Financing Public Infrastructure in Sub-Saharan Africa ... · recognized the importance of scaling...

BACKGROUND PAPER 15 (PHASE I)

Financing Public Infrastructure in Sub-Saharan Africa:

Patterns and Emerging Issues

Cecilia Briceño-Garmendia, Karlis Smits, and Vivien Foster

JUNE 2008

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

wb406484

Typewritten Text

71815 v1

© 2009 The International Bank for Reconstruction and Development / The World Bank

1818 H Street, NW

Washington, DC 20433 USA

Telephone: 202-473-1000

Internet: www.worldbank.org

E-mail: [email protected]

All rights reserved

A publication of the World Bank.

The World Bank

1818 H Sreet, NW

Washington, DC 20433 USA

The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Executive Directors of the International Bank for Reconstruction and Development / The World Bank or the governments they represent.

The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries.

Rights and permissions

The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly.

For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923 USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com.

All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street, NW, Washington, DC 20433 USA; fax: 202-522-2422; e-mail: [email protected].

About AICD

This study is a product of the Africa Infrastructure Country Diagnostic (AICD), a project designed to expand the

world’s knowledge of physical infrastructure in Africa.

AICD will provide a baseline against which future

improvements in infrastructure services can be measured, making it possible to monitor the results achieved from

donor support. It should also provide a better empirical

foundation for prioritizing investments and designing policy reforms in Africa’s infrastructure sectors.

AICD is based on an unprecedented effort to collect

detailed economic and technical data on African infrastructure. The project has produced a series of reports

(such as this one) on public expenditure, spending needs,

and sector performance in each of the main infrastructure

sectors—energy, information and communication technologies, irrigation, transport, and water and sanitation.

Africa’s Infrastructure—A Time for Transformation,

published by the World Bank in November 2009, synthesizes the most significant findings of those reports.

AICD was commissioned by the Infrastructure Consortium

for Africa after the 2005 G-8 summit at Gleneagles, which recognized the importance of scaling up donor finance for

infrastructure in support of Africa’s development.

The first phase of AICD focused on 24 countries that

together account for 85 percent of the gross domestic product, population, and infrastructure aid flows of Sub-

Saharan Africa. The countries are: Benin, Burkina Faso,

Cape Verde, Cameroon, Chad, Côte d'Ivoire, the Democratic Republic of Congo, Ethiopia, Ghana, Kenya,

Lesotho, Madagascar, Malawi, Mozambique, Namibia,

Niger, Nigeria, Rwanda, Senegal, South Africa, Sudan,

Tanzania, Uganda, and Zambia. Under a second phase of the project, coverage is expanding to include as many other

African countries as possible.

Consistent with the genesis of the project, the main focus is on the 48 countries south of the Sahara that face the most

severe infrastructure challenges. Some components of the

study also cover North African countries so as to provide a broader point of reference. Unless otherwise stated,

therefore, the term “Africa” will be used throughout this

report as a shorthand for “Sub-Saharan Africa.”

The World Bank is implementing AICD with the guidance

of a steering committee that represents the African Union,

the New Partnership for Africa’s Development (NEPAD),

Africa’s regional economic communities, the African Development Bank, the Development Bank of Southern

Africa, and major infrastructure donors.

Financing for AICD is provided by a multidonor trust fund to which the main contributors are the U.K.’s Department

for International Development, the Public Private

Infrastructure Advisory Facility, Agence Française de Développement, the European Commission, and Germany’s

KfW Entwicklungsbank. The Sub-Saharan Africa Transport

Policy Program and the Water and Sanitation Program

provided technical support on data collection and analysis pertaining to their respective sectors. A group of

distinguished peer reviewers from policy-making and

academic circles in Africa and beyond reviewed all of the major outputs of the study to ensure the technical quality of

the work.

The data underlying AICD’s reports, as well as the reports themselves, are available to the public through an

interactive Web site, www.infrastructureafrica.org, that

allows users to download customized data reports and

perform various simulations. Inquiries concerning the availability of data sets should be directed to the editors at

the World Bank in Washington, DC.

iii

Contents

Summary iv

Public infrastructure spending: the headlines v

The anatomy of public spending v

General government expenditure vii

Budget efficiency viii

The hidden cost of utilities’ inefficiencies ix

Emerging messages x

1 Motivation for this study 1

2 Description of tools: data and methodology, key definitions 5

General government expenditures sourced from budget documents 7

Expenditures of public, nonfinancial corporations, sourced from financial accounts 8

Introducing a country typology 10

3 Public infrastructure spending: headlines 11

Country efforts and purchasing possibilities 11

Anatomy of public infrastructure spending 13

4 General government expenditures 23

The macro outlook and its fiscal implications for infrastructure 23

Prioritizing within the infrastructure budget envelope 28

Budget efficiency: execution and maintenance 31

5 The hidden costs of utilities’ inefficiencies 39

6 Conclusions and policy implications 52

References 54

Appendixes 57

Appendix 1. Sector scope, functional classification 58

Appendix 2. Sources of data on infrastructure expenditures 59

Appendix 3. Country groups 59

Appendix 4. Primary fiscal balances 60

Appendix 5. Net change in central government budget: breakdown by source 61

Appendix 6. Net change in central government budget: breakdown by use 62

Appendix 7. Expenditure on main road network 63

Appendix 8. Variance around the trend line of road expenditure 64

Appendix 9. Contributions to QFCs (country aggregates)—water sector 65

Appendix 10. Contributions to QFCs (country aggregates)—power sector 66

Appendix 11. Water: efficiency and production indicators, 2006 67

Appendix 12. Electricity: efficiency and production indicators, 2006 69

Appendix 13. Annual maintenance and preservation expenditures, 2001–05 70

Appendix 14. Average annual maintenance expenditures on main road network, 2001–05 71

Summary

o be credible, any plan for scaling up infrastructure in Africa must rest on a thorough evaluation

of how fiscal resources are allocated and financed. Because in every plausible scenario the public

sector retains the lion’s share of infrastructure financing, with private participation remaining

limited, a central purpose of such an evaluation is to identify where and how fiscal resources can be better

used—if not increased—without jeopardizing macroeconomic and fiscal stability. The stakes are high,

because the magnitude of Africa’s infrastructure needs carries a commensurate potential for misuse of

scarce fiscal resources.

We analyze recent public expenditure patterns to identify ways to make more fiscal resources

available for infrastructure. We do this in three ways. First, we quantify the level and composition of

public spending on infrastructure so as to match fiscal allocations to the particular characteristics of

individual subsectors and to countries’ macroeconomic type (low-income fragile, low-income nonfragile,

oil-exporting, and middle-income). Second, we evaluate public budgetary spending for infrastructure

against macroeconomic conditions to get a sense of the scope for making additional fiscal resources

available based on actual allocation decisions in recent years. And, third, we look for ways to make public

spending for infrastructure more efficient, so as to better use existing resources.

The Government Finance Statistics of the International Monetary Fund are neither comprehensive nor

disaggregated enough to support an analysis of the fiscal costs of infrastructure for the period 2001–06.

For that reason, our analysis is based on a new, standardized cross-country dataset of fiscal indicators for

infrastructure that covers, but also extends beyond, spending from central government budgets. State-

owned enterprises (SOEs) and extrabudgetary financing vehicles are also covered, as are private

operators, as long as the assets they operate belong to the state or the operator continues to rely on public

subsidies. Expenditure by subnational jurisdictions is only partially covered, however. Data are collected

in such a way as to permit cross-classification by economic categories (including capital and current

spending) as well as functional categories—information and communication technologies (ICT), power,

roads, water, and sanitation. As far as possible, both budgeted and actual expenditures are recorded.

Any exercise of this kind encounters data limitations. First, because it was not feasible to visit all

subnational entities, some decentralized infrastructure expenditures probably have been underrepresented,

with particular implications for the water sector. Second, it was not always possible to fully identify

which items of the budget are financed by donors, and contributions by nongovernmental organizations

(NGOs) to rural infrastructure projects are likely to have been missed completely. Third, it was not

always possible to obtain full financial statements for all of the infrastructure special funds that we

identified. Fourth, accurate recording of annual changes in fixed capital formation (capital expenditure) of

SOEs remains a methodological challenge. Fifth, accurate measurement of existing public infrastructure

stock will require further methodological development.

T

FINANCING PUBLIC INFRASTRUCTURE IN SUB-SAHARAN AFRICA

v

Public infrastructure spending: the headlines

Most governments in Sub-Saharan Africa spend about 6–12 percent of their gross domestic product

(GDP) each year on infrastructure, understood as comprising ICT, power, roads, water, and sanitation

(figure A). Roughly half spend more than 8 percent of GDP, while only a quarter of countries spend less

than 5 percent, the level commonly encountered among the countries of the Organisation for Economic

Co-operation and Development. Cape Verde, Ethiopia, and Namibia spend well above 10 percent of their

GDP on infrastructure. In the few middle-income countries of the region for which comparative

information is available the level of public spending is known to be between 6 and 8 percent of GDP.

Expressed as shares of GDP,

these fiscal efforts seem larger than

when put in dollar terms. Most

countries of the region spend less

than $600 million a year on

infrastructure services—less than

$50 per person. Among landlocked

countries, whose infrastructure

needs tend to be particularly high,

the annual total is less than $30 per

capita. These annual expenditures

pale in comparison with the

amounts needed. An investment

budget of US$100 million

purchases no more than about

100 MW of electricity generation,

or 100,000 new household

connections to water and sewerage,

or 300 kilometers of two-lane

paved road.

The anatomy of public spending

Most public spending on infrastructure in Sub-Saharan Africa passes through SOEs. SOEs have a

particularly large role in the middle-income countries, where they account for over 70 percent of all

public infrastructure spending. In Namibia, for example, 90 percent of expenditures on infrastructure are

made by SOEs. In non-oil-exporting low-income countries, the share of expenditures realized by SOEs is

close to 60 percent, or just below two-thirds of total infrastructure spending.

The bulk of the fiscal resources that pass through SOEs go for current spending. Current spending

includes spending on operations and maintenance, which is essential to harness the economic returns of

capital. However, most of recorded current spending relates to so-called nonproductive expenses, namely

wages and salaries. High levels of recurrent spending may indicate that operational inefficiencies are

diverting resources away from investment.

Figure A. Fiscal flows devoted to infrastructure

-

2

4

6

8

10

12

14

16

18

20

Cote

d'Iv

oir

e

Rw

anda

Nig

eri

a

Cam

ero

on

Nig

er

Chad

Tanzania

Uganda

Benin

Madagasc

ar

Senegal

Mala

wi

Mozam

biq

ue

Zam

bia

Ghana

Kenya

Eth

iopia

Leso

tho

South

Afr

ica

Nam

ibia

Cape V

erd

e

Spen

din

g %

GD

P

-

100

200

300

400

500

600

Spen

din

g U

SD

per

cap

ita

GDP Share (%) Spending per capita

Source: Africa Infrastructure Country Diagnostic, Fiscal Baseline (2008).

Note: Based on annual averages for the period 2001–05.


vi

Governments are the

most prominent financiers

of infrastructure

investment in Sub-Saharan

Africa. Except in the

middle-income countries,

governments are

responsible for between

80–90 percent of total

capital investment,

consistently allocating at

least 80 percent of their

infrastructure budgets to

investment. In low-income

countries that are aid-

dependent or that export

oil, the prevalence of

governments as investors

is driven by their role in

channeling external funds

and/or natural resource

royalties. Most external

development funds are

earmarked by donors for

investment. The dominant

role of the central

government as an investor

is consistently found in

most subsectors:

accounting for 80 percent

of total public investment

in transport and water

supply, and about 40 percent in energy (figure B). The noticeable exceptions to this pattern are the ICT

sector and, as noted, the middle-income countries.

Even though capital budgets may fall far short of actual needs, on average, most countries are not able

to spend more than one-third of the budgeted amounts. For a number of countries we were able to

compare actual capital spending with the amounts originally budgeted. The budget execution ratios that

emerged ranged from 28 percent (Benin) to 89 percent (Madagascar), with the average being 66 percent.

This means that capital spending in the region might be 50 percent higher if only government agencies

had the capability to spend all of the resources allocated to them. The problems behind the low execution

rates include poor planning, deficiencies in project preparation, and delays in procurement. Budget

execution ratios for current spending are, on average, a little higher.

Figure B Public infrastructure spending by sector and institution

Investment

-

0.5

1.0

1.5

2.0

2.5

3.0

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

GD

P S

har

es

SOEs

Gral Govern't

Water TransportICTPower

Current Spending

-

0.5

1.0

1.5

2.0

2.5

3.0

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-NoF

ragil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-NoF

ragil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-NoF

ragil

e

LIC

-Fra

gil

e

GD

P S

har

es

SOEs

Gral Govern't

Energy ICT TransportWater

Source: AICD, Fiscal Baseline (2008).


vii

Transport and energy sectors together absorb the lion’s share of infrastructure spending—about 80

percent in low-income countries. The heavy spending on power is a response to the widely recognized

power crisis on the continent. The efforts of the middle-income countries to support energy development

contrast starkly in absolute spending terms with those of the poorer countries. Middle-income countries

spend almost 5 times more on power than do aid-dependent low-income countries. Actual spending for

water may be higher than shown here, because of difficulties in capturing spending data from municipal

water utilities.

Sectoral allocations differ markedly across different groups of countries. Aid-dependent countries

tend to show relatively high levels of investment in roads and water, which together account for 80–95

percent of donors’ allocations to infrastructure in the region. Funds from donors make up about 50

percent of water spending and 25 percent of roads spending. By contrast, donors’ commitments to the

energy sector have been low or inexistent in sharp opposition to the efforts of low-income countries that

by themselves have been allocating close to 25 percent of their public infrastructure budgets to power to

redress chronic underinvestment in that sector.

General government expenditure

For several years running, a favorable external environment (notably high commodity prices) and

sustained domestic economic growth averaging at least 4.5 percent annually have expanded the resources

available to the governments of Sub-Saharan Africa. The economies of oil-producing countries have

grown at the fastest pace (up to 15 percent a year), for obvious reasons. Non-resource-intensive countries

benefited from debt relief and successful policy reforms that offset the negative impacts of higher oil

prices. Even heavily indebted poor countries (HIPC) grew at an annual average rate of 5.5 percent.

Domestic revenues have been the largest source of additional funds for resource-intensive countries,

whereas external grants played the most significant role for the poorest countries in the region.

The favorable external environment helped many countries expand their budgets. In the period 2001–

05, Sub-Saharan governments’ budgets grew by almost 1.9 percent of GDP, with the regional average

driven largely by increases in middle-income countries (table A). Not all countries benefited, however.

Zambia’s budget contracted by more than 8 percent, while that of the Democratic Republic of Congo

chalked up a 9 percent increase.

The additional budgetary resources helped low-income aid-dependent countries to bolster capital

investments, including infrastructure. As a share of GDP, capital investment increased in the low-income

countries by more than 1 percent in 2002–05. About 40 percent of the additional resources were allocated

to clearly favored infrastructure sectors.

It is striking that the oil-exporters and middle-income countries decreased their investment despite

having more fiscal resources available. The oil-exporting countries lowered their capital expenditures on

average by 3.3 percent of GDP. In oil-exporting countries, the decrease in budgetary expenditure was

largely absorbed by a significant reduction in infrastructure expenditures. To a large extent this reflects

developments in Nigeria, where infrastructure expenditures decreased by 2.2 percentage points of GDP

during the study period. The middle-income countries appear to have chosen to devote more resources to


viii

maintenance. Most of their additional capital budget was allocated outside infrastructure, but not to health

and education, as the table shows.

Table A Net change in central government budgets by country group, financing source, and destination, 2001–06

% GDP

Financing sources Spending allocations

Country group

Net central government expenditure

budget

Of which domestic revenues

Of which donor grants

Of which infrastructure

Of which health and education

Middle-income 4.08 3.40 (0.03) 0.02 0.13

Oil-exporting (3.73) 5.25 (0.07) (1.43) (0.34)

Low-income, nonfragile 1.69 0.83 1.98 0.54 0.93

Low-income fragile 3.85 0.79 1.90 0831 0.43

Africa average 1.89 3.04 0.57 (0.14) 0.24

Source: AICD, Fiscal Database, 2008; IMF Statistical Appendixes, WB DDP.

Note: Averages weighted by national GDP. Totals may not add up.

— = data not available.

Budget efficiency

Infrastructure stocks in many of the region’s countries are sorely in need of rehabilitation after years

of poor maintenance. The percentage requiring rehabilitation ranges from 12 percent (Burkina Faso) to 48

percent (Democratic Republic of Congo)—the average for the survey group is 30 percent. Rehabilitation

needs are significantly higher for rural infrastructure (35 percent) than for other types (25 percent),

reflecting the difficulty of maintaining assets in isolated rural areas. Because rehabilitating assets is much

more costly (in present-value terms) than maintaining them well, the magnitude of the rehabilitation

backlogs indicates substantial inefficiency in lifecycle spending on infrastructure.

Maintenance is the most challenging aspect of road spending. In environments characterized by weak

fiscal management (nontransparent and politically dominated budget processes), assets often are

neglected. Because maintenance yields little observable immediate benefit and is easily deferred, its

budgetary allocations often are not protected by the executive or parliament. Furthermore, in Africa,

donors have a dominant role in channeling funds to the sector. They earmark much of their funding,

extended on concessional terms, for investment, which has the effect of making maintenance more costly

than investment, because most maintenance funds must be raised domestically. Although the share of

external financing that is allocated to road rehabilitation has increased in recent years, road spending in

Sub-Saharan Africa is dominated by new construction, leaving maintenance a secondary priority.

Roughly half of the countries in the sample have shortfalls of 40 percent or more in annual

maintenance. Expenditure shortfalls are greater than 60 percent in Chad, Uganda, and Niger. Countries

that have established well-functioning road funds tend to be more successful at maintaining their road

networks and reducing the volatility of spending.


ix

The hidden cost of utilities’ inefficiencies

Reducing inefficiencies in infrastructure operations is perhaps the most practical and realistic way of

making more resources available for infrastructure in the region. While most countries are devoting

considerable effort to improving infrastructure, they are severely constrained in what they can spend.

They have trouble raising domestic revenue and in reallocating revenue from other uses, which often

requires structural reforms. By contrast, efficiency improvements can quickly enlarge governments’

availability of funds, allowing them to provide new services. Because spending on infrastructure

consumes a significant share of GDP, even small efficiency gains can contribute large savings.

For electricity, water supply, and, to some extent, telecommunications, we measure inefficiencies by

quantifying their hidden costs. For the water and power sectors, hidden costs are estimated by using the

end-product approach. The methodology identifies three relevant quasi-fiscal activities in utilities:

underpricing (charging less than the economic cost of the good), undercollection (where bills are never

sent or allowed to go unpaid), and excessive unaccounted losses (to leaks or theft, for example). Hidden

costs are then estimated by

comparing actual indicators

of a functioning SOE

against ideal norms of cost-

recovery, collection ratios,

and distribution losses.

For telecommunications

utilities, we quantify the

hidden cost of labor

redundancies by comparing

partial labor-productivity

ratios of existing telecom

incumbents against world-

class fixed-line providers in

OECD countries.

Quasi-fiscal activities in

Africa represent average

annual hidden costs of the

following (minimum)

magnitudes: 0.5 percent of

GDP in the water sector

(figure C), 0.8 percent in

the power sector, and 0.3 in

the telecom sector. The

smaller economic size of

water utilities, together with

skewed coverage in the

Figure C Hidden costs for water and power utilities as share of GDP

Water

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

Tan

zan

ia

Nig

eria

Ben

in

Cap

e V

erde

Eth

iop

ia

Ugan

da

Nam

ibia

Ken

ya

Burk

ina

Fas

o

Rw

anda

Sudan

Nig

er

South

Afr

ica

Les

oth

o

Mo

zam

biq

ue

Cote

d'I

voir

e

Sen

egal

Mad

agas

car

Zam

bia

Mal

awi

Ghan

a

DR

CMispricing Unaccounted Losses Collection Inefficiencies

Power

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

So

uth

Afr

ica

Ben

in

Ken

ya

Mo

zam

biq

ue

Ch

ad

Cap

e V

erd

e

Mad

agas

car

Les

oth

o

Nig

eria

Bu

rkin

a F

aso

Rw

and

a

Eth

iop

ia

Ug

and

a

Cam

ero

on

Zam

bia

Tan

zan

ia

Sen

egal

Gh

ana

Nig

er

Mal

awi

DR

C C

on

go

Under-Pricing Unaccounted Losses Collection Inefficiencies

Source: Authors’ own calculations using data from the AICD Database.


x

sample because of decentralization and fragmentation, partially explains their lower hidden costs.

Underpricing is the main source of hidden costs in both power and water utilities. Not only is

underpricing inefficient, but the associated capital subsidies are hugely inequitable because access to

these services is skewed toward the better off, with substantial shares of the poor remaining unconnected

to the electrical grid and water supply network.

In middle-income countries, unaccounted losses stand out as the greatest source of inefficiency for

power utilities, particularly maintenance-deprived distribution networks. Aid-dependent countries show

slightly higher levels of hidden costs relative to their peers, largely because of mispricing, and, in the

water sector, poor collection practices. In the telecom sector, countries that have maintained state

ownership of telecommunications incumbents, thereby deterring competition, not only are forgoing future

tax revenues from expanded business activity but also are creating an additional burden of hidden costs

from inefficiency (usually a bloated workforce). Such costs can exceed 0.3 percent of GDP.

Emerging messages

The countries of the region are devoting substantial shares of their GDP to infrastructure (6–12

percent when all sources are taken into account), but that does not amount to much in absolute terms,

because the economies in question are small. On average, low-income countries are spending less than

$50 per capita per year, with public investment being only a fraction of this.

There is a marked division of labor between SOEs and central governments. While SOEs account for

the bulk of infrastructure spending in most countries, they undertake very little capital spending. Most

public investments for infrastructure continue to be made through central government budgets, with the

resulting assets often transferred to SOEs for subsequent operation and maintenance.

Despite a favorable budget environment, only aid-dependent countries seem to be allocating

additional resources to infrastructure. The combination of a commodity boom and widespread debt relief

has created substantial buoyancy in government budgets. In the case of aid-dependent countries, about 30

percent of the additional funds have been allocated to infrastructure. However, in middle-income

countries almost none of the additional resources gleaned from the recent good years have gone for

infrastructure. In oil-exporting countries infrastructure investment has actually fallen even as resource

revenues have surged.

Regardless of how windfall revenues are spent, governments in the region could substantially enlarge

their fiscal space by redressing inefficiencies in infrastructure psending. Three major sources of

inefficiency have been identified here: inattention to maintenance, failures to spend budgeted funds, and

hidden costs.

There is substantial direct and indirect evidence of undermaintenance, which leads to higher costs

over the infrastructure lifecycle. On average, almost a third of the infrastructure assets of the countries of

the region are in need of rehabilitation. With the present value of rehabilitating infrastructure exceeding

the cost of preventive maintenance, it is easy to see that, over time, countries are spending more than they

need to spend to preserve a fixed amount of infrastructure stock.


xi

Second, very low ratios of execution of capital budgets point the way to an easy and budget-neutral

increase in public investment—if only execution ratios can be raised. Addressing the causes of low

budget execution deserves very serious attention, as solving the problem could increase public investment

by 50 percent without any increase in budgeted resources. Moreover, until such deficiencies are addressed

it will remain difficult to achieve higher levels of investment, even if more external resources are injected.

Third, the hidden costs of utilities absorb some 1.8 percent of GDP, indicating a major potential

dividend in return for the right set of actions. Underpricing is by far the largest contributor to hidden costs

in power and water utilities, although, as noted, unzealous bill collection and distribution losses are also

important.

1 Motivation for this study

The coordinated efforts of African countries and the international community toward achieving the

Millennium Development Goals (MDGs) has drawn attention to an enormous funding challenge. The cost

of achieving the MDGs is estimated at 13 percent of average GDP in Sub-Saharan Africa (Sachs and

others, 2004). Looking beyond the MDGs, recent research1 estimates Sub-Saharan Africa’s aggregate

infrastructure needs—both new investment and operations and maintenance (O&M)—at $75 billion a

year for 2006–15, or 11.7 percent of average GDP (table 1.1). Estimates for power alone are about $43

billion a year, or 7 percent of GDP, half for the investments needed to overcome chronic shortages and to

propel trade in power. For low-income countries, infrastructure needs quickly add up to 20 percent of

GDP (in fragile states, this reaches an impossible 70 percent). For middle-income countries, investment

needs are two-thirds of O&M, while most low-income countries have investment needs 70–80 percent

higher than O&M costs.

Table 1.1 Sub-Saharan Africa infrastructure needs 2006–15, by sector

Water supply and sanitation Energy ICT Transport Total

US$ billion a year GDP share (%) Shares

Middle income 17.92 6.62 4.89 80.93 0.95 13.23 100.00

Oil exporting 18.73 8.97 16.84 41.97 3.14 38.05 100.00

LIC-nonfragile 24.15 21.40 16.87 48.42 3.54 31.17 100.00

LIC-fragile 16.38 42.92 10.96 56.99 2.34 29.71 100.00

Africa 74.90 11.69 13.39 56.90 2.57 27.14 100.00

Source: Africa Infrastructure Country Diagnostic, 2008

Note: Averages weighted by country GDPs. Totals may not add up because of rounding.

The private sector’s historically limited contribution to infrastructure provision and financing

underscores the importance of the public sector in meeting infrastructure needs, at least for the

foreseeable future. Despite efforts and good intentions, until recently the scale of private finance was not

as great as anticipated in the 1990s and early 2000s (up to 2006). Nor has it extended beyond the more

lucrative areas of infrastructure, such as telecommunications, power generation, railways, and ports—or

the larger and wealthier economies, such as South Africa, Nigeria, and Kenya. Since the late 1990s, a

number of Sub-Saharan African countries have raised private finance for traditionally state-funded

infrastructure. But this has amounted to roughly 0.8 percent of GDP per year—minuscule when compared

with the approximately 4 percent needed to fill the “infrastructure financing gap,” or, in other words, the

funds still needed after factoring in known financing sources (cost recovery, public expenditure, official

development assistance, and financing from the private sector and countries not in the Organisation for

1 Infrastructure needs estimates per sector within the Africa Infrastructure Country Diagnostic (2008).


2

Economic Co-operation and Development, OECD). Private sector contributions are also hindered by

stagnation in domestic capital accumulation, traceable to aggregate savings rates that are many times

lower than in other developing regions. Average saving rates from national income accounts are 11

percent in Sub-Saharan Africa, compared with 20 percent in Latin America and the Caribbean, 18 percent

in South Asia, 19 percent in the Middle East and North Africa, and 34 percent in East Asia and the

Pacific.

The characteristics that make infrastructure industries prone to government intervention—in the form

of public ownership, regulation, or both—are also widely accepted. From the supply viewpoint,

infrastructure service provision is characterized by large, fixed investments (usually site- and industry-

specific) and sharply increasing returns to scale. From the demand viewpoint, infrastructure services

range from those with the characteristics of private goods, such as telecommunications (for which prices

could be set efficiently by markets), to those that are closer to public goods, such as rural roads.

Shifting fiscal resources to and from infrastructure is not free of controversy. Easterly and Servén

(2003) point out the growing evidence that, in developing countries, fiscal stabilization has been achieved

by compressing productive public investment—notably in infrastructure—thus sapping the potential for

long-run economic growth. The authors propose a shift away from the short-term preoccupation with

fiscal deficits, toward a longer-term focus on fiscal solvency.2 But the International Monetary Fund (IMF,

2004) questions the assumption that public infrastructure spending in developing countries is necessarily

growth-enhancing, and suggests that efforts to make more resources available for infrastructure should

focus on reallocating spending within the current fiscal envelope. Recent empirical work finds that the

growth-enhancing effects of public investment in infrastructure are not always greater than those of public

investment in health and education (Estache and Muñoz, 2007).

Such controversy is attributable to the very nature of infrastructure, whose characteristics complicate

its treatment in public finances.

• Infrastructure investments are large, lumpy, and infrequent; they often take more than one budget

cycle to complete. They are therefore difficult to accommodate within a single budgetary cycle and

much better suited to a medium-term expenditure framework. In addition, budget allocations for

multiyear investment projects may not be sustained over time, thus delaying implementation and

reducing projects’ eventual rate of return. Even worse, interruptions in funding may leave a country

with a graveyard of incomplete public works that never materialize into a productive asset.

• Infrastructure assets require sustained preventive maintenance. Failure to maintain such assets

eventually leads to deterioration and the need for major rehabilitation, which costs considerably more

in present-value terms than does preventative maintenance. Nevertheless, deterioration is a gradual

process, and maintenance has low visibility, creating a permanent temptation to defer such spending

to accommodate more politically rewarding expenditures.

• Some efficiency gains from long-lived assets materialize only with time, as higher marginal returns

begin to appear. This applies to assets constructed according to 20–30 year demand projections there

2 At around the same time, Blanchard and Giavazzi (2003) advanced a similar argument with respect to the

European Union’s Stability and Growth Pact, which limited budget deficits to 3 percent of GDP and formed the

macro-economic underpinning of European Monetary Union.


3

delivered low returns early in their life cycle. Other efficiency gains depend on the successful

implementation of policy decisions aimed at a more efficient execution of multiyear projects (to

ensure that costs are kept to a minimum and that countries begin to reap their benefits as soon as

possible), and at making sure that periodic maintenance of relatively lightly used assets (such as new

roads) is not postponed.

The debate over the availability and allocation of fiscal resources has particular relevance for Sub-

Saharan Africa, which is faced with burgeoning infrastructure needs, limited private participation, and

meager public budgets. It is in this context that fiscal space becomes relevant. The IMF has defined

“fiscal space” as the budgetary room that allows a government to provide resources for a desired purpose

without imperiling its financial position.3 The concept focuses on the linkages between fiscal policy trade-

offs and the availability of resources in the medium-term to ensure fiscal sustainability.

There are several ways to make more fiscal resources available for development needs (including

infrastructure), utilizing both domestic and external sources. The Development Committee of the World

Bank Group (World Bank, 2006) has proposed a framework that defines four options for central

governments: (i) raise additional tax revenues, (ii) increase public sector borrowing, (iii) obtain more

international aid, and (iv) improve the efficiency of current expenditures. Making fiscal resources

available has an important intertemporal component, since effective use of resources today leads to

increased productivity, thus generating more resources to fund tomorrow’s policy choices. Countries that

make more fiscal resources available by cutting development expenditures may undermine long-term

growth, thereby restricting their fiscal space in the future. Aid-dependent countries face the challenge of

using donor financing to remove bottlenecks to growth and, in so doing, to expand their future fiscal

space.

This paper analyzes recent spending patterns to identify ways to increase the availability of fiscal

resources. We do this in three steps.

First, we quantify the level and composition of public infrastructure expenditures. By examining how

infrastructure spending is allocated across subsectors, institutions, and expense categories, we

characterize past patterns and levels of fiscal allocations in terms of subsector specificities and country

types (such as resource-rich, aid-dependent). This helps identify forward-looking options for fiscal

expansion and other financing alternatives

Second, we assess public spending on infrastructure against the background of overall fiscal resource

availability. The initial fiscal conditions for infrastructure spending are framed in a macroeconomic

context so as to give the reader a realistic sense of the scope for creating additional fiscal space by

increasing budgetary allocation. Fiscal resources are broken down into three observable aspects: revenue

effort, access to aid, and access to financial markets.

Third, we look for ways to increase the efficiency of public infrastructure spending. Changing the

composition and increasing the efficiency of public spending for infrastructure are two of the most

practical ways of increasing governments’ resource envelope. This study provides a cross-country

comparison of the efficiency of expenditures in the water, electricity, and communications sectors. In the

3 For details, see Heller and others (2006).


4

water and electricity sectors, potential efficiency gains are quantified by measuring the hidden costs of

state-owned service providers.

To achieve our purpose, we assembled a detailed database of public spending flows, the absence of

which had prevented systematic evaluation of the quality and impact of public spending in infrastructure.

2 Description of tools: data and methodology, key

definitions

The Government Finance Statistics (GFS) compiled by the International Monetary Fund (IMF) do not

provide a detailed or comprehensive picture of infrastructure spending. At present, the GFS constitutes

the main source of cross-country data on public finance. But the information on infrastructure presents a

number of problems, particularly for Africa. First, the GFS focuses on tracking general government

expenditure, whereas a large share of infrastructure spending passes through nonfinancial public

corporations (parastatals). Second, even within the category of general government spending, the GFS is

limited in practice to central government spending, with little reporting of subnational and special

funds—two other important channels of infrastructure spending.4 Lastly, the GFS does not break down

infrastructure spending by subsector or expense category.

In response to these limitations, we built a new database of standardized cross-country data that seeks

to give a detailed yet comprehensive picture of infrastructure spending.5 Our analysis is based on a

systematic, cross-country study of public spending on infrastructure both within and beyond the bounds

of central government budgets. The data-collection process is based on a standardized methodology

developed and explained in detail by Briceño-Garmendia (2007). To ensure the cross-country

comparability of the data, detailed templates guided the data-collection process in the field.

The methodology is designed to be comprehensive in the sense of covering all relevant budgetary and

nonbudgetary areas of infrastructure spending. The collection of data on fiscal spending was grounded in

an overview of the institutional framework for delivering infrastructure services in each of the countries

while aiming at identifying all of the channels through which public expenditure on infrastructure flows.

The work began with a detailed review of the central government budget. Thereafter, financial statements

were collected from all the parastatals and special funds that had been identified in the institutional

review. In countries where infrastructure service providers are highly decentralized (as in the case of

municipal water utilities), it proved possible to collect financial statements only from the three largest

service providers. Privatized infrastructure service providers were included if a majority of their shares

remained government owned, or if they continued to depend on the state for capital or operating

subsidies. Thus, telecommunications incumbents are typically included, whereas mobile operators are not.

In some countries, local governments have begun to play an increasing role in infrastructure service

provision. It was not possible to collect comprehensive expenditure data at the local government level.

However, in some cases the central government produces consolidated local government accounts. Where

these do not exist, an alternative source of information is the fiscal transfers from central to local

governments, which are reported in the budget and on which local governments rely heavily, given

4 Based on IMF (2001), the public sector can be roughly divided into general government and public corporations.

General government comprises central, state, and local governments. Public corporations can be grouped, according

to the nature of their activities, into financial corporations (engaged in providing financial services for the market)

and nonfinancial corporations (engaged in producing goods and nonfinancial services). 5 Soon available online at www.AfricaInfrastructure.org.


6

limited alternative sources of revenue. In some cases, transfers are earmarked for infrastructure-related

spending, whereas in others the share allocated to infrastructure could only be estimated.6

Data were collected in such a way as to permit both classification and cross-classification by

economic and functional categories. That is, a matrix was established so that spending on each functional

category could be decomposed according to the economic nature of the expense, and vice versa.

Functional classification followed as closely as possible the 4-digit category or class level of the

functional classification (COFOG) proposed in the IMF’s Government Financial Statistics Manual 2001

(GFSM, 2001),7 which allowed us to identify all the major infrastructure subsectors.8 The economic

classification of expenses also followed the IMF framework, permitting us to distinguish to some extent

between current expenditures, capital expenditures, and various subcategories thereof.9

The institutional scope of the study builds on the definition of the public sector spelled out in the

system of national accounts described by the United Nations (1993), that is, all units of general

government and all public corporations. The system of national accounts defines public corporations as

those in which the public authorities are considered the owners by virtue of owning all, or a majority of,

shares, equity, or other form of capital (table 2.1).10

Table 2.1 Institutional distribution of gross national expenditure

General government Financial corporations

Nonfinancial corporations

Nonprofit institutions serving household sector

Household sector

B. Expenditures of public financial corporations

D. Expenditures of public nonfinancial corporations

of which expenditures on infrastructure

A. Public expenditures

of which public expenditures on infrastructure

C. Expenditures of private financial corporations

E. Expenditures of private nonfinancial corporations

F. Expenditures of nonprofit institutions serving households sector

G. Household current and capital expenditures

Source: United Nations, 1993.

Note: Shaded area reflects expenditures of public sector.

6 For a quick summary of sources of data and information, see table 2.1. 7 Definitions and explanations of the infrastructure cost elements figuring in the database can be found in Briceño-

Garmendia (2007). 8 The main categories covered in the study are electricity (0435), road transport (0451), water transport (0452),

railway transport (0453), air transport (0454), pipeline and other transport (0455), communication (0460), waste

water management (0520), and water supply (0630). Irrigation spending is estimated as a share of agriculture

(0421). 9 Current expenditures are broken down into compensation of employees, use of goods and services, consumption of

fixed capital, interest, subsidies, grants and transfers, social benefits, and other current expenditure. Capital expenditures are broken down into buildings, structures, machinery, and equipment; other fixed assets; other capital

expenditure: and transfers of capital expenditures to lower levels of government. 10 See United Nations (1993) for details. The system of national accounts also states that enterprises in which the

government holds less than half of the shares may still be classified as public corporations if the government

controls the business by influencing all principal aspects of management.


7

General government expenditures sourced from budget documents

As far as possible, both budget estimates and actual expenditures were recorded for the period 2001–

05. There are usually three stages in the public expenditure process. First, resources are budgeted for

particular purposes. Second, funds are released from the Ministry of Finance (MoF) to the responsible

institutions. Finally, resources are spent by the recipient institutions. Expenditure patterns can differ

substantially across these three different stages. While actual spending is the variable of greatest interest,

and the main focus of the results presented in this report, it is also important to understand how

infrastructure spending is affected by the budget execution process. To this end, both budgeted and actual

expenditures are recorded wherever possible. Attempts were also made to capture release figures—but

with very limited success.

For countries that have not yet fully implemented GFSM 2001 standards, it was necessary to perform

a line-by-line recoding of the budget in consultation with the corresponding line ministries. In several

countries, national budget expenditures are grouped according to programs for each line ministry (such as

in Zambia). In these cases, expenditures had to be disaggregated according to infrastructure subsectors.

The advantage of using GFSM 2001 functional codes is that they reflect the purpose of the expenditure in

a consistent pattern that is not distorted, for example, by changes in institutional responsibilities over

time. Functional areas can be spread across many institutions and are subject to reallocation from one

institution to another, contingent on changes over time in institutional responsibilities and institutional

frameworks.

It is difficult to ensure that expenses are analyzed and classified according to their economic use. On

the one hand, there is no clear-cut separation between what should be considered a capital expense, a

rehabilitation expense, or even a maintenance expense. So, capital and current expenditures can be

mistakenly accounted under inappropriate budget categories by the lack of clear definition of the

expenditure’s nature. There may also be deliberate misclassification done in order to increase the chances

of budgetary approval.11 As much as possible, the economic classifications used in each country were

remapped in accord with the GFSM 2001 framework to develop a common understanding across

countries.

Efforts were made to avoid double-counting of transfers across levels of government. There is a

danger that transfers from the central government to parastatals, special funds, and subnational

governments may be reported twice: once as a central government transfer and once as an expenditure by

the recipient institution. Great care was taken to match up these line items across institutions and ensure

that they were counted only once—as an expenditure made by the recipient institution. Similar care was

taken to eliminate double-counting subsidies that originated as central government transfers to state-

owned enterprises (SOEs).

11 Unfortunately, the GFMS 2001 leaves many economic categorizations open to interpretation, particularly with

regard to the uses of the expenses that are most relevant to infrastructure services (for example, rehabilitation,

operations, and maintenance). In this first report, the analysis will be limited to establishing the broad distinctions

between current and capital expenses.


8

Expenditures of public, nonfinancial corporations, sourced from financial

accounts

Creating standardized expenditure data for public nonfinancial corporations requires using principles

established in the system of national accounts. Data on the expenditures of public nonfinancial

corporations are very heterogeneous, so this component of the data set was essentially built from scratch.

The challenge is to measure the sources and uses of parastatals’ income and changes in fixed assets

(focusing on current and cumulative accounts) in a way that permits both classification and cross-

classification by economic and functional category. According to the system of national accounts,

corporations’ transactions can be grouped into current accounts (production, distribution of income, use

of income), accumulation accounts (changes in assets and liabilities, changes in net worth), and balance

sheets (stocks of assets and liabilities, and net worth). Recurrent expenditure flows can be determined by

a relatively straightforward classification of entries in the annual financial statement accounts.

Standardizing capital expenditures is not a straightforward exercise. According to the system of

national accounts, investments can be classified as changes in financial or nonfinancial assets. In this

report, however, we focus exclusively on changes in nonfinancial assets. In particular, we seek to measure

tangible nonfinancial assets, excluding inventories, consumption of own capital, and production of other

intermediary goods (table 2.2).

Notwithstanding these efforts, it is important to recognize the inevitable data limitations in any

exercise of this kind. These limitations should be borne in mind when interpreting the results of the

analysis.

First, it is likely that decentralized infrastructure expenditures may be improperly covered, with

particular implications for the water sector. The study does not adequately capture spending by small,

decentralized service providers (for example, municipal water utilities), or spending by local governments

that is not funded through fiscal transfers. This

concern does not affect countries with highly

centralized administrations or national water

utilities (such as Benin, Cape Verde, Madagascar,

Rwanda, and Uganda). However, in some

countries, local utilities and subnational

governments have begun to play an increased role

in infrastructure services.

Second, it proved difficult to fully identify

externally financed public infrastructure

expenditures, since not all donor aid is reported in

national budget accounts. Aid disbursements are

most often channeled through the central government (via government agencies or line ministries) but in

several cases might bypass government agencies, leaving no accounting record for the central

government. An example of a direct expenditure is funds given directly to nongovernmental organizations

to support small-scale infrastructure projects in rural communities. Official government budget accounts

would accurately measure only funds that are channeled through the central government, leaving a

Table 2.2 Classification of assets in the system of national accounts

Gross fixed capital formation

Consumption of fixed capital

Changes in inventories

Acquisitions less disposals of valuables

Nonfinancial assets

Acquisitions less disposals of nonproduced, nonfinancial assets

Financial assets

Source: United Nations (1993).

Note: Shaded area includes a subset of assets that will be used to measure capital expenditure flows.


9

significant proportion of other external funds uncounted. For example, in aid-dependent Rwanda only 50

percent of donor aid is reflected in the national budget.

Data on the execution of the budget—money actually spent out of or over budget estimates—are not

available in all countries.12 In some countries, we had problems accounting for the execution of foreign-

financed expenditures.13

Third, it was not always possible to obtain full financial statements for infrastructure special funds

(for example, rural electrification funds or privatization funds, which have often been used for public

investment). We found that the level of transparency and accountability in the spending of infrastructure

special funds was highly variable across countries and sectors. In a significant number of cases, it was not

possible to obtain financial statements for these funds. However, it should be noted that this problem

generally did not apply to road funds. Only in Côte d’Ivoire was it impossible to obtain a financial

statement on the road fund.

Fourth, capturing annual changes in fixed capital formation (capital expenditure) of SOEs using

financial statements represents a methodological challenge. For one thing, financial statements often do

not distinguish clearly between expenditure on financial and nonfinancial assets, making it very hard to

measure actual capital expenditures. In some cases, financial statements lacked information on levels or

changes in spending on assets. Review of investment plans might provide a better understanding of

capital expenditures, but that would involve considerably more effort, and it would be very difficult to

make a standardized comparison across countries, since these investment plans do not always accurately

measure actual expenditure levels. Secondly, while SOEs are “off budget,” important transactions

between public budgets and these entities nevertheless occur with regularity, particularly with regard to

on-lending, capitalization, debt swaps, transfers of already built assets, and various kinds of operating

subsidies. Examples exist of public subsidies made to utilities to finance capital spending but recorded as

current spending, as well as examples of the opposite: capital infusions that are used to pay wages or

other current costs.

Fifth, there are serious constraints in supplementing expenditure flows with statistics on asset stocks.

For the purpose of this study, expenditure flows may be sufficient to give us a basic understanding of

short-term fiscal sustainability and efficiency issues. However, for a more thorough analysis of

infrastructure expenditures and growth, or to control for initial conditions in efficiency analyses, it is of

the utmost importance to be able to account for public infrastructure stocks. Accurate measurement of

existing public infrastructure stocks remains a very challenging task and a natural extension of the

exercise presented here.

12 Proxies for actuals are used in Zambia (estimates), Côte d’Ivoire (releases), Senegal (releases), Mozambique

(estimates), and Nigeria (estimates). 13 Proxies for externally financed actual expenditures are used in Ghana (estimates), Lesotho (estimates), Tanzania

(estimates), and Benin (releases).


10

Introducing a country typology

This report organizes the Sub-Saharan country sample into four nonoverlapping groups, closely

following IMF (2007): oil-exporting, low-income nonfragile, low-income fragile, and (non-oil-exporting)

middle-income countries (table 2.3).14

Table 2.3 Typology of countries

Alternative classification for low-income

Oil-exporting countries

Low-income

nonfragile

Low-income

fragile

Middle- income countries

Low-income

coastal

Low-income

landlocked

Angola

Cameroon

Chad

Congo

Equatorial Guinea

Gabon

Nigeria

Sudan*

Benin

Burkina Faso Ethiopia Ghana

Kenya

Madagascar

Malawi

Mali

Mozambique Niger

Rwanda Senegal Tanzania Uganda Zambia

Burundi

Central Africa Rep.

Comoros

Congo, Dem. Rep. of

Côte d’Ivoire

Eritrea

Gambia,The

Guinea

Guinea-Bissau

Liberia

Sao Tome and Principe

Sierra Leone

Togo

Zimbabwe

Botswana

Cape Verde

Lesotho

Mauritius

Namibia,

Seychelles

South Africa

Swaziland

Benin

Comoros

Côte d’Ivoire

Eritrea

Gambia, The

Ghana

Guinea

Guinea-Bissau

Kenya

Liberia

Madagascar

Mozambique

Sao Tome and Principe

Senegal

Sierra Leone

Tanzania

Togo

Burkina Faso,

Burundi

Central Africa Rep.

Congo, Dem. Rep. of

Ethiopia

Malawi

Mali

Niger

Rwanda

Uganda

Zambia

Zimbabwe

Source: Classification as proposed in IMF (2007), except for Sudan, which was added for completeness.

We also explored an alternative classification as proposed by Collier and O’Connell (2006), who

argue that countries’ economic behaviors are strongly affected by their endowment of natural resources

and geography. Endowment of natural resources can so define a country ‘s potential that we classify oil-

exporting countries as such even if they would otherwise qualify for another group according to their

income or geography. Geography is reflected in the landlocked/coastal split. The landlocked condition

increases the demand for—and importance of—transport services,15 as well as their costs. Higher

transport costs have a usually unaccounted negative effect on private stock creation, growth outcomes,

and returns of capital.16

14 See annex 3 for definitions. 15 See Ndulu (2004). 16 See Agénor and Moreno (2006), Limão and Venables (2001), and Collier and Gunning (1999).

3 Public infrastructure spending: headlines

Country efforts and purchasing possibilities

The vast majority of African countries have annual public expenditures on infrastructure17 in the order

of 6–12 percent of GDP. Roughly half of the countries spend the equivalent of more than 8 percent of

their GDP, while in three out of four African countries infrastructure public spending accounts for more

than 5 percent of GDP.18 Only a quarter of countries spend less than 5 percent of their GDP on

infrastructure (figure 3.1). By comparison, in most OECD countries, annual public expenditure on

infrastructure is less than 5 percent of GDP. The level is 6.4 percent in Chile, 5.7 percent in Turkey, and

7.6 percent in Indonesia.

Large fiscal efforts, as expressed in terms of share of GDP, are incommensurate with the meager

amount of infrastructure services they can provide. In dollar terms, the majority of countries spend less

than $600 million on infrastructure services—less than $50 per person per year—or less than $10 per

person per year in investment. Annual expenditures pale when compared to the amounts needed. An

investment budget of $100 million could purchase around 100 MW of electricity generation, or 100,000

new household connections to water and sewerage, or 300 kilometers of a two-lane paved road.

Figure 3.1 Public spending on infrastructure across Africa (frequency for sample)

0

10

20

30

40

50

60

below 5 % 5 - 8 % over 8%

As a % of GDP

0

10

20

30

40

50

60

70

80

below 600 m$ 600-2000 m$ over 2000 m$

In m$ per year

0

10

20

30

40

50

60

70

80

below 50 $ 50-400 $ over 400 $

In $ per capita

Source: Africa Infrastructure Country Diagnostic, Fiscal Database (2008). Based on data presented in annex 1, table 1 (separately bound).

African governments differ greatly in their ability to afford the sums they spend for infrastructure

(figure 3.2). The countries of Sub-Saharan Africa are faced with a dilemma. The tax base in most

countries is too narrow to finance the most obvious infrastructure needs and too narrow to support

subsidized access for poor consumers (Estache, 2007). That base cannot be increased, however, without

economic growth, which in turn depends on the elimination of bottlenecks caused by inadequate

infrastructure.

17 Consistently with the outlined methodology, infrastructure is understood as these 4 sectors: water and sanitation,

power, transport (including roads) and ICT. 18 Excluding expenditures by local governments.


12

Figure 3.2 Fiscal flows devoted to infrastructure

-

2

4

6

8

10

12

14

16

18

20

Cote

d'Iv

oir

e

Rw

anda

Nig

eri

a

Cam

ero

on

Nig

er

Chad

Tanzania

Uganda

Benin

Madagasc

ar

Senegal

Mala

wi

Mozam

biq

ue

Zam

bia

Ghana

Kenya

Eth

iopia

Leso

tho

South

Afr

ica

Nam

ibia

Cape V

erd

e

Spen

din

g %

GD

P

-

100

200

300

400

500

600

Spen

din

g U

SD

per

cap

ita

GDP Share (%) Spending per capita

Total Fiscal Spending

GDP share ( percent)

Spending per capita

($)

MIC 8.86 550.51

Oil-exporting 5.44 76.15

LIC–nonfragile 8.74 34.95

LIC–fragile 3.44- 46.56

LIC–coastal 8.06 46.92

LIC–landlocked 8.39 27.36

Totals 24–country sample 7.74 231.74

Totals, excluding South Africa 6.74 39.70

Totals excluding South Africa and

Nigeria 7.93 43.11


Note: Based on annual averages for the period 2001–05. Detailed data presented in annex 1, tables 1 and 2 (separately bound).

In low-income countries, fiscal efforts ranging from 4 percent of GDP in Rwanda and Côte d’Ivoire

to more than 10 percent of GDP in Ethiopia translate into annual spending of less than $45 per capita.

These averages mask enormous differences between countries and country groups. The lowest level of

per capita spending on infrastructure is found in Rwanda (about $8 per capita per year). The lowest per

capita expenditures on infrastructure are found in landlocked countries, which spend, on average, $28.

Low-income countries, of course, have narrow tax bases and are generally unable to attract significant

private investment to crowd out the lack of public resources.

In middle-income countries (including South Africa), public infrastructure spending is about ten times

higher in per capita terms than in low-income countries. Middle-income countries spend roughly $550 per

year per person, while for low-income countries (including both oil-exporting and nonfragile states) this

is roughly $45 per year. The disparities are more shocking at the country level. Rwanda’s per capita

spending is 3 and 7 percent of total spending in Namibia and Cape Verde, respectively, while Niger and

Malawi only managed to spend 2.5 percent of what South Africa does.19

A country’s income level is highly correlated with public infrastructure investment (figure 3.3). The

emerging results for Africa simply confirm assumptions. As a ratio, public infrastructure investment is

negatively associated with GDP per capita, while per capita infrastructure spending is positively related

19 Differences in per capita spending levels shrink when expenditures are adjusted for differences in purchasing

power parity. (Poor countries tend to have higher purchasing power.) Expenditure levels measured in purchasing

power parity (PPP) dollars per capita are more or less the same across different country groupings: resource-

intensive countries, non-resource-intensive countries, and aid-dependent countries.


13

with GDP per capita. These relations have been documented in the literature and essentially underscore

the endogenous nature of investment and wealth.

Figure 3.3 Infrastructure investment and country income

Zambia

Uganda

Tanzania

Senegal

Rwanda

Nigeria

Niger

Namibia

Mozambique

MalawiLesotho

KenyaGhana

EthiopiaCote d'Ivoire

ChadCameroon

Benin

-

10.0

20.0

30.0

40.0

50.0

60.0

- 1,000 2,000 3,000 4,000 5,000 6,000

GDP per Capita ($)

An

nu

al

Pu

bli

c I

nfr

ast

ructu

re I

nv

est

men

t

--U

S$

per c

ap

ita

--

ZambiaUganda

Tanzania

South Africa

Senegal

Rwanda

NigeriaNiger

Namibia

Mozambique

Malawi

Madagascar

Lesotho

Kenya

Ghana

Ethiopia

Cote d'Ivoire

Chad

Cameroon

Benin

-

1.0

2.0

3.0

4.0

5.0

6.0

- 1,000 2,000 3,000 4,000 5,000 6,000

GDP per Capita ($)

An

nu

al

Pu

bli

c I

nfr

ast

ructu

re I

nv

est

men

t

--G

DP

Sh

ares--


Low-income countries, so to speak, are doubly penalized in their possibilities for buying

infrastructure services. On the one hand, the association between infrastructure spending and national

income does not imply causality, but it does confirm that poor countries are severely limited by their own

purchasing power. On the other hand, low-income countries are more ill-equipped than middle-income

countries in existent, functioning infrastructure assets. This suggests that even if low-income countries

counted with the quantity and quality of infrastructure that middle-income countries have, they would be

at a disadvantage when trying to move forward. But they are not even there. That is why the fiscal efforts

of individual countries will not go far unless other players such as traditional donors, non-OECD

financiers (such as India, China and Gulf States) and the private sector act coordinately.

Anatomy of public infrastructure spending

Institutional background

On aggregate, state-owned enterprises (SOEs) channel the largest share of public infrastructure

expenditures in Sub-Saharan Africa. SOEs have a particularly large role in the middle-income countries,

where they account for over 70 percent of all public spending on infrastructure. In Namibia, which has the

highest absolute and relative expenditure level, 90 percent of expenditures on infrastructure are made by

SOEs. In non-oil-exporting, low-income countries, the share of expenditures realized by SOEs is close to

60 percent or just below two-thirds of total infrastructure spending (figure 3.4).

Large differences between expenditure levels are driven, to a large extent, by differences in the

spending levels of SOEs. This is a very important finding from the fiscal viewpoint for their implications

for institutional development, fiscal balances, and infrastructure planning. First, SOEs’ main stakeholder

and lender of last resort is the central government. Thus, while SOEs are expected to partially or totally


14

recover costs, they don’t, and their net operating positions affect the country’s fiscal balances when

accumulated debts trigger bail-outs, capitalizations, or simply debt-swaps between the government and

the SOEs. Secondly, SOEs are expected to function on a quasi-commercial basis, playing a key role in

matching infrastructure supply with country needs as dictated by the nature of business, carrying out most

of the investment, particularly if they handle the highest shares of public infrastructure spending and

because they are charging user fees that should cover the enterprise’s capital and operations and

maintenance (O&M) needs.

Figure 3.4 Public infrastructure spending by institution

-

2

4

6

8

10

12

14

16

18

20

Cote

d'Iv

oir

e

Rw

anda

Nig

eri

a

Cam

ero

on

Nig

er

Chad

Tanzania

Uganda

Benin

Madagasc

ar

Senegal

Mala

wi

Mozam

biq

ue

Zam

bia

Ghana

Kenya

Eth

iopia

Leso

tho

South

Afr

ica

Nam

ibia

Cape V

erd

e

GD

P S

ha

res (

%)

Nonfinancial Public Institutions

On-Budget

Composition of fiscal spending

Shares on total ( percent) On-budget

Nonfinancial public

institutions

MIC 19.3 80.7

Oil exporting 52.5 47.5

LIC–nonfragile 36.9 63.1

LIC–fragile 25.4 74.6

LIC–coastal 32.6 67.4

LIC–landlocked 47.2 52.8

Totals 24–country sample 29.2 70.8

Totals excluding South Africa 45.1 52.7

Totals excluding South Africa and Nigeria 35.8 61.1


Note: Based on annual averages for the period 2001–05. Detailed data presented in annex 1, table 2.

However, the bulk of SOE resources goes to current spending; between 70–85 percent of total current

in public infrastructure goes through SOEs. Current spending includes O&M spending. Though the latter

is essential to harness the economic returns of capital, it is often recorded as a nonproductive expense

because it does not directly translate into asset creation. In fact, the crudest proxy equates public

investment with productive expenditure even though, admittedly, not all investment is productive and not

all current expenditure is waste. Overall, however, very high levels of recurrent spending may indicate

that operational inefficiencies are diverting resources away from investment.

Oil-exporting countries are an exception for institutional spending patterns as the central government

accounts for the most significant portion of infrastructure expenditures. Boosted by record-high oil

revenues, central governments’ expenditures are covering 70 percent of total public spending. This figure

is driven by developments in Chad and Nigeria, where central governments account for close to two-

thirds of all infrastructure expenditures. The relative high importance (though not dominant role) of

central governments’ expenditures in low-income countries—accounting for about 40 percent of public

infrastructure spending—partly derives from their in channeling external development funds. In aid-


15

dependent countries, the share of externally funded expenditures absorbs more than half of all central

government spending on infrastructure.

Governments are the

most prominent financiers

of infrastructure investment

in Sub-Saharan Africa.

Excluding middle-income

countries, governments

carry out between 80 and 90

percent of total capital

investment, consistently

allocating 80 percent of

their infrastructure budgets

to investment. This pattern

can be driven in low-

income and oil-exporting

countries by the

government’s role in

channeling external

development funds and

commodities’ royalties,

respectively. The majority

of external development

funds are earmarked for

investment. The central

government is a dominant

investor in most

infrastructure subsectors: 80

percent of total public

investment in transport and

water supply, and about 40

percent in energy despite

the recent retrenchment of

public sector donors (figure

3.5). A noticeable exception

to this pattern is information and communication technology (ICT) and, as alluded, middle-income

countries.

Half of the countries invest between 3 and 4 percent of GDP; Mozambique, Uganda, Ethiopia, and

Cape Verde invest over 4 percent of their GDP. These numbers are spearheaded by investments in

transport and to some extent energy, certainly led by the central government itself. Cape Verde is the

most salient case of high investment effort essentially doubling regional averages, with reliance on SOE

investment, including the notorious publicly owned TACV, the Cape Verdian airline.

Figure 3.5 Public infrastructure spending, functional and economic breakdown

Investment

-

0.5

1.0

1.5

2.0

2.5

3.0

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

GD

P S

har

es

SOEs

Gral Govern't

Water TransportICTPower

Current Spending

-

0.5

1.0

1.5

2.0

2.5

3.0

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

GD

P S

har

es

SOEs

Gral Govern't

Energy ICT TransportWater

Source: AICD, Fiscal Baseline (2008).


16

Figure 3.6 Public infrastructure capital investment by sector and institution

-

1

2

3

4

5

6

7

Cote

d'I

voir

e

Tan

zan

ia

Cam

ero

on

Rw

anda

Ken

ya

Mal

awi

Ghan

a

Mad

agas

car

Ben

in

Nig

eria

Nig

er

Ch

ad

Zam

bia

Sen

egal

Mo

zam

biq

ue

Ugan

da

Eth

iop

ia

Les

oth

o

South

Afr

ica

Nam

ibia

Cap

e V

erde

GD

P s

har

es (

%)

WSS Power ICT Transport

-

1

2

3

4

5

6

7

Cote

d'Iv

oir

e

Tanzania

Cam

ero

on

Rw

anda

Kenya

Mala

wi

Ghana

Madagasc

ar

Benin

Nig

eri

a

Nig

er

Chad

Zam

bia

Senegal

Mozam

biq

ue

Uganda

Eth

iopia

Leso

tho

South

Afr

ica

Nam

ibia

Cape V

erd

e

GD

P s

har

es (

%)

General Government SOEs

Source: AICD, Fiscal Baseline (2008)

Note: No data were available for SOE investment in Côte d’Ivoire, Madagascar, Chad, Senegal, and Mozambique.

Sectoral background

The transport and energy sectors account for the lion’s share of infrastructure spending. For low-

income countries these two sectors combined absorb about 80 percent of their infrastructure expenditures


17

(figure 3.7). The heavy spending on power is a response to the widely recognized power crisis on the

continent. The efforts of middle-income countries to support energy development contrast starkly, in

fiscal terms, with those of the poorer countries. Middle-income countries spend almost 5 times more than

what aid-dependent and heavily indebted poor countries (HIPC) spend on power. Low shares for water

can be partly attributed to difficulties in capturing expenditure data from municipal water utilities.

Figure 3.7 Public infrastructure spending by sector

-

2

4

6

8

10

12

14

16

18

20

Cote

d'I

voir

e

Nig

eria

Nig

er

Tan

zan

ia

Ben

in

Sen

egal

Mo

zam

biq

ue

Ghan

a

Eth

iop

ia

Les

oth

o

Nam

ibia

GD

P S

har

es (

%)

WSS

ICT

Power

Transport

Composition of fiscal spending

Shares on total

( percent) WSS Power ICT Transport

MIC 12.9 21.3 24.0 41.9

Oil-exporting 14.1 44.6 7.6 33.6

LIC–nonfragile 11.5 41.1 14.9 32.4

LIC–fragile 12.6 61.8 – 25.6

LIC–coastal 8.4 39.8 21.3 30.5

LIC–landlocked 18.3 45.8 6.1 29.8

Average, Africa 12.5 30.7 19.1 37.7

Average, excluding South Africa 12.6 43.1 12.0 32.4

Average, excluding South Africa and Nigeria 12.3 38.5 17.3 31.9



Most infrastructure expenditures on energy are made by SOEs and used for current expenditure. Only

in countries that have large shares of foreign-donor-financed expenditures on electricity (Senegal and

Tanzania) and in resource-intensive countries (Chad and Nigeria) do direct expenditures by central

governments remain large (figure 3.7). The minor role of SOEs in capital investment may be partially

explained by ambiguities in asset ownership. But inefficient pricing policies are probably the most

important cause of the shortage of investment funds.


18

Figure 3.8 Aggregate expenditures on energy: sources and uses

-

1

2

3

4

5

6

7

Cam

ero

on

Rw

and

a

Ch

ad

Nig

er

Mo

zam

biq

ue

Co

te d

'Iv

oir

e

Nig

eria

Mad

agas

car

Gh

ana

Ben

in

Tan

zan

ia

Mal

awi

Ug

anda

Zam

bia

Sen

egal

Eth

iop

ia

Ken

ya

So

uth

Afr

ica

Les

oth

o

Nam

ibia

Cap

e V

erd

e

GD

P S

har

es (

%)

Budget Nonfinancial Enterprises

-

1

2

3

4

5

6

7

Cam

ero

on

Rw

anda

Ch

ad

Nig

er

Mo

zam

biq

ue

Cote

d'I

voir

e

Nig

eria

Mad

agas

car

Ghan

a

Ben

in

Tan

zan

ia

Mal

awi

Ugan

da

Zam

bia

Sen

egal

Eth

iop

ia

Ken

ya

South

Afr

ica

Les

oth

o

Nam

ibia

Cap

e V

erde

GD

P S

har

es (

%)

Investment Current



Data limitations prevent us from fully capturing aggregate expenditures on water and sanitation, but a

rough estimate is slightly more than 1 percent of GDP. Subnational entities account for a large share of

overall spending on water and sanitation, which complicates cross-country comparisons of spending

levels. The data provided in figure 3.9 represent the lower bound of spending on water supply and

sanitation services. The largest expenditures for water supply and sanitation are recorded in Niger,

Namibia, and Ghana.


19

Figure 3.9 Aggregate expenditures on water and sanitation: sources and uses

-

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Mad

agas

car

Ken

ya

Ch

ad

Co

te d

'Iv

oir

e

Tan

zan

ia

Ben

in

Nig

eria

Ug

and

a

Cam

ero

on

Rw

and

a

Mal

awi

Mo

zam

biq

ue

Gh

ana

Sen

egal

Zam

bia

Eth

iop

ia

Nig

er

Les

oth

o

Cap

e V

erd

e

So

uth

Afr

ica

Nam

ibia

GD

P S

har

es (

%)


-

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Mad

agas

car

Ken

ya

Ch

ad

Cote

d'I

voir

e

Tan

zan

ia

Ben

in

Nig

eria

Ugan

da

Cam

ero

on

Rw

anda

Mal

awi

Mo

zam

biq

ue

Ghan

a

Sen

egal

Zam

bia

Eth

iop

ia

Nig

er

Les

oth

o

Cap

e V

erde

South

Afr

ica

Nam

ibia

GD

P S

har

es (

%)

Investment Current



While the public sector retains an active role in providing communication services in several cases,

that role has been taken over by the private sector in most countries. In recent decades, many countries

have privatized publicly owned telecommunications, but communication services are still provided by

SOEs in Benin, Cameroon, Chad, Ethiopia, Ghana, Kenya, Mozambique, Namibia, and Zambia. For the

African countries with public incumbents, cost-recovery and successful productive spending mostly rely

on SOE governance and operational efficiency. Over 90 percent of public expenditure is channeled

through SOEs, and from that amount only 25 percent goes to capital. Ethiopia and Namibia are the only


20

countries where central government expenditures on communication amount to a significant share of

GDP (0.33 percent and 0.47 percent, respectively) (figure 3.10).

Figure 3.10 Aggregate expenditures on communications: sources and uses

-

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Sudan

Zam

bia

Ugan

da

Nig

er

Nig

eria

Rw

anda

Mad

agas

car

Mal

awi

Sen

egal

Ch

ad

Tan

zan

ia

Cam

ero

on

Eth

iop

ia

Mo

zam

biq

ue

Ken

ya

Ben

in

Ghan

a

Cap

e V

erde

Les

oth

o

South

Afr

ica

Nam

ibia

GD

P S

har

es (

%)


-

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Sudan

Zam

bia

Ugan

da

Nig

er

Nig

eria

Rw

anda

Mad

agas

car

Mal

awi

Sen

egal

Ch

ad

Tan

zan

ia

Cam

ero

on

Eth

iop

ia

Mo

zam

biq

ue

Ken

ya

Ben

in

Ghan

a

Cap

e V

erde

Les

oth

o

South

Afr

ica

Nam

ibia

GD

P S

har

es (

%)

Investment Current



Countries with publicly-owned telecom utilities are Benin, Cameroon, Chad, Ethiopia, Ghana, Kenya, Mozambique, Namibia, South Africa, and Tanzania.


21

Figure 3.11 Aggregate expenditures on transport: sources and uses

-

2

4

6

8

10

12

Cote

d'I

voir

e

Nig

er

Rw

anda

Nig

eria

Ben

in

Cam

ero

on

Sen

egal

Ghan

a

Ken

ya

Ugan

da

Tan

zan

ia

Zam

bia

Ch

ad

Eth

iop

ia

Mo

zam

biq

ue

Mal

awi

Mad

agas

car

Les

oth

o

Nam

ibia

South

Afr

ica

Cap

e V

erde

GD

P S

har

es (

%)


-

2

4

6

8

10

12

Cote

d'I

voir

e

Nig

er

Rw

anda

Nig

eria

Ben

in

Cam

ero

on

Sen

egal

Ghan

a

Ken

ya

Ugan

da

Tan

zan

ia

Zam

bia

Chad

Eth

iop

ia

Mo

zam

biq

ue

Mal

awi

Mad

agas

car

Les

oth

o

Nam

ibia

South

Afr

ica

Cap

e V

erde

GD

P S

har

es (

%)

Investment Current



Central government spending on the transport sector differs from other sectors. Funds are primarily

focused on investment. The exception to this pattern is Cape Verde, where there is considerable spending

on the air transport SOE. For the rest of the countries, the profile of transport spending is dominated by

roads. Overall, roads absorb between 60-80 percent of total transport budgets. That might be behind the

higher ratios of current-to-investment spending found in middle-income countries such as Tanzania and

Kenya, all with better records in road maintenance. Also, the very high importance of roads within the

transport sector underscores the impact of donors’ agendas. The limited evidence available indicates the

heavy dependence of roads investment on foreign funding, which ranges from just over 50 percent in a

country such as Senegal to almost 90 percent in a country such as Rwanda. The volatility of official


22

development assistance (ODA) flows contributes to the volatility of public investment in the sector. Thus,

the very high ratio of road investment to GDP in Chad in 2003–05, in Tanzania in 2000, and in

Madagascar in 2004–05, are all associated with short-lived surges in ODA.

4 General government expenditures

The macro outlook and its fiscal implications for infrastructure

African countries have seen their economies grow at a solid 4 percent annual average in recent years,

with half of the countries in the sample achieving growth rates in excess of 4.5 percent for the period

2001–05. The highest growth rates are observed in oil-exporting countries, which have benefited from

rising oil prices. The completion of the oil pipeline in Chad pushed the average annual growth rate in that

country over 15 percent. Côte d'Ivoire, in a postconflict situation, is the only country that has experienced

a growth slowdown. Low-income countries that benefited from debt relief and successful policy reforms

managed to offset the negative effects of higher oil prices; heavily indebted poor countries (HIPC) saw

annual average growth rates in excess of 5.5 percent. Particular robust growth rates were observed in

Tanzania and Mozambique. The middle-income countries grew more slowly than the other groups, at an

annual average growth rate below 4.3 percent.

But growth rates in Sub-Saharan Africa trail those in other developing regions with higher income

levels. Over the last five years, economic growth in Africa has lagged considerably behind growth in

other developing regions, notably East Asia, Europe, and Central Asia. Average annual growth rates in

per capita GDP were only 2.5 percent in Sub-Saharan Africa, compared with 7.3 percent in East Asia and

5.24 percent in Europe and Central Asia. Furthermore, with the notable exception of South Africa,

average per capita income remains below $600, which is about seven times lower than in Latin America

and the Caribbean and four times lower than in Europe and Central Asia. These disparities underscore the

size of the development challenge facing Sub-Saharan African countries. Furthermore, most of these

countries are likely to fall short of the Millennium Development Goals (MDGs). The World Bank’s

Global Monitoring Report for 2005 notes that in order to meet key MDGs by 2015, the region needs

average annual growth rates of more than 7 percent over the next decade. But beyond the MDGs, the

region faces important hurdles to achieve sustained growth.

Inadequate public infrastructure (notably energy and transport) has been identified as a critical

bottleneck to sustained growth in Sub-Saharan countries. Empirical work on African countries that

explicitly models infrastructure as a growth variable finds that infrastructure strongly supports economic

growth.20 Studies that address the issue of reverse causation (higher output creates higher demand for

infrastructure) are scarce, but a few examine African countries. Using a sample of 52 countries and

addressing the issue of reverse causation, Canning and Bennathan (2000) find rates of return to

investment in electrical generating capacity of around 40 percent. For most countries, this rate was

roughly the same as the return on noninfrastructure capital. But for the 11 African countries in the sample,

the average rate of return to generating capacity was 53 percent—1.3 times higher than the African rate of

return to noninfrastructure capital. Even higher rates of return were found for investment in paved roads,

the African average being 69 percent, about 1.7 times higher than returns from noninfrastructure capital.

20 For details see Estache and Wodon (2006) and references therein.


24

Furthermore, Sachs (2004) has identified high transport costs as one of the five structural reasons for

persistent poverty traps in Sub-Saharan Africa.

Efforts to provide basic infrastructure and meet development needs are frustrated by weak public

finances. Compared with other developing regions, public financing capabilities in Sub-Saharan Africa

are characterized by weak domestic revenue collection, limited borrowing capacity, and a high reliance on

external grants. Countries are prone to negative fiscal balances, especially in low-income countries.

Domestic revenue generation remains weak in comparison to other developing countries; nonetheless,

domestic revenues constitute the most important source of fiscal resources, particularly for oil-exporting

and middle-income countries (table 4.1).

Table 4.1 Primary fiscal balance (share of GDP)

Revenues

Domestic revenues Grants

Budgetary expenditures, excl.

debt service Primary balance Primary balance,

excl. grants

MIC 23.62 0.06 21.21 2.46 2.41

Oil–exporting 35.38 0.19 30.55 5.02 4.83

LIC–nonfragile 15.22 4.84 21.27 (1.21) (6.04)

LIC–fragile 12.22 1.01 12.26 0.98 (0.03)

LIC–coastal 15.50 3.25 19.23 (0.48) (3.74)

LIC–landlocked 13.24 5.79 20.45 (1.41) (7.20)

Africa average 23.29 1.33 22.57 2.06 0.72

Excl. South Africa 23.34 2.55 24.31 1.59 (0.97)

Excl. South Africa and Nigeria 15.72 3.62 19.75 (0.41) (4.03)


Note: Net change over the period 2001–06. Averages weighted by countries’ GDP. Totals may not add up.

Yet, external grants play a significant role as a budgetary financing source in the poorest countries of

Sub-Saharan Africa. External grants account for about one-third of the fiscal resources available in low-

income countries. The share of foreign grants in the overall fiscal resource envelope ranges from more

than 40 percent in Rwanda to less than 5 percent in the middle-income countries of the sample. Foreign

grants exceed 30 percent of fiscal resources in eight countries of the sample. Excluding South Africa and

Nigeria, foreign grants account, on average, for more than 3.5 percent of GDP in the study countries. The

highest levels of foreign grants are found in low-income countries recovering from conflict and in

landlocked countries.

The favorable external environment has helped countries expand their available budgets. In the period

2001–05, central government budgets grew by almost 1.9 percent of GDP, driven largely by increases in

middle-income countries (table 4.2). Paired with continued progress in economic reforms in some

countries (such as Ghana and Tanzania), the positive international environment strengthened domestic

investment and increased capital inflows, easing pressure on fiscal policy, fostering economic growth, and

increasing fiscal space in most countries of the region. Changes in available budgets differ significantly


25

across nations, ranging from a 9 percent increase in the Democratic Republic of Congo to a contraction of

more than 8 percent in Zambia.21

Table 4.2 Net change in central government budget: breakdown by financing sources (share of GDP)

Financing sources

Net expenditure

budget Domestic revenues Oil revenues Grants Net borrowing

MIC 4.08 3.40 – (0.03) 0.71

Oil–exporting (3.73) 5.25 7.74 (0.07) (8.92)

LIC–nonfragile 1.69 0.83 – 1.98 (1.12)

LIC–fragile 3.85 0.79 0.61 1.90 1.16

LIC–coastal 2.81 0.87 0.17 1.06 0.88

LIC-landlocked 0.75 0.73 – 3.59 (3.56)

Africa average 1.89 3.04 1.67 0.57 (1.71)

Excl. South Africa (0.33) 2.66 3.20 1.09 (4.07)

Excl. South Africa and Nigeria 1.49 0.93 0.19 1.54 (0.99)


Note: Net change over the period 2001–06. Averages weighted by countries’ GDP. Totals may not add up.

In oil-exporting countries, additional resources from record-high oil prices supported debt-reduction

efforts. A significant portion of oil-price-triggered increases in domestic revenues has been used to pay

back debt, narrowing in tandem the net budget available for nondebt expenditures. This arrangement

clearly makes remaining debt more sustainable in the long run, as reflected in the decrease in net

borrowing, perhaps at the risk of postponing productive investment in infrastructure.

In non-oil-exportion, low-income countries, domestic revenues have increased only marginally, but

the slow pace has been compensated by increases in foreign grant inflows. For the other low-income

countries, and to some extent middle-income countries, further increases in domestic revenue increases

are contingent to institutional reforms to make revenue collection more effective and widen the tax base.

Sub-Saharan Africa can only somewhat rely on favorable external conditions, as those seen in recent

years.

Marginal propensity to invest in infrastructure

In Sub-Saharan Africa, fiscal adjustments might be achieved though capital investments, and to some

extent, cuts in infrastructure investments. While more research is needed to prove this as a pattern, the

infrastructure investment driving fiscal balance behaviors are not a new phenomenon (figure 4.1). The

macroeconomic literature has long acknowledged that fiscal adjustment episodes tend to include

disproportionate public investment cuts.22 For example, Calderón and Servén (2004), based on averages

21 See annex 11 for details. 22 Serven (2006) and Hicks (1991) summarizes the facts on Latin American and other developing countries. For

industrialized counties, see also Roubini and Sachs (1989); De Haan, Sturn, and Sikken (1996) document the

experience of industrialized countries.


26

for eight Latin American countries, show that cuts in infrastructure investment amounted to about 40

percent of the observed fiscal adjustment between the early 1980s and late 1990s. This was remarkable

because public infrastructure investment represented less than 25 percent of overall public expenditure in

Latin American countries. The other side of this adjustment pattern is the volatility in annual capital

expenditures.

Figure 4.1 Changes in infrastructure investment, primary surplus, and government budgets

Benin

Cameroon

Cape Verde

ChadCote d'Ivoire

Ethiopia

Ghana

Kenya

Lesotho

Madagascar

MalawiMozambique

NamibiaNiger

Nigeria

Rwanda

Senegal

South Africa

Tanzania

Uganda

Zambia

(2.5)

(2.0)

(1.5)

(1.0)

(0.5)

-

0.5

1.0

1.5

2.0

2.5

(10) (5) - 5 10 15

Changes in Primary Surplus (GDP Shares)

Chan

ges

in I

nfr

astr

uct

ure

Inves

tmen

t (G

DP

Shar

es)

Zambia

Uganda

Tanzania

South Africa

Senegal

Rwanda

Nigeria

NigerNamibia

MozambiqueMalawi

Lesotho

Kenya

Ghana

Ethiopia

Cote d'IvoireChad

Cape Verde

Cameroon

Benin

(2.5)

(2.0)

(1.5)

(1.0)

(0.5)

-

0.5

1.0

1.5

2.0

2.5

(8) (6) (4) (2) - 2 4 6 8 10

Changes in Government Budget (GDP Shares)

Ch

ang

es i

n I

nfr

astr

uct

ure

Inves

tmen

t (G

DP

Shar

es)

Source: AICD, Fiscal Baseline (2008), processed.

A vast majority of African countries show strong volatility of infrastructure investment despite the

multiyear nature of infrastructure projects and the potential effect of time lags. Changes in the overall

budget envelope are correlated with changes in infrastructure investment, consistent with the well-

documented procyclical behavior of infrastructure investment (figure 4.2). In this context, half of the

countries in the sample decreased their capital expenditures,23 and with the exception of Chad, Namibia,

and Uganda, cuts in capital spending were accompanied by cuts in capital spending on infrastructure.24

Thus, the additional budgetary resources helped low-income countries to bolster capital investments.

Capital investment as a share of GDP increased, in the low-income countries in 2002–05, by more than 1

percent of GDP, about 40 percent of which went toward addressing infrastructure bottlenecks (table 4.4).

It is striking that the oil-exporting countries and middle-income countries decreased investment

despite having more fiscal resources available. Middle-income countries seem to have chosen to

strengthen maintenance. Capital expenditures in oil-exporting countries—led by Nigeria—decreased on

23 For details see annex 11. 24 In Chad, a decrease in budgetary resources and a decrease in capital expenditures as shares of GDP were

accompanied by an increase in nominal GDP owing to the significant increase in oil revenues after completion of

the oil pipeline. When measured as a share of non-oil GDP, Chad’s fiscal resources and capital expenditures

increased.


27

average by 3.3 percent as a share of GDP. In other words, the favorable external environment did not

cause sharp or even proportional increases in expenditures in oil-exporting countries.

Table 4.3 Net change in central government budgets: breakdown by economic use (share of GDP)

Economic uses

Net expenditure

budget Wages Other

current

o/w

Infrastructure Capital

expenditure o/w

Infrastructure

MIC 4.08 (0.11) 4.95 (0.02) (0.75) 0.04

Oil–exporting (3.73) (1.82) (0.23) 0.03 (1.69) (1.46)

LIC–nonfragile 1.69 0.05 0.49 (0.00) 1.15 0.54

LIC–fragile 3.85 0.79 2.49 0.09 0.58 0.22

LIC–coastal 2.81 0.16 1.00 0.04 1.65 0.45

LIC–landlocked 0.75 0.21 0.58 (0.03) (0.04) 0.55

Africa average 1.89 (0.39) 2.68 0.00 (0.39) (0.14)

Excl. South Africa (0.33) (0.63) 0.41 0.02 (0.10) (0.31)

Excl. South Africa and Nigeria 1.49 0.09 0.69 0.01 0.71 0.47


Note: Net change over the period 2001–06.

Averages weighted by countries’ GDP. Totals may not add up.

Infrastructure allocation across social sectors

Only low-income countries—particularly aid-dependent countries—have allocated additional fiscal

resources to clearly favored infrastructure sectors. Roughly one-third of low-income additional resources

were allocated to infrastructure. The education and health sectors were benefited (table 4.5). In middle-

income countries, most of the additional budget was allocated to sectors other than infrastructure and

social. In oil-exporting countries, the decrease in budgetary expenditure was largely absorbed by a

significant reduction in infrastructure expenditures. To a large extent this reflects developments in

Nigeria, where infrastructure expenditures decreased by 2.2 percent of GDP during the study period. As

previously mentioned, this might be largely explained by debt-reduction efforts.


28

Table 4.4 Net change in central government budget: breakdown by functional category (share of GDP)

Functional categories

Net expenditure budget Infrastructure Education Health Other sectors

MIC 4.08 0.02 0.12 0.01 3.93

Oil–exporting (3.73) (1.43) (0.17) (0.17) (1.96)

LIC–nonfragile 1.52 0.54 0.69 0.24 0.07

LIC–fragile 3.85 0.32 – 0.43 3.11

LIC–coastal 2.60 0.48 0.57 (0.03) 1.57

LIC–landlocked 0.75 0.52 0.56 0.80 (1.13)

Africa average 1.85 (0.14) 0.19 0.05 1.75

Excl. South Africa (0.41) (0.29) 0.28 0.08 (0.48)

Excl. South Africa and Nigeria 1.37 0.49 0.40 0.23 0.26

Source: AICD, Fiscal Database, 2008.

Note: Net change over the period 2001–06. Averages weighted by countries’ GDP. Totals may not add up. Data on education are not available for Burkina Faso, Democratic Republic of Congo, Nigeria, or Tanzania.

In conclusion, for most countries in Sub-Saharan Africa, the infrastructure sectors have benefited

from additional allocations following the expansion of countries’ fiscal budgets. This might be the result

of explicit policy making. It remains to be seen whether these gains in resources for infrastructure are

adequate to remove the bottlenecks to growth—and whether, ultimately, they are affordable. It may well

be that the tax base of most African countries is simply too narrow to permit them to cure their

infrastructure problems by investing incrementally greater shares of resources from their modest fiscal

space. Long-term reliance on a favorable external environment is probably not an option.

Prioritizing within the infrastructure budget envelope

About half of the infrastructure budget goes to the transport sector, particularly roads, which account

for the second-highest share of aggregate infrastructure spending (including that provided by the

nonfinancial sector) and the single largest item of central government spending. In low-income countries,

power captures close to one-fourth of the infrastructure budget. This is somewhat higher than the 20

percent allocated to water and sanitation. In middle-income countries, water is undoubtedly the second-

most-important focus of government; most, if not all, spending on energy is let out of the budget to

nonfinancial public enterprises.


29

Figure 4.2 Central government budget by sector and use

Budget Infrastructure Spending

-

0.5

1.0

1.5

2.0

2.5

3.0

3.5

MIC

Oil

Ex

po

rtin

g

LIC

-NoF

ragil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-NoF

ragil

e

LIC

-Fra

gil

e

MIC

Oil

Ex

po

rtin

g

LIC

-No

Fra

gil

e

LIC

-Fra

gil

e

GD

P S

har

esICT

Power

WSS

Transport

Investment Current SpendingTotal Government Budget


Note: Based on annual averages for the period 2001–05. Detailed data presented in annex 1, tables 3–6.

Budget prioritization is strongly influenced by donors’ agendas, particularly for roads and water.

Donors tend to have a bias toward investment over maintenance and significantly favor spending on water

and roads, particularly for rural areas. On average, roads and water together make up between 80 and 95

percent of donors’ allocation to the region (table 4.5). In most non-oil exporting, low-income countries,

donors provide 50–60 percent of what is spent on water and sanitation. In the case of roads, donor

funding accounts, on average, for one-fourth of total spending (table 4.6). In terms of shares of GDP, road

spending is highest in Ethiopia, Lesotho, Mozambique, and Chad, all countries with significant donor

inflows for road rehabilitation.

Donors have not focused on the energy sector, while governments are allocating increasing amounts

of domestically raised resources to the sector. The divergence between donors’ agendas and countries’

priorities is manifested in the spending profiles of the energy sector. Donors have kept a very low profile

in the sector in an environment of low cost-recovery, low affordability, and very high risks (table 4.5). On

the contrary, the governments of low-income countries have allocated close to 25 percent of their budgets

to tackle the ongoing crisis of chronic underinvestment.


30

Table 4.5 Donors’ intervention in Africa: budget relevance and composition

Share of gov’t spending financed by donors Composition of donor allocations

Ratios in percent Overall Budget Energy WSS Roads Total Energy Water Roads

MIC 2.0 0.2 2.3 4.4 100 5 21 74

Oil–exporting 5.9 1.1 2.2 11.0 100 7 7 87

LIC–nonfragile 35.9 38.1 50.6 31.5 100 18 24 57

LIC–fragile – – – – – – – –

LIC–coastal 39.6 39.4 62.2 33.4 100 22 26 52

LIC–landlocked 26.2 32.1 29.6 24.9 100 9 20 71

Africa average 18.8 10.6 24.0 21.8 100 17 23 61



Sub-Saharan countries might be paying a high price for relative overspending on investment in roads

and water to the detriment of other priority sectors and overall maintenance. While budget allocation

across sectors is as much an economic decision as a political one, undoubtedly the cost of funds is one of

the most important considerations for governments when allocating funds, and donor money is, by design,

concessional. The rational decision is to take advantage of cheap money.

Furthermore, Sub-Saharan countries are paying a higher price for funds to the energy sector—

admittedly in crisis—due to the retrenchment of donors. Each dollar raised and spent by a Sub-Saharan

government has a social value premium of almost 20 percent (Warlters and Auriol, 2005). This captures

the incidence of that tax on the society’s welfare (due to changes in consumption patterns) and the

administrative costs of raising the money, among other things. 25 In comparison, official development

assistance (ODA) and funding from the World Bank’s International Development Association (IDA) are

subsidized—with respect to domestically raised funds—26 at about 70 and 55 percent, respectively.27

Funds from countries outside the Organisation for Economic Co-operation and Development (OECD),

such as India and China, are approaching some African countries with quasi-concessional lending. The

subsidy ratio for India and China funds is estimated at around 25 percent, while for Arab funding it is 50

percent.28

25 The marginal cost of public funds measures the “change in welfare associated with raising an additional unit of

tax revenue” (Warlters and Auriol, 2005). 26 Note that, except for grants, ODA and IDA loans have to be repaid by governments using domestically raised

resource. That established the basic price tag for money. 27 This is based in the standard concessional terms For instance, IDA loans charge zero interest (though 0.75 percent

service charge) with 10 years of grace (Foster and others, 2008) 28 India, China and the Gulf States on average charge 4 percent, 3.6 percent, and 1.5 percent of interest and grant a

4–year grace period (Foster and others, 2008).


31

Figure 4.3 Costs of capital by funding source

Cost of Capital relative to the Marginal Cost of Public Funds

0.0

0.3

0.5

0.6

0.9

0.9

1.1

1.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Grants

IDA

ODA

Arabs

China

India

Private

Public

Source: Average marginal cost of public funds as estimated by Warlters and Auriol (2005). Cost of Equity for Private Sector as in Sirtaine and other (2005) and Estache and Pinglo (2004). For the rest, author’s calculations.

Budget efficiency: execution and maintenance

Budget execution

Infrastructure services provided directly through government budgets (or special extrabudgetary

funds) face institutional challenges that can often prevent spending money even if available.

Administrative and procedural difficulties can often prevent budget allocations from being ultimately

released and realized. The ratios of released to budgeted expenditure and of realized to budgeted

expenditure are usually know as budget-variation or budget-execution ratios.

Central governments’ budget-variation ratios help measure the efficiency of public administration.

Execution ratios of less than 100 percent indicate that not all of the budgeted funds were spent, either

because of inefficiencies in implementation or competing pressures for the original allocation that arose

during the fiscal year. Execution ratios of more than 100 percent indicate that the amount spent was

greater than that budgeted, which most likely indicates a cost overrun or policy realignment.

In the aggregate, African countries are not able to spend one-third of their capital budgets and one-

fourth of their recurrent budgets (figure 4.4). Poor timing of project appraisals and late releases of

budgeted funds because of procurement problems often prevent resources from being used in the budget

year. Delays in in-year fund releases are also associated with poor project preparation, leading to changes

in the original terms agreed on with contractors (such as changes in deadlines, technical specifications,

budget, costs, and so on). In other cases capital budgets are reallocated to current expenditures because of

political or social pressures.


32

Figure 4.4 Budget-variation ratios for capital and recurrent spending

Budget Execution Ratios for Capital

82.7

89.0

73.8

73.3

69.4

65.5

63.8

60.8

60.7

53.3

27.8

- 20.0 40.0 60.0 80.0 100.0

Benin

Ghana

Chad

Niger

Kenya

Average

Malawi

Uganda

Cameroon

Ethiopia

Madagascar

Budget Execution Ratios for Current Spending

115.92

137.62

98.17

82.47

82.34

65.5

72.32

69.79

67.16

64.74

33.96

- 20.00 40.0060.00 80.00 100.0

0

120.0

0

140.0

0

160.0

0

Benin

Uganda

Malawi

Niger

Average

Kenya

Ghana

Ethiopia

Chad

Madagascar

Cameroon



Maintenance: the case of roads

The roads sector is the most salient example of a service provided by the government. Roads are

typically provided without user fees, since tolling is only economic above a relatively high minimum

traffic threshold, making pricing issues less relevant. Whether due to political appeal or donor influence,

there is now a generalized sense that investment in the road network in Africa might be overextended.

The issue is that high levels of investment can be a problem if they displace essential maintenance of

existing assets, since it is much more costly over time to rehabilitate or replace poorly maintained assets

than to keep up a sound preventive maintenance regime.

In Sub-Saharan Africa, the share of assets in need of rehabilitation is a striking 30 percent overall,

reaching more than 40 percent for rural roads. Uganda, Nigeria, Rwanda, and the Democratic Republic of

Congo have the highest shares of rehabilitation backlogs (figure 4.5). Close to half of the assets of

railways and rural roads are in urgent need of rehabilitation. But roads deserve special attention. From the

financial viewpoint, roads are primarily a responsibility of governments and in most cases road budgets

are the most prominent single item in national budgets in Sub-Saharan Africa. Transportation is a proven

key ingredient of sustainable development, and roads, for the foreseeable future, are the main transport

mode in the region.


33

Figure 4.5 Infrastructure assets in need of rehabilitation (percent of total)

0%

10%

20%

30%

40%

50%G

ener

atio

n

Non-r

ura

l av

erag

e

Mai

n R

oad

s

Aver

age

Irri

gat

ion

Urb

an W

ater

Ru

ral

Wat

er

Ru

ral

aver

age

Rai

lway

s

Ru

ral

Ro

ads

Av

erag

e re

hab

ilit

atio

n i

nd

ex

0%

10%

20%

30%

40%

50%

60%

Burk

ina F

aso

So

uth

Afr

ica

Cape V

erd

e

Ghana

Cote

d'Iv

oire

Chad

Nam

ibia

Tanzania

Cam

ero

on

Eth

iopia

Kenya

Nig

er

Madagascar

Benin

Mala

wi

Zam

bia

Mozam

biq

ue

Sudan

Senegal

Lesoth

o

Uganda

Nig

eria

Rw

an

da

DR

C

Avera

ge r

ehabili

tation index

Source: AICD.

Rehabilitation ratios only underscore the need for paying increased attention to road maintenance

deficiencies, and the most challenging aspect of investing in roads is providing for their maintenance. In

environments characterized by weak fiscal management practices (nontransparent and politically

dominated budgeting processes), assets are rarely maintained. This is particularly true when, as in the case

of long-life road networks, maintenance has little observable benefits in the short term, and therefore its

budgetary allocations are not naturally protected by the executive level and the parliament.29 Furthermore,

in the context of Africa, donors have a dominant role in channeling funds to the sector. Most of this

funding, earmarked to investment, is concessional by nature, making the cost of additional investments

cheaper than the cost of maintenance, usually raised domestically.

Not surprisingly, there is a bias toward capital spending at the expense of maintenance. Except in the

middle-income countries, more than two-thirds of spending on roads goes to capital expenditures;

maintenance remains a secondary spending priority.30 Oil-exporting countries, especially Nigeria and

Chad, spend around 80 percent of all related expenditures on capital (table 4.6). In other low-income

countries, such as Ethiopia, Senegal, and Zambia, the spending on capital is 12, 24, and 36 thousand

dollars per kilometer respectively—well above the continent’s average, 10 thousand dollars per kilometer.

These amounts are also more than 15 times higher than the average capital spending in middle-income

countries. These startling differences between middle-income countries and low-income countries can be

attributed to differences in initial conditions—middle-income countries already have an established road

network. However, from the perspective of fiscal policy, how each country allocates resources toward

maintenance and capital expenditures has important intertemporal implications. Countries with a high

29 Gwilliam and Shalizi (1996) suggest another method might be to reduce the expected rate of return on projects

resulting in fewer projects crossing the hurdle rate, since inadequate maintenance leads to lower levels of return. 30 All expenditures from road funds and current expenditures by central and local governments are assumed to be for

maintenance (current expenditure), and all foreign-financed expenditures and capital expenditures by central and

local governments are assumed to be for capital rehabilitation. Furthermore, we assume that all foreign-funded and

central government capital expenditures are used on the primary road network. Central governments’ current

expenditures (adjusted for overhead) are assumed to be transfers to local governments for maintenance of the

secondary network (local roads). The distribution of road fund spending between primary and secondary networks is

based on an RMF matrix.


34

share of capital spending on roads will have to allocate greater resources for maintenance in the future,

thus straining public finances. An increase in capital expenditures undoubtedly puts additional fiscal

pressure to provide larger resources on road maintenance to preserve rehabilitated road networks.

Table 4.6 Economic classification of expenditures on road transport

Shares on total ( %) Overhead Capital expenditure (incl. rehabiliation)

Current spending

(incl. maintenance)

MIC 0.54 4.96 93.95

Oil–exporting 5.77 81.03 13.20

LIC–nonfragile 2.95 56.64 35.72

LIC–fragile 2.94 65.95 0.37

LIC–coastal 4.03 59.90 36.07

LIC–landlocked 1.01 55.39 17.58

Africa average 2.35 37.37 57.17

Excl. South Africa 4.12 67.33 22.70

Excl. South Africa and Nigeria 2.77 56.87 32.06


Note: Annual averages over the period 2001–06. Averages weighted by countries’ GDP. Totals may not add up.

High capital expenditure on roads may be justified by large rehabilitation backlogs. Using the

RONET model, it is possible to produce detailed estimates of the rehabilitation requirements for each

country’s road network, taking into account the current distribution of network condition, and working

toward a target of clearing the current rehabilitation backlog within a five-year period. On that basis, the

rehabilitation requirements can be compared with the current levels of capital expenditure to determine

whether these are high enough to eliminate the rehabilitation backlog within a reasonable period of time.

It is important to note that this calculation is only illustrative, and is based on the assumption that the

entire capital budget is devoted to network rehabilitation. While rehabilitation does tend to dominate

capital spending, other types of investment do take place, including upgrading road categories or adding

new roads; available data do not make it possible to know the exact split. However, the calculation is

helpful in illustrating whether current levels of capital expenditure would be high enough to address the

rehabilitation problem if they were fully allocated to rehabilitation works.


35

Figure 4.6 Capital expenditure as a percentage of rehabilitation requirements

-100%

-50%

0%

50%

100%

150%

200%

250%

300%

350%

400%

Ke

nya

Benin

Nig

er

Nig

eria

Mala

wi

Cam

ero

on

Uganda

Zam

bia

Lesoth

o

Rw

anda

Madagascar

Mozam

biq

ue

Senegal

Cote

d'Iv

oire

Ghana

Tanzania

Chad

Eth

iopia

Perc

enta

ge d

evia

tion fro

m r

equirem

ent

Rehabilitation Rehabilitation adjusted for capital execution

Source: Africa Infrastructure Country Diagnostic, Fiscal Baseline (2008), AICD RONET Analysis, 2008.


When compared to norms, roughly half of the countries have shortfalls of 40 percent or more in

annual maintenance (figure 4.7). It is relevant to compare current maintenance expenditures with the

norm for Africa. Using the RONET model, it is possible to produce detailed estimates of the routine and

periodic maintenance requirements for each country’s road network, taking into account the current

distribution of network condition. On that basis, the maintenance requirements can be compared with the

current levels of maintenance expenditure to determine whether these are high enough to preserve the

network in good condition. It is important to note that this calculation is only illustrative, and is based on

the assumption that the entire maintenance budget is spent on maintenance works at efficient unit costs.

However, the calculation is helpful in illustrating whether current levels of maintenance would be high

enough to preserve the network if they were efficiently spent.


36

Figure 4.7 Maintenance expenditure as a percentage of requirements

-100%

0%

100%

200%

300%

400%

500%

600%

Chad

Nig

eria

Ugan

da

Nig

er

Sen

egal

Mal

awi

Mad

agas

car

Eth

iop

ia

Leso

tho

Rw

anda

Gh

ana

Ken

ya

Tan

zan

ia

Mo

zam

biq

ue

Ben

in

Nam

ibia

Cam

ero

on

Zam

bia

So

uth

Afr

icaM

ain

tenance e

xpenditure

as %

requirem

ents

Routine plus periodic maintenance Routine maintenance

Source: Africa Infrastructure Country Diagnostic, Fiscal Baseline (2008), AICD RONET Analysis, 2008.


Maintenance shortfalls are more pronounced in countries without road funds and fuel levies. In cases

such as Chad, Nigeria, Uganda, Niger, and Senegal, maintenance spending is less than half the normative

requirements. Underspending on maintenance is evident in low-income countries (particularly the

resource-rich) and in countries with difficult geographical environments and terrain. Middle-income

countries tend to spend substantially above the maintenance norm. However, the problem of

underspending on maintenance is by far the most pronounced in countries that lack a road fund and hence

a fuel levy. Among countries with fuel levies, those with high levies do substantially better than those

with low ones.

If the definition of maintenance is narrowed to consider only the routine aspects of maintenance, the

costs are reduced by about one half. Yet around a quarter of the countries in the sample are not devoting

adequate resources to cover even routine maintenance activity.

The most important vehicle for domestic funds are the so-called road funds, which are designed to

sequester funds for use on maintenance. Domestic financing for roads is channeled mainly through a mix

of central government transfers to local governments and spending from road funds. Road fund

expenditures vary significantly in proportion and absolute amount from country to country, but, on

average, less than one-fifth of total spending for roads is channeled through road funds. Most funding

passes through central governments’ main budgets, either as transfers to regional governments or as

capital expenditure.


37

Table 4.7 Average expenditures for roads

GDP share ( %) Total spending of which external

of which road fund

MIC 0.31 0.01 0.06

Oil–exporting 1.31 0.14 0.21

LIC–nonfragile 2.36 0.74 0.51

LIC–fragile 0.66 – –

LIC–coastal 1.81 0.61 0.53

LIC–landlocked 2.51 0.62 0.22

Africa average 1.03 0.22 0.17

Excl. South Africa 1.76 0.43 0.33

Excl. South Africa and Nigeria 1.97 0.61 0.47


Note: Annual averages over the period 2001–06.

Averages weighted by countries’ GDP. Totals may not add up. Implementation of second-generation road funds seems to help protect resources for road

maintenance. Although it is possible to compute an optimal maintenance requirement (various

maintenance norms for any type of road), implementing a mechanism that ensures the availability of

sufficient funding to meet those requirements over the life cycle of the road network is a major challenge,

not only in developing countries but in many developed countries as well. Finding the right balance

between new investment and maintenance has been also the focus of extensive debate.31 In fact, during

the last decade, the majority of the countries of Sub-Saharan Africa made efforts to establish so-called

second-generation road funds, 32 based on relaying charges from users of roads33 (rather than earmarking a

proportion of general tax revenues) while incepting a separate administrative agency built upon principles

of commercializing road maintenance by collecting user charges. Countries such as Ghana, Namibia, and

Zambia that have made genuine efforts toward implementing second-generation road funds spend on

average well above $2,000 per kilometer per year on maintenance of major networks.34 Road maintenance

in these countries does not differ significantly from estimated maintenance expenditures in South Africa.

Expenditures per kilometer of primary network are below $500 in Niger and Rwanda.

With the notable exception of Malawi, maintenance expenditures via road funds have increased in the

majority of countries, but at a lower pace than overall road expenditure. These increases on maintenance

expenditures are significantly smaller than increases in capital expenditures financed by external donors

31 Gwilliam and Shalizi (1996), Benmaamar (2006), and references cited therein. 32 The predecessor to this mechanism were “first generation road funds,” which were characterized by the

earmarking of certain road-related taxes and charges to a special off-budget account. Technically, however, these

funds are nothing more than a separate budget line item with no separate oversight. 33 In most cases, these are indirect charges for access (various vehicle registration fees) and use (fuel levies), but there are a few examples of direct charges (e.g., toll roads). 34 Because of data limitations, we were able to clearly identify maintenance expenditures only for the main network.

Estimating expenditure on maintenance on the secondary network is difficult, because it is not possible to obtain a

precise estimate of the maintenance expenditures of local governments. Different degrees of data accuracy prevent

reliable cross-country comparisons of maintenance expenditures on secondary roads.


38

or central government. The share of external financing, most of which is allocated to road rehabilitation,

has increased in every country that receives external financing. Central government allocations to roads

have increased in the majority of countries, but have decreased in Malawi, Madagascar, and Nigeria.

Another important achievement of road funds has been to avoid the volatility of maintenance

expenditures. In addition to supporting higher levels of road maintenance expenditure, there is some

evidence that road funds have also helped to improve the predictability of road maintenance expenditure.

Expenditure data for 2001–05 show that volatility of road fund expenditures (measured by calculating the

standard deviation around the trend line) was only half that of expenditures arising from external funding

and one-third that of central government allocations.35 Moreover, the volatility of road fund expenditures

appears to be lower in countries that have made efforts to ensure the independence of their road funds and

increased the proportion of revenues channeled directly into the road funds.

In conclusion, countries in Sub-Saharan Africa are allocating a significant share of GDP to road

sector expenditures, which have only increased in most countries. This resource allocation to roads is

driven in part by large donor funding inflows in aid-dependent countries and robust domestic revenues in

oil-exporting countries. The vast majority of this expenditure is allocated to capital expenditures, leaving

maintenance a secondary priority. Countries that have made efforts at establishing well-functioning road

funds tend to be more successful at ensuring optimal provision of maintenance expenditures and reducing

the volatility of expenditure flows.

35 For details see annex 14.

5 The hidden costs of utilities’ inefficiencies

There are myriad approaches to the measurement of efficiency. The most widely accepted are based

on the idea that rational economic agents will pursue maximum output from a given set of inputs.36 In his

seminal work, Farrell (1957) introduced the concept of production frontiers—or a possible set of optimal

outputs—that provide a basis for measuring efficiency.37 His approach to efficiency measurement relies

on the estimation of a production frontier for which inputs and outputs can be correctly measured.

Technical efficiency refers to the inputs needed to produce a given level of output using a specific

technology. Allocative efficiency measures the extent to which factors of production are used in optimal

proportion, given their prices or economic returns.

Accurate measurement of efficiency is extremely data-intensive and requires detailed disaggregation

and precise measurement of both inputs and associated outputs. The efficiency of public spending for

infrastructure services involves the simultaneous production of multiple outputs. For instance, electricity

generation will have no benefit if access to the network is not created in parallel. Similarly, delivering an

improved water supply implies a multioutput production function. On the input side, expenditures on

infrastructure services are natural proxies for the inputs of the production functions. In the case of

services provided by public enterprises, however, money flows do not capture the costs of inefficient

administration or poor policy decisions, which are frequently hidden from central government accounting.

These so-called hidden costs have fiscal implications and reveal the implicit use of resources across

sectors and, to some extent, across economic agents. Thus, public expenditures are not always what they

seem from the input or cost perspective and should be reassessed for efficiency.

Hidden costs are estimated by monetizing the costs of four quasi-fiscal activities (QFAs): mispricing,

unaccounted-for losses, undercollection of invoiced amounts, and labor redundancies. Hidden costs are

important to gauge. Not only do they give a sense of the scope, scale, and opportunity cost of inefficient

operations, but they also help to pinpoint the sources of inefficiency, which may be policy or operational

in nature. From a macro perspective, estimating hidden costs is essential for any accurate assessment of a

country’s budget. The majority of utilities’ hidden costs are ultimately financed by subsidies, direct or

indirect. In efficiency analysis, adding hidden costs to the level of public spending provides a more

realistic proxy of public resource utilization for infrastructure provision, both within and across countries.

QFAs disguise the size and scope of the public sector in general and have sizeable macroeconomic

effects beyond the cash-flow shocks they produce in the public sector. A permanent source of concern in

public finance management, QFAs can: (i) increase financial instability, (ii) be a highly distorting factor

in resource allocation, and (iii) potentially create substantial contingent liabilities (Mackenzie and Stella,

1996). They also create distorted incentives in the economy, leading to overconsumption and waste (Petri

and Taube, 2003). From a macro perspective, QFAs affect the overall public sector balance without

explicitly affecting the budget. More often than not, QFAs escape parliamentary scrutiny, distort the size

36 Or the converse, to produce a specific level of output at the least cost. 37 For a survey of the literature on production frontiers, see Forsund and others (1980) and references cited therein.


40

of the public sector, and derail economic policy making. To allocate scarce resources properly, it is

critical to know how they are actually allocated, and this is masked by QFAs.

When utilities engage in QFAs, conventional measures of borrowing requirements based on financial

activities are misleading. In terms of fiscal discipline, it is critical to have an accurate estimate of the

magnitude of the budget and its financing. In the presence of QFAs, the overall budget deficit (or

budgetary spending in general) is understated. Cumulative debts and triggered contingent liabilities lead

to ad hoc periodic capitalization and debt swaps that usually bypass parliamentary resource-allocation

approvals. This implies that, in a very inefficient and nontransparent manner, fiscal resources are

allocated without reference to development plans and without any economic or strategic rationale. These

lumpy and ad hoc fiscal interventions divert funds from priority projects, funneling scarce resources into

uses that bring zero or negative return.

The hidden costs of QFAs partially quantify the often unintended and usually nontransparent transfers

from the producer to the consumer, while also yielding a sense of the opportunity cost of inefficient

operations at the sector and enterprise level. Hidden costs include all utility operations that could in

principle be duplicated by specific budgetary measures in the form of an explicit tax, subsidy, or other

direct expenditure (Mackenzie and Stella, 1996). Such costs are self-perpetuating: the absence of financial

sustainability and inappropriate expansion of maintenance, rehabilitation, and service lead to further

waste and create more incentives for utilities to hide their poor financial situation and seek more bailouts.

Identifying and quantifying the sources of hidden costs in utilities is prerequisite to getting them on the

policy agenda and eventually eliminating them. Special attention has been paid to this issue in the energy

sector and in transition economies (Petri and Taube, 2003; Saavalainen and ten Berge, 2006; and

Chivakul and York, 2006), motivated by the heavy burden imposed on these economies by policies that

pursued universal access to electric power regardless of cost.38 For infrastructure sectors the most

comprehensive attempt to estimate hidden costs outside the energy sector has been made by Ebinger

(2006), who estimates the hidden costs for power, natural gas, water supply, and, to some extent,

railways. She finds that for a sample of around 15 transition countries in Europe and Central Asia, QFCs

averaged 4.4 percent of GDP for power, 1 percent GDP for gas, and 1.2 percent for water.

Identifying and reducing inefficiencies in infrastructure operations is perhaps the most practical and

realistic way to free up resources for infrastructure in Africa. Most countries are making huge efforts to

improve their infrastructure. However, they are severely constrained by their ability to afford the

quantities of infrastructure needed to stimulate economic growth. This first attempt to estimate hidden

costs in Africa’s utilities uses, for the water and power sectors, the most common approach for

quantifying hidden costs—the so-called end-product approach. Described in detailed by Ebinger (2006),39

the methodology identifies three relevant QFAs in utilities: mispricing, undercollection, and excessive

unaccounted-for losses. It estimates hidden costs by comparing actual indicators of a functioning state-

owned enterprise (SOE) against ideal norms of cost-recovery, collection ratios, and distribution losses.

For the telecom utilities we quantify the hidden cost of labor redundancies by comparing partial labor

38 In transition economies, QFCs in the power and gas sector combined have been estimated, as a share of GDO, at

between 10 percent (Kyrgyz Republic, Turkmenistan) and 20 percent (Tajikistan, Uzbekistan, Azerbajan). For

details, see Petri and Taube, 2003; and Saavalainen and ten Berge, 2006. 39 See also, Petri, Gunter, and Tsyninki, 2002.


41

productivity ratios of existing African telecommunications incumbents against world-class fixed-line

providers in the member countries of the Organisation for Economic Co-operation and Development.

Hidden costs in water and power utilities

Quasi-fiscal

activities in Africa entail

average annual hidden

costs of 0.6 percent of

GDP in the water sector,

and 1.9 percent in the

power sector. These

overall aggregates mask

differences across

sectors and among

countries (figure 5.1).

Relative to GDP, hidden

costs for power utilities

are more than double

those for water utilities.

The smaller economic

size of water utilities,

together with their

miscoverage in the

sample (due to

decentralization and

fragmentation), partially

explains the lower

apparent losses. In the

water sector, hidden

costs amount to no more

than 1.5 percent of GDP

except in the

Democratic Republic of

Congo (2.6 percent

GDP), while in the power sector hidden costs are close to zero in South Africa, about 0.2 percent of GDP

in Benin, and more than 4 percent of the GDP in Malawi.

Mispricing is the main source of hidden costs in both power and water supply utilities (table 5.1). In

other words, the main source of hidden costs for African utilities is linked to policy decisions. The

implication is not that subsidies for the poor should be eliminated across the board, but rather that

subsidies and fiscal transfers must be made explicit and accounted for in budgetary appropriations. The

second main source of hidden costs is collection inefficiencies—or failure to collect bills—a cost that is

Figure 5.1 Hidden costs of water and power utilities (share of GDP)

Water

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

Tan

zan

ia

Nig

eria

Ben

in

Cap

e V

erde

Eth

iop

ia

Ugan

da

Nam

ibia

Ken

ya

Burk

ina

Fas

o

Rw

anda

Sudan

Nig

er

South

Afr

ica

Les

oth

o

Mo

zam

biq

ue

Cote

d'I

voir

e

Sen

egal

Mad

agas

car

Zam

bia

Mal

awi

Ghan

a

DR

C

Mispricing Unaccounted Losses Collection Inefficiencies

Power

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

So

uth

Afr

ica

Ben

in

Ken

ya

Mo

zam

biq

ue

Ch

ad

Cap

e V

erd

e

Mad

agas

car

Les

oth

o

Nig

eria

Bu

rkin

a F

aso

Rw

and

a

Eth

iop

ia

Ug

and

a

Cam

ero

on

Zam

bia

Tan

zan

ia

Sen

egal

Gh

ana

Nig

er

Mal

awi

DR

C C

on

go

Under-Pricing Unaccounted Losses Collection Inefficiencies



42

inextricably tied up with related to managerial and administrative practices. For that reason, political will

and capacity building must play a key role in any approach to the problem.

Aid-dependent countries show slightly higher levels of hidden costs relative to their African peers.

The chief causes are mispricing and, in the water sector, collection inefficiencies. Both sources of

inefficiency are distortionary and nontransparent mechanisms of transferring resources to users—and are

very costly from the fiscal viewpoint. In resource-rich countries, mispricing is also the predominant

source of hidden costs, whereas in middle-income countries unaccounted-for losses are the salient factor

for power utilities. The latter are quite worrisome because they are associated with maintenance-deprived

distribution networks.

What these numbers suggest is that, consciously or unconsciously, African governments are using

public utilities for quasi-fiscal purposes. The composition of hidden costs implicitly reveals rational

policy decisions that are somehow related to the overall characteristics of the economy. Tariff regimes

that admittedly do not allow for cost recovery (and that thus provide an implicit subsidy to consumers)

have existed alongside official tolerance of arrearages in customer payments or low collection rates (an

implicit tax on utilities), particularly in oil-exporting countries. Furthermore, tolerance of pilferage (an

implicit subsidy of customers) adds to deferred maintenance and underinvestment (implicit borrowing

from future taxpayers) as the main causes of excessive losses in distribution utilities.

Table 5.1 Typology of quasi-fiscal activities and hidden costs

Water sector Power sector

Mispricing Collection inefficiencies

Unaccounted for losses

Kenya

Mozambique

Nigeria

Malawi

Tanzania

Rwanda

Sudan

Zambia

Rwanda

Malawi

Nigeria

Mozambique

Ghana

Tanzania

Mor

e th

an 3

3% o

f hid

den

cost

s fr

om

Mor

e th

an 6

6 %

of h

idde

n co

sts

from

Senegal

Côte d'Ivoire

Niger

Lesotho

Madagascar

DRC

South Africa

Namibia

Benin

Ethiopia

Cape Verde

Burkina Faso

Uganda

Mispricing Collection inefficiencies

Unaccounted for losses

Lesotho

Senegal

Niger

Cape Verde

Nigeria

Uganda

Mozambique

Nigeria

Benin

Mozambique

Uganda

Mor

e th

an 3

3% o

f hid

den

cost

s fr

om

Mor

e th

an 6

6% o

f hid

den

cost

s fr

om

Zambia

Rwanda

Malawi

Tanzania

Cameroon

Chad

Madagascar

Kenya

Burkina Faso

DRC

Ghana

South Africa

Benin

Kenya

Source: AICD Database.


43

Table 5.2 Average quasi-fiscal hidden costs and their drivers according to different typologies

Total hidden cost

Unaccounted losses Underpricing

Collection inefficiencies

GDP share (Share of total hidden costs)

WATER

MIC 0.47 12.26 76.65 11.09

Oil–exporting 0.13 14.27 37.72 26.70

LIC–nonfragile 0.50 25.92 51.26 22.83

LIC–fragile 0.95 14.14 76.01 9.85

LIC–coastal 0.56 23.46 64.82 11.72

LIC–landlocked 0.60 24.49 39.23 36.28

Africa average 0.43 16.37 62.04 17.07

Excl. South Africa 0.39 19.86 48.34 23.15

Excl. South Africa and Nigeria 0.49 19.94 49.42 18.40

POWER

MIC 0.02 96.42 0.67 0.24

Oil–exporting 1.46 24.94 39.48 29.35

LIC–nonfragile 1.78 34.77 43.54 21.69

LIC–fragile 1.43 8.45 - 22.29

LIC–coastal 1.26 32.13 33.01 15.70

LIC–landlocked 2.52 26.44 40.84 32.72

Africa Average Ctry 0.84 62.33 19.62 12.99

Excl. South Africa 1.60 27.93 37.53 24.84

Excl. South Africa and Nigeria 1.68 28.69 39.26 18.32

Source: Authors’ calculations using data from the AICD, Fiscal Database, 2008.

Note: Annual averages over the period 2001–06. Averages weighted by countries’ GDPs. Totals may not add up.

Collection rates: tolerance of end-users’ arrears?

Utilities seem to be tolerant of nonpayment even though they cannot fully recover those losses via

cross-subsidies. Utilities in Africa do not “officially” report significant inefficiencies due to collection

problems. About 60 percent of African power and water utilities report collection ratios higher than 90

percent of total bills, with only 10 percent of the utilities reporting collection rates below 60 percent.

These official figures are seemingly contradicted by household data, based on which up to 40 percent

(power) and 60 percent (water) of households do not pay their utility bills (figure 5.2). Regardless,

whether the latter figures compound nonpayment by illegal connections and therefore overstate the

problem, the two sides of the story are difficult to reconcile. On the one hand, the data underscore failures

to enforce payments. On the other hand, collection ratios inconsistent with nonpayments might just reflect

cross-subsidization. Thus, cross-subsidies and relaxed collection practices seem to be joined (“If large

users pay their bills and small users leave unpaid bills, what is the problem?”). By design or by default,

neither collection index is a good proxy for miscollection or aggregate transfers from the producer to the

consumer.


44

Figure 5.2 Collection and payment rates

Water Power

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

<60 60<>90 <90

Ranges (%)

Reported by Utility Implicit Reported by HHs

0.0

10.0

20.0

30.0

40.0

50.0

60.0

<60 60<>90 <90

Ranges (%)

Reported by Utility Implicit Reported by HHs

Source: Authors’ own calculation using data from the AICD Database.

A more useful assessment of collection inefficiencies compares collected bills against tariff regimes

in place. A quick and dirty implicit collection measure is calculated as the ratio between collected

revenues and effective tariffs. The implicit collection brings a more solid ground to the story of hidden

cost by miscollection when allowing benchmarking of this type of inefficiencies against the aim of policy

makers (decision variable tariff regime and level). Estimates show that about 35 percent (water) and 45

percent (power) of African utilities have implicit collection ratios higher than 90 percent, while 75 percent

(water) and 90 percent(power) have implicit collection rates lower 60 percent of their potential bills.

Frequency charts of implicit ratios show a less extreme distribution of payers and no payers.

Hidden costs associated with collection inefficiencies are relatively small, averaging 0.1 percent of

GDP for water utilities and 0.4 percent GDP for power. In water, only Zambia and Malawi have hidden

collection costs of around 0.4 percent of GDP, perhaps the highest in region. In power, Burkina Faso,

Uganda, Ghana, and Niger have hidden collection costs of more than 1 percent of GDP.


45

Box 5.1 Subsidies for large electricity users: the case of Zambia

The average effective power tariff in Zambia, at 3 cents per kWh, is one of the lowest in Africa. This current level

does not even allow for recovering operating costs, needless to say total costs. Zambia has one of the lowest average

costs of the region (due to the combination of hydropower technologies with excess generation capacity).

This situation of inefficient pricing policies gets compounded by exceptionally favorable prices that the power utility ZESCO gives to mining companies, particularly Copperbelt Energy Corporation (CEC). A signed long-term

agreement set mining tariffs at 2 cents per kWh, not only below cost-recovery but even one-third lower than the

effective tariff for an average residential customer (100 kWh per month).

As the mining sector is the recipient of 50 percent of total ZESCO sales, this translates into a conservative estimate

of $30 million in annual subsidies, with a projected cumulative deficit of $926 million over the next 10 years.

The Zambia case is not unique to the region. Until the year 2003, Ghana’s power distribution company VRA, was

engaged in a long-term agreement with Volta Aluminum Company. VALCO was for a while VRA’s most important

customer, consuming one-third of its power generation and benefiting from an electricity preferential price estimated

to be half of the cost-recovery level.

Source: (i) Zambia Electricity Regulator Board, 2008 Press Statement on the ERB Decision on ZESCO Application to Revise Electricity Tariffs, other Charges, Fees and Penalties (ii) World Bank (2008), Zambia Growth Infrastructure and Investments: A Role for Public Private Partnership (iii) Chivakul, M. and York, R,M 2006 Implications of Quasi-Fiscal Activities in Ghana, International Monetary Fund Working Paper, Washington D.C.

Unaccounted losses: paying today for yesterday’s neglect?

Two-thirds of African power utilities have transmission and distribution losses more than twice

higher than the acceptable norm for the sector.40 The water sector has a U-shape frequency chart. About

one quarter of the utilities achieves losses within the international norm in clear contrast to a 60 percent of

the sector suffering of losses more than twice higher than the acceptable norm of the sector (figure 5.3).

Excessive power transmission and distribution losses and/or excessive nonrevenue water levels are a

predictable consequence of poor maintenance and rehabilitation as much as inadequate billing and

metering practices. They are also a very crude evidence of the consequence of poor project appraisal

(preference of investment projects over maintenance). Deferring spending on maintenance or/and

misinvesting is perhaps the most perverse and difficult to track financing mechanism of utilities’ cash

deficits. Mismaintenance and investment can disguise today’s poor financial positions but not the

permanent decay of the service—in quantity and quality—for future customers. In fact, provisioning for

subsequent maintenance and replacement investment is the most efficient way of fully grabbing the

economic returns of existing assets.

40 Norms of good performing networks: 10 percent transmission and distribution for power, 20 percent of

nonrevenue water for water.


46

Figure 5.3 Excessive distribution losses

0%

10%

20%

30%

40%

50%

60%

70%

+/- 10

%

10%

-30%

Hig

her

30%

-60%

Hig

her

60%

-100

% H

ighe

r

100%

+ Hig

her

Deviation from the Norm in Percent

Water Power

On average, hidden costs associated with unaccounted losses amount are a low 0.1 percent of GDP in

the water sector, contributing by less than 20 percent to total cost inefficiencies. Only the Democratic

Republic of Congo sticks out with 0.5 percent of GDP in hidden costs. For Mozambique unaccounted

losses are the main source of inefficiencies, accounting for close to 50 percent of its hidden costs. In

power, however, hidden costs average 0.5 percent of GDP, reaching their highest level in Niger, Ghana,

and Uganda.

Tariff regimes: aiming at cost-recovery?

With the notable exception of Cape Verde and Uganda, tariff regimes in the water sector do not allow

for cost recovery. Clearly, tariffs can be—and actually are—used as a conduit for subsidies by policy

makers. From a fiscal perspective, the budget deficit (or spending) is underestimated. From a sector

perspective, about 80 percent of AFrican countries (19 out of a 24 country sample) have in place cost-

recovery policies to recoup overall operations and maintenance (O&M) and some investment.

Nonetheless, for the large majority of water utilities, tariffs did not meet various investment-cost

threshold levels. At the highest block prices, where cost-recovery is more feasible, only one-third of water

utilities meet capital costs41 while if focusing in average (across user groups) effective tariffs only four

water utilities (out of 40 in the sample) meet simultaneously operation, maintenance and capital costs

41 Seven utilities in South Africa, and Namibia, together with utilities in Cape Verde, Benin , Niger, Rwanda,

Senegal, Burkina, Lesotho and Côte d’Ivoire.


47

(figure 5.4). This suggests that investment costs are either completely assumed or subsidized by the

governments (central and/or local), communities, and/or NGOs.42

Figure 5.4 Average effective tariffs against historical average costs

Water sector Power sector

Source: Banerjee and others, AICD 2008. Source: Briceño-Garmendia and Shkratan, AICD 2008.

Artificially low tariffs not only create additional economic costs but also induce over-consumption

and waste of already insufficient water supply resources. Altogether, in the water sector, hidden costs due

to underpricing (or mispricing) amount to an average of 0.4 percent for Africa in GDP terms and about 55

percent of total hidden costs in the sector. The hidden costs due to mispricing top 2 percent of GDP in the

Democratic Republic of Congo but also reach over 0.5 percent of GDP for Malawi, Madagascar, Senegal,

and Côte d’Ivoire.43 Admitting the existence of affordability issues that will continue if not addressed,44

the emerging policy issue is that these hidden costs accumulate from the fiscal viewpoint, eventually

inducing instability and ad-hoc spending.

The African power sector is characterized by higher prices and even higher costs. In the power sector,

6 out of 20 countries in the sample have effective tariffs aligned with average historic costs (figure 5.5).

Power mispricing entails hidden costs equivalent to about 1 percent of GDP or 60 percent of total hidden

costs in the sector. This is more than twice the cumulative mispricing hidden costs from the water sector

(0.4 percent of GDP). Malawi (3.3 percent of GDP), Zambia (2.3 percent), Niger (1.7 percent) and

Cameroon (1.6 percent) stick out as the countries creating the highest hidden costs to their economies due

to mispricing.45 Nonetheless, the main concern in power is that historic costs are extremely high by

international standards, being mostly driven by extremely operational costs (Foster and others, 2008).

Given that investment levels carried out by utilities are known to be low, cost-recover tariffs for

42 For details Banerjee, S., Foster V, at al., 2008. 43 For detailed calculations see annex 15. 44 Thoroughly covered in AICD sister piece by Banerjee, S, Wodon, Q., at al., 2008 45 For detailed calculations see annex 16.

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

0.00 1.00 2.00 3.00 4.00

Historical Average Costs ($/m3)

Av

erag

e E

ffec

tiv

e T

arif

f ($

/m3

)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Historical Average Costs ($/kWh)

Av

erag

e E

ffec

tiv

e T

arif

f ($

/kW

h)


48

admittedly high operational inefficiencies implicitly assume governments would/should finance capital

requirements (further tariff increases might just essentially be unaffordable or infeasible).

Box 5.2 Distributional incidence of capital subsidies to power and water

Widespread underpricing of power and water by utilities in Sub-Saharan Africa prevents the utilities from

recovering their capital costs, or even their full operating costs, from their operating revenues. To make up the

deficits, governments routinely provide substantial capital subsidies to the utilities.

Subsidies for basic services such as water and electricity would be socially benign if access to those services were

more evenly distributed. In fact, however, access is low and sharply skewed to wealthier households. This means

that the capital subsidies are highly regressive—they benefit the rich disproportionately.

Access to power and water provided by utilities is almost nonexistent across the bottom three-fifths of the household

income distribution in Sub-Saharan Africa—with the result that about 80 percent of the utilities’ customers are

drawn from the wealthiest 40 percent of households. Although the precise rate of subsidy distribution does not

depend on access alone but also on levels of consumption and tariff structures, the highly inequitable distribution of

subsidy suggested by the patterns of access is not substantially modified by tariff structures, often because the subsidized tariff bands are so wide that many well-off households fall within them.

0%

20%

40%

60%

80%

100%

Q1 Q2 Q3 Q4 Q5

Piped Water

Electricity

An AICD team studied subsidy incidence for power utilities in 18 Sub-Saharan countries and for water utilities in 15

countries. They devised a parameter ( ) to designate the share of the utility subsidy captured by the poor relative to

their share in the population. Thus, if the poor represent half of the population and capture half of the subsidy, the

parameter has a value of one. Values lower than one indicate a regressive distribution of the subsidy with the poor

capturing less than their share of the population, and vice versa.

The findings are sobering. In all cases studied the value of lies well below one; ranging from as low as 0.01 to a

high of 0.78. In fact, the average value of across all countries and for both services turns out to be only 0.33 for

both power and water. This means that the poor capture a share of the subsidy that is only a third of their share in the

population.

Sources: Banerjee and others, 2008; Wodon and others, 2008.

Investment plans based on more efficient existing operational technologies might lead to lower costs,

and lower cost-recovery power tariffs. A recent costing exercise of investment requirements for the East

Africa and the South Africa power pools (EAPP and SAPP) provides a reference for what incremental

costs could be were investment plans carried out under reasonable (benchmark) unit cost assumptions

(Econ Analysis, 2008). One would expect incremental costs under investment expansions to be higher

than historical costs because capital premiums are essentially nonexistent for historical costs. However,

with few exceptions, in both trade and nontrade scenarios, incremental costs are significantly lower than


49

historical costs. In the trade scenario all countries but Madagascar and Ethiopia would face lower average

costs, reduced by about 40 percent on average (figure 5.5). Salient beneficiaries of trade would be

Zambia, Uganda, and Malawi, which while facing increased average costs if carrying power investment

in isolation, would reduce about 25 percent of their current average costs if expanding with trade.

Figure 5.5 Percent improvement of “reasonable” cost investment plans over historic average costs

-100

-80

-60

-40

-20

0

20

40

60

80

100R

wanda

South

Afr

ica

Mozam

biq

ue

Tanzania

Zam

bia

Kenya

Lesoth

o

Uganda

Mala

wi

Eth

iopia

Madagascar

No-trade Trade

Source: Authors’ own calculation using data from the AICD database.

Hidden costs can be reduced by one-fourth just by improving power technologies and therefore

reducing production costs. These results suggest that current mispricing hidden costs are not a disguise

for postponed investment plans but rather a cover up for very high operational inefficiencies. In the

hopeful scenario of (good) scaling-up the sector, reduced “mispricing hidden costs” can potentially be a

significant source of fiscal space or resources in the power sector.

In summary, mispricing hidden costs in power (0.8 percent of GDP) are almost double the 0.5 percent

GDP hidden costs in water. While these aggregate levels might not seen outrageously high from a macro

perspective, they certainly are a substantial source of fiscal space. From a sectoral viewpoint, they hint at

very important issues that claim immediate attention. For instance, long-term sector sustainability would

like require revising tariff schedules and reducing cost inefficiencies by improving the quality of

investments and technologies.

In conclusion, tapping inefficiencies created by quasi-fiscal activities in power and water utilities

could increase the value for money and availability of fiscal resources by about 1.5 percent of GDP in an

average country. This translates into an estimate of over $9.5 billion across Sub-Saharan Africa. The

feasibility of raising tariffs toward cost-recovery levels is contingent on political and social

considerations. However, prices should start reflecting costs in order not to induce waste and

overconsumption, and if subsidies are needed they should be explicit, either in transfers to the utility or,

even better, subsidies targeted to those who actually need them. In the case of collection inefficiencies,

policy and administrative measures can bring short-term results and immediate cash-in benefits.

Obviously, this requires political commitment, first, in netting out payment arrears between government

agencies and utilities, and then creating without demagogy the culture of paying for services. Reducing


50

distribution inefficiencies passes through a period where more spending is inevitable before more

resources are made available. Distribution losses are a consequence of neglect because of poor

maintenance, lack of rehabilitation, and/or poor metering and billing, all of which require timely and

targeted spending to overcome.

Telecommunications utilities as social buffers

It is not uncommon that public utilities are used as social buffers redistributing wealth via excessive

employment. This is an evident sign of labor indiscipline. The dollar value of labor redundancies—or

hidden costs of excessive employment—is estimated to amount to up to 0.4 percent of GDP (Benin,

Mozambique and South Africa) or cost in excess that can reach $400 per subscriber (Chad) (table 5.4).

These estimates take as a norm the number of subscribers per employees seen in OECD fixed-line

operators.

Table 5.4 Monetary value of labor redundancies (annual average, 2001–06)

OECD benchmark 700 main lines per employee

Main lines per employee % of GDP $ per subscriber

Benin 20 0.40 91.0

Cameroon 15 0.15 47.9

Chad 63 0.14 488.9

Ethiopia 90 0.07 4.7

Kenya 19 0.00 0.4

Mozambique 138 0.41 69.6

Namibia 22 0.38 10.5

South Africa 17 0.43 0.3

Tanzania 30 0.31 123.2

Average 46 0.26 92.9

Source: Authors’ own calculation using data from the AICD Database.

Estimates only include countries with state-owned telecom utilities.

These striking results for labor inefficiencies underscore the importance of strengthening external

governance mechanisms that can impose discipline on SOEs behavior. The relation between increased

governance and lower labor costs attributed to inefficiencies is direct (figure 5.6).


51

Figure 5.6 Link between governance and cost of labor redundancy

y = -5.513x + 248.26

0

100

200

300

400

500

600

0 20 40 60 80 100

Governance Score

Cost

of

Lab

or

Red

undan

cy

(US

$/s

ubsc

riber

)

Source: AICD.

In conclusion, telecommunications labor inefficiencies are almost 0.3 percent of GDP on average,

equivalent to $2.1 billion. While public incumbents do not have to play an important role in providing

telephony services, the hidden costs of labor redundancies would be borne by governments if and when

the explosion of small private cellular providers continues and prices continue their downward trend. If

the regulatory framework is such to deter competitors and induce discriminatory practices, the hidden

labor costs would very likely then be transferred to users via higher tariffs.


52

6 Conclusions and policy implications

Sub-Saharan African countries are devoting substantial shares of GDP to infrastructure, but that does

not amount to much in absolute terms. Countries typically devote 6-12 percent of their GDP to

infrastructure, when all sources are taken into account. While these shares look relatively large compared

to middle- and high-income countries, due to very low levels of GDP, they amount to very little in

absolute terms. On average, low-income countries are spending less than $50 per capita per year, with

public investment being only a fraction of this.

There is a marked division of labor between state-owned enterprises (SOEs) and central government.

While SOEs account for the bulk of infrastructure spending in most countries, they undertake very little

capital spending. Most public investment on infrastructure continues to be undertaken through the central

government budget, with assets often transferred to SOEs for subsequent operation and maintenance.

Despite a favorable budget environment, only aid-dependent countries seem to allocate additional

resources to infrastructure. The combination of a commodity boom and widespread debt relief has created

substantial buoyancy in government budgets. In the case of aid-dependent countries, about 30 percent of

the additional envelope has been allocated to infrastructure. However, in middle-income countries almost

none of the additional budget has gone to infrastructure. Moreover, in oil-dependent countries,

infrastructure investment has actually fallen substantially even as resource revenues have surged.

Budget envelopes could be substantially increased by addressing substantial inefficiencies. Three

major sources of inefficiency have been identified in infrastructure spending: undermaintenance, budget-

execution failures, and hidden costs.

There is both direct and indirect evidence of undermaintenance leading to higher lifecycle costs of

infrastructure. On average, around 30 percent of the region’s infrastructure assets are in need of

rehabilitation. Given that the present value of rehabilitating infrastructure is substantially higher than that

associated with a sound preventive maintenance regime, this finding already indicates that, over time,

countries are spending more than they should to preserve a given infrastructure stock. In the case of the

road sector, it is possible to isolate maintenance expenditure with greater precision than elsewhere and to

compare it to rigorously determined country-specific benchmarks. Here it becomes apparent that roughly

half of the countries show shortfalls of 40 percent or more in maintenance spending.

Capital budget-execution ratios of only 66 percent on average indicate potential for a budget-neutral

50 percent increase in public investment. Deficiencies in planning, project preparation, and public

procurement conspire to create delays that prevent countries from spending more than two-thirds of the

public investment allocated to infrastructure in the budget. Addressing the causes of low budget execution

deserves very serious attention, since this alone could increase public investment by 50 percent without

any increase in budgeted resources. Moreover, until such deficiencies are addressed it will remain

difficult to achieve higher levels of investment even if more external resources are injected.

The hidden costs of power, water, and telecommunications utilities absorb around 1.8 percent of

GDP, presenting a major incentive to address inefficiency. A combination of underpricing,

undercollection of revenues, and very high distribution losses relative to technical norms lead to high


53

hidden costs of power and water. Underpricing is by far the largest contributor in both sectors, although

inefficiencies are also important. The dividend attached to solving these problems is very high—1.5

percent of GDP, and even more in some cases. Increasing efficiency would free up substantial resources

for public investment without increasing budget allocations.


54

References

Agénor, Pierre-R., and B. Moreno-Dodson. 2006. “Public Infrastructure and Growth: New Channels and

Policy Implications.” Mimeo, World Bank, Washington, DC.

Banerjee, S., V. Foster, Y. Ying, H. Skilling, and Q. Wodon. 2008. Achieving Cost Recovery, Equity and

Efficiency in Water Tariffs: Evidence From African Utilities. Africa Infrastructure Country

Diagnostic (AICD).

Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. Access,

Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa. AICD.

Benmaamar, M. 2006. Financing of Road Maintenance in Sub-Saharan Africa. Washington, DC: World

Bank.

Blanchard and Giavazzi. 2003. “Improving the Stability and Growth Pact Through a Proper Accounting

of Public Investment.” Mimeo.

Briceño-Garmendia, C. 2007. Fiscal Cost of Infrastructure Provision: A Practitioner’s Guide. AICD.

Briceño-Garmendia, C., and M. Shkaratan. 2008. Cost Recovery and Efficiency in Electricity Tariffs:

Evidence from African Utilities. AICD.

Calderón, C., and L. Servén. 2004. Trends in Infrastructure in Latin America. Policy Research Working

Paper 3400, World Bank, Washington, DC.

Canning, D., and E. Bennathan. 2000. The Social Rate of Return on Infrastructure Investments. Policy

Research Working Paper 2390, World Bank, Washington, DC.

Chivakul, M., and R. York. 2006. Implications of Quasi-Fiscal Activities in Ghana. International

Monetary Fund (IMF) Working Paper, Washington, DC.

Collier, P., and S. O’Connell. 2006. Opportunities and Choices in Explaining African Economic Growth.

Oxford: Center for Study of African Economies.

Collier and Gunning. 1999. “Why has Africa Grown Slowly?” Journal of Economic Perspectives 13(3).

De Haan, J., J. Sturm, and B. Sikken. 1996. “Government Capital Formation: Explaining the Decline.”

Weltwirtschaftliches Archiv 132 (1): 55–74.

Easterly, W., and L. Servén. 2003. The Limits of Stabilization: Infrastructure, Public Deficits and Growth

in Latin America. World Bank, Stanford University, Washington, DC.

Ebinger, J. 2006. Measuring Financial Performance in Infrastructure. Working Paper Series, World

Bank, Washington, DC.

Econ Analysis. 2008. Costing Power Infrastructure Investment Needs in Sub-Saharan Africa. AICD.

Estache, A. 2007. Managing the Interaction Between Infrastructure and Fiscal Policy for Faster Growth

in Africa. Background Note, AICD.


55

Estache, A., and M. E. Pinglo. August 2004. “Are Returns To Private Infrastructure In Developing

Countries Consistent With Risks Since the Asian Crisis?” Policy Research Paper 3373, World

Bank, Washington, DC.

Estache, A., and R. Muñoz. 2007. IMF Financial Programming versus the Growth Enhancing Power of

Education, Health and Infrastructure Investment: Lesson from Senegal and Uganda. AICD.

Estache, A., and Q. Wodon. 2006. Infrastructure and Poverty in Sub-Saharan Africa. Washington, DC:

World Bank.

Farrell, M. J. 1957. “The Measurement of Productive Efficiency.” Journal of the Royal Statistical Society,

Series A (General) 120(3).

Forsund, F., C. A. K Lovell, and P. Schmidt. 1980. “A Survey of Frontier Production Functions and of

Their Relationship to Efficiency Measurement.” Journal of Econometrics 13: 5–25, North-

Holland Publishing Company.

Foster, V., et al. 2008a. “Africa’s Power Supply Crisis: Unraveling the Paradoxes.” Chapter Contribution

to IMF Africa Regional Economic Outlook, Draft, Spring 2008.

Foster, V., W. Butterfield, and C. Chen. 2008b. Building Bridges: China’s Growing Role as

Infrastructure Financier for Africa. Public-Private Infrastructure Advisory Facility (PPIAF).

Gupta, S., K. Honjo, and M. Verhoeven. 1997. “The Efficiency of Government Expenditure: Experiences

from Africa.” IMF Working Paper, Washington, DC.

Gwilliam, K., and Z. Shalizi. 1996. Road Funds, User Charges and Taxes. Washington, DC: World Bank.

Heller, P. S., M. Katz, et al. 2006. “Making Fiscal Space Happen: Managing Fiscal Policy in a World of

Scaled-Up Aid.” IMF Working Paper 270, Washington, DC.

Herrera, S., and G. Pang. 2005. “Efficiency of Public Spending in Developing Countries: An Efficiency

Frontier Approach.” Policy Research Working Paper 3645, World Bank, Washington, DC.

Hicks, N. 1991. “Expenditure Reductions in Developing Countries Revisited.” Journal of International

Development 3(1): 29–37.

IMF. October 2007. Regional Economic Outlook: Sub-Saharan Africa. Washington, DC: IMF.

———. 2004. “Public Investment and Fiscal Policy.” Board Paper Prepared by Fiscal Affairs

Department, IMF, Washington, DC.

———. 2001. Government Finance Statistics Manual. Washington, DC: IMF.

(www.imf.org/external/pubs/ft/gfs/manual/).

Limão, N., and A. Venables. 2001. “Infrastructure, Geographical Disadvantage, Transport Costs, and

Trade.” The World Bank Economic Review 15(3): 451–479.

Makenzie, G. A., and P. Stella. 1996. Quasi-Fiscal Operations of Public Financial Institutions.

Washington, DC: IMF.


56

Ndulu B. J. 2004. “Infrastructure, Regional integration and Growth in Sub-Saharan Africa: Dealing with

the Disadvantage of Geography and Sovereign Fragmentation.” Background Paper Prepared for

the Commission for Africa, unpublished.

Park, B., L. Simar, and C. Weiner. 2000. “The FDH Estimator for Productivity Efficiency Scores:

Asymptotic Properties.” Economic Theory 16: 855–877.

Petri, M., T. Gunter, and A. Tsyninki. 2002. “Energy Sector Quasi-Fiscal Activities in Countries of the

Former Soviet Union.” IMF Working Paper, Washington, DC.

Petri, M., and G. Taube. 2003. “Fiscal Policy Beyond the Budget.” Emerging Markets Finance and Trade

39(1).

Roubini, N., and J. Sachs. 1989. “Government Spending and Budget Deficits in the Industrial Countries.”

Economic Policy 8: 99–132.

Saavalainen, T., and J. ten Berge. 2006. “Quasi-Fiscal Deficits and Energy Conditionality in Selected CIS

Countries.” IMF Working Paper, Washington, DC.

Sachs, J., J. McArthur, G. Schmidt-Traub, M. Kruk, C. Bahadur, M. Faye, and G. McCord. 2004.

“Ending Africa’s Poverty Trap.” Brookings Papers on Economic Activity.

Servén, L. 2005. “Fiscal Discipline, Public Investment and Growth.” Mimeo, World Bank.

Sirtaine, S., M. E. Pinglo, J. L. Guasch, and V. Foster. 2005. How Profitable Are Infrastructure

Concessions In Latin America? Empirical Evidence and Regulatory Implication. PPIAF.

United Nations. 2001. 1993 System of National Accounts. New York: United Nations.

(http://unstats.un.org/unsd/sna1993/).

Warlters, M., and E. Auriol. August 2005. “The Marginal Costs of Public Funds in Africa.” Policy

Research Paper 3679, World Bank, Washington, DC.

World Bank. 2005. Global Monitoring Report. Oxford University Press.

World Bank. 2006. Fiscal Policy for Growth and Development: An Interim Report. Development

Committee, Washington, DC: World Bank.


57

Appendixes

Appendix 1. Sector scope, functional classification

Appendix 2. Sources of data for data for expenditures of infrastructure

Appendix 3. Country groups

Appendix 41. Primary fiscal balances

Appendix 5. Net change in central government budget: breakdown by source

Appendix 6. Net change in central government budget: breakdown by use

Appendix 7. Expenditure on main road network

Appendix 8. Variance around the trend line of road expenditure

Appendix 9. Contributions to QFCs (country aggregates)—water sector

Appendix 10. Contributions to QFCs (country aggregates)—power sector

Appendix 11. Water: efficiency and production indicators, 2006

Appendix 12. Electricity: efficiency and production indicators, 2006

Appendix 13. Annual maintenance and preservation expenditures, 2001–05

Appendix 14. Average annual maintenance expenditures on main road network, 2001–05


58

Appendix 1. Sector scope, functional classification

Code Category Included in cross-country comparison

704 Economic affairs

7042 Agriculture, Forestry, Fishing and Hunting

Irrigation Systems (a share of 70421 Agriculture) no

7047 Other Industries

70474 Multipurpose development projects no

7043 Fuel and energy

70431 Coal and other solid mineral fuels yes

70432 Petroleum and natural gas yes

70433 Nuclear fuel yes

70434 Other fuels yes

70435 Electricity yes

70436 Nonelectric energy yes

7045 Transport

70451 Road transport yes

70452 Water transport no

70453 Railway transport no

70454 Air transport no

70455 Pipeline and other transport no

7046 Communication

70460 Communication yes

705 Environmental protection

7052 Waste water management yes

70520 Waste water management yes

706 Housing and community amenities

7063 Water supply

70630 Water supply yes


59

Appendix 2. Sources of data on infrastructure expenditures

General government sector Nonfinancial corporations sector

Source of documents Consolidated budget of general government (or separate budget books of each government unit):

Recurrent budget

Development budget

of which external development budget

Special funds

Financial statements of public nonfinancial corporations:

Balance sheet

Income statement

Cash flow statement

Stages Estimates, releases, actuals Actuals

Institutional categories Central government, local government

Economic classification GFMS2001

Source Ministry of Finance SOE

Sources of expenditures Externally funded (donor) expenditures are identified separately

Not available (e.g., what share of expenditures by borrowing)

Appendix 3. Country groups46

Oil-exporting countries

Countries where net oil exports make up 30 percent or more of total exports. Oil exporters are

classified as such even if they would otherwise qualify for another group

Middle-income countries

Countries that are not oil exporters and have per capita income higher than $905, according to 2006

gross national income per capita as calculated by the World Bank.

Non-fragile low-income countries

Countries that are not oil exporters and have per capita income equal or lower than $905 and a score

higher than 3.2 on the Country Policy and Institutional Performance Assessment of the World Bank,

following the classification in the 2007 Global Monitoring Report.

Non-fragile low-income countries

The remaining countries of the sample.

46 Adapted from IMF (2007) p.4.


60

Appendix 4. Primary fiscal balances

GDP share Domestic revenues Grants

Budgetary expenditures

excl. debt service Primary balance

Primary balance excl.

grants

Benin 16.26 2.05 19.09 (0.79) (2.84)

Burkina Faso 11.86 5.28 20.50 (3.37) (8.65)

Cameroon 16.63 0.41 13.40 3.64 3.24

Cape Verde 22.36 7.48 31.42 (1.57) (9.06)

Chad 8.94 3.18 16.88 (4.77) (7.94)

Congo, Dem. Rep. 8.64 1.92 10.46 0.10 (1.82)

Côte d'Ivoire 13.81 0.61 13.05 1.37 0.76

Ethiopia 16.95 4.99 25.27 (3.32) (8.32)

Ghana 20.90 5.27 24.69 1.48 (3.79)

Kenya 18.48 1.42 18.82 1.08 (0.34)

Lesotho 45.59 2.59 43.75 4.43 1.85

Madagascar 10.29 5.03 25.10 (9.79) (14.81)

Malawi 18.92 8.66 28.08 (0.50) (9.16)

Mozambique 12.79 9.49 24.46 (2.18) (11.67)

Namibia 31.28 0.12 32.76 (1.36) (1.48)

Niger 10.79 5.76 18.39 (1.84) (7.60)

Nigeria 41.63 - 35.24 6.39 6.39

Rwanda 12.77 9.48 22.31 (0.05) (9.54)

Senegal 18.09 1.82 20.88 (0.97) (2.79)

South Africa 23.24 - 20.67 2.57 2.57

Tanzania 11.42 5.07 17.43 (0.94) (6.01)

Uganda 9.33 5.97 16.85 (1.55) (7.52)

Zambia 17.93 9.65 25.62 1.96 (7.70)

Source: AICD, Fiscal Database, 2008; IMF Statistical Appendixes, WB DDP

Note: Annual averages over the period 2001–06

Averages weighted by countries’ GDP. Totals may not add up.


61

Appendix 5. Net change in central government budget: breakdown by source

GDP share CG net expenditure

budget Domestic revenues of which oil revenues Grants Net borrowing

Benin (1.07) 0.42 - 1.06 (2.54)

Burkina Faso (0.10) 0.63 - (0.89) 0.16

Cameroon 0.05 1.11 0.15 0.22 (1.28)

Cape Verde 0.73 (0.08) - (1.91) 2.72

Chad (6.59) 1.27 4.22 (2.75) (5.11)

Congo, Dem. Rep. 9.06 3.63 - 4.84 0.59

Côte d'Ivoire 1.54 (0.47) 0.88 0.59 1.42

Ethiopia (0.23) (0.21) - 0.93 (0.95)

Ghana 7.14 5.80 - 2.11 (0.77)

Kenya 1.12 (1.00) - 1.21 0.90

Lesotho (3.26) 6.57 - (3.06) (6.77)

Madagascar 2.87 0.77 - 1.85 0.25

Malawi 7.53 7.71 - 5.28 (5.46)

Mozambique (3.87) 1.15 - (4.01) (1.01)

Namibia (1.89) 1.27 - (0.10) (3.06)

Niger 1.83 (0.57) - 2.38 0.03

Nigeria (4.68) 6.80 10.44 - (11.48)

Rwanda 3.37 0.59 - 5.15 (2.38)

Senegal 4.73 1.42 - (0.00) 3.32

South Africa 4.32 3.45 - - 0.87

Tanzania 6.64 1.43 - 3.10 2.11

Uganda (2.79) (0.66) - (0.32) (1.82)

Zambia (6.38) (1.05) - 17.43 (22.76)


Note: Net change, 2001–06. Averages weighted by countries’ GDP. Totals may not add up.


62

Appendix 6. Net change in central government budget: breakdown by use

GDP share

CG net expenditure

budget Wages Other current of which

Infrastructure Capital

expenditure of which

infrastructure

Benin (1.07) 1.06 (2.24) 0.04 0.11 0.20

Burkina Faso (0.10) 0.13 (0.74) - 0.51 -

Cameroon 0.05 (0.58) 0.63 (0.01) 0.01 0.71

Cape Verde 0.73 1.89 (0.87) (0.04) (0.29) (0.50)

Chad (6.59) (1.21) (2.31) 0.21 (3.08) 0.18

Congo, Dem. Rep. 9.06 2.32 3.90 - 2.84 -

Côte d'Ivoire 1.54 0.11 1.86 0.13 (0.43) 0.32

Ethiopia (0.23) 0.22 (2.25) (0.04) 1.80 2.16

Ghana 7.14 (0.00) 1.26 (0.09) 5.88 0.81

Kenya 1.12 (0.03) (0.17) (0.00) 1.31 (0.30)

Lesotho (3.26) (1.17) 0.98 0.10 (3.08) 0.01

Madagascar 2.87 0.04 (0.14) (0.02) 2.98 2.04

Malawi 7.53 0.74 3.25 (2.11) 3.54 0.30

Mozambique (3.87) 0.38 0.06 0.01 (4.31) 0.44

Namibia (1.89) 0.60 (0.98) (0.04) (1.51) (0.01)

Niger 1.83 (0.31) (0.66) 0.26 2.80 0.18

Nigeria (4.68) (2.36) (0.27) 0.04 (2.05) (2.20)

Rwanda 3.37 (1.35) 0.81 0.09 3.91 0.73

Senegal 4.73 0.23 1.95 (1.61) 2.55 (0.20)

South Africa 4.32 (0.13) 5.17 (0.02) (0.71) 0.04

Tanzania 6.64 0.25 2.80 1.00 3.59 1.13

Uganda (2.79) (0.30) 0.98 (0.08) (3.47) 0.87

Zambia (6.38) (0.79) 2.33 0.73 (7.91) (1.97)


Note: Net change, 2001–06. Averages weighted by countries’ GDP. Totals may not add up.


63

Appendix 7. Expenditure on main road network

Length of main network Maintenance expenditures (**)

Rehabilitation/capital expenditures (***)

(km) Annual Averages in $ per km

Benin 4,735 $ 3,015.81 $ 4,306.67

Burkina Faso 10,231 n.a. n.a.

Cameroon 11,008 $ 2,608.66 $ 5,823.50

Chad n.a. n.a.

Côte d'Ivoire 13,291 n.a. $ 9,016.45

Ethiopia 9,721 $ 2,599.32 $ 19,524.64

Ghana 11,139 $ 2,339.67 $ 11,557.48

Kenya 22,341 $ 2,872.55 $ 1,176.95

Lesotho 3,005 $ 1,255.09 $ 8,353.15

Madagascar 9,599 $ 1,450.28 $ 11,851.26

Malawi 3,444 $ 1,960.15 $ 6,952.26

Mozambique 9,808 $ 2,371.65 $ 12,209.19

Namibia 16,362 $ 3,462.30 $ 166.70

Niger 6,055 $ 493.59 $ 4,137.90

Nigeria $ 16,963.74

Rwanda 2,836 $ 491.75 $ 6,466.14

Senegal 4,780 n/a $ 24,937.63

South Africa (*) 76,288 $ 2,074.64 $ 708.23

Tanzania 28,730 $ 1,433.37 $ 4,162.02

Uganda 9,171 n/a $ 14,367.39

Zambia 3,290 $ 3,029.45 $ 36,348.70

Averages (weighted according to the size of network)

Resource-intensive countries $ 2,852.39 $ 13,243.52

Non-resource-intensive (excluding South Africa) $ 2,001.27 $ 9,973.38

Aid dependent $ 2,020.88 $ 11,342.76

HIPC countries (end of 2005) $ 1,985.69 $ 11,470.62

Middle-income countries $ 2,134.58 $ 684.84

Average $ 2,076.51 $ 8,384.69

Average excluding South Africa $ 2,077.23 $ 10,769.11

Source: AICD Fiscal Baselines, 2007. Primary network length data from RONET. South Africa, expenditures of SANRAL.

Note: Based on annual averages for the period 2001–05

n.a. - time series were not long enough to calculate the standard errors.

(*) Excludes local governments )

(**) Includes expenditures of roads fund plus current expenditures of central government (current expenditures excluding transfers to local municipalities)

(***) Capital expenditures of central government and externally funded expenditures


64

Appendix 8. Variance around the trend line of road expenditure

Current expenditures (*) Capital expenditures (**)

Road funds Central government External Central government

Benin 16.50 44.01 36.51 62.39

Cameroon 23.91 9.87 23.17 19.17

Chad 2.06 21.84 33.42 79.40

Côte D'Ivoire n.a. 57.44 n.a. 12.86

Ethiopia 3.59 n.a. n.a. 13.48

Ghana 10.89 36.77 19.51 20.08

Kenya 4.14 3.61 n.a. 12.85

Madagascar 27.12 43.12 24.65 16.92

Malawi 10.02 78.18 n.a. 45.17

Namibia 4.00 n.a. n.a. n.a.

Niger 19.59 n.a. n.a. n.a.

Rwanda 21.79 16.16 38.18 50.64

Senegal n.a. 102.67 18.74 57.93

Tanzania n.a. 4.54 28.83 8.09

Uganda n.a. n.a. n.a. n.a.

Mozambique 7.92 13.02 32.32 23.25

Nigeria n/a 47.36 n.a. 15.10

Lesotho 11.18 9.42 9.25 35.66

Zambia n.a. 12.05 25.65

Weighted averages (according to size of network)

Resource-intensive countries 13.55 38.29 27.18 23.05

Non-resource-intensive (excluding South Africa)

10.22 22.02 29.59 27.59

Aid dependent 11.13 22.07 31.80 28.06

Weighted average (size of network)

10.59 26.61 29.30 26.36

Simple average 12.52 33.34 26.46 31.16

Source: AICD Fiscal Baselines, 2007


n.a. - time series were not long enough to calculate the standard errors.

(*) Includes expenditures of roads fund plus current expenditures of central government (current expenditures excluding transfers to local municipalities)

(**) Capital expenditures of central government and externally funded expenditures


65

Appendix 9. Contributions to QFCs (country aggregates)—water sector

Total hidden

cost Unaccounted

losses 2/ Underpricing 3/

Collection inefficiencies

4/ Unaccounted

losses Underpricing Collection

inefficiencies

Country Share of GDP (%) Share of total QFC (%) Share of GDP (%)

Tanzania 0.1 33.1 35.8 31.1 0.0 0.0 0.0

Nigeria 0.2 19.7 45.7 34.6 0.0 0.1 0.1

Benin 0.2 15.1 73.0 11.9 0.0 0.1 0.0

Cape Verde 0.3 15.3 0.0 84.7 0.0 0.0 0.2

Ethiopia 0.3 26.8 70.8 2.4 0.1 0.2 0.0

Uganda 0.3 27.7 0.0 72.3 0.1 0.0 0.2

Namibia 0.3 0.1 73.5 26.4 0.0 0.2 0.1

Kenya 0.3 28.2 60.0 11.8 0.1 0.2 0.0

Burkina Faso 0.3 0.0 22.6 77.4 0.0 0.1 0.2

Rwanda 0.3 29.7 34.4 35.9 0.1 0.1 0.1

Sudan 0.4 26.1 27.3 46.5 0.1 0.1 0.2

Niger 0.4 0.0 95.2 4.8 0.0 0.4 0.0

South Africa 0.5 12.5 77.0 10.4 0.1 0.4 0.0

Lesotho 0.5 17.0 83.0 0.0 0.1 0.4 0.0

Mozambique 0.5 49.2 50.4 0.4 0.3 0.3 0.0

Côte d'Ivoire 0.8 2.3 97.7 0.0 0.0 0.8 0.0

Senegal 0.9 0.2 99.8 0.0 0.0 0.9 0.0

Madagascar 0.9 20.3 79.7 0.0 0.2 0.7 0.0

Zambia 1.1 28.1 32.7 39.1 0.3 0.3 0.4

Malawi 1.2 23.6 43.0 33.4 0.3 0.5 0.4

Ghana 1.2 42.7 30.6 26.7 0.5 0.4 0.3

Congo, D.R. 1.3 40.8 27.2 32.0 0.5 0.4 0.4


2/ Unaccounted Losess = (End-User Consumption) * (Average Cost Recovery Price)* (Total Loss Rate - Normative Loss Rate)/(1-Normative Loss Rate)

3/ Underpricing = End-User Consumption * (Average Cost Recovery Price - Average Actual Tariff).

4/ Collection Inefficiencies = End-user Comsumption* Average Actual Tariff* (1 - Collection rate)


66

Appendix 10. Contributions to QFCs (country aggregates)—power sector

Total Hidden Cost Unaccounted

Losses 2/ Underpricing

3/

Collection Inefficiencies

4/ Unaccounted

Losses Underpricing Collection

Inefficiencies

Country Share of GDP (%) Share of total QFC (%) Share of GDP (%)

South Africa 0.00 100.0 0.0 0.00 0.00 0.00 0.0

Benin 0.24 61.7 0.0 38.26 0.15 0.00 0.1

Kenya 0.47 60.4 39.6 0.00 0.28 0.19 0.0

Mozambique 0.67 45.1 14.0 40.91 0.30 0.09 0.3

Chad 1.00 28.5 71.5 0.00 0.29 0.72 0.0

Cape Verde 1.18 26.1 45.3 28.56 0.31 0.53 0.3

Madagascar 1.31 28.9 71.1 0.00 0.38 0.93 0.0

Lesotho 1.37 22.7 61.9 15.48 0.31 0.85 0.2

Nigeria 1.42 26.1 33.4 40.51 0.37 0.47 0.6

Burkina Faso 1.47 24.6 0.0 75.38 0.36 0.00 1.1

Rwanda 1.74 23.2 76.8 0.00 0.40 1.33 0.0

Ethiopia 1.78 30.0 70.0 0.00 0.54 1.25 0.0

Uganda 1.85 40.8 0.0 59.17 0.75 0.00 1.1

Cameroon 2.22 28.2 71.8 0.00 0.62 1.59 0.0

Zambia 2.33 3.5 96.5 0.00 0.08 2.25 0.0

Tanzania 2.41 24.7 75.3 0.00 0.59 1.82 0.0

Senegal 2.47 24.9 50.4 24.73 0.62 1.24 0.6

Ghana 3.14 25.7 0.0 74.32 0.81 0.00 2.3

Niger 3.51 22.8 49.3 27.96 0.80 1.73 1.0

Malawi 4.36 17.4 76.6 6.0 0.8 3.3 0.3

DRC 4.66 27.5 0.0 72.51 1.28 0.0 3.4


2/ Unaccounted Losses = (End-User Consumption) * (Average Cost Recovery Price) * (Total Loss Rate - Normative Loss Rate) / (1-Normative Loss Rate)

3/ Underpricing = End-User Consumption * (Average Cost Recovery Price - Average Actual Tariff)

4/ Collection Inefficiencies = End-user Comsumption* Average Actual Tariff* (1 - Collection rate)


67

Appendix 11. Water: efficiency and production indicators, 2006

Country Utility

End user consumption

(M3/year)

Average cost

recovery tariff

(USD/m3)

Average effective tariff (USD/m3) 5/

NRW (%)

Implicit collection ratio (%) 6/

Benin SONEB 22,977,000 1.10 0.85 24 95

Burkina Faso ONEA 33,817,240 1.15 1.03 18 60

Cameroon SNEC — 1.06 0.67 35 —

Cape Verde ELECTRA 2,931,781 3.24 4.20 31 81

Chad STEE — 1.06 0.29 35 33

Côte d'Ivoire SODECI 128,853,000 1.03 0.09 22 100

DRC REGIDESO 137,052,336 1.10 0.91 41 75

Ethiopia ADAMA 2,660,048 0.72 0.39 43 89

Ethiopia AWSA 52,395,813 0.72 0.33 37 96

Ethiopia DIRE DAWA 2,565,138 0.58 0.24 22 96

Ghana GWC 130,463,393 1.06 0.71 53 63

Kenya KIWASCO 1,800,000 1.53 0.81 71 79

Kenya MWSC 12,516,164 0.89 0.67 38 21

Kenya NWASCO 97,869,588 0.56 0.24 38 100

Lesotho WASA 11,213,684 1.10 0.58 28 100

Madagascar JIRAMA 63,756,027 0.87 0.29 34 100

Malawi BWB 14,185,900 0.97 0.40 51 100

Malawi CRWB 4,722,000 0.81 0.32 17 57

Malawi LWB 23,600,000 0.63 0.74 22 57

Mozambique AdeM Beira 3,771,240 0.91 0.52 60 100

Mozambique AdeM Maputo 25,432,162 1.13 0.52 62 100

Mozambique AdeM Nampula 2,447,001 0.75 0.52 44 98

Mozambique AdeM Pemba 1,214,471 0.93 0.52 45 82

Mozambique AdeM Quilimane 978,095 0.82 0.52 35 100

Namibia Oshakati Municipality 1,031,884 1.84 1.98 21 81

Namibia Walvis Bay Municipality 3,626,811 1.06 0.97 16 100

Namibia Windhoek Municipality 17,500,000 2.48 1.73 14 85

Niger SPEN 33,791,483 0.86 0.49 19 96

Nigeria Borno — 1.06 0.67 35 31

Nigeria FCT 17,300,000 1.06 0.53 80 31

Nigeria Kaduna 63,690,000 1.06 0.22 21 31

Nigeria Katsina 31,000,000 0.46 0.26 14 24

Nigeria Lagos 50,405,903 0.66 0.67 57 31

Nigeria Plateau 14,660,000 1.06 0.67 23 31

Rwanda ELECTROGAZ 9,780,000 0.91 0.67 38 63

Senegal ONAS nap 1.06 0.67 35 87


68

Senegal SDE 99,640,165 1.25 0.50 20 100

South Africa Cape town metro 215,660,000 1.61 0.47 18 100

South Africa Drakenstein Municipality 13,464,003 1.10 0.35 12 100

South Africa eThekwini Metro (Durban) 200,000,000 1.96 1.42 32 79

South Africa Joburg 328,246,350 1.90 0.32 31 45

Sudan Khartoum Water Corporation 150,000,000 0.68 0.50 40 38

Sudan South Darfur Water Corporation 3,700,000 0.89 0.86 49 59

Sudan Upper Nile Water Corporation 1,788,500 1.13 0.80 29 48

Tanzania DAWASCO — 1.06 0.61 35 30

Tanzania DUWS 5,110,000 0.82 0.54 30 63

Tanzania MWSA 7,432,028 0.59 0.47 49 73

Uganda NWSC 38,000,000 1.00 1.01 34 53

Zambia LWSC 35,346,257 0.67 0.53 54 43

Zambia NWSC 73,799,000 0.60 0.35 37 33

Zambia SWSC 8,136,000 0.70 0.40 55 33



69

Appendix 12. Electricity: efficiency and production indicators, 2006

Volume electricity billed

(GWh/year)

Historical normative cost recovery tariff ($) – average

total cost

Average effective tariff (USD/kwH) 1/

System losses

(% electricity produced)

Implicit collection ratio

(%) 2/

Benin SBEE 506.6 0.14 0.15 18.25 95

Burkina Faso SONABEL 537.5 0.24 0.31 24.68 63

Cameroon AES SONEL 3263.7 0.14 0.06 30.79 100

Cape Verde Electra 161.7 0.23 0.20 17.32 90

Chad STEE 78.0 0.84 0.30 33.20 100

Côte d'Ivoire CIE 5741.9 0.34 — — 77

DRC Congo SNEL 4091.8 0.07 0.10 40.00 42

Ethiopia EEPCO 2069.2 0.09 0.02 40.00 100

Ghana VRA 3762.0 0.13 0.15 25.44 55

Kenya KPLC 4379.0 0.14 0.13 18.10 53

Lesotho LEC 353.6 0.11 0.08 20.00 89

Madagascar JIRAMA 753.7 0.17 0.10 23.74 100

Malawi Escom 1055.0 0.10 0.04 23.00 86

Mozambique EDM 1306.7 0.09 0.09 25.00 84

Namibia NamPower 3363.0 0.00 — 12.00 —

Namibia NORED 131.4 — — 18.00 —

Niger NIGELEC 333.1 0.43 0.26 27.00 61

Nigeria PHNC 21401.9 0.08 0.05 30.00 52

Rwanda ELECTROGAZ 157.4 0.38 0.20 23.00 100

Senegal SENELEC 1710.0 0.24 0.18 21.19 84

South Africa ESKOM 256959.0 0.06 0.08 10.03 100

Sudan NEC — — — — —

Tanzania TANESCO 2628.0 0.16 0.07 26.00 100

Uganda UEDCL 1806.4 0.12 0.14 36.00 63

Zambia ZESCO 3516.0 0.08 0.03 12.00 100

Source: AICD Database.


1/ Average Effective Tariff = (Effective Residential Tariff at 100 kwh)*{Residential Share on Consumption / Residential Share on Revenue} 2/ Implicit Collection Ratio = (Average Revenue/Average Effective Tariff) or (Reported Non Payment)


70

Appendix 13. Annual maintenance and preservation expenditures, 2001–05

Km of primary in fair and good condition

Maintenance expenditure (**) (US$ millions)

Preservation norms (***)

(US$ millions)

Benin 911.32 $ 16.43 $ 27.20

Cameroon 2,372.09 $ 43.05 $ 57.82

Chad 862.42 $ 3.88 $ 23.49

Côte D'Ivoire 4,545.94 $ 0.68 $ 61.28

Ethiopia 4,122.02 $ 43.82 $ 57.11

Ghana 6,284.60 $ 71.73 $ 103.03

Kenya 6,657.12 $ 110.36 $ 122.96

Lesotho 676.72 $ 9.61 $ 22.62

Madagascar 3,804.92 $ 19.15 $ 47.53

Malawi 2,567.75 $ 15.67 $ 35.42

Mozambique 4,819.70 $ 67.73 $ 75.94

Namibia 5,589.24 $ 101.81 $ 85.45

Niger 3,250.09 $ 8.24 $ 29.86

Rwanda 945.43 $ 5.79 $ 8.57

Senegal 2,141.99 $ 20.00 $ 47.53

South Africa 59,665.02 $ 390.54 $ 537.79

Tanzania 4,100.02 $ 128.48 $ 123.11

Uganda 2,119.06 $ 24.87 $ 91.10

Zambia 3,989.00 $ 86.06 $ 65.31

Notes: Primary network length and road quality data from RONET. South Africa, expenditures of SANRAL



(***) RONET data


71

Appendix 14. Average annual maintenance expenditures on main road

network, 2001–05

Length of main

network (km)

Road fund expenditure

on main roads (*) ($)

Maintenance expenditures

on main roads (**) ($ per km per year)

Rehabilitation

expenditures (***) ($)

Benin 4,735 $ 14,279,046 $ 3,015.81 $ 20,390,875

Burkina Faso 10,231 n.a. n.a. n.a.

Cameroon 11,008 $ 28,714,978 $ 2,608.66 $ 64,102,491

Chad n.a. n.a. n.a. $ 100,431,165

Côte d'Ivoire 13,291 n.a. n.a. $ 119,835,431

Ethiopia 9,721 $ 25,267,135 $ 2,599.32 $ 189,792,767

Ghana 11,139 $ 26,061,377 $ 2,339.67 $ 128,737,666

Kenya 22,341 $ 64,176,053 $ 2,872.55 $ 26,294,362

Lesotho 3,005 $ 3,770,923 $ 1,255.09 $ 25,097,111

Madagascar 9,599 $ 13,921,492 $ 1,450.28 $ 113,762,596

Malawi 3,444 $ 6,750,000 $ 1,960.15 $ 23,940,957

Mozambique 9,808 $ 23,262,139 $ 2,371.65 $ 119,752,630

Namibia 16,362 $ 56,650,000 $ 3,462.30 $ 2,727,596

Niger 6,055 $ 2,988,755 $ 493.59 $ 25,055,751

Rwanda 2,836 $ 1,394,509 $ 491.75 $ 18,336,585

Senegal 4,780 n.a. n/a $ 119,199,012

Tanzania 28,730 $ 41,180,461 $ 1,433.37 $ 119,573,801

Uganda 9,171 n/a n/a $ 131,767,055

Zambia 3,290 $ 9,966,882. $ 3,029.45 $ 119,587,211

Arithmetic average $ 2,097.22 $81,576,948

South Africa (*) 76,288 $ 158,270,950.42 $ 2,074.64 $ 119,199,012.08

Notes: Primary network length data from RONET. South Africa, expenditures of SANRAL


Notes: Primary network length data from RONET. South Africa, expenditures of SANRAL

(*) Excludes local governments)


(***) Capital expenditures of central government and externally funded expenditures

Date post:	24-Aug-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

Financing Public Infrastructure in Sub-Saharan Africa ... · recognized the importance of scaling...

Documents