Comparative Economic Studies, 2015, (31–54)
© 2015 ACES. All rights reserved. 0888-7233/15

www.palgrave-journals.com/ces/

Survey Article

The Effects of Bank Regulation on the
Relationship Between Capital and Risk
ALESSANDRA TANDA
Department of Economics, Management and Quantitative Methods, Università degli
Studi di Milano, Via Conservatorio, 7, Milan 20122, Italy.
E-mail: alessandra.tanda@unimi.it

Capital regulation acts as an external force in the determination of bank capital and risk levels. Changes in the regulatory framework can influence banks’ decisions.
Starting from the debate of the prudential regulation after the financial crisis, this paper reviews the main empirical contributions on the role of capital regulation in the determination of banks’ capital ratios and risk exposure to evaluate bank behavior. Capital and risk decisions seem to be effectively influenced by regulation, although results may vary according to factors such as time period, country, and the type of capital analyzed.
Comparative Economic Studies (2015) 57, 31–54. doi:10.1057/ces.2014.35; published online 22 January 2015

Keywords: bank regulation, capital, Basel, risk, literature review
JEL Classification: G2

INTRODUCTION
The latest financial crisis has highlighted how bank capital regulation is necessary for the stability of the financial system. But also, it appears that it is not sufficient to ensure that banks’ decisions, in terms of risk and capital, are consistent with the aims of regulation.
Regulation acts as an external force in the capital optimization process as banks set simultaneously the level of capital and the amount of risky assets to hold in order to comply with the minimum capital ratio. However, given the moral hazard and asymmetries of information characterizing the banking activity, banks might have perverse incentives that induce them to raise risk

A Tanda
Capital Regulation and Banks’ Behavior

32

when called to respond to stricter capital requirements, in order to keep their desired leverage.
Understanding the relationship between capital and risk decisions is therefore fundamental in banking, and the underlying mechanisms should be investigated to adjust regulation and correct any hazardous behavior. Additionally, as regulation evolves, incentives might change and banks might modify their decisions. For these reasons, it is also important to understand how bank behavior has evolved since the introduction of the Basel framework that set the capital ratio at an international level, and how banks reacted to changes in prudential regulation.
The aim of this paper is to provide a review of the main empirical research on the impact of regulation on capital and risk. The paper will also discuss how the recent evidence provided by these academic studies relates to the underlying theories highlighted within the literature to explain banks’ incentives, in relation to the aims of regulation. The paper contributes to the discussion on bank capital by surveying the most recent literature on the topic, providing an updated assessment on how our knowledge and understanding of the effects of capital regulation on banks’ behavior has evolved over the last decades, together with economic and market conditions.
The main conclusion that emerges from the studies is that regulation has an important role in capital and risk decisions. But the effectiveness of bank regulation depends on other factors, such as the economic cycle, country, and the type of capital considered. The contrasting evidence provided by the empirical investigations also suggests the need for further advancements, both empirical and theoretical.
The paper is organized as follows: the next section discusses how regulation enters the bank capital and risk decisions, surveying the main contributions on the evolution of regulation and its effects; the subsequent section discusses the empirical results on the role of capital regulation in the decisions of banks on capital and risk; the final section concludes.

THE EFFECTIVENESS OF THE BASEL REGULATORY FRAMEWORK ON BANK
CAPITAL
Bank activity is characterized by asymmetric information. Depositors cannot monitor the quality of banks’ assets and doubts on the solvency of banks might lead to panic and ‘bank runs’ (Llewellyn, 1999). If this should occur, depositors will be induced to withdraw their savings, causing a liquidity crisis for the bank that can potentially lead to the failure of the intermediary.
Moreover, doubts regarding the solvency of one bank can create worries about
Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

33

the solvency of other banks, leading to a generalized panic. Bank runs are considered as extreme events that are potentially highly disruptive. This was demonstrated during the latest financial crisis, when banks faced the threat of a bank run, not only by depositors, but also by institutional investors. In fact, following the 2007–2008 crisis, the interbank market almost dried up, suggesting that bank runs may move from the retail to the wholesale market.
To prevent bank runs and their effects, governments usually create implicit or explicit guarantees to protect depositors (for a concise review see Allen et al., 2009). Deposit insurance schemes however might produce unwanted effects and increase moral hazard because they can induce banks to take higher risks.1
Prudential authorities enforce capital regulation2 in order to limit bank riskiness in relation to the stability of the system, to ensure the soundness of banks in normal and in turbulent times and to minimize their probability of default3; regulation enters the bank capital optimization problem, setting a minimum level of capital that banks must hold4 (Kahane, 1977). Although capital regulation might induce banks to behave as desired by the authorities, the objectives of the two groups may not be completely aligned
(Estrella, 2004) and therefore capital regulation might generate distortions in banks’ behavior (Kim and Santomero, 1988; Blum, 1999; Calem and
Rob, 1999).

1
Various papers investigate the effects of deposit insurance on bank risk taking, but the specific topic lies outside the scope of this survey. The reader might refer to Allen et al. (2009) for a review.
2
Given the aim of the survey, the paper focuses on the latest studies, while it only briefly recalls the theoretical and empirical contributions published before the introduction of Basel I. The reader can refer to Berger et al. (1995) for a discussion of bank capital and to Santos (2001) for a comprehensive review on previous contributions.
3
With reference to capital ratios, Estrella et al. (2000) confirm that the risk-weighted capital ratio is a good measure of the probability of failure, but it does not significantly outperform other simpler ratios, for example, the leverage and the gross revenue ratio.
4
When analyzing decisions on capital and risk it is important to differentiate between minimum capital and optimum capital. The first responds to prudential regulatory goals, it is an objective and verifiable measure of capital, it can be compared across institutions and it is usually publicly known, together with the general procedure used to compute it. The second measure expresses the level of capital desired by the firm and considered optimal to achieve its objectives (Estrella, 1995). For more recent contributions on target capital ratios, see Flannery and Rangan (2006). The authors argue that non-financial firms have a target capital ratio and each year tend to adjust their actual capitalization to approach the desired level, according to the partial adjustment framework theory. With reference to banks, Flannery and Rangan (2008) explain the capital increase experienced by US banks during the
1990s with the market perception of risk. The findings support the existence of an optimal capital which is beyond the minimum capital imposed by the authorities. In fact, as the authors underline, market participants might value the bank on the basis of a market capital ratio, that is, based on market conditions, which might differ from minimum regulatory capital, based on book items.

Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

34

If banks have a desired level of leverage, they will adjust capital and risk accordingly; when there is an increase in capital requirements, banks might have the incentive to increase risk as well, in order to comply with the new regulation and, at the same time, keep their optimal leverage (Kohen and
Santomero, 1980). This behavior can be corrected if regulators impose measures to limit bank riskiness and increase their supervision and monitoring
(Kahane, 1977; Kohen and Santomero, 1980; Gennotte and Pyle, 1991).
Since the publication of the Basel Accord in 1988 (Basel I) by the Basel
Committee on Banking Supervision (BCBS), the minimum regulatory capital has been formulated as a capital ratio, computed as regulatory capital to riskweighted assets (RWA). The first Accord considered explicitly only credit risk, but it was later modified to also include market risks. The Basel Accord provided the first homogeneous framework on capital regulation at an international level and was a response to the growing internationalization in the industry and to the different level of capitalization of international banks (for a comprehensive discussion of capital requirements as in Basel I, see Dewatripont and Tirole, 1994 and for a review of the evolution of the
Basel framework, see Rochet, 2010). Empirical evidence shows that the riskweighted capital ratios increased after the release of the Accord (Wall and
Peterson, 1996), although it is not clear whether banks met the new capital regulation by increasing capital (both Tier 1 and Tier 2) and/or decreasing risk and therefore by shrinking their assets (Jackson, 1999).
Basel I had some weaknesses, which became clear after its implementation. For instance, the regulatory framework enabled banks to implement regulatory arbitrage when choosing which asset class to hold. In fact, the risk weights were defined according to the type of borrower or issuer of a security
(government, corporate, etc) and did not reflect the actual riskiness of the counterpart, nor the risk measures used by investors on the markets. The static approach on the risk weights induced banks to switch from private loans to government lending as this was classified as less risky by the regulators and therefore required a lower amount of regulatory capital.5
According to Haubrich and Wachtel (1993), risk-based capital regulation in
Basel I was one of the main determinants of this shift, which was especially strong for weakly capitalized small banks that had few opportunities to increase capital and therefore had to decrease risk, by modifying the composition of their portfolio. Diamond and Rajan (2000) document how
Basel capital ratios might have contributed to the creation of a credit crunch

5
The coefficients applied for the computation of the risk-weighted assets to securities issued by
OECD governments was equal to zero, while the weighting for any corporate bond was 100%.

Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

35
Table 1: Evolution of minimum capital requirements from Basel I to Basel III
Basel I Basel II

Basel III
2013 2014 2015

Minimum common equity ratio
Capital conservation buffer
Minimum common equity plus capital conservation buffer
Phase-in of deductions from CET1
Minimum Tier 1 Capital
Minimum Total Capital
Minimum Total Capital plus conservation buffer
Capital instruments that no longer qualify as non-core Tier 1 capital or Tier 2 capital

3.5%
3.5%
4%
8%
8%

4%
8%
8%

2016

2017

2018

4% 4.5% 4.5%
4.5% 4.5%
0.625% 1.25% 1.875%
4% 4.5% 5.125% 5.75% 6.375%

2019
4.5%
2.5%
7%

20% 40% 60%
80% 100% 100%
4.5% 5.5% 6%
6%
6%
6%
6%
8%
8% 8%
8%
8%
8%
8%
8%
8% 8% 8.625% 9.25% 9.875% 10.5%
Phased out over 10-year horizon beginning 2013

The minimum capital requirement will be incremented from the actual 8% to a potential 10.5%. At the end of the phase-in period, in 2019, the highest quality components of capital shall represent at least 6% of risk-weighted assets (RWA); more in detail, at least 4.5% of RWA should be held as common equity. This change represents an acknowledgement of the importance of the quality of capital and not only of the quantity of capital. A capital conservation buffer is being gradually introduced starting 2016. Should the bank not hold this buffer, some restrictions would be placed on its activity (eg dividends distribution).
Other provisions relate to the deductions from Core Equity Tier 1 (CET1) that were introduced in 2013 and will be gradually increased until 2018. Non-core Tier 1 or Tier 2 capital have to be eliminated from the regulatory capital base as they are being cancelled beginning 2013 over a 10-year period.
Source: Basel Committee on Banking Supervision, http://www.bis.org/bcbs/basel3.htm

in the United States in a period when the demand for loans was already decreasing due to the economic cycle.
Given the limits and weaknesses of capital regulation as designed in
Basel I, together with the evolution of the financial system and the increased complexity of banking activities, a second version of the Accord was published in 2004, after a long consultation period.
Basel II did not modify the definition of capital introduced in the previous
Accord and did not increase the minimum capital ratio (still at 8%, as shown in
Table 1). However, it did present a more complex structure, partly caused by the criticisms expressed by the industry on the first Accord. In the design of the new framework, regulators allowed an excessive involvement of banks in the design of the Second Accord that arguably resulted in ‘capture’ of the regulator by the banking system (Rochet, 2010). Another issue related to the implementation of rules in the various countries. As underlined by Barth et al. (2008), the implementation was not homogeneous and this limited the effectiveness of Basel II. The key innovation in Basel II related to the computation of
Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

36

risk-weighted assets, which now included credit, market, and operational risk.
Also, Basel II allowed banks to choose the approach to estimating their risk exposure: a standardized approach, where all the parameters are set by the authorities, and an Internal Rating Based (IRB) approach, where the parameters used in risk exposure computation are developed by the bank, in part (IRB foundation) or fully (IRB advanced). The banks using the IRB approach generally experienced a decrease in the amount of capital they had to hold, in comparison to those banks that adopted the standard approach
(Tarullo, 2008). This represented an enormous incentive to apply the IRB approach, especially for large banks (Hakenes and Schnabel, 2011). However, small banks were more likely to apply the standard approach due to the high costs of developing an internal model. Also the underlying methodology (value at risk) has been criticized because of the strong assumptions involved in the modeling and evaluation of risks. Kretzschmar et al. (2010) argue that the methodology used to integrate risks presented several weaknesses and the models were not able to integrate efficiently, in one measure, the overall exposure to the different types of risks. This caused banks to hold a level of capital which was not sufficient to guarantee their soundness.
Procyclicality of capital requirements represented another shortcoming of
Basel II. If the models used to compute capital requirements become more risksensitive, the level of capital needed to satisfy these requirements grows as the economic cycle deteriorates (Bongaerts and Charlier, 2009; Jokipii and Milne,
2008). For this reason, banks facing an increase in capital requirements, due to an increase in the riskiness of their portfolio, might decide to reduce their exposure, instead of increasing their capital base, as the latter might be more difficult to achieve. Raising new capital in unstable conditions could be challenging on the financial markets due to increased uncertainty and the ensuing high cost of recapitalization. The consequent need to reduce risk exposure might lead to a credit crunch in the real sector, with consumers and firms facing more difficulties in obtaining loans. The cyclicality of capital requirements and the reduction of lending made to preserve the capital ratio can be even more severe when banks use internal methods (Behn et al., 2013). To compensate for the cyclicality of capital ratios and soften the impact of economic conditions on banks’ behavior,
Spain introduced a countercyclical capital buffer in 2000 (later modified in 2005 and 2008) that succeeded in reducing the effects of economic boom and economic distress on capital ratios and lending decisions (Jiménez et al., 2013).
The latest financial crisis prompted a further revision of the framework on capital regulation,6 as well as the rethinking of the supervisory activity
6
The causes of the crisis and the subsequent events have been deeply analyzed by the literature, but lie outside the scope of this survey. For a comprehensive review see Dewatripont et al. (2010).

Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

37
7

exercised by the authorities. After a consultation period, Basel III was published by the Basel Committee in 2011 (BCBS, 2011) with the objective of strengthening the banking and financial sector, tackling some of the weaknesses of Basel II.
It introduced a stricter definition of capital and a revision of the methodologies used to compute capital ratios. Also, the new regulation introduces a series of measures to deal with liquidity risk, control banks’ growth, impose higher requirements on systemically important financial institutions (SIFI), and the creation of capital buffers to face periods of economic stress.
With reference to capital prudential regulation, the new version of the
Accord stresses the importance not only of the quantity, but also of the quality of capital. The Basel Committee itself recognized that the definition of capital was not harmonized and that transparency was very poor; ‘raising the quality, consistency and transparency of the capital base’ (BCBS, 2009, p. 8) became an explicit goal of the Basel Committee in the reform of the regulatory framework.
Basel III strengthens the quality and level of capital, by admitting only the highest quality instruments in the core Tier 1, revising the components of
Tier 1 and Tier 2 and eliminating Tier 3 from the regulatory capital within
10 years. As shown in Table 1, the capital ratio is now expressed not only by a single percentage to hold, but the components constituting the total capital ratio have to meet certain criteria. By 2019, the highest quality components of capital should represent at least 6% of RWA, of which at least 4.5% of RWA should be held as common equity.
Although the minimum capital ratio remains at 8%, a capital conservation buffer of 2.5% has been introduced to encourage banks to build-up capital buffers during normal times. Banks not holding the full conservation buffer will suffer limitations in the distribution of dividends or in payments to managers. A leverage ratio has also been added, with the objective to limit the growth and exposure of banks to risks, while the procyclicality of RWA has been tackled with a specific capital buffer (from 0% to 2.5%). Notwithstanding the importance of this amendment, as also underlined by Behn et al. (2013), the effectiveness of this measure relies on the capability of the authorities to correctly anticipate economic conditions and evaluate the likelihood of future distress. 7

At European level, the prudential institutional framework has been modified: the prudential regime constituted by the three commissions, (CEBS, CESR, and CEIOPS) has been modified to substitute these commissions with three authorities (ESA – European Supervisory Authorities), with more power and influence, respectively European Banking Authority (EBA) for the banking system,
European Securities and Market Authority (ESMA) and European (EIOPS). Besides, an authority for the macroprudential regulation has been created. This new framework has been in force since the beginning of 2011.
Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

38

The implementation of Basel III is expected to bring benefits to the banking system, such as a reduction in the rate of banks default, a better decisionmaking process at management level, as well as improvements in performance.8 Berger and Bouwman (2013) find that a high level of ex-ante capital tends to increase survivorship rates of medium and large banks during crises, with small banks showing a lower rate of default also during normal periods.
Cohen and Scatigna (2014) argue that the banks that have started the adjustment towards the new capital ratios have not pursued a massive reduction in lending, but rather have reinforced the capital base using mainly retained earnings and, to a lesser extent, by moving to less risky assets.
Nevertheless, a key disadvantage of Basel III is that the risk-weighting methodology will remain essentially unchanged and banks will still have the possibility to implement different methodologies to compute the capital ratio
(Vallascas and Hagendorff, 2013).
The new regulatory framework has tried to consider other aspects of banking activities which can contribute to a more stable and safer financial system. Among these, liquidity risk has received much attention, since it is a key risk characterizing banking activity and can be of considerable concern during crises.9 A Liquidity Coverage Ratio (LCR) and a Net Stable Funding
Ratio (NSFR) have been introduced and will be implemented from 2015 and
2018, respectively (BCBS, 2013). The LCR measures a bank’s ability to face any liquidity stress in a 30-day time period by holding a stock of high-quality liquid assets, while the NSFR aims at ensuring that banks have a sustainable maturity structure of assets and liabilities over a longer time horizon.
Furthermore, the new regulation considers the role of large and interconnected financial institutions, setting the basis for regulation of SIFIs, imposing stricter capital requirements for the global and domestic SIFIs, and taking a macroprudential perspective in relation to too-big-to-fail and toointerconnected-to-fail banks. The size of banks has in fact been growing in the last decades and this has created concerns for the authorities, facing larger banks with low capital ratios and less stable funding sources (Laeven et al., 2014).
8

An interesting point of view is provided by Admati et al. (2010) that analyze the arguments against stringent capital requirements (defined as equity capital requirements) and show how these criticisms can be overcome. Moreover, according to the authors, some of the cons of high capital requirements derive from the fallacies in the literature and in the industry that have produced a distortion in the way capital requirements are interpreted and enforced. A stronger capital regulation is desirable as it is necessary (but not sufficient) to have a healthy banking and financial system: social benefits deriving from high capital requirements offset any costs.
9
It is in fact difficult to differentiate an illiquid bank from an insolvent bank in times of financial distress and crisis. This creates moral hazard problems, because insolvent banks might have access to funds initially granted by the government or Central Bank to illiquid banks, increasing the costs of public intervention (Laeven and Valencia, 2008).
Comparative Economic Studies

A Tanda
Capital Regulation and Banks’ Behavior

39

During the most recent crisis, several episodes of bail out, re-capitalization and nationalization had to be implemented by governments in order to preserve market stability, especially in the case of cross-border institutions. While most of the measures were taken at a national level, authorities in Belgium, the
Netherlands, and Luxembourg cooperated to nationalize both Fortis and Dexia
(Gualandri et al., 2009; Petrovic and Tutsch, 2009).
The crisis also highlighted the issue of ‘shadow banking’, that is the net of informal relationships that exists among banks and other financial institutions, that is not captured by the supervisory authorities. Measures to cope with this issue have been debated at an international level (for a more detailed discussion see Brunnermeier et al., 2009).
THE IMPACT OF REGULATION: EMPIRICAL EVIDENCE
Capital and risk decisions in banking are influenced by regulation, private incentives, and market pressures. The focus of the survey is explicitly to provide an updated insight into the empirical evidence on the impact of prudential regulation on capital and risk, to present an assessment of the evolution of the mechanisms driving banks’ decisions. The survey is limited to empirical investigations; to have an updated discussion of the theoretical framework of the relationship between capital and risk, see the contribution by VanHoose (2007).
The impact of bank prudential regulation on capital and risk decisions has been extensively studied by empirical contributions, which commonly state that the degree of regulatory scrutiny depends on the level of the capital ratio, that is, on the extent of the regulatory pressure. Accordingly, banks with a large buffer above the minimum capital ratio should be less subject to regulatory pressure, given that their behavior would be influenced to a lesser extent by changes in regulation (assuming an increase in capital ratios is required). On the contrary, banks with low capital ratios will be exposed to more regulatory pressure because an increase in the required minimum capital ratio will necessarily imply a change in the level of capital or risks.
Table 2 presents a concise view of the most recent empirical studies on the topic and describes the sample used by each investigation, the main research question, the empirical methodology, the definition of capital, risk and regulation, as well as a summary of the results.10 It helps in understanding how the empirical studies are different in terms of the banks investigated, which aspects have been more deeply analyzed and how the results vary not only across studies, but also within the same study, when focusing on sub-samples or specific characteristics of banks (eg according to the level of ex-ante capital) or sub-periods.
10

The table extends the framework introduced by Matejašák et al. (2009).
Comparative Economic Studies

40

Authors

Year

Shrieves and
Dahl

1992 1,800 banks
Assets>
$100 million
1984–1986

Jacques and
Nigro

1997 2,750 FDIC
Insured
banks
Assets>
$100 million
1991–1992

Hovakimian and Kane

Sample and period 2000 123 listed banks first quarter 1985-fourth quarter 1994
Aggarwal and 2001 1,685 US
Jacques
banks
Assets>
$100 million
1991–1996

Research question Methodology

Test the
2SLS
algebraic sign of the relationship between capital and risk
Test the first
3SLS
effects of riskbased capital standards CAP

RISK

REG

Impact of
Relation
regulatory between CAP pressure on and RISK
RISK

Equity/RWA

RWA/Total assets; NPLs/
Total Loans

1 if CR