Capital Requirements Regulation/Capital Requirements Directive IV: A critical review. How effective are the regulations, with regards to risk-weighted assets for SMEs in German co-operative and savings banks?

(1)

Universiteit Leiden - Public Administration: Economics and Governance Track Master’s Thesis

Capital Requirements Regulation/Capital Requirements Directive IV:

A critical review. How effective are the regulations, with regards to

risk-weighted assets for SMEs in German co-operative and savings

banks?

Jonas Emanuele Fischer s2149044

Supervisor: Dr. Jeannette Mak Second Reader: Dr. Peter van Wijck

11 January 2019

(2)

Abstract

The German ‘Mittelstand’ is regarded as the engine of the German economy. These small and medium-sized enterprises (SMEs) in Germany are primarily financed by co-operative and savings banks. This paper explores whether financing this crucial segment of the economy in Germany is jeopardized by the risk-weight of assets, a nominal risk correction instrument, in the Capital Requirement Regulation (CRR) of the EU. Additionally, this paper considers whether the CRR/Capital Requirement Directive IV (CRD IV) disproportionality affects the primary financers of the ‘Mittelstand’, potentially sustaining unintended consequences. Accordingly, this paper explores two different aspects of the regulations. What is the impact of risk-weighted assets on the availability of loans to SMEs and is the regulation disproportionately affecting co-operative and savings banks? On top of that, this paper explores whether the problems identified are dealt with in the draft CRR II/CRD V revision of the legislation. As a result of the controversy on this topic in the literature, a systematic review is performed over two different periods of time, before November 2016 and after November 2016. This is because of the draft CRR II/CRD V paper the European Commission has published at that moment. Two main findings stand out. First of all certain SMEs do have problems with availability of loans, but this can be considered a minor problem that is also being addressed to a certain extent in the CRR II/CRD V. The second finding is that there is disproportionality in the current CRR/CRD IV framework. This problem is also being addressed, but arguably unsatisfactory. Additionally, although a significant part of the literature agrees on the existence of a disproportionality problem for smaller credit institutions, there is disagreement on how to tackle this. Furthermore, any solutions proposed are subject to significant controversy. This is highlighted through opposing preferences in policy outcomes of studies by, for example, interest group and supranational institutions.

(3)

List of Abbreviations

A-IRB – Advanced Internal-Rating Based Approach AEE – Advanced and Emerging Economy

BaFin – Bundesanstalt für Finanzdienstaufsicht (Federal Financial Supervisory Authority) BCBS – Basel Committee on Banking Supervision

BIS – Bank of International Settlements CET 1 – Common Equity Tier 1

CRD I-V – Capital Requirement Regulation I-V CRR – Capital Requirement Regulation

EAD – Exposure at Default

EBA – European Banking Authority ECA – European Court of Auditors

ECAI – External Credit Assessment Institution

ECOFIN – European Council’s Economic and Financial Affairs Committee ECON – European Parliament’s Economic and Monetary Affairs Committee EP – European Parliament

F-IRB – Foundation Internal-Rating Based Approach GDP – Gross Domestic Product

HI – Herfindahl index

IIF – Institute of International Finance ITS – Implementing Technical Standard IRB – Internal-Rating Based Approach

KfW – Kreditanstalt für Wiederaufbau (German development Bank) LGD – Loss Given Default

(5)

LSI – Less Significant Institution MS – Member State

NCA – National Competent Authority PD – Probability of Default

RWA – Risk-weighted Assets

RTS – Regulatory Technical Standard SA – Standardised Approach

SAFE – Survey on the Access to Finance of Enterprises SME – Small- and Medium-Sized Enterprises

(6)

List of Figures

Figure 1: Structure of the Banking Union ... 14

Figure 2: Different types of risk and the determinants of risk ... 16

Figure 3: Loan size and enterprise perceived difficult to access loans in Germany ... 26

Figure 4: Interest Rate charged to non-financial corporations in Germany ... 28

Figure 5: German SMEs most pressing problem ... 30

Figure 6: German SMEs perceived importance to finance ... 31

Figure 7: Perceived external funding gap in different Euro Area countries. ... 31

Figure 8: Possible determinants of bank lending behaviour. ... 32

Figure 9: Heat map of study by author ... 46

Figure 10: Heat map of studies by author over the type of study ... 48

Appendix Figure 1: Status quo compared to Basel III ... 108

Appendix Figure 2: Euro Area SMEs relevance of different financing sources ... 108

Appendix Figure 3: New loans issued to non-financial corporations in Germany. ... 109

Appendix Figure 4: Commission proposal to reduce reporting requirement ... 109

Appendix Figure 5: Heat map of studies by author divided into the two periods of examination ... 110

(7)

List of Tables

Table 1 – Summary of select RWA ... 20

Table 2 – Amount of studies published over the year ... 42

Table 3 – Breakdown of the studies by type ... 43

Table 4 – Possible inherent bias of studies by type ... 43

Table 5 – Data bias of study by type ... 44

Table 6 – Possible inherent and data bias by type ... 44

Table 7 – Area of study over the type ... 45

Table 8 – Summary of statistics ... 46

Table 9 – Studies exploring the SME dependent variable over the year ... 50

Table 10 – Summary of proportionality dependent variable categorisation over the year ... 51

Table 11 – SME score average (amounts of studies in brackets) ... 52

Table 12 – Average SME score excluding: possibly (amounts of studies in brackets) ... 53

Table 13 – Proportionality score average excluding: not discussed ... 54

Table 14 – Proportionality score average excluding: not discussed and possibly ... 54

Table 15 – Overview of Dependent Variables and findings of the various studies of the systematic review ... 55

Table 16 – Average score by period of examination over type. Excluding: not discussed ... 58

Table 17 – Average score by period over area of focus. Excluding: not discussed ... 63

Table 18 – Average score by period over type. Excluding: not discussed...63

Table 19 – Summary of select proposals of the various institutions...71

Table 20 – Overview of findings of the discussion...77

Appendix Table 1 – Method over Type ... 111

Appendix Table 2 – Language over Type of study ... 112

Appendix Table 3 – Breakdown by language and discussion of the SME dependent variable ... 112

Appendix Table 4 – SME Dependent Variable: Method and region over type ... 112

Appendix Table 5 – Breakdown by language of the disproportionality dependent variable (excluding not discussed) ... 113

Appendix Table 6 – Method, area of focus, and categorisation of study over type regarding the proportionality dependent variable ... 113

Appendix Table 7 – Problems (and possible problem) with proportionality and SME ... 114

Appendix Table 8 – average score by period of examination over area of focus, excluding: not discussed ... 114

Appendix Table 9 – Table showing various columns within the database, sorted by date ... 115

Appendix Table 10 – Table showing the disproportionality dependent variable response and score as well as the SME availability of loan problem dependent variable and score, sorted by date ... 117

(8)

Section I - Introduction

Section I.1 – Relevance of loan supply for SMEs

“Although commercial banks create money through lending, they cannot do so freely without limit. Banks are limited in how much they can lend if they are to remain profitable in a

competitive banking system.” - (McLeay et al., 2014, 14)

Due to the financial crisis of 2007-08, significant counter-cyclical financial measures have been put in place in order to try to reduce the impact a banking crisis may have on the real economy. The aim of the counter-cyclical financial measures is to increase money circulation. This can be accomplished through inter alia monetary policy such as decreasing interest rates and/or increasing government spending. Dario Scannapieco, Vice-President of the European Investment Bank, stated in 2015 that there is an “abundance of liquidity and the important thing is that this liquidity is directed to productive investments” (Euronews, 2015). Therefore, exploring whether this “abundance of liquidity” reaches SMEs, the main driver of economic growth in the EU, is of paramount importance. Non-financial SMEs in Europe add 56.8% of the total value added to the European economy (Muller et al., 2016). This abundance of liquidity should reach SMEs via the usual lending mechanism of banks.

The way banks refinance themselves – in order to invest/loan – is primarily determined by two sets of interest rates, the deposit interest rate and the marginal lending rate. The deposit interest rate refers to the European Central Bank (ECB) paying banks for any excess reserves held overnight. The deposit rate also refers to the interest banks charge for lending from another (Hamilton, 2018). The marginal lending rate is the rate at which banks borrow from the ECB.

What is omitted from the way banks refinance themselves is how this mechanism affects lending itself. Specifically, this lending depends on the risk-weighted assets (RWA), which has the possibility to critically impact SMEs. RWA is a correcting instrument that ‘fixes’ the nominal value of the financial asset in order to reflect the ‘true’ value of risk associated with the asset. An asset in this specific case is the lender relationship between the bank and the third party. The instrument attributes a higher risk-weight to the ‘riskier’ assets, which include

(9)

loans to SMEs. Thus, this can influence the lending mechanism described above by diverting resources to comparatively safer investments that would alternatively be allocated to SMEs. Therefore, exploring the effects, and possible shortcomings, of this component of the Capital Requirement Regulation (CRR), in light of the revision of the CRR II and the Capital Requirement Directive V (CRD V), is very important. This is because, banks are henceforth more constrained than ever to lend money to actors in the real economy. Thus, exploring the impact RWA has on loan supply to SMEs in the wake of a major revision to the legislation, the CRR II/CRD V, is particularly contemporary and relevant. Furthermore, the CRR consists of inter alia reporting requirements for all banks, which can be regarded as fixed costs. This paper will explore the possible disproportionality of those requirements for smaller credit institutions. This is because, it can increase the lending costs for SMEs to make up for the higher fixed costs, or have other unintended consequences.

Before proceeding, it is important to emphasise the difference between a directive and a regulation. A European directive has to be transposed into national law and is subject to certain interpretation by Member States. Whereas, a European regulation is directly binding throughout all Member States on the specified date set in the legislation. Furthermore, regarding the CRR/CRD IV, the main differences between the regulation and the directive is that the CRR is primarily based on quantitative demands concerned with a banks solvency whereas the CRD IV is based on qualitative requirements. Examples of the CRR include inter

alia the calculation of the leverage ratio, the capital adequacy definition or the reporting

requirements. Whereas the CRD IV inter alia stipulates how and when national regulators should step in, or how credit institutions internally assess the quality of the capital adequacy (BaFin, 2013). This paper henceforth will primarily be based on the CRR, however uses the CRR/CRD IV in tandem, as most of the stipulations of the regulations go hand-in-hand.

(10)

Section I.2 - Why Germany and what is it a case of?

“One of the main reasons the economic and financial crisis became so severe was that the banking sectors of many countries had built up excessive on- and off-balance sheet leverage. This was accompanied by a gradual erosion of the level and quality of the capital base. At the

same time, many banks were holding insufficient liquidity buffers.”

- The Basel Committee describing the causes of the financial crisis (BIS, 2009, 1).

Germany, due to its diverse banking culture, represents a good case study of the regulation for several reasons. First of all, Germany is chosen as it represents 21.3% (in 2017) of total Gross Domestic Product (GDP) of the EU (Eurostat, 2018). Thus it can be considered being the economic powerhouse of the EU. Specifically, SMEs are chosen because in Germany 61% of the working sector is employed in a business defined as an SME (Federal Statistical Office Germany, n.d.). SMEs for the purpose of this paper are defined in accordance to Commission Recommendation 2003/361/EC, where the enterprise must comprise less than 250 employees, and have an annual turnover of no more than €50 million or an annual balance sheet of not more than €43 million. In addition, according to the German statistical office, in 2016 99.3% of businesses in Germany are considered to be SMEs (i.e. the famed German ‘Mittelstand’) (ibid.). The Key Figures underscore the importance of SMEs for Germany. Additionally, Germany also provides a good case study of proportionality on smaller credit institutions, as there are over 400 savings banks and over 900 co-operative banks (Bundesbank, n.d.). Savings banks are public-sector banks comprising of Sparkassen, whereas co-operative banks are collectively owned by the depositors and consist of Volks- and Raiffeisenbanken. On top of that, generalizations on the effect of the CRR/CRD IV could be made due to the adverse impact on the European real economy. This is because of the influence RWAs can have on loan supply to SMEs.

As a result, specifically exploring if, and how, RWAs impact loan supply to SMEs is of significant importance, since bank loans are an important source of finance for SMEs. The primary financers of SMEs in Germany are savings and co-operative banks, as they have combined above 60% market share for loan supply to SMEs (DSGV, 2018, 7). Since, RWAs have the possibility to affect loan supply to SMEs due to the ‘true’ reflection of risk, this section turns to the research question, aim, and structure of the paper.

Key Figures on German SMEs: 2,467,686 Businesses in Germany 99.3% of which are SMEs 61.2% of people in Germany are employed in an SME

46.7% of Gross investment in tangible goods

47.6% of Gross value added at factor cost

(11)

The research question is:

How effective was the CRR/CRD IV, with regards to risk-weighted assets for SMEs in German co-operative and savings banks?

The aim will be to evaluate how the risk-weight of assets has impacted SMEs. In addition, this paper will also explore the proportionality aspect of the CRR. Lastly, due to the current proposed amendments to the CRR/CRD IV, i.e. the CRR II/CRD V, this paper will also establish a framework with which to assess the success of the legislation with regards to the identified problems.

Therefore, in order to explore this phenomenon, the structure of the paper is as follows. Section II explains the genesis of Basel and how it relates to the European Banking Union and CRR/CRD IV as well as the dynamic of RWAs. Subsequently, section III evaluates the literature and attempts an initial analysis. Finding no definitive conclusion in section III, section IV sets up the methodology of the data exploration employed for the purpose of this paper, which is a systematic review. Section V reports the data and interprets the results

prima facie. Section VI dives further into the data by analysing and highlighting the most

significant findings in relation to the research question. Consequently, Section VII discusses these findings with regards to the CRR II/CRD V and explores the limitations of the paper. Section VIII will round the discussion off with concluding remarks, and provides feasible policy recommendations for policy-makers as well as recommendations for SMEs.

(12)

Section II: The Genesis of Basel, the Banking Union, and the

dynamic of Risk-weighting of Assets

This section initially explores the genesis of the Basel Accord. Subsequently, this section turns to explaining the EU’s Banking Union and how the CRR/CRD IV incorporated the Basel Accord. Lastly, this section underscores the dynamic of RWA.

Section II.1 - Banking on Basel

“If the Bundesbank had knowledge of impending collapse, and they did not make this available to the banking community, then what recourse does the market have? If the

Bundesbank's not in charge, who is?”

- Unnamed American banker, Bonn 1978 (Hershey Jr., 1978)

It’s of interest to briefly outline the genesis of the CRR/CRD IV. This is because it didn’t arise from within the EU, but is the result of the Basel Accord. The spectacular Herstatt Bank failure of 1974 in West Germany is considered to be one, if not the most, important reason for the creation of the Basel Committee on Banking Supervision (BCBS), then known as Committee on Banking Regulations and Supervisory Practices (BIS, 2014; BIS, 2018). This is because, Herstatt’s foreign exchange exposure was around three times the capital it held, fuelling international turmoil especially in New York, where it came close to causing other institutions to collapse (Banque de France, n.d.; BIS, 2014). This case, alongside other notable bank failures such as: Franklin National Bank in 1974 or Banco Ambrosiano in 1982 illustrated how inadequate regulators were performing their task and how self-regulation was failing (Mourlon-Druol, 2015). Furthermore, these bank failures underscored how the illiquidity of a medium sized bank can threaten to take down the international banking system due to chain reactions (Genschel and Plumper, 1997; Banque de France, n.d.).

These bank failures were the result of banks operating on lower capital and consequently becoming more fragile (Genschel and Plumper, 1997). As a result, in 1974 the governors of banks from the G10 countries established a Committee on Banking Regulations and Supervisory Practices. The Committee numerously called upon “supervisors to resist any further erosion in capital ratios" (Kapstein, 1989, 333). The concern was that of a regulatory “race to the bottom”. This is the case where one country lowers regulatory standards thereby forcing other countries to follow suit, as otherwise banks will move to the country that offers the lowest regulatory standards. Therefore, this results in national regulators not being able to

(13)

implement policies that would be necessary for the stability of the national financial system in the respective country (Tarullo, 2008). The Committee started to negotiate in order to achieve common standards. However, as Reinicke (1995, 162) highlights: due to the differences between the Anglo-Saxon’s and continental Europeans, specifically Switzerland and Germany which argued that their national regulators were up to the task, agreement seemed rather impossible. Only in 1986 when bilaterally the United States and the United Kingdom agreed on capital adequacy, did the ball start rolling. This is because New York and London were the largest financial centres and agreeing on common capital adequacy standards signalled to other countries an understanding that could not easily be ignored (Reinicke, 1995, 168; Genschel and Plumper, 1997). Between 1986-87 negotiations continued, in the so-called Cooke Committee, and culminated in an agreement in December 1987 on the "international convergence of capital measurement and capital standards" (Kapstein, 1989, 344; BIS, 2018). The final draft of the Basel Accord was published in July 1988. This Basel framework superseded the U.S./U.K. bilateral accord and also combined the concerns voiced by Germany and Japan, due to their diverse banking structure, by introducing a two-tier capital definition (Kapstein, 1989, 344). Additionally, the Basel Accord (later dubbed ‘Basel I’) introduced one of the most fundamental principles in banking regulation: the risk weighting of assets.

Basel I, however, is not a binding document, just an advisory one. As a consequence of Basel I the EU introduced, for example, the Capital Adequacy Directive (93/6/EEC) in 1993. Furthermore, the EU in 2000 introduced directive 2000/12/EC a Banking Consolidation Directive (Practical Law Financial Service, 2007). However, the pinnacle of these regulations by the EU came after the publication of Basel II in 2004. The Capital Requirement Directive (CRD) (Directive 2006/49/EC) was adopted in 2006 and directly replaced the Capital Adequacy Directive. Additionally, Directive 2006/48/EC replaced the Banking Consolidation Directive. Therefore, technically speaking the CRD I is made up of 2 different Directives (Kattenlaw, n.d.). Consequently, future amendments in 2009, 2010 and 2013 were made, which replaced CRD with CRD II, CRD III, and CRD IV respectively. Furthermore, in 2013 the CRR (Regulation (EU) No 575/2013) was adopted. The CRR contains significant provisions that relate to the majority of the Basel III reforms. However, not all requirements of the CRR/CRD IV entered into force on the 1st of January 2014, rather requirements are phased-in taking full effect in 2019. Lastly, the current CRR/CRD IV framework is being thoroughly revisited by an amendment being discussed to date, dubbed the CRR II/CRD V.

(14)

Section II.2 - How are capital requirements strengthened?

Ever since Basel II, the financial system is tiered into 3 different categories. These tiers are also found in the CRR/CRD IV. The different tiers address varying aspects. Pillar 1 concerns the minimum capital requirements that any credit institution is supposed to hold at all times. Thus, pillar 1 addresses the different categories of risk (explained below) and risk-weight of assets. Pillar 2 concerns further capital/buffer requirements that National Competent Authorities (NCAs) are able to impose during the supervisory review process (European Commission, 2016a). Pillar 3 concerns the market discipline of banks, which are the reporting requirements to the NCAs or the European Banking Authority (EBA). Consequently, as figure 1 exhibits, the way Basel III was transposed in the EU lays the foundations on which the Banking Union is built on.

Nevertheless, turning specifically to RWAs, these are employed in the CRR as a percentage of capital required by banks to keep (Pillar 1). Specifically, the CRR stipulates, according to Article 92 of the own funds requirements, that the 8 per cent capital requirement relating to the RWA have to, at least, amount to 6 per cent of tier 1 capital, of which at least 4.5 per cent must be backed by Common Equity Tier 1 (CET 1) capital (Bundesministerium der Finanzen, 2017; EBA, n.d.(a)). For a visualization of these requirements see Appendix Figure 1.

Figure 1: Structure of the Banking Union. Source: Bank of Valletta, 2017; Regulation (EU) No 575/2013; own illustration

(15)

However, how do capital requirements relate to the research question of the impact of RWAs for SMEs in German co-operative and savings banks? The relationship between capital requirements and the question is that it establishes that the higher the risk-weight of assets, the more capital will have to be set aside in order to abide by the Pillar 1 minimum requirements. In other words, the RWAs will amount to a certain sum. This sum has to at least amount to 4.5 per cent CET 1 capital. Tier 1 capital is considered to be the “best” capital available, basically as good as liquid capital. Examples of Tier 1 Capital include: bank equity from shareholders, any reserves of the bank, inflows from sale of assets, etc. Thus, these percentages measure the degree of capitalization of the bank with regards to its assets relative to the risk that the respective bank is pursuing. In more technical terms, RWAs are defined as: the product of the exposure value of a counterparty default risk position and the risk weight of the borrower (Zschezsche, 2018). The difference between 4.5% CET 1 capital and the 6% tier 1 capital can be formed by “soft core capital” (e.g. silent deposits) (see: Zschezsche, 2018).

RWAs are only one of the tools, specified under the CRR, employed by regulators to prevent a bank sending shockwaves, in the case of failure, through the economy. Thus, as just established, banks are required to keep a pre-defined minimum amount of capital to cover the risk of the possibility of lenders defaulting on their loans. In practice, this is executed on a bank’s balance sheet by categorising each exposure either as: credit risk, market risk, or operational risk. Consequently, after determining the weight of the exposure, the sum of all the exposures is the RWA. Thus, initially an exposure is evaluated by categorising it according to the different risk types. Then, in other words, the RWA determines how much capital should balance the risk. Thus, the crux of RWA is that it is the sum of all the exposures of a bank’s balance sheet in order to offset the risk.

Since the focus of RWAs is on the first pillar of the CRR there are 3 different ways with which to categorize RWAs. These are: market risk, operational risk, and credit risk. Although the main classification of risk for the purpose of this paper is credit risk, this paper will briefly explain the 3 categories.

Market Risk relates to risk of “losses in on- and off-balance sheet positions arising from

adverse movements in market prices” (EBA, n.d. (b)). These losses are usually due to significant negative macroeconomic events that increase the risk to a trading portfolio.

Key Values:

Article 92:

a) CET 1 Capital: 4.5% b) Tier 1 Capital: 6% c) Total capital ratio: 8%

(16)

However, there are also other market risks to ponder over, such as: interest rate risk, exchange rate risk, and commodity price risk (Ord, n.d.).

Operational Risk relates to “the risk of losses stemming from inadequate or failed internal

processes, people and systems or from external events” (EBA, n.d.(c)). Thus, Operational Risk includes fraudulent activity, mistakes by employees, and even legal risk (Ord, n.d.).

Credit Risk relates to developments in “…credit and dilution risk[s] in respect of all the

business activities of an institution, excluding the trading book business” (EBA, n.d.(d)). Hence, it boils down to the default on a debt payment. However, this is closely tied to market risk as when credit risk increases market risk does too, manifested in the form of increased premiums as well as increased prices (Ord, n.d.). Furthermore, banks themselves also carry credit risk, as a possible credit rating downgrade will send the bond price spiralling lower, thereby affecting the bond investors (ibid.). A dynamic illustration of the different risk types can be examined below by figure 2.

Figure 2: Different types of risk and the determinants of risk.

As aforementioned, the primary focus henceforth is going to be credit risk. This is because, when considering loans to SMEs, credit risk is the associated exposure on a bank’s balance sheet. However, there is an administrative and reporting burden associated with the operational and market risk that will implicitly (i.e. without mentioning market/operational

Di ffer en t t yp es o f r is k an d de te rmin an ts o f r isk Credit Risk Standardised Approach (SA) Internal-Rating Based Approach (IRB/IRBA) Operational Risk Basic indicator Standardised Approach (SA) Internal-Rating Based Approach (IRB/IRBA) Market Risk Foreign-exchange risk

Position risk Inter alia Interest Rate _risk

(17)

risk by name) be discussed in the following sections. The exposures on a given bank’s balance can only be determined by either the bank employing the Standardised Approach (SA) to assess credit risk, or the Internal Rating Based (IRB) approach to determine credit risk. This is the case for all banks operating in the EU (BaFin, 2016). This is explicitly stated in the CRR (Regulation (EU) No 575/2013), where, according to Article 107 (1); “[credit] institutions shall apply either the Standardised Approach provided for in Chapter 2 or, if permitted by the competent authorities in accordance with Article 143, the Internal Ratings Based Approach provided…” the credit institutions meet the stipulations laid out previously.

The Standardised Approach for assessing credit risk relates to Article 111 of the CRR (Regulation (EU) No 575/2013). This states that: “the exposure value of an asset item shall be its accounting value remaining after specific credit risk adjustments, additional value adjustments in accordance with Articles 34 and 110 and other own funds reductions related to the asset item have been applied.” Thus, in order to determine the risk-weight of an asset, the bank is required to assign a credit rating agency this task. However, according to regulation (EC) No 1060/2009 Article 18 (3) only “credit rating agencies registered in accordance with this Regulation“ can be brought into this credit risk assessment process. As of the last available update on the 10th October 2018, there are 44 credit institutions registered or certified within the EU (ESMA, 2018). The SA is only one of the two approaches outlined in the CRR to assess credit risk, the other being the IRB approach.

The Internal Rating Based approach for assessing credit risk relates, as the SA, to the risk-weighted exposure amount of an asset item. However, as the name suggests, this method is developed internally at a credit institution by creating, through own or external databases, a mechanism in order to assess different asset items. This mechanism has the advantage of taking into account specific internal processes to determine the risk-weight of asset items. However, in order to use the IRB approach, the credit institution is required to be approved by the NCA in the country. For the case of Germany, the supervisory authority is ‘BaFin’, the Bundesanstalt für Finanzdienstaufsicht (Federal Financial Supervisory Authority). In Germany, as of August 2017 there are 22 credit institutions allowed by the supervisory authority to use the Standard-IRB approach to assess credit risk (BaFin, 2017), of which 9 are permitted to employ this on a consolidated level, and 18 banks are permitted to employ the advanced-IRB approach to assess credit risk (ibid.). The difference between the standard and advanced-IRB approach is that in the standard-IRB approach the credit institutions calculate only Probability of Default (PD) internally, whereas advanced-IRB entails that the credit

(18)

institutions calculates PD and Loss Given Default (LDG) internally (Kiene, Hessmert, and Meier, 2017). The majority of the banks operating within Germany are subject to the SA, rather than the IRB. This requires the banks to entrust the risk weighting of assets to the approved credit rating agencies. A possible consequence of this is that of a conflict of interest for rating agencies. This is due to rating fees. Rating fees by being the main source of revenue for the rating agencies, allows for the possibility of lower willingness for conservative risk estimates (Anwar, 2016). This is because the purchaser alternatively brings his business to a more liberal external credit assessment institution (ECAI). Alternatively, this is potentially the result of rating agency foresight, where the rating process just reflects the competitive distortion, rather than creating it (Hau, Langfield, and Marques-Ibanez, 2012). Textbox 1 summarizes the differences between the SA and IRB with regards to Basel I vs. Basel II.

In order to amend, change, or supplement the CRR/CRD IV, the Commission has two different methods. Specifically, these are either Delegated Acts (DAs) or Implementing Acts (IAs). DAs are granted via the delegation from the European Parliament (EP) and European Council of an already adopted EU law. It is subject to strict conditions inter alia not changing the essential elements of the law, or the EP as well as the Council revoking the delegation at any moment (EC, n.d.). IAs are directly implemented by the Commission, or in rare instances by the Council for certain legal areas (EC, n.d.). However, they have to be agreed on by a committee where every Member State is represented. IAs are usually employed as a device in order to standardise procedures (Kiene, Hessmert, and Meier, 2017). As of the 26th November 2018, the Commission has amended the CRR/CRD IV through 35 Regulatory Technical Standards (RTS), in essence DAs, and 31 Implementing Technical Standards (ITS), in essence IAs. However, more are planned. Most of the IAs/DAs are short, spanning a couple of pages, whereas a minority are hundreds of pages (for example 300 page: Commission

Textbox 1: Summary of RWA in Basel I vs. Basel II

The calculation of RWA under Basel I was based on assigning assets into different groups with pre-assigned risk-weights. Consequently, it was possible to simply calculate the capital required in order to offset the loan.

Basel II changed this by, in essence, allowing banks to choose between two alternatives. The two are: the Standardised Approach (SA), and the internally calculated Internal Ratings-Based Approach (IRB / IRBA). The SA, in essence, is mostly the same as Basel I with fixed risk-weights. Whereas, within the IRB, there are two supplementary models that banks can be granted to employ, the Foundations IRB (F-IRB) where the bank calculates Probability of Default for the counter-party, while the other elements, such as: the Exposure at Default (EAD) or Loss Given Default (LGD) are given by the regulator. The other model is the Advanced IRB (A-IRB) where the bank estimates everything; LGD, EAD, and PD in order to obtain the risk-weight of an asset.

Sources: BIS, 2005; Behn, Haselmann, and Vig, 2016

(19)

Implementing Regulation (EU) 2015/1278). For a full list of IAs and DAs, see: European Commission, n.d. Since these amendments to the CRR/CRD IV require all banks to read these supplementary documents, it puts a higher disproportionate burden on the smallest of the credit institutions. The reason is because these acts have to be interpreted and adopted. This is also the case if the documents introduce more proportionality.

This paper now turns to current the state of affairs regarding the CRR II/CRD V. After the Commission published its amended proposal in November 2016, recently both the Council and EP published its own revisions. Specifically, in May 2018 the Council’s Economic and Financial Affairs (ECOFIN) Committee published its stance (Economic and Financial Affairs Committee, 2018; Tomasik, 2018). Only a month later in June 2018, the EP’s Economic and Monetary Affairs Committee (ECON) published its decision on this topic (Committee on Economic and Monetary Affairs, 2018). Furthermore, a Euro-Summit statement in June 2018 further raised expectations of potentially concluding the banking reform by the end of 2018. Additionally, by the 22nd November 2018 seven trilogues between the Commission, EP, and Council on the CRR II/CRD V were held (Permanent Representatives Committee, 2018). However, certain issues, such as remuneration, still remained open. The final statement of the Summit declared “the agreement in the Council on the Banking package should allow the co-legislators to adopt it before the end of the year while preserving the overall balance“ (General Secretariat of the Council, 2018). The latest development on this is that on the 4th December 2018 the European Parliament published a press release declaring that the EP and Council negotiators reached an agreement (Kolinska, 2018). Consequently, the plenary of the EP will table a vote on this matter in spring 2019.

(20)

Section II.3 –Credit risk exposures and the dynamic of RWA

Chapter 2 Section II Articles 114 up to and including Article 134 of the CRR stipulates an exhaustive list of different risk-weights for different exposures, concerning the SA to assess credit risk. Table 1 highlights a summary of different RWAs when employing the SA to assess Credit Risk (see the CRR Section II for further clarification and examples). As can be identified by the table, sovereign exposures are privileged when compared to the other types of exposures, especially up to and including credit quality step 3. One of the primary motivations for researching RWAs is, in theory, being able to ask the question: why banks would lend to SMEs, if they can just invest, nearly, without limits into government debt?

Textbox 2 (below) provides an illustrative example of the potential problem resulting for SMEs in the current regulatory environment. Specifically, it highlights why, in theory, SMEs can be significantly disadvantaged when it comes to loan supply when compared to other types of exposures on a bank balance sheet. Regulators recognising this disadvantage and in 2013 with the adoption of the CRR introduced an SME “Supporting Factor” (SF). Accordingly, recital 44 of the CRR states that “capital charges for exposures to SMEs should be reduced through the application of a supporting factor equal to 0,7619 to allow credit institutions to increase lending to SMEs” (Regulation (EU) No 575/2013). However, this SF is only applicable if the total exposure of the SME does not exceed 1,5 million euro (Article 501 (2c)). As a result, the SF clearly has the aim to offset any negative consequences that SMEs experience resulting from the introduction of the CRR/CRD IV. This highlights the importance the EU places on SMEs access to loans. Despite the SME SF the question still remains why banks would lend to SMEs. Furthermore, in the original legislative text, the

Table 1- Summary of select RWA

According to the SA

E.g. Standard & Poor assessment AAA to AA- A+ to A- BBB+ to BBB- BB+ to BB- B+ to B- Below B- Unrated (for sovereign: flat 100% or central government exposure rating, whichever is higher)

Credit Quality Step 1 2 3 4 5 6

RWA Sovereign 0% 20% 50% 100% 100% 150% 100%

RWA Corporate 20% 50% 100% 100% 150% 150% 100%

RWA Retail

75% (if 1. Natural person/s or SME. 2. Exposures have...similar characteristics so risks... are substantially reduced. 3. Sum of exposures does not exceed 1 million euro)

(21)

SME SF is not mentioned to be in place on a permanent basis. Rather, by the 2nd January 2017 should report on this in order to determine whether to extend, revoke, or make the SF permanent, see Article 501 (4). In the current draft of the CRR II/CRD V the SME SF has been made permanent.

Given the potential problems associated with the CRR/CRD IV, this section turns to developing a framework with which to assess legislation with. This is done in order for this paper to ultimately be able to assess the success of the CRR/CRD IV. Thus, in order to assess whether the CRR/CRD IV has been a successful legislation for the case of German co-operative and savings banks, the UK’s ‘better regulation principles’ (Better Regulation Council, 2010, 51) are employed. The reason being, that the UK’s principles build upon Christopher Hood’s (1986, 21-2) pre-conditions, namely: i) knowable and stable, ii) acceptable and visible, iii) consistent, iv) verifiable, and v) provide a robust category. It is noteworthy to mention that the UK’s ‘better regulation principles’ are meant to provide guidance (Lodge and Wegrich, 2012, 54). Nevertheless, the UK’s principles are: I) Proportionality: where intervention occurs when necessary, remedies should be appropriate to

Textbox 2: Sample calculation of loans to SMEs.

Banks are required to hold 8% of total capital ratio in order to guarantee minimum solvency (see Key Values Textbox). For the sake of simplicity, this example assumes that the bank is compliant with the other stipulations of Article 92 of the CRR (e.g. CET 1). Furthermore, this example calculates a large corporate loan, a retail loan with the SME SF, and a sovereign exposure rated AA+ (thus investment grade). The underlying mechanism, simply put, in order to determine how much capital the bank has to hold is to multiply the risk-weight by the capital (i.e. 20%, 75% or 0% RWA multiplied by 8% capital).

However, the supporting factor for SMEs reduces the risk-weight for qualifying SMEs. This supporting factor is 0,7619 (found by dividing 8%/10,5% (i.e. the old/new higher capital requirements)).

Given a 100.000 euro loan/exposure in all cases:

The loan to large company may require only around 1,6% of the loan amount (20% risk weight multiplied by 8% capital), thus 1.600 euro.

A retail loan to an SME would require the bank to hold 4,5714% of the loan amount (75% multiplied by the SME SF of 0,7619 multiplied by 8% capital), thus 4.571,40 euro.

The sovereign exposure rated AA+ requires 0 capital to offset, because of 0% risk weight multiplied by 8% capital.

Therefore, this brief example highlights that despite the supporting factor, banks have to hold significantly more capital to offset the risk SME loans exhibit. This higher capital could have been invested elsewhere by the bank and yielded higher returns.

(22)

the risk posed, and costs identified are minimised. II) Accountability: justifiable decisions, and subject to public scrutiny. III) Consistency: rules and standards must be joined up and implemented fairly. IV) Transparency: regulators should be open, and keep regulations simple and user-friendly. V) Targeted: regulation should be focused on the problem, and minimise side effects.

This section highlighted the genesis of Basel and how it relates to the Banking Union and the CRR/CRD IV. Furthermore, this section highlighted the dynamic of RWA and the potentially limiting factor it has on SME lending. Thus, the subsequent section is a targeted literature review on RWAs, availability of loans to SMEs, and the possible disproportionality of the CRR/CRD IV. Furthermore, an initial analysis is attempted in order to answer the latter two problems.

(23)

Section III –The state of the literature

"There are large levels of diversity and this makes Europe so strong. It is clear that this diversity has to be maintained.”

- Johannes Rehulka, European Association for co-operative banks representative and

Executive Director of Fachverband der Raiffeisenbanken, 8th November 2017 (EACB, 2017).

This section aims to evaluate the CRR’s/CRD IV’s impact on co-operative and savings banks ability to lend to SMEs. Firstly, Germany’s decentralised banking system is laid out. Subsequently, the reasons why the Bundesbank sees no threat to SMEs financing are explained. Consequently, this section explores SMEs demand, RWAs and other difficulties smaller credit institutions face regarding the CRR/CRD IV. Lastly, this section attempts an initial analysis addressing the research question. While reading the following sections, it’s important to bear in mind that although this paper focuses on the CRR/CRD IV, parts of the literature use Basel III and the CRR/CRD IV rather interchangeably.

Section III.1 – Germany’s banking system

Germany’s banking system, comprising nearly 1,823 banks in total in 2017 (Bundesbank, n.d.) can be divided into three distinct pillars: private commercial banks, public-sector banks, and co-operative banks. The ownership of the banks as well as its legal form can distinguish these different pillars. The first pillar, private commercial banks, is the largest accounting for 40% of the total assets held by German banks (EBF, 2018). This banking pillar contains sizable banks, such as: Deutsche Bank, or Commerzbank. Private commercial banks offer all standard-banking services (CBG, 2015). The second pillar, public-sector banks, comprises Sparkassen (savings banks), and Landesbanken. Landesbanken act as ‘central banks’ for the Sparkassen (Choulet, 2016). Savings banks differ from private commercial banks in the sense that they do not compete with each other – the regional principle – and have a regional focus. Furthermore, savings banks “emphasize maximizing the welfare of their members or stakeholders rather than making profit” (Cermak, 2017). Thus, they give back to the locality that owns them through inter alia loans. The final, third, pillar of the German banking sector, comprises banks belonging to the Volks- and Raiffeisenbanken (co-operative banks). This pillar has its origin in the 19th century (ibid.). The idea behind co-operative banks is that customers run the bank. Furthermore, co-operative banks also tend to have a regional focus (EBF, 2018). Given that the specifically savings banks and co-operative banks are the primary financers of SMEs in Germany these banks are the essence of the analyses (DSGV, 2018).

(24)

Section III.2 – Outlook of SME financing

Ab initio it is important to outline the reason why the Bundesbank sees no threat to

availability of loans to SMEs. The Bundesbank reiterated from the very start of the announcement of Basel III that savings banks and co-operative savings banks “meet the Basel III requirements on average well” (Lautenschläger, 2012). In particular, “that is what at least all our investigations suggest” (ibid.).

This can be complimented by the Bundesbank’s own Financial Stability Report of, for example, 2010. In the Financial Stability Report, the Bundesbank ran different scenarios in order to demonstrate that Basel III does not negatively impact SME financing. In its simulation, the Bundesbank found that there was an increase in interest rates of around 50 basis points, resulting from the higher capital requirements (Bundesbank, 2010). However, the Bundesbank underscores that the aggregate capital requirements vary significantly for each banking group, where savings and co-operative banks need comparatively less capital than other groups. The Bundesbank notes that this increases costs to businesses slightly. Nevertheless, despite the increase of basis points of interest rates, the simulation shows a maximum negative deviation of GDP from the baseline of 0.1% to around 0.4% in the period from 2011 to 2016. The Bundesbank also notes that costs may be over- or under-estimated. This is because, on the one hand it is more expensive to raise additional capital, especially for less capitalized banks, which can adversely affect the real economy due to higher interest margins. On the other hand there is the potential of risk premiums on stocks and bonds of banks decreasing in the course of the greater capitalization in the future. Slovik and Cournède (2011), two OECD researchers, study complements certain of the Bundesbank’s findings such as the increases in base points of around 50 points and the impact this will have on GDP. The OECD researchers find a decrease of 0.08-0.23% over 5 years for the Euro Area.

However, what is omitted from both analyses is the implication of an increase of 50 base points on demand for loans. A Research Paper by BBVA Research found, for the specific case of Spain, that SMEs are highly price sensitive when it comes to prices. The research paper finds a decrease of 8% for every increase of 100 basis points (Izquierdo, Muñoz, Rubio, and Ulloa, 2017). Therefore, if applied to the Bundesbank’s or OECD’s analyses, the increase in base points decreases demand for loans from SMEs due to the price elasticity and can possibly adversely affects the real economy.

(25)

During the simulation the Bundesbank made two assumptions that can be challenged. First of all, it assumes a low need to adjust to the new rules, specifically for lending to SMEs. Although there is an SME SF in place, a significant part of the relief is conditional on the SME loan receiving a good rating by the ECAI (Meier, 2014). This is because, a worse rating results in more capital having to be set aside by a bank to offset the loan with, see Textbox 2. Additionally, this is also the result of banks compliance with the CRR’s/CRD IV’s higher capital requirements and tighter capital definitions. The second assumption that the Bundesbank makes is that the financial situation of SMEs has not deteriorated significantly during the crisis compared to that of large companies. The reason why the Bundesbank assumes this is because many small businesses are less involved in the global economy than most large companies. Although this is largely true, as German SMEs on average did better than, for example, US SMEs (Weltman, 2018), German SMEs still got setback by the financial crisis (Hordorogel, 2011).

(26)

Section III.3 – SME demand, Risk-Weighted Assets, and the CRR/CRD IV

The findings of the Bundesbank and OECD are not undisputed, as other studies suggest more severe outcomes. Specifically the Institute of International Finance (IIF) study suggests that the impact could total a 3 per cent reduction of GDP over the next five years for Europe (IIF, 2011, 10). Additionally, when examining the credit to the non-financial sector the IIF found a negative impact of credit availability in the Euro Area. This suggests that SMEs and other higher risk rated businesses find it harder to borrow. The KfW, Kreditanstalt für Wiederaufbau (German Development Bank), complements this (Zimmermann, 2018). The KfW’s survey finds that there are large differences in access to credit between small and large companies, exhibited by figure 3. Figure 3 illustrates that the smallest SMEs experience the highest difficulty of accessing loans. This is because loans up to 1 million euro can be employed as a proxy for SME loans. Loans up to 1 million euro is employed as a proxy for SME loans, as otherwise, these businesses would not qualify for the SME SF. This proxy is also employed by inter alia the EBA, making it accepted in the literature (see: EBA, 2016).

Figure 3: Loan size and enterprise perceived difficult to access loans in Germany. Source adapted from Zimmermann (2018).

However, although some SMEs experience difficulties, banks experience problems of their own. Paul and Lange (2016) point to an alternative explanation how banks try to comply with the Basel III. Banks instead of trying to raise capital in order to finance new loans reduce their share of loans with comparatively high RWA. The Bundesbank’s Basel III monitoring

37,2 48,9 60,5 69,2 76,8 54,3 38,4 38 34 27,4 21,4 33,2 24,4 13,1 5,5 3,4 1,8 12,5 Up to 1 million euro (640) > 1 million and up to 2.5 million euro (313) > 2.5 million until 10 million euro (403) > 10 million up to 50 million euro (318) > 50 million euro (224) Average (2044)

(27)

exercise complements this development. This is because the Bundesbank states that the CET 1 capital has increased from 8.4% to 13.5% for Group-2 banks between 2011 and 2015, but the increase cannot solely be attributed to an increase in CET 1 capital. The increase has been achieved by reducing the overall RWAs by about 17.8% (Bundesbank, 2016). Thus, this suggests a more selective and lower-risk action on the part of the credit institutions. Consequently, suggesting fewer loans to SMEs, as SMEs have a higher RWA (ibid.). This shift in modus operandi is also found in the IIF report. The IIF report states that banks are encouraged “to hold more lower-yielding claims on public sector entities and thus biasing banks against lending to private sector borrowers, especially at the more risky end of the spectrum” (IIF, 2011, 38). This was underscored in Textbox 2. This change in modus

operandi can occur in the form of closing braches. This development is reflective of the

current situation in Germany, as the number of overall bank branches has decreased year-on-year from 40,276 in 2010 to 31,949 in 2017 (Bundesbank, n.d.).

However, this contrasts to a study by Hackethal and Inderst (2015). Hackethal and Inderst (2015) find that many of the banks surveyed consider a move towards lower-volume business, as these loans are easier off-set with the SME SF, despite the need of higher capital. This is due to the large exposure requirement. The large exposure requirement stipulates that a loan, which reaches 10% of the institution’s eligible own funds, should be treated as a large loan and that this loan may not exceed 25% of eligible own capital. This will naturally have a particularly stronger impact on smaller institutions. This is because smaller credit institutions have, by definition, lower absolute amount of own funds.

A 2015 call for evidence performed by the Commission suggests smaller credit institutions experience difficulty with the higher administrative costs associated with increased disclosure requirements (EC, 2015, 12). It’s important to note that this does not reflect the position of the Commission, but those of the stakeholders. Hackethal and Inderst (2015) underscore this finding, as data collection and maintenance leads to higher costs. This suggests a disproportionate burden on smaller banks. Adding to the conflicting evidence is the IIF (2011) identifying that the higher costs incurred by banks are passed on to borrowers through higher lending rates. However, German banks might be constrained to do so. This could be due to higher rates being turned down by price sensitive SMEs. Alternatively, savings and co-operative banks have their own incentive not to offer higher rates. This is because income from interest rates accounts for around 75% of all their earnings (Sander, 2015).

(28)

Section III.4 – Initial Analysis

The literature review on SMEs being constrained to access finance was contradictory and inconclusive. This was also the case for the possible disproportionate impact the CRR/CRD IV has on smaller credit institutions. Thus, this paper attempts to answer these questions quantitatively. Accordingly, this is done in two ways by first of all exploring whether banks incentivise lending, through the price mechanism of the end-consumer. This means passing on the historically low interest rates, which are in place since early 2016, to the end-consumers in order to attract more customers. Furthermore, other indicators are employed assisting this mechanism making up interest rates charged to the end-consumer (e.g. Euribor 3 months). Additionally, other databases such as the ECB’s Statistical Data Warehouse as well as the ECB’s Survey on the Access to Finance of Enterprises (SAFE) are utilised in order to address the questions of loan constraint to SMEs and the disproportionality of the CRR/CRD IV.

Figure 4: Interest Rate charged to non-financial corporations in Germany. Source: ECB Statistical Data Warehouse, n.d.; own calculations

Figure 4 clearly demonstrates that interest rates charged on small and large loans are falling over time. The line “up to and including EUR 1 million” is a proxy for SME loans. Employing this as a proxy has its limitation, such as possibly being imprecise. The lack of a better proxy, and the data itself not being available, in addition to inter alia the EBA employing this as well as a proxy make this method a fair and accepted one (EBA, 2016). Curiously the interest rate does not fall below 2% in Germany as well as the whole of the EU (see: EBA, 2016). The interest rate not falling underneath 2% could be due to the inflation target set by the EU (European Central Bank, n.d.). This suggests that banks are unwilling to,

(29)

in real terms, ‘lose’ money by lending to SMEs. This is particularly acute for co-operative and savings banks. The reason for this is that these banks primarily earn money through interest rate returns (Lang, Signore, and Gvetadze, 2016, 37; International Monetary Fund, 2016). Therefore losing money in ‘real’ terms could drive these smaller credit institutions out of the market. Another interesting development is that the interest rate spread, the difference between the large business proxy and the SME proxy, is increasing over time. This might suggest that banks are trying to disincentives SMEs from taking up a loan by charging higher interest rates. This possibly prima facie suggests that the price mechanism is employed for the largest of loans by making these cheaper. This would also be reflective of the lower risk-weight of the larger corporations. Nevertheless, this figure by itself does not offer any indication on loan availability to SMEs, or disproportionality of the CRR/CRD IV for smaller credit institutions.

Figure 5 illustrates that access to finance remains over time the most pressing concern for, on average, 8% of the SMEs surveyed. In Section I, Dario Scannapieco made a reference to an abundance of liquidity. Figure 5 would suggest that this liquidity reaches SMEs, since less SMEs highlight a problem with access to finance. However, this data contrasts to the KfW’s. The KfW data attributes a higher concern for SMEs access to finance, namely 12,5% (Zimmermann, 2018). In addition, the KfW exhibits that the percentage of businesses responding that access to finance is the most pressing concern seems to increase as the business size decreases. Figure 5 therefore compliments figure 4 insofar as it suggests that as interest rates charged to SMEs are decreasing so does SMEs most pressing concern being access to finance. However, figures 4 and 5 do not paint a very convincing picture.

Textbox 3: Exploration of a potential reason why SMEs are charged comparatively higher interest rates

Banks require liquidity to lend money to inter alia SMEs. This liquidity can be acquired from the central bank or other banks through the deposit or marginal lending rate. This liquidity is assigned a rather low risk weight, e.g. 0%, due to Article 114(4) of the CRR.

However, since SMEs have a higher RWA than the bank that issued the borrowed money, more capital has to be set aside from the borrowing bank in order to offset the loan. In order to recuperate the higher capital being set aside to offset the loan and to repay the borrowed money, a higher interest rate is charged to the lender, such as the SME.

In other words, in order for the bank to make up for the risk-weight discrepancy for lending to a SME, and the higher capital it has to set aside, it may consider charging higher costs to the borrower manifested in the form of higher interest rates in order to increase returns on the investment. This sample exploration does not take into account other factors determining the interest rate or the awarding of the risk-weight such as maturity, credit quality etc.

(30)

Figure 5: German SMEs most pressing problem. Source: European Central Bank, 2018

Despite access to finance being the most pressing concern for a relatively low amount of SMEs, figure 6 exhibits that access to finance overall remains important. The SAFE survey measures this through a 1-10 scale. The numbers in brackets at the bottom of figure 6 represent the importance of access to finance where 1 is the lowest and 10 the highest score that can be answered. Consequently, these values were grouped into categories of 1-3 for low importance, 4-6 for medium importance, and 7-10 for high importance. Thus, specifically, access to finance is of high importance for nearly a third of SMEs, whereas for another third of relative importance. This suggests that if availability of loans to SMEs were hampered, this would result in adverse effects for SME growth and German real economy growth. Therefore, figure 6 does paint a more serious picture than figure 4 and 5, due to the possible significant adverse effects on the real economy.

(31)

Figure 6: German SMEs perceived importance to finance. Source: European Central Bank, 2018

Furthermore, appendix figure 2 illustrates that other financing sources, outside of bank loans and bank overdrafts, are not a significant source of finance for SMEs in Europe. Consequently, this underscores the relevance of the research question. Nevertheless, this restricts the funding options for SMEs, due to the dependency on bank loans.

Figure 7: Perceived external funding gap in different Euro Area countries. Source:European Central Bank, 2018

However, figure 7 paints a different picture to the previous figures. Figure 7 highlights the perceived percentage change in the financing gap in Germany for SMEs is steadily

(32)

decreasing. This might suggest prima facie that a part of the abundance of liquidity is flowing to SMEs. Despite figure 7 highlighting that the percentage change in the external funding gap for SMEs, specifically in Germany, has been decreasing there still remains a problem. Concretely SMEs are still experiencing an external funding gap. This suggests, at the very least, that not all of the “abundance of liquidity” is arriving to SMEs. This can be underscored by figure 3 where the smaller the size of the loan, and assumingly the size of the business, the harder it becomes to access the loan. Nevertheless, figure 7 complements figure 5 that the most pressing concern is clearly not access to finance, thus the negative percentage change in the external financing gap perceived by SMEs in Germany. Therefore, figures 3, 4, 5, 6, and 7 cumulatively paint a rather mixed picture. This is because access to finance remains important, although not the most pressing problem. Additionally, interest rates are falling for all loan sizes, and the external funding gap in Germany is decreasing, albeit it still existing for certain SMEs.

Figure 8: Possible determinants of bank lending behaviour. Source: ECB Statistical Data Warehouse, n.d.; own calculations

Figure 8 demonstrates that despite significantly decreased costs (Euribor 3 months) of interbank borrowing, these rates are not passed on to the SME proxy. Rather, this suggests that banks are passing their higher costs from inter alia the CRR/CRD IV onto the end-consumer, specifically on the riskier end of the spectrum, as highlighted by figure 4. Furthermore, German government bond yield decrease over time, highlighting increased demand for government debt. Accordingly, this suggests that banks are investing heavily into government debt, as government bond returns decrease as demand increases. Furthermore, banks while investing into government debt are not decreasing interest rates charged on loans at the same rate as the decrease in interbank lending rate, further suggesting that the costs are

(33)

passed onto the end-consumer. Therefore, although this could possibly suggest that SMEs are discouraged from taking on loans, this is development is nuanced by appendix figure 3. Appendix figure 3 exhibits the new loans issued to the SME proxy is starting to recover to pre-crisis levels, suggesting that the “abundance of liquidity” is reaching SMEs.

Therefore, what do these graphs and results of the SAFE survey suggest? A very mixed picture. Although interbank lending is at the lowest point, interest rate charged to the end-consumers, specifically SMEs are rather unreflective of this. Thus, banks might be passing on the higher costs incurred through the CRR/CRD IV to the end-consumers by disproportionately charging them more than the decreasing interbank lending (Euribor 3 months). Furthermore, the interest rates on loans up to and including 1 million euro, the proxy for SME loans, highlights that banks in Germany are not willing to go beneath the 2% inflation target thereby ‘losing’ real money, despite the inflation line jittering significantly. The relationship between interest rates and inflation is significant for two reasons. First of all, smaller credit institutions primarily earn money from interest rate returns. Therefore, if inflation is higher than the interest rate return, the bank is losing money in real terms, i.e. the money the banks earn is worth less than the money from the returns. This effect is potentially exacerbated since retail-focused banks have a higher average risk-weight exposure, meaning these institutions have to hold higher capital reserves (de Groen, 2016, 14). This capital could have been invested elsewhere more profitably. Secondly, the relationship between interest rates and inflation is that in theory they are inversely correlated. This is because when interest rates are low there is more money in circulation, which means that inflation is ‘high’ and vice-versa. However, as can be identified by figure 8, this theory does not translate into practice well.

Nevertheless, are these developments reflective of lower lending to SMEs? Appendix figure 3 does not suggest this, as new loans to the SME proxy are consistent and complemented by figure 5 and 8. A possible explanation as to why German SMEs do not see a pressing problem in accessing finance might be due to the complementary nature of the SME SF and the regional focus of co-operative and savings banks. Furthermore, the SMEs might also benefit from recital 44 of the CRR, which states, with regards to the SME SF, that: “…credit institutions should effectively use the capital relief produced through the application of the supporting factor for the exclusive purpose of providing an adequate flow of credit to SMEs” (Regulation (EU) No 575/2013). Therefore, this initial analysis of the available data does not

(34)

provide any conclusive findings to the question of the availability of loans to SMEs and whether the CRR/CRD IV disproportionately affects smaller credit institutions.

This section highlighted the disagreement within the literature regarding the possibly constrained access to finance for SMEs. Additionally, the literature was inconclusive regarding the disproportionality of the CRR/CRD IV for small credit institutions. Nevertheless, there are overlaps. With respect to the disproportionality of the CRR/CRD IV these overlaps are that the reporting requirements increases costs, the size of the institution matters, there is increased pressure to merge for smaller credit institutions, and there is significantly more red tape. The overlap in the literature regarding the potential problem SMEs experience when trying to access finance are the possible adverse effects on the real economy and the possibility of SMEs being loan constrained. Additionally, the initial analysis’ evidence was contradictory as certain figures suggested a problem whereas others did not. Thus, the initial analysis is inconclusive. Given this information, whether the CRR/CRD IV does disproportionately affect smaller credit institutions can only be hypothesised at this point. Overall the answer to the questions regarding availability of loans to SMEs and disproportionality of the CRR/CRD IV to smaller credit institutions cannot be adequately determined. Therefore, due to the broad literature available and the inconclusive findings of a preliminary scan of the state of the literature, the following section sets up a systematic review. This is done in order to evaluate the possible disproportionality of the CRR/CRD IV for smaller credit institutions and the potential problem with access to finance for SMEs.

(35)

Section IV: Methodology

This section explains why a systematic review is performed in the subsequent section, what variables are included and why, how they are operationalized, the motivation for this design, and the execution.

Section IV.1 – Design and motivation

Although systematic reviews are usually performed in clinical trials (see: Buchwald et al., 2004; Khan et al., 2006; Polanczyk, de Lima, Horta, Biederman, and Rohde, 2007), there are studies in the banking, finance, or risk sector that attempt this sort of approach (e.g. Hanafizadeh, Keating, and Khedmatgozar, 2014; Falkner and Hiebel, 2015; de Carvalho Ferrei, Sobreiro, Kimura, and de Moraes Barboza, 2016). Systematic reviews allow for a presentation of the field of research where there is a plethora of information, studies, or significant disagreement. This disagreement aspect is crucial, as even the most rigorous studies rarely remain unchallenged or not contradicted by subsequent studies throughout time, even in highly cited medical clinical trials (Ioannidis, 2005).

Thus, the motivation for performing a systematic is threefold. First and foremost the reason why a systematic review is applicable in this case is due to the plethora of information, studies, and contradiction within this field of research, at all levels. A systematic review contributes more to this field than simply another quantitative study exploring this relationship, which would get buried by more comprehensive studies. Second of all, due to the specific area of research within the CRR, namely RWA, and the focus on smaller credit institutions the aggregate data needed to perform either small-n or large-n is confidential. Lastly, although theoretically it is possible to extract the risk-weighting of assets from smaller credit institutions, it would need a whole team of researchers, or significant time, in order to generate a representative sample of the banks under consideration. Therefore, resource constraints make up the third reason for motivation in performing a systematic review. However, performing this exercise alone, as well as having a limited amount of time of 2 months to do so, can narrow the extent of the literature included. Additionally, this has the chance to exclude vital studies of the literature, which again, has the potential to limit the validity of the study. Nevertheless, the readily available plethora of studies available, a systematic review presents an opportunity to step back and review a part, given the size of the literature on the CRR/CRD IV, of the literature and to contribute in a meaningful way to the growing field of research.

Capital Requirements Regulation/Capital Requirements Directive IV: A critical review. How effective are the regulations, with regards to risk-weighted assets for SMEs in German co-operative and savings banks?