation of the regulatory framework of Basel II fully introduced in 2008.
3
where previous research and literature seem to lack of studies. Absence of research especially in terms of effects that Basel II, aiming to increase risk awareness for financial institutions, had in respect of cross-border exposure. The decision toward a deep focus on this specific financial framework is due to the stringent control it applies in terms of risk recognition through the introduction of the newly computational formula of risk-weighted assets. Risk approach which I believe can drive also the strategies toward the level of internationalization.
In order to be able to investigate the aforementioned implications, I construct my analysis in three structured steps based on assets level. I firstly consider the ratio of risk-weighted assets to total assets, namely risk-weighted assets density, with the aim to investigate possible variations. Secondly, I conduct an analysis toward the level of total assets, the denominator of the
risk-weighted asset density, since I believe that variations and adjustments can occur on this item instead of the numerator due to profitability reasons. Thirdly and lastly, I am able to study the variation of the level of internationalization as the ratio of foreign assets to total assets. To conduct the research, I adopt a difference-in-difference statistical approach which permits me to investigate variations in the dependent variables after the introduction of a treatment, which in this is case is represented by Basel II. In a time sample from 2005 to 2012, through the creation of a control composed by 50 European plus Liechtenstein, Norwegian and Swiss banks and a treatment group represented by 12 European Globally Systemically Important Banks, I am able to catch differences in effects and behaviors in the context of the treatment.
Challenging the market risk and portfolio diversification theories, what I found is an absence of any statistical significant variation in terms of risk-weighted asset density after the introduction of the Basel II, which can be motivated by further adjustments in the upper part of the risk-adjusted capital ratio as the level of equity (Heynderickx, Cariboni, & Giudici, 2016). In terms of total assets, the whole sample reported a significant increase after 2008 demonstrating the common trend directed toward the increase in size. Strategies which represents a new finding since the previous literature did not consider this eventuality possible and, furthermore, highly criticized for the implication it could have on the leverage ratio. The last finding is that the financial institutions subjected to Basel II significantly increase the cross-border exposure. This result stems for the relevancy that the context has in terms of banks approach toward internationalization.
The contribution of this paper relies in filling the gap of previous literature in terms of research of the behavior of European financial institutions in the scenario of Basel II regarding cross-border exposure and risk control. Research so far did not study and interpret the approach a bank can adopt when facing a highly regulated environment in terms internationalization as a mitigating or
4
The remainder of this paper is structured as follows. Section 2 relates to the literature review. Section 3 presents the hypothesis formulated. Section 4 describes the methodology adopted. Section 5 is devoted to data and summary statistics. Section 6 reports the empirical result and discussion. Section 7 embodies the conclusion.
2. Literature review
2.1. Basel II and current channels of adjustments
In early 2008 Basel II has been fully implemented in the major world economies. The aim of the new framework is to assess and provide an effective alignment with the capital of financial institutions subjected to regulation and the economic risk (Heid, 2007).
In order to achieve the aforementioned alignment, the accord provides a further tight control and incentive to increase the stability and soundness of the banking system through the employment and adoption of more effective risk management standards and operations.
The new Basel II accord is composed by three main pillars which address the minimum eligible requirement in terms of capital a bank should own, namely the supervisory control and review, and the practices and standard toward the increase of market discipline.
Consistent with the capital adequacy framework of 1988, banks are obliged to hold a minimum capital equal of 8% of the amount of risk-weighted assets (Basel Committe on Banking Supervisor, 2004).
The framework implemented in 2008 has modified the charges applied to capital from previous type-based assets to quality-based assets. A further change from the initial framework is the implementation of two different approaches in terms of risk assessment, namely the standardized approach and the Internal Rating Based Approach (IRB). The former relies on public ratings while the latter gives the possibilities to financial institutions to implement in the risk model the internal ratings that have internally developed in order to define the credit eligibility of their clients (Heid, 2007).
In despite of the aim of the new capital requirement framework to align banks capital with the economic risk in order to increase system stability, the literature provides strong concerns regarding the effective application, highlighting the possibilities that new developed accord could increase the riskiness of institutions subjected to capital requirements (VanHoose, 2007).
5
assessment models the required ratio between risk-weighted assets and total assets (Blundell-Wignall, Atkinson , & Roulet, 2014). Further concerns regarding the establishment of risk-adjusted capital requirements is to induce institutions to establish the value of bad assets at unrealistic levels in order to comply with the leverage ratio (Capital/Total Assets) and through the engagement of elaborated and misleading contracts terms and conditions diluting the apparent clienteles risk (Blundell-Wignall, Atkinson, & Roulet, 2014).
The adoption of Basel II could also have the consequences of an increase in the exposure to non-performing assets. This result is explained by an arbitrage behavior. Banks can increase the level of capital in order to comply with the leverage ratio established by Basel II. In doing so, the cost of equity will inevitably increase and will induce financial institutions to extend their exposure at risk, aiming to increase premiums and gain back the return required by the issue of new equity (Gual, 2011).
In order to evaluate the relationship between the level of risk-weighted assets and
internationalization, I have investigate if previous researches considered cross-border exposure as a mitigating or amplifying factor in terms of risk. The current literature proposes different
adjustments a financial institution can adopt to lower its constraints carried by the level of risk-weighted assets but none of them mentioned internationalization. I will briefly present a summary of the previously researched channel of adjustments.
The literature currently highlights four main strategy a financial institution, willingly to adhere with the regulation, can undertake. Firstly, banks can increase the level of retained earnings (Myers, 1984). Secondly, they can consider to issue new equity in order to counterbalance the ratio between capital and risk-weighted assets (K0/RWA0). Even though this option carries several concerns such
as the decrease of the market value of shares, agency costs of equity jointly with the issue regarding the moment of issuing stock and signals to investors (Morgan Stanley, 2005), financial institutions have demonstrated to drastically increase the level of equity after the implementation of Basel II framework (Heynderickx, Cariboni, & Giudici, 2016). Thirdly, banks can undertake a decrease of the level total assets and, lastly, a reduction of the level of risk-weighted assets. This paper will investigate the existence of a fifth strategy, namely the increase of the exposure to foreign markets expressed by the ratio of Foreign Assets to Total Assets. Option that the previous and the current literature did not focus on.
2.1.1. Risk-adjusted capital ratio decomposition and risk-weighted asset density
6
framework has introduced a ratio which represents a monitoring tool for risk levels, namely capital to risk-weighted assets (Beltratti & Paladino, 2016). The aforementioned ratio has been under study of several researches due to its effects on bank behavior. For the purpose of this paper, the
decomposition in the change in risk-weighted capital requirements will be considered and further explained in the methodology section.
A specific focus is devoted on the numerator of decomposed ratio, namely risk-weighted asset density which is risk-weighted assets to total assets. This concept has been under focus of several previous researches (Gual, 2011; Slovik, 2012; Jokipii & Milne, 2011; Blundell-Wignall, Atkinson , & Roulet, 2014; Heynderickx, Cariboni, & Giudici, 2016; Banco de Espana; Bianchi, Farina, & Fiordelisi, 2016; Mariathasan & Merrouche, 2014; Ferri & Pesic, 2016). The role covered by the aforementioned ratio, identified both as risk-weighted assets density and risk-weighted assets dispersion, is crucial due to its nature of risk assessment.
Basel II relates to it as a measure to identify the degree of risk, which is developed by a financial institution assets. The decision to conduct an analysis to measure the risk of financial institutions after the implementation of Basel II through the use of risk-weighted assets ratio is supported by its capacity to represent both the allocation and quality of financial institutions portfolio risk (Jacques & Nigro, 1997).
Previous research also adopted different measures to identify the riskiness of financial institutions. One of these ratio is commercial and industrial loans to total assets, since this specific class of loans has usually a higher component of risk than other loans (Gorton & Rosen, 1995). The choice to do not engage and discard the use of the aforementioned ratio is due to the limited nature of the outcome it can provide since it only focuses on two specific assets classes, while the risk-weighted assets ratio is performed taken into account all the different assets classes with their respective risk ratings (Jokipii & Milne, 2011).
7
2.2. Internationalization within Basel II framework 2.2.1. Macro-economic implications
The implications that the full implementation of Basel II Accord can produce in respect of the level of international activities of financial institutions, have developed several concerns from the financial and macroeconomic literature.
In the late ‘80s, the first Basel Accord (1988) established a policy meant to encourage the
development of international activities through the allocation of 20 per cent risk-weight for lending towards emerging countries characterized by short maturities. This risk structure and incentives increased the willingness of financial institutions from developed countries to engage in high level activities deployed cross-border (Reisen, 2001).
The development of the Basel Accord of 1988, namely Basel II Agreement, implemented a new risk-weight percentage structure, which is significantly influenced by country origins with the effect of an increase on the divergence between developed and developing countries.
This assumption could support the concept that the new framework will induce financial institution to decrease their assets in foreign countries in order to decrease the risk-weight percentage
allocation, aiming to comply with the risk-adjusted capital requirement.
A risk-weight percentage allocation based also on the probability of default of the country’s origin of an asset could further lead to macro-economic implications. The rates of default and their volatility are influenced by the macro-economic shocks intended as financial crisis and wealth periods. During negative financial periods the volatility of rates can reach an astonishing 10 per cent fluctuation compared to a moderate 1 per cent recorded in steady periods (Moody's , 2001).
Therefore, risk-weight policy based on countries’ probability of default, jointly with pro-cyclical fluctuation of rates, could discourage the undertaking of international asset allocation by banks. The aforementioned concept has been further elaborated in the financial literature, reporting and supporting a strong macro-economic implication in associating the level of equity and risk-weighted assets in terms of cross-border exposure since bad and good time influences, on the one hand, the level of financial institutions lending and, on the other hand, affects the ratings of the borrower. These two separate effects, when jointly considered, increase the assumption that Basel II framework can represent an intensifier for macro-economic fluctuations in terms of banks cross-border exposure (Blum & Hellwig, 1995).
2.2.2. Probability of default and asset correlation
8
risk of their assets. Due to the opposite nature of these two elements, the relationship between risk and cross-border exposure can have contrary effects.
Operating in the financial institutions sector implies facing several loss in terms of interest rates and principals since there is always the presence of bankruptcy at borrowers’ level.
The magnitude of losses occurred in a specific time frame are influenced by the implications and amount of default shocks. Therefore, it is not technically acknowledgeable to identify the proper level of losses in future forecasting.
There is a further factor which can undermine the credit eligibility of borrowers, namely the geographical distance. Previous researches demonstrated that holding foreign assets positions can lead to an absence of sound monitoring practices and high information asymmetry. Elements that can interfere with the proper assessment of the probability of default of foreign customers (Carling & Lundberg, 2005).
In order to implement capital requirements on which financial institutions should comply with, an average of the value of the conceptually forecasted losses has to be in place, namely Expected Losses (EL).
The amount of Expected Losses is the result of the multiplication of three factors on which the framework of Basel II is constructed upon.
EL= PD * EAD * LGD
According to the formula, PD stems for the probability of default, EAD represents the exposure at default and LGD is the loss in terms of percentage in case a borrower run into bankruptcy, namely loss given default.
The second significant factor for analyzing the relationship between risk-weighted assets and internationalization is the asset correlation. This element has a strong influence in the formula of risk-weighted assets.
The asset correlation represents the degree of dependency between the value of the assets of a borrower A and the value of assets of borrower B. Since the diversified portfolio of assets a financial institutions holds, the concept of asset correlation can be soundly extended to the dependency of borrowers’ assets value with the general state of the economy. In order to
accomplish and achieve the aforementioned dependency, the Basel II framework has introduced, in the computational formula of the capital requirement, a single factor stemming for systemic risk (Basel Committee on Banking Supervision, 2005).
9
Capital requirement (K)= [LGD * N [(1-R)-0.5 * G (PD) + (R/(1-R))0.5 * G (0.999)] – PD * LGD]
* (1 – 1.5 * b(PD))-1 * (1+ (M – 2.5) * b (PD))
The allocation of the asset correlation factor is performed on assets classes’ level, hence different classes with an expected losses level (EL) of the same value can represent and gather a higher risk percentage and consequently demanding an increasing capital requirement share.
There is supporting literature in terms of variation of asset correlation during the cycles of financial economy stability.
During economic financial distress periods, the level of volatility and correlation tend to shift from the forecasted long-period averages, incrementing the risk estimations. Concept highly
demonstrated in the equity classes, which reported significant increase in volatility and correlation in time frames affected by turbulent financial economic factors (Bauer, Haerden, & Molenaar, 2004).
Therefore, considering both the magnitude asset correlation covers in the computation of the formula of risk-weighted assets and the easiness with which domestic assets classes, apparently uncorrelated, can increase their correlation during financial stress time, it is predictable and assumable that financial institutions are willing to decrease the level of correlation through an higher exposure to foreign market aiming to achieve a more diversified asset portfolio. Assumption consistent with the Portfolio diversification theory described in the following paragraph.
2.3. International diversification and risk taking theories
The effect of undertaking international activities that increase cross-border exposure has divided the literature in terms of implications regarding risk taking outcomes.
Currently, previous findings, on the one hand, elaborated results in favor of a decrease of risk level through the engagement of cross-border expansion and geographically diversified activities, which is defined as Portfolio diversification theory. On the other hand, there is a large share of researches acknowledging a positive relationships between the increase of international exposure and the degree of risk, represented by the Market risk theory.
2.3.1. Portfolio diversification theory
10
environment where new and developing technologies are up and running in order to keep a solid level of innovation, which can be translated in higher and more updated risk management tools and efficiency practices (Berger, DeYoung, Genay, & Udell, 2000).
A further incentive for banks to expand abroad is to engage in solid relationships with significant customers and applying expansion policy based on the “follow your customer” concept, requiring the establishment of foreign assets positions in countries where the main borrowers live in order to export knowledge and sound practices (Stanley & McManis, 1993). Following customers abroad also includes several advantages in terms of opportunities and risk reduction as it provides the possibility to exploit inefficient financial and banking systems of host countries, benefits from weaker and less tightening regulations, jointly with a chance to enter less saturated markets where higher growing rates are expected (Berger, Dai, Ongena, & Smith, 2003).
Another significant advantage in augmenting foreign assets exposure relies on the concept of portfolio diversification where the idea to allocate assets in several distanced “baskets” could lead to a lowering risk concentration and decreasing the correlation between the asset portfolio, therefore considered a straightforward risk decreasing practice (Amihud, DeLong, & Sauders, 2002).
The aforementioned concept, namely diversification, has been considered one of the main drivers when engaging in foreign asset allocation especially in term of mergers and acquisitions (Repullo, 2001). The role of assets correlation, when analyzing a cross-border expansion strategy, represents a crucial element. Through the adoption of the decision to expand the international exposure,
financial institutions can benefit of a higher diversification of assets when the assumption that domestic and foreign returns are featured by a low or negative correlation (Berger, 2000).
2.3.2. Market risk theory
If on the one hand geographical diversification is considered as a strategy that can lead to a mitigation of risk, the other share of financial and macro-economic literature perceives internationalization as a risk-carrier element.
When establishing new business lines in a foreign country, financial institutions may face fierce competition and suffer from an adverse selection leading to a an underperforming loan product in respect to home country benchmark (Winton, 1999).
Expanding the geographical allocation of assets has the potential to lead to monitoring issues since the ability and effectiveness in control and manage both distant and diversified assets could decrease and eventually switch to higher level of risk, caused by the cost of information and monitoring constraints (John, John, & Senbet, 1991; John, Saunders, & Senbet, 2000). The
11
Effects which are also catalyzed by the eventuality where the geographic sector, where the financial institution has directed its new assets, suffers of loans high downside which can induce a further decrease in monitoring incentives leading to an increase in both the operational and credit risk (Winton, 1999).
Market specific factors further play a role in defining the risk of a foreign expansion strategy, namely facing cultural barriers, host country competition, regulatory framework structure, unstable political environment and the feature of being a foreigner (Méon & Weill, 2005).
For instance, the host country financial system may present a saturated business environment leading to significant efforts in entering and gain sufficient market shares in order to be able to assess and create long-lasting lending relationships with foreign borrowers (Chari & Gupta, 2008) (Berger, Klapper, & Udell, 2001). Significant other market factors can undermine the aim of risk mitigation and, on the contrary, augmenting it. The regulatory framework and the monetary and law policy has been proved to increment the risk level of an expansion strategy (Alibux, 2007) as well as the political environment of the host country jointly with asymmetries both in the market and information (Buch & DeLong, 2004) (Gleason, Mathur, & Wiggins, 2006).
Due to the expensive nature of capital, banks move in favor of debt as financing tool as term notes, deposits and commercial papers which can provide also the advantage and benefits in terms of tax shield (Orgler & Robert, 1983). The choice of equity as a priority is also discard by factors as agency problems (Besanko & Kanatas, 1996) and issues regarding information asymmetry evolving in constraints in holding equity positions (Patrick & Freixas, 1996). In the scenario where
monitoring incentives reach low levels, jointly with being unable to soundly control the foreign assets under management due to geographical localization, the risk will eventually increase and it will require higher level of capital than a non-geographically diversified financial institution (Winton, 1999).
2.3.2.1. Home field advantage theory
12
The aforementioned issues will imply a lower capacity in offering the same quality of products and services that are equally offered in the home country leading to low competitive position in respect to domestic financial providers.
3. Hypothesis
3.1. First hypothesis (H1)
The aim behind the development and implementation of Basel II framework is to better assess and evaluate the level of risk attached to a financial institution assets. As stated in the previous
paragraphs, the measure adopted by the literature (Gual, 2011; Slovik, 2012; Jokipii & Milne, 2011; Blundell-Wignall, Atkinson , & Roulet, 2014; Heynderickx, Cariboni, & Giudici, 2016; Banco de Espana; Bianchi, Farina, & Fiordelisi, 2016; Mariathasan & Merrouche, 2014; Ferri & Pesic, 2016) and the regulative structure in order to identify the level of risk, is the risk-weighted assets density (RWA/TA).
Due to the stringent requirements and regulations, this paper expects a significant decrease of the risk-weighted assets density for the financial institutions subjected to the treatment, namely Basel II framework.
H1: After the implementation of Basel II framework, the financial institutions subjected to the New
Accord requirements show a significant decrease in risk-weighted assets density.
3.2. Second hypothesis (H2)
The formulation of the second hypothesis relies on the nature of the first hypothesis, considering the strong implications of the components of the risk-weighted assets density. Complying with the regulators requirements, in terms of risk-weighted assets to total assets, can be achieved through the reduction of the level of risk-weighted assets maintaining equal the amount of total assets or, on the other hand, increasing the magnitude of the denominator, namely total assets. The second option carries with itself an interesting concept since all the previous literature did not mention this eventuality.
On the contrary with the previous literature in terms of channel of adjustments, this paper believes that banks subjected to the treatment, namely the implementation of Basel II, have adopted the second option, increasing total assets, instead of reducing the level of risk-weighted assets.
13
The introduction of the new Basel Accord requires a significant increase in the level of capital leading to a forced increase of the cost of equity. Therefore, a further reduction in the current business related to risk-weighted assets would consequently decrease once more the return on equity since the return on risky assets, namely the assets classes carrying a higher percentage of risk-weight, are definitely higher than non-risky positions.
The result would be the creation of a low performing equity position for financial institutions (Gual, 2011).
Due to the aforementioned implications, this paper expects a variation in terms of denominator instead of numerator, namely and increase of the level of total assets for financial institutions subjected to the treatment.
H2: After the implementation of Basel II framework, the financial institutions subjected to the new
Accord requirements show a significant increase in the level of total assets.
3.3. Third hypothesis (H3)
In order to investigate the relationship between risk-weighted assets and foreign assets from a risk level perspective, it is necessary to analyze the change of the total assets composition in terms of domestic and international assets.
Considering the structure of the formula of risk-weighted assets, there are two main factors that could drive financial institutions toward opposite strategies in terms of internationalization. According to the presence of the probability of default factor (PD), financial institutions could recall their foreign assets and concentrating their exposure toward the domestic market, as expressed and supported by the Market risk theory. This decision could also be driven by the constraints that a market contagion could imply for the overall inner risk.
On the other hand, there is a solid presence in the capital requirement formula of the element of asset correlation. Element which this paper considers crucial in determining and expecting the future behavior of financial institutions. The allocation of assets, spread on multiple “baskets”, namely different asset classes in terms of geographical determination, can enhance the possibility to diversify a bank’s asset portfolio, mitigating the effect of a domestic financial shock (Winton, 1999).
14
believe of this paper that complying with the regulators requirements, should imply an expansion of the level of foreign assets to total assets aiming to decrease the level of asset correlation expressed in the risk-weighted asset formula.
The rationale behind this assumption relies on the beliefs that the asset correlation factor (R)
represents and embodies a higher magnitude in the computation of risk-weighted assets level, hence the interest of financial institutions toward the embrace of the Portfolio Diversification Hypothesis instead of the Market Risk Hypothesis.
Therefore, here follows the formulation of the third hypothesis of this paper.
H3: After the implementation of Basel II framework, the financial institutions subjected to the New
Accord requirements show a significant increase in the ratio of foreign assets to total assets.
4. Methodology
4.1. Sample and data collection
The sample I adopted in order to investigate the relationship between the level of risk-weighted assets and internationalization of European financial institutions is composed by two defined groups.
The decision to create two different group is motivated by the selection of a difference-in-difference model which will be explained in the following paragraph.
The first group is the one subjected to the requirements expressed by the Basel II accord, which is considered as a “treatment”, therefore the creation of a treatment group. This group is composed by twelve Globally Systemically Important Banks (GSIB) holding the headquarters in Europe, namely BNP Paribas, Deutsche Bank, HSBC, Barclays, Credit Agricole, ING Bank, Nordea, Royal Bank of Scotland, Santander, Societé General, Santander Chartered and Unicredit Group.
I have defined the control group with fifty publicly listed non-systemically important banks, always with the headquarters in Europe. The decision to select fifty banks is consistent with the concept of the difference-in-difference model where significant variations in terms of sample elements
between two groups could affect the reliability of the statistical outcome. In constructing the control group the decision whether to concentrate merely in considering banks belonging to European countries (twenty-eight) or to resemble the composition of the European Economic Area (EEA) has been evaluated.
15 (ECB).
In terms of adequacy in reaction to inflation, both central banks are applying a strong focus mainly on the low frequency in the variables of interests, only considering medium frequency when significant macroeconomic shocks appear. Further similarities are tangible in terms of monetary indicators and definitions of price stability (European Central Bank, 2017).
Analyzing the Swiss Financial Market Supervisory Authority (FINMA) it appeared a clear coordination and cooperation with several European authorities as the European System of Financial Supervision (ESFS) and also the international collaboration and supporting, through bilateral agreements, as the IOSCO MMoU (Multilateral Memorandum of Understanding) aiming to achieve and put in place an international standard in terms of supervision (Intenrational
organization of securities commissions, 2002).
Switzerland banking systems is proving a consistent risk mitigation behavior compared to European Central Bank (ECB) practices such as the implementation of operations and controls aiming to prevent laundry of money such as AML (Anti-Money-Laundering) and a persistent check on borrowers activities through the KYC (Know-Your-Customer) procedures (FINMA, 2016). Significant similar procedures with the European banking system is the awareness that the Swiss financial system has to protect itself from the effect of Too-big-to-fail (TBTF) putting in place emergency plans aiming to mitigate possible macroeconomic implications.
In the Goal 4 of the targets of FINMA there is a clear recall to the concept of the European Central Bank (ECB) in protecting creditors and insured persons through an increase responsibility toward financial institutions (FINMA, 2016).
The decision to include Swiss banks in the control group is also motivated and supported by an increasing interest by the Swiss financial system to cooperate and be integrated to the European one as demonstrated by the establishment of the Supervisory Colleges aimed to implement a
cooperation platform with foreign financial system supervisory regulators (FINMA).
Nevertheless, Switzerland is currently a member and the founder of the European Free Trade Area (EFTA) (The European Free Trade Association, 2013).
In terms of the introduction of the Lichtenstein financial institutions, the European Economic Area
Council decided the 10TH March, 1995 to accept Lichtenstein as member of the EEA and previously
was already a member of the European Free Trade Area since 1991 (EFTA, 2017).
After I have analyzed the relationship between the European Union and the Nordic countries, I have decided to include Norwegian financial institutions.
16
which the European Parliament is a member and founder of.
The collaboration between these two actors dates back to 1992 when the Council of the Baltic Sea States was created aiming to repair and bond the trade environment following the disruption of USSR. A further reason to include Norwegian banks is the role, this country, covers as member of the European Economic Area (EEA) (European Parliament , 2017).
I took the second decision to exclude Russian financial institutions from my control group sample. The reason behind it is that, even though the Norther Dimension Agreement is facilitating
cooperation between Russia and the European market, during the time frame of the analysis
undertaken by this paper, Russia was not a member neither of the European Economic Area (EEA), neither of the European Free Trade Area (EFTA), therefore the exclusion of the Russian financial institutions (EEA, 2016).
4.2. Models
The aim of this paper is to investigate the relationship and behavior of banks in the context of Basel II Agreement in respect of Risk-weighted assets and internationalization. In order to accomplish a sound analysis of the effect of the New Accord, I adopted a model which has the capability to highlight the difference in variations between the introduction of a policy in a specific time frame. Based on the aforementioned necessities, I have developed a difference-in-difference model, which utilizes panel data considering multiples entities during multiple time periods.
The underlying framework of a difference-in-difference approach is to investigate the effect of the introduction of a treatment on a specific group, namely the treatment group, compared to the performance of a group which is not touched by the aforementioned treatment, namely the control group.
The treatment group is composed by twelve European Globally Systemically Important Banks (GSIB) while the control group by fifty non-systemically important European banks, plus Norway, Switzerland and Lichtenstein.
A difference-in-difference model requires to create groups which are as similar as possible in order to catch possible variations in terms of variables, therefore the decision to limit the size of the control group to fifty entities. Due to the requirement of the model to provide groups as similar as possible, I performed the natural logarithm of the dependent variables in order to standardize the values. Action which was necessary especially when considering the difference in terms of size, namely total assets.
17
full adoption of Basel II in the late 2008 and consequently I have considered four years after and before the aforementioned treatment introduction aiming to achieve a sound estimate.
As introduced previously, I adopt the decomposition of the risk-adjust capital ratio which permits me to isolate the risk-weighted asset density.
Cohen et al. (2016) developed a sound decomposition of the ratio concerning risk-weighted assets and capital which has been further used by researchers to assess risk after the implementation of Basel II (Heynderickx, Cariboni, & Giudici, 2016).
The aforementioned risk-weighted capital requirements decompositions is expressed as follow:
/
/ =
K= capital
RWA= risk-weighted assets TA= total assets
Inc= net income Div= dividends
Oth= other changes to capital computed as residual
The decision to implement the use of a decomposed ratio relies on the necessity of this paper to investigate, in the first part of the study, the relationship between the denominators elements.
This paper formulates three distinct hypothesis, therefore the necessity to create three models in order to catch the expected variations in terms of variable after the treatment.
(1) RWA/TA ij = β0 + β1 Afteri, + β2 Treated i, + β3Afteri X Treatedi + β4 Bank1 +…+
β65Bank61 +
ε
i,t(2) TA ij = β0 + β1 Afteri, + β2 Treated i, + β3Afteri X Treatedi + β4 Bank1 +…+ β65Bank61 +
ε
i,t
(3) Internationalization ij = β0 + β1 Afteri, + β2 Treated i, + β3Afteri X Treatedi + β4 Bank1 +…+
18
I constructed the three models based on the framework of difference-in-difference changing the dependent variable for each hypothesis I plan to test. RWA/TA embodies the risk-weighted-assets density which is the ratio between the level of risk-weighted-assets and total assets, TA represents the level of total assets and Internationalization is the proxy stemming for the level of cross-border exposure namely the ratio of foreign assets to total assets. The remainder of the models is composed by β1 Afteri which considers the sample of banks from a time perspective therefore i assumes the
value of 0 if the time period is comprehended before the treatment namely from 2005 till 2008, while assumes the value of 1 if the time period is after 2008. β2 Treatedi is the factor defining the
groups, i assumes the value of 1 if the bank is subjected to the treatment or 0 if the bank is part of the control group. β3Afteri X Treatedi is the interaction between the two previous factors identifying
the difference-in-difference items. β4 Bank1 +…+ β65Bank61 represent the introduction of the bank
fixed effects therefore the exclusion of the 62nd banks.
ε
i,t finally represents the error term.5. Data and summary statistics
I performed the collection of the data of interest through the database provided firstly by OrbisBank Focus for the gathering of all the isin codes of financial institutions. Secondly, I adopted Datastream source to download the variables data. All the data are gathered on a yearly basis due to the nature of my reported variable. The currency which I used in order to perform my analysis is euro and the conversion of currencies has been already performed by the database.
The total assets embodies the sum of cash & due from financial institutions, the total amount of the investments, net loans, customer liability on acceptance, the investment in unconsolidated
subsidiaries, real estate assets, net property, plant and equipment and other assets. Foreign assets represent the value of total or identifiable assets of foreign operations before the adjustments and eliminations. Risk-weighted assets are defined as the total of the attached value of every asset class multiplied by their respective assigned risk weighting, as defined by the Basel Committee.
19 Table 1. Summary statistics and t-test.
Treatment Group Control Group
Pre Basel II After Basel II Diff. Pre Basel II After Basel II Diff.
Variable Mean Obs. Mean Obs. Variation Mean Obs. Mean Obs. Variation RWA density 0.38 44 0.34 44 -0.04* 0.56 166 0.50 180 -0.06* Total assets 1240.46 48 1341.93 48 101.46 148.48 199 154.06 200 5.58 Internat. 0.58 41 0.47 40 -0.11*** 0.19 190 0.15 190 -0.04
Note: this table is a report of the summary descriptive statistics for the key variable I adopted to empirically test the relationship between risk-weighted assets and the level of internationalization of European financial institutions. ***, **, * represent the significance at 99%, 95% and 90% confidence levels.
Table 1 reports the summary statistics and the t-test of the dependent variables before (2005-2008) and after (2009-2012) the implementation of Basel II for both the treatment and control group. In terms of risk-weighted assets density, the treatment group shows the lowest average values in both periods demonstrating the implications of a regulated environment. Testing for the significance of the variation between the two periods, I find that the treatment group is subjected to a negative variation of level of risk attached to the total assets owned by financial institutions. Result which stems for a reaction in favor of the compliance in respect to the newly implemented financial agreement significant at 90% confidence level. Considering the control group, even though Basel II is not directed toward non-systemically important banks, a significant variation after the
implementation occurred. Variation which reduced the level risk-weighted assets density at a 90% significance level. The second variable of interest, namely total assets, reports an increase in value after the treatment for both the treatment and control group but the variation is not supported by statistical significance at any confidence level. The third and last variable I focus on, is the ratio of foreign assets to total assets, namely the level of internationalization. Analyzing the average values for all the period, it is clear that the treatment group has a higher cross-border exposure with almost half of its assets localized abroad compared to the control group which has a minor international exposure. With the introduction of the financial agreement in 2008, the treatment group
20
6. Empirical results and discussion
The aim behind the implementation of the new regulatory framework, namely Basel II, was to increase the awareness regarding the risk attached to financial institutions assets, therefore the introduction of defined thresholds to comply with.
As expressed in previous paragraphs, the first concept this paper focuses on is the risk-adjusted capital ratio which has been decomposed in order to further analyze the variation of the factor of interests. Risk-weighted assets density represents one of the pillar of the risk-adjusted capital ratio and it is composed by the ratio of the level of risk-weighted assets to total assets of a financial institution. According to the formulation of the first hypothesis (H1), this paper expects a significant
decrease of the ratio after the implementation of the treatment within the treatment group.
Table 2. Difference-in-difference regressions results
Model (1) Model (2) Model (3) RWA Density Total Assets Internationalization
(RWA/Total Assets) (Foreign Assets/Total Assets)
After -0.05 0.28*** -1.16*** (-0.62) (12.04) (-5.53) Treated -0.14 2.79*** 0.58 (-0.25) (22.32) (0.47) After X Treated -0.12 -0.10 0.73* (-0.69) (-0.20) (1.67) Cons -0.42 17.01*** -1.05 (-0.84) (190.01) (1.29) Obs. 434 495 327 R2 0.58 0.94 0.49
These coefficients are parameters estimations of the Difference-in-Difference regressions explained in Eq. (1), (2), (3). T-statistics are reported in parentheses as robust standard errors. ***, **, * represent the significance at 99%, 95% and 90% confidence levels.
As shown in table 2, the factor After represents the variation of the overall sample after the implementation of Basel II.
In respect of the concept of risk-weighted asset density, after the introduction of the treatment the whole sample does not report a dramatic change in terms of variation reporting a negative
21
adjustments in terms of this parameter after 2008. When analyzing the difference in value of ratio between treatment and control group, the results show that in the whole time period there is no significant difference between financial institutions subjected to Basel II and the non-systemically important banks. Result which is shown in the line of factor Treated under model 1, reporting a slightly negative variation of -0.14 which does not represent any significant change at confidence level. The implementation of the financial agreement did not have a significant change in terms of treatment group since the factor of difference-in-difference, namely After X Treated, shows a minor variation of -0.12 but is not significant at any statistical level.
Since the risk-weighted assets density is one of the components of the risk-adjusted capital ratio, possible adjustments directed to a compliance strategy may have been undertaken in the numerator of the formula. Previous studies (Heynderickx, Cariboni, & Giudici, 2016) found out that in that period, numerous financial institutions increased the level of equity which is consistent with the adequacy of the ratio. Even though a negative variation in treatment group after Basel II is reported, it is not statistical relevant therefore it is not possible to confirm the first hypothesis (H1).
After performing the difference-in-difference regression of the second model in order to evaluate if a variation in terms of total assets occurred, I found interesting results. The outcome reported in table 2 shows that the overall bank sample increased significantly in size at a 99% confidence level with a positive coefficient of 0.28. Result which is in contrast with previous researches that were considering an increase in total assets after Basel II extremely unsuggestable considering the repercussion that it could have on the leverage ratio. In terms of treatment group referred to the whole time period, namely from 2005 to 2012, the group reports a higher level of total assets significant at 99% confidence level. Result which is clearly expected due to the nature of being a European systemically important bank, therefore the positive coefficient of 2.79 confirms the higher size and magnitude that the treatment group represent in the European banking system. Analyzing the difference-in-difference factor of model 2, it seems that the treatment group after the
implementation of the financial agreement does not show a significant variation compared to the control group with a negative coefficient of -0.10 not comprehended in any confidence level. The aforementioned result does not permit to confirm the second hypothesis (H2) which was testing and
increase of the level of total assets after the implementation of Basel II within the treatment group. The overall result of the regression of the second model represent an initial interesting findings since the overall sample shown a significant increase in size after 2008, which is opening a new insight in terms of implication of the financial agreement considering the previous literature and researches.
22
internationalization changed significantly since previous researches did not faced this specific implication. The financial institutions sample considered as whole showed a dramatic change in terms of cross-border exposure, reporting a negative variation of the ratio of -1.16 significant at 99% confidence level. The result represent a variation of the all sample considering both the control and treatment group. Outcome that suggests that after the introduction of the treatment the banks in the European sample adjusted their cross-border exposure negatively, possibly confirming the implications behind the market risk theory. The treatment group, without a distinction of time, reported no significant higher or lower exposure to external markets, evidence represented by a positive coefficient of 0.58 which is not significant at any confidence level. The interesting and core result is the difference-in-difference factor which scored a positive significant variation of 0.73 at 90% significance level confirming the third hypothesis (H3). This results represent the trend of
European financial institutions subjected to the regulatory framework of Basel II to have undertaken a strategy toward a significant increase in terms of internationalization degree. Decision that
confirms my expectation to consider internationalization as a risk mitigation tool under a specific regulative context.
23
Table 3. Difference-in-difference regressions results without Swiss, Liechtenstein and Norwegian banks
Model (1) Model (2) Model (3) RWA Density Total Assets Internationalization (RWA/Total Assets) (Foreign Assets/Total Assets) After 0.07 0.32*** -1.13*** (0.97) (11.97) (-4.76) Treated -0.53 6.56*** 0.61 (-1.08) (57.03) (0.53) After X Treated -0.25 -0.04 0.69 (-1.60) (-0.88) (1.60) Cons -0.55 14.38 -1.09 (-1.28) (180.58) (-1.39) Obs. 331 375 258 R2 0.59 0.99 0.54
These coefficients are parameters estimations of the difference-in-difference regressions. T-statistics are reported in parentheses as robust standard errors. ***, **, * represent the significance at 99%, 95% and 90% confidence levels.
Both in terms of risk-weighted asset density and total assets, the estimation results are not subjected to deviation from the main regressions after the exclusion of the aforementioned countries.
24
implementation of the financial agreement in 2008 disappear.
Performing the regression of the three models without Liechtenstein, Norwegian and Swiss banks provided the same results as the all-countries regression demonstrating a check in terms of
robustness of the assumption of including these countries. The main deviation in terms of outcome is only reported in the difference-in-difference factor of the third model.
The overall concepts of table 2 and 3 are that, firstly, after the introduction of Basel II, the whole sample of banks, considering also the three including countries, did not adjust their ratio of risk-weighted assets to total assets suggesting the possibilities that further adjustments took place in numerator part of the risk-adjusted capital ratio formula. Variation which also is not reported when considering the difference of value between the treatment and control group after 2008. Secondly, according to table 2 and 3, after the introduction of the financial regulation the two groups
considered jointly, have significantly increase in size as demonstrated by positive coefficients in the first row of model 2. On the other hand, the treatment group did not reported a significant variation in terms of size after the treatment compared to the control group therefore the impossibility to confirm the second hypothesis. The interesting outcome of the second model however is the overall increase in size after Basel II of the whole sample which to some extents introduce a newly reaction of banks to the introduction of the financial regulation of 2008. Thirdly, the core result is the implication that the treatment had in terms of level of internationalization for European financial institutions. On the one hand, the whole sample, without a distinction of groups, demonstrated that a clear and significant reaction was undertaken toward a reduction of the level of cross-border
exposure as demonstrated by the negative variation of the coefficient in table 2 of -1.16 and -1,13 in table 3. Outcome that can be interpreted as a consideration of internationalization as a risk
increasing factor supporting the market risk theory. On the other hand, the treatment group reported a contrary trend in terms of strategy after 2008. When considering the difference in value of the ratio foreign assets to total assets, the financial institutions subjected to the treatment increased significantly the level of international exposure compare to the control group with a significance coefficient of 0.73. Variation which confirms in first place the third hypothesis (H3) and
25
enhance the overall level of risk attached of the assets of a financial institution that has to comply with risk regulation compared to one which has not.
7. Conclusion
Previous literature and research did not focus on the relationships between risk and level of internationalization of financial institutions after the implementation of Basel II. This thesis has tried to fill the gap by studying the implications that the financial agreement of 2008 had in respect of the level of risk-weighted assets density and cross-border exposure. In order to be able to catch the variations of value after the introduction of the framework, I adopted a difference-in-difference estimation strategy which permitted me also to create a control group as a comparable unit. The sample of banks has been collected among the European countries plus Liechtenstein, Norway and Switzerland due to similarities in terms of monetary and financial policies. The main empirical outcome is that the introduction of Basel II clearly increased the level of internationalization of financial institutions subjected to treatment, while considering the effect on the whole sample, a contrary trend is reported, namely a significant negative variation. The results confirm the different approach banks adopted in terms of internationalization, which is driven by being subjected to the financial regulation or not. This result has been reached through a first analysis of the effect of Basel II on the risk-weighted assets density which did not reported significant variations which can be motivated by adjustments in the numerator elements of the risk-adjusted capital ratio as the increase in equity. I further tested the change in value of its denominator, namely total assets, which presented a significant positive variation in terms of whole sample demonstrating the common trend to increase in size after the treatment. The financial institutions subjected to the financial agreement did not reported significant difference compared to the control group in terms of size variation. These two analysis finally led me to test whether the ratio between foreign assets and total assets changed after 2008.
The contribution of this paper is firstly an explorative study of the implications that Basel II had in terms of level of internationalization for European financial institutions, creating a starting basis for further research in field. Secondly, the demonstration that the environment in which cross-border exposure is considered can change the approach to risk. As I discovered, a strong financial
26
2008, the overall sample increased in level of total assets. Trend which was not expected and suggested by the previous literature due to the implication that such positive variation could have on the leverage ratio. Further research can be directed, firstly toward an analysis in terms of the
27
References
Alibux, A. (2007). Cross-border mergers and acquisitions in the European banking sector. Erasmus University of Rotterdam. Dissertation. .
Amihud, Y., DeLong, G., & Sauders, A. (2002). The effects of cross-border bank mergers on bank risk and value. Journal of International Money and Finance, vol. 21, pp 857-877.
Banco de Espana. (n.d.). Towards more consistent, albeit diverse, risk-weighted assets across banks. Estabilidad Finaciera.
Basel Committe on Banking Supervision. (2010). Basel III: A global regulatory framework for more resilient banks and banking systems. bank for International Settlements.
Basel Committe on Banking Supervisor. (2004). International Convergence of Capital Measurement and Capital Standards. Bank for International Settlements.
Basel Committee on Banking Supervision. (2005). An explanatory note in the Basel II IRB risk weight functions. Bank for International Settlements.
Bauer, R., Haerden, R., & Molenaar, R. (2004). Asset allocation in stable and unstable times. Journal of Investing, pp 72-80.
Beltratti, A., & Paladino, G. (2016). Basel II and regulatory arbitrage. Evidence from financial crises. Journal of Empirical Finance, vol. 39, pp 190-196.
Berger, A. (2000). The big picture of bank diversification. Conference on Bank Structure and Competition Proceedings (pp. pp 162-174). Federal Reserve of Chicago.
Berger, A., Dai, Q., Ongena, S., & Smith, D. (2003). To what extent will the banking industry be globalized? A study of bank nationality and reach in 20 European nations. Journal of Banking & Finance, vol. 27, pp 383-415.
Berger, A., DeYoung, R., Genay, H., & Udell, G. (2000). Globalization of Financial Institutions: Evidence from Cross-Border Banking Performance. Brooking-Wharton Papers on Financial Services, pp 23-120.
Berger, A., Klapper, L., & Udell, G. (2001). The ability of banks to lend to informationally opaque small businesses. Journal of Banking & Finance, vol. 25, pp 2127-2167.
Besanko, D., & Kanatas, G. (1996). The regulation of bank capital: Do capital standards promote bank safety? Journal of financial intermediation, vol. 5, pp 160-183.
Bianchi, N., Farina, V., & Fiordelisi, F. (2016). Risk Culture in Banks: Just Words? Working Paper, 1-38.
Blum, J., & Hellwig, M. (1995). The macroeconomic implications of capital adequacy for banks. European Economic Review, vol.39, pp 739-749.
28
Blundell-Wignall, A., Atkinson, P., & Roulet, C. (2014). Bank business models and separation issue. Financial Market Trends, vol. 2013/2, pp 69-87.
Buch, C., & DeLong, G. (2004). Cross-border bank mergers: What lures the rare animal? Journal of Banking & Finance , vol. 28, pp 2077-2102.
Carling , K., & Lundberg, S. (2005). Asymmetric information and distance: an empirical
assessment of geographical credit rationing. Journal of Economics & Business, vol. 57, pp 39-59.
Chari, A., & Gupta, N. (2008). Incumbents and protectionism: The political economy of foreign antry liberalization. Journal of Financial Economics, vol. 88, pp 633-656.
Chiang, T., Jeon, B., & Li, H. (2007). Dynamic correlation analysis of financial contagion: Evidence from Asian markets. Journal of International Money and Finance , vol. 26, pp 1206-1228.
Cohen, B., & Scatigna, M. (2016). Banks and capital requiements: Channels of adjustment. Journal of Banking & Finance, vol. 69, pp s56-69.
EEA. (2016). Agreement on the Europena Economic Area. Brussels: EFTA States' official gazettes. EFTA. (2017). EFTA through the years. EFTA.
European Central Bank. (2017, June 22). Swiss monetary policy as viewed by the European Central Bank.
European Parliament . (2017). The European economic area (EEA), Switzerland and the North. European Parliament .
European Parliament. (2016). Non-performing loans in the Banking Union: stocktaking and challenges.
Ferri, G., & Pesic, V. (2016). Bank regulatory arbitrage via risk weighted assets dispersion. Journal of Financial Stability, pp 1-15.
FINMA. (2016). Strategic goals 2017-2020. Swiss Financial Market Supervisory Authority. FINMA. (n.d.). Supervisory colleges provide a key forum for cooperation between FINMA and
foreing supervisory authorities involved in group supervision. Swiss Financial Market Supervision Authority.
Focarelli, D., & Pozzolo, A. (2005). where do banks expand abroad? An empirical analysis. The Journal of Business, vol. 78, No 6.
Fratzscher, M. (1998). Why are currency crises contagious? A coparison of the Latin American Crisis of 1994-1995 and the Asian Crisis of 1997-1998. Review of World Economic, vol. 134, pp 664-691.
Gleason, K., Mathur, I., & Wiggins, R. (2006). The use of acquisitions and joint ventures by U.S. banks expanding abroad. Journal of Finacial Research, vol. 29, pp 503-522.
29
Gorton, G., & Rosen, R. (1995). Corporate control, portfolio choice, and the decline of banking. Journal of Finance, vol. 50, pp 1377-1420.
Gual, J. (2011). Capital requirements under Basel III and their impact on the banking industry. "la Caixa" Economic Papers .
Heid, F. (2007). The cyclical effects of the Basel II capital requirements. Journal of Banking & Finance , vol. 31, pp 3885-3900.
Heynderickx, W., Cariboni, J., & Giudici, M. (2016). Drivers behind the changes in European banks' capital ratios. European Commission.
Intenrational organization of securities commissions. (2002). Multilateral memorandum of understanding concerning consultation and cooperation and the exchange of information. OICV-IOSCO.
Jacques, K., & Nigro, P. (1997). Risk-based capital, portfolio risk, and bank capital: A simultaneous equations approach. Journal of Economics and Business , Vol. 49, no. 6, pp 533-547. John, K., John, T., & Senbet, L. (1991). Risk-shifting incentives of depository institutions: A new
perspective on federal deposit insurance reform. Journal of Banking & Finance, vol. 15, pp 895-915.
John, K., Saunders, A., & Senbet, L. (2000). A theory of bank regulation and management compensation. Review of Financial Studies, vol. 13, pp 95-125.
Jokipii, T., & Milne, A. (2011). Bank capital buffer and risk adjustment decisions. Journal of Financial Stability, vol. 7, pp 165-178.
Koehn, M., & Santomero, A. (1980). Regulation of Bank Capital and Portfolio Risk. The Journal of Finance , vol. 35 no. 5, pp 1235-1244.
Mariathasan, M., & Merrouche, O. (2014). The manipulation of Basel risk-weights. Journal of Financial Intermediation, vol. 23, pp 300-321.
Méon, P.-G., & Weill, L. (2005). Can mergers in Europe help banks hedge against macroeconomic risk? Applied Financial Economics, vol. 15, pp 315-326.
Moody's . (2001). Testing for Rating Consistency in Annual Default Rates. Moody's Investors Service.
Morgan Stanley. (2005). Capital structure, payout policy, and the IPO process. Journal of applied corporate finance, vol. 17, pp 8-18.
Myers, S. (1984). The capital structure puzzle. The Journal of Finance, vol. 39, pp 575-592. Orgler, Y., & Robert, A. (1983). Implications of corporate capital structure theory for banking
institutions. Journal of MOney, Credit and Banking, vol. 15, pp 212-21.
Patrick, B., & Freixas, X. (1996). A dilution cost approach to financial intermediation. Universitat Pompeu Fabra, working paper.
Reisen, H. (2001). Will Basel II Contribute to Convergence in Intenrational Capital Flows.
30
Repullo, R. (2001). A model of takeovers of foreing banks. Spanish Economic Review, vol. 3, pp 1-21.
Slovik, P. (2012). Systemically Important Banks and Capital Regulation Challenges. Paris: OECD Economics Department .
Stanley, T., & McManis, C. (1993). The effects of foreing ownership of US banks on the availability of loanable funds to small business. Journal of Small Business Management, vol.31, pp 51-66.
The European Free Trade Association. (2013). Establishing the European Free Trade Association. (pp. pp 1-30). Geneva: The European Free Trade Association.
VanHoose, D. (2007). Theories of bank behavior under capital regulation. Journal of Banking & Finance , vol. 31, pp 3680-3697.
