Financing Africa Through the Crisis and Beyond-2011

8/2/2019 Financing Africa Through the Crisis and Beyond-2011

1/322

Through The Crisis

and Beyond

Thorsten Beck, Samuel Munzele Maimbo,

Issa Faye, Thouraya Triki

FINANCING

AFRICA


2/322

Financing AfricaThrough the Crisis and Beyond


3/322


4/322

Financing AfricaThrough the Crisis and Beyond

Thorsten Beck

Samuel Munzele Maimbo

Issa FayeThouraya Triki

Washington, DC


5/322

2011 The International Bank or Reconstruction and Development / The World Bank

1818 H Street NW

Washington DC 20433

Telephone: 202-473-1000Internet: www.worldbank.org

All rights reserved

1 2 3 4 14 13 12 11

This volume is a product o the sta o the Arican Development Bank, the German Federal Ministry

or Economic Cooperation and Development, and the International Bank or Reconstruction and

Development / The World Bank. The ndings, interpretations, and conclusions expressed in this vol-

ume do not necessarily refect the views o the Executive Directors o the Arican Development Bank

and the World Bank or the governments they represent or o the German Federal Ministry or Eco-nomic Cooperation and Development.

The Arican Development Bank, the German Federal Ministry or Economic Cooperation and

Development, and the World Bank do not guarantee the accuracy o the data included in this work.

The boundaries, colors, denominations, and other inormation shown on any map in this work do not

imply any judgment on the part o the Arican Development Bank, the German Federal Ministry or

Economic Cooperation and Development, or the World Bank concerning the legal status o any terri-

tory or the endorsement or acceptance o such boundaries.

Rights and Permissions

The material in this publication is copyrighted. Copying and/or transmitting portions or all o this

work without permission may be a violation o applicable law. The International Bank or Reconstruc-tion and Development / The World Bank encourages dissemination o its work and will normally

grant permission to reproduce portions o the work promptly.

For permission to photocopy or reprint any part o this work, please send a request with complete

inormation to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA;

telephone: 978-750-8400; ax: 978-750-4470; Internet: www.copyright.com.

All other queries on rights and licenses, including subsidiary rights, should be addressed to

the Oce o the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; ax:

202-522-2422; e-mail: [email protected].

ISBN: 978-0-8213-8797-9

eISBN: 978-0-8213-8798-6

DOI: 10.1596/978-0-8213-8797-9

Library of Congress Cataloging-in-Publication Data

Beck, Thorsten.

Financing Arica : through the crisis and beyond / Thorsten Beck . . . [et al.].

p. cm.

Includes bibliographical reerences and index.

ISBN 978-0-8213-8797-9ISBN 978-0-8213-8798-6 (electronic)

1. Financial institutionsArica. 2. FinanceArica. 3. Monetary policyArica. I. Title.HG187.5.A2B43 2011

332.096dc23

2011034178

Cover design by Debra Naylor o Naylor Design.


6/322

v

Foreword xi

Acknowledgments xiii

Abbreviations xv

1. Financing Africa: Setting the Stage 1

Introduction 1

Financing Africa, the Book 3

An Analytical Framework 3

The Main Messages and a Caveat 7

Financial Sector Development: Why Do We Care? 9

Time or New Solutions to Old Problems 16

The Outline o the Book 23

2 Landscaping African Finance 27

Introduction 27

The Eect o the Crisis 27

Aricas Financial Systems in International Comparison 36

Explaining Financial (Under)Development in Arica 59

The Firm and Household Usage o Financial Services 68

Conclusions 74

3 Expanding Financial Systems 77

Introduction 77What Do We Know about Access to Finance? 78

Broadening Finance: A Reality Check 85

Landscaping the Providers: How to Get to the Frontier 88

Contents


7/322

vi Contents

Demand-Side Constraints 102

Technology: The New Silver Bullet? 105

Pushing Out the Frontier: The Role o Governments and Donors 111The Role o Government: Looking beyond Institution Building 113

From Agricultural Credit to Rural Finance 119

SME Finance: Continuing Challenges 127

Conclusions 137

4. Lengthening Financial Contracts 141

Introduction 141

Aricas Long-Term Financing Gap 142Optimizing the Current Possibilities or Expanding Long-Term Finance 145

Tapping International Markets 164

The Changing Global Environment: Tales o Dragons and Elephants 178

Pushing toward the Frontier and Beyond: A Long-Term Agenda

with Tricky Shortcuts 180

Conclusions 190

5. Safeguarding Financial Systems 193Introduction 193

Stability: We Have Come a Long Way 194

Bank Regulation and Supervision: New Challenges in a Changing

Environment 197

Looking beyond Banks: How to Regulate Which Segments o the

Financial System 213

Focusing on Users: Consumer Protection 218

Conclusions 223

6. All Financial Sector Policy Is Local 227

Introduction

Modernist Reorm Policies: Aricas Achilles Heel 227

The Activist Reorm Agenda Revisited: Larger, More Ecient,

and Stable Financial Markets 229

The Politics o Financial Sector Reorm 234

The Stakeholders 240

Redene the Role o Government with the Necessary Saeguards 245

One Size Does Not Fit All 247

Conclusions 256


8/322

Contents vii

References 259

Index 279

Boxes

1.1 Whats New? 4

2.1 Perormance on Selected Stock Markets during 200809 31

2.2 The Real Economy Impact o the Financial Crisis in Zambia 32

2.3 Bank Consolidation: Learning rom the Nigerian Experience 45

2.4 The Egyptian Exchange 55

2.5 Benchmarking Financial Development 66

3.1 Measuring Access to Financial Services: Recent Advances 78

3.2 Financial Capability Programs 106

3.3 Examples o Transormative Technology-Based Products 110

3.4 Malswitch 116

3.5 Financial Innovation: The Opportunities and Risks in

Expanding Outreach 117

3.6 Examples o Value Chain Finance 1223.7 Malawi: Development Finance Institutions 128

3.8 NAFINs Productive Chains 135

4.1 The Housing Finance Frontier: An Approximation or Arica 146

4.2 The Role o the Housing Bank in Mortgage Finance in Tunisia 149

4.3 International Labour Organization Best Practice Guidelines or

Board Members in Arica 160

4.4 The Experience with a Secondary Board in Egypt: Nilex 1614.5 Capital Markets in Latin America 163

4.6 Private Equity Funds: The Benets and Experience in Arica 166

4.7 Sovereign Wealth Funds 175

4.8 Diaspora Bonds 181

5.1 Competition and Stability 203

5.2 Bank Resolution 206

5.3 The Consumer Protection Framework in South Arica 220

6.1 Learning rom the East Asian Miracle 280

6.2 Financial Journalism 236

6.3 What to Do with State Financial Institutions? 248


9/322

viii Contents

Figures

1.1 Finance and Growth across Countries, 19802007 10

1.2 Finance and Poverty Alleviation across Countries 131.3 GDP across Countries, 2009 18

1.4 The Inormal Economy across Countries, 2007 18

1.5 Governance across Countries, 2008 20

2.1 Remittance Flows through the Crisis and Beyond, 200410 29

2.2 Stock Market Perormance through the Crisis, 200810 29

2.3 Arican Exports through the Crisis and Beyond, 200610 32

2.4 Bank Stability in Arica, 200510 342.5 Absolute Size o Arican Banking Systems, 2009 37

2.6 Relative Size o Arican Banking Systems, 2009 38

2.7 Intermediation Eciency Versus Financial Depth, 2009 39

2.8 Asset Composition o Banks across Regions, 2009 40

2.9 Financial Deepening across Arica, 19902009 40

2.10 Private Credit across Arica, 19902009 41

2.11 Branch Penetration across Countries, 2009 42

2.12 The Maturity Structure o Deposits and Loans across Arica,

200509 42

2.13 Sectoral Lending Share Relative to GDP Share, 200509 43

2.14 Banking Sector Concentration across Countries, 2006 44

2.15 Banks Market Power across Countries, 2006 46

2.16 Foreign Bank Shares across Regions, 19952009 47

2.17 Foreign Bank Ownership across the Continent, 19952009 48

2.18 Interest Rate Margins across Regions, 2009 49

2.19 Small Scale Comes with High Margins, 2009 52

2.20 Low Governance Comes with High Margins, 2009 532.21 Stock Market Capitalization across Countries, 2009 56

2.22 Listed Firms across Countries, 2009 56

2.23 Stock Market Liquidity across Countries, 2009 57

2.24 Bond Market Structure in Arica 58

2.25 Lie Insurance Penetration across Countries, 2008 59

2.26 Financial Development and GDP Per Capita, 2009 60

2.27 Domestic Savings across Regions, 2008 62

2.28 The Ratio o Oshore to Domestic Deposits across Regions, 2009 622.29 The Ratio o Oshore to Domestic Deposits over Time,

19952009 63

2.30 Capital Flows across Regions, 200009 64


10/322

Contents ix

2.31 Population Density across Countries, 2008 64

2.32 Rule o Law over Time in Arica, 19962009 65

2.33 Access to Bank Loans across Countries, 200610 692.34 The Bank-Based Nature o Arican Finance rom the

Perspective o Users 70

2.35 Private Equity Penetration across the World, 200910 70

2.36 Account Fees across Countries 71

2.37 Card Fees across Countries 72

2.38 Documentation Requirements across Countries 72

2.39 Remittance Costs across Countries 73

3.1 Access to Financial Services across Southern, Eastern,and Western Arica 81

3.2 Overlap among Users o Formal and Inormal Financial Services

in Kenya, 2009 84

3.3 Increasing the Use o Banking Services: A Framework 86

3.4 The Access Possibilities Frontier o Payment and Savings

Services 87

3.5 Total Borrowers and Depositors across Arican Subregions,

200009 933.6 Borrower and Depositor Growth Rates and Average Loan

and Deposit Balance, 200509 94

3.7 The Funding Structure o Arican MFIs 96

3.8 Asking Potential Users in Kenya about the Barriers to Access 103

3.9 The Size Gap in Corporate Finance and in Deposit Services 129

3.10 Business Obstacles in Arica and Elsewhere 137

4.1 The Size o Mortgage Markets in Arica 144

4.2 Bank Loan Maturities in Selected Countries, 2008 147

4.3 The Cost to Register a Mortgage or a Title Transer 150

4.4 The Protability and Loan Impairment o Arican Development

Banks, 2009 152

4.5 Sub-Saharan Arica: Private Equity Fund-Raising

and Investment, 200610 168

4.6 Private Equity Investment Activity, Sub-Saharan Arica, 200910 168

4.7 Factors Deterring Limited Partners rom Investing in Arica

and Latin America 1704.8 The Size o Sovereign Wealth Funds, December 2009 173

4.9 Infation and Infation Volatility, 19902009 175

4.10 The Eects o Guarantees on Loan Maturity 187


11/322

x Contents

5.1 Systemic Banking Crises in Arica, 19802009 195

5.2 The Capitalization o Banks: Delevering during the Crisis,

200510 197

5.3 The Median Basel Core Principle Assessment across Arica 199

Maps

1.1 Reliance on Natural Capital across the World 19

4.1 Trade Links between the BRIC Countries and Arica, 19802009 179

Tables

1.1 Income Elasticities across Dierent Segments o theFinancial System 15

2.1 Decomposition o Weighted Average Spreads, All Banks, Uganda,

200508 50

2.2 Explaining the Interest Margins and Overhead Costs in Arica 51

2.3 Interest Spreads in Uganda: Domestic Versus Foreign, 2008 54

2.4 Bond Markets across the Region 58

3.1 The Use o Formal Banking Services across 18 Arican Countries,

2009 80

3.2 Total Borrowers, Depositors, and Penetration Rates, 2008 92

3.3 The Awareness o Micronance across Arican Countries, 2009 104

3.4 The Regulatory Framework or Branchless Banking in

Selected Countries 108

3.5 Enterprise Credit Demand: Comparing Arica with

Non-Arican Developing Countries 136

4.1 Inrastructure: Comparing Arica with Non-Arican Developing

Countries 142

4.2 Pension Fund Portolios 157

4.3 Risk Premiums or Private Equity Funds Worldwide 169

6.1 Innovation Can Come rom Unexpected Quarters 240

6.2 Country Characteristics and the Primary Areas or Financial

Sector Policies 252


12/322

xi

For too long, rms and households in Arica have aced severe constraints to aord-

able nancecostly ees and commissions, insucient and inecient nancial

inrastructure, and short tenor, to name a ew. But things are changing, albeit

slowly. This book, a joint eort o the Arican Development Bank, the German

Federal Ministry o Economic Cooperation and Development, and the World

Bank, demonstrates that Arica is making progress in relaxing these constraints.New players and products, enabled by new technologies and business models, have

helped broaden access to nancial services, especially savings and payment prod-

ucts. Critically, Arican nance has been stable or quite a while now; ater a peak o

banking crises in the 1980s, there have been ew systemic banking crises since then.

Despite the recent global nancial crisis, banks in Arica are, on average, well capi-

talized and liquid.

At the same time, Arica aces persistent but not insurmountable challenges,

namely its small scale, inormality, volatility, and poor governance. Many rms and

most households are still excluded rom access to nancial services, especially long-term nance. Inrastructure nancing needs remain largely unmet. Agricultural

nance has been ignored by commercial nanciers or being too high-cost and

high-risk, aggravated by the same challenges enumerated above.

Financing Africa: Through the Crisis and Beyond is a call to arms or a new

approach to Aricas nancial sector development. First, policy makers should ocus

on increasing competition within and outside the banking sector to oster innova-

tion. This implies a more open regulatory mindset, possibly reversing the usual

timeline o legislation-regulation-innovation or new players and products. It also

implies expanding traditional inrastructure, such as credit registries and paymentsystems beyond banks. Second, the ocus should be on services rather than exist-

ing institutions and markets. Expanding provision o payment, savings and other

Foreword


13/322

xii Foreword

nancial services to the unbanked might mean looking beyond existing institu-

tions, products, and delivery channels, such as banks, traditional checking accounts,

and brick-and-mortar branches. Third, we should ocus on the demand constraints

as well as the supply ones; expand nancial literacy programs or households and

enterprises; and address nonnancial constraints, especially or small enterprises

and in rural areas.

All nancial sector policy is local. To reap the benets o globalization, regional

integration, and technology, policy makers have to recognize the politics o nan-

cial deepening and build constituencies or nancial sector reorm. While the chal-

lenges o expanding access, lengthening contracts, and saeguarding the nancial

system are similar, the ways o addressing them will depend on the circumstances

and context o each country.

With its cautiously optimistic tone, this book creates an opportunity or Aricas

policy makers, private sector, civil society, and development partners to harness the

progress o the past as a way to address the challenges o the uture and enable the

nancial sector to play its rightul role in Aricas transormation.

Shantayanan DevarajanChie Economist, Arica Region

World Bank

Mthuli Ncube

Chie EconomistArican Development Bank

Thomas AlbertDirector Arica

German Federal Ministry o Economic Cooperation and Development


14/322

xiii

The principal authors o this report are Thorsten Beck, Samuel Munzele Maimbo,

Issa Faye, and Thouraya Triki. The report also draws on drat material specially

provided by Mike Coates, Robert Cull, Florence Dae, Michael Fuchs, Robin Ho-

meister, Thomas Losse-Mller, Margaret J. Miller, Thilasoni Benjamin Musuku,

David Porteous, Lemma Senbet, Mircea Trandar, Simon Christopher Walley,

Makayo Witte, and Alice Zanza. Excellent research assistance has been provided byMohammad Hosseini, Thierry Kangoye, Ines Mahjoub, Pranav Ramkrishnan,

Tania Saranga, David Symington, Radomir Todorov, and, in particular, Alexandra

Jarotschkin, who also contributed with textual revisions on the completion o the

consultation process. We are grateul to Stephen McGroarty and Susan Graham

and the World Bank Oce o the Publisher or coordinating the book design, edit-

ing, and production process. The data collection eort on nancial structure and

regulations in Arica was carried out by the Arican Development Banks Depart-

ment o Statistics led by Charles L. Luumpa (Director) and, especially, the team o

Beejaye Kokil (Division Manager), Letsera Nirina (Statistician), and Tarak Hasniand Slaheddin Saidi (Statisticians-Consultants).

External peer reviewers or the study were Patrick Honohan, (Governor, Central

Bank o Ireland), Perks Ligoya (Governor, Reserve Bank o Malawi), Louis Kasek-

ende (Deputy Governor, Bank o Uganda), Laurence Harris (Proessor o Econom-

ics, School o Oriental and Arican Studies, University o London), Allen Franklin

(Proessor o Finance and Economics, The Wharton School), Mohammed Omran

(Vice Chairman, Insurance Holding Company, Arab Republic o Egypt), Ghada

Waly (Assistant Resident Representative, UN Development Programme, Egypt),

Ismail Douiri (Co-Chie Executive Ocer, Attijariwaa Bank, Morocco), ZiadOueslati (Founding Partner Tuninvest-Finance Group), Bouakez Haedh (Proes-

sor o Economics, HEC Montreal), and Lemma Senbet (Proessor o Finance,

University o Maryland).

Acknowledgments


15/322

xiv Acknowledgments

The Arican Development Bank and World Bank reviewers were Simon C. Bell,

Mohamed Damak, Asli Demirg-Kunt, James Emery, Michael Fuchs, Devinder

Goyal, Leonce Ndikumana, Gabriel Victorien Mougani, Charles Muthuti, Sebas-

tian O. Okeke, Tilahun Temesguen, and Dsir Vencatachellum.

We are grateul or the insights o ocials and market participants visited or

the study in Egypt, Kenya, Libya, Mauritius, Morocco, Senegal, South Arica, Tuni-

sia, and Uganda. We have beneted rom comments during consultation events in

Dakar, Frankurt, Johannesburg, Nairobi, Tunis, and Washington, DC. At these

events, we appreciate the institutional support we received rom the Kenya School

o Monetary Studies in Nairobi, the Bankers Association o South Arica, and Fin-

Mark Trust in South Arica. We are also grateul or the comments and assistance

provided by Ed Al-Hussainy, Abayomi A. Alawode, Henry K. Bagazonzya, Gunhild

Berg, Johannes Braun, Irma I. Grundling, Juliet Kairuki, Zachary A. Kaplan, Maya

Makanjee, Hannah Messerli, Stephen N. Ndegwa, Ismail Radwan, Oliver Reichert,

David Scott, Riham Shendy, and Smita Wagh.

The study was carried out under the overall guidance o Shantayanan Deverajan

(Chie Economist, Arica Region, World Bank) and Ncube Mthuli (Chie Econo-

mist and Vice-President, Arican Development Bank). It also beneted rom the

support and guidance rom management: Gabriela Braun and Karen Losse (both

Deutsche Gesellschat r Internationale Zusammenarbeit), Leonce Ndikumana

(Director at the Operations Policy Department, Arican Development Bank),

Marilou Uy (Director, Arica Finance and Private Sector, World Bank), and Dsir

Vencatachellum (Director o the Research Department, Arican Development

Bank). Coordination and management support was ably provided by the Secretar-

iat o the Making Finance Work or Arica Partnership, led by Stean Nalletamby

(Coordinator) and his team: Habib Attia, Alessandro Girola, Hugues Kamewe,

Sarah Mersch, and Carlotta Saporito.

On behal o the Arican Development Bank and the World Bank and on their

own behal, the authors are especially grateul to the German Federal Ministry o

Economic Cooperation and Development (BMZ) or providing nancial support

or the preparation o the study.


16/322

xv

AIDS acquired immune deciency syndrome

AML antimoney laundering

ATI Arican Trade Insurance

ATM automated teller machine

BCP Basel Core Principle or Eective Banking Supervision

BH Banque de lHabitat (Tunisia)

BRIC Brazil, Russian Federation, India, and China

CDG Caisse de Dpt et de Gestion (Morocco)

CEMAC Economic and Monetary Community o Central Arica

CFT Combating the Financing o Terrorism

CGAP Consultative Group to Assist the Poor

DBSA Development Bank o South Arica

DFI development nance institutionFDI oreign direct investment

G-7 Group o Seven

G-20 Group o Twenty

GDP gross domestic product

ICT inormation and communication technology

LBC licensed buying company (Ghana)

LIA Libyan Investment Authority

m-banking mobile bankingMFI micronance institution

MIX Micronance Inormation Exchange

Abbreviations


17/322

xvi Abbreviations

MOU memorandum o understanding

MRFC Malawi Rural Finance Corporation

NAFIN Nacional Financiera (Mexico)NGO nongovernmental organization

NMB National Micronance Bank (Tanzania)

OPCR organisme de placement en capital risque(venture capital

investment und)

PPP public-private partnership

PPT Pesewa Power Trust (Ghana)

RBM Reserve Bank o Malawi

SACCO savings and credit cooperativeSIM subscriber identication module

SME small and medium enterprise

SWF sovereign wealth und

TPS Tenant Purchase Scheme

UEMOA West Arican Economic and Monetary Union

Note: All dollar amounts are U.S. dollars (US$) unless otherwise indicated.

Country Abbreviations

ABW Aruba

ADO Andorra

AFG Aghanistan

AGO Angola

ALB Albania

ANT Netherlands AntillesARE United Arab

Emirates

ARG Argentina

ARM Armenia

ASM American Samoa

ATG Antigua and Barbuda

AUS Australia

AUT Austria

AZE Azerbaijan

BDI Burundi

BEL BelgiumBEN Benin

BFA Burkina Faso

BGD Bangladesh

BGR Bulgaria

BHR Bahrain

BHS Bahamas, The

BIH Bosnia and

Herzegovina

BLR Belarus

BLZ Belize

BMU BermudaBOL Bolivia

BRA Brazil

BRB Barbados

BRN Brunei

BTN Bhutan

BWA Botswana

CAF Central Arican

Republic

CAN Canada

CHE Switzerland

CHI Channel IslandsCHL Chile

CHN China

CIV Cte dIvoire

CMR Cameroon

COG Congo, Rep.

COL Colombia

COM Comoros

CPV Cape Verde

CRI Costa Rica

CUB Cuba

CYM Cayman IslandsCYP Cyprus

CZE Czech Republic

DEU Germany

DJI Djibouti

DMA Dominica

DNK Denmark

DOM Dominican Republic

DZA Algeria

ECU Ecuador

EGY Egypt, Arab Rep.

ERI EritreaESP Spain

EST Estonia

ETH Ethiopia

FIN Finland

FJI Fiji

Code Country name Code Country name Code Country name


18/322

Abbreviations xvii

FRA France

FRO Faeroe Islands

FSM Micronesia, Fed. Sts.

GAB Gabon

GBR United Kingdom

GEO Georgia

GHA Ghana

GIN Guinea

GMB Gambia, The

GNB Guinea-Bissau

GNQ Equatorial Guinea

GRC Greece

GRD GrenadaGRL Greenland

GTM Guatemala

GUM Guam

GUY Guyana

HKG Hong Kong, China

HND Honduras

HRV Croatia

HTI Haiti

HUN Hungary

IDN Indonesia

IMY Isle o ManIND India

IRL Ireland

IRN Iran, Islamic Rep.

IRQ Iraq

ISL Iceland

ISR Israel

ITA Italy

JAM Jamaica

JOR Jordan

JPN Japan

KAZ KazakhstanKEN Kenya

KGZ Kyrgyz Republic

KHM Cambodia

KIR Kiribati

KNA St. Kitts and Nevis

KOR Korea, Rep.

KWT Kuwait

LAO Lao PDR

LBN Lebanon

LBR Liberia

LBY Libya

LCA St. Lucia

LIE Liechtenstein

LKA Sri Lanka

LSO Lesotho

LTU Lithuania

LUX Luxembourg

LVA Latvia

MAC Macao, China

MAR Morocco

MCO Monaco

MDA Moldova

MDG Madagascar

MDV Maldives

MEX Mexico

MHL Marshall Islands

MKD Macedonia, FYR

MLI MaliMLT Malta

MMR Myanmar

MNG Mongolia

MNP Northern Mariana

Islands

MOZ Mozambique

MRT Mauritania

MUS Mauritius

MWI Malawi

MYS Malaysia

MYT MayotteNAM Namibia

NCL New Caledonia

NER Niger

NGA Nigeria

NIC Nicaragua

NLD Netherlands

NOR Norway

NPL Nepal

NZL New Zealand

OMN Oman

PAK PakistanPAN Panama

PER Peru

PHL Philippines

PLW Palau

PNG Papua New Guinea

POL Poland

PRI Puerto Rico

PRK Korea, Dem. Rep.

PRT Portugal

PRY Paraguay

PYF French Polynesia

QAT Qatar

ROM Romania

RUS Russian Federation

RWA Rwanda

SAU Saudi Arabia

SDN Sudan

SEN Senegal

SGP Singapore

SLB Solomon Islands

SLE Sierra Leone

SLV El Salvador

SMR San Marino

SOM Somalia

STP So Tom and

Principe

SUR Suriname

SVK Slovak RepublicSVN Slovenia

SWE Sweden

SWZ Swaziland

SYC Seychelles

SYR Syrian Arab Republic

TCD Chad

TGO Togo

THA Thailand

TJK Tajikistan

TKM Turkmenistan

TMP Timor-LesteTON Tonga

TTO Trinidad and Tobago

TUN Tunisia

TUR Turkey

TWN Taiwan, China

TZA Tanzania

UGA Uganda

UKR Ukraine

URY Uruguay

USA United States

UZB UzbekistanVCT St. Vincent and the

Grenadines

VEN Venezuela, RB

VIR Virgin Islands (U.S.)

VNM Vietnam

VUT Vanuatu

WBG West Bank and Gaza

WSM Samoa

YEM Yemen, Rep.

YUG Serbia and

Montenegro

ZAF South Arica

ZAR Congo, Dem. Rep.

ZMB Zambia

ZWE Zimbabwe

Code Country name Code Country name Code Country name


19/322


20/322

Introduction

Cautious hope is in the air or nance in Arica. While the global crisis mayhave dented some o the progress made since the beginning o the 21st cen-tury, one eels the optimism and sees the positive trends. A deepening o -

nancial systems can be observed in many Arican countries, with more nancial

services, especially credit, provided to more enterprises and households. New play-ers and new products, oten enabled by new technologies, have helped broadenaccess to nancial services, especially savings and payment products. Innovativeapproaches to reaching out to previously unbanked parts o the population go be-yond cell phonebased M-Pesa in Kenya and basic transaction accounts, such asMzansi accounts in South Arica. Competition and innovation dominate Aricannancial systems, and, or every ailure, there is now at least one success. However,many challenges remain, and the journey toward deeper, more-ecient, and more-inclusive nancial systems will be long and raught with many dicult choices in

many countries in Arica.Aricas nancial systems have progressed over the past 20 years. Yes, the prom-

ise o the eorts at liberalization, privatization, and stabilization in the 1980s hasonly been partly ullled, though Arican nance has been stable or quite a whilenow. Since the peak o the banking crises in the 1980s, there have been ew systemicbanking crises, though pockets o ragility persist, oten related to political crisis ordeciencies in governance. On average, banks in Arica are well capitalized andliquid. Still, the benets o deeper, broader, and cheaper nance have not yet beenreaped. Finance in Arica still aces problems o scale and volatility. And the same

liquidity that helps reduce volatility and ragility in the nancial system is also asign o the limited intermediation capacity on the continent. Nonetheless, as wediscuss below, globalization, technology, and increasing regional integration mayprovide new opportunities or nance in Arica. 1

Chapter 1

Financing Africa:

Setting the Stage


21/322

2 Financing Arica: Through the Crisis and Beyond

As a consequence o recent positive trends, Arican nancial sectors entered thecrisis with a cushion o high levels o capitalization and liquidity. Financial institu-tions on the continent largely evaded the direct impact o the global nancial crisis.Low levels o integration with international nancial markets limited the exposureto toxic assets and to the volatility o international markets. With a number o im-portant exceptions, nonperorming loans remained stable despite the slowdown ineconomic growth. Notwithstanding the limited impact o the global nancial crisisand the improvements in overall stability, several countries have been sueringrom homegrown nancial ragility related to governance challenges and socio-political unrest.

Aricas economies have been hit by the ensuing Great Recession through re-duced trade fows and reduced portolio fows and remittances. The all in demandand, consequently, in the prices o commodities has hit commodity-based econo-mies signicantly. The rapid decline in global trade that started in late 2008 (by upto 45 percent in real terms year on year) aected all Arican economies and, to alarge extent, explains the lower growth the region experienced in 2009. The in-creasing spreads and reduced maturities resulting rom the shortage o liquidity inthe global nancial system have made investment and trade more dicult as well.Arica has also experienced a rapid reduction in capital fows, which has depressedstock exchange indexes throughout the continent and orced governments andcompanies to cancel bond and stock issues. Nonetheless, the overall impact o theGlobal Recession seems to have been milder on Arica than on other regions o theworld, and the continent is already expected to match precrisis growth rates o5 percent in 2011 (IMF 2011).

More importantly, Arica will be aected by long-term trends that started beorethe crisis and have been reinorced by the crisis, especially the shits in the distribu-tion o global economic power. The shit o weight away rom the North (the G-7)toward the East (especially China and India) and the South (to the G-20) has beenanother consequence o the crisis not only or Arica, but also or the global nan-cial and economic system. In the context o globalization, the BRIC countries, es-pecially China and India, but, more recently, also Brazil, are playing a growing rolein Arica.1 This is refected in capital fows and also the structure o banking sys-tems. While Indian banks have long had a presence in East Arican countries, thepurchase o a 20 percent stake by a Chinese bank in Standard Bank in South Aricarepresents a new trend or China. The oreign direct investment o Brazil, China,and India has been increasing across the continent. It began mostly in natural re-source extraction and agriculture, but has now extended to other sectors. Addi-tional unds have been orthcoming rom the Gul region, oten in the orm osovereign unds. This shit in capital fows and international governance oers op-portunities and challenges or Arica: opportunities in terms o urgently neededresources and challenges in terms o managing the resources properly.

This chapter opens with a description o the books objectives and contribu-tions, including the main policy messages. It then develops the basic analytical


22/322

Financing Arica: Setting the Stage 3

ramework through which we view nancial sector development in Arica, distin-guishing among beneciary concepts and groups so as to ocus our policy mes-sages. We highlight the importance o nancial sector development or economicgrowth and poverty alleviation and conclude with a summary o the persistentproblems in Aricalimited scale, inormality, volatility, and governance issuesthat require new solutions. Opportunities or solutions are possible within the newtrends o globalization, regional integration, and technology.

Financing Africa, the Book

This book targets the stakeholders in Aricas nancial systems. Stakeholders areunderstood widely: policy makers, regulators, practitioners, development partners,academics, and others. The book includes a stocktaking and orward-looking exer-cise that indicates viable paths to nancial sector deepening and broadening. Itrepresents an eort to document new and existing trends in Aricas nancial sec-tors, taking into account Aricas many dierent experiences. It ocuses on generaltrends and, thus, does not encompass an exhaustive, detailed discussion o the de-velopment and structure o each o the 53 Arican nancial systems (or example,see Allen, Otchere, and Senbet 2010).

The book intends to contribute to the eorts o Arican policy makers to cap-ture opportunities and overcome challenges. It outlines broad policy messages ornancial systems in Arica on the premise that one size does not t all. It does notoutline strategies or the nancial sector in every country across the continent,but, rather, oers general policy messages. It also discusses specic segments o thenancial sector, such as rural and housing nance; it does not, however, oer anexhaustive and conclusive coverage o these segments. We leave that to more spe-cialized publications in these areas.

The book builds on and extends substantially the World Bank publication Mak-ing Finance Work or Arica, which drew attention to the opportunities and chal-lenges o nancial system development across Arica (see Honohan and Beck 2007).First, it relies on a much broader array o data than the previous publication. Sec-ond, it includes North Arican countries in the analysis, which, along many dimen-sions, have ollowed a dierent path o nancial sector development. Finally, it ex-pands on the analysis o the previous publication, including a thorough discussiono the regulatory challenges o nance in Arica. Critically, the world is dierent in2011 rom the world in 2007. Box 1.1 summarizes the main dierences between thetwo publications.

An Analytical FrameworkIn theory, nancial institutions and markets exist to help overcome market ric-tions that make direct exchanges between economic agents dicult. Academicstypically distinguish between specic unctions o nancial service providers, such


23/322


Financing Africais a ollow-up to

Making Finance Work for Africa. What has changed sinceearly 2007, and what distinguishes this publication rom the previous one?

The environment has changed

The environment in which Arican nancial systems operate has changed dramatically over

the past our years. The number o Arican countries experiencing a systemic banking crisis

has allen rom a peak o 15 in the mid-1990s to a sporadic outlier in the 2000s. Credit to the

private sector (as a ratio o gross domestic product [GDP]) has risen by more than 20 per-

centage points since 1990. There has been a growing trend toward regional integration

within the continent in recent years, though this trend started well beore 2007. Kenyan, Mo-

roccan, Nigerian, and South Arican banks are rapidly expanding operations in the region.Over the past our years, the transormational impact o the deepening and broadening o -

nancial system technology has become clear as well. With over 13 million clients in Kenya,

M-Pesa is the worlds most widely used telecommunications-led mobile money service.

Globally, too, the environment is dierent. We might be at the tail end o the rst global

nancial crisis o the 21st century and the Great Recession, but the global nancial system

has changed dramatically. The center o economic and nancial power has shited to the

South and East, which is also refected in the replacement o the G7 by the G20 as the major

international policy coordination body.

The set of information and experiences is larger

Relative to our years ago, we have a much richer and more detailed set o data available.

Specically, we can draw on a systemic data collection eort by the Arican Development

Bank and the Deutsche Gesellschat r Internationale Zusammenarbeit o indicators on the

development and structure o nancial systems across Arica, as well as the regulatory

ramework. Furthermore, the international community has made enormous progress in col-

lecting data on the outreach o nancial systems and the barriers to access among enter-

prises and households; we draw on this experience.

A critical dierence with respect to the previous publication is the inclusion o the North

Arican subregion. The inclusion o these countriesdierent in income level and economic

and nancial structureand the comparison with other parts o Arica enrich the discussionin the book and provide additional insights into the process o nancial sector deepening

and broadening. The recent turmoil in this part o Arica, however, makes many conclusions

on the related nancial systems appear tentative.

While Arica can learn rom the rest o world, the world can learn rom Arica, as we lay

out in this publication. The experience with mobile phone banking, or example, shows the

power o technology and the potential that payment-led inclusion strategies possess relative

to credit- or savings-led inclusion strategies. We thereore reer to experiences in other re-

gions, but also experiences in dierent Arican countries and how these experiences may

be used across the continent.

The focus has expanded

The altered global environment also calls or a somewhat dierent emphasis. In light o the

recent regulatory reorm debate in Europe and the United States and in the context o the

Box 1.1 Whats New?


24/322


as (1) acilitating the exchange o goods and services by providing a medium oexchange; (2) pooling societys savings or investment in large investment projectsbeyond the savings capacity o small individual savers; (3) screening potential in-vestment projects, thus putting societys savings to the best use; (4) monitoringenterprises and thus making sure money is used or the best purpose; and (5) in-vesting in risk management services, such as diversiying across dierent projectsor smoothing volatility over time.2 These unctions overlay with the practitionersdistinction among (1) payment and transaction services, (2) deposit and savings

services, (3) credit services, and (4) insurance and risk management services. Theseservices are oten provided by dierent institutions or in dierent markets. Yet an-other, partly overlaying distinction is based on dierent beneciary groups andtime horizons. Expanding on a distinction made by Honohan and Beck (2007),this book distinguishes between three concepts: Finance or Markets, Finance orGrowth, and Finance or All. This distinction helps us rame our discussionthroughout the book.

Finance or Marketsrelates to nancial services that underlie short-term com-mercial market transactions, such as trade nance, remittance payments, andvarious types o short-term credit acilities. This concept relates primarily to thenancial system unction o enabling market-based transactions within theeconomy and across borders. By acilitating commerce, nancial systems allowthe market-based exchange o goods and services beyond the immediate amily

G20 process, we ocus more prominently on the regulatory ramework. We argue that thereorm suggestions developed in response to the recent crisis have to be adapted with cau-

tion to the Arican context and, even within the Arican region, to the level o development o

dierent nancial systems.

Including North Arica also reemphasizes the benets o ocusing on dierences across

Arica. Thus, needs and policy options vary between small and large and between low- and

middle-income countries, while landlocked, resource-rich, and ragile states ace yet an-

other set o challenges.

Honohan and Beck (2007) present two main policy recommendations: (1) strengthen

credit and property registries and streamline court procedures and (2) establish independent

supervisors. This publication ollows a dierent path by presenting three main general mes-

sages, which are then ne-tuned in each o the thematic chapters and also detailed or di-

erent country groups.

Some themes and contrasts are maintained rom the previous book. The contrast be-

tween the modernist and the activist approach is also used in this publication to highlight

the advantages o a careul assessment o the role o government, a role that should help

create and develop markets rather than replace them. We also build and expand on the dis-

tinction between Finance or All and Finance or Growth by highlighting the importance o

nance or basic market transactions beyond ostering long-term investment activities.

Box 1.1 Whats New? (continued)


25/322


and community. Finance or Markets reers to nancial services or enterprisesand households, thus cutting across all possible beneciary groups. The conceptcovers transaction and payment services, including remittances rom emigrantworkers to their amilies back home. It covers deposit services or householdsand enterprises, as well as short-term credit acilities or enterprises o all sizes,including trade credit. These basic services are provided by almost every nan-cial system in the world, even the most rudimentary ones, although at dierentdegrees o eciency. They are mostly provided by banks, but may also be pro-vided by nonbank nancial service providers, including telecommunicationscompanies. The recent crisis and the reduction in the supply o trade nanceunderline the importance o Finance or Markets.

Finance or Growth relates to the nance or enterprises, households, and govern-ments that supports medium- and long-term activities (longer than 12 months).Finance or Growth is nance mainly or investment purposes, and it is here thatnancial institutions and markets ulll their key unction o the maturity trans-ormation o short-term liquid claimsbe they deposits or marketable securi-tiesinto long-term investment nance. Finance or Growth thus involves a keyunction o nancial systems: pooling societys savings and putting them to theirbest use. This comprises risk management techniques and the screening andmonitoring o entrepreneurs and projects. It relates to large-scale nance, in-

cluding or inrastructure and agriculture, and nance or small and mediumenterprises and to debt and equity instruments, as well as hybrid instruments,such as mezzanine debt and guarantees. These services are provided by an arrayo institutions, including banks, insurance companies, pension unds, mutualunds, and private equity unds, and relate to activities on dierent nancialmarkets, including stock and bond markets. Moving rom Finance or Marketsto Finance or Growth constitutes a major challenge or many low-income coun-tries, including in Arica.

Finance or All relates to the process o expanding nancial services both or

markets and or growth to the largest possible segment o the population, in-cluding households, small enterprises, and large rms. Finance or All overlapsthe concepts o Finance or Markets and Finance or Growth, but reers to theprocess by which short- and long-term nancial services, including payment,savings, credit, and insurance services, are pushed out to previously unservedsegments o the population. It overlaps with Finance or Markets to the extentthat access to basic transaction services is being extended to all segments o thepopulation. It overlaps with Finance or Growth to the extent that more seg-ments o the population gain access to contractual savings services, while mi-

croenterprises gain access to investment nance. In discussing Finance or All,we reer to all types o ormal nancial institutions, but also semiormal nancialinstitutions such as cooperatives or savings and credit cooperatives. Finance orAll has been a challenge throughout the world not only or low-income coun-


26/322


tries, but also or many middle-income countries that have made substantialprogress in the dimensions o Finance or Markets and Finance or Growth.

The three concepts overlap, including in policy prescriptions, but it is importantto keep in mind the dierent ocus o each. Countries at dierent levels o eco-nomic and nancial development might ocus on dierent concepts. Postconfictor low-income economies might ocus mainly on the basic services implied by Fi-nance or Markets, and middle-income and socioeconomically more stable coun-tries might ocus on Finance or Growth strategies.3 Finance or All has remained achallenge or low- and middle-income countries and even or some high-incomecountries, such as the United Kingdom or the United States.

It is important to note that these three concepts do not involve a trade-o, least

o all the Finance or All approach. It is more about sequencing than trading o.For instance, the existence o ecient nancial services or market exchange is thebasis or longer-term nancial contracts.

The Main Messages and a Caveat

In Arica, distinguishing among these three concepts is critically important orpolicy design. While Arican economies and nancial systems share many eatures,there are critical dierences along notable dimensions. Financial systems ace di-

erent challenges in low- and middle-income countries across the continent. Basicnancial services or commercial transactions and short-term credit (Finance orMarkets) characterize the nancial systems o many low-income countries, whereormal nancial services are oten limited to a small share o enterprises and house-holds. Middle-income countries are characterized by a much larger outreach obanking systems to households and enterprises, a larger variety o nancial servicesand products, and a diversication o nancial institutions and markets. Size mat-ters: even among low-income countries, larger economies are able to sustain largerand more diversied nancial systems.

There are also important geographical dierences. North Arican nancial sys-tems are dominated by government-owned nancial institutions to a much largerextent than systems in Sub-Saharan Arica, where many systems are weighted to-ward oreign-owned banks. Even there, though, governments are preponderant inother segments o the nancial system, such as the pension sector and the bondmarket. However, there are also important dierences in the challenges that nan-cial systems in densely populated economies, such as Rwanda and Uganda, acerom systems in countries with more dispersed populations, such as Ethiopia orTanzania. There is an important distinction between common law and civil code

countries. Common law countries typically have a more fexible legal and regula-tory ramework that oers more room or innovation, while civil code countriesrely more steadily on written codes and oten take longer to adjust the legislativeand regulatory ramework to new developments. Finally, postconfict countries


27/322


and economies with abundant natural resources ace their own unique set o chal-lenges in achieving nancial deepening and broadening.

We reer to these distinctions throughout the book and provide policy recom-mendations or the various subgroups in chapter 6.

Taking into account these large dierences across the continent, we use theramework above to develop the three main messages resulting rom our analysis,as ollows:

Competition is the most important driver o fnancial innovation that will help A-

rican fnancial systems deepen and broaden. Competition, in this context, isbroadly dened and encompasses an array o policies and actions. On the broad-est level, it implies a nancial system that is open to new types o nancial service

providers, even i they are nonnancial corporations. It allows the adoption onew products and technologies. The example o cell phonebased payment sys-tems across the continent is one o the most powerul illustrations in this cat-egory. Within the banking system, competition implies low entry barriers ornew entrants, but also the necessary inrastructure to oster competition, such ascredit registries that allow new entrants to draw on existing inormation. Toachieve more competition in smaller nancial systems, more emphasis has to beplaced on regional integration. However, this might also mean more active gov-ernment involvement by, or example, orcing banks to join a shared payment

platorm or contributing negative and positive inormation to credit registries.While it is important to stress that the ocus on innovation and competitionshould not lead to the neglect o nancial stability, there has been a tendency inmany Arican countries to err too much on the side o stability.

A second and related message is that there should be an increasing ocus on fnan-cial services rather than on specifc institutions. Across all three dimensions dis-cussed above (Finance or Markets, Finance or Growth, and Finance or All), wecare primarily about the necessary nancial services and, only in a second in-stance, about the institutions or markets that provide the services. Banks are an

important component o every nancial system, but i nonbanks are better atproviding certain nancial services, they should be allowed to do so. I the smalleconomies o Arica cannot sustain organized exchanges, the emphasis shouldbe placed instead on alternative sources o equity nance, such as private equityunds. I the local economy is not suciently large to sustain certain segments oa nancial system, then the import o such services should be considered. Onesize does not t all: smaller and low-income countries are less able than largerand middle-income countries to sustain a large and diversied nancial systemand might have to rely more heavily on international integration.

Finally, there is a need or increased attention on the users o fnancial services.Turning unbanked enterprises and households into a bankable population andultimately banked customers involves more than pushing nancial institutionsdown-market. Achieving such a change requires nancial literacy, that is, knowl-


28/322


edge about products and the capability to make good nancial decisions amonghouseholds and enterprises. It also means that nonnancial constraints must beaddressed, such as, most prominently, in agriculture. It includes a stronger em-phasis on equity nancing or oten overleveraged enterprises. It also includes aconsumer protection ramework, though what suits South Arica, or example,may be too costly in resources and skills or Malawi.

Financial Sector Development: Why Do We Care?

The provision o nancial services or specic beneciary groups is important, butthe ultimate goal is economic development. What is the role o nance in the devel-opment process in Arica? How important is nancial sector development relativeto development in other policy areas? Where should the emphasis lie: in banks ormarkets? Broad cross-country comparisons, but also experiences in the region,have provided insightul evidence in this respect.

Ultimately, nancial deepening and broadening can contribute to Aricas moveout o poverty and low-income status toward middle-income and emerging mar-ket status. The vision is o a nancial system that ullls the three concepts dis-cussed above by providing a sound and eective platorm or the market-basedexchange o goods and services, attracting and intermediating the necessary re-sources or long-term private and public investment, and expanding nancial ser-vices to larger segments o the population so as to oer, at least, access to transac-tion services.

Two decades ago, nancial system development was an aterthought in the mindo a development economist designing a policy agenda. Today, nancial sector pol-icies have become a centerpiece in the debate on how to oster growth in low-income countries, reduce stark poverty levels, and, ultimately contribute to theachievement o the Millennium Development Goals. Over this period, ample evi-dence based on various levels o aggregation and distinct methodologies has beenaccumulated on the growth-enhancing eect o nancial sector development. Evenaccounting or reverse causation, research has established the robust positive im-pact o nancial sector deepening on economic development. Figure 1.1 illustratesthe conclusion o a well-established body o empirical evidence: countries withhigher levels o credit to the private sector relative to GDP experienced higher aver-age annual real GDP per capita growth rates over the period 19802007. The rela-tionship holds not only or a broad cross-section o countries, but also withinArica. The conclusion is conrmed by cross-country, panel, and time-series esti-mation techniques.4

The eect o nance on growth is not only statistically, but also economicallysignicant. To illustrate the eect o nancial deepening, compare Ethiopia withThailand. Over the period 19802007, the ratio o private credit to GDP averaged18 percent in Ethiopia, but 87 percent in Thailand. The cross-country comparisonsillustrated in gure 1.1 suggest that Ethiopias real GDP per capita would have


29/322


grown by 1.3 percentage points more had the country been at the same level o -nancial development as Thailand, or 1.4 percent instead o the actual 0.1 percent.Under this scenario, GDP per capita would have been over 40 percent greater in2007.5 We can also compare the nancial development and corresponding growthperormance in Arica with that in low- and middle-income countries in East Asia.While nancial development measured according to the ratio o private credit toGDP stood, on average, at 21 percent across Arica over the period 19802007, itwas 32 percent in East Asia. During the same period, the East Asian economiesgrew 2.3 percent per year on average, while the Arican economies grew 0.7 percenton average. The estimates illustrated in gure 1.1 suggest that 0.4 o a percentagepoint o this dierence in average annual growtha quarter o the dierencewas caused by the lower level o nancial development. Thus, the estimates suggestthat, today, Arica could have a GDP per capita greater by 13 percent than the ac-tual GDP per capita. This is, indeed, a signicant loss.

The positive impact o nancial development on growth does not mean thatgrowth has no infuence on nancial deepening and broadening. On the contrary,by helping to increase incomes, nancial deepening can create additional demandor nancial services, thus generating a positive eedback loop. Policies that help

Source:Author calculations.

Note:Sample size: 99 countries. The gure shows a partial scatter plot o private credit to GDP and GDP per capita

growth averaged over 19802007 and controlling or initial GDP per capita, government consumption, infation, trade

openness, and education. For a complete listing o 3-letter country codes and the respective country names, see

pages xvixvii.

DZARWA

ZMB

SWZ

avera

geGDPgrowth(logresidual)

private credit/GDP (log residual)

Africa all other regions

residual trendline

BDI

BEN

BWA

CAF

CIV

CMR

COG

EGY

GAB

GHA

GMB

KEN

LSO

MAR

MLI

MOZ

MRT

MUS

MWI

NER

SDN

SEN

SLE

TGO

TUN

ZAF

0.04

0.02

0

0.02

0.04

2.00 1.00 0 1.00 2.00

Figure 1.1 Finance and Growth across Countries, 19802007


30/322


oster nancial sector development ultimately also help establish a virtuous growthcycle. Moreover, many o the policies that oster nancial sector deepening andbroadening, including an eective contractual and inormation ramework andmacroeconomic stability, also have a direct positive impact on economic develop-ment and poverty alleviation.

What are the channels through which nancial development helps increase eco-nomic growth? While nancial systems assist in pooling savings, transorming ma-turity, and converting savings into capital accumulation, it is ultimately throughimprovements in resource allocation and productivity growth that nance helpseconomies grow more quickly (Beck, Levine, and Loayza 2000; Love 2003; Wurgler2000). The unctions o attracting deposits and investment and transormingshort-term claims into long-term assets, thereby nancing investment, should ob-viously not be ignored; they are the basis or the ultimate unction o nance, whichis to put the savings o society to the best use, that is, put savings where they canreap the highest (expected) returns, thus translating into growth. Financial deep-ening especially helps industries that rely heavily on external nance, but it alsohelps reduce the nancing constraints on enterprises, particularly smaller rms(Rajan and Zingales 1998; Beck, Demirg-Kunt, and Maksimovic 2005). Financialdeepening thus has a transormative eect on economies by shaping industrialstructure, distribution by rm size, and even organizational structures (Demirg-Kunt, Love, and Maksimovic 2006). It is the acilitating role o nancial systemsthat helps oster economic growth. Finance provides opportunities or new entre-preneurs and osters innovation and competition as well.

Providing external nance to enterprises in the orm o equity, debt, or somehybrid thus seems critical to the positive impact o nance on growth. Recent cross-country comparisons have indeed ound that it is enterprise credit, rather thanhousehold credit, that explains the positive impact o nance on growth (Beck etal. 2009). This does not mean that credit services or households are not important;the growth eect o nancial development, however, seems to come mainly romenterprise nance. One thereore has to look beyond credit services to other nan-cial services in discussing the welare impact o nancial service provision onhouseholds. This casts doubt on the credit-led inclusion strategy oten propagatedby microcredit institutions and puts a premium on enhancing access to savings andtransaction services. We return to this topic in chapter 3.

Financial sector development is important not only or ostering the economicgrowth process, but also or dampening the volatility o the growth process. Asshown by Aghion et al. (2010), nancial systems can alleviate the liquidity con-straints on rms and acilitate long-term investment, which ultimately reduces thevolatility o investment and growth. Similarly, well-developed nancial marketsand institutions can help dampen the negative impact that exchange rate volatilityhas on rm liquidity and thus investment capacity (Aghion et al. 2009). This is es-pecially important in economies that depend heavily on natural resources and arethus subject to high terms o trade and real exchange rate volatility.


31/322


What has the recent crisis taught us about the importance, but also the risks onancial deepening?6 First and oremost, it has shown us the enormous risks thatnancial system ragility can create in the overall economy and or peoples liveli-hoods. The global crisis and the ensuing Great Recession have put in doubt theparadigm that nancial deepening is good or growth under any circumstance.Consumer credit booms in several European countries and the United States,ueled by the combination o regulatory neglect, the eeling that this time is dier-ent, and the liquidity glut linked to global macroeconomic imbalances, ended inthe global nancial crisis. International links through global nancial marketshelped propagate the shock, rst, through nancial markets, while trade links ulti-mately resulted in the propagation o the real sector slump. For students o nan-cial systems, the bright (growth-enhancing) and dark (instability) sides o nancialdevelopment go hand in hand. The same mechanism through which nance helpsgrowth also makes nance susceptible to shocks and, ultimately, ragility. Speci-cally, the maturity transormation rom short-term savings and deposit acilitiesinto long-term investments is at the core o the positive impact o a nancial sys-tem on the real economy, but also renders the system susceptible to shocks. The rolethat nance has as a lubricant or the real economy likewise exacerbates the eecto nancial ragility on the real economy. However, the externalities that the ailureo nancial institutions and markets impose on the real economy and the heavygovernment support or incumbent nancial institutions that nancial ragilitythereore typically triggers are taken into account by the stakeholders in nancialsystems and give these stakeholders an incentive to be aggressive in the ace o risks.It is thus critical to harness nancial market orces or the benet o the real econ-omy and the population at large, rather than ocus on nance or its own sake.

Instead o throwing out the baby with the bathwater, it is thereore important toconstruct a regulatory and governance ramework that minimizes the risk o ragil-ity and provides policy makers with better possibilities or managing bank ailuresin a way that is incentive-compatible. I there is a lesson to be learned in Arica romthe crisis, it seems to be that the growth benets o a well-developed nancial sys-tem can only be reaped in a stable macroeconomic environment protected by anappropriate regulatory and supervisory ramework and strong internal bank gover-nance. This means there should be more transparency and accountability in bankmanagement, less direct government intervention in the regulatory and supervisoryprocess, and a ocus on building up mechanisms o market discipline. However, thesituation also highlights the demand-side constraints in terms o nancial literacyand consumer protection, a topic we take up in several parts o the publication.

Ultimately, the nancial systems o Arica are signicantly less sophisticatedthan systems elsewhere, and most are ar rom becoming overheated as several -nancial systems in Europe and North America did beore the crisis. This does notmean that there is no ragility (see chapter 5). Nonetheless, most o the ragility inrecent years has not arisen because o too much nance, but because o misallo-cated nance generated by governance challenges. In a nutshell, Aricas nancial


32/322


systems stand to gain signicantly rom deepening and broadening. I there is adecreasing marginal benet rom nancial deepening or even a threshold wheremore nancial deepening may have a negative eect, Aricas nancial systems arear rom reaching it.7

Who benets the most rom nancial deepening? While theory provides con-ficting analyses about whether it is the rich or the poor who benet most rom -nancial sector development, cross-country comparisons indicate that nancialdeepening has a pro-poor eect (Beck, Demirg-Kunt, and Levine 2007). Figure1.2 illustrates the pro-poor eect o nance: countries with deeper nancial sys-tems see poverty levels drop more rapidly. As in the case o economic growth, theeconomic eect o nancial deepening on poverty reduction is strong. Again, acomparison between Ethiopia and Thailand illustrates. Specically, the cross-country comparisons shown in gure 1.2 suggest that, instead o a reduction in thepoverty headcount rom 33 to 23 percent over the period 1981 to 2000, a level onancial development similar to that o Thailand would have allowed a reductiono the headcount to 9 percent in Ethiopia.8

Financial deepening can have broader eects on socioeconomic developmentthan those captured by GDP per capita and the poverty headcount. While eradicat-

Source:Beck, Demirg-Kunt, and Levine (2007).

Note:Sample size: 68 countries. The gure shows a partial scatter plot between the ratio o private credit to GDP

and growth in the headcountthe share o the population living on less than a dollar a dayaveraged over the

period 1980 to 2003 and controlling or the initial headcount. For a complete listing o 3-letter country codes and the

respective country names, see pages xvixvii.

MAR

ave

rage

reduction

inh

ead

count

(log

residual)

DZA

BWA

BFA

BDI

CMR

CIV

EGYETH

GMB

GHA

KEN

LSO MDG

MWI

MLI

MRT

NER

NGARWA

SEN

ZAF

TUN

UGA

ZMB ZWE

0.40

0.20

0

0.20

2.00 1.00 0 1.00 2.00

private credit/GDP (log residual)

Africa all other regions

residual trendline

Figure 1.2 Finance and Poverty Alleviation across Countries


33/322


ing extreme poverty by 2015 is one o eight Millennium Development Goals ad-opted in 2000, nancial development may also be linked to the other seven goals,which reer to education, gender equality, health, the environment, and global part-nerships.9 In the case o education and health, one important outcome o access tonancial services occurs through the income eect: better access to nancial ser-vices improves incomes and thereore the possibility o accessing health and educa-tion services, while, at the same time, reducing the need to rely on children as la-borers in the household. Allowing women direct access to nancial services mightimprove the possibility they could become entrepreneurs, thus increasing their in-dividual incomes and their chance to be more independent. This is refected ingreater emale participation in amily and community decision making. There isalso an important insurance eect: better access to credit, savings, or insuranceservices reduces the need to use child labor as a buer in the case o seasonal in-come fuctuations and transitory income shocks and allows consumption smooth-ing in the case o transitory income reductions caused by health shocks. It alsoallows more rapid attention to health problems. Finally, there is an aggregate inra-structure eect: more ecient nancial institutions and markets allow more pri-vate and public investment in the construction o schools and health acilities.

What are the mechanisms o this poverty-reducing impact o nancial deepen-ing? Theory suggests dierent channels. On the one hand, providing access to creditamong the poor might help the poor overcome nancing constraints and allowthem to invest in microenterprises and human capital accumulation (Galor andZeira 1993; Galor and Moav 2004). On the other hand, there might be indirect e-ects through enterprise credit. By expanding credit to new and existing enterprisesand allocating societys savings more eciently, nancial systems can expand theormal economy and pull larger segments o the population into the ormal labormarket. The rst explorations o the channels through which nance aects in-come inequality and poverty levels point to an important role or such indirect e-ects. Specically, evidence rom Thailand and the United States suggests that animportant eect o nancial sector deepening on income inequality and poverty isan indirect one. By changing the structure o the economy and allowing more entryinto the labor market by previously unemployed or underemployed segments othe population, nance helps reduce income inequality and poverty, but not by giv-ing access to credit to everyone (Beck, Levine, and Levkov 2010; Gin and Townsend2004). It is important to stress that this is preliminary evidence to be conrmed orreuted by uture research, but it has centered the debate on an important question:should policy makers ocus on deepening or on broadening nancial sectors? It hasalso helped widen the debate on nancial services or the poor beyond microcreditto other nancial services, such as savings services, payment services (especially inthe context o remittances rom amily members who have emigrated to other partso the country or outside the country), and insurance services.10

A long-running discussion has centered on whether policy makers should ocusmore on banks or on capital markets. While both provide important nancial ser-


34/322


vices, the related technologies are dierent. Banks create proprietary inormationabout their clients, especially borrowers, while capital markets collect and processinormation rom dierent sources and refect this inormation in prices. Banksoer better intertemporal risk diversication tools, while markets are better in di-versiying risk cross-sectionally. Markets are better at oering standardized prod-ucts, while banks are better at oering tailored solutions. However, banks and mar-kets can also be complementary through the application o instruments such assecuritization, by allowing exit strategies or venture capitalists, and by providingcompetition with each other. However, cross-country comparisons have shownthat it is not really the structure o the nancial system that matters, but rather theprovision o nancial services, whether these are supplied by banks or markets(Levine 2002; Beck and Levine 2002; Demirg-Kunt and Maksimovic 2002). Theattempts o policy makers to push articially or the development o a specic seg-ment o the nancial system over another are typically not ruitul. Yet, comparingnancial systems across countries, one can discern clear patterns. Such a compari-son suggests that there is an ordering in the development o dierent segments othe nancial system. Systems in low-income countries are typically based muchmore on banks, while capital markets and contractual savings institutions, such asinsurance companies, develop at a later stage.11 Beck et al. (2008) estimate an in-come elasticity or dierent components o the nancial system, which illustratesthe dierent speeds at which dierent segments develop as GDP per capita rises(see table 1.1). Given the level o GDP per capita on the continent, it is not surpris-ing that all Arican nancial systems are based heavily on banks and exhibit under-developed markets. Not only is the capital market segment o these nancial sys-tems underdeveloped, but also the contractual savings component (insurance,pensions, and mutual unds) is small in most Arican countries. Finally, there is a

Table 1.1 Income Elasticities across Different Segments of the Financial

System

Variable Rank Income elasticity

Public bonds to GDP 1 0.20

Bank deposits to GDP 2 0.35

Bank assets to GDP 3 0.44

Pension unds to GDP 4 0.45

Bank credit to GDP 5 0.49

Stock market capitalization to GDP 6 0.56

Insurance assets to GDP 7 0.66

Institutional investor assets to GDP 8 0.77

Mutual unds to GDP 9 0.88

Private bond capitalization to GDP 10 1.20

Value traded to GDP 11 1.30

Source:Beck et al. (2008).


35/322


scale element to the development o capital markets, and small economies there-ore have dicultyeven in the developed worldin sustaining liquid markets.

The above discussion does not imply that policy makers in Arica should becontent with the underdeveloped nonbank segments o the nancial system. Thereis a need to diversiy the nancial center away rom a heavily bank-dominated sys-tem, but it is also important to recognize that articially creating certain compo-nents o the nancial system without the necessary demand and inrastructure willhave limited economic benet.

Time for New Solutions to Old Problems

Financial institutions and markets exist to help overcome market rictions relatedto transaction costs and risk. However, the eciency with which nancial institu-tions and markets can overcome these market rictions is critically infuenced bycountry characteristics. Fixed transaction costs in nancial service provision resultin decreasing unit costs as the number or size o transactions increases. These xedcosts exist at the level o the transaction, client, institution, and even nancialsystem. Processing an individual payment or savings transaction entails costs thatare, at least in part, independent o the value o the transaction. Maintaining anaccount or an individual client also implies costs that are largely independent othe number and size o the transactions the client makes. At the level o a nancialinstitution, xed costs span a wide rangerom the brick-and-mortar branch net-work to computer systems, legal and accounting services, and security arrange-mentsand are independent o the number o clients served. Fixed costs also ariseat the level o the nancial system, including regulatory costs and the costs o pay-ment, clearing, and settlement inrastructure, which are, up to a point, indepen-dent o the number o institutions regulated or participating in the payment sys-tem. The resulting economies o scale at all levels make it unprotable to stay in thebusiness o nancial service provision unless the associated scale economies arecaptured in some orm.12

In addition to costs, the outreach in the supply o nancial services, especiallycredit and insurance services, is constrained by risks, particularly the risk o de-ault. The risks can be either contract specic or systemic. Systemic risk can be de-ned as risk that is nondiversiable within a given economy and that, as a conse-quence, aects all nancial contracts. Systemic risk typically stems rom highmacroeconomic uncertainty (refected in high infation and exchange rate volatil-ity), weaknesses in the contractual and inormational environment, or geographi-cal limitations. Regardless o its origin, systemic risk hinders the supply o nancialservices because it raises the deault probability or the loss, given deault, or allcontingent contracts written in a given jurisdiction. This leads to a higher cost orunds and, hence, a higher foor or the interest rate required to grant a loan, shortermaturities as risk increases with the loan horizon, or higher premiums to write in-surance policies. As systemic risk increases, it enlarges the set o borrowers andprojects that nd the cost o credit unaordable and are thus priced out o the


36/322


credit market. Similarly, this makes insurance policies unaordable or larger seg-

ments o the population.

Idiosyncratic credit risks are specic to individual borrowers or projects and

thereore are not correlated with systemic risk. As a result, the cost o nance and

the availability o credit or insurance services dier across debtors and projects

depending on the related dierences in idiosyncratic riskiness. Importantly, how-

ever, the ability o the lender to manage idiosyncratic risk is infuenced by the sys-

temic risk environment. Two actors are particularly important in explaining the

dierences in interest spreads across debtors (or a given type o loan) that are in-

duced by idiosyncratic risk: agency problems and limits to the diversication o

risks that are not related to agency problems. The rst can be linked to the lack o

inormation, but also to volatility, while the second can be linked to diseconomieso scale. Agency problems arise rom inormation asymmetries between debtors

and creditors, whereby a debtor is privy to relevant inormation about hersel and

her project that the creditor may not be able to secure or only at a prohibitively

high cost; this can lead to two conceptually distinct sources o credit risk: adverse

selection and moral hazard.13 The ormer reers to higher interest rates that attract

riskier borrowers and projects, while the latter reers to the borrowers incentive to

use the proceeds o the loan in endeavors that are riskier than those specied in the

credit contract, while concealing this behavior rom the creditor.

A lack o scale and a lack o tools to deal with idiosyncratic and systemic risk canlimit the capacity o nancial systems to eectively serve the host economy, oster

growth, and reduce poverty. However, the characteristics o the host economy pro-

vide the backdrop beore which nancial institutions and markets have to operate.

Arican economies are characterized by several adverse circumstancespointed

out by Honohan and Beck (2007)that make it more dicult to overcome the

two market rictions o size and risk, as ollows:

The small scaleo many economies does not allow nancial service providers

to reap the benets o scale economies. The small size o Arican economies,as shown in gure 1.3, is driven by the low income level across the continent,

but also by the small size o countries. The limited demand or savings, insur-

ance, credit, or even simply payment transactions means that large parts o the

population o Arican economies are not commercially viable customers. The

dispersed populations in many Arican countries means that nancial service

provision outside urban centers is not cost-eective. Despite the increasing

trend toward urbanization, large parts o populations in Arica still live in rural

areas. The small size o nancial systems does not allow nancial institutions to

recover the xed costs o basic systems and might undermine competition i thesystem does not sustain more than a small number o institutions.

As documented in gure 1.4, large parts o the economy and a large share o all

economic agents operate in the inormal sectorand do not have the necessary

ormal documentation, such as enterprise registration, land titles, or even ormal


37/322


addresses. This increases the costs and risks or nancial institutions and ex-cludes large segments o the population rom ormal nancial services.

Volatilityon the individual and aggregate levels increases costs and underminesrisk management. At the individual level, volatility is related to inormality and

Source:World Bank (2010a).

Note:Sample size: 188 countries. GDP is measured in current U.S. dollars.

30

20

10

0

log

(GDP)

countries

rest of the worldAfrican countries

Figure 1.3 GDP across Countries, 2009

Source:Schneider, Buehn, and Montenegro (2010).

Note:Sample size: 145 countries.

80

60

40

20

0shareofpopulationworkinginthe

in

formalsector

countries


Figure 1.4 The Informal Economy across Countries, 2007


38/322


the consequent fuctuations in the income streams o many microenterprisesand households. This means these agents are less attractive or nancial institu-tions. At the aggregate level, volatility reers to the dependence o many Aricaneconomies on commodity exports, which makes economies vulnerable to thelarge price swings characteristic o commodities (see map 1.1). Volatility at theaggregate level also reers to political and social unrest, rom which Arica hassuered over the past 50 years o independence. Volatility increases the costs, butespecially the risks aced by nancial institutions and markets.

Governanceproblems continue to plague many private and government institu-tions throughout the continent and undermine not only the market-based pro-

vision o nancial services, but also reorm attempts and government interven-tions aimed at xing market ailures. These governance challenges are widespreadand aect many nancial institutions, ranging rom banks, micronance institu-tions, and cooperatives to government institutions, including development -nance institutions. Governance problems have been at the root o many nancialcrises on the continent. They also aect directly the ability o nancial institu-

Source:Gylason (2010).

Map 1.1 Reliance on Natural Capital across the World

< 15%

15%30%

30%50%

50%65%65%80%

> 80%

Share of naturalcapital over totaltangible capital (%)

IBRD 38788


39/322


tions and markets to manage idiosyncratic and systemic risks. The governancechallenge and the related agenda contain a large number o dimensions, rangingrom political stability and accountability in the control o grat to the rule olaw. Along all these dimensions, Arica ranks signicantly below other countries,although there are positive examples o reorms (gure 1.5).

Not all countries show these our characteristics. Arica has large economies,such as Kenya and Nigeria. Middle-income countries such as Mauritius and SouthArica are able to benet rom their higher income levels in terms o scale econo-

mies. The prevalence o the inormal economy is signicantly lower in the middle-income countries o North Arica than in Sub-Saharan Arica. Noncommodity ex-porters are subject to much less volatility than commodity exporters, and countriessuch as Botswana and Mauritius have shown signicant and increasing levels ogovernance. However, many countries across the continent are aected by at leastone o these characteristics, which make them less receptive to hosting a thrivingnancial system.

In analyzing the long-standing challenges described above, the book draws at-tention to three recent trends and phenomenaglobalization, regional integra-

tion, and technologythat oer new solutions, but also represent new challenges.

Globalization: Integration in international nancial markets has been an impor-tant, but controversial aspect o nancial sector policy throughout the world inpast decades and even more so since the recent crisis. While most Arican coun-

Source:Kauman, Kraay, and Mastruzzi (2009).

Note:Sample size: 197 countries.

2

1.5

1

0.5

0

0.5

1

1.5

2

co

mpositegovernanceindicator

countries


Figure 1.5 Governance across Countries, 2008


40/322


tries have opened up their nancial systems to the entry o oreign banks, capitalaccount restrictions are still in place in many countries, although oten more de jure than de acto. Capital account liberalization has long been considered animportant component o the modernist agenda o the Washington Consensus(Rodrik 1998). Yet, the crisis experience in East Asia and other emerging marketsin the 1990s has led to a more cautious approach. The underlying economicmodel has also aced severe criticism: cross-country comparisons do not yieldconsistent results on the benets o capital account liberalization. In addition,there may be a threshold value in economic, institutional, and nancial develop-ment below which countries do not benet rom capital account liberalizationbecause capital infows critically depend on nancial markets and institutions.14Foreign investors need capital markets to invest in equity or debt, unless theycreate subsidiaries or joint ventures. Even portolio investors need institutions ormarkets in which to invest. Governance is thereore important so that countriesmay reap the benets o international capital fows, as shown by the example oresource-based economies in which the corresponding capital infows are notalways properly accounted or or used in the public interest. Overall, a cautiousapproach is called or that ocuses more on long-term capital infows (oreigndirect investment) rather than short-term portolio fows and that imposes ad-ditional saety lines on macroeconomic management. This debate has gainedresh signicance in light o the current attempt o emerging markets to use cap-ital fow restrictions to counter the negative repercussions o the capital infowsrom developed markets that are a consequence o the application o quantita-tive easing policies.

Regional integration: Regional integration has been on the agenda o Aricanpolicy makers since many countries achieved political independence. The suc-cessul example o Europe in creating a large regional market with a joint cur-rency, joint institutions, and coordinated policy making has been inspiring,although the recent euro crisis may have dampened the enthusiasm somewhat.

Prima acie, there is an enormous potential or Arica in overcoming scale dis-economies by coming together. Not surprisingly, there have been numerousattempts at such cooperation. However, the results have been limited. Apartrom three currency unions, two joint bank regulation and supervision au-thorities, a joint insurance regulatory authority, and two regional stock ex-changes, most eorts have o

Financing Africa Through the Crisis and Beyond-2011

Documents

Transcript of Financing Africa Through the Crisis and Beyond-2011