table 1: minimum total capital to risk weighted assets ...€¦  · web viewan initial minimum...

52
Basel III and the global reform of financial regulation: how should Africa respond? a bank regulator’s perspective Louis Kasekende, Justine Bagyenda and Martin Brownbridge 1

Upload: vohanh

Post on 27-Aug-2018

221 views

Category:

Documents


0 download

TRANSCRIPT

Basel III and the global reform of financial regulation: how

should Africa respond? a bank regulator’s perspective

Louis Kasekende, Justine Bagyenda and Martin Brownbridge

1

Abstract

The global financial crisis has spurred an urgent review of bank regulatory frameworks to

enhance the resilience of financial systems and institutions. It revealed the strong inter-linkages

between the financial systems of the developed countries and those in emerging and developing

markets and exposed the weaknesses of existing prudential regulations in safeguarding systemic

financial stability. As a result, reforms are necessary to rectify flaws in the current regulatory

framework. The Basel Committee on Banking Supervision (BCBS) is leading efforts to reform

the global regulatory framework. In December 2010, BCBS announced proposals that have been

dubbed Basel III, which national regulators and regional supervisory organizations are reviewing

to evaluate their suitability to conditions in their own financial systems.

Key lessons from the global financial crisis revolve around leverage, capital and liquidity. Many

banks entered the crisis with too much leverage, too little capital and inadequate liquidity

buffers. Therefore Basel III emphasizes these three issues. This paper assesses the relevance of

Basel III reforms for Africa. The focus of the Basel III reforms is strengthening capital standards.

Most African banks already hold capital in excess of global standards. However, given that

banks in Africa face much greater volatility in the value of their assets, an effective prudential

regulatory regime must also include restrictions on the risk exposures in banks’ asset portfolios.

Basel III introduces macroprudential measures for the first time in global bank regulation.

Macroprudential regulation is relevant for Africa but the proposals in Basel III are unlikely to be

adequate to mitigate systemic risks on the continent, especially because they do not address

systemic threats arising from cross border capital flows mediated through the banking system. As

a consequence, the regulatory challenges facing Africa require a wider array of instruments than

those presented in Basel III, such as regulations to restrict large loan concentrations and foreign

exchange exposure. This may call for a more intrusive regulatory regime to ensure a more robust

and resilient financial system in Africa.

2

1. Introduction

The global standards which have shaped bank regulation over the last two decades have been

found wanting, unable to prevent the most severe systemic global financial crisis since the 1930s

(Blinder, 2010). This has prompted efforts to reform bank regulation, to forestall any recurrence

of financial crisis. In December 2010, the Basel Committee on Banking Supervision (BCBS)

issued reforms to global regulatory standards, dubbed Basel III, which include the strengthening

of capital adequacy requirements, as well as the introduction of liquidity requirements and

countercyclical macroprudential measures (Basel Committee on Banking Supervision, 2010A).

Other reforms are still under discussion, such as more stringent prudential requirements for

systemically important financial institutions (SIFIs).

Banking is an international business. Finance flows across borders and in many countries

international (cross border) banks hold a prominent share of the banking market. Africa is not an

exception to this. The international character of banking provides a clear rationale for global

minimum regulatory standards, especially to avoid regulatory arbitrage, whereby banks locate in

the jurisdictions with the least onerous regulations, spurring a race to the regulatory bottom.

Hence key regulatory innovations – notably the Basel Capital Accords – have been drawn up at

the global level (by the BCBS) and most governments around the world, including those in

Africa, have incorporated them into their own national banking regulations. Similar principles

will apply to Basel III and any other proposals for the reform of bank regulation which are

eventually agreed upon at the global level. Although these reforms were developed in response

to the financial crisis in advanced economies, and African bank regulators had little influence in

shaping them,1 they will be regarded as a global minimum standard which should be eventually

adopted by all countries.

The aim of this paper is to assess the implications of Basel III for bank regulation in Africa. We

address two questions. First, are the specific proposals in Basel III relevant for Africa and are

they likely to be helpful or counterproductive? Second, are there aspects of regulatory reform 1 The BCBS’s membership comprises 27 countries but the only African country which is a member is South Africa.

3

which are important for Africa, because of the specific nature of the regulatory challenges it

faces, which are not included in Basel III? Hence are the Basel III reforms in someway

incomplete from the standpoint of Africa’s needs? Our paper aims to help fill a gap in the

literature on the latest global reforms to bank regulation; very little of which has assessed these

issues from the standpoint of developing countries.2

The paper is organised as follows. Section 2 briefly reviews the Basel III reforms and highlights

the other reforms currently under discussion. Section 3 analyses the relevance for Africa of the

proposed reforms to the quantity and quality of regulatory capital. Section 4 examines the

relevance of the liquidity requirements proposed in Basel III. Section 5 evaluates the Basel III

reforms which are intended to introduce macroprudential regulation into the regulatory toolkit,

and asks whether they are likely to be adequate to protect African financial systems against

systemic threats. Section 6 addresses issues pertaining to the regulation of international banks.

Section 7 provides a conclusion.

Before proceeding a few remarks about the nature of the regulatory model which is being

reformed are pertinent. Over the last two decades, bank regulation in the advanced economies

has focussed mainly on setting capital adequacy requirements for banks, in line with the global

minimum standards set out in the Basel I Capital Accord, which was issued in 1998, and then

Basel II, issued in 2006, which aimed to more closely align capital requirements with the risk

incurred by a bank. Bank regulations have largely avoided imposing quantitative restrictions on

the composition of banks’ assets and liabilities or their business activities; in many countries

regulatory reforms have radically liberalised restrictions on banking activities, as for example

with the enactment of the Gramm-Leach-Bliley Act in 1999 in the United States.3 Instead bank

regulators have put more emphasis on the supervisory monitoring of banks’ risk management.

This has reflected the evolution of regulation from a framework which is compliance driven (i.e.

regulators focus on checking that banks comply with regulations) to one which relies more on

2 Exceptions include Agenor and Pereira da Silva (2010), Calice (2010) and Taylor (2010).3 The Gramm-Leach-Bliley Act removed the regulatory barriers to banks providing “non bank” financial services, thereby allowing a single financial institution to undertake commercial banking, investment banking, securities business and insurance business.

4

the judgement of regulators, especially in regard to the risk management of a bank and the

quality of its directors and senior managers.4 The focus of Basel III is mainly (although not

entirely) on the strengthening of capital adequacy requirements as the primary quantitative tool

for ensuring both microprudential and macroprudential regulation.5

Most African countries have implemented the Basel I Capital Accord.6 Bank regulation in many

African countries has differed from the model practised in most advanced economies in two

respects. First, stricter minimum capital adequacy requirements have been imposed. More than a

third of African countries impose a higher total capital to risk weighted assets requirement than

the 8 percent minimum in the Basel Capital Accords (table 1). This is not incongruous with the

Basel Capital Accords, because the accords are intended to be global minimum standards and

national regulators have the discretion to impose higher standards if warranted. Secondly, a

broader range of restrictions on the composition of banks’ assets and liabilities and their business

activities have been retained in Africa. For example, many Africa countries impose restrictions

on banks’ large loan concentrations, foreign exchange exposures and business activities which

fall outside of traditional commercial banking. They also impose minimum liquid asset

requirements and more stringent loan loss provisioning requirements than is the case in advanced

economies. As a consequence, one can argue that African bank regulation is more robust than

that which prevails in the advanced economies in that the former relies less exclusively on just

one regulatory instrument, the capital adequacy requirement, which in the advanced economies

proved very vulnerable to “gaming” by banks to lower the amount of capital they had to hold.

Table 1: Minimum Total Capital to Risk Weighted Assets Requirement (percent)

Country CAR Country CAR

Algeria 8 Kenya 8Angola 10 Lesotho 8Benin 8 Malawi 10

4 See Financial Services Authority (2009; pp86-88) for an account of how this approach was applied by the Financial Services Authority in the United Kingdom.5 Microprudential regulation refers to regulation which is designed to ensure sound management of individual financial institutions. Macroprudential regulation aims to safeguard the stability of the financial system. 6 Not all African countries have implemented the capital charge for market risk under Basel I. South Africa, Mauritius and Morocco have implemented Basel II.

5

Botswana 15 Mali 8Burkina Faso 8 Mauritius 10

Burundi 8 Morocco 8Cameroon 8 Mozambique 8

Central African Rep 8 Niger 8Chad 8 Nigeria 10

Congo Republic 8 Senegal 8Cote d’Ivoire 8 Sierra Leone 15

Egypt 10 South Africa 10Equatorial Guinea 8 Tanzania 12

Gabon 8 Togo 8Ghana 10 Uganda 12

Guinea Bissau 8 Zimbabwe 10Source: 2006 World Bank Survey of Bank Regulation and Supervision Around

the World and central bank websites

The main contention of this paper is that Basel III, while useful in some respects for African

countries, does not provide an adequate framework for bank regulation at either the

microprudential or macroprudential levels, largely because it places too much emphasis on

capital adequacy. We argue that a more holistic approach is necessary in Africa. Capital

adequacy requirements are useful regulatory tools, but they need to be supplemented by other

tools to provide a robust regulatory framework.

2. Basel III and other proposals for reform of bank regulation

The Basel III reforms to global regulatory standards were agreed by the G-20 in November 2010

and were then issued by the BCBS in December 2010. They include measures which aim to

strengthen microprudential regulation and to introduce macroprudential tools. The emphasis of

the reforms is the strengthening of the capital adequacy requirements, in terms of both the

quantity of capital which must be held by banks to absorb losses and its quality (the capacity of

capital to actually absorb losses). They also include a leverage ratio and two separate liquidity

requirements. The reforms are motivated by an analysis of the financial crisis which views

excessive leverage, inadequate and low quality capital and insufficient liquidity buffers as the

6

main weaknesses in the banking sector. The revisions and new measures will be phased in over

the period 2013-2019. The appendix shows the schedule for their introduction. As with the

existing Basel capital requirements, Basel III is intended to be a global minimum standard which

allows national authorities to impose stricter standards if they see fit to do so. The details and

rationale for the reforms are set out in Basel Committee on Banking Supervision (2010A). Below

is a brief summary.

Basel III will raise the minimum requirement for tier 1 capital from the current level of 4 percent

of risk weighted assets (RWA) to 6 percent. It will also raise the requirement for common equity

capital (a component of tier 1 capital) from 2 percent to 4.5 percent of RWA. In addition there

will be more stringent definitions of common equity capital and tier 1 capital; for example banks

will have to deduct intangible assets from common equity and tier 1 capital. Minimum total

capital will remain at 8 percent of RWA. Banks will also be required to hold a capital

conservation buffer, composed of common equity, of 2.5 percent of RWA. They will be allowed

to run down this buffer in periods of stress but if they do so they will face restrictions on

earnings distributions, designed to preserve capital. Basel III will, therefore, raise the minimum

requirements, inclusive of the capital conservation buffer, for common equity, tier 1 and total

capital to 7 percent, 8.5 percent and 10.5 percent of RWA respectively. In addition, banks will

face higher capital charges for the market risk in their trading books, exposures to off balance

sheet vehicles and derivatives and for the credit risk of trading counterparties. These reforms are

essentially microprudential in nature; i.e. they are intended to strengthen the financial resilience

of individual banks, although the capital conservation buffers can also contribute to

macroprudential objectives.

The reforms to the capital adequacy requirements will be supplemented by a non risk based

leverage ratio defined as tier 1 capital to total assets. An initial minimum leverage requirement of

3 percent will be set on a test basis beginning in 2013. The motivation for the leverage ratio is

both microprudential and macroprudential; it is intended to provide a safeguard against errors in

measurement and modelling risk pertaining to the risk based capital requirement and to constrain

the build up of leverage during booms, and hence the risk of a subsequent de-leveraging.

7

Two liquidity standards will be introduced. The Liquidity Coverage Ratio (LCR) will stipulate

that banks should hold sufficient liquid assets to meet all potential demands for liquidity over a

30 day period under stressed conditions. The Net Stable Funding Ratio (NSFR) will aim to

curtail liquidity mismatches over the longer term by ensuring that banks use stable sources of

funding. This is the first time that liquidity standards have been established at the global level.

Basel III introduces, for the first time in global bank regulatory standards, a specific

macroprudential measure, designed to address threats to systemic stability; the countercyclical

capital buffer. This is intended to moderate the amplitude of the credit cycle and especially to

avoid sharp contractions of credit during cyclical downswings which can be highly damaging for

the real economy. National regulators will have the discretion to impose a countercyclical capital

buffer of 2.5 percent of RWA, on top of the minimum capital adequacy requirement inclusive of

the capital conservation buffer, in periods when bank credit growth is deemed excessive. By

imposing a higher capital requirement during the upswing of the credit cycle, the countercyclical

capital buffer aims to constrain credit growth. When the credit cycle turns down, the

countercyclical buffer will be removed to reduce the pressure on the banks to shed assets in order

to remain in compliance with their capital adequacy requirements.

In addition to the measures described above, a number of other measures are under consideration

but have yet to be finalised or agreed upon. The current methodology for loan loss provisioning

has a pro-cyclical bias, because it is based on the current performing status of the loan portfolio

which varies with the business cycle. To counter this pro-cyclicality, the BCBS is working with

the International Accounting Standards Board to develop an expected loss approach to

determining loan loss provisions, which take account of expected credit losses over the entire life

of the credit portfolio.

The BCBS and the Financial Stability Board are considering macroprudential measures to

strengthen the resilience of SIFIs whose failure could have systemic consequences for the

8

financial system and which, therefore, should have a greater capacity to absorb losses than

financial institutions which are not systemically important. Such measures could include

imposing capital surcharges on SIFIs and requiring them to hold contingent capital; debt

instruments which have a contractual obligation to be converted into equity in the event of bank

distress. Work has also been started on legal and policy reforms to enhance the cross border

resolution of failed financial institutions in a manner which minimises systemic risk to financial

systems.7

3. How relevant are the Basel III capital requirements for Africa?

At the core of Basel III are reforms to the level and quality of capital which banks must hold to

absorb losses. As noted in the introduction, the global standards for bank regulation have placed

great emphasis on capital adequacy requirements as the primary regulatory instrument. At a

theoretical level, capital requirements are justified as a microprudential instrument because of the

incentives this gives to the owners of a bank to avoid taking excessive risks which would

jeopardise the value of the claims of its debt holders, such as its depositors (Dewatripont and

Tirole, 1993); see footnote 8 below.

To illustrate the issues discussed in this section, consider a highly simplified bank balance sheet,

in which the bank holds two classes of assets, a risky asset (Ar) and a safe asset (As) and two

classes of liability, deposits (D) and equity capital (E). We assume that the value of As is certain,

whereas that of Ar is uncertain, with a variance θ. The greater is θ, the higher the probability of

losses to deposits.8

Assets Liabilities

7 Vinals et al, (2010) provide more detailed analysis of these proposals. 8 The higher is θ, for a given mean value of assets, the higher are the expected gains to shareholders, because shareholders downside risk is limited by the finite value of their capital, whereas the upside risk is not. Hence owners have an incentive to raise θ (invest in riskier assets) which is greater the lower is the value of their capital. This provides the theoretical rationale for imposing a minimum capital requirement to restrain risk taking by the bank.

9

Ar DAs E

On the bank’s financial statement, E is measured as Ar + As - D; the difference between the

value of a bank’s assets and its deposits. If the value of Ar falls, E is eroded, and if the losses on

Ar are sufficiently large, E becomes negative and the deposits lose some of their value. Capital

adequacy is measured, for regulatory purposes, as E/Ar (for simplicity we assume that As has a

zero risk weighting and that Ar has a 100 percent risk weighting for the computation of capital

adequacy). The larger is the bank’s capital adequacy, the greater the losses a bank can absorb on

its portfolio of risky assets without being rendered insolvent and thus imposing losses on its

deposits. This also implies that a larger θ requires a higher capital adequacy to protect a bank

against insolvency. The capital adequacy requirement (CAR) puts a statutory floor under the

bank’s capital adequacy, with the regulator empowered to intervene in the bank if its capital

adequacy falls below the statutory minimum.

The approach taken by Basel III to strengthening the resilience of banks against insolvency is to

raise the statutory CAR, as well as to ensure that the composition of E is actually capable of

absorbing losses in the event that they occur.9 The resilience of banks to insolvency could also

have been strengthened by regulations which reduce θ or raise the share of As in the bank’s total

asset portfolio, but Basel III does not pursue this approach. Clearly, the efficacy of a given CAR

to ensure that a bank remains solvent depends on θ. If the volatility of asset values is relatively

small, a feasible CAR will protect banks against insolvency with a very high probability of

success.

To evaluate the minimum level at which the CAR should be set, the BCBS used several

approaches, including empirical analysis of the volatility of the rate of return of RWA as a

measure of the potential losses a bank might incur, estimates of losses incurred by banks during

the recent financial crisis and stress tests (Basel Committee on Banking Supervision, 2010B).

Cecchetti (2010) defends the Basel III increases in capital adequacy requirements by noting that

9 For example, by circumscribing the eligibility, for inclusion in the computation of capital, intangible assets whose value might not be realisable in the event of bank failure.

10

the losses incurred by 99 percent of globally active banks during the recent financial crisis were

less than, or equal to, 5 percent of their RWA. Hence if banks were to hold equity capital of 7

percent of RWA, as required under Basel III, inclusive of the capital conservation buffer, 99

percent of globally active banks would remain solvent if there was a repeat of the losses incurred

during the 2007-09 financial crisis.

How relevant are the proposed changes to the level and composition of bank capital requirements

to Africa? African regulators will have to revise their banking legislation to incorporate the

capital conservation buffer of 2.5 percent of RWA. Some African countries already impose a tier

1 capital ratio of 6 percent or more of RWA, and so will not have to adjust this simply to comply

with Basel III; countries which currently impose a tier 1 capital requirement of less than 6

percent of RWA will need to raise this to at least 6 percent to comply with Basel III. However,

the Basel III capital requirements should only be considered as a minimum standard by African

regulators; it would be appropriate to establish higher minimum capital adequacy ratios in

national regulations because of the greater risks facing banks in Africa. African banks face

greater volatility in their asset values than do their counterparts in advanced countries, as a result

of many factors. They operate in a more volatile macroeconomic environment (Loayza et al,

2007), African economies are less diversified and hence banks’ loan portfolios are less

diversified sectorally (Narain et al, 2003), and weaknesses in the legal system impede banks’

ability to recover loans and foreclose on loan security.

As shown in table 1, several African countries already apply higher minimum capital adequacy

ratios than the existing (Basel I and II) global standards, by between 2 and 7 percentage points. It

would be appropriate for African regulators to at least maintain the differential between their

own capital requirements and the global minimum standards once Basel III comes into effect.

Hence countries which currently impose a minimum total capital requirement of 12 percent of

RWA (four percentage points higher than the Basel I and II minimum) should consider raising

this to 14 percent and to 16.5 percent inclusive of the capital conservation buffer.

11

Table 2: Bank Regulatory Capital to Risk-Weighted Assets (percent)

Source: Global Financial Stability Report (October, 2010), Statistical appendix table 22

Will higher minimum capital adequacy ratios induce banks to actually hold more capital? In

most African countries, the regulatory capital held by the banking system in aggregate exceeds

the statutory minimum requirement. A comparison of tables 1 and 2 indicates that aggregate

banking system capital exceeded the statutory minimum by between 4 and 16 percentage points

in 2009, with the average being about 9 percentage points. Hence even a significant rise in

capital adequacy requirements in Africa could be absorbed without most banks having

necessarily to increase the amount of capital they hold. However, the data in table 2 are

aggregate numbers for each national banking system, within which there are banks which hold

less capital than the average for that banking system: such banks may be forced to raise more

capital to comply with higher capital adequacy requirements. Furthermore, banks may

voluntarily hold a buffer above the statutory minimum capital requirement, to avoid facing

constraints if they were to face a shock to their capital adequacy. If this is the case, raising the

12

2005 2006 2007 2008 2009

Egypt 13.7 14.7 14.8 14.7 15.1Gabon 19.8 17.8 14.3 19.4 24.0Ghana 16.2 15.8 14.8 13.8 18.2Kenya 16.4 16.5 18.0 18.9 19.5Lesotho 22.0 19.0 14.0 12.0 15.0Morocco 11.5 12.3 10.6 11.2 11.8Mozambique 13.4 12.5 14.2 13.9 15.1Namibia 14.6 14.2 15.7 15.5 15.0Nigeria 17.8 22.6 21.0 21.9 21.5Rwanda 14.0 13.7 16.6 15.9 19.0Senegal 11.1 13.1 13.6 13.9 16.5Sierra Leone 35.5 36.0 38.7 43.5 33.4South Africa 12.7 12.3 12.8 13.0 14.1Swaziland 17.3 26.3 23.6 33.8 26.3Tunisia 12.4 11.8 11.6 11.7 12.4Uganda 18.3 17.9 19.3 20.7 20.9

statutory minimum capital adequacy requirement might induce these banks to hold more capital.

Overall, it is plausible to conclude that raising the statutory minimum capital adequacy

requirements in Africa will be some positive impact on banks’ capital adequacy. However,

because banks already have substantial excess capital, it is possible that bank capital ratios will

rise by less than the percentage point increase in the statutory minimum.

Table 3: Bank Return on Equity (percent)

Source: Global Financial Stability Report (October, 2010), Statistical appendix table 27

Banks can increase their capital adequacy ratios by raising more capital or by reducing their risk

weighted assets. If banks were to reduce their lending to the private sector, this would have

detrimental effects on economic growth, given that private sector companies in Africa rely

heavily on bank credit for their external finance. How serious a problem this will be is difficult to

assess. Banking systems are relatively profitable in Africa. The average return on bank equity

during 2005-2009 for the 16 African countries shown in table 3 was 22 percent. Hence banks

should have the resources to increase their capital from their earnings, without having to cut back

on their lending.

13

2005 2006 2007 2008 2009

Egypt 10.2 14.3 15.6 14.1 13.0Gabon 21.1 23.5 32.3 20.8 17.2Ghana 23.6 39.6 35.8 23.7 17.5Kenya 25.0 28.6 27.5 25.2 24.8Morocco 6.3 17.4 20.6 16.7 14.5Mozambique 26.9 60.8 50.7 44.7 36.6Namibia 45.6 19.9 44.9 52.1 38.4Nigeria 7.1 10.4 13.1 22.0 10.0Rwanda 9.9 27.0 15.5 18.5 5.0Senegal 15.8 14.6 15.3 13.0Sierra Leone 28.0 21.1 10.3 7.2 3.8South Africa 15.2 18.3 18.1 28.7 15.9Swaziland 19.0 21.2 14.8 22.7 14.4Uganda 28.7 27.3 27.8 27.7 12.7

Even if banks in Africa are induced by the application of higher statutory minimum capital

adequacy ratios to hold more capital, will they be safer (less vulnerable to financial distress) as a

result? Unfortunately, it is doubtful if statutory minimum capital adequacy requirements alone,

even if set at higher levels than the minimum global standards, can provide adequate protection

for banks in a very volatile environment. Banks in developing countries face potential losses of a

different order of magnitude than those which afflict banking systems in the advanced

economies, in the face of which the capital adequacy requirements in Basel III would be far too

low to prevent financial distress.10 Table 4 presents data on non performing loans in Africa. In

2009, non performing loans as a share of total loans at the level of the banking system ranged

from 2 percent to 19 percent, with an average of about 9 percent. This was three times as high as

the level in advanced economies in 2009.11

In such circumstances, the minimum capital requirements should be supplemented with

regulations which curb θ, such as restrictions on large loan concentrations, on insider lending and

on foreign exchange exposures, etc. In addition, the volatility of a bank’s overall asset portfolio

can be reduced by imposing regulations which raise the share of riskless assets in the total

portfolio. This is, in effect, what minimum liquidity requirements do, because the liquid assets

eligible to fulfil this requirement are almost always generally safe assets, such as government

securities or bank reserves held by the central bank, even though minimum liquidity

requirements are usually formulated as a tool to protect bank liquidity rather than solvency.

Table 4: Bank Nonperforming Loans to Total Loans (percent)

10 For example, Majnoni and Powell (2005) estimated the unexpected credit losses which would be incurred by banks in Argentina, Brazil and Mexico, using a simulated distribution of credit losses. To cover 99 percent of the distribution of unexpected losses, banks would need to hold capital equivalent to about 15 percent of their risk assets. 11 The unweighted average for the advanced economies in 2009 was 3.1 percent, excluding Iceland which is an outlier with a non performing loan to total loan ratio of 61 percent. Data from the Global Financial Stability Report, October 2010, Statistical annex table 24.

14

Source: Global Financial Stability Report (October, 2010), Statistical appendix table 24

As noted in the introduction, African bank regulations have retained a much broader set of asset

portfolio restrictions than is typically the case in developed countries. The main drawback of the

Basel III reforms for African countries relates to the over-emphasis placed on capital

requirements as a regulatory tool and the lack of attention paid to other regulations which could

complement capital in strengthening the resilience of banks.

4. Basel III liquidity requirements

Basel III includes two new liquidity requirements, as noted in section 2. Many African bank

regulations already include a statutory liquid asset requirement, usually defined in terms of

eligible liquid assets as a minimum percentage of deposits. The rationale for this liquid asset

requirement is that the main source of liquidity pressures facing a bank is withdrawals of

15

2005 2006 2007 2008 2009

Egypt 26.5 18.2 19.3 14.8 13.4Gabon 14.1 10.7 7.6 8.5 9.8Ghana 13.0 7.9 6.4 7.7 16.2Kenya 25.6 21.3 10.9 9.0 7.9Lesotho 3.0 3.0 3.0 4.0 4.0Morocco 15.7 10.9 7.9 6.0 5.5Mozambique 3.5 3.1 2.6 1.9 1.8Namibia 2.3 2.6 2.8 3.1 2.7Nigeria 18.1 8.8 8.4 6.3 6.6Rwanda 29.0 25.0 18.1 12.6 13.1Senegal 11.9 16.8 18.6 19.1 18.7Sierra Leone 26.8 26.8 31.7 17.9 16.5South Africa 1.5 1.1 1.4 3.9 5.9Swaziland 7.0 7.7 7.5 7.6 8.1Tunisia 20.9 19.3 17.6 15.5 13.2Uganda 2.3 2.9 4.1 2.2 4.2

customer deposits. The LCR in Basel III is more comprehensive and sophisticated in that it

requires banks to hold sufficient high quality liquid assets to cover all possible sources of

liquidity pressures over a 30 day period, under stressed conditions; including a partial withdrawal

of deposits, a loss of unsecured wholesale funding and calls on committed credit facilities. Banks

will have to hold much more liquid assets to cover wholesale liabilities maturing within 30 days

than retail deposits, because wholesale liabilities are a much less stable source of funds than

retail deposits.

African countries could usefully adopt the LCR into their own banking legislation because, as a

measure to safeguard liquidity, it is probably superior to the existing liquid asset ratios. Banks in

Africa have mostly relied on retail deposits to fund their assets, but this is likely to change in the

future. As domestic money and capital markets develop and African financial systems become

more globally integrated, banks are likely to make greater use of wholesale funds, from domestic

or foreign sources, especially if they want to fund rapid growth without having to establish large

branch networks. For banks which mobilise wholesale funds, a liquid asset requirement based on

retail deposits provides inadequate protection against liquidity pressures, whereas the LCR takes

explicit account of the need to hold sufficient liquidity to protect against a withdrawal of

wholesale funding.

5. Macroprudential Policy Measures

The global financial crisis was a systemic crisis. Most financial regulators have always regarded

preserving financial stability as a key objective of regulatory policy, but the approach taken prior

to the crisis was that systemic banking crises could best be prevented by microprudential

regulations together with deposit insurance and central bank “lender of last resort” facilities, to

protect against “bank runs”. The global financial crisis exposed the flaws in this approach.

Systemic risks to the financial system are not simply the aggregation of the risks facing each

individual bank in the banking system. The actions taken by individual financial institutions to

protect their own solvency or liquidity and to comply with microprudential regulations can have

16

negative externalities which create or exacerbate systemic risks to the financial system (Caruana,

2010A). Consequently, to guard against systemic instability it is necessary to implement specific

macroprudential policies, which are distinct from microprudential policies.

Systemic risk should be a major concern for regulators in Africa. Although most African banking

systems avoided a systemic crisis during the recent global financial crisis, historically African

economies have proved as vulnerable to systemic crises as those elsewhere in the world. Laeven

and Valencia (2008) provide a database on 124 systemic banking crises around the world,

covering the period 1970-2007, which includes 44 systemic banking crises in 37 African

countries.12 Systemic crises have enormous costs, in terms of lost output, employment and fiscal

costs (Reinhart and Rogoff, 2009). Daumont et al (2004) analyse 10 banking crises in SSA in the

1980s and 1990s which involved losses of between 3 percent and 25 percent of GDP. Hence,

macroprudential policies which are effective in mitigating systemic risks are potentially very

valuable. The question we address in this section is how relevant are the Basel III

macroprudential policy measures for Africa?

Threats to systemic stability in Africa probably arise from four main sources. The first is

contagion arising from financial shocks emanating from outside of the domestic economy; such

as volatile external capital flows or financial distress in cross border banks which have

subsidiaries in Africa. The second is the pro-cyclical characteristics of financial intermediation

which can amplify the economic cycle, through rapid credit growth during economic upturns and

credit contraction during downturns. The third entails common exposures of the financial system,

such as credit exposures. In less diversified economies, the constraints on diversification mean

that all banks face similar credit concentrations; hence a shock to the dominant sector of the

economy poses a potential systemic risk to the banking system (Narain et al, 2003). The fourth

involves the financial distress of a systemically important financial institution (SIFI). These four

sources of systemic risk may be connected. For example, external capital inflows can fuel a

credit boom and a reversal of capital flows can force a credit contraction, while many of the

SIFIs in Africa are subsidiaries of international or regional banks. 12 A systemic banking crisis in this database is defined as one in which non performing loans increase sharply and almost all of the capital of the banking system is destroyed by losses.

17

The specific macroprudential policy measure in Basel III is the countercyclical capital buffer,

which is intended to counter the time dimension of systemic risk by moderating the amplitude of

the credit cycle. Basel III gives regulators the option of imposing a countercyclical capital buffer

of up to 2.5 percent of RWA during periods of excessive credit growth. Regulators would have

to evaluate whether credit growth is excessive in relation to some macroeconomic variable, such

as GDP. By imposing the buffer during the upturn of the economic cycle, regulators would aim

to restrain excessive credit growth. By removing the buffer during the downturn of the cycle,

they would aim to ensure that banks are not forced to deleverage rapidly (by contracting credit

extensions) because of a shortage of capital.

Credit booms have been relatively common in Africa in recent years, hence measures to

moderate the credit cycle are relevant for Africa.13 However, the countercyclical capital buffer of

2.5 percent of RWA is likely to be too low to bind as a constraint to excessive credit growth in

African banking systems, and hence will be ineffective. As was noted in section 3, the aggregate

level of total regulatory capital in the banking systems of most African countries exceeds the

statutory minimum capital requirement by several percentage points of RWA. If banks in Africa

continue to hold regulatory capital at levels which are substantially higher than the statutory

minimum, temporary countercyclical capital buffers are unlikely to be effective in restraining

credit growth during lending booms.

To restrain excessive credit growth, regulators in Africa may have to employ other tools which

impose a more direct constraint on the ability of banks to expand their loan portfolios. For

example, regulators could impose direct controls on aggregate lending by each bank or on

lending to specific sectors which are regarded as posing most risk to systemic stability, such as

loans for real estate.14 Imposing regulatory caps on loan to value ratios for real estate lending 13 Iossifov and Khamis (2009) identify 13 SSA countries which experienced rapid credit growth during 2003-07, defined as annual average real growth of bank credit to the private sector of over 20 percent in this period.14 One of the reasons why loans for real estate might be regarded as posing a greater threat to systemic stability than loans to other sectors is that the former stimulates demand for real estate, which pushes up prices, given that supply is finite. In turn the higher real estate prices raise the value of collateral for lending and may stimulate further demand if investors expect to make capital gains in a rising market. Hence real estate lending can generate a self

18

may also be useful. Both direct credit curbs and loan to value ratios have been used by bank

regulators in East Asia to dampen the procyclicality of credit (Caruana, 2010B). Regulatory

requirements for forward looking provisioning could also contribute to moderating the credit

cycle. Forward looking provisions are based on the expected loss of a loan portfolio over the

entire life of the portfolio, which encompasses the upswings and downswings of the economic

cycle, hence they are less procyclical than provisions based on incurred losses.15 Some African

countries already include an element of expected loss in their provisioning requirements through

a requirement for a general provision which is determined as a fixed percentage of total

outstanding loans.

Agenor and Pereira da Silva (2010) argue that a simple leverage ratio (which is part of Basel III)

would be a useful complement to capital requirements to prevent excessive credit growth and to

dampen the procyclicality of bank lending, especially in countries where weaknesses in bank

regulation allow banks to disguise the riskiness of their asset portfolio. They also propose that

the leverage ratio applied to each bank should be adjusted upwards on the basis of that bank’s

share of total banking system assets, so that larger banks must hold more capital as a percent of

their total assets; thereby restricting risk taking by banks which are potentially systemically

important because of their size.

Table 5: Bank Capital to Assets (percent)

reinforcing spiral of rising asset prices and a credit boom. 15 Incurred losses, which are related to the performing status of the loan, tend to fall in an economic upturn because borrowers can more easily service their loans, but rise sharply during an economic downturn. Hence provisioning based on incurred losses becomes procyclical.

19

Source: Global Financial Stability Report (October, 2010), Statistical appendix table 23

Table 5 depicts a measure of bank capital to total assets in a selection of African countries. The

minimum capital to asset ratio for the banking systems in these countries is about 6 percent, and

about half of the countries have capital to asset ratios in the region of 10 percent or above. These

ratios are far above the minimum leverage ratio of 3 percent of total assets proposed in Basel III.

As such the Basel III leverage ratio is unlikely to bind as a constraint to credit growth at the level

of the banking system. African regulators will have to impose higher leverage ratios than the

Basel III minimum if this tool is to be an effective constraint to excessive asset growth.

The macroprudential policies of Basel III aim to control the growth of the banking system’s

assets. Researchers such as Shin and Shin (2010) argue that macroprudential policy should focus

instead on the non core liabilities of the banking system. The core liabilities are customer retail

deposits, which provide a relatively stable funding base for banks. Non core liabilities are

wholesale funds, such as interbank borrowing and external borrowing, which are a much more

volatile source of funding for banks. The growth of core liabilities is constrained by the income

of depositors; hence during credit booms banks must rely more on mobilising non core liabilities

20

2005 2006 2007 2008 2009

Egypt 6.1 6.3 5.5 6.2 6.4Gabon 11.1 10.2 7.0 10.7 16.2Ghana 7.2 15.0 13.6 12.8 17.0Kenya 12.1 12.4 12.6 11.4 12.7Morocco 7.7 7.4 6.9 7.3 7.6Mozambique 6.6 6.3 7.2 7.5 7.7Namibia 7.8 7.5 7.9 8.0 7.9Nigeria 12.4 14.7 16.3 18.0 18.4Rwanda 9.4 9.3 10.3 7.1 13.0Senegal 7.6 8.3 8.3 9.1 9.3Sierra Leone 10.3 17.0 16.7 18.7 18.9South Africa 7.9 7.9 8.0 5.6 6.7Swaziland 14.4 13.7 17.3 17.6 16.9Uganda 10.3 10.9 10.3 13.1 13.4

to fund their rapid asset growth. If access to non core funding is suddenly curtailed, as in a

sudden stop of external capital flows, banks are faced with liquidity shortages and will have to

contract credit extension.

This analysis is relevant for Africa. Although African banking systems still rely heavily on retail

deposits to fund their assets, several African countries are now classified as “frontier markets”

and have begun to attract institutional investment from abroad. Most of the external capital flows

from institutional investors to frontier markets are intermediated through the latter’s domestic

banking systems and some involve sophisticated financial technology, thereby strengthening the

potential for such flows to contribute to systemic risk (Nellor, 2008).

Systemic risk related to external capital flows can arise through several channels. Domestic

banks may exploit the carry trade to mobilize short term external borrowing which is then

invested in domestic assets. The volatility of short term private capital flows exposes the

domestic banks to liquidity risk if they cannot refinance their external borrowing. A sharp

depreciation of the real exchange rate will also expose the domestic banks to either exchange rate

losses, if they have currency mismatches on their balance sheets, or to credit risk if they have

extended foreign currency denominated loans to domestic borrowers whose revenues are

generated from the sale of non traded goods. Furthermore, many African banking systems are

dollarized; the average share of foreign currency deposits in total deposits in the frontier markets

of SSA is about one third. Ragan and Tokatlidis (2005) argue that dollarized banking systems

could be vulnerable to generalised bank runs on foreign currency deposits, in circumstances of

foreign exchange shortfalls.

Basel III does not address, in a comprehensive manner, the systemic risks that can arise from the

banking system’s reliance on foreign currency denominated liabilities.16 For African regulators,

16 The only reference to foreign exchange risk in Basel III is in the section on monitoring tools for the Global Liquidity Standard which states that the Liquidity Coverage Ratio should be assessed in each significant currency, in order to monitor and manage the overall level and trend of currency exposure at a bank (Basel Committee on Banking Supervision, 2010A, p10).

21

this is a major lacuna in the new global regulatory framework. What could African regulators do

to mitigate these risks?17

Some of the existing regulations in Africa designed to protect the liquidity of banks could

actually have a perverse effect on financial stability. Both statutory cash reserve ratios and liquid

asset ratios are levied on a bank’s deposit base (i.e. its core, stable liabilities) but not usually on

its non core liabilities (or not all of its non core liabilities); hence they provide an incentive for

banks to increase their non core funding. Consequently regulators should broaden the base for

the cash reserve and liquid asset requirements to include all bank liabilities, other than capital. It

would also be worth considering imposing higher cash reserve and liquid asset requirements on

non core liabilities, because these pose a greater liquidity risk than core liabilities.

From the macroprudential standpoint, the non core liabilities which pose the greatest risk are

short term foreign currency liabilities, because these are vulnerable to “sudden stops”, which in

turn exposes banks to liquidity, exchange rate and credit risks. Regulators could discourage

banks from mobilising these liabilities by imposing an unremunerated cash reserve requirement

with a minimum stay period.18 The rate at which the reserve requirement is applied need not be

constant over time to achieve macroprudential goals; instead it could be applied at higher rates

during periods when external capital inflows are very strong. Regulators should also consider

imposing restrictions on banks’ foreign currency derivatives because these could be used to

circumvent controls on foreign currency liabilities. To address the credit risk emanating from

bank lending in foreign currency to domestic residents, stricter prudential requirements, such as

higher capital charges and/or loan loss provisions could be imposed. Moreno (2011) discusses a

range of policy measures which regulators could use to mitigate the risks arising from banks’

foreign currency denominated transactions.

17 Regulators in Africa already impose some restrictions on the foreign exposures of banks, but these are microprudential in nature.18 The experience of Columbia with non-remunerated reserve requirements on capital inflows is analysed in Uribe (2005).

22

To summarise this section, the macroprudential policy measures in Basel III only target one of

the sources of systemic risk which African economies face, and the specific policy measure

proposed in Basel III, the countercyclical capital requirement, will probably be ineffective to

curb excessive credit growth, given the characteristics of banking systems in Africa. The

systemic risks emanating from external capital flows are not tackled in Basel III. Also missing

from Basel III are measures to address the cross sectional aspects of systemic risk. As noted at

the start of this section, the cross sectional aspects of systemic risk have two components which

are relevant to Africa: large common credit exposures of banking systems in economies which

are not well diversified, and the potential financial fragility of SIFIs. The latter is recognised as a

problem which must be addressed at the global level, but agreement has not yet been reached on

the precise measures to do this; several proposals are being considered, such as the imposition of

stricter prudential regulations and resolution regimes for SIFIs. Many of the SIFIs in Africa are

cross border banks, hence we return to this issue in the following section.

6. The regulation of cross border banks

Cross border banking poses a number of challenges for bank regulation, not least because many

cross border banks are systemically important. As noted in section 2, the BCBS is considering

imposing stricter regulations, such as higher capital charges, on globally important SIFIs, which

operate in multiple countries, to make them more resilient to shocks and hence less likely to fail.

These banks are supervised at the consolidated level by their home country regulator (e.g. the

regulator in the country in which they are headquartered) which will be responsible for ensuring

compliance with any higher standards imposed on SIFIs. Stronger prudential regulations

imposed on the globally consolidated SIFI should make the failure of any of its subsidiaries less

likely.19

19 The Financial Stability Board is drawing up a list of global SIFIs. According to the Financial Times of 9 th

November 2010, the list of around 20 SIFIs includes several banks which have subsidiaries in Africa, including Citigroup, Barclays, Standard Chartered, Societe Generale and BNP Paribas. There will also be a second list of banks which are systemically important in their home countries but not at a global level.

23

The reforms to the regulation of SIFIs are potentially beneficial to Africa, because large

international banks, which are likely to be deemed SIFIs at the global level, have a major market

share in many African countries; moreover in some of these countries the market share of the

global SIFI is sufficiently large for it to be regarded as systemically important in that country.

Consequently, the imposition of higher regulatory standards on global SIFIs at the consolidated

level should ensure that some of the most systemically important banks in Africa are more

resilient to failure, and hence it will strengthen the stability of African banking systems. To

provide additional safeguards, African bank regulators should also impose the stricter prudential

regulations designed for SIFIs on any financial institution operating in their own financial

markets which they judge to be systemically important in these markets, even if it is not deemed

to be a SIFI at the global level.

The resolution of failed cross border banks is extremely complex and involves bank resolution

legislation in multiple countries. The Cross-border Bank Resolution Group of the BCBS is

examining how to strengthen the resolution of cross border banks, including how national

resolution frameworks should be adapted for cross border resolutions. This is also relevant for

African countries; their priority should be the global implementation of a “fair” resolution

framework: i.e. a framework which does not favour creditors in one country of the failed bank’s

operation at the expense of creditors in another country.

For the regulation of international banks, African bank regulators require the cooperation of their

counterparts in the other jurisdictions in which these banks operate, and especially their

counterparts in the home country of the bank. Supervisory colleges for international banks have

been set up to coordinate supervision and share information between home and host regulators,

in which African supervisors are participating. Under the Basel core principles, the home country

regulator has clear responsibilities to the host country regulator to provide information about the

consolidated operation of the banking group, including the extent to which the subsidiary could

rely on the parent banking group for financial support and arrangements for crisis management.

But because the subsidiaries in small countries are unimportant in terms of the health of the

overall banking group, home country regulators do not always comply with their responsibilities

24

to the host regulators. African bank regulators should demand better cooperation from the host

regulators of international banks, in line with the latter’s responsibilities under the BCPs.

Cross border banks headquartered in African countries have expanded their market share in

Africa in the last few years and are systemically important in some countries. The regulatory

treatment of these banks should be the same as that applied to other cross border banks which are

systemically important. However, there are some gaps in the regulation and supervision of these

banks which need to be rectified. Their consolidated supervision by the home country regulator

is deficient in some cases, supervision of the non banking activities of these banks is not

adequate, outside of South Africa, and cooperation between home and host country regulators is

weak (Lukonga, 2010).

7. Conclusions

In the introduction we posed two questions. First, are the specific proposals in Basel III relevant

for Africa and are they likely to be helpful or counterproductive? Second, are there aspects of

regulatory reform which are important for Africa, because of the specific nature of the regulatory

challenges it faces, which are not included in Basel III? Hence are the Basel III reforms in

someway incomplete from the standpoint of Africa’s needs?

The answer to the first question is that the Basel III reforms can contribute to improving

prudential regulation of banks in Africa. Raising capital adequacy requirements and introducing

a capital conservation buffer would be useful in Africa as a microprudential measure. This will

probably have a limited positive impact on capital adequacy in Africa, given that banks in

aggregate already hold substantial excess capital. Nevertheless, it will establish a higher floor

under banks’ capital adequacy ratios beneath which regulators will have to instigate remedial

action. Therefore African bank regulators should raise capital adequacy requirements and

introduce the capital conservation buffer. African regulators who already impose higher capital

requirements than the global minimum in the Basel Accords, to take account of the greater risk

25

facing banks in their economies, should consider maintaining the differential between their own

capital adequacy requirements and the global minimum standards when the Basel III capital

standards are implemented.

Introducing the Basel III liquidity coverage ratio (LCR) in Africa should also be useful because

it provides a more comprehensive safeguard against liquidity risk than the existing liquidity

regulations in Africa, which are focussed primarily on the liquidity risk arising from deposits.

The LCR takes account of the liquidity risk from all components of the balance sheet, as well as

from off balance sheet items, some of which, such as wholesale funding, constitute a much

greater source of liquidity risk than deposits.

Basel III places too much emphasis on capital adequacy requirements as a regulatory tool to

ensure the resilience of banks. Capital is important, but in banking systems in which asset values

are very volatile, capital requirements cannot realistically shoulder all the burden of prudential

regulation. Instead it is necessary to regulate the asset side of the bank’s balance sheet, to control

excessive risk taking and improve the quality and accurate valuation of bank assets and the

associated provisioning for losses. From the standard of microprudential regulation in Africa, the

most serious lacunae in Basel III is the omission of global standards to regulate bank assets.

Many African countries already impose a range of regulations to curb risk in bank asset

portfolios, such as loan concentration limits and limits on foreign currency exposures. These

regulations should remain in place, but they need to be amended and updated to keep pace with

the evolving nature of banking in Africa. Bank regulation also needs to be backed up by stronger,

more intrusive bank supervision: hence strengthening supervisory capacities and supervisory

methodologies should be accorded priority by African regulators (Briault, 2009). Stronger

supervision will probably contribute more to safeguarding bank soundness in Africa than

strengthening the bank regulations.

The macroprudential policy measures in Basel III fall short of what Africa needs to prevent

systemic risks to the financial system. Basel III includes a countercyclical capital buffer of 2.5

26

percent of RWA, to be imposed at the discretion of the national regulator, and a leverage ratio of

3 percent of total assets. Neither is likely to be large enough to provide an effective constraint to

excessive asset growth during economic booms, which could generate systemic vulnerabilities.

Furthermore, Basel III includes no macroprudential measures to tackle what is likely to be a

major source of systemic risk for African financial systems in the future, the intermediation of

short term external capital flows through the banking system, which can entail exchange rate,

credit and liquidity risks. Consequently, an effective macroprudential policy toolkit must include

a much wider range of instruments than those included in Basel III, including possible curbs on

banks’ use of short term foreign exchange liabilities and derivatives.

International banks have a major presence in many African banking markets and some of these

banks are large enough to be systemically important in these markets. Consequently, the

proposals to impose more stringent prudential regulations on a group of globally systemically

important financial institutions, to reduce the likelihood of their failure, will be potentially

valuable to Africa if they lead to concrete measures being agreed at the global level.

To summarise, Basel III will provide a new set of global minimum standards for banks, primarily

pertaining to their capital adequacy. The Basel III reforms have been formulated to deal with the

deficiencies of financial regulation as they are perceived in developed countries. They are less

relevant for African economies, whose banking systems have different characteristics and face

different challenges. Nevertheless, African bank regulators will want to adopt the Basel III

reforms into their own banking regulation, for example by raising their own statutory capital

adequacy requirements, not least because they do not want to be perceived as applying weaker

regulatory standards than the global norms. Adopting the Basel III reforms will be useful, to

some extent, in Africa, but it will not be sufficient to deliver either the microprudential or the

macroprudential policy objectives. Safeguarding the resilience of banks and of banking systems

in Africa will require that bank regulators take a much broader approach to regulation than that

of Basel III, and in particular focus on three critical areas: setting regulations on the assets and

business activities of banks which complement capital adequacy requirements as a

microprudential tool; strengthening the supervision of banks to ensure that regulations are

27

enforced; and creating an effective macroprudential toolkit to address the multifaceted risks to

the systemic stability of the financial system.

28

References

Agenor, Pierre-Richard and Pereira da Silva, Luiz A. (2010), “Reforming International Standards

for Bank Capital Requirements: A Perspective from the Developing World”, International

Finance Review, Vol 11, pp191-254.

Basel Committee on Banking Supervision (2010A), “Basel III: A global regulatory framework

for more resilient banks and banking systems”, Bank for International Settlements.

Basel Committee on Banking Supervision (2010B), “Calibrating regulatory minimum capital

requirements and capital buffers: a top down approach”, Bank for International Settlements.

Blinder, Alan S. (2010), “It’s Broke, Let’s Fix It: Rethinking Financial Regulation” International

Journal of Central Banking, Vol 6, no 4, pp277-330.

Briault, Clive (2009), “Trust less, Verify More”, Crisis response note no 5, World Bank.

Calice, Pietro (2010), “A Preliminary Assessment of the Implications of Financial Regulatory

Reform for Africa”, Policy Brief No 3/2010, African Development Bank.

Caruana, J. (2010A), “Systemic risk: how to deal with it?” Bank for Interntaional Settlements.

Caruana, J. (2010B), “Macroprudential policy: working towards a new consensus” remarks at the

high-level meeting on “The Emerging Framework for Financial Regulation and Monetary

Policy”, organised by the Financial Stability Institute and the IMF Institute, Bank for

Interntaional Settlements.

29

Cecchetti, Stephen (2010), “Financial reform: a progress report”, remarks prepared for the

Westminster Economic Forum, Bank for International Settlements.

Daumont, Roland; Le Gall, Françoise and Leroux, François (2004), “Banking in Sub-Saharan

Africa: What Went Wrong?” Working Paper WP/04/55, International Monetary Fund.

Dewatripont, Mathias and Jean Tirole (1993), The Prudential Regulation of Banks, MIT Press,

Cambridge, MA.

Financial Services Authority (2009), “The Turner Review: A regulatory response to the global

banking crisis”, Financial Services Authority, London.

International Monetary Fund (2010), Global Financial Stability Report, October.

Iossifov, Plamen and May Khamis (2009), “Credit Growth in Sub-Saharan Africa-Sources, Risks

and Policy Responses”, Working Paper WP/09/180, International Monetary Fund.

Kaminsky, G. L., and C. M. Reinhart (1999). “The Twin Crises: Causes of Banking and Balance

of Payments Problems”, American Economic Review, Vol. 89, No. 3, pp473-500.

Laeven, Luc and Fabian Valencia (2008), “Systemic Banking Crises: A New Database”,

Working Paper WP/08/224, International Monetary Fund.

30

Loayza, Norman V., Romain Ranciere, Luis Serven and Jaume Ventura (2006),

“Macroeconomic Volatility and Welfare in Developing Countries: An Introduction”, The World

Bank Economic Review, Vol 21, No 3, pp343-357.

Lukonga, Inutu with Kay Chung (2010), “The Cross Border Expansion of African LCFIs):

Implications for Regional Financial Stability and Regulatory Reform”, Research Paper,

International Monetary Fund.

Majnoni, Giovanni and Andrew Powell (Reforming Bank Capital Requirements: Implications of

Basel II for Latin American Countries”, Economia, Spring, pp105-149.

Moreno, Ramon (2011), “Policymaking from a “macroprudential” perspective in emerging

market economies”, BIS Working Papers No 336, Bank for International Settlements, Basel.

Narain, A., Rabanal P. and Byskov S. (2003), “Prudential Issues in Less Diversified Economies”,

Working Paper WP/03/198, International Monetary Fund.

Nellor, David C. (2008), “The Rise of Africa’s Frontier Markets”, Finance and Development.

Vol 45, no 3.

Ragan, G. Raghuram and Ioannis Tokatlidis (2005), “Dollar Shortages and Crisis”, International

Journal of Central Banking, Vol 1, No 2, pp177-220.

Reinhart M. Carmen and Kenneth Rogoff (2009), “The Aftermath of Financial Crises” American

Economic Review Papers and Proceedings, Vol 99, No 2, pp466-72.

31

Rojas-Suarez, Liliana (2002), “Can international capital standards strengthen banks in emerging

markets”, Journal of Financial Transformation, Vol 5, pp51-63.

Shin, Hyun Song and Kwanho Shin (2010), “Procyclicality and Monetary Aggregates”, paper

presented at the 2010 Bank of Korea Research Conference.

Taylor, Michael (2010), “Basel II fails the younger cousins”, Financial World, October.

Uribe, Jose Dario (2005), “Capital controls and foreign exchange market intervention in

Columbia”, in Globalisation and Monetary Policy in Emerging Markets, BIS Papers no 23, Bank

for International Settlements, pp128-140.

Vinals, J., Fiechter, J., Pazarbasioglu, C., Kodres, L., Narain, A., and Moretti, M. (2010),

“Shaping the New Financial System”, IMF Staff Position Note, International Monetary Fund.

32

Appendix 1: Time frame for the implementation of revisions to capital adequacy requirements, the introduction of a leverage

ratio and liquidity ratios

2011 2012 2013 2014 2015 2016 2017 2018 2019

Minimum Common Equity

Capital Ratio

2.0% 2.0% 3.5% 4.0% 4.5% 4.5% 4.5% 4.5% 4.5%

Capital Conservation Buffer - - - - - 0.625% 1.25% 1.875% 2.5%

Minimum Common equity

plus capital conservation

buffer

2.0% 2.0% 3.5% 4.0% 4.5% 5.125% 5.75% 6.375% 7.0%

Minimum Tier I capital 4.0% 4.0% 4.5% 5.5% 6.0% 6.0% 6.0% 6.0% 6.0%

Minimum Total Capital 8.0% 8.0% 8.0% 8.0% 8.0% 8.0% 8.0% 8.0% 8.0%

Minimum Total capital plus

capital conservation buffer

8.0% 8.0% 8.0% 8.0% 8.0% 8.625% 9.25% 9.875% 10.5%

Capital instruments not

qualified for Tier I or Tier II

capital treatment

Phased out (2013-2023)

33

Leverage (3.0%) Supervisory

Monitoring

Parallel Run Phase (Public Disclosure as 0f

January 2015)

Effective January

2018 (migration to

Pillar 1)

Liquidity Coverage Ratio Observation

period begins

Introduce

Minimum

Standard

Net Stable Funding Ratio Observation

period begins

Introduce

Minimum

Standard

Source: Bank for International Settlements (2010A)

34