The Impact of Warnings and Recommendations issued by the European Systemic Risk Board on Bank Performance 

The Impact of Warnings and Recommendations issued by the European

System Risk Board on Bank Performance

Tom van Eekhout, 11248351

Business Administration, specialization Finance Faculty of Economics and Business

University of Amsterdam Magdalena Rola-Janicka

Statement of Originality

This document is written by Student: Tom van Eekhout who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

Abstract

The effect of warnings and recommendations issued by the European Systemic Risk Board towards specific countries is analysed in relation to the return of banks within these countries. Using data from two different 4 month periods, in 2016 and 2019, to conduct the analyses. This study utilises the event study methodology pioneered by Fama et al. in 1969 and constructed in a useful way by De Jong in 2007. The study computes the cumulative average abnormal return of the banks per ESRB warning or recommendation to test for significance. The mandate of the ESRB in a bank-based economy could have had its effects on cumulative bank performance. However, literature suggests that the

developments of regulations since the 2008 financial crisis already had their stabilizing effect due to centralized supervision and control. Causing individual regulatory effects on financial performance to be reduced. The results of this study find no deviation from normal performance after an ESRB announcement.

Table of Contents TITLE PAGE 1 STATEMENT OF ORIGINALITY 2 ABSTRACT 3 TABLE OF CONTENTS 4-5 1.0 INTRODUCTION 6-7 2.0 BANKING INDUSTRY 7-16 2.1 INTRODUCTION 7

2.2 EUROPEAN SYSTEMIC RISK BOARD 7-10

2.2.1 GENERAL MANDATE OF THE ESRB 7-8

2.2.2 RISK FACTORS 8

2.2.3ESRB ANNOUNCEMENTS 9-10

2.3 BANKING REGULATION 10-14

2.3.1 INSTITUTIONAL BACKGROUND 12-13

2.3.2 CENTRALIZED EFFECT ON COUNTRY BASIS 13-14

2.4 ECONOMICAL BACKGROUND 14-15

2.4.1 GENERAL ECONOMIC DEVELOPMENT 14

2.4.2 ECONOMIC DEVELOPMENT BANKING INDUSTRY 14-15

2.5 EMPIRICAL EVIDENCE 15-16

2.5.1ESRB 15

2.5.2 BANKING REGULATION & THEORY 15-16

3.0 RESEARCH QUESTION & HYPOTHESES 16-18

3.1 HYPOTHESIS 1 17 3.2 HYPOTHESIS 2 17 3.3 HYPOTHESIS 3 17 4.0 DATA 18-21 4.1ESRB WARNING 2016 18-19 4.2ESRB WARNING 2019 19-20 4.3ESRB RECOMMENDATION 2019 20-21 5.0 METHODOLOGY 21-23 6.0 RESULTS 23-25 7.0 DISCUSSION 25-27 7.1 FINDINGS 25-26 7.2 LIMITATIONS 26-27

7.3 SUGGESTIONS FOR FURTHER RESEARCH 27

APPENDIX 29-51

REFERENCES 29-32

TABLES 33

FIGURES 34-39

1.0 Introduction

The financial crisis in 2008 started a significant development in the financial industry. The European Union realized it was necessary to control it, by creating a supervisory body that all member states answer to. This indicates that the industry since this crisis is moving from a decentralized supervisory system to a centralized supervisory system. Committees and regulatory agencies were created in order for the centralization to succeed. The regulatory devices entered in the European Union since 2008 have significant impact on the industry and therefore the effect of many regulators in the industry have been analysed (S&P, 2015). The individual impact of major players in the financial supervisory industry could be relevant for the financial markets in the EU, as it is described as a bank-based economy (Szczepanski, 2019). This paper is an analysis of the effect of a possibly significant regulatory device within this system, the European Systemic Risk Board (ESRB). Whose effect on the industry has not been thoroughly included in the analysis of previous research into regulators.

The European Systemic Risk Board is a regulatory body within the financial supervisory system in the European Union, and can be considered a major player. It assesses systematic risks in the financial industry and warns countries if found necessary (Mazzaferro & Dierick, 2018). Its mandate is not supported by law. However, research does suggest that significant regulatory devices have found authoritative power since the development of the EU supervisory system. The European Systemic Risk Board has issued 2 groups of warnings and recommendations. These are based on assessed risk factors since its establishment and are issued in 2016 and in 2019. The announcements are based on residential real estate risk factors, which is a focus area of the ESRB (ESRB, 2016). This industry is important in the European banking economy and therefore needs continuous supervision.

The question arises what the impact is of these warnings on the market since the

establishment of the centralized regulatory system. Therefore, this research tests the effect of ESRB warnings and recommendations on bank performance. Specifically; what is the impact of an ESRB warning or recommendation on the cumulative performance of banks around the announcement dates. This question is tested using the event study methodology pioneered by Fama et al. (1969) and

modernized for this study by De Jong (2007). The method is statistically tested with Stata software, using a guide by Princeton University (2007). These provide the t-test statistics on the average cumulative return for all banks as the basis for the conclusion of each event.

The result of these tests do not show significant abnormal performance. Indicating that the announcements by the ESRB do not negatively influence the performance of banks operating in countries recognized as having significant risks. I advocate that the mandate of the ESRB in a heavily financially based economy could have had (reduced) effects on cumulative bank performance. However, the developments of regulations since the 2008 financial crisis already had their stabilizing effect due to centralized supervision and control. Causing results in this study that indicated no

deviation from normal performance after an ESRB announcement, as indicated by the results of the tests.

To provide context around the tests a comprehensive analysis of literature is done. The analysis of literature is divided into an informational background on the ESRB, followed by an analysis of the financial industry in the European Union. Moreover, a review of economic

developments surrounding the events is provided. Finally, empirical evidence is given on the most important factors influencing the research question. This provides a basis for the hypothesis statements and for the analyses conducted later in the study. The data and method used for the analyses are described in detail before the results of the analyses are provided and discussed in relation to the literature.

2.0 Banking Industry 2.1 Introduction

This section provides an informational background on the impact of warnings and

recommendations issued by the European Systemic Risk Board and the impact it can have on bank performance. Moreover, this review summarizes a collection of materials on specific regulations and economic factors that can have an effect on bank performance. The review will flow from an

informational background on the warnings and recommendations issued by the ESRB into more general concepts concerning the impact of regulation on bank performance. After reviewing the regulatory literature an insight is given into the economic state and development during the period of the ESRB warnings and recommendations. In order to give context to the regulations and their possible effects. Concluding the review with a section on empirical evidence on the collection of materials of the regulations, economic effects and ESRB warnings and recommendations

2.2 European Systemic Risk Board

This research presents the effect of the European Systemic Risk Board (ESRB) on bank performance of banks subject to their mandate. The European Systemic Risk Board was established in 2010 and became operational in January 2011 (ESRB, 2020). An introduction of the mandate of the ESRB provides a basis in this section, after which the most significant tasks and effects are described regarding the banking industry relevant for this research.

2.2.1 General Mandate of the ESRB

The ESRB is an independent body, with the macro prudential oversight over banks (Mazzaferro & Dierick, 2018). However, their supervision has extended towards most of the

European financial system. Their mandate consists of assessing systemic risks, and where necessary, issue warnings and recommendations to countries (ESRB, 2016).

Mazzaferro & Dierick (2018) describe the role of the ESRB since its establishment as establishing a coherent macro prudential framework for the financial markets in the EU. Moreover, they state that the ESRB helps the member states make this framework operational. This role was created in organized steps, firstly, a groundwork was laid by setting up national macro prudential authorities in all member states and providing their tasks and directive (Mazzaferro & Dierick, 2018). Secondly, Mazzaferro & Dierick (2018) mention that after establishing the framework, the macro prudential supervision could be made more precise by identifying objectives and designing and delegating macro prudential instruments towards the member states. Mazzaferro & Dierick (2018) continue in their report that the real estate sector in an important area for the ESRB regarding macro prudential policy, also due to the fact that this area has often been the cause for significant risks and vulnerabilities. Since the beginning the ESRB has taken initiative to improve the real estate sector in the EU (ESRB, 2016).

Even though the ESRB assesses systemic risk in general in the entire banking industry, the warnings and recommendations issued are focused on the residential real estate (RRE) sector in the researched events (ESRB, 2016). The residential real estate sector is of importance because real estate markets have been major factors in systemic financial crises (Hartmann, 2015). Moreover, the

establishment of the ESRB relates to this specific systemic risk (ESRB, 2016). It is therefore noticeable that the issued warnings and recommendations are focused on the RRE sector.

Ferran & Alexander (2011) raise questions about the power of the ESRB, because the

institution is set up without any legally-binding powers. However, Ferran & Alexander (2011) suggest that the lack of formal power does not necessarily prevent the ESRB, and other systemic risk bodies, to act in a credible and authoritative matter.

Thus, the ESRB has the task of assessing significant systemic risks in the financial sector, together with establishing a framework for this industry to tackle those risks. These tasks are performed with limited power, however, the centralized organizations cause a certain authority. The following section focuses on the systemic risks associated with the residential real estate sector, i.e. the risks associated with the warnings and recommendations issued by the ESRB

2.2.2 Risk factors

The risks associated with the issue of ESRB warnings differ per member state. The ESRB report (2016), mentions three major risk factors for the member states that are identified residential real estate vulnerabilities (or stretches). The “stretches” are Collateral, Household, and Banking. The ESRB (2016) states that collateral indicates the price levels and dynamics in residential real estate, household stretch is the implication of household borrowers’ debt for their consumption and other behaviour, and banking stretch indicates the potential impact of residential real estate developments on lenders. The analysis of the vulnerabilities by the ESRB (2016., 2019) showed that several

risks sectors, or “stretches” described above are important to factors that help the ESRB determine the consequences for each member state (ESRB 2016., ESRB, 2019). These factors are in turn used to issue to warnings and recommendations to each member state if the ESRB decides this is necessary. An extensive elaboration of these risks sectors is provided in Appendix J, to provide context to the ESRB announcements.

The specific risk factors that the ESRB bases on stretches described in Appendix J provides the evidence to issue certain warnings and recommendations to the member states of the ESRB. The following section describes these announcements.

2.2.3 ESRB announcements

I expect that 2 ESRB warnings and 1 recommendation have had a significant negative effect on bank returns after their announcement. Both of these warnings concern residential real estate measures taken by the ESRB. I expect this, based on the negative effect other regulations have, the continuation of a struggling banking economy, and theory that states the effect of announcements to be negative on bank performance. These statements will be further explored in following sections, this section states the specific ESRB announcements in 2016 and 2019.

The first announcement occurred on the 28 of November 2016. The ESRB issued eight country-specific (Table 1) warnings on medium term residential real estate vulnerabilities and a recommendation on closing real estate data gaps (ESRB, 2016). By issuing these warnings the ESRB uses its mandate to warn the countries for significant systemic risks. The key vulnerabilities presented by the ESRB are of medium-term nature and relate to the rising indebtedness and ability of

households to repay mortgage debt or to the valuation or price dynamics of residential real estate (ESRB, 2016). According to ESRB (2016), most of the key vulnerabilities are regarding the level of indebtedness and growth of mortgage credit. Moreover, regarding the valuation, some countries have significant vulnerabilities concerning the rate of price growth.

Table 1: ESRB Warnings; 28 November 2016, and ESRB recommendations; 23 September 2019 per country

FocusC AUT* BE DK EE FI LU MT NL SK SE UK

2016w X X X X X X X X

2019r X X X X X X

Note: Let FocusC be all the focus-countries the ESRB noted as vulnerable to systemic risk, let 2016w be the ESRB warnings issued in 2016 marked by “X”, let 2019r be the ESRB recommendations in 2019 marked by” X”.

*All countries are in their country code.

The second announcement relevant for this research occurred on the 23rd of September 2019. Where the ESRB published country-specific warnings and recommendations on medium-term vulnerabilities in the residential real estate sector (ESRB, 2019). The ESRB (2019) issued five

warnings (Table 2) and six recommendations (Table 1) to members of the European Economic Area (EEA). These warnings arose from vulnerabilities that could result in systemic risk to financial stability on the medium-term according to the ESRB report (2019). The recommendations were a continuance of warnings issued by the ESRB in 2016, where the vulnerabilities were not properly addressed. Regarding the remaining EU Member States, ESRB (2019) states that there was no indication of risk and vulnerability related to the residential real estate sector to issue warnings and recommendations. This was the case for Estonia, Ireland, Malta, Austria, Portugal, Slovakia and the UK (ESRB, 2019).

Table 2: Countries ESRB warning; 23 September 2019

AFocusC CZ* FR DE IS IE NO PT SI

2019w X X X X X

Note: Let AFocusC be the added focus-countries at the ESRB announcement of warnings and recommendations in 2019, let 2019w be the warnings issued in 2019 marked by “X”

*All countries are in their country code

The warnings in the residential real estate sector are based on numerous factors. Firstly, this area is of importance because it can have significant impact and implications for financial stability and the real economy in the European Union (ESRB, 2016). This sector in the banking industry accounts for a large part of households’ wealth and is important collateral for lenders. It therefore makes up a large percentage of banks’ balance sheets (ESRB, 2016). As mentioned before the banking industry, and with that the residential real estate sector, is an important source for

employment, investment and growth. If the residential real estate sector where to be organized in a wrong matter, significant risks may arise.

The importance of the residential real estate sector creates the need for macro prudential authorities like the ESRB to analyse vulnerabilities in these markets to translate them to actions for actors in these industries. This analysis is based on a comprehensive approach (ESRB, 2019). The ESRB and the ECB develop an indicator-based cross country framework first, to identify a set of focus countries for further analysis. Secondly, a country-specific analysis of the focus countries is performed, relating to structural, institutional and policy measures. Finally, these eleven focus-countries were assessed further, the ESRB concluded with eight focus-countries (Table 1) with certain medium-term vulnerabilities as a source of systemic to financial stability (ESRB, 2019). Either one, or a combination of risks associated with the stretches mentioned previously resulted in the countries receiving warnings from the ESRB.

2.3 Banking regulations

Eurozone regulations on banking have changed quite significantly since the financial crisis over a decade ago. This review will not go into the details of the crisis. However, it will list the most

important regulations impacting the European banking industry since. This is included to provide an overview of regulations that are significant in their effect on banks and bank performance. This will be followed by a description of regulations and other measures from the most significant European regulatory institutions besides the European Systemic Risk Board. Moreover, the general concepts are related to country specific rules of countries affected by ESRB warnings and recommendations.

The first Eurozone regulation that has influence in the banking sector are the Basel lll reforms on capital requirements. This international regulatory accord is organized after the financial crisis in 2008 to strengthen global capital and liquidity rules, aiming for a more resilient banking sector (BIS, 2010). According to Standard and Poor (2015), the four main aspects to Basel lll are capital

requirements, the leverage ratio, liquidity coverage ratio (LCR), and the net stable funding ratio (NSFR), a detailed review of these requirements can be found in Appendix A.

These reforms to banking regulation formed a real danger to the availability of credit and could possibly reduce economic activity within the banking sector (Allen et al., 2012). Moreover, Gavalas & Syriopoulos (2014) state that country-by-country estimations imply a reduction in the volume of loans and economic activity. For the banking industry in the European Union, Roulet (2018) found that the Basel lll requirements have significant negative impacts on European bank retail. These effects create an important aspect to consider in this research, as Basel lll causes

significant effects to happen in the banking industry regarding regulatory effect on bank performance. Secondly, the European Banking Union has a significant role in the banking industry since the 2008 financial crisis (ECB, 2020). The EU Banking Union aims to bring all large Eurozone banks under a single supervisor, which is the European Central Bank (ECB, 2020). According to Véron (2015) is one of the most transformative institutional responses to the crisis. The European Banking Supervision (2020) states three important elements for the EU Banking Union; the single supervisory mechanism, the single solution mechanism, and the single rulebook, details devices and effects can be found in Appendix B. Véron (2015) continues by stating that the EU Banking Union enabled the power of the European Central Bank in the European banking industry, causing centralized control.

Two regulatory devices that influence a banks loss absorbing capacity are the total loss absorbing capacity (TLAC) and a minimum requirement for own funds and eligible liabilities (MREL) according to the S&P report in 2015. Quaglia & Spendzharova (2018) state that previously the lack of regulatory coherence hampered the effective resolution of large banks, undermining financial stability. By implementing these regulatory devices by the Financial Stability Board (2020) and the European Central Bank (2020), coherence is promoted. This creates significantly more consistency between banks, which proves to be important for the banking industry because the centralized rules show stabilizing effects (Quaglia & Spendzharova, 2018). Details of these devices are found in Appendix C.

The next regulations have a key effect on the banking industry, however, they are smaller regulatory devices created for stability in the industry compared to previously mentioned regulations (S&P, 2015). The regulations include ring-fencing and a Capital Markets Union (CMU), details can be found in Appendix D. Together, they attempt to stabilize the banking industry and create a single market for capital across the European Union (S&P, 2015). Similar to previously mentioned regulations, the introduction of these devices has a negative effect on the followed profitability for banks in the European Union (S&P, 2015).

This subsection reported the most important regulations for the banking industry since the 2008 financial crisis. The regulations that were implemented since that time had significant impact on the industry (Véron, 2015) and are therefore included to provide a context that the ESRB operates in.

2.3.1 Institutional background

Besides regulations affecting the banking industry, there are different committees and other agencies governing banking on the European level. The following section will provide an overview of significant agencies that have an impact on banking regulations and performance. The supervision that is done by these agencies has changed since the 2008 financial crisis, similarly to the regulations. According to De Rynck (2016) the European Union adopted a transformational change to its banking policy for the Eurozone. He states that the EU replaced the policy model of decentralized supervision and regulatory competition with a single supervisor and a more harmonized approach, by transferring banking supervision to the European level. The following section explores this supervisory network.

Supervision on the European financial system (ESFS) is organized in a network centred around the European Systemic Risk Board (ESRB), national supervisors, and three European Supervisory Authorities (ESAs) according to the European Central Bank (2020). The ECB (2020) mentions that the ESAs are the European Banking Authority (EBA), the European Securities and Markets Authority (ESMA), and the European Insurance and Occupational Pensions Authority (EIOPA).

These agencies cooperate closely with the European Central Bank and have acquired major control in the European decision-making process (Lo, 2013). Moreover, Lo (2013) states that the ESAs are responsible for micro prudential supervision and the ESRB is responsible for macro prudential supervision. Lo (2013) states that the supervisory agencies are created to improve an existing framework of decentralised financial control. Besides, Lo (2013) continues that this system of supervision can propose regulatory standards and force implementation of technical standards in the financial industry.

This means that the ESFS has significant control on the consistent application of EU rules by having the power to structurally enforce procedures and regulations in the industry, this includes the European Systemic Risk Board.

Another important regulator in the banking industry is the European Commission (EC). According to the European Commission (2020), they are an agency creating strategies for the European Union. Its role is setting priorities, and implement policies throughout the EU. The European Commission has the right to initiate financial regulation in many but not all financial sectors (Vander Stichele, 2008). It therefore has significant power regarding most financial sectors in the EU. Vander Stichele (2008) mentions the European Parliament (EP) as another important

regulator in the European financial sector. Compared to the European Commission, however, the EP is less powerful. The EP has power in the form of co-decision making (Vander Stichele, 2008). Moreover, Vander Stichele (2008) mentions the EP makes these decisions with the EC. The EC is the more powerful committee in this relationship, as the EP tries to convince the Commission to initiate some regulatory decisions. However, both institution form important regulators in the financial industry, having an effect on bank performance. More details can be found in Appendix E.

Related to these two regulatory agencies, is the European Central Bank (ECB), and the System of European Central Banks (SECBs). The ECB (2020) describes its directive on financial stability and supervision in three parts: it monitors financial stability, it provides advice, and it promotes the cooperation between central banks and supervisors in the European Union.

The relationship between these last three committees is important for this report, as together they control most of the supervisory and regulatory powers in the European Union regarding the banking industry and financial sectors.

Lastly, a noteworthy financial institution is the Bank for International Settlements (BIS). The BIS is an international institution owned by central banks, its attempt is to foster international

monetary and financial cooperation, as mentioned in the statutes of BIS (2016). It serves as a bank for central banks, by being an agent in international financial operations. Therefore, BIS does not

specifically have a regulatory or supervisory role, it does have an advisory role.

The combination of the institutions regulating the European financial industry shows how strictly banks need to adhere to regulations. Moreover, it shows the centralized control the banking industry is still moving towards.

2.3.2 Centralized effect on country basis

As is evident, since the financial crisis banking authority and supervision in the Eurozone has moved in a centralized direction, creating a more controllable supervisory system (De Rynck, 2016). The countries participating in the Eurozone and that fall within the view of the European Systemic Risk board have comparable systems. This comes from the fact that the ECB is responsible for the supervision of banks in the Eurozone since 2014 (ECB, 2020). It supervises significant banks, and has national competent authorities supervise less significant banks within the country, subject to control by the ECB (ECB, 2020). The ECB has this role for all Eurozone countries and all countries subject to ESRB warnings. The following section moves to specific supervision within countries. Due to the

centralization there will be significant similarities between the countries. The system described in Austria in Appendix F is an extensive example of how the Eurozone regulates its national financial markets since the 2008 financial crisis. The system described is similar to other European authorities and works as an example for the European supervisory system.

The agencies described in Appendix F create an environment of strict supervision in Austria and have an important role in overall bank performance that are subject to the ESRB warnings of this research. Each country has implemented this system with a Financial Supervision Authority, working with the ECB, and a National Central Bank regulating monetary policies and financial stability within the financial sector. These two institutions have the biggest impact on financial supervision in each of the individual countries important for this research, they are similarly constructed and have similar tasks supervised by the ECB (ECB, 2020).

2.4 Economical Background

Besides regulations and the ESRB implementation, another factor that could influence bank performance in a significant way are certain economic developments. Large institutions such as banks can influence the economy with their strategies. Nevertheless, bank institutions are subject to

economic developments and shocks too. Economic developments during the time of ESRB warnings could have had an effect on bank performance and will therefore be described in the following section.

2.4.1 General economic development

General economic development is most commonly described in terms of gross domestic product (GDP), which may be used to measure macroeconomic activity and growth, as well as providing the basis for comparisons between regions (Eurostat, 2016). At the time of the first ESRB warning in November 2016 most of European countries seemed to have reached the level their economy had before the 2008 financial crisis according to the Eurostat report (2016). However, some individual countries had not yet performed as well as before. For example, Ireland had surpassed its economy before the 2008 crisis with more than 10%, however, Greece was still looking at a decline of almost 30% (Eurostat, 2016). The economy was in a state of recovery up until 2016, afterwards expansion and growth were the main components for a significant part of the European economy (Eurostat, 2016). 2 years from 2016 saw the most significant and stable growth since the global financial crisis. According to a report of the European Banking Federation (2019) the economy in the Eurozone continues to expand after 2018 and into 2019 as well. However, they mention that this is paired with a slowdown in growth. Global economic conditions arising from multiple global challenges and uncertainties create an atmosphere of stagnation in the economy (EBF, 2019).

As said, banks have a significant role in the industry, the structure and economic contribution of the banking sector will be explored in the next section. There has been a reduction in credit

institutions in Europe as a direct effect of economic developments since 2009, one out of 4 banks have disappeared since the financial crisis (EBF, 2019). Consolidation plays an important role in this reduction and helps the banking sector to reduce overcapacity and enhance profitability within the developing regulations in the European Union (EBF, 2019). This has been one way for banks to improve their overall performance according to the European Banking Federation (2019). The banking industry still represents a large part of the European economy, for example, about one in every one-hundred jobs was a banking job in 2017 (EBF, 2019) and between 3% and 4% of gross value added to the economy has come from financial services. Because of the position the banks are in in the economy, the create a strong dependence of European economies on banks, evident in the financial crisis and its resulting regulations (Szczepanski, 2019). The dependence on banks in the European Union and the importance of bank financing is a reason why the EU can be considered a bank-based economy (Szczepanski, 2019). The implementation of centralized regulation and supervision of this economy creates a similar line in the organization of these industries across European countries.

2.5 Empirical evidence

This section serves as an elaboration on the previously mentioned concepts. To assess the full effect of these concepts on bank performance, empirical evidence is provided. This section provides a contextual basis for the discussion and the formation of the hypotheses. Starting with an introduction of how the ESRB comes to its decisions on assessing risk factors and a review of the possible effects. Followed by literature on the effect of general regulations on bank performance. This is related to literature on the banking industry.

2.5.1 ESRB

The European Systemic Risk Board provides their advice based on these steps and empirical facts (ESR, 2016, 2019). Moreover, it assesses risk factors starting from a market perspective moving to a country-specific perspective. These steps are further described in Appendix G and can aid in the understanding of the possible effect the ESRB has on bank performance. As mentioned the ESRB then issues warnings or recommendations based on these risk factors. According to Smaga (2013) these tools can have an effect on the functioning of the financial system, reducing systemic risk and stabilizing the industry. The effect of stabilization in the banking industry supported by previously mentioned authors (De Rynck, 2016., Lo, 2013., Véron, 2015). This suggests that the effect of the warnings and recommendations issued by the ESRB is reduced over this period of time. The changes in regulations and their suggested effects are described in the following paragraphs.

2.5.2 Banking regulation & theory

The 2008 financial crisis caused significant changes in banking regulation and in the structure of the banking industry. According to Angkinand (2009), empirical results show that the enforcement of strict capital requirements in the banking industry cause a reduction in the effect of shocks.

Creating an industry that could control and cope with significant shocks and changes. Angkinand’s (2009) research does not show, however, that regulation changes have a significant impact on the results of banks. Moreover, Zulfikar et al. (2017) state that risk disclosure does not affect a bank’s financial performance. The ESRB issues their warnings based on risk factors (ESRB, 2016). If it is expected that the announcement of risk exposure to banks will not affect a bank’s financial

performance, the returns of banks have a more limited chance of decreasing significantly. Besides, the effect of negative news is expected to be lower compared to positive news based on Stankevičiene &

Akelaitis (2014). Therefore, the announcement, or expected announcement of an ESRB warning or recommendation could be reduced in effect.

Generally, Uhde & Heimeshoff (2009) suggest that capital regulations support financial stability across the entire European Union. The changes made since the 2008 financial crises significantly increased capital requirements in the financial industry. Therefore, based on Uhde & Heimeshoff (2009), stability across the member states of the ESRB is expected to decrease the effect of certain announcements (Angkinand, 2009).

Some theory suggests however that announcements that impact firm fundamentals will have significant effects on bank performance (Bodie et al., 1999). These jumps in prices are a direct effect of announcement and related to the efficient market hypothesis (Malkiel, 1989). He states prices reflect the market efficiently and all relevant information that is known to the public is priced in and reflected in the market prices. These announcements could have been expected, therefore reducing the effect on the announcement day (Malkiel, 1989).

All in all, theory suggests that the announcements of risk factors and vulnerabilities proposed by the European Systemic Risk Board will have an effect on bank performance, but reduced due to stabilizing factors in the financial industry. However, empirical evidence suggests that the mandate of the ESRB could have a reduced effect on the financials of banks.

3.0 Research Question and Hypotheses

After reviewing the literature on the European Systemic Risk Board, major regulations in the banking industry, and the economical background in this industry a research question is formed to study missing information in the literature. They question that arises out of literature is whether ESRB warnings or recommendations issued towards specific countries have an effect on bank performance of banks within these countries?

The stock return of banks is dependent on many factors, including the supervision of the European Systemic Risk board. The hypotheses are formed based on reviewed literature and follow

the trends of the most important factors and theories. Below a consensus of factors that form the basis behind the hypotheses.

Firstly, research on the European Systemic Risk Board suggest their mandate comes with limited, but authoritative power (Ferran & Alexander, 2011). Indicating that countries respect the warnings and recommendations addresses by the ESRB. However, the limited power can reduce pressure on countries to make fast and significant changes. This suggest a response slightly limited response from countries.

Secondly, empirical evidence regulatory devices in the banking industry suggest that the effect new announcements have on banks is reduced (Angakind, 2009., Zulfikar et al. 2017.,

Stankevičiene & Akelaitis 2014). This is related to the stabilizing effect in the banking industry due to centralized control (Quaglia & Spendzharova, 2018., De Rynck, 2016., Lo, 2013., Véron, 2015).

Finally, according to Malkiel (1970, 1989), stock prices should reflect all information available to the public in an efficient market. Therefore, at the time of announcement the information is public and should have an effect on the stock returns of the banks. This theory is significant in forming the hypotheses, because it states the initial effect of the ESRB announcements. However, Malkiel (1989) also states that the timing of knowledge is of importance, if the knowledge of an announcement comes before the announcement itself the expectations could already be priced in.

The factors above, combined with literature, suggest the effect of the ESRB announcements to be negatively correlated with bank performance. It can therefore be hypothesized that ESRB warnings issued towards countries are (reduced) negatively correlated with bank performance within these countries. Specifically, this means that it is expected that the cumulative abnormal returns of the banks are negatively affected by an ESRB announcement for a given country. Giving the following hypotheses:

3.1 Hypothesis 1

For the eight warnings issued on the 28th of November 2016 it can be hypothesized that;

- H0: The cumulative abnormal returns for banks are not affected by an ESRB announcement. - H1: The cumulative abnormal returns for banks are negatively affected by an ESRB

announcement.

3.2 Hypothesis 2

For the five warnings issued on the 23rd of September 2019 it can be hypothesized that;

- H0: The cumulative abnormal returns for banks are not affected by an ESRB announcement. - H1: The cumulative abnormal returns for banks are negatively affected by an ESRB

announcement.

For the six recommendations issued on the 23rd of September 2019 it can be hypothesized that; - H0: The cumulative abnormal returns for banks are not affected by an ESRB announcement. - H1: The cumulative abnormal returns for banks are negatively affected by an ESRB

announcement.

4.0 Data

The data necessary for the analysis included bank returns for banks within ESRB warning receiving countries, a market index return to regress the bank returns against, an event date, an estimation window, and an event window. In order to test the hypothesis data is collected from Wharton Research Data Services (WRDS). All data is analysed in STATA 14, a codebook for each ESRB warning and its analysis is provided Appendix H in the form of a Stata do file.

The bank returns are collected from: WRDS – Compustat – Capital IQ – Global daily – Security daily, and the company codes are retrieved from the GIC Sub-Industry category using the industry code for banks stated by Compustat (40101010). This database provided data for each specific date and all within country banks. The benchmark index market returns were collected from Investing.com, using equally weighted returns on MSCI Europe Historical data over the specified periods.

Table 3: Summary Statistics on key data variables (mean and standard deviation)

ESRB Warning 2016 ESRB Warning 2019 ESRB Recommendation 2019 Bank Return Mean 0.19743 0.02133 0.14867 Standard deviation 3.99973 1.53955 11.5564

MSCI Europe Market Index return

Mean -0.03709 -0.0394 -0.04054

Standard deviation 0.79855 0.78059 0.782

*Numbers are given as percentages based on descriptive statistics on daily returns over the specified period.

4.1 ESRB Warning 2016 data

For the ESRB announcement on November 28, 2016. Data is collected on closing stock prices for all eight countries that received ESRB warnings. The date range is from three months before the

event date, until thirteen days after the event date, from the 29th of August until the 15th of December 2016.

The bank specific return data consists of 84 different banks ranging over 8 different countries with a mean return of 0.0019743 across all banks within countries and a standard deviation of

0.0399973 across banks within these countries (Table 3). The market index return shows a lower mean than the return of the banks and is -0.0003709, with a standard deviation of, 0.0079855 (Table 3). This data is analysed on a total of 6,652 observations, which is the total number of trading days recognised as return dates divided over 84 banks (Noted as N in future reference). These lower ranges in means and standard deviation can be expected of such a significantly large index, as it shows the effect of a large section of the market.

The event date (t0) is equal to the date of announcement. The estimation window, used to calculate the normal returns across the banks, ranges from 10 days before the event (T2) date until 60 days before the event date (T1). The estimation window is before the event window, this follows the majority of event studies. Moreover, the estimation period is wide enough to avoid contamination of the regression. And there are no indicators to choose another estimation period (Henderson, 1990). The event window, where abnormal returns are calculated, had a total number of 11 days including the event date, with 5 days before (t1) and 5 days after the event date (t2). All observations that do not fulfil the estimation window, or event window requirements are dropped from the analysis. The number of observations dropped from analysis is 387. More details on the estimation window and event window follow in the method section.

The data was also controlled for outliers on a company cumulative abnormal return basis. The data for the 2016 ESRB warnings shows 1 significant outlier for company 24 (Figure 6). Which is due to a jump from 125.6 to 6.3 in the stock price. The stability of the stock price for company 24 indicates that this is an error in the data. Therefore, this observation was removed from the data. Besides this significant outlier, company 45 and 84 show deviating cumulative abnormal returns (Figure 6), .821 and 1.078 respectively compared to an average of .2197407 (lowered after removing significant outlier, Figure 9). The data shows no indication for errors and therefore these outliers show no justification for dropping the observations.

4.2 ESRB warning 2019 data

For the ESRB announcement on September 23rd, 2019. Data is collected on closing prices for all five countries receiving a warning from the ESRB and all six countries receiving a

recommendation from the ESRB. The date range is from 3 months before the event date, until thirteen days after the event date, from the 24th of June until the 10th of October 2019. The bank specific return data consists of 44 different banks.

For the 5 countries that received a warning from the ESRB in 2019, the mean return is 0.0002133 across banks within these countries and the standard deviation across banks within these

countries is 0.0153955 (Table 3). The market index returns show a negative return in this period and is -0.000394, with a standard deviation of 0.0078059 (Table 3). This data is analysed on a total of 3,621 observations, which is the total number of trading days recognised as return dates divided over 45 banks (N).

The parameters are similar to the ESRB warning 2016 parameters (Appendix I). A total of 47 observations are dropped based on parameter requirements.

This data is also controlled for outliers on a company cumulative abnormal return basis. The data for the 2019 ESRB warnings shows 1 outlier for company 25 (Figure 7). Company 25 shows a deviating cumulative abnormal return, .28 compared to the average of -.0005. The data shows no indication for errors and therefore these outliers show no justification for dropping the observations.

4.3 ESRB recommendation 2019

For the 6 countries that received a recommendation from the ESRB in 2019, the mean return is 0.0014867 and has a standard deviation of 0.115564 (Table 3). The market index return is lower, with a mean return of -0.0004054 and a standard deviation of 0.00782 (Table 3). The difference between the 2019 ESRB warning and recommendation is due to the number of observations in the analysis. This data was analysed on a number of 3,465 observations, which is the total number of observations recognised as trading days divided over 44 banks (N).

The parameters are similar to the ESRB warning 2016 parameters (Appendix I). For this data no observations are dropped from analysis based on parameter requirements.

This data is also controlled for outliers on a company cumulative abnormal return basis. The data for the 2019 ESRB recommendations show 5 outliers for company 4, company 27, company 29, company 35, and company 36 (Figure 8). All show a deviating cumulative abnormal return, -1.12, .277, -.168, .566, .464 compared to the average of -.0073. Only the data for company 4 show indication for errors, a price drop from 194.1 tot 25.8 is recorded. Similar to the outlier in the 2016 data, there is indication that this is an error and therefore the outlier is removed from the analysis. The other outliers show no justification for dropping the observations.

Finally, the estimation period is 50 days instead of the initial 55 days, because the number of dropped observations moves down significantly. For the ESRB warning in 2016 the amount of dropped observations halves by reducing the estimation period with 5 days. For the ESRB warning in 2019, the number drops from 121 to 47 observations. For the ESRB recommendation in 2019, the number drops from 701 to 0. By reducing the estimation window, the number of dropped observations equals the number of observations that would be dropped only due to the event window period

requirement. Therefore, it is chosen to reduce the estimation period.

After dropping the observations and looking for outliers, the analysis could continue into the preparation of the data and testing. All steps taken to prepare data for analysis, which includes

creating of variables, creating returns, combining data, dropping data, “market_return” preparation, “ret” preparation are taken using a Stata note provided by Princeton University (2007).

5.0 Methodology

The event methodology is an important tool in finance (De Jong, 2007). It can be used as a tool in many different types of financial research. According to De Jong (2007) empirical finance literature has taken this approach to test the significance of abnormal returns around event days. This research utilizes this method to study the significance of cumulative abnormal returns around the event days of ESRB announcements based on a market model residuals benchmark model. The event study was pioneered by Fama et al. in 1969, who studied the behaviour of stock prices around stock splits. De Jong (2007) states 3 steps for an event study, which are followed to describe the

computations used to form the analyses. Including the market single index model, cumulative abnormal returns and the test statistic used. Finally, limitations and problems of the analyses are addressed.

Step 1 in De Jong (2007) is identifying the event of interest. In this study there are 2 specified events that involve 2 warnings and 1 recommendation issued by the European Systemic Risk Board. Care has to be taken into interpreting the results of the event study, as the event dates carry some uncertainty in their timing (De Jong, 2007).

Step 2 involves specifying a benchmark model for normal stock return behaviour (De Jong, 2007). This study utilizes the Single Index Market Model (SIMM) method. The method assumes that the “beta” of each stock has differences and is not equal to 1 (De Jong, 2007). Abnormal returns and the normal returns are calculated based on the formulas in Figure 1.

Figure 1: Normal performance estimation and definition of abnormal return based on Single Index market model.

In the model the normal return is defined by multiplying the market return with each firm’s individual beta factor and adding this to the alpha intercept that includes the risk free interest rate. This method considers each bank’s individual risk. Where alpha and beta are OLS estimates of the regression coefficients used to estimate normal performance. This model assumes that the risk free rate included in the alpha is constant, which contrasts the fact that market returns vary over time (Binder, 1998). Brenner (1979) found that SIMM does essentially as well as other regression models used in event studies.

In step 3 the cumulative average abnormal returns are calculated and analysed based on the estimates in step 2. Average abnormal returns are used, because the analysis of individual stock returns often is not very informative due to stock price movements caused by unrelated events (De

Jong, 2007). De Jong (2007) states that the results of the average analysis are significantly more informative. Besides averaging, a longer event period is used, which is studied using cumulative abnormal returns from the start of the event period (t1), up to the end of the event period (t2). Again the cross sectional average (Figure 2) is taken from the cumulative abnormal returns to improve the in formativeness of the information.

Finally, the significance of cumulative abnormal returns is tested using a t-test (Figure 2). According to De Jong (2007) G (Figure 2) follows a standard normal distribution if the assumptions that abnormal returns are independent and have the same mean and variance are maintained. He states that this is a result of the Central Limit Theorem, concluding that, if N is large enough, the quantiles of normal distribution can be used as critical values for a t-test. In the case of event studies N should typically be larger than 30, where N is the number of banks per specific warning or recommendation in this study. The null hypotheses of no abnormal performance are rejected for a lower-sided test at 5% confidence level if G exceeds the critical value of -1.645. So, G < -1.645.

Another test is conducted based on the Princeton (2007) Stata event study guide. The cumulative abnormal return for all banks as a group is calculated, similarly to the statistic used in G (CAAR) and is tested on significance. This statistic results from a regression on cumulative abnormal return on the event date for each bank (Appendix H), therefore testing across all events. With a significance level of 5% on the one-tailed hypotheses.

Figure 2: Cumulative Abnormal Return: Average, Standard Deviation and t-test formulas

De Jong (2007) states four possible problems of the tests utilized in this research. The problems De Jong (2007) addresses are econometric issues related to the assumptions of the regressions used to calculate normal returns. More specifically, Henderson (1990) states that the model assumes that the residuals are normally distributed with a mean of zero, are not serially correlated, have a constant variance, and are not correlated with explanatory variables. Fama et al. (1969) and Boehmer et al. (1991) in De Jong (2007), state that by using the robust test procedure that calculates the abnormal returns (Figure 2) the tests will perform well even if there is event-induced variance. Therefore, these specific calculations and tests are utilized to conduct the analyses. Besides, Henderson (1990) states that the event study appears to be robust enough to limit the issues associated

with these problems. Moreover, De Jong (2007) mentions that with an N large enough (N >30), the issues are limited.

However, using daily data increases the significance of serial autocorrelation and nonsynchronous trading issues (Henderson, 1990). He states that this could result in betas of

infrequently traded stocks to be downward biased, and upward bias for more frequently traded stocks. Henderson (1990) and De Jong (2007) mention suggested techniques to correct for biases, however, in most cases the corrections do not seem to strengthen event study results.

Another possibly significant problem for this event study is event (or industry) clustering. Henderson (1990) mentions that multiple events occurring at near the same time can reduce the power of the analyses. The OLS regression in this analysis does not account for this issue and it could therefore reduce the power of the conclusions.

Limitations and problems with this event study analysis are considered when discussion the computations and results that follow in section 6.

6.0 Results

The results show an overall insignificant result for all event periods. The significance of the average abnormal return for each bank is based on the “G” statistic computed in this analysis. The results are significant if the absolute value of the “test” statistic exceeds the limit of -1.645, which indicates significance at the 5% level. The value -1.645 comes from the standard normal distribution with mean 0 and standard deviation of 1. Indicating that 95% falls between these values.

When testing across all events, the P-value for the ESRB announcement show no significant results at the 5% level (Table 5). Similarly, the G-statistic test result indicates no abnormal

performance for all events studied (Table 4).

Table 4: Test for Significance results (ESRB warning 2016/2019, ESRB recommendation 2019)

ESRB Warning 2016 ESRB Warning 2019

ESRB Recommendation

2019

G-Statistic Test Result 0.2566281 -0.0713791 -0.233184

Where G is the test statistic, calculated by √𝑁𝐶𝐴𝐴𝑅_𝑠  𝑁 (0,1) as described in the methodology. t (G) < -1.645

Table 5: Result from testing across all events, regression on cumulative abnormal returns

ESRB Warning 2016

ESRB warning 2019

ESRB recommendation

2019

Constant

0.0020994

0.0002511

-0.0068792

Error term

(0.0168)

(0.0083)

(0.0318)

P-value

0.901

0.976

0.830

Observations

84

45

44

R-squared

0.000

0.000

0.000

Robust standard errors in parentheses

p < .05 one-tailed

For the ESRB warning in 2016 when testing across all events, the P-value on the constant does not show a significant result for abnormal performance, with a value of 0.901. This value indicates that the ESRB announcement in 2016 does not affect bank returns significantly. Moreover, with an R-squared of 0 (Table 5), the model does not show that it explains variance in the results.

The G statistic to test for significance is equal to 0.2522 (Table 4), which does not indicate abnormal performance at the 5% level, where G must exceed -1.645.

For the ESRB warning in 2019 when testing across all events, the P-value on the constant does not show significance at the 5% level, with a value of 0.976 (Table 5). Indicating that the ESRB warning announcement in 2019 does not affect bank returns significantly after announcement. This result is strengthened by the value of R-squared, which is equal to 0 (Table 5).

The G statistic to test for significance is equal to -0.0713791 (Table 4), which does not indicate abnormal performance at the 5% level, where G must exceed -1.645.

For the ESRB recommendation in 2019 when testing across all events, the P-value on the constant is not significant at the 5% level, with a value of 0.830 (Table 5). Showing that the ESRB recommendation in 2019 does not affect cumulative bank returns around the announcement date. This result is exemplified by the R-squared with a value of zero (Table 5), indicating that the model does not explain any of the variance in the results.

The G statistic to test for significance is equal to -0.233184 (Table 4), which does not indicate abnormal performance at the 5% level, where G must exceed -1.645.

A graphical representation is given in Figure 9 – Figure 11 of cumulative abnormal return (CAR) on an individual bank basis per indicated event. These figures strengthen the result of the test statistic and the result of the regression. They show a steady distribution of cumulative abnormal return around zero, with occasional deviation that provide their negative direction.

Figure 9 shows the distribution of CAR for the ESRB warning in 2016. This figure shows a steady distribution besides 2 higher CAR’s, creating the positive test statistic G (Table 4). Figure 10

shows the distribution of CAR for the ESRB warning in 2019. The figure indicates a steadier distribution around zero than 2016. No major deviations show in the table, following the low (negative) test statistic G (Table 4). Figure 11 represents the distribution of CAR for the ESRB recommendation 2019. Again, the CAR is distributed steady around zero with minor deviations. The largest being negative deviation, causing the negative result of the test statistic G (Table 4).

The results indicate that the announcement of an ESRB warning or recommendation do not significantly influence the cumulative abnormal returns.

7.0 Discussion

The study set out to analyse the cumulative returns of banks that operated in countries

receiving a warning or recommendation in either 2016 or 2019 by the European Systemic Risk Board. An event study based on the single index market model around the three announcements of the ESRB indicated no significant cumulative abnormal returns for all events. These findings are placed into context of previously mentioned literature, theory and practice in the following paragraphs. Moreover, limitations of the findings are discussed to form suggestions for further research.

7.1 Findings

The warning in 2016 was the first major announcement of the European Systemic Risk Board (ESRB, 2016). The intention of this announcement was to warn specific European countries that had significant risk factors in the residential real estate sector and to provide them with resources to address these issues (ESRB, 2016). The announcements that followed in 2019 were the second major issue of the ESRB and required similar changes form countries (ESRB, 2019). This paper expects that the announcements would have had significant negative effect on the cumulative abnormal return of banks. However, the results indicate that the announcement is not directly reflected in the cumulative return of banks within the warning or recommendation receiving countries.

Noticeable in literature is amount of regulatory devices in the European Union that influence the financial industry in some way since the 2008 financial crisis. De Rynck (2016) states that these devices cause a centralized direction in the supervision of the financial industry. Uhde & Heimeshoff (2009) indicate that this causes a more stable financial system. Provided the consensus of stability in the European financial system, the result of insignificant change in the cumulative returns of banks follows this stabilizing effect. The results also follow the empirical results of research conducted on other regulatory devices. Angakind (2009) shows that regulatory changes do not have significant effect on bank performance. Besides, as the ESRB bases their warnings and recommendations on risk factors, Zulfikar et al. (2017) indicates that disclosing such risks has no effect on a bank’s financial performance. Therefore, not rejecting the null hypothesis follows a line in the literature and theory.

However, the importance of the residential real estate sector in the financial economy (Hartmann, 2015., Mazzaferro & Dierick, 2018) and the focus of the warnings and recommendations

by the ESRB on this sector (ESRB, 2016., ESRB, 2019) suggest a reaction after the announcements. Moreover, the European Banking Federation (2019) states that the EU economy relies on the banking industry. Therefore, if significant risk factors are announced by the ESRB and changes are required (ESRB, 2016., ESRB, 2019), no negative abnormal performance is an unexpected result of this study. The fact that the ESRB has no legally binding powers in addressing financial risk factors according to Ferran & Alexander (2011) could explain the lack of effect in this case. Because countries might slack in responding immediately to these announcements and markets expect no significant effect on bank returns accordingly. Moreover, De Jong (2007) states that the results can be biased in a direction because the event was expected and therefore already priced in, causing no abnormal performances in the event window of the conducted study. This effect is a limitation also described in section 7.2.

All in all, the mandate of the ESRB in a heavily financially based economy could have had its effects on cumulative bank performance. However, literature suggests that the developments of regulations since the 2008 financial crisis already had their stabilizing effect due to centralized supervision and control. Causing results in this study that indicated no deviation from normal performance after an ESRB announcement.

7.2 limitations

Extensive research into related literature and empirical evidence surrounding the

announcements by the European Systemic Risk Board provide a solid context for this study to operate in and base its conclusion on. However, there consist certain limitations to this study that constrain the interpretation of the results and their generalizability to practice. This sections describes limitations to the event study first, and secondly limitations in research found in this study.

The first limitation is related to the timing of the event date (Chen, 2017). In this research the announcement date is directly specified by the European Systemic Risk Board and therefore the identification of the event date formed no problem. However, the announcement could have been expected by the market before it was made, which indicates that the effects of the announcement happen in an earlier stage. Therefore, showing no significant results during the actual announcement and relating event windows (De Jong, 2007). This a common limitation in all event studies

(Henderson, 1990) and is minimized here by choosing the best possible event date.

The second limitation according to Chen (2017) is the methodology used to calculate cumulative abnormal returns that could induce upward bias. Chen (2017) states that this bias arises from the observation by observation rebalancing to equal weights needed to calculate the aggregate cumulative abnormal return. By using the SIMM model provided by De Jong (2007) and the Princeton University (2007) Stata analysis method this effect is minimized. As described in the method section, the market model method is robust against the biased effects (Henderson, 1990).

A third limitation to this method is the use of cumulative average abnormal return to test for significance. This test analyses the average effect of the ESRB warning or recommendation over all

the affected countries and banks. If the overall result of the analyses is insignificant, an individual bank could still have significant results. This research was interested in the cumulative average result and therefore this method is chosen instead of testing individually.

Lastly, normal returns in an event study are calculated using an estimation window, by increasing this window, the estimation results usually become more reliable. However, other factors need to be considered when determining the estimation window. The window is carefully considered in this research.

Another limitation to this study is the financial industry itself. Literature indicates that this industry has become vastly regulated since the 2008 financial crisis. Countries and banks operating in the industry have to adhere to a wide range of regulatory devices. Besides, it is important to note that no specific controls are used in this research. Based on the literature it is noticeable that bank returns are affected by a significant number of factors. These factors could in turn influence bank return and cause biased results. Moreover, research is limited on the power of the ESRB in relation to these other regulatory devices. When interpreting the results, the relative power of other devices is important, as it could indicate that the results are caused by not only the ESRB.

The limitations to the event study methodology are common among all these type of studies (Henderson, 1990) and are carefully considered when decisions were made on its aspects. To provide results as reliable as possible within the constraints of the method. However, the limitations to an event study and the financial industry need to be considered when interpreting the results in a context outside of this research.

7.3 suggestions for further research

Based on the limitations of this specific study and the literature, suggestions are made for further analyses on this topic.

The European Systemic Risk Board is one of many regulatory devices in the European financial industry. However, few research is conducted on its power in this industry and the effect of its mandate. The gap that exists on relative power in the financial industry is interesting and definitely important in all research on specific regulations and their effects. Therefore, I suggest that research is conducted on relative power of regulatory devices within the European Union, before further research is done on individual effects. This might significantly increase the reliability and strength of research conducted on the effect of individual regulations and institutions in this industry.

8.0 Conclusion

The European Systemic Risk Board assesses risk warnings in the European financial industry. It issued 1 warning in 2016, 1 warning in 2019, and 1 recommendation in 2019. This paper suggested a negative effect of each event on the cumulative performance of banks. The expectation of

player in the financial industry, monitoring major risk factors that will influence a bank’s performance. Therefore, a warning concerning these factors was expected to result in significant changes in cumulative abnormal returns for banks within receiving countries.

However, the results indicated that the announcements by the ESRB have no significant effect on the cumulative return for all events studied.

Research suggests that the influence of regulatory devices have decreased. The development of many institutions and regulations since the 2008 financial crisis has caused an increasing stability in the industry. Therefore, it is suggested that new changes in the system have no significant influence on a bank’s financial performance due to this effect. Besides, the ESRB does not show sufficient power in their mandate regarding the announcements to cause concern and immediate effects.

All in all, the recent stabilization in the financial sector suggest that regulatory institutions independently do not affect performance significantly, including the announcements made by the European Systemic Risk Board.

References

- About BIS – overview. (2020). bis. Retrieved May 28, 2020, from

https://www.bis.org/about/index.htm

- Allen, B., Chan, K. K., Milne, A., & Thomas, S. (2012). Basel III: Is the cure worse than the disease?. International Review of Financial Analysis, 25, 159-166.

- Angkinand, A. P. (2009). Banking regulation and the output cost of banking crises. Journal

of International Financial Markets, Institutions and Money, 19(2), 240-257.

- April 2015, Standard & Poor’s Rating Services, McGraw Hill Financial: Key Regulations Impacting the European Banking Industry

- Basel Committee on Banking Supervision (BIS). (2017, December). High-level summary of Basel lll reforms. Retrieved May 28, 2020, from

https://www.bis.org/bcbs/publ/d424_hlsummary.pdf

- Bank for International Settlements (BIS). (2010, revised, June 2011). Basel lll: A global regulatory framework for more resilient banks and banking systems. Basel Committee on Banking Supervision.

- Banking Union. (2020). Banking supervision Europa. Retrieved May 26, 2020, from

https://www.bankingsupervision.europa.eu/about/bankingunion/html/index.en.html

- Binder, J. (1998). The event study methodology since 1969. Review of quantitative Finance

and Accounting, 11(2), 111-137.

- Bodie, Kane, Marcus (1999). Investments. Edition 11, McGraw Hill Financial. - Brenner, M. (1979). The sensitivity of the efficient market hypothesis to alternative

specifications of the market model. The Journal of Finance, 34(4), 915-929.

- Brown, S. and J. Warner (1985). Using daily stock returns: The case of event studies. Journal of Financial Economics, 14, 3-31. 


- Chen, C. (2017). Limitations to Event Studies and How They Apply. Available at SSRN

2982219.

- Darvas, Z., Mazza, J., & Midoes, C. (2019, June 3). A european atlas of economic success and failure. Retrieved from https://www.bruegel.org/2019/06/a-european-atlas-of-economic-success-and-failure/

- De Jong, F. (2007). Event studies methodology. Lecture Notes.

- De Pauw, P. (2019, May 1). Prudential regulation of banks in belgium. Lexology.com. Retrieved from https://www.lexology.com/library/detail.aspx?g=8fd1923f-f7c7-4d92-9673-170f98a7f7e3

- De Rynck, S. (2016). Banking on a union: the politics of changing eurozone banking supervision. Journal of European Public Policy, 23(1), 119-135.