Component2 LGRevenueStudy Final 1

Report No.
AUS5340
ETHIOPIA
Ethiopia Local Government Revenue Study

Part II – A Situational Analysis of Urban Local Governments in
Ethiopia: The Institutional, Governance and Fiscal Context of
Local Governments
June 2014
Urban Development and Services Practice 1 (AFTU1)

Sustainable Development Network
Country Department 3, AFCE3
Africa Region
Document of the World Bank

For Official Use Only
Ethiopian Local Government Revenue Study
Part II - A Situational Analysis of Urban Local Governments in Ethiopia
Abbreviations and Acronyms

AACG Addis Ababa City Government
AF Additional Finance
ARAP Abbreviated Resettlement Policy Framework
BOFED Bureau of Finance and Economic Development (Regional)
BTIUD Bureau of Trade, Industry and Urban Development (Regional)
CB Capacity Building
CBDSD Capacity Building for Decentralized Service Delivery Project
CD Capacity Development
CIP Capital Investment Plan
COA Chart of Accounts
CSA Central Statistics Agency
DDCA Dire Dawa City Administration
EC Ethiopian Calendar
EFY Ethiopia fiscal year
EMP Environmental management plan
EOI Expression of Interest
ERCA Ethiopian Revenues and Customs Authority
ESMF Environmental and Social Management Framework
ETB Ethiopian Birr
GC Gregorian calendar
GIZ German Society for International Cooperation (Deutsche Gesellschaft für Internationale
Zusammenarbeit)
IBEX Integrated Budget and Expenditure System
ICB International Competitive Bidding
ICR Implementation Completion and Results Report
IDA International Development Association
IFR Interim Financial Report
IGFTS Intergovernmental Fiscal Transfer System
IMF International Monetary Fund
LG Local governments
MOFED Ministry of Finance and Economic Development (Federal)
MTR Midterm review
MUDC Ministry of Urban Development and Construction (Federal)
NBE National Bank of Ethiopia
O&M Operations and maintenance
OFED Office of Finance and Economic Development (city level)
OM Operational Manual for ULGDP
PBG Performance-Based Grant
PBGS Performance-Based Grant System
PBTS Performance-Based Transfer System
PM Performance Measures
PPA Participation and Performance Agreement
PSCAP Public Sector Capacity Building Program Support Project
QCBS Quality and Cost Based Selection
ii
RAP Resettlement action plan

REOI Request for Expressions of Interest
REP Revenue Enhancement Plan (AACG)
REPA Regional Environmental Protection Agency
RPF Resettlement Policy Framework
SNNPR Southern Nations, Nationalities, and People’s Region
SOE Statement of Expenditure
TOR Terms of Reference
TP Technical Proposal
UGCBB Urban Governance and Capacity Building Bureau (Federal)
ULG Urban Local Government
ULGDP Urban Local Government Development Project
UNCDF United Nations Capital Development Fund
UNDB United Nations Development Business
USD United States Dollar
WB World Bank
Regional Vice President: Makhtar Diop

Country Director: Guang Zhe Chen
Sector Director: Jamal Saghir
Sector Manager: R. Mukami Kairuki
Task Team Leader: Onur Ozlu
iii
Contents
1. Introduction ................................................................................................................................... 1
1.1 Background ........................................................................................................................................................................... 1
1.2 Objective ................................................................................................................................................................................. 2
1.3 Methodology ......................................................................................................................................................................... 3
1.4 Organization of this Report ............................................................................................................................................ 4
PART I: INSTITUTIONAL AND GOVERNANCE CONTEXT ....................................................... 5
2. Overview of the institutional relationships between federal, regional and local governments ....... 5
2.1 Overview of the intergovernmental institutional structure of the public sector ......................................... 5
2.2. An overview of functional assignments and local regulatory powers ............................................................. 8
2.3 Overview of the intergovernmental administrative structure of the public sector ................................. 11
3. Relations between regional and city governments ..................................................................... 15
3.1 De jure autonomy of urban local governments versus de facto interdependence ................................... 15
3.2 Regional and City/Municipal legislation on municipal revenues .................................................................... 16
3.3 Regional Oversight and Support for City Financial Management .................................................................... 18
3.4 Regional Oversight Over Urban Development ......................................................................................................... 18
3.5 Regional Auditing of Urban Local Governments Accounts ................................................................................. 19
PART II: THE FISCAL POSITION OF LOCAL GOVERNMENTS .............................................. 22
4. An overview of the intergovernmental fiscal system and sub-national finances in Ethiopia ........ 22
4.1 Local Public Sector Expenditure Profile ..................................................................................................................... 22
4.2 Local Public Sector Revenue Profile............................................................................................................................. 25
4.3 Spending on Urban Development ............................................................................................................................. 31
4.4 Issues arising for the next phase of ULGDP .......................................................................................................... 36
4.5 Chapter Annex (Data tables in US$) ............................................................................................................................. 37
5. An analysis of intergovernmental fiscal flows ............................................................................... 39
5.1 Overview of federal-regional intergovernmental transfer system ................................................................. 39
5.2 Overview of regional-local intergovernmental transfer system ...................................................................... 46
5.3 Overview of urban grants under ULGDP .................................................................................................................... 48
5.4 Issues arising for the next phase of ULGDP .............................................................................................................. 53
6. An Analysis of Municipal Own Source Revenues......................................................................... 54
6.1 Overview of municipal revenue assignment ............................................................................................................ 54
6.2 Regional differences in the local revenue context ............................................................................................. 56
6.3 An analysis of municipal own source revenue collections ................................................................................. 57
6.4 Conceptual discussion: local revenue potential ...................................................................................................... 67
6.5 Issues arising for the next phase of ULGDP .............................................................................................................. 69
7. A review of local government borrowing systems ........................................................................ 71
7.1 Borrowing by regional governments........................................................................................................................... 71
7.2 Local borrowing by sub-regional urban local governments .............................................................................. 73
PART III: CITY MANAGEMENT PRACTICES............................................................................. 74
8. Participatory planning, formulation of CIPs and Transparent Governance ................................... 75
iv
8.1 Policies and Guidance for Urban Planning / CIPs................................................................................................... 75

8.2 Measurement of CIP and Planning Procedures in APA ........................................................................................ 76
8.3 City Achievements and Challenges for Improved Planning ............................................................................... 79
8.4 Measurement of Transparency and Accountability in APA ............................................................................... 80
8.5 Issues Arising for Future Improvement of APA ...................................................................................................... 81
9. Local Government Financial Management .................................................................................... 83
9.1 Key policies and Guidance for PFM in Cities............................................................................................................. 83
9.2 Key Finding on Financial Management practices in cities .................................................................................. 85
9.3 Internal audit ......................................................................................................................................................................... 88
9.4 Measurement of Financial Management in APA ..................................................................................................... 89
9.5 Issues Arising for Future Improvement of APA ...................................................................................................... 90
10. An assessment of Asset Management and O&M Planning ......................................................... 92
10.1 Current procedures of assessment of asset management and O&M in APA ............................................ 92
10.2 Observations on Asset management and O&M from the Field Visits .......................................................... 93
10.3 Issues Arising for Future Improvement of APA.................................................................................................... 94
11. An assessment of procurement systems ....................................................................................... 95
11.1 Current Assessment of Procurement in the APA ................................................................................................. 95
11.2 Issues Arising for Future Improvement of APA.................................................................................................... 96
12. Local revenue management ......................................................................................................... 97
12.1 Key policies and Guidance for Revenue Management in Cities ..................................................................... 97
12.2 Observations on Revenue Management Practices in cities ............................................................................. 98
12.3 Variation in Local Tax Rates and City Policy Decisions on Local revenue................................................. 99
12.4 Measurement of Revenue Management in APA ................................................................................................ 100
12.5 Issues Arising for Future Improvement of APA................................................................................................. 102
PART IV: INITIAL IMPLICATIONS FOR THE DESIGN OF URBAN GRANT ....................... 105
13. Initial Implications for the design of a Performance-Based Urban Grant System..................... 105
13.1 Sustainability of the performance-based urban grant program ................................................................. 107
13.2 Integrating ULGDP further into the intergovernmental finance architecture ...................................... 108
13.3 Managing funding flows across all levels of government .............................................................................. 111
13.4 Building the capacity at the local level .................................................................................................................. 114
13.5 Stimulation of local governments performance by urban development grants .................................. 118
13.6. Implications for the role of regional-level institutions.................................................................................. 121
Annexes ............................................................................................................................................ 123
Annex 1: References ................................................................................................................................................................ 123
Annex 2: Institutions Consulted ......................................................................................................................................... 126
Annex 4: Example of City Tariffs as Regulated by a Region.................................................................................... 139
v
Tables
Table 1 Sample Local Governments for Component 2...................... Error! Bookmark not defined.
Table 2.1. Regional population and urban population (July 2013) ...................................................... 8
Table 3.1 Degree of Autonomy for City Management of Revenues ................................................. 17
Table 4.1: Public Sector Expenditures in Ethiopia, EFY 2004 .......................................................... 23
Table 4.2. Revenue profile for Ethiopia (actual revenue collections, EFY 2004) ............................. 26
Table 4.3. State and municipal revenue collections by regional state (EFY 2004, ETB mn) ........... 28
Table 4.4. State and municipal revenue collections by regional state (EFY 2004, ETB per capita) . 30
Table 4.5: Urban development spending at different government levels, EFY 2004 ........................ 31
Table 4.6: Regional and Local Urban Development Spending by Region, EFY 2004 (ETB mn) .... 33
Table 4.7: Regional and Local Urban Development Spending by Region, EFY 2004 (ETB per
capita) ................................................................................................................................................. 33
Table 4.8: Regional and Local Urban Development Spending by Region, EFY 2004 (ETB per urban
resident) .............................................................................................................................................. 34
Table 5.1: Revenue Sources of Regional States: State Revenues, Subsidies (Grants), and External
Funding (EFY 2003) .......................................................................................................................... 39
Table 5.2: Regional revenue composition, EFY 2003 (ETB mn) ..................................................... 44
Table 5.3: Regional revenue composition, EFY 2003 (ETB per capita) .......................................... 45
Table 5.4: Regional revenue composition, EFY 2003 (as share of total) ......................................... 46
Table 5.5: ULGDP Allocations and transfers by year, EFY 2001-2005 ........................................... 49
Table 6.1: Municipal Revenue Collections for Selected Urban Local Governments (EFY 2004) .... 57
Table 6.2: Municipal revenue by major category (as pct. of total municipal revenue), EFY 2004 ... 60
Table 6.3 Revenue comparison 2004-2005, selected ULGs (ETB mn)............................................. 65
Table 6.4: Municipal Revenue Performance in Selected ULGs, EFY 2004, 2005 (ETB mn) .......... 67
Table 9.1: Summary assessment of Audit reports for GIZ Supported new ULGDP Cities .............. 88
Table 12.1: Examples of range of municipal tax and tariff rates ....................................................... 99
Figures
Figure 1.1: Regional States of Ethiopia ............................................................................................... 6
Figure 1.2: Institutional structure of the public sector in Ethiopia ...................................................... 7
Figure 4.1 The size and composition of local public sector expenditures: selected countries .......... 25
Figure 4.2: Municipal revenue collections (US$ per capita): Comparative analysis ......................... 27
Figure 4.3. Municipal Revenue Collections by Region (excluding Addis), EFY 2004 (ETB mn) ... 29
Figure 4.4. Municipal revenue collections by region (ETB per urban resident, EFY 2004) ............. 30
Figure 6.1: Municipal revenues per capita and population size (EFY 2004) .................................... 59
Figure 6.2: Municipal revenue collections and other economic and fiscal indicators, EFY 1999-
2004 .................................................................................................................................................... 66
Figure 8.1: Summary score APA 2004: City Planning and Citizen Involvement ............................. 77
Figure 8.2: Summary Score APA 2004: O&M Specified in AMP .................................................... 78
Figure 8:3 Summary Score APA 2004: Linkages between CIP, AAP and PP .................................. 79
Fig 10.1 Score of the Cities in for Asset Management Plan on 0-5 scale (EFY 2004) ..................... 93
vi
Government Fiscal Year

July 8 – July 7
The Gregorian and Ethiopian calendars are used throughout this document. The preceding abbreviations used are: GC
= Gregorian Calendar, EC = Ethiopian Calendar. The current year according to the Ethiopian calendar is 2006, which
began on September 11, 2013 AD of the Gregorian calendar.
The Ethiopian Financial Year commences on 1st Hamle (8th July) and ends on 30th Sene (7th July) each year. In
government documents the financial year 1st Hamle EC 2005 (GC 2012) to 30th Sene EC 2005 (GC 2013) is referred to
as the Ethiopian Fiscal Year (EFY) 2005.
Unless otherwise stated the years referred to are the Gregorian year.
USD – ETB Exchange Rate
The exchange rate as per August 2013 $1= Birr 18.6 is used throughout this document unless otherwise stated.
vii
Acknowledgements
The preparation of the situational analysis of urban local governments in Ethiopia was led by Onur
Ozlu (Senior Economist and Task Team Leader) and benefitted from the written contributions,
technical analysis and other inputs from a team comprised of Abebaw Alemayehu (Senior Urban
Specialist), Dinkneh Tefera (Urban Specialist), and technical specialists from Dege Consult,
Denmark, Urban Institute, US and SuDCA Development Consultants, Ethiopia. The analysis
benefitted from peer review comments at various stages of preparation from Serdar Yilmaz (Senior
Economist) and Berhanu Legesse Ayane (Senior Public Sector Management Specialist). The report
was prepared under the overall guidance of Mukami Kariuki (Sector Manager, AFTU1) and
benefitted from the advice of Guang Zhe Chen (Country Director for Ethiopia).
The authors would like to express their gratitude for the valuable assistance provided by a number
of core ministries, including especially, MUDHCo and MOFED, and representatives from the urban
local governments visited during the field mission in August 2013. The authors would especially
like to thank Tmuzghy Berhe Fikadu, and Mr. Fekadu Abera from the Project Management Unit in
MUDC for invaluable input, support to organization of the work, feedback, and guidance during the
start-up of the study for guidance and support.
viii
Executive Summary
This study on the institutional, governance and fiscal context of urban local governments in
Ethiopia is undertaken in order to inform the design of the future ULGDP and to enhance a detailed
understanding of the Ethiopian inter-governmental fiscal architecture. The report is divided into
four main parts.
Part I: Institutional Background and Governance Context
Chapters 2 and 3 provide a broad overview of the institutional and governance features of
Ethiopia’s public sector with an emphasis on the fiscal aspects of the intergovernmental
relationships. This overview includes a discussion of the role played by different government levels
in Ethiopia’s system of (fiscal) federalism, including the regional role in legislating and regulating
municipal revenues.
A key conclusion from this analysis is that local governments below the regional states only have a
limited degree of autonomy. Cities manage “state functions” such as education and health services
largely as deconcentrated units of the regional states, but have significantly more autonomy in the
delivery of “municipal services” (urban roads, sanitation, markets, etc.) Basic regional legislation
(such as the “City Proclamations” in each regional state) generally provides some measure of
autonomy to cities, for instance, with regards to their powers to (formally) determine local taxes and
rates. In reality, however, the regions issue various instructions and guidelines (in particular the
“City Tariff Regulations”) that in most cases nullify or significantly reduce cities’ ability to adopt or
update local tax rates. The main recommendation arising from this analysis is to support the
ongoing implementation of the Government’s municipal revenue strategy to introduce a prototype
“City Tariff Regulation” that each state can adopt and which will allow each city some autonomy in
setting local taxes and rates within a band of options.
Part II: The Fiscal Position of Local Governments
Chapters 4 through 7 provide an overview of the intergovernmental fiscal system and sub-national
finances in Ethiopia. These chapters consider the distribution of expenditures across different
government levels; trends in the collection of “state revenues” and “municipal” (own source)
revenues; the allocation of intergovernmental fiscal transfers as well as a brief discussion of
borrowing. Special emphasis is placed on urban development spending and on the financing of
urban local governments.
Our analysis of intergovernmental finances in Ethiopia revealed that the country’s current
intergovernmental fiscal system continued to be informed by the formerly deconcentrated state
structure under previous regimes. As such, the only grants that are consistently accounted for (and
reported on) as true intergovernmental fiscal transfers in Ethiopia are federal-regional general-
purpose grants. In fact, there is limited space in the Chart of Account for earmarked (special
purpose) grants. In practice, most “earmarked grants” are actually sub-national development
projects funded by international (external) grants or loans.
Key findings and conclusions with respect to municipal revenue collections of direct relevance to
the ULGDP design include:
ix
• Municipalities on average collect around ETB 200 per urban resident in municipal revenues
(see Figure E-1)– this is a figure that is roughly comparable to other countries in the region.
• There is considerably variability in the municipal revenue patterns, with great variation
when it comes to the major revenue sources. The three most important revenue sources for
the sample ULGs were: 1) licenses, fees and charges, 2) municipal rent and 3) land lease.
• Although municipalities may choose to spend municipal revenues on “state functions”,
municipal revenue is almost always spent on municipal development activities; and it is
largely spent on municipal capital expenditures.
• In general, larger (more populous) cities collect more municipal revenue than smaller cities
(when revenue collection per capita is analyzed), but the larger cities also tend to rely more
on land lease revenue, which is not a sustainable source of income for smaller cities.
Figure E-1: Municipal revenues per capita and population size (EFY 2004)
Note: Data from fieldwork in municipalities, supplemented with fieldwork data from GIZ/GOPA. Existing
ULGDP cities are identified with a blue diamond; non-ULGDP cities are indicated with red triangles.
While municipal revenue growth has been robust in nominal terms, municipal revenue growth has
been less impressive in real terms. Indeed, annual growth of municipal revenue has been slow
compared to overall public revenue growth, inflation and GDP growth (Figure E-2). Even the
ULGDP supported cities have not been able to demonstrate significant growth in municipal
revenues growth over the past few years. This is in large part due to the structure of municipal
revenues: municipal revenues (especially fees and charges) tend to be eroded by inflation as they
are typically specified in nominal terms, whereas federal and regional revenue sources (e.g. the
VAT or taxes on personal and business income) are generally specified in percentage terms.
x
Figure E.2: Municipal revenue collections and other economic and fiscal indicators, EFY 1999-
2004
Source: MOFED (IBEX) and IMF.
Part III: City Management Practices
Chapters 8 through 12 provide an assessment of the city management practices with regards to local
participatory planning and the formulation of CIPs, local government financial management, asset
management and O&M planning, procurement systems – and importantly: local revenue
management. The chapters discuss fieldwork and other observations in relations to results from the
Annual Performance Assessments (APAs) with a view of refining these for the future ULGDP. The
main recommendations that arise from these five chapters are summarized below according to
functional areas of city management.
Planning: Two areas could be improved upon:
• “Participation” is currently measured rather crudely – foremost with figures on “total number of
citizen meetings” and “number of groups consulted with”. It would be unreasonable to expect
the cities to continuously report on growing numbers of meetings and groups. Instead it could
be more useful to establish a general standard for type (and number) of meetings and other
opportunities for citizens to air their views. This could e.g. include standards for public hearings
with standards for how agenda and other information should be shared in advance, etc.
• Transparency is currently measured with regards to publication of accounts, budgets, etc.

However, at present there is no clear format for how such information can be summarized in a
meaningful way that effectively will empower citizens. It is recommended that clear standards
be developed for such information. In addition it is recommended that the information is also
made available on the Internet. All municipalities could upload all documents required for the
APA process to an online system managed by MUDC and/or regional urban bureaus.
Documents would be time-stamped when uploaded by municipalities. This would ease the
assessment process and also put relevant documentation into the public domain for greater
transparency.
xi
Financial management: Currently PFM issues are almost entirely assessed by use of the Audit
General report – however, the audit opinion is a very crude measurement of performance. If the
APA should continuously spur cities to improve their performance in areas related to financial
management, it is necessary to include other/additional indicators than just the audit report. This
could for instance include:
• On site analyses of bank reconciliation,

• On site analyses of appropriateness of coding – in particular of municipal revenues,
• Other on-site analyses of selected financial management practices deemed critical (exclusive
use of numbered vouchers etc.),
• Analyses of adherence to good practices for internal audits (clear standards/good practice
benchmarks need to be developed),
• Submission of final accounts on time (perhaps even as a minimum access condition),
• Indicators on follow up on audit queries,
• Independent Internal Audit functional (reporting directly to Mayor) with significant audit of
capital works (e.g. 25% of all contracts annually or other measurement of the scope of internal
audits).
Asset Management and O&M: It is recommended to sharpen the focus on actual implementation of
O&M activities rather than just mapping of infrastructure and assets. Putting more emphasis on
comprehensiveness of the actual O&M plan and budget, and subsequent monitoring of actual
budget utilizations will sharpen the focus of assessment on actual O&M performance.
Procurement: The only problem with the current assessment system is with the indictor on
“procurement audit”. This is only undertaken as an integral element of the general audits with
various levels of emphasis - however most often without separate chapters, specifically thus it has
been difficult for the assessment team to verify. This indicator is not recommended to be included
in a scoring system until there is a common agreement with Government on how (and if) to
undertake separate procurement audits at city level.
Revenue enhancement: In order to improve significantly on cities municipal revenue management it

is required that the federal government elaborates on the guidance and standards for revenue
management at both regional and city level. However, some adjustment can also immediately be
made of the APAs with the above weaknesses in mind in order to feed into the soon forthcoming
APA (tentatively to be undertaken from January 2014). The recommended indicators for city level
are to focus on the two basic issues as today, but with some modification:
1) Improved Revenue Enhancement Plan: This could follow the existing way of measuring its
quality but inclusion of sub-indicators that are given more significant weight to measure:
• City analysis of previous years’ revenue performance with detailed analyses of each main
source of revenue including discussion of its revenue potential (revision of current sub-indicator
No. 2);
• City strategies for revenue enhancement (current sub-indicator 6).
2) Improved revenue collections and management:

• Proper coding of municipal revenue as per chart of accounts (some additional guidance on
coding should be issued),
xii
• Percentage increase of each main category of municipal revenue (less land lease): Business
taxes, municipal rent, charges and fees,
• City publication of how municipal revenues have been spent in format suitable for citizen
budgets/accounts,
• Registers on revenue collection are up-to-date,
• Clear procedures and evidence on follow up on tax defaulters,
• Public consultations on tax issues especially prior to issuing of new rates and bases,
• Inclusion of counterpart funding on the ULGDP/capital projects as a performance measure to
promote sustainable spending of OSR on development.
Recommended medium term revisions of the assessment: In order to incentivize more structural
reforms of municipal taxes it is necessary that the Federal Government/MUDC provide clearer
guidelines and specific standards for the Revenue Enhancement Plan than currently is the case. The
type of guidance should include:
• Guidance to regions in the form of “Model Tariff Regulations” and guidance on how regions
can assist cities to undertake relevant studies for tax reforms. It should be noted that the MUDC
led municipal tax task force has already initialed work that suggest recommendations in this
direction,
• More specific guidance and standards to cities in the REP manuals regarding how cities can
estimate coverage ratio, assessment ratio and collection ration for main revenue sources (as
discussed in Chapter 6),
• Guidance to cities on coding of revenues sources in the chart of accounts and guidance to
regions for subsequent annual analysis of municipal revenues,
• Working manuals on municipal taxes including simple standard formats for computerization of
tax registers, etc. (at present this is in part included in TOR for the regional TA however it is
more appropriate to establish some national basic standards),
• Compilation and sharing of good practices in own source revenue mobilization.
Part IV: Initial Implications for the design of a Performance-Based Urban Grant System
The final chapter of the report (Chapter 13) summarizes key findings of direct relevance to the
design of ULGDP II (and beyond). It should be noted that this last chapter only provides a
preliminary discussion of design options, whereas Component 2B reports will provide the more
detailed and specific recommendations.
Sustainability of the performance based urban grant. Based on an analysis of intergovernmental

finances in Ethiopia and international comparisons, it is argued that it is unrealistic to expect that
future municipal development can be fully funded from cities’ own source revenues alone. The
Government of Ethiopia, World Bank and other possible development partners and stakeholders
should therefore think of urban development grants as a (relative) permanent feature of Ethiopia’s
intergovernmental fiscal transfer system. The current approach of accounting for ULGDP in a
projectized manner (following Channel 1 funding as a non-subsidy cash transfer) is not an
appropriate solution if the long-term ambition is to achieve an urban development grant— with
regional/city contributions—that is fully integrated in government processes at all levels of
government as a real intergovernmental grant. Instead, whenever possible, the grant should be
xiii
accounted into the regular budget and PFM processes as a proper earmarked grant at the federal,
regional and local level.
Counterpart funding from cities and regions have to date not been systematically monitored, and it
is recommended to strengthen such monitoring in future phases of support. The level of co-
financing has appeared for the majority of cities—in particular the larger cities—to be achievable.
However, for some of the smaller and newly proposed cities it may be difficult to reach (or go
beyond) the current level of 20% co-financing. Cities in general (with few exceptions) already use
most of their revenues for municipal capital expenditures. With future increased integration of the
ULGDP into city financial management systems it may be more relevant to monitor cities total
spending on municipal infrastructures rather than narrowly focus on specific “ULGDP co–
financing”. Furthermore, several regions have developed their own schemes for co-financing city
level infrastructure development; some of this has taken the form of “home grown” incentive-based
grant systems. As the ULGDP is increasingly (also beyond ULGDP II) becoming integrated into
regional budgets it may (over time) be appropriate to take a more holistic view of regional
contributions to urban infrastructure development and also consider such schemes and contributions
as part of regional co-financing.
Managing funding flows across all levels of government – The future ULGDP is recommended to
be fully integrated (“proclaimed”) as an on-budget earmarked grant at all three levels of
government: federal, state and city. The current ULGDP is replenished during the budget year as
cities provide documentation of their expenditures (use of funds) to the higher level. This is an
administratively cumbersome approach and will only become more problematic as the universe of
ULDGP municipalities expands. Treating ULGDP as a “real” grant (which is bestowed upon the
urban local government jurisdiction, similar to the general block grant) would allow for a more
streamlined disbursement process. As a real grant, each municipality’s grant allocation could
simply be disbursed in two equal, semi-annual tranches at the beginning of Quarter I and Quarter
III. Compliance with all grant conditions would be verified as part of the annual performance
assessment and audit processes. It should be noted that these issues are further discussed in the
Component 2b Report.
Capacity Building of cities has hitherto benefitted from PSCAP that will no longer continue at a
similar level. Other capacity building under ULGDP has primarily targeted new cities for inclusion
in ULGDP through centrally managed TA arrangements. The future ULGDP II aims to include
more significant ongoing support to cities included for capital investments in ULGDP II. This is
recommended to be supported through more demand-led capacity delivery arrangements, financed
by a capacity building grant.
Improvement of performance indicators under the APA. Tentative suggestions for improving the
performance indicators under the APA for ULGDP cities was discussed above in Part III, where the
report provides suggestions for refinement of existing indicators as well as for a few new targeted
indicators.
As the future program will address targeted activities at the regional level as well, a few targeted
interventions and performance areas are discussed in the final parts of the report.
Implications for the role of regional level institutions. The analysis has indicated that regional level
institutions play very important roles for the development of the urban local governments. This
xiv
report has indicated weaknesses in several aspects of regional involvement and support to date.
These will require strengthening and also improved monitored as part of a future DLI verification
framework for the regional level.
The relevant performance measures at regional level could include indicators of oversight and
capacity building support in various areas. The suggestions below are very tentative, and should be
further defined (and prioritized) during the coming design work:
• Municipal revenue:
o Regional governments pass legislative framework for municipal finance that
facilities cities an appropriate level of autonomy in setting local tax, tariff and fee
rates suitable for each city,
o Regional oversight ensures that municipal revenues are properly coded in all revenue
reports,
o Regional government provides an online platform for sharing municipal Revenue
Enhancement Plans,
o Regional officials provide guidance, backstopping and capacity building support for
revenue planning and enhancement,
o Regional oversight ensures local adherence to fairness in revenue collections and
other good revenue practices.
• Planning:
o Regional officials provide guidance and ensure adherence to standards for physical
planning, master plans, cadasters, etc.,
o Regional officials provide assistance and quality assurance to cities for publication
of relevant plans (CIPs; O&M plans; etc.) on a regional web-site.
• Financial management:
o BOFED disburses grants to ULGs in a timely manner,
o Regional oversight ensures that municipal accounts are managed properly in
accordance with laws/regulations,
o Regional officials ensure that adequate internal audits are performed,
o BOFED ensures that municipal and regional budget accounts accurately reflect
municipal revenues and expenditures, including ULGDP transfers and expenditures.
• Audits:
o Adequate and timely external audits are performed.
• Transparency and service delivery:
o Regional officials provide guidance to cities with regard to inclusive service delivery
planning, transparency of citizens’ rights to municipal services, and support to
citizens’ efforts to monitor local services
o Establishment of basic service delivery monitoring (and publication of municipal
performance on the regional website).
All indicators for performance will have to be guided by clear federal instructions, guidelines and
standards, which provide guidance across all government levels about the respective roles and
responsibilities of each level in promoting effective, equitable and sustainable urban infrastructure
and services.
xv
1. Introduction
1.1 Background
Ethiopia is rapidly urbanizing. With an estimated population growth rate of about 4.1 percent per
year1, Ethiopia’s urban areas will expand, in nominal and relative figures, from an estimated 16.5
million people and an urban share of population of 17.5 % in 2013 to 21.9 million people and
19.0 % of the total population in 2020.
Government of Ethiopia (GoE) acknowledges the important role of the urban sector in overall
economic growth and has established policies, strategies and programs that support investment in
urban development. Municipal authorities were established in Ethiopia in the 1940s, but fully
democratic urban local governments were established after 2000, when proclamations to
establish urban local governments were first issued by Regional States, and Federal government
in the cases of Addis Ababa and Dire Dawa. Combined with a commitment to fiscal
decentralization, the proclamations were intended to give local governments more direct and
transparent control over public spending. The objective was to create and strengthen urban local
governments so as to ensure public participation in making choices and enhanced service
delivery in cities and towns. The challenge has been to find ways to support urban local
governments in developing the capacities, incentives, and financial resources needed to deliver
infrastructure and services to residents effectively and efficiently.
Urban local governments in Ethiopia have dual responsibilities. These are the provision of “state
services,” such as education, health, justice, and security, and the provision of “municipal
services,” such as urban roads, drainage, solid waste collection and disposal, and sanitation. State
functions are financed through regional block grants, which are often barely enough to cover the
recurrent needs in these sectors. All municipal functions are expected to be funded from own
local revenues for both recurrent and capital expenditures.
In this context, the government introduced the Urban Local Government Development Project
(ULGDP) in 2008 which had two components: (1) performance grants for 19 cities totaling
approximately USD 287 million; and (2) implementation support and unallocated funds of
approximately USD 13 million. 2 The ULGDP was preceded by a series of Bank supported
interventions, which aimed to build capacity at urban local government. Based on local
government capacity enhanced through these projects, the government introduced the ULGDP as
a performance-based programmatic fiscal transfer to urban local governments. The overall goal
of the government program is to support improved performance in the planning, delivery, and
sustained provision of urban services and infrastructure by local governments. It aims to fulfill
this goal by providing grants to urban local governments based on their performance across eight
performance areas. The program funds are allocated for institutional performance and are
1
World Bank Development Indicators (2012 figures) based on official estimates.
2
Effective date: November 10, 2008; closing date: December 31, 2014.
1
earmarked for expenditure on local infrastructure (including the cost of associated design and
supervision). The IDA funding for ULGDP performance grants is supported by Regional States
and the participating cities who each contribute an additional 20% of the IDA performance grant
amounts. Therefore, sum investment funds applied by the ULGDP total approximately USD 400
million.
The Government of Ethiopia will now extend its performance based fiscal transfer program to 26
additional urban local governments, with the support of the World Bank. The government’s
rationale in expanding the urban performance grant system to new cities stems from its
acknowledgement of the key role that cities play in economic growth and development and from
its desire to expand ULGDP’s successful approach to additional fast-urbanizing cities. With the
addition of the new 26 local governments, the program will comprise a total of 44 urban local
governments – which together comprise 51% of all urban centers in Ethiopia, with a population
of 20,000 or more (according to the 2007 census). It is believed that these 26 new cities will be
ready to participate in the performance based fiscal grant system because 8 out of 26 new cities
have been receiving technical assistance provided by the German Society for International
Cooperation (Deutsche Gesellschaft für Internationale Zusammenarbeit, GIZ) and the remaining
18 will shortly receive capacity building support from ULGDP.
The planned expansion of ULGDP is part of the overall fiscal decentralization reform of the
government supported by the World Bank. Under this reform, urban local governments’ own
revenue sources have been consistently increasing. It is important to understand this rise in order
to determine the long-term sustainability of own financial resources. Analysis will serve two
distinct purposes; firstly to help local government decision makers to develop policies and
practices (as necessary) to ensure sustained enhancement of revenues, and secondly to inform the
planned expansion of the performance based fiscal grant system.
Own source revenues are key to funding urban public services for a number of reasons,
including: to provide funding for urban investments; to fund the effective operation and
maintenance of investments in infrastructure and service; to support access to capital markets or
infrastructure funds for capital investment in city-wide infrastructure and services; to provide a
local financial autonomy; and to serve as an accountability mechanism between local authorities
and local citizens. For these reasons, ULGDP provides a strong financial incentive for local
governments to enhance local own source revenues. It is believed that as local governments in
the country enhance their revenue base and fiduciary systems, they will in principle be able to be
fiscally self-sufficient and will be able to finance their infrastructure needs by borrowing from
the national capital market. At this point, a clear understanding of the overall fiscal picture, as
well as the financial, institutional and organizational capacity at the local government level is
needed.
1.2 Objectives
The objective is to study the institutional and fiduciary governance context and fiscal position of
local governments, with a view to provide recommendations for the government’s performance
2
based fiscal grant system and enhance the detailed understanding of the Ethiopian inter-
government fiscal architecture.
1.3 Methodology
The study team has applied a mix of methodologies for data collection:
• Review of secondary data such as previous analyses of the Ethiopian inter-governmental
fiscal architecture. However, most of these studies are very old and tend to focus on the
broader federal-Regional issues or the rural local governments (woredas). See Annex 1.
• Review of various primary data, collected in relation to the management of the ULGDP.
This includes data from the Annual Performance Assessments (APA) as well as various
benchmarking studies – these all focus on the existing ULGDP supported cities.
• GIZ supported analyses of the new ULGDP cities.
• Review and analysis of existing primary data on consolidated government accounts, as
well as regional data on regional and local (woreda) revenues and expenditures from
MOFED/IBEX. The limitations of consolidated government accounts and other IBEX
reports in analyzing intergovernmental finances in Ethiopia are discussed in Chapter 2
(Box 2.2).
• Interviews with key stakeholders involved in implementation of the existing system for
delivery of investment to, and capacity building of cities, at federal level (MUDC,
MOFED, World Bank office, etc.). See Annex 2.
• Fieldwork in 12 selected urban local governments as specified in the TOR. These
constitute a mix of existing and planned ULGDP supported cities as well as a regional
mix that also include very significant variations in population size (see table below).
Table 1 Sample Local Governments visited
Population 2007
City Name Region Ongoing program (as indicated in TOR)
Adama Oromia ULGDP – Grant 220,212
Jijiga Somali GIZ – TA 125,876
Mekele Tigray ULGDP – Grant 215,914
Bahir Dar Amhara ULGDP – Grant 155,428
Hawassa SNNP ULGDP – Grant 157,139
Harar Harar ULGDP – Grant 99,368
Hosaena SNNP ULGDP – TA 69,995
Debre Brehan Amhara ULGDP – TA 65,231
Zeway / Batu Oromia ULGDP – TA 43,660
Adwa Tigray GIZ – TA 40,500
Gambella Gambella GIZ – TA 39,022
Benishangul
Asosa Gumuz GIZ – TA 24,214
1,256,559
3
The availability and reliability of data is always a challenge in this type of study. The main text
comments on data issues where relevant, however some general commentary needs to be made
up-front. In general it can be observed that cities have made very significant improvements over
a short time with regards to establishment of reasonable well-organized accounts through IBEX.
However, there has to date been limited or no interest (beyond ULGDP specific monitoring of
the ULGDP supported cities) from the regional states and federal government in compilation,
monitoring, and analyzing municipal revenue and expenditure patterns. Most municipal finance
data therefore has to be extracted from the individual cities. These cities have received little
guidance or oversight on data recording and the teams found variations in the degree to which
local officials adhere to the budget classification codes across budget documents. It is also not
uncommon for different data sources to indicate different figures. The team has nevertheless
tried to triangulate various data sources and has where relevant indicated the specific data
challenges in order to represent a reliable and hopefully informative picture of “the institutional,
governance and fiscal context of urban local governments in Ethiopia”.
1.4 Organization of this Report
The report is divided into four main parts:
Part I: Institutional Background and Governance Context

Provides a broad institutional and governance overview with emphasis on the fiscal aspects of
the interrelationships. This includes a discussion of the regional regulation of municipal
revenues.
Part II: The Fiscal Position of Local Governments

Provides an overview of the intergovernmental fiscal system and sub-national finances in
Ethiopia including trends of own source revenues and fiscal transfers as well as a brief
discussion of borrowing.
Part III: City Management Practices

Provides an assessment of the city management practices with regards to local participatory
planning and the formulation of CIPs, local government financial management, asset
management and O&M planning, procurement systems – and importantly: local revenue
management.
Part IV: Initial Implications for the design of a Performance-Based Urban Grant System
Summarizes key findings of direct relevance to the design of ULGDP II (and beyond). It should
be noted that this last chapter only provides a preliminary discussion of design options, whereas
Component 2B reports will provide the more detailed and specific recommendations.
4
PART I: INSTITUTIONAL AND GOVERNANCE CONTEXT
2. Overview of the institutional relationships between federal,

regional and local governments
The context within which urban local governments operate in Ethiopia is to a large extent
defined by the country’s sub-national institutional and governance architecture. As such, this
chapter provides an overview of the institutional, fiscal and administrative relationships between
federal, regional and local governments in Ethiopia. The sub-national organizational-governance
structure has been shaped to a large extent by domestic political forces, although support from
the international community in building the capacity of the public sector (and urban local
governments in particular) has contributed significantly to shaping the functioning of sub-
national governments.3
2.1 Overview of the intergovernmental institutional structure of the public sector
The Constitution defines the federal-

Regional relationship. The current structure Figure 1.1: Regional States of Ethiopia
of Ethiopia’s public sector was formulated
during the transitional period from 1991-
1994 following the defeat of the Mengistu
regime in 1991. Ethiopia’s Constitution—
adopted in 1994 and formally promulgated
in 1995—establishes a federal, democratic
government structure taking into
consideration the ethno-linguistic and multi-
cultural diversity of Ethiopian nations,
nationalities and peoples. As such, the
constitution represents a fundamental shift
in the organization of the public sector and
in the political system, which was organized
in a highly centralized manner under the previous regime.
The Constitution recognizes and assigns powers, functions and revenues between the federal
government and the nine member states of the Federation (Tigray; Afar; Amhara; Oromia;
Somalia; Benishangul Gumuz; the State of the Southern Nations, Nationalities and Peoples
3
Details about Ethiopia’s public sector reforms are provided by World Bank. April 2013. Building the
Developmental State – A Review and Assessment of the Ethiopian Approach to Public Sector Reform, Report No.
ACS3695. A synopsis of support to urban local governments is provided by GOPA Consultants, July 2013. Ethiopia
- Urban Governance and Decentralisation Programme (UGDP), Municipal Finance Reform (Annual Report
2012/13).
5
(SNNP); Gambela and Harar), which were delimited on the basis of the settlement patterns,
language, and ethnic identity. These entities are generally referred to as ‘regional states’. In
addition, two cities—Addis Ababa and Dire Dawa—are each chartered by federal proclamations
(laws) and are treated in some respects as state-level City Governments.
Article 52 of the Constitution assigns states the power to levy and collect taxes and duties on
revenue sources reserved to the States. Articles 95 through 100 define the specific assignment of
revenues between the federal level and the regional level. Although important specific revenue
sources are assigned either to the federal and regional levels, 4 respectively, many important
revenue sources are assigned concurrently to the federal and regional, with the Constitution
directing the federal government and the states to jointly levy and collect profit, sales, excise and
personal income taxes on enterprises they jointly establish.
State constitutions and proclamations define regional-local relationships. Although the federal
constitution only formally establishes two government levels, in practice, Ethiopia is a federal
state with three main government levels: the federal level, the regional state level, and the local
government level.5 Article 50 of the federal Constitution states that “[t]he Federal Democratic
Republic of Ethiopia comprises the Federal Government and the State members” and “State
government shall be established at State and other administrative levels that they find necessary.
Adequate power shall be granted to the lowest units of government to enable the People to
participate directly in the administration of such units”. In addition, the States “shall have the
following powers and functions (a) To establish a State administration that best advances self-
government, a democratic order based on the rule of law, to protect and defend the Federal
Constitution…”. While this language is supportive of state-to-local devolution, the Constitution
falls short of specifically recognizing a local government level, or providing constitutional rights
or protections to (urban) local governments.
In accordance with the constitutional

Figure 1.2: Institutional structure of the public
guidance, each regional state has
sector in Ethiopia
adopted a proclamation establishing its
local government structure. Although
there are variations in the sub-regional
structures among the regional states,
the most prevalent local government
level is formed by elected woredas or
district-level governments. Regional
states are typically subdivided into
administrative zones, which is a
deconcentrated territorial level (see
Figure 1.2). 6 The international Note: Dark (blue) indicates government jurisdictions with their
own budget sources; light (grey) indicates deconcentrated entities.
4
For instance, the Federal Government shall levy and collect custom duties, taxes and other charges on imports and
exports (Article 96.1).
5
The Constitution refers to “member states”. In this document we will use the terminology “regional states”,
“states” and “regions” interchangeably. As further discussed in the Component 1 report, we note that Addis Ababa
(and Dire Dawa) are not regional states. However, they are often treated as such in the PFM reporting process.
6
In some regions (such as SNNP Regional State), the zonal level is more relevant than in other regional states.
6
development community has played an important role in advocating for a strong role of the
woreda level in delivering public services, through programs such as Protection of Basic
Services (PBS) Program and a series of public sector capacity building projects.
Woredas are generally semi-autonomous local government entities, which have a separate legal
status as a corporate body with their own political leadership (council) and their own budget
accounts. The woreda council’s members are directly elected to represent each kebele (ward) in
the district. In practice, the terminology used to designate local governments varies: some use the
term woreda strictly to refer to rural districts, whereas others consider district-level urban units
also to be woredas.
Regional legislation typically allows for urban governments to be established at the woreda-level
as well as at the sub-woreda level. Generally, the main urban areas are covered by woreda-level
City Administrations, which are responsible for both woreda-level responsibilities (i.e.,
concurrent state functions) as well as city affairs. Depending on the regional state and on the size
of the city, woreda-level City Administrations may be further sub-divided into sub-cities/city
districts, and/or kebeles. Smaller municipalities—which exist in rural areas below the woreda
level—may be governed by Town Administrations or Municipalities, which are exclusively
responsible for municipal functions and revenues.
Given the focus of this report, all local entities below the regional level will generally be referred
to as local governments. The term “urban local governments (ULGs)” will be used to generally
refer to the main urban jurisdictions or City Administrations, which are typically recognized as
district-level local governments. These larger urban local governments form centers of economic
and social activity in the country, and play a special role in the development and transformation
of the country. This point notwithstanding, it should be acknowledged that there are many
smaller urban areas below the woreda level that are recognized as municipalities.
“Interdependent (or collaborative) federalism”. Although Ethiopia’s intergovernmental system

has features that resemble many other federal countries, there are some unique features in
Ethiopia’s public sector structure that suggest that the different government levels in Ethiopia are
less autonomous than federal systems in more industrialized countries (e.g, see Yilmaz and
Venugopal, 2008; Gebre-Egziabher and Dickovick, 2011). In fact, Ethiopian political leaders and
policy makers readily point to capacity limits at lower levels to justify the more limited
autonomy at lower government levels. Keeping this in mind, one might term Ethiopia’s approach
to federalism as “collaborative federalism” or “interdependent federalism”. The institutional
interdependence—rather than autonomy—of different government levels is an important feature
of Ethiopia’s federalist system that should be taken into account when discussing the political
economic space for municipal local governments.
Currently, this dual subordination is facilitated by the fact that the same political party is in
charge of all government levels and jurisdictions. When Ethiopia moves towards a more
competitive political environment, this may pose a challenge to the current approach to
collaborative and interdependent federalism.
7
Population and urbanization pattern. While Ethiopia is composed of nine regional states and two
regional-level City Administrations (Addis Ababa and Dire Dawa), these regional(-level)
jurisdictions range considerably in terms of population size, geographical size and urbanization.
Almost 90 percent of the country’s population is resident in five regional states: Tigray, Amhara,
Oromia, SNNPR and Addis Ababa.
Some regions are considerably more urbanized than other regions. For instance, Tigray has an
urbanization rate nearly twice higher than SNNPR. By their very nature, Dire Dawa and Addis
Ababa are highly urbanized (68-100%).
Table 2.1. Regional population and urban population (July 2013)

Urbanization
Population Urban Population rate
Tigray 5,061,991 1,120,456 22.1
Afar 1,649,999 280,101 17.0
Amhara 19,211,994 2,703,588 14.1
Oromia 32,220,001 4,262,970 13.2
Somali 5,318,000 755,639 14.2
Benishangul 1,027,994 177,989 17.3
SNNPR 17,887,005 2,147,944 12.0
Gambela 406,004 128,800 31.7
Harari 215,000 112,781 52.5
Dire Dawa 395,000 269,134 68.1
Addis Ababa 3,103,999 3,103,673 100.0
TOTAL 86,613,986 15,100,075 17.4
Source: Central Statistical Agency of Ethiopia, 2013 (Population Tables: Table B2; Table B.4).
Note: Urban population is equal to the population of “towns” in each region, as reported by the CSA. Special
Enumeration areas are not reported in the table (but included in total).
2.2. An overview of functional assignments and local regulatory powers
The division of functions between the federal and regional (state) level. Article 51 of the
Constitution defines the powers and functions of the federal government. These federal functions
include functions commonly assigned to the national level in a federal system, such as national
defense, foreign affairs, macro-economic policy, and the regulation of inter-state trade. In
addition, the federal government is assigned the power to “formulate and implement the
country’s policies, strategies and plans in respect of overall economic, social and development
matters”, as well as the power to “establish and implement national standards and basic policy
criteria for public health, education, science and technology as well as for the protection and
preservation of cultural and historical legacies.”
The powers and functions assigned to the regional states (Article 52) are considerably more
vague: for instance, no responsibility is assigned to the state level for education, health services,
agriculture, or any other specific function. Instead, the powers and functions of the regional
8
states include the broad responsibility “[t]o enact and execute the state constitution and other
laws” and to “formulate and execute economic, social and development policies, strategies and
plans of the State”. 7 However, “[a]ll powers not given expressly to the Federal Government
alone, or concurrently to the Federal Government and the States are reserved to the States”.
In practice, the main regional government functions and expenditure responsibilities include
most important social and economic functions of the public sector, including the provision of
primary and secondary school education; the provision of basic health services; intra-state and
rural road construction; rural water supply; agricultural extension services; regional
administration, general public administration service, and justice.
However, the degree of autonomy for the regional level to implement key social services (such
as education, health, and so on) is limited by the constitutionally mandated role for the federal
government to set policy objectives in these areas.
The federal role in urban development. There is no explicit constitutional or legal basis for a
federal role in urban development in Ethiopia. However, as urban development is obviously
important to the nation as a whole, a potential constitutional justification for a federal role in
urban development could be based on the broad federal power and function to “formulate and
implement the country’s policies, strategies and plans in respect of overall economic, social and
development matters” (Article 51.2).
Federal Ministries are empowered by Proclamation 610/2010, which defines the powers and
duties of the executive organs of the Federal Democratic Republic of Ethiopia. All Federal
Ministries are granted common powers and duties; for example, in their area of jurisdiction, to:
initiate policies and laws, prepare budgets and plans, undertake study and research, enforce
federal laws, undertake capacity building activities, provide assistance and advice, enter into
contracts and international agreements and direct and coordinate the performance of executive
organs. The Ministry of Urban Development and Construction is given powers and duties, for
example, to: undertake studies and set criteria for grading urban centers; provide all round and
coordinated support to urban centers; provide capacity building support; cooperate with regional
states to integrate urban and rural development; follow up the activities of city administrations
accountable to the federal government; and promote expansion of micro and small enterprises.
MUDC is organized into a number of bureaus, include including the Urban Planning, Sanitation
and Beautification Bureau and the Land Development and Management Bureau. The Urban
Good Governance and Capacity Building Bureau (UGGCBB) is one of the main bureaus
organized under the ministry. Among others, this Bureau coordinates and administers the
7
Detailed functions and powers of the regional state are: a. To establish a State administration that best advances
self-government, a democratic order based on the rule of law; to protect and defend the Federal Constitution; b. To
enact and execute the state constitution and other laws; c. To formulate and execute economic, social and
development policies, strategies and plans of the State; d. To administer land and other natural resources in
accordance with Federal laws; e. To levy and collect taxes and duties on revenue sources reserved to the States and
to draw up and administer the State budget; f. To enact and enforce laws on the State civil service and their
condition of work; in the implementation of this responsibility it shall ensure that educational; training and
experience requirements for any job, title or position approximate national standards; g. To establish and administer
a state police force, and to maintain public order and peace within the State.
9
financial aid and loans obtained from Urban Local Government and Development Project
(ULGDP) and the Urban Governance Decentralization project (UGDP) to capacitate cities and
build standard infrastructures. MUDC is currently not providing earmarked urban development
grants to urban local governments other than grant schemes supported by externally funded
projects. 8
The federal government supports urban development in more direct ways as well. For instance,
the federal government actively supports the construction of a light rail system in Addis Ababa
through the federal Ethiopian Railways Corporation.
The division of functions between the regional and local level. Not only does the federal
constitution not explicitly define the exact responsibilities of regional states, an equally
significant challenge is that the federal constitution does not define the division of
responsibilities between the regional and local level.
Following Article 50 of the Constitution, municipalities or “city administrations” are recognized

in most regional states in Ethiopia, proclaimed either by the state constitutions or by subsequent
laws (proclamations). The division of functional responsibilities between the regional level and
sub-regional entities is equally guided by regional constitutions and subsequent regional-level
proclamations. (For further details see Chapter 3.)
State Functions versus Municipal Functions. A unique feature of the expenditure assignments in
Ethiopia is the distinction between so-called “state functions” and “municipal functions”. In
urban areas, both functions are administered by the ULGs, but the state functions are delegated
from the region to the local authorities, whereas municipal functions are considered to be their
exclusive functions. State functions include most social services (education, health, and so on),
which were provided in a deconcentrated manner under the previous regime, whereas basic
urban services (urban streets, solid waste management, abattoirs, and so on) are considered to
fall under “municipal functions”. The main definition of urban services is found in the Chart of
Accounts, where municipal services are defined under Head 500.
Consistent with the institutional evolution described in Box 2.1 (next page), the main functional
assignment in Ethiopia is not actually between functions assigned to different government levels,
as in reality most service delivery functions are the concurrent responsibility of the regional
states and the local level. In contrast, the division of functional responsibilities of local
governments is defined primarily in terms of the two main types of responsibilities that are
assigned to the local level: first, concurrent “state functions”, and second, exclusive “municipal
functions”.
The conceptual distinction between state services and municipal services has important funding
implications. Rural woredas that lack any urban centers only provide state functions, and receive
a block grant from the regional level to fully fund these functions, as they essentially lack any
8
The federal level is currently not providing direct counterpart funding for ULGDP. It should be noted that in due
time the federal level will be responsible to repay the IDA loan funds. The role of federal co-finding will be
discussed in greater detail in Chapter 13.
10
own source revenues. 9 Similarly, ULGs generally receive transfers (block grants) from the
regional level in order to fund their state functions. However, they are (in principle) expected to
fully fund their municipal functions from municipal revenues.
Box 2.1: State functions versus Municipal functions: A brief historical perspective
Under the previous regime (and under the imperial rule that preceded it), municipalities were semi-
autonomous local government entities, which were legally separate from the (previously) deconcentrated
local offices of the state at the local level. As part of the decentralization reform process, deconcentrated
offices at the woreda level (and in some cases, sub-woreda city administrations) were given autonomous
status. Under this previous system, deconcentrated structures provided “state services” (mainly social
services), whereas municipalities delivered only municipal-type services (such as solid waste, urban
roads, and so on). As part of the decentralization reforms that took place from 2000 to 2003 these two
separate organizations at the local level were integrated into a single organizational unit, providing both
types of local functions (state functions and municipal functions). As a subsequent step, five years ago,
municipal budgets and deconcentrated local “state” budgets were further integrated, by combining all
local finances at the local level under a single financial management system.
Local revenue autonomy and regulatory autonomy with respect to urban planning. To varying
degrees, local governments (as discussed further in Chapter 6) can set their own local tax rates
within the context of regional law. They are also generally empowered to adopt their own urban
land use plans and make local regulations pertaining to urban affairs.
The regional role in urban development. The extent and role of regional-level governments in
urban development is different in different regions. As further explored in section 4.3 (in Chapter
4), various regions directly fund urban development activities; provide co-funding for ULGDP
projects; and/or implement their own urban development grants (outside of ULGDP) to urban
local governments within their respective jurisdictions.
2.3 Overview of the intergovernmental administrative structure of the public sector
Although Ethiopia has adopted a federal structure, its federal system and administrative and
organizational structures are uniquely shaped by the country’s social, political and economic
transition path. Its system is federal in the sense that each of the main government levels is
formed by semi-autonomous corporate bodies with their own political leadership (council).
Funding flows between the different levels are characterized as intergovernmental fiscal transfers
9
This is not the only remnant of the previous deconcentrated system that remains part of Ethiopia’s current system.
Another unique feature of Ethiopia’s blend of deconcentration and devolution is the collection of “state revenues”
by local revenue offices, and the resulting offsetting of transfers. Since local revenue offices (i.e., revenue offices
that formally fall under the local governments) are assigned the responsibility of collecting “state revenues”
(revenues which in principle belong to the regional state) in addition to collecting “municipal revenues” (revenues
which in principle belong to the local government itself), regional states generally prevent offsetting funding flows
in opposing directions by allowing local governments to retain the “state revenues” collected within their
jurisdiction, up to the budgeted (estimated) amount of state revenue collections within that jurisdiction. The regional
level will pre-emptively reduce the local government’s block grant by the same amount. This issue is discussed
further in Section 5.2.
11
as opposed to sub-national budget allocations, and sub-national (especially regional) councils

have a meaningful degree of discretion over their spending priorities.
At the same time, Ethiopia’s public sector structure is fairly typical of many “transitional”
countries moving from a highly centralized state dominated by a single party system to a more
market-orientated and democratically pluralistic decentralized state. For instance, the
intergovernmental administrative structure retains many features of a deconcentrated system.
There is a degree of dual subordination of local units, who are expected to respond not only to
their own local government organization but are also expected to respond to technical
instructions from the higher-level government. As such, the head of the local Office of Finance
and Economic Development (OFED) in a ULG reports not only to the Mayor of his (or her) City
Administration but also has a clear line of accountability to the regional level Bureau of Finance
and Economic Development (BOFED). Similarly, the Head of the regional BOFED reports to
the regional President and Council, while at the same time reporting upwards and receiving
technical instructions from the Federal Ministry of Finance and Economic Development
(MOFED). Such dual subordination was common in the (previous) highly centralized French
administrative system and reminiscent of the dual subordination held under centrally planned
systems, which were common under the Soviet system (and by extension, other planned
economies).
Despite the de jure federal nature of Ethiopia’s public sector, similar vertical reporting and
administrative structures exist among the (federal) line ministries, (regional) bureaus and (local)
offices in each sector. Most relevant to this study, MUDC’s counterpart at the regional-state level
is the Bureau for Urban Development and Construction (previously the bureau responsible for
trade, industry, construction and urban development), and—at the local level—the City
Manager’s Office (in urban local governments).
Within ULGs, the City Manager’s Office reports to the Mayor and captures all the “municipal”
functions. However, the City Manager is only one department head in the executive municipal
cabinet, which further includes the heads of the local education department, health department,
the head of OFED, and so on.
Administrative interdependence in financial management. As already noted, in Ethiopia’s system

of “interdependent federalism”, despite their formal legal and political autonomy, different
government levels are not given full autonomy over a wide range of public sector functions. In
contrast, federal government is given the constitution power, as further defined and described in
Proclamation 691/2010, to plan and set policy standards for key social services (including
education, health services, and so on), whereas constitutionally the delivery of these services is a
regional (state) affair. In turn, regional states typically delegate the delivery of key social
services to the woreda level, to which they pass on the policy standards and priorities from the
federal level.
Similarly, federal instructions guide the financial management and procurement processes at all
government levels, which can result in higher-level governments being designated the
responsibility to procure capital goods on behalf of lower government levels. One practical
benefit of this collaborative, interdependent arrangement is that government entities at all levels
12
use the same Chart of Accounts and all use a single integrated budget and expenditure system
(IBEX) to report on public sector finances (Box 2.2).
Box 2.2: Consolidated Government Accounts, Municipal Expenditures, Municipal Revenues and
the Integrated Budget and Expenditure (IBEX) reporting system
All government levels and jurisdictions in Ethiopia are expected to use a single Chart of Accounts and
report on their revenues and expenditures through a common Integrated Budget and Expenditure (IBEX)
reporting system (IBEX). As such, IBEX is used to prepare the governments’ consolidated financial
accounts. In this context, information extracted from the IBEX system (at the federal, regional as well as
local level) informs various parts of the current study. As data is being drawn from IBEX for different
parts of the analysis, it is useful to briefly note the strengths and weaknesses of the system. The
discussion here focuses on IBEX as a tool for preparing consolidated financial reports (a secondary
function of IBEX), rather than assessing IBEX a tool for integrated financial management information
system (IFMIS) for each individual government jurisdiction (its primary function).
IBEX is a robust, domestically developed budget and expenditure reporting system. The most obvious
strength is the relatively high data availability for public expenditures and revenues at different
government levels based on (at least, notionally) a consistent Chart of Accounts. In this role, IBEX plays
an important part in achieving a relatively robust public financial management system and high degree of
fiscal transparency in Ethiopia.
Detailed reports from IBEX not publicly available. Given the consistency of the IBEX coding structure,
and its nation-wide roll-out and use at all government levels, it should be possible to readily produce—
and make publicly available—detailed reports on local/municipal finances. For instance, all City
Administrations are required to report all revenue collections through IBEX (both state revenue
collections and municipal revenue collections) to BOFED. If properly provided by municipalities, and if
consistently consolidated by regional and federal authorities, the resultant IBEX database should be able
to provide an instant, detailed picture of municipal revenues and expenditures for each local jurisdiction
in the country. There is no reason not to make such data on public finances widely available (as is done
in other countries, for instance, in Mozambique). However, we were unable to get such detailed,
disaggregated data from MOFED in time for analysis in this report. As such, whenever necessary, this
report relies on municipal financial data collected directly from the field.
Exclusion of municipal expenditures. Despite the wealth of data (potentially) contained in IBEX,
analysis of available IBEX data suggests that the system is not used to its full potential in terms of
capturing municipal-level finances. For some reason, it appears that MOFED concentrates its reporting
exclusively on expenditures for state functions—despite the merger of local deconcentrated entities and
city administrations some years back—by maintaining the historical separation between state functions
and municipal functions. Instead, as the steward of all public finances, it would be appropriate for
MOFED to report on all public finances through IBEX.
For instance, on the expenditure side, there would be two ways to capture “municipal expenditures” from
IBEX. First, municipal expenditures are categorized in the Chart of Accounts under Head 500. In the
disaggregated IBEX reports supplied to the study team, it appeared that a number of states simply did not
report on Head 500. Second, another approach to uncover municipal expenditures might be to analyze all
public expenditures at the local level that are funded by municipal revenues. Although all funding now
flows (or should flow) through a single account at the local level, analyzing expenditures by funding
source should give an idea how municipally collected revenues are actually spent. However, we were
13
unable to obtain such expenditure reports from MOFED; the only available reports were expenditure
reports for expenditures funded from ‘treasury sources’ and/or from international development agencies.
Consolidated government accounts focus on expenditures and revenues, not on intergovernmental

transfers. There is also a concern with regard to the way in which IBEX data are used to prepare
consolidated government accounts. These accounts focus heavily on showing the total amount of
revenues and expenditures—disaggregated by government level—but do not reflect flows of
intergovernmental fiscal transfers.
This practice may stem from Ethiopia’s prior (de facto) deconcentrated state structure, in which
“transfers” are merely cash-flow transfers (“non-subsidy financial transfers”) provided to funnel resources
to and from the central treasury within a single government unit, depending on where they collected and
are where they spent. In contrast, in Ethiopia’s nascent fiscal federalism, intergovernmental fiscal
transfers should be considered grants or subsidies, which are outlays of the higher-level government and
receipts of the lower-level government (IMF, 2001). While the consolidated general government accounts
report on expenditures and revenues at the federal, regional and local/woreda levels (with the caveats
noted above), these general government accounts (as made available by MOFED) tend to ignore
intergovernmental transfers from the federal to the regional level (presumably on the argument that they
are unimportant, as the outlay of federal block grants is offset by the receipt of these grants at the regional
level). Similarly, regional-level financial accounts fail to report on the flow of state-level transfers to the
local level.
Absence of earmarked transfers in the budget classification structure. A final broad area of concern
with the way in which the general government accounts and IBEX capture intergovernmental finances is
the absence of “space” in the budget classification system for earmarked grants. Whereas budget codes
exist in the Chart of Accounts for federal-regional block grants as well as regional-local block grants, but
codes are generally not available for earmarked block grants (with the exception of transfers from the
Road Fund). To the extent that most “earmarked grants” are in fact projects or programs funded from
external (international) grants or loans, these funding flows are recorded not as grants (under Code 1600)
but rather as externally-funded financing (under revenue Code 2000 or 3000). In other cases, the
consolidated financial reports may actually fail to properly capture the financing for urban local
governments provided by international development agencies.
Other anomalies and concerns with regard to the capturing and reporting of public sector finances
(including expenditures, revenues and transfers) at different government levels in IBEX are noted
throughout this report.
14
3. Relations between regional and city governments
3.1 De jure autonomy of urban local governments versus de facto interdependence
Ethiopia had a tradition of centralist authoritarian government until 1991. Municipal structures
were established in some cities, but the mayors were centrally appointed and municipalities were
treated as branches of central government. When the current government came to power in 1991,
it proclaimed a decentralized form of government and developed a constitution that established a
Federal Democratic Republic, consisting of nine Regional States, the federal capital city Addis
Ababa, and the special administrative region of Dire Dawa. The government structure has four
tiers—federal, regional, woreda (or city/municipal) and kebele (neighborhood). The nine
regional states have their own constitutions. The country has introduced a dramatic change in
terms of its traditions of governance. This legal framework has enabled more participation of the
regional states in matters that concern them.
Each region has developed specific “City Proclamations” that specify the powers, duties and
responsibilities of the cities.
Legal and Institutional Framework for Cities

Addis Ababa City Government was originally established by the Addis Ababa City Government
Charter Proclamation No 87/1997, and the proclamation revised in 2003 by Proclamation 361/2003.
Dire Dawa City Administration was established by federal Charter Proclamation 416/2004.
Amhara Regional State: Amhara passed its first City Proclamation No. 43 in 2000. The urban reform
process in Amhara Regional State is now guided by the Revised Proclamation for the establishment,
organization and definition of powers and duties of urban centers in the Amhara Region, No. 91/2003.
Oromia Regional State passed its first City Proclamation as no. 65/2003. The revised City
Proclamation 116/2006 now guides the urban reform process in Oromia Regional State.
Southern Nations Nationalities and Peoples Regional State passed its first City proclamation as no.
51/2002. The urban reform process in SNNP Regional State is guided by the revised proclamation
103/2006.
Tigray Regional State: Tigray’s first regional city proclamation was no. 65/2003. The urban reform
process in Tigray Regional State is guided by the Tigray Regional State revised City Proclamation
107/2006.
All five remaining National Regional States - Afar, Benshangul/Gumuz, Gambela, Harari and
Somali have, in 2006/07 developed legal and institutional frameworks (City Proclamations) to
establish local urban government authorities. Some of these have subsequently been amended by
(number and years of City Proclamations):
Gambela: 73/2008, Somali 90/2011, Benishangul Gumuz 69/2007.
Under the highly centralized Derg regime (1974-1991), Ethiopia’s municipalities were
marginalized and did not function as independent local authorities. Since 2000, national
decentralization policies have formed part of a large-scale reform of government resulting in
creation of institutional and legal frameworks for urban local government authorities. The
objective has been to create and strengthen urban local government that will ensure public
15
participation, democratization and enhance decentralized service delivery through institutional

reforms, capacity building, systems development, and training.
In all regions, the Bureaus of Urban Development and Construction are responsible for urban
management and development issues within the regional government. All regions that have
enacted legislation creating urban local government (or city) authorities have adopted an urban
governance model that follows the elected council, elected mayor, Mayor’s Committee and city
manager system.
The 2000-2003 City Proclamations were developed region by region (starting with Amhara) in a
fairly robust participatory process initiated by MUDC with consultancy inputs from GTZ (now
GIZ). There was strong debate around principles – elected mayor/non elected, etc. In the end all
regions followed, more or less, the result arrived at in Amhara and Tigray – Amhara then revised
theirs to included additions other regions introduced.
The City Declarations are therefore relatively similar in scope and form. Generally, in principle
they provide significant autonomy for cities to set their own standards, policies, plans, budgets,
etc. – but other parts of the city proclamations also specify that the region will issue guidelines
and directives. Section 3.2 below explores the degree of “autonomy” in more detail with focus
on municipal revenue management.
3.2 Regional and City/Municipal legislation on municipal revenues
The basic regional legislation that regulates cities’ powers and responsibilities is found in the
respective City Proclamations. However, various subsidiary legislations – most importantly the
“City Tariff Regulations” - provide more detailed guidance. The City proclamations give only
very broad and generally wide-ranging authority to the cities10. For instance in Oromia, cities are
bestowed with powers to “introduce, adjust and ensure the collection of taxes, service charges
according to law”, in Tigray, Gambella and Benishangul Gumuz the cities are empowered to
“determine and collect taxes”. However, the practice that regional governments determine local
tariff and tax rates has continued regardless and they continue to provide very detailed guidance
in the form of City Tariff Regulations. Typically the tariff regulations include a prescriptive
rating of the various urban areas within the region according to their size and economy (typically
five grades)11. For each of these grades of urban areas the tariff regulations will subsequently
either set a fixed rate or a band for each of the revenue sources.
Annex 3 provides a more detailed overview of the legislation and tariff regulations for each of
the regions.
10
For detailed review of the legislation and tariff declarations – please see Annex 3.
11
In the case of the more recent regulations from Benishangul Gumuz the criteria include: A) Annual
income/revenue of the woreda in which the town/city is located; B) Annual revenues collected by the town/city; C)
Level given to them under Proclamation No. 69/2007; D) Administrative importance; E) Interest of the region; and
F) Accessibility of towns/cities.
16
Table 3.1 below summarizes the extent to which the regional legislation and regulations
effectively allows each city to set its own tariff and tax rates. As indicated by the table, there is
some variation of the extent to which regions grant cities autonomy – but the general rule is that
the regions set the tariff and tax rates with limited or no autonomy for cities to adjust rates.
Table 3.1 Degree of Autonomy for City Management of Revenues
Region Degree of autonomy set by City Tariff Regulations
Tigray Tariffs and rates are set by region – no autonomy

Benishangul -
Gumuz
Gambela
Amhara Regional council sets the tariff rates but provides a range within which the
Somali cities have some limited discretion.
SNNP
Oromia ULGs appear to have full autonomy to select a rate appropriate for their
circumstances and/or to introduce a new tax base other than what is
suggested by the regional revenue Bureau. In practice, Cities decide within
a recommended range established by the region.
Harari Special case as the City is almost the same as the state (mayor = president,
council, revenue office – although with separate legislation for the city,
urban bureau and city administration, etc.)
Annex 4 provides an example of how regions set the tariffs – as can be seen from the Annex, the
tariffs, taxes and fees are spelled out in great detail for each and every city in the region – and in
the case of this particular region (Benishangul Gumuz) with no scope for city level adjustments.
In the cases where cities are given some scope for adjustment of rates, it appears from the
fieldwork that cities generally prefer to opt for lowest possible tax rates in order not to “disturb”
tax-payers (for a comparison of actual tax rates in cities see Chapter 12). In some cases – as in
Benishangul Gumuz – tax-payers react strongly to increased tax rates: out of the approximately
900 businesses in Assosa City, 850 had filed a complaint against the new tax rates and the city
didn’t enforce the regulations but await amendment by the regional council.
17
3.3 Regional Oversight and Support for City Financial Management
Regions provide the basic legal framework and guidelines for financial management to the cities.
The legislative process has varied significantly across regions although the federal government
has issued various model (“prototype”) laws and regulations (e.g. for general city financial
management – but not model tariff declarations). It is difficult to get a precise overview of the
legislation and regulations that are effectively in place as all laws and regulations are not well
published by regions – e.g. no publication of legislation on website.
Some legislation, e.g. “the Tigray National Regional State Municipalities’ Trade, Professional
Service and Service Tax and Tariff Determination, Council of the Regional Government
regulation 6/1997” is outdated. Although a revised legislative instrument has been in process for
years it is still to be approved.
In addition, the regional BOFEDs provide technical support in the form of training and other
capacity building for financial management –in particular the use of IBEX.
The regional BOFEDs receive regular financial reports and compile financial reports on both
state and municipal expenditures through IBEX.
The Region (BOFED and Revenue Authority) monitors revenue collections with particular
emphasis on state revenue. Regional level monitoring of municipal revenue is limited with some
minor efforts by BUDC and no substantive effort is made to analyze trends beyond the aggregate
level.
3.4 Regional Oversight Over Urban Development
In all regions, the Bureaus of Urban Development and Construction are responsible for urban
management and development issues within the regional government. All regions that have
enacted legislation creating urban local government (or city) authorities have adopted an urban
governance model that follows the elected council, elected mayor, Mayor’s Committee and city
manager system.
The Bureaus of Urban Development and Construction generally support cities with development
of urban plans, monitor urban development and also initiate support for infrastructure
development. This may take the form of direct project implementation on behalf of the cities –
such as, e.g., development of asphalt roads or other infrastructure or initiation of other financing
schemes such as the “Incentive Based Grants” for city development.
18
3.5 Regional Auditing of Urban Local Governments Accounts
The regional Auditor General’s Offices audit cities as well as woreda and other entities within
the regions. In general the AG’s Offices apply the Ethiopian Audit Standard, which is adopted
from AFROSAI-E 2012.
In general the regions regularly audit all the ULGDP supported cities, but are not necessarily
able to cover all smaller cities on a regular basis – priority is given to larger entities. Staff
vacancies limit ability to complete audit of all entities in all regions. For instance Somali
Regional State Office of Auditor General has total 68 staff (40 technical and 28 administrative
staff): Technical staff are not adequate to cover all Government entities in the region. According
to the approved structure an additional 50 employees are needed to cover all entities.
As part of the assessments of the local government fiduciary systems, the TOR requests a sample
based assessment of the quality and coverage of the internal and external audits of the local
governments. The study has therefore carried out interviews with audit staff in selected local
governments and reviewed audit reports and management letters to assess the quality of these
audits.
The following audits of local governments were included in the sample:

• Amhara Regional State Office of Auditor General,
• Auditor General, Oromia Region,
• Somali Regional State Office of Auditor General,
• Addis Ababa City Administration, and
• Adwa, Adama, Hawassa, Hosanna, Mekele, SNNPR, Tigray, and Ziway Batu cities.
The findings and assessments are outlined in the following sections.
The scope, comprehensiveness and frequency of audit

All funds and entities of the local governments are covered by the external audit. The focus is on
financial and compliance audits. Performance audit is only carried out on a limited scale. Special
audits such as audit of frauds are carried out on request.
Annual audit is given the highest priority and undertaken in all local governments. Interim audit
is not yet carried out in all local governments because it is a relatively new type of audit and
sufficient manpower is not yet in place to carry out this type of audit. The largest cities and
woreda are given the highest attention by the audit units, due to the complexity of the tasks. The
size of cities is in some regions relatively minor compared to the larger rural woreda and cities
will therefore be given less priority.
Compliance with audit standards

The external audit units use standards and manuals, which are in full compliance with
international standards to carry out audit. The audit units are independent from external and
internal interference in the audit. The code of ethics is used as a standard for the behavior of the
audit unit. Annual audit plans are used and a detailed planning of the audit is undertaken based
on locally collected information. The planned audit is supervised, documented, reviewed and
followed up by audit managers and senior staff. Objectives of the audit are defined before the
19
audit execution. Team members are assigned to the audit tasks based on the planning of the audit
to be carried out. An audit plan based on risk assessments and materiality is established and
applied for the audit. An approach is selected for the audit to be executed. The evidence to be
provided is included in the planning. Methodologies with audit criteria are applied and evidence
is collected to draft conclusions, recommendations and to form a clear opinion in the reports. The
audited entities are consulted to discuss audit findings before the final recommendations and
opinions are drawn.
Internal control
The evaluation of the internal control system is given high priority. The internal control
procedures are examined. An evaluation of the reliability of the internal control system is
undertaken with the support of the internal audit units. Evaluations of typical control procedures
such as segregation of duties, authorization and approval, and asset management are included.
Information system audit has commenced in some of the local governments but only to a limited
extent because this requires special competences and IT knowledge.
Evidence
Audit findings, conclusions, and recommendations are based on documents and evidence
collected during the audit. Evidence tracking sheets are used. Sampling techniques for data
collection are applied based on risk and materiality assessments. The audit of the internal audit
unit is controlled by the external audit to ensure due diligence of the audit. The findings of the
internal audit are used as an integrated part of the audit of the external audit unit.
Supervision and review

Audit tasks are undertaken according to established laws and regulations. There is a system of
quality assurance for the audit of the auditors based on audit objectives. Peer to peer evaluations
are applied. The audit is supervised including oversight of documentation of the evidence behind
the audit opinion. Potential and actual instances of fraud, embezzlement, misuse, and corruption
are investigated during the financial and compliance audit. Insufficient actions taken by the
administration to eliminate these occurrences and the risks of these are identified and
documented. Legal actions and internal control measures taken to safeguard assets and prevent
similar problems are recommended.
Reporting standards and reports

Standard report formats, including objectives, the scope of audit, applied methodology,
documentation, audit findings and recommendations, are applied by the audit unit. Working
papers are used to document the audit. Clear references to documents and evidence are found in
the audit queries. The annual audit report forms a clear opinion in a standard form relating to the
financial statements. The report sets out in a structured form the findings, recommendations, and
actions to be taken. Compliance of transactions and activities with laws and regulations, and
major matters on inadequate systems of control, illegal acts and fraud are adopted.
Management letters directed to the management directly responsible, as to systemic and internal
control issues such as weaknesses in administrative procedures for cash and bank accounting,
periodic payments, transfers, account receivables are issued. There is a clear demarcation of the
20
contents between the management letters and the audit reports12. Materiality considerations as to
what should be taken from the management letters and included in the audit report are identified.
However, some examples of insufficient consideration of materiality are also found in the audit
reports. A disclaimer for the areas where the auditor is unable to arrive at an opinion regarding
the financial statements is provided.
Interaction and audit follow-up

The timeliness of the audit has improved over the years. However, delays in delivery of audit
reports of up to six months have been found, for example in Hossana. At times, annual accounts
are closed late, which delays the audit and presentation of the audit reports. The follow up on the
audit reports has generally been slow, but has, reportedly, improved. Follow up actions have
been taken to improve responses to findings and recommendations. Steps are taken to check if
corrective measures in previous audit reports are, or are not followed up. The audit of the
previous year is examined as part of the audit of the current year. Oromia has created a team
specifically to follow up on the previous year’s recommendations. The elimination of prior
years’ audit report backlogs has improved. Internal audit is following up on the findings of the
external audit. In most instances measures to rectify negative findings are taken by the cities’
cabinet, the mayor or regional BOFED. A high degree of responses to audit findings has been
reported.
Legislative scrutiny of external audit reports

Budget and audit standing committees at both city and regional levels are reportedly scrutinizing
findings and recommendations in audit reports and undertaking legislative review of follow up
on findings within the requested time frame. Consolidated annual audit reports are sent to the
regional council finance and audit committee and presented to the council. Amhara regional
council’s finance and audit committee takes action, based on examination of reports, against
entities, which have repeated disclaimer or adverse opinions.
The capacity of the external audit unit

The capacity of the audit units is not sufficient to carry out all the planned activities because the
operations and complexity of audit tasks have increased. The lack of adequate manpower to cope
with all the planned audits, the insufficient number of high qualified auditors to plan, supervise,
and make quality assurance of the audit, and the insufficient number of staff with sufficient skills
within IT, law, engineering, water supply, environment, social sectors such as education and
health, sufficient number of computers, and a high degree of staff turnover have contributed to
underperformance in the execution of the audit services.
It is difficult to attract sufficiently qualified staff from the market and to retain existing staff
because salaries are low and benefits are not attractive compared to the private market. Private
auditors have been used to conduct the external audit in some regions to fill in the staff gaps.
Training and capacity building need to be strengthened to ensure continued and sufficient
adherence to international standards.
The arrangements and procedures for internal audit in the cities are discussed in Part III of this
report.
12
However, it is reported to the team that local governments outside the sample are not always using management
letters for the reporting to the management
21
PART II: THE FISCAL POSITION OF LOCAL GOVERNMENTS

The second part of this report describes the overall fiscal position of local governments in
Ethiopia’s system of intergovernmental fiscal relations, with an emphasis on urban local
government finance. This part is divided into four chapters. Chapter 4 provides a “big picture”
overview of the intergovernmental fiscal system and sub-national finances in Ethiopia. Chapter 5
provides an analysis of intergovernmental fiscal flows, including recurrent, capital and block
transfers to local governments. Chapter 6 provides and analysis of local government revenues
(with an emphasis on municipal revenues), while Chapter 7 considers the role of sub-national
borrowing in Ethiopia.
4. An overview of the intergovernmental fiscal system and sub-

national finances in Ethiopia
This chapter presents an overview of public sector expenditures across different government
levels, as well as an overview of public sector revenues at different levels. Analysis here largely
focuses on EFY 2004 (coinciding with Gregorian calendar years 2011-2012). Section 4.1
presents an expenditure profile of the public sector. Section 4.2 presents a revenue profile of the
public sector. Section 4.3 looks somewhat more closely at urban development expenditures at
different government levels. Details about specific elements of the sub-national (and in
particular, municipal) finances are dealt with in subsequent chapters.
The tables and figures presented in this table largely rely on data extracted from the
government’s IBEX system. Before proceeding further, it should be highlighted that there is
some reason to be concerned about the quality of the source data. For instance, variations in
reported expenditures and revenues in different iterations of (in principle, the same) data sets
were found. In other cases, expenditure reports or revenue reports were incomplete—for
instance, omitting data for one or more states). Despite the concerns, with noted exceptions, it is
believed that the data present a reasonable “big picture” of sub-national finances in Ethiopia.
4.1 Local Public Sector Expenditure Profile
Table 4.1 provides a “local public sector expenditure profile” or an overall picture of the public
sector expenditures in Ethiopia. Based on the consolidated government accounts (prepared by
MOFED based on IBEX data), the table breaks down all (on-budget) public expenditures into
five categories: (1) federal expenditures; (2) externally funded capital expenditures (which are
not assigned to either federal or regional levels); (3) regional (Bureau) expenditures; (4) woreda
level expenditures (including municipal expenditures funded from state revenues and grants);
and (5) municipal expenditures from municipal own source revenues.
Box 2.2 (in Chapter 2) notes the imperfections of the IBEX system as a data source for
consolidated public sector finances. These issues include the fact that municipal expenditures
(generally: local spending of own municipal revenues) are not reported through IBEX, and thus,
are excluded from the consolidated government accounts. This observation is very relevant to the
second phase of ULGDP: as argued further in Chapter 13 of this report, a well-functioning
22
intergovernmental fiscal system in Ethiopia will need to be supported by consolidated public

sector financial accounts that represent the major financial flows for all levels of government.
In order to nonetheless arrive at the “big picture” of the vertical distribution of public
expenditures, the broad assumption was made for Table 4.1 that all municipal revenues are spent
25-75 on recurrent and capital spending, respectively, and are all spent in the same fiscal year as
they are collected.13
Table 4.1: Public Sector Expenditures in Ethiopia, EFY 2004

ETB millions Recurrent Capital Total
Federal spending 18,679.5 34,214.4 52,893.8
Federal/Regional (external) capital spending 0.0 15,532.2 15,532.2
Regional (bureau level) spending 9,145.6 19,388.6 28,534.2
Woreda level spending 23,620.4 3,836.2 27,456.5
Municipal spending (OSR) 787.0 2,361.0 3,148.1
Total 52,232.5 75,332.3 127,564.8
ETB per capita Recurrent Capital Total
Federal spending 215.7 395.0 610.7
Federal/Regional (external) capital spending 0.0 179.3 179.3
Regional (bureau level) spending 105.6 223.9 329.4
Woreda level spending 272.7 44.3 317.0
Municipal spending (OSR) 9.1 27.3 36.3
Total 603.0 869.7 1,472.8
In millions of US$ Recurrent Capital Total
Federal spending 1,004.3 1,839.5 2,843.8
Regional (bureau level) spending 491.7 1,042.4 1,534.1
Woreda level spending 1,269.9 206.2 1,476.2
Total 2,808.2 4,050.1 6,858.3
As % of total public spending Recurrent Capital Total
Federal spending 14.6 26.8 41.5
Regional (bureau level) spending 7.2 15.2 22.4
Woreda level spending 18.5 3.0 21.5
Total 40.9 59.1 100.0
Source: Computed by authors based on data from consolidated government accounts (IBEX/MOFED). Note that
in these report, externally-funded capital expenditures are not specifically assigned to any government level:
they are reported here as “Federal and Regional Government (External) Capital Expenditures.”
13
While this ratio is based on municipal expenditure profiles collected for our twelve sample municipalities, it
should be understood that this is a broad assumption based on a small sample, resulting in an estimated allocation of
municipal resources between recurrent and capital expenditures.
23
The vertical allocation of public sector expenditures presented in Table 4.1 is not complete, as
the first two categories (both federal and externally-funded federal/regional spending) may
include funds that are spent at the regional (bureau) or local level. In addition, the consolidated
budget accounts (and PFM processes) do not appropriately record earmarked intergovernmental
fiscal transfers.14 For instance, given the fact that ULGDP-funded spending is not captured in
municipal budget accounts, ULGDP spending is not included in the municipal spending figures
reported here (which are solely based on local spending of own municipal revenue collections).
Instead, ULGDP spending is captured in the consolidated government accounts (and therefore, in
Table 4.1) as “federal and regional (external) capital spending”.15
The analysis in Table 4.1 suggests that a minimum of 46.4% of public expenditures is being
spent sub-nationally (the total sum of regional, municipal and woreda-level spending in the
bottom panel of Table 4.1: 22.4%, 21.5%, and 2.5%, respectively). In addition, as noted above,
some resources categorized as federal (or unassigned) may in fact be spent at the sub-national
level or trickle down to the sub-national level.16 However, if the view is taken that regional level
spending mostly consists of “overhead” spending that does not produce a direct benefit for
citizens, then only around 24% of public spending trickles down to the local level where front-
line services are provided.17
The imperfections of the data sources aside, the share of overall public spending that takes place
at the sub-national level in Ethiopia compares reasonably favorably with similar other countries.
Figure 4.1 places the vertical breakdown of public expenditures in an international context.
Because consistent government finance statistics for state and local government finances are
virtually unavailable for developing countries (outside of the LAC region), Figure 4.1 draws on a
recent international comparison of local public sector expenditures developed by the Urban
Institute’s Local Public Sector Initiative. An advantage of the LPSI methodology is that its
approach takes into account not only devolved expenditures, but also deconcentrated
expenditures and other localized public sector spending (for instance, spending on the
construction of schools and clinics, even if this spending is done by central line ministries rather
than by local authorities). Based on comparative analysis, the level of expenditure
14
This issue is discussed in greater detail in Section 5.
15
Instead, ULDGP is vertically reported upward through a parallel reporting mechanism for so-called Channel 1
funding, where “transfers” of ULGDP funds are recorded as “non-subsidy transfers” or “cash transfers” instead of
intergovernmental fiscal transfers that are recorded on-budget. If we were to adjust the figures in Table 4.1 by
counting ULGDP expenditures at the local level, this would shift roughly one additional percentage point of total
public expenditures from the central government to the local government level. Total ULGDP allocations for EFY
2004 equaled US$69.6 million, whereas actual disbursements
16
In addition to ULGDP, there are potentially other, large externally funded programs (which are recorded here as
federal/regional spending) that should be counted as sub-national expenditures, such as Promoting Basic Services
(PBS). Since a detailed LPS Expenditure Profile falls beyond the scope of the current study, Figure 4.1 below
assumes that 50% of all unassigned federal/regional capital expenditures take place at the sub-national level. This
results in an estimated 52.5% of total public sector spending taking place below the federal level.
17
Local-level spending would include woreda-level state expenditures (21.5%) plus spending from municipal own
source revenues (2.5%).
24
decentralization of Ethiopia’s public expenditures is among the highest among developing

economies in Africa and Asia.18
Figure 4.1 The size and composition of local public sector expenditures: selected countries
Sources: The size and composition of Ethiopian public expenditures is based on Table 4.1. For a detailed description
of the methodology and comparative country information, see (Boex 2012, 2013).
4.2 Local Public Sector Revenue Profile
Ethiopia’s Constitution defines the division of main revenue sources between the federal and
state level, with most high-yielding revenue sources being assigned to the federal level or jointly
to the federal and regional state levels. With respect to jointly assigned revenues, large taxpayers
(e.g., corporations) pay their taxes to the (deconcentrated offices of the) Ethiopia Revenue and
Customs Authority (ERCA). Smaller tax-payers pay their state-level taxes to the regional
revenue bureau or to the local revenue offices, which act as agents for the regional revenue
bureau in the collection of regional “state” revenues. In addition, local revenue offices within
municipalities (City Administrations) collect municipal revenues, which are in principal
municipal own source revenues (and historically retained in municipal account).19
The integrated budgeting and expenditure reporting system, IBEX, also records revenue
collections data for all public jurisdictions. However, in the same way that there is limited
18
Boex (2013) finds that there appears to be a relatively strong positive correlation between the size of local public
sector expenditures and government effectiveness.
19
Historically, municipal revenues were collected and controlled by the City Administration, whereas state revenues
at the local level were managed by the deconcentrated state administration. Since (Gregorian year) 2008, these
organizations were consolidated and all revenues are now supposed to be routed through a single account.
25
inclusion of municipal finances on the expenditure side of the system, IBEX is largely focused
on recording consolidated government revenues, which are often understood to include only
federal and regional (state) revenue. In recent years, efforts have been made to also record and
report municipal revenue collections through IBEX.
Table 4.2. Revenue profile for Ethiopia (actual revenue collections, EFY 2004)
ETB (mn) Percent
Consolidated (Fed / Regional) Revenue 102,863.7 97.0
o/w Federally Collected Revenues 85,879.7 81.0
o/w Regionally Collected Revenues 16,983.9 16.0
o/w Collected at Bureau Level 3,910.2 3.7
o/w Collected at Woreda Level 13,073.7 12.3
Municipal Revenue 3,148.1 3.0
Total Public Revenues (excluding Grants) 106,011.7 100.0
Source: MOFED/IBEX; Oromia municipal revenue reported by BOFED/Oromia
Table 4.2 presents a revenue profile for Ethiopia, based on actual revenue collections for EFY
2004 as reported through IBEX. 20 The table suggests that about 15.3% of all public revenue
collections takes place at local level, with approximately 3 percent of revenue collections
consisting of municipal revenues. Based on these aggregate figures, on average, municipal
revenue is about 20% of revenue collected at the local (woreda) level.21 Of course, this share is
lower in (mostly) rural woredas, and is likely to be higher in woreda-level municipalities. As
discussed further in Chapter 6, in the 12 cities visited, the share of municipal revenues (as a
percentage of state revenue collections plus municipal revenue collections) was approximately
30 percent. A more detailed discussion of revenue trends and patterns is contained in Chapter 6.
In making any judgments about whether urban local governments in Ethiopia collect an adequate
amount of revenue, the analysis should take into account that the level of municipal revenue
collections in any municipality (or in any country) depends on a variety of factors, including the
economic base of the urban area(s), the revenue sources assigned to the municipal level, the
compliance of local taxpayers, as well as the revenue collection and enforcement efforts of
municipal officials. In addition, since local revenues are much higher in urban areas than in rural
areas, urbanization patterns have a major influence on local (per capita) revenue collection
efforts. For instance, all else equal, a country that has an urbanization rate of 30% is likely to
have almost twice higher per capita local revenue collections compared to a country with a 15%
urbanization rate.
20
The data limitations should be taken into account in interpreting these figures. It was noted that in the IBEX data
provided by MOFED, no municipal revenue was reported for Oromia Regional State. In order to provide a more
complete fiscal picture, we included data on Oromia municipal revenue collections received from Oromia
BOFED/Revenue Bureau.
21
I.e., municipal revenues (ETB 3.1 billion) expressed as a share of all revenues collected at the woreda (and
municipal) levels (ETB 13.1 billion plus ETB 3.1 billion).
26
Since the nature and structure of urbanization and local revenues differs from country to country,
it is hard to compare local revenue levels across countries. However, when municipal own source
revenue collections in Ethiopia are placed alongside other countries in the region, it appears that
the (per capita) level of own source revenue collections in Ethiopia is similar (or at least, within
the same range of experiences) as other countries in the region (see Figure 4.2).
Figure 4.2: Municipal revenue collections (US$ per capita): Comparative analysis
Notes/Source: Ethiopia data drawn from Table 4.4. Comparative figures drawn from Boex, 2010. Domestic
Resources Mobilization for Poverty Reduction: The Role of Fiscal Decentralization and Local Revenue Mobilization
(The Urban Institute), and Boex, et al., 2011. An Analysis Of Municipal Revenue Potential In Mozambique.
Comparative figures represent municipal revenues for 2008 or later years.
Even the comparison of (per capita) municipal revenue collections in Ethiopia to other countries
in the regions (which are arguably more or less similar in economic structure) is made difficult
by underlying factors.22 For instance, it should be noted that Kenya, Mozambique, Rwanda and
even Tanzania are more urbanized than Ethiopia: Mozambique has an urbanization rate of 31.2%
whereas Kenya is 24% urbanized (compared to 17.4% in Ethiopia) (UN, 2009). In addition,
Ethiopia has a lower per capita GDP than any of the comparator countries: for instance, Kenya’s
GDP per capita is 80% higher than Ethiopia’s (World Bank, 2013). As such, care should be
taken not to infer too much from the comparison of nominal per capita revenue collections across
countries, even within the same region.
Instead of an international comparison that risks comparing “apples with oranges”, given the
federal nature of Ethiopia’s public sector, it might be more meaningful to compare local revenue
collections between different regional states (although the same caveats should be noted as
above, since Ethiopia’s regional states are by no means uniform).
22
It is even more difficult to compare local revenue patterns in Ethiopia with the situation in countries from other
regions, such as Europe, Latin America, China or even other parts of Asia. Differences in GDP, urbanization rates,
spatial economic structure, and revenue assignments vary considerably across global regions, thus hiding the role
that policy choices play in stimulating local economic development through local revenue policy. Furthermore, as a
result of the wide variations in country practices, policy lessons do not easily carry from one region of the world to
another.
27
Table 4.3 presents aggregate municipal revenues by state (in millions of Birr) broken down by
revenue type (i.e., state revenue versus municipal revenues). Table 4.4 (further below) presents
the same data in per capita terms. The tables reveal considerable variation among states in the
(nominal and relative) levels and composition of local revenue collections. For instance, local
governments in Tigray and Oromia seem to collect a relatively high proportion from municipal
revenue sources.
Table 4.3. State and municipal revenue collections by regional state (EFY 2004, ETB mn)
Municipal Muni. Rev.
Municipal (as pct of (as % of muni.
State Revenue Revenue Total total) rev. total)
Tigray 1,453.5 423.5 1,877.0 22.6 13.5
Afar 196.6 1.0 197.6 0.5 0.0
Amhara 2,409.1 223.1 2,632.2 8.5 7.1
Oromia 3,297.5 975.3 4,272.8 22.8 31.0
Somali 353.1 44.0 397.1 11.1 1.4
Benishangul 191.7 1.1 192.8 0.6 0.0
SNNPR 1,844.9 216.6 2,061.5 10.5 6.9
Gambela 109.3 4.4 113.7 3.9 0.1
Harari 76.7 0.7 77.4 0.9 0.0
Dire Dawa 162.2 57.4 219.6 26.2 1.8
Addis Ababa 7,864.7 1,200.7 9,065.4 13.2 38.1
TOTAL 16,983.9 3,148.1 20,132.0 15.6 100.0
Source: MOFED/IBEX; Oromia municipal revenue reported by BOFED/Oromia
There are notable asymmetries in local revenue collection patterns, and the importance of
different regional states in the municipal revenue policy environment in Ethiopia. For instance,
excluding Addis Ababa as being a separate case in terms of municipal revenue and finances,
Figure 4.3 suggests that 94.4% of municipal revenues are collected in the four largest regional
states (Oromia, Tigray, Amhara, and SNNPR).23
23
Of course, Dire Dawa and other smaller states have a different dynamic; depending on ease of collections, it may
make more sense to rely more heavily on state revenues.
28
Figure 4.3. Municipal Revenue Collections by Region (excluding Addis), EFY 2004 (ETB mn)
In per capita terms, there are also considerable variations between the states. Table 4.4 shows
state and municipal revenues per capita, as well as presenting municipal revenues per urban
resident. 24 The per capita (per urban resident) variation in municipal revenue collections is
further graphically presented in Figure 4.4. Whereas Addis and Tigray collect about ETB 380 per
urban resident, this reflects poorly with Amhara (ETB 82 per urban resident) and even worse
with Somali, Harar, and Afar. It should again be noted that variations in municipal revenue
collections should not automatically be attributed to poor local revenue effort. Local revenue
collections are determined by underlying differences in (taxable) economic activities, the
assignment of revenue sources, local tax rates, as well as local revenue administration and
enforcement practices. As further discussed in Chapter 6, regional legislation and practices with
respect to municipal rate-setting authority differ. As such, it is not possible to say much in
meaningful way about the level of local revenue effort in different municipalities or regions from
these figures alone.
24
Please refer back to Table 2.1 for data on population and urban population by regional state for July 2013.
29
Table 4.4. State and municipal revenue collections by regional state (EFY 2004, ETB per capita)
Muni. Rev.
Per Capita Per Capita Muni. Rev. (as % of
State Municipal Per Capita (Per Urban muni rev
Revenue Revenue Total Resident) total)
Tigray 287.1 83.7 370.8 378.0 13.5
Afar 119.2 0.6 119.8 3.7 0.0
Amhara 125.4 11.6 137.0 82.5 7.1
Oromia 102.3 30.3 132.6 228.8 31.0
Somali 66.4 8.3 74.7 58.2 1.4
Benishangul 186.4 1.1 187.5 6.3 0.0
SNNPR 103.1 12.1 115.3 100.8 6.9
Gambela 269.2 10.9 280.1 34.4 0.1
Harari 356.5 3.4 359.9 6.5 0.0
Dire Dawa 410.6 145.4 555.9 213.4 1.8
Addis Ababa 2,533.7 386.8 2,920.6 386.9 38.1
TOTAL 196.1 36.3 232.4 208.5 100.0
Figure 4.4. Municipal revenue collections by region (ETB per urban resident, EFY 2004)
Further analysis of local (urban) revenue patterns in Ethiopia—including some analysis of local
revenue collection patterns over time—is presented in Chapter 6 of this report.
30
4.3 Spending on Urban Development
Some information is available from consolidated government accounts (through MOFED/IBEX)

on expenditures on urban development by different government levels in Ethiopia, as urban
development spending constitutes a category of spending under ‘Economic Activities’ in
Ethiopia’s budget classification structure. Table 4.5 provides an overview of spending on urban
development at different government levels.25
Table 4.5: Urban development spending at different government levels, EFY 2004
Urban Total Public Urban dev. as Urban dev. as pct
development Spending % of total of total urban
(ETB mn) (ETB mn) dev.
Federal 232.7 52,893.8 0.4 4.5
Fed/Reg External Capital 223.7 15,532.2 1.4 4.4
Regional (Bureau) 869.6 28,534.2 3.0 17.0
Woreda Level 658.3 27,456.5 2.4 12.8
Municipal (OSR) 3,148.1 3,148.1 100.0 61.3
TOTAL 5132.3 127,564.8 3.7 100.0
Source: Computed by authors based on data from IBEX/MOFED.
As noted in sub-section 4.1, however, the consolidated budget accounts in Ethiopia only report
on spending from treasury sources (i.e., federal/state revenues and block grants) and external
sources (international grants and loans), but exclude local expenditures funded from municipal
own source revenues. In order to get a view that is as comprehensive as possible, it is assumed
here that municipal own sources are fully spent on urban services and other urban development
activities.26
Table 4.5 suggests that the federal government only makes a small contribution to urban
development spending, with domestically-funded federal spending representing only 4.5% of
urban development spending.27 Most urban development spending takes place at the regional and
25
As is the case with all reports generated by IBEX, caution should be used in interpreting these data. For instance,
it is possible (in fact, it is likely) that inconsistencies in the categorization and/or coding of the underlying budget
data make it difficult to consistently isolate urban development spending. For instance, recurrent federal spending on
urban development was computed as federal recurrent spending on “Urban Development and Construction”
excluding spending on road construction. Federal capital spending (and federal/regional external spending) on urban
development was categorized as spending on “Urban Development and Housing” (with Road Construction being a
separate spending category).
26
As also already noted, ULGDP appears to be recorded in the consolidated government accounts as externally-
funded federal/regional capital expenditures. Because the capital grants are not spent at the federal or regional level,
but rather, provided to urban local government authorities as cash transfers, it is quite likely that the expenditure of
these funds at the local level is not recorded as urban development spending within the consolidated government
accounts.
27
An additional ETB 223.7 million of externally-funded capital expenditures (4.4% of total urban development
expenditures in Ethiopia) are made on urban development, without indication whether these expenditures are made
at federal, regional (bureau) or woreda levels.
31
local levels, with the regional level representing 17% of (domestically-funded) urban
development spending. State-level urban development spending takes one of several forms
(beyond the recurrent cost of the urban development bureau): regional states co-fund ULGDP
spending at the local level; some regional states provide direct state-level spending on urban
development projects; while some states provide ULGDP-inspired incentive grants to
municipalities for urban development.
Close to 13% of urban development spending takes places at the woreda level funded by state
revenues, whereas almost two-thirds (61.3%) of urban development spending is actually funded
from municipal own source revenues. In fact, when ULGDP is left out of consideration, and if it
is assumed that all (or the large majority of) municipal revenues are spent on urban development,
more than eighty percent (82.5%) of local spending on urban services and urban development in
Ethiopia is funded from municipal own revenue sources.28 This is a considerable share. While it
is possible that some municipal revenues fund the recurrent provision of urban services
(sweeping streets, transporting solid waste to landfill, operational costs of abattoir, and so on),
fieldwork suggests that a considerable majority of municipal revenues are typically spent on
urban capital infrastructure.
Tables 4.6 and 4.7 present the regional breakdown of urban development spending. 29 Patterns
for urban development seem to be different for different regions. Of course, Addis Ababa and
Dire Dawa are (per capita) outliers as city-regions, spending more on urban development per
capita) than other regional states. Another clear pattern is the higher level of urban development
spending of the four bigger, advanced regions (Oromia, SNNPR, Amhara and Tigray) versus the
“emerging regions” (Gambella, Afar, Somali and Benishangul Gumuz).
In per capita terms, on-budget urban development expenditures range from less than US$1 per
capita (i.e., less than ETB 18.6) in several states (including Somali, Amhara and SNNPR), to
around US$2 (ETB 34.4) per capita for Oromia, and over US$6 (ETB 121.5) per capita in
Tigray. This variation is not surprising given the significant role played by own source revenues
in funding municipal infrastructure, and the relatively limited urban population of the emerging
regions.
The per capita figures presented in Table 4.7, however, may understate the effective level of
urban development spending, as the denominator is regional population, rather than urban
population.
28
This share is computed as municipally-funded urban development spending (ETB 3,148 million) expressed as a
share of total local urban development spending (ETB 3,148 million plus ETB 658 million).
29
It is important to note that these subsequent tables only include domestically-funded recurrent and capital
expenditures on urban development. To facilitate international comparisons, the Chapter 4 Annex contains Tables
4.6, 4.7 and 5.8 expressed in US dollars.
32
As such, Table 4.8 (further below) presents the level of urban development spending per urban
resident.
Table 4.6: Regional and Local Urban Development Spending by Region, EFY 2004 (ETB mn)
Regional (Bureau) Level Woreda Level Regional
Recurrent Capital Total Recurrent Capital Muni OSR Total Total
Tigray 23.1 114.0 137.1 26.7 27.4 423.5 477.7 614.8
Afar 6.3 29.4 35.6 0.5 0.0 1.0 1.6 37.2
Amhara 23.7 16.0 39.7 2.8 0.0 223.1 225.9 265.6
Oromia 53.5 80.2 133.6 0.0 0.0 975.3 975.3 1,108.9
Somali 3.8 26.4 30.2 0.0 0.0 44.0 44.0 74.2
Benishangul 0.0 0.0 0.0 0.0 0.0 1.1 1.1 1.1
SNNPR 3.4 1.1 4.5 24.5 31.6 216.6 272.7 277.2
Gambela 3.5 0.0 3.5 0.1 0.0 4.4 4.5 8.0
Harari 5.3 5.5 10.8 0.0 0.0 0.7 0.7 11.6
Dire Dawa 0.0 0.0 0.0 0.0 0.0 57.4 57.4 57.4
Addis A. 18.6 456.4 474.9 149.2 395.4 1,200.7 1,745.4 2,220.3
Total 141.1 729.0 870.1 203.9 454.4 3,148.1 3,806.3 4,676.4
Note: In order to achieve a more complete picture of urban development spending, municipal OSR is assumed to be
spent on urban services and urban development activities during the year it was collected. Note: The table does not
include (semi) off-budget municipal expenditures, such as ULGDP.
capita)
Tigray 4.6 22.5 27.1 5.3 5.4 83.7 94.4 121.5
Afar 3.8 17.8 21.6 0.3 0.0 0.6 1.0 22.6
Amhara 1.2 0.8 2.1 0.1 0.0 11.6 11.8 13.8
Oromia 1.7 2.5 4.1 0.0 0.0 30.3 30.3 34.4
Somali 0.7 5.0 5.7 0.0 0.0 8.3 8.3 14.0
Benishangul 0.0 0.0 0.0 0.0 0.0 1.1 1.1 1.1
SNNPR 0.2 0.1 0.3 1.4 1.8 12.1 15.2 15.5
Gambela 8.6 0.0 8.6 0.2 0.0 10.9 11.1 19.7
Harari 24.6 25.8 50.4 0.0 0.0 3.4 3.4 53.8
Dire Dawa 0.0 0.0 0.0 0.0 0.0 145.4 145.4 145.4
Addis A. 6.0 147.0 153.0 48.1 127.4 386.8 562.3 715.3
Per cap total 1.6 8.4 10.0 2.4 5.2 36.3 43.9 54.0
33
urban resident)
Tigray 20.6 101.8 122.4 23.9 24.5 378.0 426.3 548.7
Afar 22.4 104.8 127.2 1.9 0.0 3.7 5.7 132.9
Amhara 8.8 5.9 14.7 1.0 0.0 82.5 83.6 98.3
Oromia 12.5 18.8 31.3 0.0 0.0 228.8 228.8 260.1
Somali 5.0 35.0 40.0 0.0 0.0 58.2 58.2 98.2
Benishangul 0.0 0.0 0.0 0.0 0.0 6.3 6.3 6.3
SNNPR 1.6 0.5 2.1 11.4 14.7 100.8 126.9 129.0
Gambela 27.0 0.0 27.0 0.5 0.0 34.4 35.0 62.0
Harari 46.8 49.2 96.0 0.0 0.0 6.5 6.5 102.5
Dire Dawa 0.0 0.0 0.0 0.0 0.0 213.4 213.4 213.4
Addis A. 6.0 147.0 153.0 48.1 127.4 386.9 562.4 715.4
Per urban
resident 9.3 48.3 57.6 13.5 30.1 208.5 252.1 309.7
The fieldwork conducted for this study suggests that the preceding tables significantly understate
the level of urban development spending that has taken place in ULGDP cities. Based on the
available data in four ULGDP cities visited by field teams, Table 4.9 presents municipal
(recurrent and capital) expenditures (Head 500: Municipal Expenditures) made during EFY
2004, which is generally not reported through IBEX.30 In addition, the table presents the ULGDP
allocations received during the same financial year. These rough figures suggest that ULGDP
allocations potentially account for a considerable share (35-78 percent) of spending on municipal
services and urban development in cities that are included in the ULGDP universe; in three out
of four ULGDP cities included in the table, ULGDP allocations doubled municipal expenditures.
Municipal Recurrent Municipal Capital ULGDP ULGDP (as %

Expenditures Expenditures Allocations Total)
Hawassa 34,426,937 40,235,327 80,712,170 51.9
32,354,197 113,108,702 77.8
Adama
Mekelle 9,032,451 193,161,300 110,901,086 34.4
Bahir Dar 11,614,082 62,083,698 92,543,760 55.7
30
Note, Table 4.9 does not report on expenditures on state functions (which are funded from retained state revenue
or from regional block grants). The expenditure data that the team received from Adama does not provide a
distinction between recurrent and capital.
34
Aggregate size of urban investment needs in Ethiopia

It is not likely that the expenditure variations in the table above reflect differences in expenditure
needs for urban development across cities and regions in Ethiopia. It is much more likely that the
variations highlighted above are driven by resource availability, and possibly the degree to which
‘urban’ is a regional priority.
Given the disproportionate population growth of urban areas in Ethiopia, the level of unmet
urban infrastructure needs is considerable, and current spending levels on urban development—
as revealed in the preceding tables—do not come close to filling the gap in urban infrastructure
and services.
It is difficult to accurately estimate the requirement for urban infrastructure across urban areas in
an objective manner. This task is complicated by the fact that the provision of a desirable level of
urban infrastructure typically far exceeds the available resources in a context such as Ethiopia. 31
In an environment where data availability is limited, a rough estimate of aggregate urban
infrastructure needs can be computed using a more general methodology. 32 This approach to
estimating the size of urban investment needs is based on the notion that when rural residents
migrate to urban areas, they impose an additional demand on the stock of urban infrastructure
without taking their previous stock of public infrastructure with them. Without expanding the
level of urban infrastructure correspondingly, roads become congested, transportation networks
become overloaded, solid waste piles up in unmanaged locations, informal markets sprout up
where population growth occurs, and so on. Additional infrastructure is required both within the
Central Business District –which serves an increasingly large daytime population- as well as
across the rest of the urban area –which witnesses increases in population and population
density- and on the urban fringes, where the footprint of the urban area is expanding on lands
that were not previously served in terms of municipal services.
Ignoring, for a moment, the fact that the existing level of urban infrastructure is already
potentially inadequate for the population that it is serving, what is the unmet need for urban
infrastructure solely based on urban population growth? Based on an average urban growth rate
of 4 percent for a city of, say, 200,000 people, urban infrastructure is needed for approximately
8,000 new urban residents each year. If the urban infrastructure need per new resident is
assumed to equal $400,33 then the total annual requirement for new urban infrastructure would
31
One could use a norm-based approach (relying on physical norms) to compute the infrastructure needs of different
urban areas. However, such norm-based approaches require considerable data availability and generally ignore
considerable variations in the costs and demand for urban infrastructure and services. As such, norm-based estimates
of urban infrastructure needs often end up being rather subjective. In addition, coming up with a highly detailed and
ideal level of urban infrastructure need is irrelevant if this level is altogether unaffordable. Arriving at a highly
detailed estimate of aggregate urban infrastructure needs falls beyond the scope of the current study.
32
For instance, see The Urban Transition in Tanzania, World Bank ESW, 2008, Table S.1.
33
As an example, let us assume that an average city of this population size has 20 km of hardened (paved) road, or
about 10cm per urban resident, and let us assume a very rough estimate for the cost of procuring paved (or cobble-
stoned) road at US$2 million per kilometer, so that if we wanted new residents to contribute an average level of new
road infrastructure (in order to extend the road network in peri-urban areas and prevent congestion on the radial
access roads and in the city center), the cost for the required additional road infrastructure for each new resident
would approximately equal US$200 per new resident. It should be noted that this assumes that no congestion point
is reached and that there is no diminishing marginal need for urban infrastructure as the urban population grows. If
paved roads make us x percent of the urban infrastructure asset base, then the total infrastructure cost of a new
35
equal approximately $3.2 million per year (or about US$ 15 per urban resident, or around ETB
270 per person), only to maintain the infrastructure-to-resident ratio in urban areas.
It is important to note that this (very rough) estimate ignores the ongoing cost for delivering
municipal services and the cost of maintaining and rehabilitating and replacing—as needed—
existing urban infrastructure. Given the wide range of municipal services that in principle fall
under the category of urban development and urban services—ranging from construction and
maintenance of roads, to abattoirs and markets, to the collection and disposal of solid and liquid
waste, to urban water systems—an average actual spending level of ETB 300 per urban resident
(around US$ 17) merely to maintain the existing level of services for existing urban residents
seems quite modest.
4.4 Issues arising for the next phase of ULGDP
To the extent that the next phase of ULGDP aims not only to provide targeted urban
development funding and to improve urban local governance—but rather, also aims to improve
the overall intergovernmental fiscal system within which urban local governments operate—
there “big picture” overview of the intergovernmental fiscal system presented in this chapter
raises a number of issues for consideration in the design of the second phase of ULGDP:
• The consolidated government accounts in Ethiopia do not address urban local

government finances adequately. In particular, it is suggested that (a) consolidated
government accounts should take into account federal, regional as well as local-level
expenditures and revenues (i.e., woredas and City Administrations); (b) municipal
expenditures (Head 500) and revenues (Revenue Code 1700-) should be fully captured in
IBEX reporting; (c) ULGDP expenditures should be fully integrated into municipal
budget processes and expenditure reports.
• As further noted in Chapter 5, the budgetary accounting system should properly account
for intergovernmental fiscal transfers.
• From a policy standpoint, Ethiopia would stand to benefit considerably from the
preparation of an annual analysis of intergovernmental finance, such as South Africa’s
Intergovernmental Fiscal Review. Such an analysis would provide an ongoing evidence-
basis for policy discussions on urban (municipal) finance in Ethiopia.
resident would equal (100/x) * $200. For the purpose of this exercise, we will conservatively assume that paved
roads make up 50 percent of the municipal asset base of urban areas in Ethiopia, and therefore, that the estimated
urban infrastructure requirement for a new resident is $400. An alternative way of calculating the cost of new
residents would be to take into account the cost of site preparation for land plots for new urban residents. Such costs
can easily reach much higher; in the range of $1000-$2000 per new urban resident.
36
4.5 Chapter Annex (Data tables in US$)
Table A4.1: Regional and Local Urban Development Spending by Region, EFY 2004 (US$ mn)
Tigray 1.2 6.1 7.4 1.4 1.5 22.8 25.7 33.1
Afar 0.3 1.6 1.9 0.0 0.0 0.1 0.1 2.0
Amhara 1.3 0.9 2.1 0.2 0.0 12.0 12.1 14.3
Oromia 2.9 4.3 7.2 0.0 0.0 52.4 52.4 59.6
Somali 0.2 1.4 1.6 0.0 0.0 2.4 2.4 4.0
Benishangul 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1
SNNPR 0.2 0.1 0.2 1.3 1.7 11.6 14.7 14.9
Gambela 0.2 0.0 0.2 0.0 0.0 0.2 0.2 0.4
Harari 0.3 0.3 0.6 0.0 0.0 0.0 0.0 0.6
Dire Dawa 0.0 0.0 0.0 0.0 0.0 3.1 3.1 3.1
Addis A. 1.0 24.5 25.5 8.0 21.3 64.6 93.8 119.4
Total 7.6 39.2 46.8 11.0 24.4 169.3 204.6 251.4
Table A4.2: Regional and Local Urban Development Spending by Region, EFY 2004 (US$ per
capita)
Tigray 0.2 1.2 1.5 0.3 0.3 4.5 5.1 6.5
Afar 0.2 1.0 1.2 0.0 0.0 0.0 0.1 1.2
Amhara 0.1 0.0 0.1 0.0 0.0 0.6 0.6 0.7
Oromia 0.1 0.1 0.2 0.0 0.0 1.6 1.6 1.9
Somali 0.0 0.3 0.3 0.0 0.0 0.4 0.4 0.8
Benishangul 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1
SNNPR 0.0 0.0 0.0 0.1 0.1 0.7 0.8 0.8
Gambela 0.5 0.0 0.5 0.0 0.0 0.6 0.6 1.1
Harari 1.3 1.4 2.7 0.0 0.0 0.2 0.2 2.9
Dire Dawa 0.0 0.0 0.0 0.0 0.0 7.8 7.8 7.8
Addis A. 0.3 7.9 8.2 2.6 6.8 20.8 30.2 38.5
Per cap total 0.1 0.5 0.5 0.1 0.3 2.0 2.4 2.9
37
Table A4.3: Regional and Local Urban Development Spending by Region, EFY 2004 (US$ per
urban resident)
Tigray 1.1 5.5 6.6 1.3 1.3 20.3 22.9 29.5
Afar 1.2 5.6 6.8 0.1 0.0 0.2 0.3 7.1
Amhara 0.5 0.3 0.8 0.1 0.0 4.4 4.5 5.3
Oromia 0.7 1.0 1.7 0.0 0.0 12.3 12.3 14.0
Somali 0.3 1.9 2.2 0.0 0.0 3.1 3.1 5.3
Benishangul 0.0 0.0 0.0 0.0 0.0 0.3 0.3 0.3
SNNPR 0.1 0.0 0.1 0.6 0.8 5.4 6.8 6.9
Gambela 1.5 0.0 1.5 0.0 0.0 1.9 1.9 3.3
Harari 2.5 2.6 5.2 0.0 0.0 0.3 0.3 5.5
Dire Dawa 0.0 0.0 0.0 0.0 0.0 11.5 11.5 11.5
Addis A. 0.3 7.9 8.2 2.6 6.8 20.8 30.2 38.5
Per urban
resident 0.5 2.6 3.1 0.7 1.6 11.2 13.6 16.7
38
5. An analysis of intergovernmental fiscal flows
Both at the regional level as well as the local (woreda) level, intergovernmental fiscal transfers
form a critical component of sub-national finances in Ethiopia. Regional states receive the
majority of their financial resources as intergovernmental fiscal transfers from the federal level
(Section 5.1). In turn, regions provide major intergovernmental fiscal transfers to the local level
(Section 5.2). In the absence of major federally-funded mechanisms to provide grant funding for
urban development activities at the local level, allocations provided under the Urban Local
Government Development Project (ULGDP) form the main urban development grant in Ethiopia
(Section 5.3).
Even as part of a background review for a new urban development funding scheme, it is
important to understand the funding of local entities in Ethiopia’s (fiscal) federalism (including
funding provided through block grants), as the actions of all local governments may be
influenced by incentives that are caused by (both specific and general purpose) transfers
provided to the local level.
5.1 Overview of federal-regional intergovernmental transfer system
Basic nature of the federal-regional intergovernmental transfer system. Despite the

decentralization of revenue resources to the regional level in the constitution, regions in Ethiopia
are not able to cover their expenditure needs from their own revenue sources. This is due to the
fact that regions have extensive responsibility for service delivery, insufficient economic base,
and regional inequality in resource mobilization. The mismatch between expenditure
responsibilities and revenue assignments has been financed through intergovernmental transfers
from the federal government.
Table 5.1 presents the funding sources of regional state budgets for EFY 2003, broken down
between state revenues, grants (subsidy revenue), and external funding, showing that close to
two-thirds of regional state resources come from intergovernmental fiscal transfers from the
federal level.
Table 5.1: Revenue Sources of Regional States: State Revenues, Subsidies (Grants), and
External Funding (EFY 2003)
ETB US$ ETB per Percent of
million million capita total
State-Collected Revenues 12,999.1 698.9 150.1 33.1
Subsidy Revenue 25,967.8 1,396.1 299.8 66.1
External Assistance / Loans 311.5 16.7 3.6 0.8
Total 39,278.4 2,111.7 453.5 100.0
Source: MOFED/IBEX.
39
Notionally, two types of grants exist in the Ethiopian fiscal system: specific purpose grants
(SPGs) and equalizing block grants or General Purpose Grants (GPGs). In practice, considerable
emphasis is placed on the (equalizing, unconditional) general-purpose block grants. In contrast,
specific-purpose grants appear to be mostly (although not exclusively) aligned with externally
funded projects.
Specific-purpose grants. Federal Government provides specific-purpose grants for expenditure

assignments that are under its mandate but executed by lower tiers of government. Specific
purpose grants may be provided to fulfill the goals of national priorities that would not be
pursued sufficiently by other tiers of government, for instance, as a result of capacity weaknesses
at the sub-national level. Among others, specific purpose grants are (or have been) provided
under the Public Sector Capacity Building Program (PSCAP), the Food Security Program, the
Urban Local Government Development Project (ULGDP), the MDG Performance Fund, and the
Road Fund. 34
Tracking of specific-purpose grants through the financial management system is (unnecessarily)

weak. In fact, the Ethiopian Chart of Accounts is almost exclusively geared towards recording
the general-purpose block grants. Other than (recurrent and capital) block grants, the only
revenue that is recorded from a “real” specific-purpose grant within the revenue classifications
by regional states is the Road Fund Transfer (revenue code 1604). There is no space within the
current budget classification system to meaningfully record special-purpose grants; in other
words, from a budgetary viewpoint, special purpose grants basically do not exist in Ethiopia’s
intergovernmental fiscal system. As a result, unsurprisingly, data on special purpose grant
expenditure is not readily available. Indeed, Ethiopia’s Public Finance Review 2010 (World
Bank, 2010: p. 22) acknowledges that “full-fledged data on SPG expenditure is not readily
available”.
It is indeed virtually impossible to produce a detailed breakdown of donor projects and programs
that could be considered as “special purpose grants” to the regional or local level, as these
“transfers” are generally not recorded as intergovernmental transfers (under code 1600). Instead,
externally-funded intergovernmental resource flows from the federal to the regional level are
generally accounted for as External Assistance or Loans (revenue codes 2000/3000). 35 Data
availability, consistency and quality all stand in the way of meaningfully tracking special
purpose grants based on consolidated budgetary accounts. For instance, ULGDP transfers made
in EFY 2003 (US$62.5 million) far exceed the total amount of External Assistance/Loans
funding that was recorded to be received by the regional states in that year (Table 5.1).
Although it is virtually impossible to determine the actual volume of (quasi) specific-purpose

grants, it is clear that the magnitude of financial resource through special purpose grants is highly
dependent on donor funding: whereas prior to 2011 the Food Security grant was a substantial
34
For a deeper discussion and analysis, see: World Bank. 2010. Ethiopia Public Finance Review 2010. Report No.
54952-ET.
35
In selected cases, it appears that these special purpose grants to the regional level are recorded by the regional
states under revenue code 1603. In other cases, as noted further below in this chapter, “grant” flows may be recorded
as non-subsidy (i.e., cash) transfers—in other words, as a financial transaction rather than as a budgetary transaction.
40
share of the “special purpose grant” portfolio, this funding source has declined significantly over
the past few years. In contrast, the MDG Performance Grant is a relatively new entrant in the
intergovernmental finance system: since its introduction in 2011/12, the size of this grant has
rapidly increased and is now reported to be about half the size of the unconditional GPG.36
Main features of the federal GPG.37 The main justification for the central government to give
unconditional grants to states and localities is, first, to correct vertical fiscal imbalances (to
ensure adequate sub-national funding for the provision of a minimum or reasonable level of
public services), and second, to achieve horizontal fiscal balance (that is, to equalize fiscal
capacities of different states). Although a detailed analysis of the GPG in Ethiopia is beyond the
scope of the current study, it is worthwhile to highlight the key features of the grant scheme, as
this funding modality provides the majority of funding to the regional level (and furthermore, has
an important impact on the design of regional-local fiscal relations).
Policy objective: To equalize the fiscal conditions among regions so that, despite differences in
regional expenditure needs and revenue potential, they are fiscally capable to provide public
services of comparable type and quality as mandated by the constitution.38
Size of pool: The resource pool for the envelope of constitutionally mandated federal-regional
transfers is set annually by the proclamation of the federal budget. It is not determined by a
formula such as a percent of national revenues (as is the case, for instance, in Ghana or Kenya)
or as a percentage of revenues from a specific tax (such as is the case in Morocco).
As noted in Table 5.1., the size of the federal-regional block grant in Ethiopia for EFY 2003 was
approximately US$ 1.4 billion annually. (For reference, as noted below, ULGDP transfers over
the five-year period from EFY 2001-2005 were $210 million, or roughly $42 million per year).
Surprisingly, given the importance of the GPG block grant in Ethiopia’s fiscal system,
comparative data for the GPG is not readily available, preventing us from further analyzing the
vertical (and horizontal) allocation of block grants over time.39 Discussions with federal budget
officials indicate that the implicit rule is not to cut the size of the GPG pool, and if possible,
promote some growth in the vertical allocation of resources (Vaillancourt, 2013).
Equalization formula: The distribution of the federal-regional transfers is determined by a

formula developed and approved by the House of Federation. The allocation formula is revisited
36
See: House of Federation Team. Ethiopia Equalization Grant Distribution Experience, Presentation at the World
Bank. Washington, August, 2013.
37
This section is based in part on information provided by the House of Federation, as well as draft comments
prepared by Prof. François Vaillancourt (July 2013).
38
Relevant Constitutional provisions include Article 89/2 (which states that Government has the duty to ensure that
all Ethiopians get equal opportunity to improve their economic conditions and to promote equitable distribution of
wealth among them); Article 41/3 (which states that every Ethiopian national has the right to equal access to
publicly funded social services); and Article 89/4 (which states that Government shall provide special assistance to
Nations, Nationalities and Peoples least advantaged in economic and social development).
39
The consolidated government accounts look at consolidated public sector expenditures and revenues at all
government levels, and therefore ignore intergovernmental transfers. Most IBEX revenue reports at the regional
level focus exclusively on revenue collections (rather than transfers and external grants/loans). Comparative data is
available for EFY 2003 (presented in Table 5.1) and EFY 2002. The IBEX data available for EFY 2004 is
incomplete, thereby making it impossible to produce a comparison from 2002-2004.
41
and revised every few years; the current formula is to be used for a five-year period (FY 2012-
2013 to 2016-2017).40
The overall design of the block grant formula takes into account both a region’s revenue
potential as well as its expenditure needs. In fact, each region’s block grant is computed in
proportion to its gap:
GPG i = Expenditure needs i – Revenue potential i.
Since the revenue potential of Addis Ababa arguably exceeds its expenditure needs, Addis does
not receive a GPG allocation.
Each region’s revenue potential is calculated using a so-called Representative Tax System
(RTS). The basic idea underlying a RTS is to calculate the amount of revenue that a region
would collect given its tax bases if it were to exert an average level of fiscal effort (or the
average effective tax rate, AET) for each main revenue source. This is done by collecting data
on revenue collections and tax bases for each of the taxes under consideration for every sub-
national region. 41 Based upon information on all tax bases for every region, as well as the
national average, one can compute the amount of revenue that each jurisdiction would collect
under an average level of fiscal effort. This amount is then considered to quantify the fiscal
capacity of each jurisdiction.42
In the specific case of Ethiopia, the House of Federation compiles data for regional collections
and (proxy) tax bases for seven main revenue sources in each regional state: the pay roll tax; the
agricultural income tax; land use fee; the livestock tax; profit tax; turnover tax; and the Value
added tax (VAT).43 For each of these seven revenue sources, the federal government computes
the average effective tax rate (AETR j) by dividing the aggregate revenue collections for each
revenue source (across all regional states) by the aggregate tax base for that revenue source. It
then computes the revenue potential for each revenue source in each region by multiplying the
average effective tax rate by the actual tax base in each regional state, so that the revenue
potential for each regional state is determined as:
Revenue Potential i = Σ AETR j * Tax Base i,j

Similarly, the GPG formula determines the level of each region’s expenditure needs by first
dividing a notional resource envelope among the different functional responsibilities of the
regional level (based on a pre-determined percentage of the total resource envelope).
Expenditure needs are then determined for each functional grant window based on a sectoral
formula, including for general administration (28%); primary and secondary education (33%);
40
The GPG formula is also used to distribute the MDG Grant among regional states, which supports capital
spending in specific functional areas. The choice to use the GPG formula rather than one more aligned with the
specific nature of these transfers was a political one.
41
It is reported that Ethiopia’s RTS includes the following tax sources: Pay roll tax; Agricultural income tax; Land
use fee; Livestock tax; Profit tax; Turnover tax; and Value added tax (VAT).
42
For a more detailed discussion on “Designing Intergovernmental Equalization Transfers with Imperfect Data:
Concepts, Practices, and Lessons” (Jamie Boex and Jorge Martinez-Vazquez). In: Challenges in the Design of Fiscal
Equalization and Intergovernmental Transfers. Jorge Martinez-Vazquez and Robert Searle (ed.), Springer, 2007.
43
See House of Federation (2012) for further details.
42
public health (11%); agriculture and rural development (11%); drinking water development
(5%); and urban development (4%). Remaining functions attract 8 percent of the allocation
formula. The allocation factors used for each of these windows include (House of Federation,
2012):
Expenditure Allocation Factors Weight (%)

General Admin, Population size, pastoral share, % of slope, in formula 28
Extreme weather in ad hoc fashion (hardship allowances and
cooling expenditures)
Nationalities Unit cost * number of nationalities
Costs linked to cross border security issues are addressed in ad
hoc but reasonable fashion
Education Student enrollment and population density for those enrolled; 33
Unit costs of new students * projected enrollment:
(Number of languages of instruction in region-1) * unit costs
of text book development
Two regions (Harari, DD) receive a 10% premium on the sum of
these items for teaching students of bordering regions
Rural roads Number of kilometers to be constructed; 3
Number of kilometers to be maintained; sloppiness; cost
differentials
Drinking water Population with water access (CSA Welfare Monitoring 5
Survey)
Unit cost of provision (Ministry of Water and Energy)
Health Unit costs: average expenditure 2005-2010/ # of users 11
Unit cost of poor: twice this amount
Fixed costs for new health posts * planned health posts
Agriculture Data for 1995-2002 from various sources 11
Coefficient from estimation used to project
Environmental Regional Environmental expenditures 2005-2009 (RES) 2
protection Productive Safety Net Program beneficiaries (PNSPB)
Protected forest area; unit cost of forestry guard
Large and medium enterprise value added
Micro and Small Three year average urban unemployment rate from CSA. 2
enterprises Unit costs: #s from Tigray, Amhara, Oromia, SNNP and DD
Urban development Urban population 4
Unit cost from 2005-2009 data
For instance, the education “needs” of each region are determined by its student enrollment (and
the population density of enrolled pupils); the projected cost to attract school-aged children into
school that are currently not enrolled; and compensation for cross-regional enrollment. Similarly,
expenditure needs for urban development are assumed to be proportional to the size of the urban
population that resides in each regional state.
Conditionalities: In principle, the grant is unconditional and regional governments are free to
allocate the grant resources in accordance with local needs and priorities. However, the federal
government has an important (constitutionally-defined) role in identifying standards and
43
priorities for the main social sectors. As such, the federal government is able to use non-fiscal
channels (policy channels and administrative mechanisms) as needed to guide the spending on
sub-national resources, rather than relying on grant conditionalities.
Analysis and Discussion: Tables 5.2, 5.3, and 5.4 present the resource composition of regions
for EFY 2003 in millions of Birr, in Birr per capita, and as a share of total regional resources,
respectively.44 To the extent that subsidy (grant) allocations represent general-purpose grants, the
tables reflect the vertical and horizontal allocation patterns of the GPG.45
Table 5.2: Regional revenue composition, EFY 2003 (ETB mn)

External
State-Collected Subsidy Assistance /
Revenues Revenue Loans Total
Tigray 937.2 1,797.4 100.5 2,835.1
Afar 155.9 891.8 0.0 1,047.7
Amhara 1,561.6 6,053.2 0.1 7,614.9
Oromia 2,601.9 8,450.5 0.0 11,052.4
Somali 245.8 2,164.1 0.0 2,409.9
Benishangul 117.4 508.4 56.5 682.4
SNNPR 1,169.4 5,166.3 19.8 6,355.5
Gambela 85.5 405.4 0.0 491.0
Harari 53.8 228.7 2.9 285.5
Dire Dawa 108.0 260.2 0.0 368.2
Addis Ababa 5,962.5 41.6 131.8 6,135.8
TOTAL 12,999.1 25,967.8 311.5 39,278.4
Source: MOFED/IBEX. Note: Addis Ababa receives specific purpose grants, but no GPG/block grant.
44
Unfortunately, the comparable IBEX data set for 2004 which we received was not complete.
45
It is not clear to what extent donor-funded (specific-purpose) grant schemes are captured consistently by IBEX.
As noted in Chapter 4, in some cases, these grants may be reported as federal (or unallocated) capital expenditures,
rather than being reported as regional or local expenditures.
44
Table 5.3: Regional revenue composition, EFY 2003 (ETB per capita)
External
Tigray 185.1 355.1 19.9 560.1
Afar 94.5 540.5 0.0 635.0
Amhara 81.3 315.1 0.0 396.4
Oromia 80.8 262.3 0.0 343.0
Somali 46.2 406.9 0.0 453.2
Benishangul 114.2 494.6 55.0 663.8
SNNPR 65.4 288.8 1.1 355.3
Gambela 210.7 998.6 0.0 1,209.3
Harari 250.4 1,063.8 13.6 1,327.8
Dire Dawa 273.4 658.8 -0.1 932.1
Addis Ababa 1,920.9 13.4 42.4 1,976.8
TOTAL 150.1 299.8 3.6 453.5
Source: MOFED/IBEX. Note: Addis Ababa receives specific purpose grants, but no GPG /block grant.
The total allocation levels in Table 5.3 are difficult to interpret, since total allocations are to a
large degree driven by differences in the size of each region’s population. After taking this into
account (by specifying the figures in per capita terms), Table 5.4 suggests that—based in part on
the GPG formula—there are considerable variations in the (per capita) grant levels provided to
different regions, with some states (such as Harari or Gambela) receiving considerably greater
transfers than other states (e.g., Oromia or SNNPR). The variations in grant allocations, in fact,
are often considerably larger than the variations in revenue collections. This suggests that
equalization of expenditure needs is a driving force behind the observed allocation patterns.
The right-hand-side column of Table 5.3 also indicates that considerable (per capita) resource
variations continue to exist after subsidy resources are distributed. This pattern points to large
(perceived) variations in expenditure needs across regions. In particular, the high levels of grant
resources being allocated to Gambela, Harari and Dire Dawa are notable. Although it is not
surprising that smaller (less populous) states might be subject to scale economies and require
higher overhead expenditures on regional government administration (and therefore tend to have
higher per capita expenditure needs), the size of the per capita resource allocations being directed
towards these three smallest states is considerable.46
46
There are not only economic/fiscal but also important political economy considerations why smaller states might
receive greater intergovernmental fiscal transfers. For instance, see: “The determinants of the incidence of
intergovernmental grants: A survey of the international experience” (Jamie Boex and Jorge Martinez-Vazquez),
Public Finance and Management, Vol. IV(4), December 2004.
45
Table 5.4: Regional revenue composition, EFY 2003 (as share of total)
External
Tigray 33.1 63.4 3.5 100.0
Afar 14.9 85.1 0.0 100.0
Amhara 20.5 79.5 0.0 100.0
Oromia 23.5 76.5 0.0 100.0
Somali 10.2 89.8 0.0 100.0
Benishangul 17.2 74.5 8.3 100.0
SNNPR 18.4 81.3 0.3 100.0
Gambela 17.4 82.6 0.0 100.0
Harari 18.9 80.1 1.0 100.0
Dire Dawa 29.3 70.7 0.0 100.0
Addis Ababa 97.2 0.7 2.1 100.0
TOTAL 33.1 66.1 0.8 100.0
Source: MOFED/IBEX.
Finally, Table 5.4 confirms the importance across the board of intergovernmental fiscal transfers
in Ethiopia’s system of intergovernmental fiscal relations.47 In fact, only one region receives less
than two-thirds of its financial resources from the federal level through intergovernmental grants.
5.2 Overview of regional-local intergovernmental transfer system
There are similarities as well as differences between the federal-regional transfer system in
Ethiopia on one hand and the intergovernmental transfers at the regional-local level on the other
hand. At the regional / sub-regional level, the intergovernmental fiscal transfer system in
Ethiopia has an asymmetric element: some regions have transfers to zones for all or part of their
territories, while others exclusively rely on transfers to the local (woreda) level. In fact, in the
case of some city-regions, one finds transfers to woredas and directly to schools and health posts
from such zones (Vaillancourt, 2013).
Given the incomplete transition to a federalist system, the budget reporting system provides
inadequate information on regional-local block grants, making it nearly impossible to engage in a
systematic analysis of regional-local fiscal transfers. Regional states do not prepare state-level
budget reports that transparently reveal funding flows from the regional state level to the local
level. In fact, in line with the historical relationship between the regional and local levels, there
appears to be only a weak recognition that the woreda level is semi-autonomous. This is reflected
in the regional-local transfer system (as discussed further below).
As already noted earlier, consolidated government accounts present data on spending and
revenues at each level (federal, state, and woreda level) but essentially ignore intergovernmental
47
We have already acknowledged that Addis Ababa is an exception in this regard.
46
fiscal flows. This would be appropriate if all government levels were administrative tiers of the
same government entity (as is typically the case in centrally planned economies and in
deconcentrated systems). In a devolved system, however, intergovernmental fiscal transfers from
one government level to another should systematically be captured—both one the revenue side
as well as the expenditure side of each government level—in government financial reports.
The distribution of the regional-local transfers is determined by a set of formulas in each region
by the regional council (parliament). These formulas may be (and often are) modified from year
to year. To some extent, the larger regional states follow the same methodological approach as
the federal formula, which fills the gap between expenditure needs and revenue potential. The
state-level measures of sectoral expenditure needs vary from state to state, but tend to be similar
in nature to the expenditure norms used in the federal formula.
At the regional-local level, however, the nature of revenues is different—both conceptually, as

well as technically—from the federal-regional relationship. Whereas regional states collect their
own revenue sources (upon which federal revenue equalization is based), local revenue offices
typically collect state revenues as an agent of their region (rather than collecting state revenue as
on own revenue source). 48 , 49 Furthermore, from a technical viewpoint, in most states the
necessary data are unavailable to rely on a Representative Tax System to estimate local (woreda-
level) revenue potential. Instead, it appears that in practice, some regions use an incremental
approach to estimating local revenue potential (i.e., increase last year’s revenue collections by a
fixed percentage) or even use a fully discretionary (negotiated) approach to determining expected
revenue collections/potential.
A “cleaner” (but not necessarily preferable) transfer approach might have been used which
would simply require all local governments to deposit all state revenues collected by the local
revenue office in the regional treasury, and would subsequently distribute all local grant
resources exclusively in proportion to local needs. This approach would be consistent with the
notion that regional and local government levels are legally and politically separate government
levels. However, to prevent the off-setting funding flows associated with such an approach, an
alternative “gap-filling” approach is taken: the block grant allocation from the regional level to
the local level is determined as the difference between the norm-based expenditure need of the
local government, reduced by the (actual or expected level of) revenue collections.
In other words: local governments retain the collections from state revenues in their own
accounts, while the regional state reduces the block grant allocation to the locality by the
estimated annual revenue collections. In principle, if actual local revenue collections at the end
of the financial year are higher than was estimated at the beginning of the year, the local revenue
office would have to transfer the excess revenue collections to the regional level at the end of the
year. Of course, there is limited incentive to collect state revenues at the local level beyond the
48
In addition, as discussed further in Chapter 6, local revenue offices also collect municipal revenues. Since it is
recognized that municipal revenues are to be used for the funding of municipal functions, they seem to be excluded
from the computation of revenue collections (or revenue potential) in the regional-local allocation formula.
49
In practice, it is not relevant whether locally collected state revenues are legally assigned to the regional level or to
the local level: as long as the regional state government claws back any local revenue collections by subtracting
locally-collected state revenues from the local government’s estimated level expenditure needs in computing its
block grant, for all intents and purposes, these revenues belong to the state level.
47
planned revenue threshold if local governments do not get to retain some portion of these
resources. Fieldwork suggests that different regional states appear to have somewhat different
practices, with some regions allowing municipalities to retain the full balance of excess revenue
collections (but will then increase revenue expectations accordingly next year).
5.3 Overview of urban grants under ULGDP
The main funding source for urban infrastructure development is the World Bank-funded
ULGDP project. Despite its intent to be a more fully integrated government program, the first
phase of ULGDP retained many features of a more projectized approach. For instance, ULGDP
is disbursed in a projectized manner and relies on projectized parallel monitoring and reporting
systems. This approach was typical for the “first generation” of urban local government
development projects (compare for instance, with the Bangladesh Municipal Services Project).
Since the initial design of ULGDP, the World Bank has evolved considerably in its ambition to
work through country systems whenever possible (and the P4R modality should allow for this).
In addition, as the capacity of the (local) public sector in Ethiopia has increased over time, the
second phase of ULGDP can increasingly aim to provide World Bank support to a government-
owned federal earmarked intergovernmental fiscal transfer for urban development that flows
from MOFED through the regions to a select set of urban local governments in Ethiopia.
The discussion here will thus treat ULGDP as a (quasi-) intergovernmental grant and describe its
features as part of the country’s overall intergovernmental fiscal architecture of Ethiopia. 50 When
the projectized features deviate from good transfer practices, this will also be noted.
Objective. ULGDP was designed to support the government’s Urban Development Program and
Urban Good Governance Program. 51 The specific development objective of the project is to
support improved performance in the planning, delivery and sustained provision of priority
municipal services and infrastructure by urban local governments.
The project is expected to result in better performing cities that carry out:
• Effective and responsive planning to meet service delivery priorities identified by citizens
(allocative efficiency/participation objective);
• Improved financial management and mobilization of own resources and more effective
operations and maintenance of infrastructure assets (sustainability objective);
• Improved dissemination to the public of budgets/plans and performance measures
(accountability objective); and
• Effective implementation of Capital Investment Plans (service delivery improvement
objective).
50
The discussion here is based on a review of ULGDP project documents, including, for instance, the PAD, the
ULGDP Operational Manual (Revised November 2011) as well as quarterly and annual project reports. In addition,
this discussion is further informed by stakeholder feedback from official at all government levels.
51
ULGDP was designed and implemented by the Ministry of Works and Urban Development in 2006 as the urban
component of the Government’s Plan for Accelerated and Sustained Development to End Poverty 2005/06 to
2009/10 and now continued as part of the Growth and Transformation Plan 2010/11 to 2014/15.
48
Size of pool. The annual size of the ULGDP grant pool is determined in a discretionary or ad hoc
fashion.52 Since the Government of Ethiopia does not contribute financial resources itself, the
size of the transfer pool from year to year is fully determined by the amount of IDA funding
agreed between the government and the World Bank.
The size of the ULGDP grant pool has fluctuated over the project period, which has reduced the
stability with which urban planning has been possible. For instance, during the first years of
ULGDP, the allocations from ULGDP were not aligned with the Ethiopian financial year at all.
In EFY 2003, ULGDP switched to more regular annual disbursements, when the initial tranche
of funding was (largely) absorbed by ULGs. For EFY 2004 and 2005, the grant pool has been
steady at around US$ 12 per capita. The fluctuations in the size of the grant pool over time are
clearly visible in Table 5.5.53
Table 5.5: ULGDP Allocations and transfers by year, EFY 2001-2005 (US$)
EFY EFY EFY EFY EFY Cumulative
2001 2002 2003 2004 2005 2001-05
Total Allocation (US$ mn) 99.0 0.0 26.0 69.7 68.2 262.9
Total Transfer (US$ mn) 15.0 35.9 62.5 29.4 67.6 210.5
Total Alloc. Per Capita 16.8 0.0 4.4 11.8 11.6 44.7
(US$)
Disbursement perf. (%) 15.2 -- 240.5 42.2 99.1 80.1
Source: Computed by authors based on ULGDP data. Note: per capita allocations were determined
based on census population estimates for July 2013. See footnote 51 for 2001 allocations.
As noted in Table 4.9, ULGDP grants represent a considerable share of capital funding for cities
in the ULGDP universe, with ULGDP resources roughly making up from one-third to three-
quarters of municipal expenditures in the selected cities in our sample. This is unsurprising,
given that the average annual per capita ULGDP allocation is roughly US$ 9 per year, which is
close to the average level of per capita municipal revenue collections in ULGDP cities (as further
discussed in Chapter 6). These per capita figures are somewhat biased by the inclusion of Addis
Ababa in the analysis, which has a large population but only receives a relatively small (per
capita) allocation from ULGDP. If we exclude Addis Ababa from the analysis, the average
annual per capita ULGDP allocation increases to roughly US$ 16 per year. The relative
importance of ULGDP is likely to be even larger in the cities which will be added to the project
during its second phase, given their relatively smaller size and relatively smaller revenue
potential.
52
At the beginning of the project, the size of the allocation was related to the demand (i.e., the investment required
to fund the 3-year Capital Investment Plans developed by cities). Subsequently, absorption was taken into account.
No fixed funding rule (e.g., a fixed percent of the development budget, or a fixed amount per capita) is used to
determine the size of the grant pool from year to year.
53
Delays in the annual performance assessment in EFY 2004 resulted in late adoption and late implementation of
the ULGDP budgets that year. These delays are likely a major contributing to the weak budget performance in EFY
2004.
49
Basic flow of funds. The basic operation of the ULGDP grant program is described in the
Operational Manual, and several related project manuals. Unlike some experiences in other
countries with urban grant programs that are similar in design, Table 5.5 reveals that allocations
and transfers each year are not (at all) the same under ULGDP.
While subject to the allocation formula and performance criteria discussed below, ULGs are
awarded a certain “grant” allocation, this does not mean this is actually the amount that is
disbursed (for instance, in two semi-annual installments). Instead, after the initial disbursement,
subsequent disbursements are provided to replenish the ULG’s ULGDP account only after a
significant share of the original tranche has been spent. As such, ULGDP almost seems to works
on a projectized reimbursement model (within formula-based reimbursement ceiling for each
ULG) rather than providing a true sectoral block grant for urban development.
Stakeholder feedback from both the regional as well as the local level suggests that the basic
flow of funds mechanism for ULGDP is one of the most cumbersome features of the scheme,
especially as fund replenishment is not directly from the federal level to the local level, but
rather, all quarterly reports have to be aggregated at the regional level. Inconsistent budget
execution at the local level has resulted in major carry forwards from year to year. Rather than
enhancing predictability and transparency (both within years as well as between years), the high
transaction costs associated with the current funding flow mechanism seem to have resulted in
considerable delay, uncertainty, and less stability and transparency.
Allocation formula. The allocation formula for ULGDP is based exclusively on population, so
the ULGDP grant pool is distributed in proportion to the number of residents in each urban local
government. In principle, this means that all local governments (potentially) get the same amount
of grant funding per capita. This measure is conceptually consistent with the design of the
federal-region GPG, which relies on urban population of each region to measure urban
development needs.
The only exception to this funding formula under the first phase of ULGDP has been Addis
Ababa, which was allocated a fixed amount that was much lower in per capita terms. This was
justified conceptually as well as in practical terms: since Addis is expected to have a much
higher revenue potential than other cities in Ethiopia, it would be unfair to provide it with the
same per capita grant as all other urban areas. At the same time, the nominal distribution of
ULGDP resources would be completely biased in favor of Addis if it was funded on par with all
other ULGs, which would have reduced the project’s ability to incentivize good urban local
governance across the country.
Although the allocation formula used by ULGDP may seem simplistic, it has the benefit of being
objective and transparent. Since the number of urban residents is the primary driver of urban
needs, all in all, it should be seen as a reasonable indictor of municipal expenditure needs. As
noted in Section 4.3, when compared to measures of expenditure needs in social sectors, it is in
fact hard to develop more nuanced measures for the urban infrastructure requirements of
different urban areas. Pursuing a more complex allocation formula would most certainly reduce
the objectivity and transparency of the grant system.
50
One potential consideration should be noted as ULGDP is set to expand to a larger set of
municipalities during its second phase. It is envisioned that in ULGDP II, many of the new
municipalities will be smaller (less populous) in nature. In addition, these “new” municipalities
will presumably have a lower revenue capacity when compared to municipalities that have
already been exposed to revenue enhancement support and performance conditions for several
years. One possible way to accommodate this bifurcation in the “universe” of municipalities
under ULGDP II would be to provide new municipalities with a different (presumably lower)
grant amount than older municipalities. However, this would artificially and unnecessarily make
a distinction between different municipalities in a projectized fashion, and would also “punish”
new municipalities for not having been part of ULGDP I.
Furthermore, an alternate argument could be made that smaller municipalities, despite potential
concerns about their absorptive capacity, have a higher need for urban development grants, not
lower. This may be the case for two reasons. First, in general, smaller local governments are
likely to have lower revenue potential than larger ULGs and thus might be provided higher
grants to offset their lower revenue potential (see Figure 6.1 in Chapter 6). Second, less
populous municipalities have a larger overhead cost and are subject to economies of scale, and
thus, require higher per capita allocations in order to achieve the same level of urban services. As
such, it would be prudent to consider whether ULGDP II should rely on an allocation formula
that incorporates both a fixed lump sum amount in addition to a population-based allocation.
Annual Performance Assessment Process. Cities that participate in ULGDP are only provided
with formula-based funding envelopes for urban infrastructure investments if they meet certain
minimum access criteria. In addition, an annual performance assessment measures the
performance of each city in their use of ULGDP funds and in implementation of good
governance practice in accordance with seven performance measures. The performance
assessment also assesses the cities’ compliance with procurement, environment and social
safeguards, as described in the ULGDP operational manual.
The areas for which ULGs’ performance was assessed for the first three rounds of performance
assessment included the following areas:
1. Progress in the utilization of the Annual Capital Investment Budget;

2. Timely submission of all progress reports to regions, including Progress on Procurement
and Safeguards;
3. Timely submission of annual updates of three-year rolling plans/budgets and Annual
Action Plans, including evidence of effective public participation in defining the annual
updates;
4. Achievement of annual targets in revenue generation and implementation of accounting
system improvements;
5. Annually updated Infrastructure Asset Inventories, clearly showing both increases in
asset base and changes in quality of assets;
6. Timely submission of external audits and compliance with audit recommendations;
51
7. Evidence of effective public dissemination of information on budgets (including

investment plans), expenditures, physical progress of investments, and results of bid
evaluations and contract awards; and
8. Compliance with Procurement, Social and Environment Safeguards Requirements.
Following the agreement on Additional Financing between the World Bank and the Government
of Ethiopia, a slightly different set of performance measures was used for the fourth annual
performance assessment (EFY 2004) and going forward.54
To the maximum extent possible, the annual performance assessment process uses objective and
meaningful performance indicators in order to measure good local government practices. These
performance measures, however, were never defined in the Operational Manual and evolved
over time in a rather ad hoc manner in pursuit of greater refinement and consistency.
Nonetheless, the lack of consistency and transparency as regards the indicators used in the
annual performance assessment process is a weakness.
Performance incentives. Performance scores are awarded for different indicators, upon which a
composite score is determined. Based on the overall performance of a ULG in each of the
different categories, for the first three performance cycles, the following consequences were
attached to a ULG’s performance:
Performance rating Scores Reward or sanction

Suspension/reallocation LG has not achieved adequate progress in No capital grants
CIP implementation (below 80% CIP
utilization) and/or scored more than 65% only
in 2 or less performance measures.
Full allocation or basic LG has scored 65% on 3 or 4 performance Full allocation
allocation for next fiscal measures
year
Accelerated allocation / LG has scored 65% on 5 or 6 performance + 20 percent capital
performance bonus measures grant allocation
From EFY 2004 forward, the performance categories and incentives for ULGDP were again
modified slightly. The performance scoring is now based on the total performance score attained
by a ULG (from 0-100 points): if a ULG receives a performance score of under 50 points, the
city will not be awarded a capital grant under ULGDP; a score of 50-74 gets rewarded with a full
basic allocation; while scores above 75 points attract an additional 20 % reward.
Have performance incentives worked? Performance grants should be seen as an important tool
to incentivize good local government performance, as they draw the local government’s attention
to good practices in different areas of local government administration and provide some degree
of quantitative feedback in these areas. There is considerable anecdotal evidence to suggest that
performance grants are able to encourage local governments to adopt certain sound local
54
The new performance areas are: 1) progress in CIP, 2) financial management, 3) assets management and plan, 4)
revenue enhancement, 5) planning and budgeting, 6) transparency and accountability, 7) procurement, 8)
environment and safeguard management.
52
management practices. For instance, the widespread adoption of CIPs, REPs, asset management
plans, and other good local practices should be credited to the introduction of performance-based
grants under ULGDP.
However, it should be recognized that performance grants can become a rather blunt instrument,
as the performance measures used are almost certainly will be inadequate to fully capture the
underlying behaviors that the higher government levels seeks to incentivize. To the extent that
local governments (and their constituents, if there is a degree of horizontal accountability) do not
see a benefit in improving local government practices, there is a risk of the performance
assessment process becoming a “check-the-box” exercise and/or a cat-and-mouse game with the
performance assessors.55
While the performance element of the ULGDP grant has made important contributions to
incentivizing the adoption of good local governance practices in Ethiopia, the ULGDP
assessment process in Ethiopia faces the same types of obstacles that are faced in other countries
with such performance grant schemes. These challenges are discussed in more details in Part III
(chapters 8-12) of this report with preliminary recommendations for improvement of the system.
A more extensive discussion of the performance-based grant system, including a detailed
discussion of recommended indicators, will be included in the report for Component 2b.
Beyond merely improving the financing of urban development through ULGDP II, the second
phase of the project provides a significant opportunity to strengthen the system of
intergovernmental fiscal relations as a whole in Ethiopia:
• ULGDP II might instigate improvements in the Chart of Accounts and the budgetary
accounting and reporting processes, so that (both block grants as well as earmarked)
intergovernmental fiscal transfers are properly and consistently accounted for as revenue
and expenditure source at each government level as per IMF budget classification.
• ULGDP II might encourage more transparent regional-level budget practices, including
proper inclusion of ULGDP funding as a revenue (subsidy) to the regional state, while
also recording the grant allocations to City Administrations (including the federal/IDA
portion of ULGDP funding, as well as state-level co-funding) as intergovernmental
transfers in the regional state budget.
• More detailed recommendations regarding the design of the second phase of ULGDP are
made in Chapter 13 of this report and in the subsequent Component 2b report.
55
In this regard, we should note that the independence of the performance process is critical. When the performance
assessment is conducted by the hierarchical reporting structure itself, this removes the checks and balances that are
required for an objective assessment process.
53
6. An Analysis of Municipal Own Source Revenues

Urban local governments in Ethiopia collect two sorts of revenue: first, state revenue, and
second, municipal revenue. In principle, state revenue is collected by local revenue offices,
essentially as agents on behalf of the state-level government. 56 In practice, however, local
governments are allowed to retain state revenues in their own accounts in order to prevent cash
from flowing in two directions between regional and local governments. Instead, the regional
level generally reduces the block grant to the local level by the amount of (estimated or actual)
state revenue collections.
Municipal revenues, on the other hand, are fully considered to be the local government’s own
revenue sources. As such, these revenues are the primary financial resources available to fund
urban services and urban infrastructure development.
Over the course of the past decade, urban (capacity) development projects (including ULGDP)
have been supporting the improved collection of municipal revenues. Recognizing the
importance of municipal revenues, MUDC (in collaboration with GIZ/UGDP) recently prepared
a draft Municipal Revenue Enhancement Strategy (May 2013). In addition, MUDC is pursuing
support for the modernization of property taxation in Ethiopia.
This chapter seeks to analyze and discuss the level and composition of municipal own source
revenues. Section 4.2 of the current report already provides some quantitative background with
respect to municipal revenues, revealing that the national average level of municipal revenue
collections for EFY 2004 was ETB 208 per urban resident (i.e., around US$11 per urban
resident). Excluding Addis, municipal revenues per urban resident equal ETB 163 per person.
The analysis in Chapter 4 further found considerable variation in municipal revenues across
regions (based on aggregate municipal revenue collections at the regional level), with some
regional governments collecting virtually no municipal revenues whatsoever, whereas other
regions have robust municipal revenue collections. Municipal revenues exceed ETB 200 per
urban resident in Dire Dawa, Oromia, Tigray and Addis Ababa, whereas municipal revenues fall
between ETB 50-100 in SNNPR, Amhara and Somali Region; municipal revenues fall below
ETB 50 per urban resident in the remaining regions.
6.1 Overview of municipal revenue assignment
Given Ethiopia’s federal structure, municipal revenue assignments vary from state to state, as
local revenue is not a federal function, but rather, a state function. Nonetheless, the federal
government does involve itself—generally indirectly—in municipal revenue issues. For instance,
land lease revenues are supposed to be administered and collected in accordance with a federal
proclamation. In addition, the federal government (MOFED, acting in concert with MUDC)
56
As already noted in Chapter 5 as part of the discussion of the intergovernmental grant system, in practice, it is not
relevant whether locally collected state revenues are legally assigned to the regional level or to the local level: as
long as the regional state government subtracting locally-collected state revenues from the local government’s
estimated level expenditure needs when computing its block grant, for all intents and purposes, these revenues
belong to the state level.
54
prepares prototype proclamations or regulations on municipal finance issues for consideration by

regional authorities. In Ethiopia’s “collaborative federalism” model, this appears to be an
effective means for achieving vertical coordination on sub-national fiscal issues.
Perhaps the closest attempt to standardize local revenue practices was the prototype “Regulations
To Provide For City Government Finance Administration” prepared by MOFED and transmitted
to the regions in 2005 (and gradually adopted by different regional states from 2006-10). This
prototype regulation suggests that municipal revenues might include the following fifteen
municipal revenue sources:
(a) taxes on real property57

(b) residence tax (a “head tax” or “poll tax”)
(c) taxes on fuels and utilities
(d) taxes on telephone, fax, telegram and other electronic services
(e) environment protection taxes
(f) tax on business
(g) sur-tax on tobacco, alcohol, soft drinks
(h) cattle tax
(i) boat tax
(j) bicycle tax
(k) cart tax
(l) emblems & sign board tax
(m) market stall
(n) fees and charges
(o) such other taxes/fees/charges as may be levied from time to time.
Only very limited guidance is provided in the regulatory framework as to the exact tax base and
administration of the local taxes. This means that, notwithstanding the issuance of prototype
municipal financial management regulations, local revenue systems may vary considerably in
their design and implementation. For instance, there is no standardized guidance on the tax base
of a municipal “tax on business”.
Fees and charges (noted under “n” above) are suggested for (1) vehicles; (2) certification of
qualification; (3) information dissemination; (4) entertainment; (5) justices; (6) education; (7)
registration; (8) quarantine and control; (9) health and sanitation; (10) cemetery and ambulance
services; and (11) land use fee. Again, there is no detailed guidance on the specific nature or
revenue base for these fees and charges.
The prototype regulations clearly recognize the conceptual difference between local taxes and
non-tax revenues: whereas local taxes are mandatory payments that generate general (untied)
revenue for the municipality for which no direct quid pro quo is provided, fees and charges are
collected to fund specific municipal services on a fee-for-service (full cost recovery) basis.58 This
57
The prototype regulations define this as a tax or rating on the valuation of land, the rating of the value of
buildings, or some combination of both.
58
It should be noted that full cost recovery would include not only recovery of the operation and maintenance costs,
but also the capital depreciation (i.e., saving for future rehabilitation and/or replacement of capital investment).
55
interpretation of the role of fees and charges is explicitly recognized in the prototype municipal
financial management regulations.
It is unclear to what degree these prototype regulations have been adopted and remain valid. For
instance, in which regional states are urban local governments legally authorized to collect a
property tax? In addition, to the extent that regions have adopted municipal financial regulations
based on the prototype, subsequent legislation is likely to have re-centralized some of the
(revenue) autonomy of City Administrations. Regional variations in local revenue legislation are
discussed in greater detail in Chapter 3.
6.2 Regional differences in the local revenue context
Observers have noted considerable differences in the local revenue context from region to region
in Ethiopia. Indeed, these differences seem to be confirmed by earlier findings of considerable
variations in municipal revenue collections across regional states. The TORs that guide the
current study imply that an important determinant for such differences may be differences in the
degree of municipal revenue autonomy provided by regional proclamations.
Differences in regional proclamations with respect to municipal revenues and rate-setting ability
were discussed at length in Chapter 3 (Section 3.2). The main conclusion from the review was
that cities in principle were granted significant autonomy in the basic legislation (city
proclamations, etc.), but that regions in practice provide no or very narrow set of tax policy
options for the cities through the various tariff declarations. There is a great deal of similarity of
the various tariff declarations– as the regions have learned from each other – but the federal state
has never issued a general “model” that could provide “best practices”.
Differences in other regional factors related to revenue collection. In addition to potential

differences in municipal revenue effort, differences in municipal revenue collections across
regions are likely to stem from differences in geographic and demographic conditions, and
differences in the economic structure of different regional states in Ethiopia. As noted below (in
Section 6.3), the average size of cities in each region is an important determinant of municipal
revenue collections.
Although different regions defy generic classification, Addis Ababa, Dire Dawa, and Harar have
a unique dynamic as city-states, where the urban area coincides (nearly or completely) with the
state-level jurisdiction; this brings with it a stronger economic base, but also unique trade-offs in
terms of collecting state revenues versus municipal revenues. Likewise, because there is no
separate regional versus local level, the institutional and political dynamics which might be used
to encourage greater municipal revenue collection in other states (for instance, greater regional
co-funding of ULGDP grants) are not an option in these regions.
However, depending on the existence of financial management controls and proper oversight, fiduciary risk may be
involved when saving large sums of money in dedicated capital replacement accounts. This is true whether these
funds are saved by the municipality directly or by municipal enterprises (e.g., a municipally-owned abattoir
enterprise). A discussion should be had how best to fund long-term capital replacement.
56
At the other end of the spectrum, Somali and Afar are land-rich regions with a relatively smaller
economic base (and as a result, lower public revenue collections overall).
Although there are considerable economic, demographic and geographic differences between the
bigger, more advanced regions (Oromia, SNNPR, Amhara and Tigray), these regions have been
on the receiving end of the bulk of institutional capacity building support—both at the regional
as well as the local level. This leaves Gambella and Benishangul-Gumuz, which represent the
smallest (non-city) regions in the country (in terms of population), therefore representing only a
tiny share of the country’s total urban population.
6.3 An analysis of municipal own source revenue collections
In order to get a better view of municipal revenues in Ethiopia (beyond the aggregate view which
is provided by national-level IBEX figures), fieldwork was conducted in 12 urban local
governments in Ethiopia (as described in the introductory chapter of this document). For each
municipality, data were collected regarding municipal revenues and interviews were conducted
with key municipal staff, including the head of the local revenue office. In addition, other city-
level revenue figures were extracted whenever available (e.g., from ULGDP Annual
Performance Assessment documents). Based on these disaggregated data sources, this section
provides an analysis of municipal own source revenue collections. Whereas the current analysis
is largely quantitative in nature, issues related to municipal revenue administration and collection
processes are discussed in greater detail in Chapter 12.
The impact of economic base and population size on municipal revenues. It is likely that a city’s
population size is highly correlated with the size of the city (formal) economic base, which will
have an important impact on own source revenue potential and own source revenue collections.
Table 6.1 and Figure 6.1 highlight this relationship for the twelve selected urban local
governments included in our sample, along with the ULGDP cities for which data were
available. The figure shows a relatively strong, positive relationship between city size and per
capita municipal revenue collections: for every ten thousand increase in population size, a city
might be able to roughly collect around two Birr per person more.59 All else equal, this means
that smaller cities (and many of the ‘new’ municipalities in ULGDP II are smaller in terms of
population) are likely to have a considerably lower degree of revenue potential. However, it
should be noted that the sample which is being analyzed is not a representative sample of ULGs
in Ethiopia, given the considerable oversampling of ULGDP cities versus non-ULGDP cities.
Table 6.1: Municipal Revenue Collections for Selected Urban Local Governments (EFY 2004)
59
Note that this is simply a basic bivariate correlation. Other factors may be (and in fact are) at play in determining
the revenue potential and revenue collections in municipalities.
57
Municipal
Revenues Per Capita Per Capita
Population (ETB mn) Revenues (ETB) Revenues (US$)
Hawassa (*) 225,686 59.0 261.3 14.0
Hosanna 100,528 11.1 110.0 5.9
Ziway Batu 56,104 7.7 138.0 7.4
Adama (*) 282,974 89.7 316.9 17.0
Mekelle (*) 286,624 114.6 399.9 21.5
Adiwa 53,763 10.8 201.5 10.8
Bahir Dar (*) 276,218 67.1 243.0 13.1
Debre Berhan 83,461 6.7 80.8 4.3
Asosa 40,686 1.1 27.1 1.5
Gambella 64,602 5.2 81.2 4.4
Jijiga 152,674 19.3 126.4 6.8
Harar (*) 112,781 18.6 165.2 8.9
Gonder (*) 264,964 35.8 135.0 7.3
Kombolcha 75,078 8.7 116.1 6.2
Dessie (*) 153,691 27.9 181.4 9.8
Axum (*) 59,269 18.5 312.3 16.8
Adigrat (*) 76,447 14.2 185.7 10.0
Shire Enidasilase (*) 62,769 20.1 320.6 17.2
Sodo (*) 109,225 13.2 120.8 6.5
Dila (*) 84,952 12.4 145.4 7.8
Arba Minch (*) 107,542 7.9 73.8 4.0
Bishoftu (*) 128.408 59.0 459.5 24.7
Jimma (*) 155,434 65.3 420.1 22.6
Shashemene (*) 129,084 31.1 241.1 13.0
Dire Dawa (*) 269,134 57.4 213.4 11.5
Addis Ababa (*) 3,103,673 1,242.9 400.5 21.5
Total/Average 6,515,771 2,025.5 210.6 11.3
Source: Computed by authors based on data collected from municipalities. Refer to Table 2.1 for population
data. Note: ULGDP cities are indicated with (*).
What are the differences in (per capita) revenue collections between ULGDP cities and non-
ULGDP cities. Under the assumption that all else is held equal, Figure 6.1 suggests that smaller
cities (in terms of population) have a considerably lower degree of revenue potential. However,
ULGDP cities are clearly not equal to all other cities: the cities selected for inclusion into the
initial phase of ULGDP were not only larger (more populous), but also administrative more
capable than other municipalities in the country. Furthermore, they have benefitted
disproportionately from capacity building support and performance incentives over the past five
years to increase their revenue collections.
As such, the pattern shown in Figure 6.1 may be the result of a non-random sample that reflects
selection bias: it is unsurprising to find that ULGDP cities (with the exception of Harar) have
58
considerably higher revenue collections than non-ULGDP cities.60 Given the small size of the
sample, it is impossible to determine to what extent higher revenue collections in the ULGDP
cities in the sample reflect greater revenue effort versus a larger (initial) economic base vis-à-vis
the non-ULGDP cities.61
Figure 6.1: Municipal revenues per capita and population size (EFY 2004)
Note: Existing ULGDP cities are identified with a blue diamond; non-ULGDP cities are indicated with
red triangles.
What are the most important municipal revenue sources? What patterns exist in terms of
municipal revenue collections? First and foremost, from the list of possible municipal revenue
sources noted above (in the regulatory framework), which are the most important revenue
sources? Tables 6.2 and 6.3 sheds some light on this situation for the twelve selected urban local
governments included in the sample: Table 6.2 presents the composition of municipal revenues
in millions of Birr; Table 6.3 shows the same municipal revenues composition, but expressed as
a percentage of total municipal revenues for each municipality.
For the purpose of the current analysis, five broad categories of municipal revenues are
considered:
60
For 2004, the (unweighted) average per capita municipal revenue collections for the ULGDP cities in our sample
was ETB 248 versus ETB 109 in non-ULGDP cities. For 2005, the average per capita revenue collections for these
two sub-samples were ETB 271 and ETB 169, respectively.
61
Further below, the analysis will consider revenue trends over time for ULGDP cities. Comparative data over time
are not available for non-ULGDP cities.
59
• 1701-1719: Municipal taxes (1701-1719). According to the Chart of Accounts (CoA),

this category includes urban building taxes and taxes on real property; entertainment
taxes, and other taxes.
• Municipal rents (1721-1729). Includes house and land rents, rent of shops and market
stalls, and other municipal rental income.
• Urban land lease income (1731).
• Licenses, fees and charges for city services (1740-1782). This includes a wide range of
fees, from business license fees and registration fees (for contracts, births and marriages,
and so on), to sanitation service fees, water service fees, sewer service fees, slaughter
fees, to land permit registration fees.
• Other (capital) revenues and investment income (1732-1739; 1789-1799). This includes
the sale of municipal property and other capital receipts.
Before proceeding, it should be noted that the CoA classification of municipal taxes is not
consistent with the prototype municipal financial administration regulations cited above, and thus
it is hard to effectively categorize revenue sources into meaningful categories. For instance, the
residence tax (head tax) included in the regulations is not explicitly recognized in the category of
municipal taxes. An added concern is that these revenue categories may not be consistently
adhered to by municipal authorities in their revenue collection reporting.
Table 6.2: Municipal revenue by major category (in ETB million), EFY 2004
revenue
land lease
charges
taxes
fees,
rents
Total
Urban
Licenses,
Other
Municipal
Municipal
Hawassa 0.7 8.7 23.0 24.0 2.6 59.0

Hosanna 1.9 2.3 3.2 3.6 0.0 11.1
Ziway Batu 2.6 3.6 - 0.9 0.6 7.7
Adama 0.0 25.6 18.4 39.5 6.2 89.7
Mekelle 3.1 15.3 59.3 36.3 0.7 114.6
Adiwa 2.2 1.6 4.8 1.9 0.3 10.8
Bahir Dar 1.0 4.6 27.0 29.9 4.6 67.1
Debre Berhan 4.6 0.5 - 0.3 1.4 6.7
Asosa 0.2 0.8 - 0.1 - 1.1
Gambella 0.4 1.1 1.1 2.6 0.0 5.2
Jijiga 0.5 0.2 - 18.1 0.4 19.3
Harar - 5.3 2.4 10.2 0.7 18.6
Total 17.2 69.6 139.1 167.3 17.7 411.1
Source: Computed by authors based on data collected from municipalities.
60
Table 6.3: Municipal revenue by major category (as pct. of total municipal revenue), EFY 2004
revenue
land lease
charges
taxes
fees,
rents
Total
Urban
Licenses,
Other
Municipal
Municipal
Hawassa 1.2 14.7 39.0 40.6 4.5 100.0
Hosanna 16.9 21.0 28.9 32.8 0.4 100.0
Ziway Batu 33.5 46.7 0.0 12.1 7.7 100.0
Adama 0.0 28.5 20.5 44.1 6.9 100.0
Mekelle 2.7 13.3 51.7 31.6 0.6 100.0
Adiwa 19.9 14.4 44.7 17.7 3.2 100.0
Bahir Dar 1.6 6.8 40.2 44.5 6.9 100.0
Debre Berhan 67.7 7.7 0.0 3.9 20.8 100.0
Asosa 15.0 75.7 0.0 9.4 0.0 100.0
Gambella 8.5 21.1 20.3 49.5 0.6 100.0
Jijiga 2.8 1.1 0.0 93.7 2.3 100.0
Harar 0.0 28.6 12.7 54.8 3.8 100.0
Average 14.1 23.3 21.5 36.2 4.8 100.0
Source: Computed by authors based on data collected from municipalities.
A number of conclusions with regard to municipal revenue collections can be drawn based on
overview of the patterns revealed by the preceding table:
• There is considerable variability in municipal revenue patterns. Some ULGs rely heavily
on taxes, whereas others rely heavily on rental revenues, whereas yet other ULGs rely
heavily on land lease or license incomes. This variability is caused—at least in part—by
significant differences in local revenue structures. One pattern that appears to be clear is that
land lease revenues are very important for larger urban local governments but less so
for smaller municipalities. The observed variability may also be partially due to inconsistent
assignment of revenue sources to different categories/codes.
• Municipal taxes appear under-utilized as a municipal funding source. The municipal
finance regulations note a series of potentially important municipal revenue sources,
including a personal (residence) tax, a property (land/building) tax, and a variety of taxes on
business and trade. However, this revenue category contributes more than 25% of municipal
revenues in only two out of twelve urban local governments reviewed.
• Licenses, service fees and charges are an important revenue driver. For seven out of twelve
ULGs, service fees and charges represent the single-largest municipal revenue category. It is
good that collection of service charges is robust, as charges and fees ensure a strong link
between municipal service provision and revenue collections. However, these revenues are in
principle tied to funding specific services, and therefore, do not provide the municipality with
general revenues to do important things such as the construction of and maintenance of roads
61
and drains, and provide other municipal amenities for which user charges are not a practical
funding solution.62
Revenue collection trends over time. Although IBEX does not consistently track municipal
revenue collections prior to EFY 2004, the following tables and figures nonetheless give an
useful indication of the relative performance of municipal revenue collections over time in
Ethiopia.
First, Table 6.4 presents municipal revenue collections for ULGDP cities from 2001 through
2005 in millions of Birr. In addition, the table shows the average year-on-year growth rate for
municipal revenues and concludes that the aggregate amount of municipal revenue collections
across all ULGDP cities has increased an average of 16% each year for the period from 2001-
2005. We also note that increases in municipal revenue collections are not linear; while some
years witness relatively large increases in aggregate municipal revenue collections versus the
previous year (e.g., EFY 2004), other years see a small much smaller increase, while from EFY
2005 even shows a decline in municipal revenue collections.
Table 6.4: Own source municipal revenue collections for ULGDP cities (in nominal ETB millions), EFY
2001 – 2005
EFY EFY EFY EFY EFY Average
2001 2002 2003 2004 2005 Year-on-Year
Addis Ababa 600 637 797 1,242 935 116%
Dire Dawa 23 36 48 57 72 134%
Adama 86 43 90 88 123 124%
Bishoftu 38 29 76 59 77 137%
Shashemene 30 32 29 31 31 101%
Jimma 36 31 33 31 25 92%
Hawassa 34 31 47 53 62 118%
Dilla 6 7 11 12 13 123%
Wolaita Sodo 7 17 21 13 19 144%
Arbaminch 6 5 6 8 13 125%
Mekelle 67 67 87 115 147 122%
Adigrat 7 7 10 14 19 130%
Shire 8 8 13 20 22 132%
Axum 7 6 16 19 16 139%
Bahir Dar 43 65 44 68 118 137%
Gondar 24 28 32 36 40 114%
Dessie 15 23 24 32 35 125%
Kombolcha 4 7 10 9 13 138%
Harar 12 14 16 18 24 119%
Total 1,053 1,093 1,410 1,925 1,804 116%
62
It is possible that despite their categorization, some of these service fees are actually collected in the shape of
taxes. For instance, a sanitation charge might be collected as a surcharge to a (simple) property tax. In fact, some of
these charges may even be collected to fund non-excludable public goods, such as road drains.
62
Source: Computed by authors based on ULGDP project data.
As noted earlier in section 4.2 (and discussed in greater detail below), revenue collection levels
are not merely determined by local revenue administration effort; instead, many factors combine
to determine the level of revenue collections. As such, without further analysis, it is impossible
to say whether an average year-on-year increase in municipal revenue collections of 16% (or
10% or 20%) is an adequate amount of revenue growth.
In order to explore the adequacy of municipal revenue growth, Table 6.5 presents the same data
as in Table 6.4, but adjusts the collections for changes in price levels over time, expressing all
revenues in inflation-adjusted Birr (2001). After making this adjustment, the analysis shows that
in real terms, own source revenue growth over time is a lot less robust. For six municipalities in
our sample, rather than reflecting an increase in revenue collections, municipalities actually are
facing an overall decline in the value of revenue collections when municipal revenues are
expressed in real terms. In fact, although the majority of ULGs in our sample show (small)
increases in real revenue collections, the average level of inflation-adjusted own source revenues
appears to have declined over time.
Table 6.5: Own source municipal revenue collections for ULGDP cities (in inflation adjusted ETB
millions, 2001=100), EFY 2001 – 2005
Average
2001 2002 2003 2004 2005 Year-on-Year
Addis Ababa 600 594.6 649.3 744.5 448.1 96%
Dire Dawa 23 33.6 39.1 34.2 34.5 113%
Adama 86 40.1 73.3 52.7 58.9 103%
Bishoftu 38 27.1 61.9 35.4 36.9 115%
Shashemene 30 29.9 23.6 18.6 14.9 84%
Jimma 36 28.9 26.9 18.6 12.0 77%
Hawassa 34 28.9 38.3 31.8 29.7 98%
Dilla 6 6.5 9.0 7.2 6.2 103%
Wolaita Sodo 7 15.9 17.1 7.8 9.1 124%
Arba Minch 6 4.7 4.9 4.8 6.2 103%
Mekelle 67 62.5 70.9 68.9 70.4 102%
Adigrat 7 6.5 8.1 8.4 9.1 107%
Shire 8 7.5 10.6 12.0 10.5 109%
Axum 7 5.6 13.0 11.4 7.7 117%
Bahir Dar 43 60.7 35.8 40.8 56.5 113%
Gondar 24 26.1 26.1 21.6 19.2 95%
Dessie 15 21.5 19.6 19.2 16.8 105%
Kombolcha 4 6.5 8.1 5.4 6.2 117%
Harar 12 13.1 13.0 10.8 11.5 100%
Total 1,053 1,020.2 1,148.8 1,153.9 864.5 96%
Source: Computed by authors based on ULGDP project data.
63
Conceptually, there are a number of potential explanations for declining (real) revenues over
time. The most obvious explanation for reductions in revenue collections over time is simply a
contraction of the urban tax base, although this does not appear to be a terribly likely explanation
in the case of Ethiopia (since Ethiopia’s economy has been growing in both nominal and real
terms). Another possible explanation might be the nature of the taxes assigned to the municipal
level: it is possible that municipal revenue sources are not buoyant, meaning that they do not
grow responsively in proportion to the size of the underlying economy.
Table 6.6 provides a test of the comparative buoyancy of municipal revenues and state revenue
collections from EFY 2004 to 2005, based not only on the data for all ULGDP cities, but also on
the data collected from the twelve sample municipalities visited during field work. Since the
underlying economic base for municipal revenues and state revenues is essentially the same, any
differences in the revenue growth rated should predominantly be due to differences in the
buoyancy of different revenue sources.63
Similar to the previous table, revenue collections in Table 6.6 have been inflation-adjusted and
are expressed in real terms (normalized in 2004 Birr). Although the tables shows that there is
virtually no change in state revenues from EFY 2004 to 2005, municipal revenue collections in
the same set of municipalities reportedly decreased by 25%. This indeed confirms the hypothesis
that there is something about the municipal revenue structure that makes municipal revenues
non-buoyant. Indeed, the fact that the majority of number of municipal fees and charges are set
in nominal terms—rather than determined as a share of their tax base—makes municipal
revenues extremely prone to low revenue growth. Even merely in order to offset inflation,
municipal leaders would have to make the unpopular decision to increase municipal fees and
charges each year. In contrast, since most the tax rates for state revenue sources are often
expressed in percentage terms, this means that the increase in revenues from these sources will
be automatically (more or less) proportionate to the changes in the size in the underlying
economy. This conclusion seems to be borne out by the patterns in Table 6.6.
63
This is not a perfect test, as local tax administration officials may also exert different levels of tax effort in
collecting municipal versus state revenues over time.
64
Table 6.6 Comparison of revenue collections for selected ULGs, EFY 2004-2005 (in inflation
adjusted ETB million, 2004=100)
State Revenues Municipal Revenues State % Municipal
Change % Change
2004 2005 2004 2005
Hawassa 196.81 478.63 58.98 47.28 94.4% -35.9%
Hosanna 26.01 43.55 11.06 15.68 33.9% 13.3%
Ziway Batu 25.52 32.38 7.74 8.63 1.4% -11.0%
Adama 151.63 177.59 89.69 120.84 -6.4% 7.7%
Mekelle 323.93 497.86 114.61 120.79 22.9% -15.7%
Adiwa 25.14 35.59 10.83 15.82 13.2% 16.8%
Bahir Dar 184.21 311.16 67.13 121.11 35.0% 44.2%
Debre Berhan 30.71 44.47 6.74 10.53 15.8% 24.8%
Asosa 14.32 16.38 1.10 3.01 -8.6% 118.3%
Gambella 8.30 7.20 5.25 5.52 -30.6% -15.9%
Jijiga 3.78 6.55 19.30 44.73 38.3% 85.3%
Harar 86.19 131.90 18.63 24.51 22.3% 5.2%
Addis Ababa 8,504.00 7,992.57 1,242.00 747.53 -6.0% -39.8%
Dire Dawa 162.00 172.69 57.00 57.56 6.6% 1.0%
Bishoftu 55.00 61.56 59.00 61.56 11.9% 4.3%
Shashemene 51.00 54.37 31.00 24.78 6.6% -20.1%
Jimma 46.00 44.77 31.00 19.99 -2.7% -35.5%
Dilla 27.00 31.18 12.00 10.39 15.5% -13.4%
Wolaita Sodo 30.00 47.17 13.00 15.19 57.2% 16.8%
Arba Minch 28.00 38.38 8.00 10.39 37.1% 29.9%
Adigrat 39.00 49.57 14.00 15.19 27.1% 8.5%
Shire 34.00 38.38 20.00 17.59 12.9% -12.1%
Axum 26.00 35.18 19.00 12.79 35.3% -32.7%
Gondar 102.00 130.32 36.00 31.98 27.8% -11.2%
Dessie 88.00 105.53 32.00 27.98 19.9% -12.6%
Kombolcha 40.00 48.77 9.00 10.39 21.9% 15.5%
Total/Average 10,308.56 10,276.13 1,994.05 1,493.81 -0.3% -25.1%
Of course, since Table 6.6 only reflects data for a single year and for a relatively small sample of
municipalities, this pattern is not necessarily reflective of overall municipal revenues over time.
Therefore, the final analysis of municipal revenue buoyancy (in Figure 6.2) uses aggregated
IBEX data for municipal revenues over a longer period, notably from EFY 1999 through EFY
2004. As already noted earlier, the problem with IBEX-based municipal revenue data for this
time span is that municipal revenues were not consistently reported through IBEX throughout
this period. For instance, municipal revenue collection figures for Oromia Regional State are
missing for all years, and the number of regional states for which municipal revenues is reported
increases over this period from three states to eight states. As a result, the increase in reported
65
municipal revenues is considerably biased upward over this period. Despite this upward bias,
Figure 6.2 suggests that municipal revenues have fared relatively poorly over the period under
investigation.
Figure 6.2: Municipal revenue collections and other economic and fiscal indicators, EFY
1999-2004
Source: MOFED (IBEX) and IMF.
Figure 6.2 suggests that reported municipal revenues have increased threefold in nominal terms
over the past five years in Ethiopia. However, despite the upward bias in municipal revenue
collections noted above, when municipal revenue collections over time are compared with other
economic and fiscal indicators (all presented as indices, with EFY 1999=100), a failure of
municipal revenues to keep up with overall economic growth over time can be noted. While
aggregate municipal revenues have managed to increase somewhat faster than inflation over the
past five years (perhaps as an artifact of the upward bias), state revenue collections have still
performed considerably better over the same period. Given the high levels of economic growth
and inflation suggested by Table 6.2, it is likely that it is the inherently non-buoyant structure of
municipal revenues—and not necessarily the lack of municipal revenue effort—that is the
driving force behind lagging revenue collections. Unless this point is effectively recognized—
both at a policy level, as well as at the local level and as part of the municipal Revenue
Enhancement Plans—it is not likely that merely incentivizing municipal revenue collections
through a performance-based grant is going to solve the erosion of municipal revenues.
Incentives for improved municipal revenue management have to address wider (regional and
possibly federal level) policy direction, rate and tariff settings, revenue planning and monitoring.
In addition wider municipal tax reforms – such as introduction of property tax has to be
supported and effectively implemented.
66
6.4 Conceptual discussion: local revenue potential
Although some cities collect more municipal revenues than others, this does not necessarily
reflect on the collection effort of each city. As already noted in Chapter 4, there are a number of
factors that influence municipal revenue collection, including the city’s tax base, local tax rates
(which, to a greater or lesser degree, could be determined by regional authorities), local
collection and enforcement efforts, and taxpayers’ compliance. Unlike existing Revenue
Enhancement Plans reviewed by the study team—which generally reflect cut-and-paste
discussions of municipal revenue issues, minimally compliant with ULGDP requirements—a
sound revenue enhancement plan should be informed by an actual analysis of revenue collection
patterns.
The most “sophisticated” analysis currently used in local revenue enhancement plans in Ethiopia
is to express revenue effort as the ratio of actual collections expressed as a share of budgeted or
planned revenue collections. On this score, most municipalities (perhaps unsurprisingly) do a
reasonably good job collecting revenues (Table 6.7).
Table 6.7: Municipal Revenue Performance in Selected ULGs, EFY 2004, 2005 (ETB mn)
Budgeted Actual Perf. Budgeted Actual Perf.
2004 2004 (%) 2005 2005 (%)
Hawassa 69.2 59.0 85.3 81.1 47.3 58.3
Hosanna 14.2 11.1 78.0 14.9 15.7 105.4
Ziway Batu 8.4 7.7 92.7 9.1 8.6 95.2
Adama 108.0 89.7 83.0 124.0 120.8 97.5
Mekelle 134.5 114.6 85.2 180.0 120.8 67.1
Adiwa 10.4 10.8 104.2 16.4 15.8 96.3
Bahir Dar 94.3 67.1 71.2 123.9 121.1 97.7
Debre Berhan 16.3 6.7 41.5 15.0 10.5 70.3
Asosa 0.9 1.1 121.7 1.9 3.0 158.2
Gambella 7.4 5.2 70.9 7.8 5.5 70.4
Jijiga 25.9 19.3 74.7 35.2 44.7 127.2
Harar 35.3 18.6 52.8 50.5 24.5 48.6
Total/Average 524.5 411.1 80.1 659.7 538.4 91.0
It should be noted that the ULGDP annual performance assessment tracks and rewards revenue
performance based on municipal revenue performance. This might (partially) explain the timid
municipal revenue growth over the past five years: municipalities can simply lower their
municipal revenue target in order to achieve “good” revenue performance.
Although the sample is too small to make meaningful conclusions, it is interesting to note that
non-ULGDP cities seem to have a somehow higher level of municipal revenue performance
(when comparing planned versus actual revenue collections) in both EFY 2004 and EFY 2005
67
(83.4% and 103.3%, respectively) when compared to ULGDP cities (75.5% and 73.8% for 2004
and 2005, respectively).64
Local revenue officials in ULGDP cities interviewed as part of the fieldwork were generally
unable to articulate how the revenue estimate was prepared. Certainly, these estimates are only
weakly based—if at all—on the anticipated number of taxpayers for next year for a particular
revenue source. In most cases, it appeared that the revenue estimate was determined as a
negotiation between local officials on one hand (seeking to limit revenue increases) and the
regional revenue authority on the other hand (putting upward pressure on local revenue
collection plans). This is not a good municipal revenue practice.
In contrast to current practices, a meaningful revenue enhancement plan would be informed by

an analysis of the different stages of revenue administration, and the determination of what stage
of the revenue collection process most revenues get “lost”. A detailed estimation of municipal
tax effort and the degree of unutilized municipal revenue potential can be computed based on the
identity that posits that the level of revenue collections for any particular revenue instrument is a
function of tax policy decisions (including the tax base and tax rate) as well as a number of tax
administration factors. 65 For instance, revenue collections for a specific local revenue source
could be defined as:
R = (TB * TR) * (CVR * AR * CLR) (1)
Where the variables have the following meaning:
R Actual amount of revenue collected

TB The value of the statutory tax base, as defined according to government policy in
terms of what is and is not taxed.
TR The statutory (legal) tax rate.
CVR The Coverage Ratio (CVR) is defined as the share of the statutory tax base that is
covered by the municipal tax administration. For instance, in the case of a property
tax, the Coverage Ratio would be calculated as the number of taxable properties,
which are captured in the cadastre, divided by the total number of taxable property in a
jurisdiction. The Coverage Ratio measures the accuracy and completeness of the roll
of tax-payers or taxable properties.
AR The Assessment Ratio (AR) is defined as the share of the statutory tax rate that is
assessed by the municipality. A municipality could either under-assess a tax by simply
lowering the tax rate that it applies (below the statutory rate), or alternatively, by
64
It is possible that this pattern results from the incentives faced by ULGDP cities, who may simply aim to meet the
ULGDP threshold (rather than maximize revenue effort). If this is the case, the performance assessment may
unintentionally lower revenue effort. An opposite argument, however, could be made: it is equally possible that
revenue expectations for non-ULGDP cities are lower (in the absence of revenue-enhancement support). Because
the bar may be set lower for non-ULGDP cities, it may be subsequently easier for them to meet regionally-
negotiated revenue expectations.
65
Roy Kelly. 2000. Property Taxation in East Africa: The Tale of Three Reforms, Lincoln Institute of Land Policy,
Working Paper.
68
under-valuing the value of the tax base. The Assessment Ratio measures the accuracy
with which the value of the tax base is assessed and the accuracy with which the tax
rate is imposed.
CLR The Collection Ratio (CLR) is defined as the tax revenue collected over the total tax
liability billed for that year, measuring the collection efficiency. The Collection Ratio
reflects the vigor with which the municipality administers and enforces municipal
revenue collections, as well as the degree to which municipal tax-payers comply with
municipal revenues.
A meaningful local Revenue Enhancement Plan should estimate the coverage ratio, assessment
ratio, and collection ratio—at least for the main local revenue sources. Then, the plan should
identify one or more specific interventions to be pursued and predict the revenue impact of the
proposed interventions on improvements on the coverage, assessment or collection ratios (and
therefore, on revenue collections). This allows for a meaningful cost-benefit analysis (and value-
for-money) in terms of ranking revenue enhancement interventions.
It should be possible not only to revise the performance incentives under ULGDP II to require
municipalities to produce a REP (as is currently the case), but to make sure that each REP meets
a minimum level of quality. Whereas the latest Revenue Enhancement Manual (prepared by
MUDC with German support) provides some guidance along the lines above (in terms of
analyzing the causes for revenue shortfalls), the current manual does not compel municipalities
to implement such analysis-driven revenue enhancement practices. As such, municipalities
would have to be provided with specific standards with which to prepare their REP in order to
meet a more meaningful revenue enhancement performance criterion under ULGDP II.
To the extent that ULGDP wishes to stimulate municipal owns source revenues as an important
element of sustainable urban finance, the design of ULGDP II should take into account the
strengths and weaknesses of municipal revenue collections.
Need for stronger reporting on municipal revenue collections at all government levels. Previous
chapters suggested that the budgetary and PFM systems in Ethiopia do a poor job reporting on
(and monitoring) municipal expenditures and intergovernmental fiscal transfers. Similarly, the
analyses in the current chapter have revealed the need for stronger reporting on municipal
revenues at all government levels. There is a need to systematically collect data on municipal
revenues for all urban local governments—not just ULGDP ULGs.
Higher expectations for urban local governments. At the local level, there is a need for more
systematic monitoring and reporting of municipal revenues. For instance, in order to ensure
clarity and consistency, there is a need for municipalities to consistently use the revenue
classification codes contained in the Chart of Accounts as they prepare their budget documents,
their revenue enhancement plans, and the revenue reports produced upward to the regional level.
69
In addition, enhanced municipal revenue transparency towards local taxpayers is a critical

hallmark of sound local revenue practices.
However, current municipal revenue practices do not analyze how many taxpayers actually
comply with municipal revenues, nor is any such analysis made public as an indicator of revenue
collection performance of the municipality. Since municipal officials are given the privilege to
collect municipal revenues from their residents, there is also a need to set high standards in terms
of transparency and accountability. On behalf of urban residents, federal and regional authorities
should thus impose no just projectized incentives for good revenue practices, but rather, a clear
expectation and commitment from municipalities to transparency, fairness and efficiency in
municipal revenue administration.
Strengthening the regional role. In order to achieve stronger municipal compliance with good
local revenue administration principles, there is a need to strengthen the role of the regional level
in regulating and monitoring municipal revenues practices. (This is especially true as the number
of municipalities to be managed under ULGDP expands). This is a role clearly assigned to
BOFED in the prototype municipal finance regulations. However, in practice, BOFED and the
regional revenue authority seem to focus almost exclusively on state revenues.
Similarly, regional urban development bureaus often do not involve themselves extensively with
municipal revenues (although they backstop municipalities in the preparation of REPs for
ULGDP). Site visits suggest that the regional urban development bureaus are generally not
appropriately staffed—nor do they have the necessary technical skills—to effectively champion,
backstop and monitor municipal revenue collections within their regional state. This gap between
BOFED and the BUD needs to be filled.
Given that local governance and urban development are functions that are constitutionally the
responsibility of the regional level, it is important for regional states to take this functional
responsibility seriously. For instance, it would behoove regions to make serious efforts to
compile and prepare regional-level reports on local government revenues, and provide consistent
monitoring and backstopping. In fact, it might be appropriate for ULGDP II to include one or
more specific Disbursement-Linked Indicators (DLIs) to encourage regions to play a more
proactive role in managing not just project-related reports, but to become active advocates for
increased and improved municipal revenue collections.
Institutional gaps at the federal level. In addition, a gap should be noted at the federal level in the
policy leadership with regard to municipal revenues: it does not appear that this role is
adequately played either by MOFED or by MUDC. On the part of MOFED, to the extent that the
Ministry is the steward for all public finances in Ethiopia, including municipal finances, where is
the focal point for municipal finances? Who within MOFED will champion a much-needed
revision of the CoA with regard to municipal revenue classifications? Who within MOFED
produces a document similar to South Africa’s intergovernmental fiscal review that provides
policy-driven analysis regarding municipal finances?
On the side of MUDC, the main involvement in the sphere of municipal revenues appears to be
the championing of a modern property tax. This seems to be a curious priority given the basic
70
improvements in municipal revenue administration in Ethiopia that could be made by providing

municipalities across the country access to basic guidance and support on a range of local
revenue administration issues to enhance their revenue performance, and the overall transparency
of municipal revenue collections in the country.
Better targeting of specific revenue-enhancement interventions. If the main cause for lagging
municipal revenues is the non-buoyant nature of municipal revenues (as suggested by the
analysis), then incentivizing municipal revenue collection efforts will be an imperfect solution.
Instead, encouraging municipalities (and the regulatory frameworks at the regional level) to
automatically make annual inflation-adjustments for fees and charges that are determined in
nominal terms can have a much more meaningful impact on stopping the erosion of municipal
revenues.
7. A review of local government borrowing systems

Intergovernmental finance is widely recognized to have four pillars: the assignment of
expenditure responsibilities; revenue assignments; intergovernmental fiscal transfers; and sub-
national borrowing and debt. As discussed in the preceding chapters, sub-national revenue
collections and intergovernmental fiscal transfers are important funding sources for regional and
local governments in Ethiopia.
Although the role of sub-national borrowing is quite limited in the case of Ethiopia, this study
would be incomplete without a discussion of the role of sub-national borrowing. In particular,
there is a need to ensure that a hard budget constraint exists at the sub-national level to ensure
that sub-national governments operate within their means. In this light, this chapter provides a
brief discussion on the role of borrowing by regional governments (section 7.1) as well as the
context for borrowing by (urban) local governments (section 7.2).
7.1 Borrowing by regional governments
National borrowing framework for local government. The Constitution does not mention
borrowing by the regions except to state that the Federal Government may itself grant to the
regions, “…emergency, rehabilitation and development assistance and loans” (Art. 94.1).
However, a framework for sub-national borrowing was already in place at the time of the
adoption of the current Constitution. Borrowing from the domestic market by the regions for
specific projects is governed Proclamation 33/1992 of the Transitional Government of Ethiopia
(Article 10) which defines revenue sharing and financial relations between the national and
regional governments. The Proclamation defines some small details of the borrowing process /
application to be submitted to MoFED, which includes:
• loan amount requested

• statements of the region’s revenue collection forecast and economic indicators
• copy of the region’s consolidated budget
71
• feasibility study of the project that the requested loan is intended to finance.
The Ministry evaluates the accuracy of information provided by the Region, as well as the
potential impact of the loan on the national budget deficit. A borrowing ceiling is to be set for
each Region, within the national borrowing authorization limit. The text of Proclamation
33/1992 seems to imply that the National Bank of Ethiopia (NBE) would authorize disbursement
of the loan.
Borrowing by Addis Ababa City Government (AACG) is regulated specifically by Proclamation

361/2003 (AACG Revised Charter), which provides for the city’s borrowing authority (Article
54). Borrowing from domestic resources as loans or bonds and for short or long term requires
authorization from the Federal Government (Art. 54.1). Among the considerations, that the
Federal Government should take into account in its decision to authorize said borrowing, is the
non-endangerment of macro-economic stability, as well as the financial capacity of the city to
repay the loan, without defaulting or entering bankruptcy (Art. 54.2). The City Council must
approve any loan agreements from domestic sources that have been authorized by the Federal
government (Proclamation 361/2003, Art.14.2.q). Sub-cities and kebeles of Addis are not
authorized to borrow (Proclamation 361/2003, Art. 54.4).66
Borrowing from international financial institutions. The authorization for regions to borrow from
international sources is not provided for in Proclamation 33/1992.
In the case of AACG, and as per Proclamation 361/2003, the city may request the Federal
government to borrow on its behalf from international financial institutions (art. 54.1) and places
on the city the responsibility to identify such sources. In this case and as per FNG Proclamation
648/2009 to provide for the Financial Administration of the Federal Government of Ethiopia, any
money borrowed by or on behalf of the Federal Government must have the authorization of the
House of People’s Representatives (Art. 40.1).
According to discussions with MoFED, not only does the Government authorize the regions and
AACG to borrow, it also provides a guarantee to the Commercial Bank of Ethiopia (source of
domestic loans). In the case of default or a missed payment by a region, MoFED can intercept its
transfer to the region and use the funds to honor the debt payment vis-à-vis the CBE. There do
not appear to have been any cases of missed payments or defaults by Addis; in such a case, as the
Federal Government does not provide transfers to the city, another form of encouragement to the
city to meet its debt obligations would have to be applied.
Examples of domestic borrowing by the regions include loans to finance housing construction
and loans to finance fertilizer. Addis has also borrowed to finance housing construction – these
are a form of medium-term relay loans, which are then converted to mortgage loans to the final
beneficiaries. The loans were provided by the Commercial Bank of Ethiopia with a guarantee
provided by the FGE.
66
Further detail about the role of borrowing in the financing of Addis Ababa City Government is covered in greater
detail in the Component 1 report (Chapter 5).
72
7.2 Local borrowing by sub-regional urban local governments
Legal environment for local government borrowing. Borrowing by (urban) local governments is
regulated by regional proclamations and regulations, and thus, varies from region to region.
Furthermore, the policy view of local government borrowing does not seem to be static over
time.
The prototype Regulations to Provide for City Government Finance Administration (prepared by
MOFED in 2005 and generally adopted by regional councils from 2006-2010) were reasonably
permissive with regard to local borrowing. These regulations recognized the right of city
governments to incur both short term debt (within the financial year) as well as long-term debt.
These regulations restrict both types of debt to specific purposes (short-term cash flow
management in the case of short-term debt, and capital investment in the case of long-term debt),
require the local council to specifically approve the borrowing, and subject the borrowing to
BOFED approval.
In recent years, BOFEDs in some regions have moved to restrict the financial management of
urban local governments by insisting that cities should be treated like rural woredas. As a result,
some states have adopted regional financial administration proclamations that apply regional
financial management processes on urban local governments, likely restricting their ability to
engage in borrowing.
Local government borrowing: de facto situation. From a legal viewpoint, a hard budget
constraint exists given that the legal framework—when borrowing is allowed at all—restricts
local government borrowing to specific situations and requires the approval of the higher
government level. However, what matters more than the legal situation for achieving a hard
budget constraint is the actual practice, as it is possible for local governments to engage in
certain types of (quasi-)borrowing without permission from the higher government level.
Fieldwork did not suggest that extensive borrowing was going on at the local level. However,
there was some anecdotal evidence of “creative” borrowing in the sample. For instance, one of
the City Administrations visited by the study team has a line of credit with a local truck
dealership, which allows the city to purchase trucks and pay them off over time. Similarly, in one
of the other cities visited by the study team, the municipality had borrowed funds from the water
authority to construct an asphalt road. In this instance, internal auditors flagged this arrangement
based on the argument that the municipality did not have the legal authority to borrow these
funds.
73
PART III: CITY MANAGEMENT PRACTICES

This part of the report will analyse in more detail selected City management practices with
regards to planning, financial management, asset management, O&M, procurement, urban
service delivery and revenue management.
The analysis is primarily based on four key sources:
• Fieldwork in 12 urban local governments (the sample as determined by TOR – see Table
1, for overview). – This in turn was based on fairly short visits to each of the 12 cities (2
days) plus visits to the related regional offices. In each of the cities the team visited the
city managers office, the finance departments (including units responsible or accounts,
internal audit and procurement) and the local revenue office. At regional level, the team
visited the BUDC, BOFED, the revenue office and the Auditor.
• Data from the Annual Performance Assessments (APAs) over the last three years (that
covers Addis Ababa + 18 other Urban Local Governments),
• Data from the GIZ Supported Urban Governance and Decentralisation Programme
(UGDP). This includes annual reports as well as detailed assessments of selected cites.
• Data from ULGDP supervision missions.
The main purpose of the review is to inform future design of the ULGDP – in particular the
procedures for performance assessments of cities. Each of the chapters include where relevant a
brief outline of the polices and other guidance offered by the Federal state and regions to cities
followed by a discussion of city practices and current systems for performance assessments.
Each of the chapters concludes with an assessment of areas for possible improvements for the
future ULGDP assessments.
74
8. Participatory planning, formulation of CIPs and Transparent

Governance
8.1 Policies and Guidance for Urban Planning / CIPs
The Government of Ethiopia strives for public participation and transparency in the planning
process at all levels of Government. At local government level this policy ambition is also
inscribed in law as all City proclamations include various requirements to that effect (see e.g. the
box below).
Box 8.1: Requirements for participation and transparency in City Charter

The City budget shall be prepared transparently and in a manner that ensures the participation of
residents. Budgetary offices and other concerned organs shall give their opinions on a budgetary
proposal before it is approved. Mass media shall be employed to maximize the participation of
residents. The meetings of the Council deliberating over budget proposal shall be open to
residents67
All cities in Ethiopia have development plans, consisting of strategic development objectives and
structure plans. City development plans encompass both physical and spatial objectives.
The federal government’s Growth and Transformation Plan, which outlines guiding objectives
for cities, is guiding the cities in development of their own local, five- year GTP integrating both
federal and local objectives. This plan is thereafter translated into the yearly Action Plan of city
line-offices.
The MUDC is assisting the cities with development of three corresponding plans:
1. The Revenue Enhancement Plan (REP),

2. The Asset management Plan and
3. Capital Investment Plan (CIP).
These plans are closely related and refers essentially to three steps of planning68:
The Revenue Enhancement Plan (REP) shows how much revenue the city administration can
generate from its own resources in three to five years’ time. A key output is the Capital
Investment Budget for the upcoming three years, as well as the Revenue Collection Plan, which
articulates how the city will collect revenue.
67
Proclamation No. 416/2004 the Dire Dawa Administration Charter Proclamation
68
Ministry of Works and Urban Development, UGDP: Capital Investment Plan Guide for Ethiopian Cities (July
2011).
75
The Asset Management Plan translates strategic objectives into concrete project proposals, such
as infrastructure extension, or one-time investments, such as vehicles. A second function of the
AMP is to manage existing infrastructure by determining maintenance requirements and costs,
which are then consolidated in the maintenance budget.
The CIP derives from both these plans. Their key functions are to link the CIP with the city’s
physical and financial objectives.
MUDC has guided this process in ULGDP participating cities with TA support and use of
training and manual developed by GIZ (and in earlier phases of support from UI).
8.2 Measurement of CIP and Planning Procedures in APA
During the Annual Performance Assessment planning and budgeting was assessed with regards
to the planning process and linkages of the plans.
This includes:
1) Clear description of the planning process with evidence of the involvement and
participation of citizens,
1. Extent to which O&M and recurrent costs as specified in the Asset Management Plan
(AMP) are reflected in Annual Action Plan (AAP) and budget for EFY 2004; and
2) Evidence of clear linkages between the CIP, the Annual Action Plan (AAP) and the
Procurement Plan (PP).
The results from the most recent assessment are summarized below69:
Description of the planning process & evidence of the involvement of citizens: The result of the
assessment, Figure 8.1, shows that the majority of the cities (15 out of 19) have shown good
performance demonstrating clear and comprehensive evidence of planning process including
evidence of effective public participation. On the other hand 3 cities (Dessie, Kombolcha and
Axum) do have evidences for this indicator, but it is not comprehensive, while there is no clear
and adequate evidence of planning process & public participation in Addis Ababa city.
69
MUDC, SUDCA 2013: ULGDP Annual Performance Assessments for EFY 2004 (2011/12) Draft July 2013. The
summaries of results are ad-verbatim quotes from the report.
76
Figure 8.1: Summary score APA 2004: City Planning and Citizen Involvement
77
Extent to which O&M and recurrent costs as specified in the AMP are reflected in AAP &
Budget: The result of the assessment, Fig 8.2, shows that only about half of the cities (10 out of
19) have shown good performance demonstrating coherence of O&M and recurrent costs
specified in the Asset Management Plan with that of Annual Action Plan and budget for EFY
2004. The good performing cities are Adama, Bishoftu, Dilla, Dire-Dawa, Harar, Jimma,
Kombolcha, Mekele, Shashemene and Wolaita Sodo. On the other hand 9 cities (Addis Ababa,
Adigrat, Arba Minch, Axum, Bahir-Dar, Dessie, Gondar, Hawassa and Shire) did not fully
reflected O&M and recurrent costs specified in the Asset Management Plan in Annual Action
Plan and budget for EFY 2004.
Figure 8.2: Summary Score APA 2004: O&M Specified in AMP
78
Evidence of clear linkages between the CIP, the Annual Action Plan (AAP) and the Procurement
Plan (PP): The result of the assessment, Fig 8.3, shows that the majority of the cities (12 out of
19) have shown good performance demonstrating clear linkages between the CIP, the Annual
Action Plan (AAP) and the Procurement Plan (PP) for EFY 2004. The good performing cities
are Adigrat, Arba Minch, Axum, Bahir Dar, Bishoftu, Dilla, Dire-Dawa, Hawassa, Jimma,
Mekele, Shashemene and Wolaita Sodo. On the other hand, for 7 cities (Adama, Addis Ababa,
Dessie, Gondar, Harar, Kombolcha, and Shire) linkages between the CIP, the Annual Action
Plan (AAP) and the Procurement Plan (PP) for EFY 2004 is partial.
Figure 8:3 Summary Score APA 2004: Linkages between CIP, AAP and PP
8.3 City Achievements and Challenges for Improved Planning
There is no doubt about the achievements of cities with regards to improvement of their CIPs
under ULGDP. Part of the improvements is directly linked to the fact that ULGDP for the first
time has allowed the cities a meaningful capital budget for planning which in itself is a
significant trigger for improved practices for planning. However, capacity building and the
incentive structure of the grant have also begun to institute important improvements within each
of the three areas measures by the APAs.
In general it can be observed that CIPs all are in place in all the cities sampled for the study.
However, the quality of CIP planning is naturally higher in the existing ULGDP cities than in the
new cities. The differences are most significant with regards to the technical competency of CIP
drafting as well as internal coherence between CIP, AMP and REP, etc.
79
There is a marked difference in quality of city performance between the GIZ and so-called
ULGDP supported new cities; the ULGDP support has not yet started effectively whereas GIZ
has supported some of the new cities for 2 years – their plans are naturally more developed,
although they bear witness of significant external support.
Both the old ULGDP cities and the new pursue similar strategies for participatory planning
foremost in the form of public hearings at kebele level but also with participatory meetings with
specific stakeholder groups such as “professional associations, women associations, youth
associations, faith based organizations, disable, Bahir Dar University, vulnerable groups,
business community, rural community, government officials, civil servants and political
parties”70.
Some of the most common weaknesses in the CIPs that are captured by the APA include lack of
integration between CIP and other documents (REP, AMP, budget). However, there are also
other weaknesses not captured by the current APA:
• Databases for planning appear not well developed – especially with regards to broader
development objectives and targets as discussed in the CIPs. It is for instance very
difficult to assess relative coverage of cobble stonework or relative coverage of other
municipal services. The plans typically mentioned that e.g. cobble stonework should be
expanded from x to y rather than initially assess the relative coverage. Data on e.g. solid
waste collection.
• Selection of projects are seemingly based on consultations but more explicit technical
comparisons of project proposals are not presented (e.g. some form of cost-benefit
analyses).
8.4 Measurement of Transparency and Accountability in APA
During the APA transparency and accountability was assessed based on two indicators:
1) Posting of summary annual budgets, approved projects, expenditures, audited accounts

and results of procurement decisions in city offices and other public places; and
2) Timely submission of quarterly financial and physical progress reports for the previous
fiscal year.
The most recent APA summarizes the results as follows:
Posting of budgets, projects, expenditures, audited accounts and results of procurement

decisions: The result of the assessment, shows that only 11 out of 19 cities have shown good
performance demonstrating full evidence for posting summary of annual budgets, approved
70
Example from Bahir Dar CIP, EFY 2004.
80
projects, expenditures, audited accounts and results of procurement decisions in city offices and
other public places for EFY 2004. The good performing cities are Adama, Adigrat, Arbamich,
Bahir-Dar, Bishoftu, Dessie, Dilla, Hawassa, Kombolcha, Shashemene and Shire. On the other
hand 6 cities (Addis Ababa, Axum, Dire Dawa, Gondar, Mekele, and Wolaita Sodo) do have
evidences for posting of summary annual budgets, approved projects, expenditures, audited
accounts and results of procurement decisions in city offices and other public, but done partially,
specially most failed to post information on expenditure on quarterly basis. On the other hand 2
cities (Jimma and Harar) do not have evidences or have scanty evidences for this indicator.
These cities are, particularly, failed to post results of procurement decisions and expenditure
reports in city offices and other public places. Overall, it has to be noted that most of the cities
failed to post audited accounts in city offices and other public places, saying that it is
confidential.
During fieldwork in the 12 ULGs it was difficult to locate the posted plans and reports. In fact,
the team didn’t find one single city where all the supposed information was clearly published. It
is possible that the publication is made mainly around the time of the visit of the APA team. It is
also clear that some of the information only is made as very crude summarized with limited
practical use for the public.
Timely submission of quarterly financial and physical progress reports. The result of the
assessment shows that the majority of the cities (14 out of 19) have shown good performance in
submitting all quarterly financial and physical progress reports on the time for the EFY 2004
(2011/12). The timely reporting cities are Adama, Arbaminch, Axum, Bishoftu, Dilla,
DireDawa, Gondar, Hawassa, Jimma, Kombolcha, Mekele, Shashemene, Shire and Wolaita
Sodo. On the other hand four cities (Adigrat, Bahir Dar, Dessie and Harar) they do submit
quarterly financial and physical progress reports but miss to meet deadlines for one quarter;
while Addis Ababa never met the deadlines of all of the quarters.
8.5 Issues Arising for Future Improvement of APA
The APA has three main indicators related to the quality of planning:
• Planning process with evidence of the involvement and participation of citizens

• Linkages between CIP, Asset Management Plan, Annual Action Plan and budgets
• Linkages between the CIP, the Annual Action Plan (AAP) and the Procurement Plan,
In addition the APA have one indicator on transparency (posting of budgets and accounts etc.
and another on reporting).
These indicators are all relevant and the data collected for their measurement generally provides
a valid assessment; e.g. the analysis of internal coherence between CIP, AAP and REP and
accounts is important and fairly straight forward to undertake when the cities provide the
required base documents – and these may in future be published on-line and therefore not require
fieldwork in future for assessments.
81
However, at least two areas could be improved upon:
1) “Participation” is currently measured rather crudely – foremost with figures on “total

number of citizen meetings” and “number of groups consulted with”. It would be
unreasonable to expect the cities continuously to report on growing number of meetings
and groups. Instead it could be more useful to establish a general standard for type (and
number) of meetings and other opportunities for citizens to air their views. This could
e.g. include standards for public hearings with standards for how agenda and other
information should be shared in advance etc.
2) Transparency is currently measured with regards to publication of accounts, budgets etc.

However, at present there is no clear format how such information in a meaningful way
can be summarized in a way that effectively will empower citizens. It is recommended
that clear standards be developed for such information. In addition it is recommended that
the information also is made available on the Internet. More specifically it is suggested
that all municipalities could upload all documents required for the APA process to an
online system managed by MUDC and/or regional urban bureaus. Documents would be
time-stamped when uploaded by municipalities. This would ease the assessment process
and also provide relevant documentation into the public domain for greater transparency.
Documents could include:
• CIP
• Citizen service charter / O&M information
• REP / quarterly revenue reports.
• Budget proposed by executive (including state functions, OSR-funded muni
function, and ULGDP-funded investments)
• Budget adopted by council (including state functions, OSR-funded muni function,
and ULGDP-funded investments)
• Final Financial Statements, and
• City Structure Plan.
82
9. Local Government Financial Management
9.1 Key policies and Guidance for PFM in Cities
The legal framework for PFM in the cities is based on the specific regional legal framework for
PFM: Financial Administration Proclamations, accompanying Financial Regulations, Annual
Budget Proclamations (including proclamations for supplementary budgets), Tax proclamations,
Procurement Proclamations and Proclamations concerning the Office of the Auditor General
(ORAG). There is a great deal of uniformity across cities and regions, except for the specific
legislation and regulations that pertain to municipal revenue management (as discussed in
Chapter 3) but also variation in other elements of PFM legislation (see Box below).
Box 9.1: Regional Variation in Local PFM legislation

Oromia: The Finance Administration Regulation for the cities that has been pending for the last year is
still being approved by the regional Cabinet. BOFED confirmed in writing to have no objection to the
issuance of the Regulation. The draft Regulation has been submitted to the Bureau of Justice in May 2012
in order to be passed on to the Cabinet for enactment. Finally, after a long delay, the Justice Bureau has
returned this letter to BIUD with comments. BIUD has responded to the comments and has sent the
document back to the Justice Bureau. The regulation is planned for final approval within the coming
months.
SNNPR Regional State: The adoption and implementation of the new City Financial Management
Regulation in the region since June 2008 has paved the way for the subsequent improvements of the
financial management system. The regulation has installed a unified financial management system with
the introduction of a new chart of accounts. Furthermore, the new chart of accounts has enabled
modernizing the financial management system of the city administration by making use of automation
(i.e. the implementation of IBEX) possible.
Amhara Regional State: Following the “Revised Proclamation for the Establishment, Organization and
Definition of Powers and Duties of the Urban Centers of the Amhara National Regional State”
(Proclamation No. 91/2003) in 2003, a Financial Administration Regulation (Regulation No. 37/2005)”
was issued by the Regional Administration Council in 2005 with the objective of establishing a
permanent financial resource administration system for the cities. The Amhara Urban Centers’ Revenue
Tariff Regulation was one of the most important legal documents established in 2009; this latter provides
for cities to mobilize their own revenues.
Tigray Regional State has three proclamations on the basis of which the cities perform their day to day
activities, namely: (1) Cities’ Establishment Proclamation No.107/1998 E.C (2) Cities Finance
Administration Regulation No. 50/200 E.C, and (3) Procurement and Property Management Proclamation
No. 174/2002 E.C. Moreover, a new tariff regulation has been prepared three years ago and submitted to
the cabinet for approval. At the beginning of 2005 EFY, training was given to revenue officers assuming
that the regulation would soon be approved. For an unknown reason, the RRA issued a letter notifying
officers to continue with the old tariff. The cities were forced to revise their REPs and annual revenue
plans accordingly. However, according to recent information, the regional government is now ready to
approve and implement the tariff regulation in the coming fiscal year in order to replace the 15 year old
one.
83
Gambela: One of the major achievements of the reform is the approval and implementation of the City
Financial Administration Regulation No. 36/2003 E.C., which was enacted in October 2010. This
regulation was the first legal document approved by the regional government and has paved the way for
the unification of the financial management system. The previously issued tariff, in use for many years by
the municipality of Gambella, was not proportional to the services delivered. Thus, efforts went into
developing tariffs that are comparable to the current market situation and equivalent to the services the
city administration delivers. Therefore, following the Financial Regulation No. 36/2003 E.C., Gambella’s
Regional State Urban and Tariff Regulation No. 42/2004 was issued by the region as a legal instrument to
set forth sources of revenue and expand the tax base for Gambella and eight other municipalities.
Afar Regional State: The Semara-Logia city administration is operating based on the city establishment
Proclamation No./1998 issued by the regional council. This proclamation has recently been amended to
give more power and responsibilities to the CA. Due to lack of urban related financial legal framework,
the cities were using the Regional Financial Administration Regulation for many months. In April 2012,
the regional council issued the City Finance Administration Regulation. Furthermore, the new tariff
regulation has been approved recently and is ready to be used by Semara CA. Recently the city
administration has raised some questions on the issued proclamations and regulations regarding power
distribution and responsibilities. Some parts of the proclamation are currently being revised.
Somali Regional State: The “Financial Administration Regulation” as a legislative basis for the financial
management system was approved by the regional Cabinet in July 2011. Somali Regional State Urban
and Tariff Regulation was developed and issued by the region in December 2011 as a legal instrument to
set forth sources of revenue and enable cities to deliver the efficient and quality social, economic, and
administrative services required by the residents.
Benishangul-Gumuz Regional State: In 2007, the council of Benishangul-Gumuz Regional State

announced Proclamation No. 69/2007 providing for the establishment and organization of Urban Centers
of the Benishangul-Gumuz Regional State, and a definition of their power and duties. Following this,
Asossa was to be organized at the city administration level. The administration now accommodates 13
state function offices. The Financial Administration Regulation (No. 54/2011) and the tariff regulation
were both approved in November 2011.
Source: Urban Governance and Decentralization Programme (UGDP) annual report (July 2013)
PFM reforms have gathered pace
Use of IT in PFM systems is now common in all regions and the majority of cities. It started out
in the early 2000s through the development of a computerized Budget Information System (BIS)
for reporting on budget performance and a Budget Disbursement and Accounting System (BDA)
at MOFED and BOFED level, both systems being stand-alone. These two modules were then
merged under the umbrella of an Integrated Budget and Expenditure Management (IBEX)
system, consisting of the following modules: budget, accounts, budget adjustment, budget
control, accounts consolidation and administration. The accounts module manages the tracking
of revenues and expenditures of public bodies: specifically, it records the financial transactions
of budgetary institutions, captures the aggregated monthly accounting reports and provides
accounting reports in the form of ledgers, financial statements, management reports and
transactions listings. GoE has plans to roll out a comprehensive IFMIS system within the coming
3-4 years. However, based on international experiences, it will take time for all cities to be
enrolled in this.
84
The previous separate municipal and state financial management systems were since 2008
unified with a corresponding introduction of a new chart of accounts.
IBEX was rolled out to BOFEDs and during 2008/09 and has since then gradually been rolled
out in regions to cities. All the cities visited use IBEX for state and municipal revenues, whereas
accounts for ULGDP are kept either manually, using Peachtree software or stand-alone versions
of IBEX. In addition, expenditure cannot be integrated into the IBEX system with the current
system of budget proclamations (currently only done at the Federal (MUDC)71 level, and the
manner in which funds are reported for using a parallel reporting system).
Business Process Re-engineering (BPR)

BPR has been undertaken in all regions since 2007/08 and included various studies to determine
how to improve administrative efficiency in support of strengthened PFM. Implementation of the
recommendations commenced in 2008/09 and is still ongoing. Implementation activities so far
have included
• At both state and city levels: re-organization of offices; e.g. in BOFED, the procurement,
property management and financial administration departments were combined, to form
the Financial Administration and Property Management Department, and departments
have been re-designated as “business processes” for the core functions, such as budget
preparation and financial administration, and “supportive processes” for supporting
functions such as human resource management;
• Creation of ERCA and the establishment of the related offices at regional and city level.
In the cities this has included transfer of veracious revenue collection functions from the
city managers office into a consolidated Revenue Authority at city level – this process is
still ongoing in a few cities – e.g. in Debre Berhan the team found that the office of the
city administration still was responsible for collection of some fees.
• Reduction in the number of signatures required for the implementation of a process, such
as procurement; and
• Establishment or increase in the numbers of positions for in internal audit (at both
regional and city levels) departments in order to increase their effectiveness,
9.2 Key Finding on Financial Management practices in cities
City financial management practices are monitored under ULGDP with particular emphasis on
ULGDP specific practices. It should be noted that these differ to some extent from how cities in
general manage their finances from state and municipal revenues. The most important
differences are:
1. Separate bank accounts are kept for ULGDP funds, collected state revenues and the
municipal revenues.
71
IBEX (only) covers state and municipal revenues (fully proclaimed revenues and expenditures), whereas most
donor funded projects are reported through additional stand-alone systems.
85
2. State revenues and municipal revenues are accounted for through IBEX from city levels,
whereas ULGDP is accounted for through a mix of systems: manual, Peachtree software
or stand alone versions of IBEX,
3. Cities are for ULGDP required to prepare IFR according to format in Operational Manual
(reported manually or in standalone system), whereas they submit monthly and quarterly
reports on revenues and expenditures to BOFED through use of IBEX.
From the field visits and other sources (e.g. a recent ULGDP Financial Management Review
(FMR) July 2013) the following strengths and weaknesses of the cities financial management
can be observed:
• Cities all maintain regular accounts (from state and municipal revenues) in IBEX,
• IFR for ULGDP are prepared on quarterly basis (however, mainly for upwards reporting
and not internal management purposes),
• The FMR mission classified the overall financial management rating for the project as
Satisfactory but risks were classified as substantial with reference to e.g. the manual
accounting practices for ULGDP expenditures,
• Internal audits are now functional in most cities but capacities are weak (see section 11.3
below),
• Internal controls have some general weaknesses reflected in e.g. problems with bank
reconciliations. The FMR noted for instance that “In Dessie the city reconciliation is 1yr
behind, there are reconciling items dating as far back as 2003, the city accounts have
revenue amounting to Birr 21.4m received in 2003 and to date they still have not
reconciled this because they do not know where this revenue came from. This means they
cannot use these funds, and for lack of proper reconciled accounts the decision making
and management is made without proper information.”
• Bank reconciliations are mentioned as problems in many of the reports from the Auditor
general (see Annex 4).
• ULGDP reporting is in some cities maintained with manual accounts, the FMR noted
problems e.g. in Kombolcha where the manual accounts revealed instances of figures
overwritten without any counter signature to show that the supervisors are aware of the
alterations made in the books.
• Cities recording in IBEX is done according to a standard chart of accounts (with some
room for modification by the cities) however revenue coding is not uniform across cities
which makes it very difficult to undertake meaningful analysis of the composition of
municipal revenues,
• Cities occasionally fail to deduct required withholding taxes or VAT72,
• Procurement regulations not always followed73,
• Manuals (in particular ULGDP specific but also government manuals) are not always
readily available for responsible staff,
72
See Annex 4
73
Ibid
86
• Some fragmentation of work processes – e.g. under ULGDP the M&E persons are to
some extent monopolizing work on ULGDP budgets and reporting without adequately
informing other OFED staff,
• External Audit Reports; whereas the backlog issue seems to have been resolved for
ULGDP cities, all ULGDP cities’ audits are qualified (see also section 11.4). For several
of the new ULGDP cities, they have much more problematic Audits – see table below.
87
Table 9.1: Summary assessment of Audit reports for GIZ Supported new ULGDP Cities74
External audit opinion in External audit opinion in General level of

City administration
2003 EFY 2004 EFY compliance with 2004
audit findings (in %)
Nekemte Qualified Unqualified 100%
Yirgalem Audit not available Audit not available Not applicable
Debre Markos Disclaimer Audit not available Not applicable
Adwa Qualified Adverse ~ 75%
Gambella Qualified Audit not available Not applicable
Semara-Logia Audit not available Adverse* Less than 50%
Jigjiga Qualified Adverse ~ 75%
Asossa Audit not available Audit not available Not applicable
Some audit findings are recurring, but both the ULGDP APAs and GIZ TA teams have started to
monitor the extent to which cities follow up on specific audit findings. However, the type of
findings are recurring which points to significant weaknesses in city financial management. Staff
turn over is frequently stated as impediment for improved management.
9.3 Internal audit
Internal audit functions have been established within all cities visited. However, the internal
audit function is generally undertaken as completely integrated within the City OFED with very
limited (if any) autonomy.
During fieldwork, the team only found one case where the internal audit team had some
independence as the internal auditor reported directly to the Mayor rather than trough OFED.
74
UGDP Annual Report.
88
Box 9.1: Key features of Internal Audit in Assosa City
• Assosa internal audit unit is accountable to the mayor - the unit sends its audit reports to the
mayor and copies to OFED,
• Number of staff are four (one degree-, and three diploma-qualified),
• Internal audit unit was established in 2003 E.C,
• Even if the internal audit is given some attention by the mayor, it has problems in terms of
capacity and experience. Internal auditors do not have adequate training,
• No specific budget is allocated to the unit. It is funded as part of OFED with limited budget
for stationery, etc.,
• Focus of internal audit is post audit, financial audit - including procurement issues, salaries,
and operational expenses,
• In 2005, EFY internal audit unit conducted audits in two kebeles by order of the Mayor.
Major findings of the audit were: a) kebeles have no adequate or reliable financial records b)
they collected cash for ID cards, birth certificates, marriage certificates and court fines
without issuing receipts,
• OFED takes a long time to take corrective actions on the findings.
Source: Fieldwork August 2013.
Common observations of the Internal audit units were:
• Relatively low levels of staffing,

• Limited or no computerisation,
• Generally reporting through head of OFED
• Very limited fieldwork and typically not analysing more than a fraction of finance
entities in any one year. E.g. audit of 1-2 schools and one kebele in any one-year.
The above suggest the need for establishment of some minimum standards for internal audits –
e.g. in terms of staffing and scope of internal audits.
9.4 Measurement of Financial Management in APA
The APA has two main indicators for financial management:
• Audit backlog cleared, and

• Auditors opinion for the most recent audit,
Status of Audit Backlog Clearance by Cities: The majority of the cities (11 out 19) have cleared
audit backlogs of the last 9 years (EFY 1995-2003). Among the 8 cities that have not fully
cleared the backlogs, cities in the Amhara region account for 50%, and Cities in SNNP accounts
for 25%. Addis Ababa and Shire are also categorized among the cities that have not fully cleared
the audit backlogs. Concentration of non-compliant cities in Amhara and SNNP implies
inadequate guidance/monitoring from higher level (Regional Office). Temporal analysis of audit
backlogs shows that all the Cites have cleared audit backlogs from EFY 1999 on-wards. The 8
cities that were failed to fully clear the audit backlogs, they are unable to clear the earlier years
89
(1995-1998), mainly because of lack of proper documentation and absence of personnel that
knows where about of the document, due to high turnover of staff and officials.
Investigation into ‘the Outcomes of the external audit opinion’ shows that all cities received
qualified audit opinion for EFY 2003 accounts. The same is also true for EFY 2000 accounts.
Temporal analysis of external audit opinion from EFY 1995 to EFY 2003 shows that Cities
receiving qualified audit opinion have been increasing from year to year except for EFY 2002
accounts, where 6 out 19 (32%) received ‘adverse or disclaimer’ external audit opinion. Analysis
of frequency of ‘Adverse or Disclaimer’ external audit opinion by Cites in the past 9 years (EFY
1995-2003) shows that there are 7 cities out of out of the 19 cities that were repeatedly received
‘Adverse or Disclaimer’ external audit opinion. This includes Harar, Mekele, Adigrat, Shire,
Dilla Bishoftu & Dessie. Among the seven cities Harar received ‘Adverse or Disclaimer’
external audit opinion, six times, while Mekele received three times, the rest received two times
each, except Dessie which received Adverse or Disclaimer’ external audit opinion only once.
In addition the APA also makes a very useful record of major audit findings for each city as well
as a scoring (from 0-2) of the extent to which the city has or is in the process of addressing the
audit queries. This scoring is at present not included in the formal performance assessment.
Currently PFM issues are almost entirely assessed by use of the Audit General report – this has
some advantages:
• It is relatively simple to assess – in principle it can be done without a field visits but
through review of the Audit reports,
• The Auditors reports is an authoritative document with significant standardisation and
objectivity of assessments,
• The status of the Auditor Generals report – and cities efforts for follow up on audit
findings may be further stringed.
However, the current APA of financial management also has some challenges:
• Audit backlog – and timeliness of audits (as long as financial reports have been submitted
on time, which is not always the case) – is not really (only) a city responsibility and it
could be viewed as unfair to penalise cities for problems in this area. It might in future be
more appropriate include this as part of the DLI for regions.
• The Audit Opinion is a very crude measurement of financial management practices in
cities. In this years assessment all cities received “qualified audit opinion” – yet a more
detailed assessment of selected issues such as e.g. bank reconciliations, strengths of the
internal audit system and procedures, etc. would probably have demonstrated more
variation across the cities.
90
• The Audits provide a slightly out-dated picture of financial management practices – the
most recent APA (completed July 2013)75 is reviewing Audit reports for EFY 2003 and
financial management practices may have improved or worsened since then, hence need
to review follow up on this after audit reports etc.
If the APA continuously should spur cities to improve their performance – also in areas related to
financial management, it may be necessary to include other/additional indicators than just the
audit report. This could for instance include

• Other on-site analyses of selected financial management practices deemed critical
(exclusive use of numbered vouchers etc..)
• Analyses of adherence to good practices for internal audits (clear standards/good practice
benchmarks need to be developed),
• Submission of final accounts in due time (perhaps even as a minimum access condition)
• Indicators on follow up on audit queries.
• Independent Internal Audit function (reporting directly to Mayor) with significant audit
of capital works (e.g. 25% of all contracts annually).
The challenge with the above is of course that
• On site analysis in general is time consuming and not near as comprehensive assessments
as a full audits (however, it should be noted that the costs of the APAs still only is
minimal - about 0.3 % of the grants).
• It will require some work to establish clear standards for e.g. internal audits.
75
However, part of this delay is also due to the delays in the APAs, which should have been conducted nearly a year
earlier for APA 2004.
91
10. An assessment of Asset Management and O&M Planning
10.1 Current procedures of assessment of asset management and O&M in APA
Asset management is during the APA assessed using two basic performance indicators
1. Extent to which asset management plan was updated:

2. Utilization of operation and maintenance budget of the CIP.
The first of these indicators is assessed based on the following five sub-indicators.
1 Did 'Asset Inventory', updated for EFY 2004, featuring a tabular and spatial database of all
infrastructure, with specifications and characteristics, at least for the five categories of Municipal assets
(Road and Drainage, Solid & Liquid waste, Socioeconomic Infrastructure, Greenery, Utilities, Public
buildings)
• Road and drainage network: Amount, surface, and condition of existing physical infrastructure
• Solid and liquid waste: Facts and figures on waste production and collection, existing infrastructure,
vehicles, collection scheme, etc.
• Socioeconomic infrastructure and greenery: parks and greenery, markets, industrial zones and
abattoirs
• Utilities: Facts and figures on water & streetlights
• Public buildings: Facts and figures and conditions on-(public gathering hall, youth center, public
recreation and sport center, etc.)
2 Did condition of existing assets, which describes the current status of deterioration of the asset base,
reflected in Asset Inventories (Fill Annex- 3 and judge accordingly)
3 Did the inventory show an asset value and deficit, which calculates the remaining asset value,
maintenance, and rehabilitation deficit based on annual depreciation rates
4 Did the City update the AMP according to the AMP 10 steps, elaborating its implementation strategy,
which details individual activities and their respective budgets over the course of the year (2004 EFY)
(Fill Annex-1 and 2, judge accordingly)
5 Did the AMP clearly show the related budget for asset maintenance, rehabilitation and new assets,
which lists all necessary infrastructure development for the city (Fill Annex 2 & 4 and judge
accordingly)
The results of the most recent APA are summarised below.
With regards to “updating asset management plan” almost all the cities have updated the Asset
Inventory and Asset Management Plan. Overall, as depicted in Fig. 10.1, 8 cities (Hawassa,
Kombolcha, Addis Ababa, Shashemene, Bishoftu, Gondar, Desse and Bahir Dar) showed good
performance and the rest performed satisfactorily. However there are considerable variations on
the comprehensiveness of the asset inventory and planning process. Most of the cities have
comprehensively inventoried and developed management plan for road and drainage network,
and Streetlights. However, almost all cities missed inventory of socioeconomic infrastructure
such as: parks and greenery, markets, industrial zones and abattoirs; and public buildings such
as: public gathering hall, youth center, public recreation and sport center, etc. There are also
some cities that missed inventory of water facility and solid and liquid waste facilities. Most of
the cities followed the 10 steps of AMP, except that some missed step 2 ‘‘Build the Asset
Management Templates) and Step 10 ‘Develop the Implementation Strategy for the Asset
92
Management Plan. Moreover, the AMP of some of the cities did not clearly show the related
budget for asset maintenance, rehabilitation and new assets for some of the necessary
infrastructure of the city.
Fig 10.1 Score of the Cities in for Asset Management Plan on 0-5 scale (EFY 2004)
With regards to utilization of operation and maintenance budget of the CIP, the cities vary
considerably. Among the 19 cities 4 cities (Hawassa Arba Minch, Kombolcha and Gondar)
reported they have utilized over 100% of O&M budget. On the other hand Harar and Adama
have utilized nothing. Overall 9 cities utilized above 80%, while 5 cities utilized below 50% and
the remaining 5 cities utilized 50-80%. However, most of the cities don’t have expenditure
records disaggregated by category of assets, which makes it challenging to assess even if the
recorded fund utilization and quantity maintained make sense.
10.2 Observations on Asset management and O&M from the Field Visits
All cities visited during fieldwork – also the new ULGDP cities – had some form of Asset
Management Plan. However, asset management and operation and maintenance of capital work
and facilities also appeared to have many problems that were not clearly addressed by the AMPs:
Several drainage channels were poorly maintained and some were half blocked during rains.
Existing cobble stone works were not given much attention by city authorities for maintenance
“as other parts of the city without cobble stone are in need of more support…”. In addition it was
observed that user fees for facilities (such as abattoirs, city bus, etc.) were set rather arbitrarily
(in several cases by the regions rater than the concerned cities) without regards to true
maintenance costs.
93
It is recommended to sharpen the focus on actual implementation of O&M activities rather than
just mapping of infrastructure and assets. Putting more emphasis on comprehensiveness of actual
O&M plan and budget and subsequently monitoring of actual budget utilizations will sharpen the
focus of assessment more on actual O&M.
94
11. An assessment of procurement systems

11.1 Current Assessment of Procurement in the APA
There are four main indicators used to assess the procurement performance and compliance of
the cities during the APA. Each of these four indicators includes several sub indicators discussed
below:
A. Preparation and implementation of annual procurement plans for ULGDP according to the
timeline;
B. Adherence to procurement procedures;
C. Availability of annual procurement audit reports that include physical inspections and
compliance checks;
D. Maintenance of adequate relevant auditable records on the procurement process
A. Preparation and implementation of annual procurement plans for ULGDP according to the
timeline: The assessment for this indicator is based on five sub-indicators: i) number of
items procured versus number of planned items (60%), ii) Variation of planned Bid Floating
Dates and Actual Bid Floating Dates (10%), iii) Variation of planned award date and actual
award date (10%), iv) Variation b/n Bid floating and Bid closing (10%) and v) Variation of
b/n Bid closing and award date (10%). The result of the assessment. The analysis of results
from the recent assessment showed that only 6 cities (Adama, Axum, Bahir-Dar, Bishoftu,
Gondar Shire) have shown good performance in preparing and implementing annual
procurement plans for ULGDP according to the timeline for the EFY 2004 (2011/12). The
remaining cities performance is satisfactory, except Addis Ababa that failed to implement the
annual procurement plans for ULGDP for EFY 2004.
B. Adherence to procurement procedures: Adherence to procurement procedures were assessed

against 12 sub-indicators: (1) adherence to the thresholds limit for responsibility of
procurement decision, (2) Adherence to the thresholds limit for World Bank prior/ post
review, (3) Adherence to threshold limit for procurement method, (4) Adherence to SBDs
prepared including NCB, DC, LCB, RFQ, RFP, (5) Invitation of bid is made using
appropriate Public media, (6) Adherence to Bid Evaluation report and approval obtained,(7)
Notification to unsuccessful bidder/regret letter disseminated, (8) Award of contract is made
to the winner of bid within the period of the validity of bid stated in the bid document, (9)
Bid security and performance bound is collected as stated in the bid document (not be
applicable for LCB and shopping), (10) Complete contract agreement signed, (11) Under
shopping procurement method, price quotations are collected from the minimum of three
suppliers (only For shopping), (12) Adherence to the complaints handling system. The
assessment is made by taking 10 samples –encompassing all categories of items procured
(goods, works, and services). Beside this Contract effectiveness (timeline, cost and quality)
was also assessed under this indicator. The result of the recent assessment indicated that only
8 cities (Adigrat, Axum, Bishoftu, Dire-Dawa, Hwassa, Mekele, Shashe and Wolaita Sodo)
have fully adhered to procurement procedures of ULGDP for the EFY 2004 (2011/12). The
remaining cities performance is satisfactory, showing only minor non-compliance. The
commonly bypassed procedures are notification to unsuccessful bidder (most failed to send
95
regret letters when require). Other rarely violated procedures include not awarding contract
within bid validity period stated in the bid document, non- adherence to bid evaluation report
and approval procedures, non-adherence to the complaints handling system. A few cities was
not properly preparing and/or not using SBDs including NCB, DC, LCB, RFQ, RFP.
C. Availability of annual procurement audit reports: Some cities in Amhara, DireDawa

Administration, and Addis Ababa, reported that annual procurement audit is part of the
regional government fiduciary policy, vested into the Procurement Agency’s mandates. On
the other hand the Regional Auditor General Offices are saying that their role is to do
comprehensive audit in which procurement is just treated as one element of audit procedures.
Consequently, there were no full-fledged procurement audit carried out by the city in the last
five year, and hence, there is no annual procurement audit reports made available to the
assessment team.
D. Maintenance of adequate relevant auditable records on the procurement process:

Maintenance of adequate relevant auditable records on the procurement process were
assessed using checklists that contain seven items to be checked. The result of the recent
assessment indicated that only 13 cities (Addis Ababa, Adigrat, Arba Minch, Axum, Dilla,
Gondar, Hwassa, Kombolcha, Mekele, Shashe, Shire, W. Sodo) have shown good
performance in maintaining adequate relevant auditable records on the procurement process
for the EFY 2004 (2011/12). The remaining cities (Adama, Bahir-Dar, Bishoftu, Dessie,
Dire-Dawa, Harar, Jimma) performance is satisfactory. The most common missing records
on the procurement process is ‘Folders for bidders document’. In some cites file boxes or
drawers according to the procurement categories: works, goods and consulting services/
training were also not created.
The assessment of procurement modalities is very detailed, but the assessment process has
generally proved feasible with useful results that have spurred cities to improve their
performance. The only problem with the current assessment system is with the indictors on
“procurement audit”. These are only undertaken as integral element of the general audits with
various levels of emphasis - however most often without separate chapters specifically thus it has
been difficult for the assessment team to verify. This indicator is not recommended to be
included in scoring system until there is a common agreement with Government on how (and if)
to undertake separate procurement audits at city level. Finally, in the future, if audit is expanded
to cover specific procurement audit, the result of this could be applied in the APA, instead of
some of the performance measures currently checked.
96
12. Local revenue management

Earlier parts of this report have discussed several aspects of municipal revenues:
• The regional/local policy framework for municipal revenues (Chapter 3),

• The broader fiscal picture including analysis of regional variations in municipal revenues
(Chapter 4),
• More detailed analyses of municipal own revenues (trends in per capita collections, the
composition of own revenue collections, etc.), trends in municipal revenue collections
compared to others fiscal indicators as well as discussion of revenue potentials (Chapter
6).
This part of the report will look a bit closer at cities management practices with regards to local
revenue management and how the APAs assess these practices with the aim of providing
suggestions for possible future improvements.
12.1 Key policies and Guidance for Revenue Management in Cities
As discussed in Chapter 3, cities administer municipal revenues within the legal framework set
by each of the regions (city proclamations, tariff proclamations, etc.) that provides rather limited
range of policy choices. The federal government has through MUDC sought to influence this
legal framework with e.g. model financial proclamations (that only very partially have been
adopted) and guidance in the forms of e.g. various manuals for Revenue Enhancement
Planning76, but not (yet) model tariff proclamations.
The most advanced version of the REP Manual (MUDC/GIZ 2012) outlines seven steps for
revenue enhancement planning:
1. Considerations of the legal framework (that cities are instructed to follow) and
establishment of REP task force,
2. Analysis of past performance,
3. Examination of expenditure management,
4. Strategies to increase revenue administration efficiency which includes discussion of tax
base, tax coverage, tax rate and ratio, valuation and assessment, collection effectiveness,
payment procedures, enforcement mechanisms and development of human resources and
capacities.
5. Setting up cost accounting system,
6. Analysis of other means of for increasing municipal revenues (this included discussion of
public private partnerships as well as discussion of “increase tax rate and adjustment of
user charges and fees”)
7. Borrowing
76
The most recent is the MUDC 2012: Revenue Enhancement Plan Guide for Ethiopian City Administrations (with
assistance from Urban Governance and Decentralization programme/GIZ), earlier guidelines like Ministry of Work
and Urban Development Guidelines and procedures for revenue Enhancement Plan for Municipalities (GIZ-IS May
2006) were developed as part of regional TA.
97
12.2 Observations on Revenue Management Practices in cities
All the ULGDP supported cities have REPs in place. This includes also the new cities visited by
the team that are preparing for inclusion in ULGDP II. The cities generally adhere to most of the
formal requirements of REP planning (discussed in section 15.4) but revenue management
reforms have not been transformative in any of the cities yet.
The challenges with municipal revenue management are common in the cities visited and are
generally well recognized by the cities themselves and often reflected in the city REPs:
• Cities are generally operating with revenue targets set by regional authorities. Targets are
seen as “the usual top-down way of planning; the plan is given as a quota without detail
analysis and consensus”77.
• Outdated tariffs and rates – both for items set by regions as well as rates set by the city
council (such as rental value of its properties).
• Lack of proper analyses of tax potentials from specific sources; lack of basic date bases on
businesses, rental properties and other revenue sources,
• Very limited use of ICT. Registers of businesses and other municipal tax-payers are
typically maintained as manual ledgers,
• Inconsistent coding of revenues – making it difficult for proper analysis of trends.
• Municipal revenue collections generally have been decentralized to kebele level, but these
offices have not received any significant support in the form of equipment or capacity
building for management of municipal revenues.
• Problems with tax collection efficiency and poor recording of tax defaults and delayed
payment taxes.
• Limited enforcement mechanisms.
• Some REPs also record lack of performance based incentives for tax officers.
Revenue enhancement plans generally make observations on all or most of the above challenges,
but reforms are still to be properly implemented.
Some good practices that have been introduced so far includes (areas that could be included in
future APA):
• Tax information campaigns have started in various forms which has begun to raise a
discussion and importance of municipal revenue,
• Tax rates are publicly disseminated using microphone and by posting tax obligations on
notice boards,
• Compliance for business taxes are enforced as trade and professional licenses have not
been renewed unless taxpayers meet their tax obligations,
• Tax assessments and tax appeal procedures are undertaken in collaborative manner with
business community.
77
Adama REP EFY 2005.
98
12.3 Variation in Local Tax Rates and City Policy Decisions on Local revenue
Most city tax rates and tariffs are, as discussed in Chapter 3, regulated by regional
directives/proclamations, but cities have also in some cases leeway for adjustment of rates or
setting these relatively independently (such as the rental income from municipal property). The
table below summarizes some of the most common tax and tariffs rates across cities.
Table 12.1: Examples of range of municipal tax and tariff rates

City Revenue item
Business and Land rent for Abattoir services Loading and For erection
professional prime per cattle unloading of of land mark
service areas/Commerc goods that enter and sign
charges per ial and leave city boards for
annum promoting
business
Bahir Dar Max Birr Birr 0.25/sq m Max Birr 96 Max Birr Max Birr 405
18,000 5/quintal
Min Birr Birr 0.12/sq m Min Birr 28 Min Birr Min Birr 18
45 0.5/quintal
Debre Berhan Max Birr Birr 0.20/sq m Max Birr 150 Max Birr Max Birr 270
6,000 5/quintal
Min Birr 0 Birr 0.06/sq Min Birr 100** Min 0.5 Min Birr
45 m Birr/quintal 20***
Asossa Max Birr Birr 1.5 /sq m Max Birr 60 Max Birr Max Birr
35,000& 1.75/quintal 200/sq m
above*
Min Birr Birr 0.90 sq m Min Birr 30 Min Min Birr
150 Birr0.50/quintal 75/sq m
Maximum Maximum Birr Birr 250/ m2
Jigjiga Birr 9000 Birr 1.50/sq m 60 Birr 2.50/Quintal one side
Maximum
Birr 11,750 + 4% of value of Birr 120/m2
0.2% of Maximum Birr goods loaded or for one side
Gambella turnover Birr 1.75/sq m 35 un loaded promotion
not available on
Maximum tariff regulation Maximum Birr
Harar Birr 8000 No 28/2001 50 Birr 1.00/Quintal Birr 200/sq m
Max Birr
Zeway/Batu Maximum 140 2.0 Birr /m2 Maximum 125 5/quintal Max 150 Birr
Max10,000 B=1.76 300 (500 for
Mekelle Min 50 Birr R=0.62-1.06 30-35 n.a. two sides)
*Birr 35,ooo for organizations whose annual sales is 4.9 million and for additional one million sales 2.5%. So far it is not
implemented
** Abattoir service in Debre Berhan is outsourced, however the regulation says max Birr 60 and min 36
*** So far the city has not collected from this revenue item even if the regulation says
99
Most of the variation is a result of differences in the regional tariff proclamations. Within
Amhara for instance, the differences between business fees in Bahir Dar and Debre Berhan is the
result of different classification of the two cities by the region. In the cities there were different
complaints regarding the regional guidance/instructions. In some (relative rare) cases the cities
had specific requests for upwards adjustment of the rates (e.g. in Bahir Dar this was by the city
management expressed with specific reference to the maximum tax rates they could impose on
larger businesses), but cities are generally also very sensitive towards possibly levy of “high”
taxes on its citizens. It is rather common that cities, when they have an option for rate setting,
will go for the lowest possible rates. This is for instance reflected in the relative low rates for rent
of municipal properties that are managed by cities.
The by far highest level of business taxation is found in Assosa and is based on the Benishangul-
Gumuz Regional State Urban Revenue and Tariff Regulation No. 53/2011. This tariff regulation
was drafted by private consultancy firm under BIUD/GIZ. Neither the city nor the Regional
Revenue Authority participated significantly in the development of the regulations 78 . The
business taxes are far higher than in more developed cities like Mekelle (Tigray) or Debre
Berhan (Amhara). As a result almost all the concerned taxpayers (approximately 800) have
appealed the tax assessments and Assosa does not yet collect the new taxes.
Land rent is widely recognised as low; plots are typically small and typical annual rents are not
more than 2 USD annually for prime business areas (and generally the land rent remains
unchanged even when significant infrastructure improvements are made – such as cobble stones
in the areas). Still there is significant variation from 0.2 to 2.0 Birr/m2 in business areas. Some of
the local tariff proclamations make reference to “property tax”/”roof tax” but it is not generally
collected.
Service fees for fairly comparable services such as abattoirs services ranges from 35-150 Birr.
12.4 Measurement of Revenue Management in APA
The performance of the cities in Revenue Enhancement is assessed during the APA according to
two main criteria:
1. Revenue Enhancement Plan (REP) updating and approval processes and

2. Percentage of revenue collected against planned for the EFY 2004.
Regarding updating of the REP – the latest APA summarized the results as follows: “Among the
18 cities, excluding Addis Ababa, 13 cities have updated the city’s Revenue Enhancement Plan
(REP) in a comprehensive & complete manner passing through all the processes. As the result
they were categorized as cities with good performance in this regard. On the other hand 4 cities
fall within satisfactory performance category; i.e. they did the updating but missed some
important updating and approval process. The only city whose REP is not comprehensive and
missed a lot of important planning process is Dire Dawa. For Addis Ababa there is no adequate
information to judge the process”.
78
Interviews in Assosa August 2013.
100
The APA uses 11 underlying indicators for assessment of the REP:
Is the REP, updated in EFY 2004, different from the one updated in the EFY 2003 in terms of amount and sources
1 of revenue
Did the REP, updated in EFY 2004, shows that the ULG reviewed past performance, indicating the following
2 Amount of collected own revenues
3 Amount of recurrent expenditures, also called operating expenditures.
4 Capital investments (expenditures) split into different investments.
5 Did the REP, updated in EFY 2004, shows analysis of needs for capital expenditures
Did the REP, updated in 2004, elaborate ways of increasing the revenue base and analysis of capacity to attain the
6 plan?
Did the REP, updated in EFY 2004, forecast revenues, recurrent expenditures and the operating surplus clearly for
7 the coming three years?
Date of Council Meeting held for approving the Revenue Enhancement plan (dd/mm/yyyy)
Minutes of city council meeting referring to or indicating in its agenda EFY 2004 revenue enhancement plan
8 available, (Yes or No)
9 Revenue Enhancement plan of EFY 2004 approved by the council and signed by the mayor (Yes or No)
10 Revenue Enhancement plan of EFY 2004 approved by the Regional Bureau/ Government (Yes or No)
11 Revenue Enhancement plan of EFY 2004 approved by the Federal MuDC (Yes or No)
In terms of revenue collections the lasts APA summarizes the results as follows: “most of the
cities (16 out 19) have managed to collect 80% and above, against planned overall revenue of the
cities for the EFY 2004. The only cites that failed to collect 80% of overall city’s revenue are
Dessie (46%) Gondar (50%) and Harar 79%. Comparison of revenue by two majorly categories
depicts that most of the cities (13 out of 19) tend to achieve 100% or more of the planned state
revenue while only 2 cities (Bishoftu and Dilla) achieved 100% and above in collecting planned
municipal revenue. On the average, excluding Addis Ababa, municipal revenue accounts for
about 34% of the total revenue (ETB 2.35 billion) collected by 18 cities during the EFY 2004.
The only city whose collected municipal revenue exceeds 50% of the total revenue is
Shashemene. Finally the PA report states: “The low level of municipal revenue as compared to
state revenue hint at poor financial sustainability when ULGDP project phase out, and
consequently failure in meeting raised demand of the citizens”.
The APA collects data on both state and municipal revenue generation – however it is the
aggregate figures of overall total tax collections compared to total targets that is utilised for
scoring of the cities. Thus a city may have stagnant or declining municipal revenue collections –
but if total collections grow as projected (because of significant collections of state revenue) they
city will still qualify for a bonus.
101
The procedures for assessment of municipal revenue management have a number of challenges.
The most important are:
1. The assessment of REP is rather formalistic – even if the cities fulfill most or all
indicators it will not necessarily imply that the cities are undertaking meaningful revenue
enhancement reforms and good practices in revenue collection. It is of course relevant
that REP is approved by council meeting, etc. and that REP has basic analysis of past
performance. However the cities adherence to these performance indicators are far from
sufficient for real improvements in revenue management.
2. The current focus on adherence to total (state and municipal) revenue targets is
problematic for several reasons:
a. The focus on combined revenue target ignores the significance of municipal
revenues: if state revenue is substantially increasing, a city will qualify for
rewards even if municipal revenue is stagnating or declining79. However, only
municipal revenues are funding municipal services.
b. The targets (for both state and municipal revenues) are currently set very crudely
and assume general growth of all revenue sources without substantive analyses of
revenue potential – it is therefore difficult to assess revenue efforts as different
revenue performance in different cities foremost may be a reflection of very
different underlying economic factors.
c. Municipal revenues are substantially composed of income from land lease that
mainly is in the form of one-off incomes from city “sale” of land. Such
occasional substantive increases should not be rewarded and should have been
excluded from the review.
12.5.1 Recommended Short Term revision of the assessments of revenue management:
In order to improve significantly on cities municipal revenue management it is required that the
federal government elaborate on the guidance and standards for revenue management at both
regional and city level as discussed further in section 15.2. However, some adjustment can also
immediately be made of the APAs with the above weaknesses in mind in order to feed into the
soon forthcoming APA (tentatively to be undertaken from January 2014).
This should include both indicators for regional as well as city levels.
It is agreed (Project Concept Note) that there will be a DLI for the regional level of government.
The recommended DLI should be sufficiently broadly defined in order to capture the key
79
This has been the case for several cities. For instance Adama municipal revenue declined from 112 million in
EFY 2000 to 88 million in 2004 while its state revenue in same period grew dramatically from 34 to 151 million
(source: Adama REP).
102
requirements for regional support and supervision to cities for improved revenue management
(and possibly even other support for improved urban development).
Possible general DLI for region:
• DLI 7: Regional governments support and monitor cities municipal revenue enhancement
and provide capacity building to cities within core areas of the objectives of the program.
Suggested sub-indicators
• The region has established a consultative forum with cities and has at least once during the
year discussed possible need for revision of tax and tariff proclamations, (minutes of
meeting),
• Regions have established possible bands within granted cities a certain degree of autonomy
in setting local tax rates and tariffs – at least within a recommended band of rates.
• The region has developed a capacity development plan with special focus on (but not only
on) improved city municipal revenue management, (basic requirements for plan to be
developed with regards to type of need assessment, type of other analysis and format for plan
and delivery of capacity building),
• The region monitors cities municipal and state revenue collections and publishes an annual
analytical report with related policy recommendations.
The recommended indicators for city level are recommended to focus on the two basic issues as
today, but with some modification:
Improved Revenue Enhancement Plan
This could follow the existing way of measuring its quality but inclusion of sub-indicators that
are given more significant weight to measure:
• City analysis of previous years revenue performance with detailed analyses of each main
source of revenue including discussion of its revenue potential (revision of current sub-
indicator 2).
• City strategies for revenue enhancement (current sub-indicator 6).
Improved revenue collections and management:
• Proper coding of municipal revenue as per chart of accounts (some additional guidance on
coding should be issued),
• Percentage increase of each main category of municipal revenue (less land lease)
o Business taxes,
o Municipal rent,
o Charges and fees,
103
• City publication of how municipal revenues has been spent in format suitable for citizen
budgets/accounts,
• Registers on revenue collection are up-to-date;
• Clear procedures and evidence on follow up on tax defaulters.
• Public consultations on tax issues especially prior to issuing of new rates and bases
• Inclusion of counterpart funding on the ULGDP/capital projects as a performance measure to
promote sustainable spending of OSR on development.
12.5.2 Recommended medium term revisions of the assessment
In order to incentivize more structural reforms of municipal taxes it is necessary that the Federal
Government/MUDC provides more clear guidelines and specific standards for REP than
currently is the case.
The type of guidance should include:
• Guidance to regions in the form of “Model Tariff Regulations” and guidance on how
regions can assist cities to undertake relevant studies for tax reforms. It should be noted
that the MUDC led municipal tax task force already have initialed work that suggest
recommendations in this direction,
• More specific guidance and standards to cities in the REP manuals regarding how cities
can estimate coverage ratio, assessment ratio and collection ration for main revenue
sources (as discussed in Chapter 6),
• Guidance to cities on coding of revenues sources in the chart of accounts and guidance to
regions for subsequent annual analysis of municipal revenues,
• Working manuals on municipal taxes including simple standard formats for
computerization of tax registers, etc. (at present this is in part included in TOR for the
regional TA however it is more appropriate to establish some national basic standards),
104
PART IV: INITIAL IMPLICATIONS FOR THE DESIGN OF

URBAN GRANT
13. Initial Implications for the design of a Performance-Based

Urban Grant System
The need for a substantial and reliable source of investment in urban infrastructure in Ethiopia is
undeniable. Discussions with policy makers at all government levels indicate that there is broad-
based recognition of the importance of urban development for the success of the Ethiopian
economy. The ULGDP has substantially expanded the pool of resources available for urban
infrastructure investments that have a direct impact on the quality of urban life, including
improved road access to households and businesses in the urban area; improving the cleanliness
of cities; and street lights for improved safety and job creation to improve livelihoods. This does
not mean, however, that there is no room for further improvement.
This final chapter of Component 2A report provides some preliminary conclusions and
recommendations regarding the future design of ULGDP. While the Component 2B reports will
provide more coherent and detailed discussions of the ULGDP design. In this chapter only
selected design issues that have emerged directly from the Component 2A analyses will be
highlighted. This chapter in particular will:
• Briefly summarize general views of the stakeholders consulted in the 12 urban local
governments and corresponding 8 states (in Box 13.A below),
• Discuss the concept of “sustainability” of the ULGDP as a performance based grant
system (section 13.1),
• Discuss options and challenges with regards how best to integrate an (earmarked) urban
development grant into Ethiopia’s intergovernmental finance architecture (section 13.2)
• Discuss issues related to management of funding flows across all levels of government
(section 13.3),
• Reflect on issues related to capacity building at the local level (section 13.4),
• Provide a general discussion of how (better) to stimulate performance of (urban) local
governments (with a more detailed and ULGDP specific discussion in Report 2B) – in
section 13.5,
• And finally a discussion of the implications for the regional level institutional
arrangements – not least a discussion issues related to definition of most appropriate type
of DLI to be applied for regions.
105
Box 13.1: Observations and stakeholder views of the ULGDP / performance-based urban grant
system
Stakeholder consultations at the local level suggest a significant level of appreciation of the ULGDP; it
has proved critical as a financing source for municipal capital funding and introduced many good
practices: labor intensive cobble stone work, improved capital investment planning and management, the
annual performance assessments, etc. However, several areas were flagged by one or more municipal
officials as areas for possible improvement:
Several local officials mentioned the timing of the ULGDP cycle as problematic. It was noted that the
annual resource ceiling for ULGDP is generally not known at the time of budget preparation. As such, it
is difficult to plan and approve the ULGDP investments together with the rest of the annual budget.
Instead, a budget revision is required late in the fiscal year in order to approve relevant ULGDP-funded
investments.
The lack of timeliness of disbursements from BOFED (and/or the difficulty in securing payment from the
higher level) was also mentioned by numerous municipal-level stakeholders as a significant challenge.
This is related to many different aspects of the ULGDP management including the modalities for final
approval of procurement plans, progress reports, etc.
Challenges with municipal counterpart funding were also flagged. It was reported that ULGDP requires
municipalities to deposit their co-funding share (as well as the regional counterpart funding) up-front at
the beginning of the financial year, as a requirement to receive the IDA contribution. This poses particular
challenges for municipal co-funding, since they receive their funding on an ongoing basis throughout the
financial year (rather than at the beginning of the financial year) – in practice this requirement is not fully
adhered to (nor well monitored).
The financial management of ULGDP is not well-integrated with ‘regular’ municipal expenditures.
Different financial management practices were observed in different ULGs, with ULGDP spending being
recorded manually or using private-sector accounting software (e.g., Peachtree) or using a stand-alone
version of IBEX (rather than the regular version of IBEX).
Several local officials noted limitations in the capacity support provided by the project or regions. It was
remarked by some that capacity building support (e.g., in the area of Revenue Enhancement Planning)
was one-off in nature. The absence of ongoing regional backstopping support was also flagged. Both the
cities and regions face challenges with regards to retaining qualified staff – staff turnover is high and
practical solutions are not yet put in place. Some of the newly recruited project staff at regional levels
complained of not yet having been paid salaries (several months after appointment).
Overall the local governments were very appreciative of the annual performance assessment. However,
even though many municipalities seemed to be meeting the pro-forma performance criteria (e.g.,
preparation of a Revenue Enhancement Plan), it is clear that there still is substantive room for
improvement of city management practices in a more transformative sense. In addition some municipal
officials flagged their inability to view the performance of other municipalities, which meant they were
unable to assess their own performance vis-à-vis other municipalities in their regional state.
106
13.1 Sustainability of the performance-based urban grant program
Some policy makers expect that municipalities should be able to cover all their expenditures—at
least, for municipal functions—from own source revenues. Indeed, ULGDP raises the
expectation that its performance-based grants will sufficiently increase municipal revenue
collections in Ethiopia, so that, in the future, municipal development can be fully funded from
municipal own source revenues (ULGDP PAD). In rapidly urbanizing environments, this is not
an appropriate expectation.
From a conceptual viewpoint, municipal jurisdictions are engines of economic growth, which
produce considerable positive spillovers or externalities for the rest of the economy. If federal
and regional governments retain most of the significant revenue instruments for themselves
(typically including the value-added tax, the personal income tax, and the corporate income tax),
it may be necessary to provide municipalities with (targeted) grants to ensure that municipalities
are able to afford the public infrastructure (road infrastructure, local public transport systems,
solid waste management, water supply, and so on) necessary to accommodate continued
economic growth. For instance, cities in most developed economies receive intergovernmental
grants for these kinds of functions.80
In practice, the argument in favor of ongoing intergovernmental grants to municipalities is

especially strong in rapidly urbanizing countries. When rural residents move to an urban area,
these new urban residents place additional pressures on the existing municipal infrastructure and
services, as they do not bring additional infrastructure with them as they move to the urban area.
Since recent in-migrants to an urban area are often lower income households that fall outside the
formal economy, and/or who locate in peri-urban areas—and thus outside the reach of the
municipal revenue system—it is impossible for the municipality to tax in-migrants appropriately
for their use of urban infrastructure. This means that as urban areas are expanding rapidly,
subsidies from a higher government level are needed in order to allow urban areas to fund the
capital investments needed to ensure an optimal level of municipal infrastructure development.81
These arguments suggest that it is not realistic to assume that an urban development grant is only
a temporary feature of the intergovernmental finance system of Ethiopia, which can be dealt with
80
One of the reasons why it is not unusual for local governments (even in urban areas) to receive considerable
funding is because it is usually more efficient to collect state revenues at the center and deliver services at the local
level. Following the subsidiarity principle, this means that (most) revenue collection should be centralized.
However, when revenues are relatively centralized (as is the case in Ethiopia), a mechanism is needed for
transferring revenues from the higher-level government to those local jurisdictions that deliver services, including
urban jurisdictions. It is further useful in this regard to differentiate between funding for recurrent expenditures and
capital expenditures. Whereas local revenue sources may be sufficient to finance the operation and maintenance
costs for urban functions and services, unless revenue systems and borrowing are well-developed (and unless
revenue sources adequately decentralized), it is rare for urban jurisdictions to have adequate financial resources at
their disposal to fund (larger) capital infrastructure project.
81
It should be noted that higher-level governments should not consider the provision of such transfers as altruism. In
fact, since the large majority of taxpayers live in urban areas, higher-level governments have a considerable
(revenue) stake in making sure urban areas are economically successful.
107
in a projectized manner. Instead, GOE, the World Bank (and other development partners and
stakeholders) should be thinking of a (performance-based) urban development grant as a
permanent feature of Ethiopia’s intergovernmental architecture. This means that as GOE and the
Bank look ahead to the design of ULGDP II, they should consider the second phase of ULGDP
to be the platform for the introduction of a permanent urban development grant in Ethiopia. This
represents a significant philosophical change in the role that ULGDP would play in its second
phase: rather than a temporary uptick in urban development funding to incentivize good local
government practices (as was the case in its first phase), the project’s second phase would have
to put in place the necessary mechanisms within public sector systems at different levels to
accommodate a permanent and sustainable municipal development grant.
Another consideration in the need for a permanent municipal grant mechanism is that urban
jurisdictions of different sizes have considerably different revenue structures. The analysis in
earlier chapters of this study suggests that municipal revenue collections are quite uneven, and
that many smaller urban local governments have considerably lower municipal revenue
collections (and arguably, lower revenue potential) than many of the larger urban local
governments, which have already been included in the first phase of ULGDP. The evolving
composition of the ULGDP ‘universe’ (the ULGs included in ULGDP) is an important element
that should be taken on board as the set of potential recipients of urban development grants
expands to include secondary cities.
13.2 Integrating ULGDP further into the intergovernmental finance architecture
As discussed in Chapter 2 and especially in Chapter 5, Ethiopia’s intergovernmental fiscal

architecture is driven by sizeable “unconditional” block grants, both from the federal to the
regional level as well as from the regional to the local level. While these grants are unconditional
in nature, the success of Ethiopia’s federal system is helped by the fact there are strong vertical
linkages in policy, planning and implementation, especially within the “state functions”. In
contrast, earmarked grants have received little or no attention in Ethiopia.
Since the federal government has a clear interest in achieving robust urban development
throughout the country, a strong argument could be made that a permanent, earmarked urban
development grant should be a permanent feature of the country’s public finances. ULGDP II
would be a step in the process towards such an earmarked grant. In that case, the project design
for ULGDP II would have to be informed by the vision of how such a permanent federal urban
development grant scheme would function.
If federal policy makers agree that promoting urban development is a priority of the federal level
(and legally within its purview), one implication of this argument is that we should expect the
federal government to make an incrementally larger financial contribution to such a grant
scheme, in addition to any contribution expected from the regional and local levels.
108
Box 13.2: The vertical and horizontal allocation of urban development grants
Chapter 4 and Chapter 5 of this report address the vertical and horizontal allocation or urban development
resources in Ethiopia, respectively. The relevant sections in these chapters provide guidance with respect
to the optimal allocation of urban development resources.
Analysis suggests that the proposed grant pool amount for a second phase of ULGDP ($315 million over
a five year period) should fall within a range where these resources are able to provide a meaningful
impact on urban development, while also falling within a range where municipalities are generally able to
absorb the resources. As noted in the foregoing discussions, it should be possible (given reasonable
increases in revenue efforts) for municipalities to maintain the additional investments from own sources.
However, it is not likely that municipalities would be able to enhance their own municipal revenue
collections in the near future to the point where they would be able to dedicate substantial own revenue
sources to infrastructure development at a scale comparable to their ULGDP grants.
It is recommended that the horizontal allocation of urban development grants under ULGDP II should be
based on a simple grant allocation formula, such as the current population-based allocation. It is not
recommended to use a differentiated per capita amount for “old” or “new” ULGDP cities. One suggestion
for the consideration of the World Bank and/or the GoE is whether it would be appropriate to include a
“fixed lump sum amount” in the formula in addition to the population-based allocation (and/or whether to
institute a minimum grant amount).
Next, an important question is how an earmarked urban development grant would be anchored in
Ethiopia’s budgeting and PFM processes. Although funds currently flow from MOFED through
BOFED to urban local governments, the regional level essentially acts as a disbursement agent
for the project, without any real governance ownership. Unlike the general purpose block grant
to local governments, regional states do not plan for the receipt of urban development grants as a
specific grant in their budget, nor do they plan for the provision of grants to urban local
governments in their budgets (beyond the co-funding provided by state authorities). This
significantly reduces the transparency of the public budgeting and financial management
processes. As such, the current approach of treating ULGDP in a projectized manner (following
Channel 1 funding) would not be an appropriate solution if the long-term ambition is to achieve
an urban development grant – with regional/city contributions - that is fully integrated in
government processes at all levels of government. Instead, whenever possible, the grant should
be included into the regular budget and PFM processes at the federal, regional and local level.
Two basic funding flow options seem to present themselves in this regard. One option is for the
federal government to budget for an earmarked urban development grant to ULGs as part of the
federal budget (presumably under the Ministry of Urban Development and Construction), which
would flow directly from MOFED (upon instruction from MUDC) to the accounts of the urban
local governments, thus by-passing the regional level. Urban local governments would be
informed by MUDC as to the size of their grant allocation—based on the total size of the grant
pool, formula, and their performance assessment—well before the beginning of the financial year
budget, so that they are able to include the expected grant resources as part of their revenue
109
budget (presumably under Chart of Accounts Code 1600, which covers intergovernmental
transfers).82 In turn, municipalities would plan the municipal infrastructure projects to be funded
from such a grant (in addition to their co-funding requirement) in an on-budget manner as part of
their regular planning and budget processes, while adhering to all requirements imposed as part
of the urban development grant scheme. 83 As such, the grant would be fully reflected in the
federal budget as well as at the municipal level.
The complication presented by Ethiopia’s sub-national architecture is its federal structure.

Whereas in principle, the federal government could provide municipal grants directly from the
federal level to the municipal level,84 it may be more appropriate in Ethiopia for such grants to be
routed through the regional level. In particular, to the extent that it is desired for regional states
to champion and co-fund urban development within their territorial jurisdiction, it would be
appropriate to more fully include regional states in the urban development process.
Box 13.3: Counterpart contributions for ULGDP
The requirement for regional and local counterpart contributions is an important design feature to be
considered as part of the design of ULGDP II. Compared to similar programs in other countries, it is
interesting to note that the federal government has fully passed the obligation to provide counterpart
funding on to the regional states and the municipalities, as the federal government considers its loan to be
a federal contribution. However, given that the federal government itself has an important stake in urban
development in the country, one would imagine that it could make its own—if nothing else, token—
counterpart contribution to the program. The federal government provides considerable counterpart
support for the woreda-level funding provided by the Provision of Basic Services Program (PBS II).
Under the current ULGDP arrangements there is not a stringent system in place for monitoring (as part of,
e.g., annual assessment of minimum conditions) counterpart contributions. However, selected monitoring
reports (from, e.g., World Bank supervision missions) indicate that counterpart funding has been provided
by most – but not all - regions and cities.
For the future ULGDP II (as well as for long term planning of a successor to ULGDP II) it is important to
note the following points that arose from analysis discussed in details in Part II and III of main report:
• Co-funding requirement for the larger cites is definitely achievable – however, for some of the
82
The higher-level government would presumably budget the grant under Chart of Account Expenditure Code 6400,
which includes grants to lower-level governments. Since the current intergovernmental grant system is not well
developed, there may be a need to include new domestic revenue codes for earmarked transfers under Category
1600 (for recording the receipt of earmarked grants as revenues) and Category 6400 (to record the provision of such
grants on the expenditure side of the budget).
83
It should be noted that although we are speaking here of a conditional or an “earmarked” transfer scheme, there is
no need for higher-level governments to include any detail of the specific sub-projects to be funded at the municipal
level. The higher-level budget would simply include the grant amount to be provided to the municipal level
(potentially broken down by region and/or municipality, for greater transparency). This amount would be coded as
an intergovernmental transfer rather than as a budget outlay to prevent double-counting.
84
There is nothing preventing federal governments from providing grants directly to local governments. This is not
unusual in other federal countries, such as the United States, Canada or Australia, or quasi-federal countries (such as
South Africa).
110
new and smaller urban local governments it appears as if 20% co-financing will be more
challenging. From the 12 ULG sample it is in particular Assosa city that is challenged as it will
be required to utilize almost 80% of its current own revenue for co-financing leaving little for
O&M and other services – however, as co-financing requirements are likely to trigger
improvements in revenue generation then 20% co-financing is not deemed an unreasonable
requirement even for the smaller ULGs
• Cities in general (with few exceptions) already use most of their revenues for municipal capital
expenditures. With future increased integration of the ULGDP into city financial management
systems it may be more relevant to monitor cities total spending on municipal infrastructures than
a narrow focus on specific “ULGDP co–financing”.
• Several regions have developed their own schemes for co-financing city level infrastructure
development; some of this has taken the form of “home grown” incentive based grant systems. As
the ULGDP is increasingly (also beyond ULGDP II) becoming integrated into regional budgets it
may (over time) be appropriate to take a more holistic view of regional contributions to urban
infrastructure development and also consider such schemes and contributions as part of regional
co-financing.
As such, the second general option for an urban development grant that is integrated into country
systems is for the grant to flow from the federal level through the regional level to the municipal
level. In such a configuration, the federal government would budget its municipal development
grant, which would be provided to the regional state during the next financial year; each regional
state would in turn include these grants as a revenue source on the income side of its budget,
while budgeting the onward provision of urban development grants on the expenditure side of its
budget. To the extent that each regional state would provide a financial contribution to the
program, the expenditure on urban development grants would exceed the planned earmarked
grant by its own contribution. As the final stage in the three-step process, municipalities would
plan for the grant on the revenue side of their budget, as well as on the expenditure side as noted
above.
13.3 Managing funding flows across all levels of government
The Study’s TOR asks for recommendations on adjustments (if needed) to the timing for the
flow of transfers to LGs (including LG self-assessment, deployment of the independent
assessment, verification of results and actual disbursement of funds) taking into account the
Ethiopian local and regional government budgeting cycles.
Capturing ULGDP in regional budgets. The obvious complication or disadvantage of

incorporating the regional level into a federally-funded urban development scheme 85 is that
budget planning processes would have to be aligned across all three government levels. Although
this will require a degree of coordination and advanced planning on behalf of all officials
involved, this should not be an impossible exercise if the system is designed with such
coordination in mind.
85
In this formulation, the IDA contribution is seen as a federal commitment in addition to any additional co-funding
that the government may put into the scheme directly – after all, it will be the federal level that will be repaying the
IDA commitment in due time.
111
The main advantage of including all three government levels into the implementation of an urban
development grant is that each level’s role in the system is properly incorporated. In the current
ULGDP implementation mechanism, the regional level is somewhat overlooked. While regional
states are expected to provide (and thus, budget for) co-funding for the municipal grants, they are
not asked to include the actual grants in the regional budget (either on the revenue side or on the
expenditure side). However, in practice, the grant funds do flow through regional accounts,
inviting complications in regional financial management and making it hard to track the funds
through the public sector’s regular expenditure management systems. Therefore, placing a
nascent urban development grant on-budget at all levels would allow a matching of urban
development related revenues and expenditures at all government levels: federal, regional and
local. As such, a more comprehensive approach could actually simplify—rather than
complicate—the monitoring and tracking of urban development grants as they flow through the
different government levels to reach their final destination at the local level.
Reporting, replenishment and disbursements. A related discussion is to be had as part of the

design of ULGDP II on how the disbursements of a future (performance-based) urban
development grant ought to take place. The current ULGDP grant is released in a projectized
manner, with funds being “replenished” as municipalities spend down the existing resources.
This is an administratively cumbersome approach, as municipal spending from ULGDP funds
have to be reported and consolidated at the regional level and then forwarded to the federal level.
While in practice, disbursements should take place on a quarterly basis, delays in the finalization
of the report for one municipality may actually result in the delay of quarterly disbursements for
all other municipalities in the region. This issue will only become more problematic as the
universe of ULDGP municipalities expands.
The variable municipal performance with respect to budget execution under ULGDP should be
viewed in light of the evolving practices under the first phase of ULGDP. It is not unusual for
(urban) local governments that receive development grants to carry forward relatively large
balances (as various obstacles make it difficult to spend the bulk of development resources
before the end of the financial year). For instance, during the first few years of ULGDP, grant
allocations were made based on project periods that were not aligned with Ethiopia’s fiscal year,
making the carrying forward of funds all but a necessity. Then, due to the successful absorption
of ULGDP funds in its initial years, municipalities were provided with accelerated grant
allocations, thereby “upping the ante” in terms of the ability of municipalities to absorb
resources. Several other factors complicated the ability of municipalities to absorb urban
development resources. For instance, annual performance assessments were regularly conducted
(too) late, preventing ULGs from planning their expenditures as part of the regular budget
process. It is recommended that ULGDP II should aim for a simpler, more stable, consistent and
less projectized planning and budget execution process, which should be possible under a P4R
modality.
The substantial carry-forward of ULGDP balances from year to year is not necessarily a problem
that should be prevented per se; rather, it should be seen as a fiduciary risk that should be
112
managed. 86 In particular, it is unclear whether robust financial management processes are in

place in Ethiopia to ensure proper accounting of these funds; if appropriate processes are in place
in order to ensure that developed projects which have been procured in one year are
appropriately completed and paid in the following financial year, then the carry-forward of
balances is not a major in itself. A review of this issue should be taken into account in the design
of ULGDP II.
A more streamlined disbursement process would simply disburse each municipality’s grant in
two equal, semi-annual tranches at the beginning of Quarter I and Quarter III.87 Removing the
within-year disbursement triggers would allow a much more predictable and transparent
disbursement process. To the extent that poor budget execution at the municipal level is a
concern, this could be incorporated into the annual performance assessment process; for instance,
by giving a “penalty” (reducing the upcoming year’s grant allocation) if budget performance is
below a certain threshold.88 Likewise, to the degree that the replenishment-based provision of
allocations is a mechanism to reduce fiduciary risk, to some degree, this risk could be placed at
the regional and/or municipal level. For instance, if ex post financial reports reveal that
municipal spending of ULGDP II funds failed to adhere to certain minimum standards (in terms
of adherence to certain financial management or procurement standards, for instance), municipal
and/or regional authorities may be required to offset such unaccounted-for spending. As needed,
next years’ grant disbursement to the region may be reduced by an equivalent amount.
What should the role of the BUDC be? One issue that remained largely unaddressed in the first
phase of ULGDP is: what role should regional BUDCs play?
In many ways, the regional urban development bureaus are currently the missing link in the
chain, and have a critical role. Whereas MUDC at the federal level is too far away to provide
meaningful support and guidance to ULGs, this role should be taken on at the regional
government level. BUDCs should be the champions for urban local governments at the regional
level; provide technical backstopping to ULGs; mentor local officials; monitor local government
performance on an ongoing basis; and so on.
Whereas it is best for the annual performance assessment for ULGDP II to be done by a third
party for the foreseeable future (rather than by regional officials), performance information is not
just needed as a carrot for the performance-based grant system. Instead, performance information
in various areas of performance—ranging from governance inclusiveness to revenue effort to
service delivery quality—will provide necessary information for local decision-makers as well as
the communities who have to hold those local officials accountable. As such, the key role of the
BUDCs, other than specific project-related tasks- is to promote the transparency of all aspects of
86
This is typically a problem unique to local governments, since central budget agencies are typically required to
return any outstanding financial balances to the treasury at the end of the financial year.
87
The timely disbursement by federal (and potentially, regional) authorities to the lower level could be one of the
disbursement-linked indicators (DLIs) that could incentivize effective management of the urban development grants
at the federal and/or regional levels.
88
To the degree that the budget execution weaknesses have been due to project-related delays or due to changes in
the “rule of the game” for ULGDP, municipalities should not necessarily be faulted. Indeed, a certain degree of
carry-forward of funds should be expected. Penalizing ULGs for not spending resources down fast enough may
actually result in weakening municipal PFM practices.
113
local governance within each region. This connects well to a key strategic commitment by the
MUDC under GTP to develop a performance measurement system for all cities with populations
of above 20,000 during the GTP period.89 BUDCs would become strong regional partners of
MUDC in this ambition to prepare regular and transparent performance reports. In fact, BUDCs
could be made the stewards of online regional municipal performance reporting systems, which
cover key areas of municipal operations.
13.4 Building the capacity at the local level
13.4.1 General Observations on ULG Capacities
Observations during fieldwork suggest that the capacity to perform key municipal functions
exists at a basic level, including participatory planning and identification of priorities;
procurement; financial management; asset management; operation and maintenance; and revenue
collection; and so on. There is of course variation across ULGs (as captured by the annual
performance assessments and described further in details in Part III of this report) and more
significant differences between the cities included in ULGDP and the new cities intended for
ULGDP II. However, the cities that have been supported by GIZ have already responded
significantly to the capacity support and other new cities will most likely rapidly catch up to the
basic level of capacity as witnessed under ULGDP I.
Field level consultations and review of project literature also indicate that capacity building
interventions in these areas have been useful and that they - in combination with the ULGDP
grant incentive structures – have continuously improved functional performance of the cities.
However, there is room for considerable improvement in specific areas of municipal

performance. For instance, despite the priority given to enhancement of own revenue sources in
ULGDP, the municipal capacity to pursue revenue enhancement is generally weak. Although
REPs were present in all municipalities that were visited, most REPs appeared to be prepared by
external consultants, with limited linkages between the REP documents and actual revenue
enhancement practices. Few municipalities—if any—were meaningfully computerized in terms
of revenue collection efforts. Another important problem that was uncovered during analysis
was weak and inconsistent adherence to basic revenue management practices with regard to
municipal revenues. Although the collection of state revenues attracted substantial backstopping
and oversight from the regional state level, the supervision and backstopping provided was much
weaker (and institutionally more confused) when it comes to municipal revenues. As a result,
considerable inconsistency was detected in the basic reporting of municipal revenues through
IBEX. For instance, even though municipal revenues are increasingly being tracked through
IBEX reports, it appears that different municipalities follow different practices in terms of
reporting different revenues: for instance, land rent collections from business entities may
alternatively be reported as land rent collections or as investment income. Such inconsistencies
point to the weak or unclear regulatory framework and guidance from higher levels.
89
To some degree, the performance information prepared and made public by BUDCs is less likely to be biased or
misreported, as there is no incentive funding tied to this performance report.
114
13.4.2 Past Approaches for Capacity Building of ULGs

The approach for capacity building for the cities delivered through MUDC has so far been fairly
top down and dominated by comprehensive TA packages delivered by GIZ (and in earlier stages
UI/GTZ). It should be noted that these capacity building efforts at the local level have in the past
been combined with more demand driven initiatives supported through PSCAP – for various
policy developments as well as basic financial management improvements (including use of
IBEX), restructuring of offices and departments, improved business processes, etc. However,
PSCAP has come to an end and partly for this reason, the ULGDP II concept note has set aside a
much larger amount of funding for capacity building than under ULGDP I.
Box 13.4 PSCAP Contributions to Capacity Building in ULG90:
PSCAP included several sub-programs that impacted upon ULGs. This included in particular the
following three:
Sub-program 2—District-level decentralization: The objective of the sub-program was to deepen

the devolution of power to the lower tiers of regional government and institutionalize decision-
making processes at the grassroots level. This was achieved through the following sub-
components: i) refining Woreda institutions and organizations; ii) training and Woreda planning;
iii) strengthening grassroots participation; iv) improving program development; v) building
capacity for Woreda fiscal transfer and own revenue; vi) strengthening Woreda planning and
financial control; vii) developing and implementing minimum service standards; and viii)
supporting Woredas with necessary equipment for effective service delivery.
Sub-program 4—Urban management capacity building: This sub-program aimed to enhance the
capacity of municipalities in the delivery of services, and enable urban centers to play a more
effective role in social and economic development. These objectives would be achieved through
the following sub-components: i) formulating and/or adapting enabling policy frameworks for
the improvement of urban services; ii) deepening the process of decentralization; and iii)
restructuring and capacity building of urban-local governments.
Sub-program 5—Tax systems reform: The sub-program aimed to develop a modern tax
administration that would increase tax revenues, ensure equity and fairness, and encourage
private sector investment and development. These objectives would be achieved through the
following sub-components: i) tax policy and legislation; ii) presumptive taxation; iii) taxpayers
identification number; iv) value added tax; v) operational program, systems and procedures; vi)
organization and tax payers’ education; and vii) customs reform.
Large Technical Assistance (TA) teams to cities for inclusion in ULGDP II (and not covered by
GIZ) deliver a very significant part of local level capacity building under ULGDP. The contracts
90
For details on PSCAP achievements and lessons see World Bank 2013b: (PSCAP) Implementation Completion
and Results Report. Report No: ICR00002697
115
have only recently been signed (one for Oromia and SNNP regions and one for Amhara and
Tigray). Each of the two contracts includes approximately 236 man months of technical
expertise. The main objective of the assignment is to ensure that the new cities will fulfill
minimum conditions and performance measures under ULGDP. Thus the emphasis of ULGDP
has primarily been on capacity building of new cities to facilitate their entry into the ULGDP and
its performance assessments – much less emphasis has been on gradually improving and
strengthening the practices of the cities already included under ULGDP.
GIZ has in addition to TA for cities and regions also supported strengthening of teaching of
urban management at Ethiopian University of Applied Sciences for Public Administration as
well as support for the establishment of the Ethiopian Cities Association.
One final important observation regarding the current approach to capacity building of the cities
relates to the monitoring system. At present, the main system for monitoring the impact of the
various capacity building activities at local government level under the ULGDP is the Annual
Performance Assessments (APAs). The advantage of this approach is that it monitors the actual
performance of local governments (rather than monitoring capacity building outputs such as
number of manuals developed, people trained, etc.). However, the APAs need to be
supplemented with more qualitative assessments that can explore the extent to which local
governments scoring on indicators have also led to real meaningful changes in performance.
How the APAs can be improved is further discussed below in section 13.5.
13.4.3 Future plans for capacity building of ULGs
MUDC has embarked on several plans for future capacity building of ULGs. The MUDC is
working on the development of its “Ethiopian Resilient and Green Cities Development and
Governance Programs Package”91 which describes how the Ministry intends to implement its
core mandates within the overall Ethiopian Growth and Transformation Plan. The plan describes
ten pillars of the strategy each with its own programs (job creation, capacity building and good
governance, urban land use planning, land development and management, housing development
and management, construction industry and management, integrated urban infrastructure
development, green inclusive and safer cities, urban construction and sector strategic leadership
and policy implementation, research and development). The program 2: “Capacity Building and
Good Governance” includes 6 subprograms of which “sustainable financial management and
organization” is one. Within this sub-component “revenue enhancement” is mentioned as one of
the issues to work on. The future plans of MUDC also include considerations for the
establishment of a Centre of Excellence for Urban Governance and Capacity Building92. The
Centre is intended to undertake three key functions (1) capacity building, (2) provide inputs to
policy formulation and (3) develop and maintain an urban M&E system. “Urban governance” is
interpreted broadly as governance within the core urban mandates as they fall under MUDC –
thus it includes a wide range of issues related to urban planning and service delivery but without
explicit inclusion of policy and capacity issues related to municipal revenue management.
91
MUDC Ethiopian Resilient and Green Cities Development and Governance Programs Package – unofficial draft
15 March 2013.
92
MUDC 2013: Proposal for a Centre of Excellence for Urban Governance and Capacity Building.
116
The (draft) Ethiopian Resilient and Green Cities Development and Governance Programs
Package also discuss the implementation framework where the concept of “capacity building
grants” is introduced. However, details are to be developed in the form of an operational manual.
In summary it can be noted that MUDC has a wide and demanding agenda for capacity building
of ULGs but that plans are under development for future more comprehensive and demand led
interventions. However, significant challenges should also be noted:
• Uncertainty about future support by GIZ to the urban sector. GIZ has for long been a key
partner to MUDC in development and delivery of capacity building to ULGs, but their
future engagement in the sector is uncertain because of GIZ need for focus on fewer
sectors.
• Significant need for development of systems to deliver and manage demand driven
capacity building – design of capacity building grants etc. – discussed also in section
below.
Capacity building of ULGs in the area of revenue management will require improved
institutional collaboration between MUDC, MOFED and ERCA – something that has
been initiated by the MUDC led work on the “Municipal Revenue Strategy”
13.4.4. Preliminary Recommendations for future capacity building of ULGs
Capacity building to cities under ULGDP has to date been quite centrally managed. Cities and
the regional institutions have played only a minor role in planning or management of capacity
building to-date but the regional institutions as well as the cities themselves have the potential to
take a more active role in management and delivery of capacity building activities in a future
support program. Capacity development is considered a “core process” of both regional and city
administrations but their budgets are nil or very meager. For these reasons it is recommended to
include a capacity building grant as part of ULGDP II – an approach that has worked effectively
elsewhere93.
The design of such a grant is beyond the scope of this report, but will inter alia have to consider:
• Investment menu (what areas of capacity development can ULGs spend the grant for?):
this should primarily follow the main functional areas that ULGs also are assessed for
during the APAs,
• The degree and modalities for central guidance of capacity contents in the form of
standardized training, certification of training providers etc,
• Guidance and minimum requirement for ULGs capacity development plans,
• The respective roles of MUDC, other federal institutions, the regions and cities in terms
of policy development, standard settings, guidelines, certification, planning, monitoring
93
A Market-Based Approach To Capacity Development: How Uganda’s Local Governments Are Breaking New
Ground By Mark Nelson, World Bank Institute Capacity Building Brief number 22 – 2007, For technical design of
this approaches see, e.g., PAD for the Tanzania LGSP or the Uganda LGDP II.
117
13.5 Stimulation of local governments performance by urban development grants
What institutional performance should be stimulated through the urban development grant
system and how can it best be done? Analysis indicates that the set of priority areas targeted by
the ULDGP performance grants was about right from the start, as it focused on internationally
well recognized areas of key importance for municipal performance: Planning (CIP), own source
revenue enhancement, (REP), investment utilization, asset management and O&M planning, as
well as various aspects of fiduciary management and public participation/consultation.
However, analysis (in part III of the report in particular) also points to some weaknesses of the
system. Currently the assessment system is based on many performance sub-indicators and the
assessment process is cumbersome and rarely timely. Many of these weaknesses can be corrected
if the assessment procedures and guidelines for scoring were clearly described in a manual and if
the procurement of the consultants for the assessments was made timelier.
Some of the weaknesses in the systems are more fundamental as it is not clear that they actually
achieve what is intended. For instance, the performance measurements related to achievements
of revenue targets may in fact introduce an incentive to reduce planned revenue collections,
rather than increasing actual revenue collections. Municipal revenue growth seems somewhat
limited (Chapter 6), and obstacles to revenue growth may or may not be within the control of the
local level. Thus with regard to revenues, it might be possible to monitor and incentivize
adherence to certain revenue principles (e.g., a commitment to revenue transparency; ensuring
fairness in local revenue collections by ensuring a compliance ratio of 80% or better; and so on),
rather than (just) monitoring the level of revenue collections relative to revenue plans. These
principles need, on the other hand, to be very clearly communicated to the cities in the form of
manuals and guidance posted on the MUDC website, so that urban local governments are able to
integrate the intended outcomes in their operational structures in a transformative (rather than
superficial) manner.
It is important to encourage competition between municipalities about adopting good practices –

and also set clear standards for these good practices in order to enable objective scoring of ULG
performances. It is important to look beyond the presence of a REP and an AM plan, or the
number of NGOs attending municipal meetings, but to look at the depth of interactions, and if
possible, improved outcomes. This is of course challenging and not immediately possible to
resolve. However, one issue across the board is that standards should be set higher and be more
clearly specified in the form of clear directives or “best practices” about the quality of REP, AM
plan, etc. If municipal documents do not meet the required standard, they should not be counted.
All guidance and manuals for ULGDP should be provided in a clear way on the MUDC website.
At present there are different versions of manuals in circulation (e.g., manuals for revenue
enhancement).
Part III of this report (chapters 8, 9, 10, 11 and 12) explored city management practices and how
these at present were measures in the APAs. The summary recommendations from this analysis
are outlined in the box below, whereas recommendations for regional level performance
indicators are discussed in chapter 13.6.
118
Box 13.5: Summary Recommendation of Improvement of Performance Indicators in APA
Planning: Two areas could be improved upon:

• “Participation” is currently measured rather crudely – foremost with figures on “total number of
citizen meetings” and “number of groups consulted with”. It would be unreasonable to expect the
cities continuously to report on growing number of meetings and groups. Instead it could be more
useful to establish a general standard for type (and number) of meetings and other opportunities for
citizens to air their views. This could e.g. include standards for public hearings with standards for
how agenda and other information should be shared in advance etc.
• Transparency is currently measured with regards to publication of accounts, budgets etc. However, at
present there is no clear format how such information in a meaningful way can be summarized in a
way that effectively will empower citizens. It is recommended that clear standards be developed for
such information. In addition it is recommended that the information also is made available on the
Internet. All municipalities could upload all documents required for the APA process to an online
system managed by MUDC and/or regional urban bureaus. Documents would be time-stamped when
uploaded by municipalities. This would ease the assessment process and also provide relevant
documentation into the public domain for greater transparency.
Financial management
Currently PFM issues are almost entirely assessed by use of the Audit General report – however, the audit
opinion is a very crude measurement of performance. If the APA continuously should spur cities to
improve their performance in areas related to financial management, it may be necessary to include
other/additional indicators than just the audit report. This could for instance include

• Other on-site analyses of selected financial management practices deemed critical (exclusive use of
numbered vouchers etc..)
• Analyses of adherence to good practices for internal audits (clear standards/good practice benchmarks
need to be developed),
• Submission of final accounts in due time (perhaps even as a minimum access condition)
• Indicators on follow up on audit queries.
• Independent Internal Audit functional (reporting directly to Mayor) with significant audit of capital
works (e.g. 25% of all contracts annually or other measurement of the scope of internal audits).
Asset Management and O&M:

It is recommended to sharpen the focus on actual implementation of O&M activities rather than just
mapping of infrastructure and assets. Putting more emphasis on comprehensiveness of actual O&M plan
and budget and subsequently monitoring of actual budget utilizations will sharpen the focus of assessment
more on actual O&M.
Procurement:
The assessment of procurement modalities is very detailed, but the assessment process has generally
proved feasible with useful results that have spurred cities to improve their performance. The only
problem with the current assessment system is with the indictors on “procurement audit”. These are only
undertaken as integral element of the general audits with various levels of emphasis - however most often
without separate chapters specifically thus it has been difficult for the assessment team to verify. This
indicator is not recommended to be included in scoring system until there is a common agreement with
Government on how (and if) to undertake separate procurement audits at city level.
119
Revenue enhancement:
In order to improve significantly on cities municipal revenue management it is required that the federal
government elaborate on the guidance and standards for revenue management at both regional and city
level. However, some adjustment can also immediately be made of the APAs with the above weaknesses
in mind in order to feed into the soon forthcoming APA (tentatively to be undertaken from January 2014).
The recommended indicators for city level are should focus on the two basic issues as today, but with
some modification:
Improved Revenue Enhancement Plan

This could follow the existing way of measuring its quality but inclusion of sub-indicators that are given
more significant weight to measure:
• City analysis of previous years revenue performance with detailed analyses of each main source of
revenue including discussion of its revenue potential (revision of current sub-indicator 2).
• City strategies for revenue enhancement (current sub.-indicator 6)..
Improved revenue collections and management:

• Proper coding of municipal revenue as per chart of accounts (some additional guidance on coding
should be issued),
• Percentage increase of each main category of municipal revenue (less land lease): Business taxes,
Municipal rent, Charges and fees,
• City publication of how municipal revenues has been spent in format suitable for citizen
budgets/accounts,
• Registers on revenue collection are up-to-date;
• Clear procedures and evidence on follow up on tax defaulters.
• Public consultations on tax issues especially prior to issuing of new rates and bases
• Inclusion of counterpart funding on the ULGDP/capital projects as a performance measure to promote
sustainable spending of OSR on development.
Recommended medium term revisions of the assessment
In order to incentivize more structural reforms of municipal taxes it is necessary that the Federal
Government/MUDC provide clearer guidelines and specific standards for REP than currently is the case.
The type of guidance should include:
• Guidance to regions in the form of “Model Tariff Regulations” and guidance on how regions can
assist cities to undertake relevant studies for tax reforms. It should be noted that the MUDC led
municipal tax task force already have initialed work that suggest recommendations in this direction,
• More specific guidance and standards to cities in the REP manuals regarding how cities can estimate
coverage ratio, assessment ratio and collection ration for main revenue sources (as discussed in
Chapter 6),
• Guidance to cities on coding of revenues sources in the chart of accounts and guidance to regions for
subsequent annual analysis of municipal revenues,
• Working manuals on municipal taxes including simple standard formats for computerization of tax
registers, etc. (at present this is in part included in TOR for the regional TA however it is more
appropriate to establish some national basic standards),
120
13.6. Implications for the role of regional-level institutions
The analysis Parts I – III has indicated that regional level institutions play very important roles
for the development of the urban local governments. This includes:
• Bureau of Urban Development and Construction (BUDC) is responsible guidance and

oversight to urban local governments for planning, infrastructure development as well as
several aspects of basic institutional development issues (city proclamations, etc.),
• Bureau of Finance and Economic Development (BOFED), - oversees financial
management in cities, compiles financial reports, provides capacity building/training on
financial management issues,
• Regional Revenue Authority (RRA) – oversees the revenue collections by the cities (city
revenue authority) and provides also capacity building/training on revenue management
issues,
• Regional Auditor General (ORAG): undertakes external audits of the cities and report
findings to the Regional council,
• Regional State Ethics & Anti Corruption Commission – watchdog and awareness rising.
• Regional Environmental Protection Agencies (REPAs) – ensure social and environmental
safeguards for all ULGDP investments.
• Legislative oversight from the Regional Council
These regional institutions have to a varying degree been involved in the ULGDP but under the
future program will have to be further involved and probably strengthened. This report has
indicated weaknesses in several aspects of regional involvement and support to date. These will
require strengthening and also to be monitored as part of a future DLI at regional level.
The relevant performance measures at regional level could include indicators of oversight and
capacity building support in various areas. The suggestions below are very tentative, and should
be elaborated further during the final design:
• Municipal revenue:
o Regional governments pass legislative framework for municipal finance that
facilities cities an appropriate level of autonomy in setting local tax, tariff and fee
rates suitable for each city,
o Regional oversight ensures that municipal revenues are properly coded in all
revenue reports,
o Regional government provides an online platform for sharing municipal Revenue
Enhancement Plans,
o Regional officials provide guidance, backstopping and capacity building support
for revenue planning and enhancement,
o Regional oversight ensures local adherence to fairness in revenue collections and
other good revenue practices.
• Planning:
121
o Regional officials provide guidance and ensure adherence to standards for

physical planning, master plans, cadasters, etc.
o Regional officials provide assistance and quality assurance to cities for
publication of relevant plans (CIPs; O&M plans; and so on) on a regional web-
site.
• Financial management:
o BOFED disburses grants to ULGs in a timely manner,
o Regional oversight ensures that municipal accounts are managed properly in
accordance with laws/regulations,
o Regional officials ensure that adequate internal audits are performed,
o BOFED ensures that municipal and regional budget accounts accurately reflect
municipal revenues and expenditures, including ULGDP transfers and
expenditures.
• Audits:
o Adequate and timely external audits are performed.
• Transparency and service delivery:
o Regional officials provide guidance to cities with regard to inclusive service
delivery planning, transparency of citizens’ rights to municipal services, and
support to citizens’ efforts to monitor local services,
o Establishment of basic service delivery monitoring (and publication of municipal
performance on the regional website).
All indicators for performance will have to be guided by clear federal instructions, guidelines
and standards, which provide guidance across all government levels about the respective roles
and responsibilities of each level in promoting effective, equitable and sustainable urban
infrastructure and services.
122
Annexes
Annex 1: References
Jamie Boex and Jorge Martinez-Vazquez. 2004. “The determinants of the incidence of
intergovernmental grants: A survey of the international experience”, Public Finance and
Management, Vol. IV(4), December 2004.
Boex, Jamie, and Jorge Martinez-Vazquez. 2007. “Designing Intergovernmental Equalization

Transfers with Imperfect Data: Concepts, Practices, and Lessons”. In: Challenges in the Design
of Fiscal Equalization and Intergovernmental Transfers. Jorge Martinez-Vazquez and Robert
Searle (ed.), Springer.
Boex, Jamie, Ozias Chimunuane, Minoz Hassam, Sven Hindkjær, Uri Raich, and Bernhard
Weimer. 2011. An Analysis Of Municipal Revenue Potential In Mozambique. Maputo: Banco
Mundial.
Boex, Jamie. 2010. Domestic Resources Mobilization for Poverty Reduction: The Role of Fiscal
Decentralization and Local Revenue Mobilization. Washington: The Urban Institute.
Boex, Jamie. 2012. Measuring the Local Public Sector: A Conceptual and Methodological
Framework. Washington: The Urban Institute.
Boex, Jamie. 2013. Analyzing the role of the local public sector in achieving sustainable
development. Washington: The Urban Institute.
Central Statistical Agency of Ethiopia, 2013. Population: Table B2; Table B.4.
Chanie, Paulos. 2007. “Clientelism and Ethiopia’s post-1991 Decentralisation”. Journal of

Modern African Studies 45(3): 355–384.
CIDA. 2005. Grassroots empowerment Review of Progress and Prospects Phase I.

(http://sitersources.worldbank.org/INTETHIOPIA/resources/Final_GER_MA)
Constitution of the Federal Democratic Republic of Ethiopia, 21 August 1995, available at:
http://www.refworld.org/docid/3ae6b5a84.htm. [Accessed September 2013]
Dereje Feyissa. 2006. “The Experience of Gambella Regional State”, in David Turton, ed. Ethnic
Federalism: the Ethiopian Experience in Comparative Perspective, pp. 208–230. Oxford: James
Currey.
Dickovick, Tyler, and Gebre-Egziabher, Tegegne. 2010. Comparative Assessment of

Decentralization in Africa: Ethiopia Desk Study. Washington: USAID.
Freedom House. 2010. Ethiopia Country Report. Available at:

http://www.freedomhouse.org/template.cfm?page=363&year=2010
123
Garcia, Marito and Andrew Sunil Rajkumar. 2008. “Achieving Better Service Delivery Through
Decentralization in Ethiopia”. World Bank Working Paper (Africa Human Development Series),
No. 131. Washington, D.C.: World Bank.
GOPA Consultants, July 2013. Ethiopia - Urban Governance and Decentralisation Programme
(UGDP), Municipal Finance Reform (Annual Report 2012/13).
House of Federation. 2012. The Federal Budget Grant Distribution Formula 2012/13 - 2016/17,
House of Federation Secretariat Research and Decision and Implementation Follow Up
Directorate.
Kelly, Roy. 2000. Property Taxation in East Africa: The Tale of Three Reforms, Lincoln
Institute of Land Policy, Working Paper.
Ministry of Capacity Building and Ministry of Works and Urban Development. 2008. Woreda
and City Benchmarking Survey: Citizens’ Report Card. Unpublished report.
MOFED. 2007. Manual 3 Federal Accounting System [Chart of Accounts].
MOFED. 2013. Ethiopia: Summary of Quarterly General Government Finance, Performance

2004/05-2011/12.
MUDC. 2011. ULGDP Operational Manual. (Revised November 2011).
Tegegne Gebre-Egziabher and Kassahun Birhanu. 2007. “A Literature Review of

Decentralization in Ethiopia”, in Taye Assefa and Tegegne Gebre-Egziabher, eds.,
Decentralization in Ethuopia. Addis Ababa: Forum for Social Studies.
Vaillancourt, François. 2013. Inter-Governmental Fiscal Transfer in Ethiopia. Draft report.
World Bank. 2008. Project Appraisal Document – Ethiopia Urban Local Government
Development Project.
World Bank. 2013a. Building the Developmental State – A Review and Assessment of the
Ethiopian Approach to Public Sector Reform, Report No. ACS3695.
World Bank 2013b: Implementation Completion and Results Report (Ida-38990 Ida-46690) on a
credit in the amount of (Sdr85.76 Million (Us$150 Million Equivalent) to the Federal
Democratic Republic of Ethiopia for a Public Sector Capacity Building Program Support Project
June 25, 2013 - Poverty Reduction and Economic Management 2 Country Department AFCE3 -
Africa Region - Report No: ICR00002697
Yilmaz, Serdar and Varsha Venugopal. 2008. “Local Government Discretion and Accountability
in Ethiopia”, (International Studies Program Working Paper 08-38). Atlanta: Georgia State
University. (http://aysps.gsu.edu/isp/files/ispwp0838.pdf)
124
125
Annex 2: Institutions Consulted
Federal Institutions:
MUDC
MOFED
International Organizations
World bank
GIZ
IMF
Regional Institutions
For each of the Regional States of Amhara, Tigray, Oromia, SNNP, Somali, Harari, Gambella and
Benishangul-Gumuz, the team me with representatives from:
• Regional Bureau for Urban Development

• Regional Revenue Authority
• Regional BOFED
• Regional Audit General
City level consultations
For each of the 12 cities (Adama, Jijiga, Mekele, Bahir Dar, Hawassa, Harar, Hosaena, Debre
Brehan, Zeway/Batu, Adwa, Gambella and Assosa), the team met with representatives of:
1. City Managers Office (and ULGDP team)

2. City Revenue Office
3. City Office for Finance and Economic Development (OFED)
4. Procurement staff (OFED)
5. Internal Audit (OFED)
In addition selected meetings were made at few kebele with staff involved in local revenue
management.
126
Annex 3: Tariff setting Autonomy in City regulations of 6 states
The basis for city organization and structures across the country is city proclamations enacted by
regional states and, in the cases of Addis Ababa and Dire Dawa, the federal government. The
contents of these are similar in the way they determine revenue sources and financial administration
in general terms, although some have detailed provisions while others are more concise.
There is not much detailed information on the tariff setting in the proclamations, save for the
statement that city administrations and municipalities94 have the mandate to levy, determine and
collect taxes, dues and service charges, lift taxes and impose penalties. Besides this, there is the
common designation of property and land taxes as being under the jurisdiction of city
administrations.
For this reason, reference is made below to the tariff regulations and regulations enacted by six of
the regional states.
Oromia
The main proclamation, i.e., “Urban Local Government Proclamation No. 65/2003 of the Oromia
National Regional State”, under Art. 14/2/c, gives the city councils the power to introduce, adjust
and ensure the collection of taxes and services charges, according to law.
Regulation no. 107/2008 on implementation of proc. No. 65/2003 has some interesting provisions
regarding autonomy. It states that cities designated from grade 1-4 95 can legislate on taxes and
charges/ fees based on the power given to them under Art. 14 /2 of Proc. 65/2003 on powers and
functions of city councils. 96 This power includes the mandate referred to in the above paragraph on
taxes.
With regard to the legislative procedure for this, for cities designated as grade 1 & 2, a draft must be
submitted to the Mayor’s committee, which will then submit it for approval to the council of the
city. Whereas for cities designated as 3rd and 4th grade, the draft will similarly be presented to the
Mayor’s committee, then to the Woreda council for approval. 97 Furthermore, publication of these
and availability of them in three languages is envisaged. Lastly, the amendment process follows the
same procedure. 98
A central issue which makes Oromia different from other states is: its liberal stance in providing
model tariff regulations with tariff tables for cities to adjust and use. This distinctive model tariff
law of Oromia, as described in the regulation, is prepared by the Oromia Bureau of Industry and
Urban Development. The latter can take it as a whole or adjust it for their purpose. 99 It is not clear
from the text whether a city can change the model substantially (or as a whole for that matter) or is
only allowed to make minor changes on it. Judging by the wording of the provision (of the
regulation) and lack of any proof to the contrary, the former view appears acceptable. However,
field interviews in, e.g. Ziway/Batu (Oromia) indicted that city felt that they had to stay within a
specified band guided by the revenue study conducted by Oromia. Interviews of the responsible
officer from the Revenue Office in Adama also indicated that the office was unclear about the exact
degree of autonomy to set rates.
94
There is also the emerging city category.
95
There are criteria for the categorization
96
Art. 17 of Regulation
97
It is a must that the legislations are consistent with federal and state laws.
98
See Art. 17-22 of the Regulation
99
Art. 23 of Regulation.
127
Amhara state
The Amhara state cities proclamation, i.e., Amhara state Proclamation No. 91/2003 “The Revised
Proclamation for the Establishment, Organization and Definition of Powers and Duties of Urban
Centers of the Amhara National Region State”, like all the other city proclamations, provides for a
few articles on; the general mandate of cities, sources of revenue100 and financial administrative
provisions. At the same time, the legislation empowers the Council of the regional government to
issue a regulation on levying urban service charges and taxes.101
“The Revised Amhara National Region City Administrations and Municipalities’ Revenue Title
and Tariff Determination, Council of Regional Government Regulation No.69/2002 ” contains
provisions that are relevant to the autonomy of cities in revenue matters.
It states that cities, having determined their tariff based on the objective situation of their locality,
without prejudice to the minimum and maximum ceiling tariff rates indicated parallel to revenue
sourcing titles of the tariff tables, may collect taxes, dues and service charges from those receiving
services.102
Imposition of service execution charges, taxes and dues and collection of revenue from those
receiving services shall be on the basis of those specifically indicated with regard to class and
objective situation of each city.103
Some autonomy is given in setting new service charges. In the regulation, City administrations or
municipalities having assessed their tariff rates, have the power to request service charges where
they provide various services, depending upon the objective situation of their cities.104
Besides, City administrations or municipalities are able to ask for payments from services that are
aimed at recouping capital. They, pursuant to Regulation 69/2002 Art. 2/B: as deemed necessary (as
soon as the provision of the service is commenced or if maintenance is made on the service delivery
institution), have the power to impose tariffs of service charges to be, fully or partly, refund capital,
including to maintenance, and working capital expenditures which are related to service delivering
projects on beneficiaries or adjust the rate thereof.
In connection with income to be collected from rent of resource and property of cities, the basis is
the contractual agreement they have with users, though cities shall have the power to impose ad
valorem tax/betterment levy on fixed assets, which have transferred through lease or in any
contractual agreements, as well as on urban land holding held in the rental land holding system.105
An important matter that says a lot about autonomy is the legislation process for such powers. The
procedure is for the mayor or manager of the municipality, based on the framework annexed at the
end of the regulation, to prepare their own tariff and income to be collected from rent of resource
and property, contractual agreement as well as service charge rates, and thereby submit same to
their respective city or Woreda council for approval.106
One point that is worthy of mentioning is some of the provisions on ”City Administrations and
Municipalities Finance Administration, Council of the Regional Government regulation No.
100
E.g., Art. 8/J states: as regards financial matters which fall under the jurisdiction of the city, …levy, determine and
collect taxes, dues and service charges; lift taxes and impose penalties
101
See Art. 49/2 of Proclamation No. 91/2003 “The Revised Proclamation for the Establishment, Organization and
Definition of Powers and Duties of Urban Centers of the Amhara National Region State
102
See Art. 4/1 of The Revised Amhara National Region City Administrations and Municipalities’ Revenue Title and
Tariff Determination, Council of Regional Government Regulation No.69/2002
103
See Art.5/1 of regulation.
104
Art. 2/a of regulation.
105
See Art. 3/A of regulation
106
For issues that do not include Urban land lease holding and Tax for land and building title since the Amhara tariff
rates do not include these.
128
37/2005”, which is no longer applicable, unlike the regulation above, go to some extent to explain
city tariff setting, including condition of determining tax by city administration and municipalities
and lays down principles for assessing the rate of fees and charges.
Tigray State
The autonomy of cities is dependent on “the Revised Proclamation no. 107/2006 to Provide for the
Organization and Definition of Powers and Responsibilities of Cities in the Tigray NRS”. The
proclamation confers city administrations with the power of preparation of their budget,
determination and collection of taxes and service charges, and lifting of penalties on areas reserved
to them by law. Besides, they can participate in revenue enhancement activities and receive
donations and support. The city council, in particular based on the limits set for it in the regulation,
is mandated to levy taxes and determine service charges. 107
Proclamation No. 98/2012, on the organization and Definition of Powers and responsibilities of
Mekele City, also gives the Mekele city administration the power to determine and collect taxes and
service charges on revenues that have not been allocated for the state. 108
Another law that is crucial in assessing autonomy for Tigray cities is: the Tigray National Regional
State City Administrations Finance system and Administration Regulation No. 50/2008, which lays
down the principles and considerations those city administrations have to ponder to determine taxes
and service as well as goods charges. 109This has some similarity to the Benishagul Gumuz state
tariff regulation.
The tariff rate of Tigray state is contained in an older legislation entitled “The Tigray National
Regional State Municipalities’ Trade, Professional Service and Service Tax and Tariff
Determination, Council of the Regional Government regulation 6/1997”, which does not say much
on tariff setting but has provisions on tariff assessment.
The tariffs need to be updated and there is a draft but it has not been approved.
In conclusion, Proclamation 107/2006 and Regulation No. 50/2008 seem to give a wider mandate to
cities although the tariff regulation No. 6 seems restrictive.
Benishangul Gumuz state

Typically, the state’s city proclamation: “A Proclamation to provide for the Establishment,
Organization and Determination of Powers and Duties of urban centers of the Benishangul Gumuz
Regional State Proclamation No. 69/2007”, allows city administrations to levy, determine and
collect taxes, duties and service charges, lift taxes and impose penalties within their jurisdiction.110
107
See Art. 13/2 of the revised Proclamation no. 107/2006 to Provide for the Organization and Definition of Powers and
Responsibilities of Cities in the Tigray NRS and Art. 42 lays down the details of the powers of the city administration.
It has the mandate to determine and collect: employment tax from the city (excluding federal employees), on Urban
land usage, on agricultural income tax on profit as well as excise and turn over tax on traders in the city, on urban land
rent and house tax, on income tax on rented houses and other property in the city, on stamp duty on agreements and
contracts, on profit as well as excise and turn-over tax on enterprises/corporations owned by the city administration, on
Urban road use fee, on rent from government houses and other property owned by the city administration, on royalty,
mineral income tax and land rent fees from minor road constructions, on permits and services offered by the city, on
city related taxes and service charges; and capital gains tax on property in the city.
108
Art. 10/7 of Proclamation No. 98/2012. Unlike this provision, the wording in other states gives cities taxation power
on financial matters that fall under their jurisdiction, but Proc.no. 98/2012’s wording gives the impression that residual
taxation power is reserved for the city.
109
Art. 6 & 7 of the Tigray National Regional State City Administrations Finance system and Administration
Regulation No. 50/2008
110
Art. 8/2/h of the Proclamation to provide for the Establishment, Organization and Determination of Powers and
Duties of urban centers of the Benishangul Gumuz Regional State Proclamation No. 69/2007
129
The Benishangul-Gumuz regional state urban revenue and tariff regulation no. 53/2011 is
exemplary in that it separately lays down principles for the tax administration.111 This regulation
further elucidates the powers given to city administrations in the proclamation.
An area of autonomy in the regulation is in the case of revenue sources that are not covered in the
regulation (including tariff schedules); these will be endorsed and enacted by the mayor’s
committee, municipality council or Woreda administration cabinet of emerging towns.112 However,
the administration council of the region has regulation issuing powers on levying urban service
charges and taxes.113
Another interesting issue is the authority given to Towns/cities to determine and collect tariffs
based on the current market price on rental of their own assets and property and on payments to
maintain road dug to install or maintain other infrastructures and payable per square meter.114
Other than this, there is not much information on tariff setting as such, which is separate from tax
assessment.
Harari state
In the Harari city Proclamation no. 58/2006, the city is given the power of collection and
determination of taxes, service charges and other revenues according to the law. 115
The procedure of this is: for the manager of the municipality to present the tariff amendment
proposals to the mayor’s committee, 116which has the authority to decide: on the revenues of the
city and taxes and service charges based on the law, and to give its verdict on it. This, as a matter of
procedure and division of power, is presented for approval by the City council under Art. 12/2 of
Proc. 58/2006.117
The Harari People Regional State, Harari City Municipality Revenue, and Tariff Regulation
no.28/2009 (Revised) reiterates what has been referred to in the Proclamation by stating powers of
allocation and collection.118 There is not much detail on tariff setting separate from assessment in
the regulation.
The unique nature of this state, i.e., a state, which is closely attached to a city, makes autonomy of
cities from states irrelevant to some degree.
Gambella state
The Gambella legal framework has many similarities with the Amhara municipal revenue
legislations.119 The main city proclamation of Gambella state is: “ Gambella Proclamation 73/2008
for the Establishment, Organization and Definition of Powers and Duties of Urban Centers of the
Gambella Peoples National State”. The law states: “A city administration On matters which fall
under its jurisdiction shall levy, determine and collect taxes, dues and service charges; it can also
lift taxes and impose penalties.”120
111
Art. 4 and the Gambela state tariff regulation also has similar provisions.
112
Art. 29 of the Benishangul-Gumuz regional state urban revenue and tariff regulation no. 53/2011
113
See Art. 42/2 of proclamation cited at note 16
114
Art. 24 of the regulation
115
See Art. 2/C of the Harari city Proclamation no. 58/2006
116
Art. 15/3 of the proclamation
117
It might be a case of bad translation into Amharic but the mayor’s committee seems to be given the power to
approve the revenues of the city and taxes and service charges, which might be interpreted as usurping the power of the
council. A thorough rechecking of the original provisions is needed.
118
See Art. 6/1 of The Harari people Regional State, Harari city municipality revenue and tariff regulation no.28/2009
(Revised)
119
Although the Amhara tariff rates do not include Urban land lease holding and Tax for land and building titles.
120
Article 8/2
130
What's more, “the Gambella Regional State Revised Cities’ Revenue Title and Tariff
Determination, Council of the Regional Government Regulation No. 42/2012” permits any city
administration, municipality and emerging cities in the Region to determine their tariff, (depending
on the objective situation of their locality, based on the tariff rates indicated in parallel to revenue
sourcing titles as indicated in the table attached to the regulation) and collect revenue of monthly or
yearly taxes, dues and service charges from the service seeking society.121
Another instance where similarities with the Amhara state are evident is the revenue titles of cities.
Here, the law calls for city administration, municipality and emerging cities, pursuant to this
regulation, to impose service execution charges, taxes and dues and collect revenue on the basis of
those specifically indicated with regard to class and objective situation of each city indicated in the
regulation.122
Even though there is limitation on autonomy in the above provisions, cities shall, having assessed
their tariff rates, have the power to request service charges where they provide various services. 123
One more area of autonomy that Cities or municipalities enjoy is: they can assess their own service
charge rates and can determine themselves service sectors on which they may request a charge
depending upon objective situation of their cities.124
A mandate that enables City administrations or municipalities to demand timely, commensurate
payment for service is another area that shows some autonomy. Here, they, as deemed necessary (as
soon as the provision of the service is commenced or if maintenance is made on the service delivery
institution), have the power to impose tariffs of service using charges to be, fully or partly, refund
capital, including to maintenance, and working capital expenditures which are related to service
delivering projects on beneficiaries or adjust the rate thereof. 125
With regard to income to be collected from rent or resource and property of cities or municipalities,
it shall be on the basis of any contractual agreement they make. Despite this, they shall have the
power to impose ad valorem tax/betterment levy on the fixed asset, which has transferred thereto
through lease or in any contractual agreements, as well as on urban land holding held in rental land
holding system. 126
Last but not least, identical to Benishangul-Gumuz state, for revenue sources that are not covered in
the regulation; these will be ratified by the mayor’s committee, municipality councils or Woreda
administration cabinet of emerging towns and will be applied.127
The legislative process of Gambella state is also similar to the Amhara state. The mayor or manager
of the municipality founded on the framework envisaged in the regulation will prepare their own
tariff, income to be collected from rent of resource and property, contractual agreement as well as
service charge rates and thereby submit same to their respective city or Woreda council for the
approval. 128
Conclusion
Although there is a maximum and minimum ceiling for setting tariffs for city administrations and
municipalities in Amhara and Gambella, there is some room for setting charges for: new services,
121
Part two under the title “the revenue titles of cities” and the reference to “without prejudice to the minimum and
maximum ceiling” which is found in the Amhara version of the regulation is not included here.
122
Part two under the title “the revenue titles of cities”
123
Ibid.
124
Part two under the title “the revenue titles of cities” lit. a
125
Ibid, lit. b
126
Ibid, Part two under the title “the revenue titles of cities” lit. a
127
Ibid, Part five under the title “miscellaneous” of
128
Part two under the title “the revenue titles of cities”
131
service delivering projects and betterment levy on the fixed asset as well as on urban land holding
held in the rental land holding system.
The Oromia case is different since the regional state offers model tariff ceilings as a proposal to
cities so that they can formulate (including altering the model proposal substantially) their city tariff
rates. This, obviously, is the most autonomous of the states discussed here.
The case of Tigray state is different because the regional city “Proclamation No. 107/2006
Providing for the Organization and Definition of Powers and Responsibilities of Cities in the Tigray
NRS”, and the Tigray National Regional State City Administrations Finance system and
Administration Regulation No. 50/2008 bestow wider powers to city administrations while the older
tariff regulation seems to indicate narrower autonomy.
As far as Benishagul Gumuz state is concerned, unless it is for revenue sources that are not covered
in the regulation as well as determination and collection of tariffs on rental fees of their own assets
and property and on payments to maintain roads dug to install or maintain other infrastructures,
there is limited autonomy.
On the other hand the Harari situation is a bit different because the city and state are more or less
identical which makes it difficult to talk of autonomy.
Looking at all the laws, and although there is some autonomy in setting city tariff rates there has not
been much evidence of that. Reasons such as elected officials being careful of burdening the
electorate with new tariffs so that these will not inconvenience their chances in future elections and
regional states being wary of macroeconomic implications of wider autonomy to cities are some of
the concerns raised by some towards the limited exercise of autonomy.
In general, balancing the different concerns and justifications with wider autonomy to cities calls
for a thorough understanding and analysis of the detailed issues, which will be the basis for a lucid
approach in putting in place a legal framework that grants meaningful autonomy for city
administrations.
132
Annex 3A: Comparison of selected Municipal Revenue clauses in City Proclamations and regulations of Tigray, Amhara, Oromia and Southern Nations
nationalities and Peoples states.
Amhara Proclamation 91/2003 The Revised Urban Local Government City Administrative Proclamation No 65/2003 to Provide for
Proclamation for the Establishment, Proclamation No. 65/2003 of the Proclamation of the Southern the Organization and Definition of Powers
Organization and Definition of Powers and Oromia National Regional State Nations Nationalities and and Responsibilities of Cities in the Tigray
Duties of Urban Centres of the Amhara Regulation no. 107/20008 on Peoples National Regional State NRS
National Region State implementation of proc. No. No. 51/2002 Tigray revised City Proclamation
Amhara National Regional State City 65/2003 107/2006 (EC 1998).
Administrations and Municipalities Finance
Administration, Council of Regional
Government Regulation No. 37/2005 The Tigray National Regional State
The Amhara National Regional State Municipalities’ Trade, Professional
Revised Cities’ Revenue Title and Tariff Service and Service Tax and Tariff
Determination, Council of the Regional Determination, Council of the Regional
Government Regulation No. 69/2009 Government regulation 6/1997
133
Proc. 91 Art. 49. Urban Finance and Proc. No. 65 City Finance City Finance
Budget Administration 37. Revenue Sources of the Urban 31 Revenue Sources of the Proc. No. 107
49. Sources of Income, 1. Urban Centres Local Government City 30 Revenue Sources of the City
shall have the following sources of income; 1) The urban local government may 1. A city administration may, in 1. A city government may, in close
a) charges and taxes collected from service introduce, adjust and collect taxes, close consultation with its consultation with its residents, introduce
delivery, b) block grant from Regional rentals and service charges in line residents, introduce and collect and collect new taxes not assigned to other
Government; c) Loan, d) subsidy granted by with Federal and Regional policies new taxes on items assigned to levels of government, rentals and service
the Regional State or, as deemed necessary, and laws and use the revenue for it by the regional council. It may charges and adjust existing ones.
from the concerned woreda administration; the development of the city. also determine, adjust, and 2. A city may raise funds from own
Reg. 37 Art. 5 & 6 2) Revenue from urban land and collect rentals and service income generating schemes
e) Budget subsidy to be granted by the property tax shall be the exclusive charges in line with federal and 3. The region may adopt an arrangement
Regional State for the implementation of income of the concerned urban state laws. whereby a city is entitled to receive a
special purpose; local governments. 2. A city may raise funds from portion of the revenue collected within its
f) Aid to be obtained from governmental 3-7) Other sources: Own income own income generating borders.
and non-governmental organizations either generating schemes, voluntary schemes. 4. Funds from regional government for
in kind or in cash; contributions and donations, 3. The region may adopt an works it carries on behalf of the regional
g) City land rental fee and lease revenue to subsidies, funds from regional arrangement whereby a city is government
be collected pursuant to formerly held urban government for works it carries on entitled to receive a portion of 5. Within the limits of macroeconomic
lands and lands in lease exempted towns; behalf of the regional government the revenue collected within its stability set by the federal government, a
h) Payment to be requested and collected & domestic borrowing borders. city shall have the legal competence to
pursuant to other urban land use directives; 6) With the prior approval of the 5. Subject to Government borrow from the Federal and Regional
i) Revenue to be collected as a dividend of Regional Government, urban local financial regulations, a city shall governments. It may also borrow through
development enterprises administrated governments can borrow money have the legal competence to the sale of stocks and floating bonds.
under any city administration or municipal from Federal and Regional borrow from the Federal and However a city may not take any loan
city. Governments as well as accredited Regional governments and from without the prior consent of its residents
financial institutions for capital accredited financial institutions where due to the loan sought, the annual
expenditures in observing the limits for capital expenditures. loan repayment size would exceed 25% of
of macroeconomic stability set by However, it shall secure the its annual revenue. Details shall be
the Federal Government. Details of prior consent of the concerned provided by law.
borrowing and repayment Federal and Regional 6. The Federal and Regional Governments,
procedures shall be defined by Government organs before cities and the donor community may
regulations of the Regional drawing any loan from local and together form a city fund from which cities
Executive Council. external sources. It may also may borrow. Details shall be provided by
borrow through the sale of the instrument, which establishes the fund.
stocks and floating bonds. 7. Based on the law of the regional state,
6. The Federal and Regional revenue collected from land and property
Governments, cities, and the tax shall be the income of the city.
donor community may together
form a city fund from which
cities may borrow. Details shall
be provided by the instrument,
which establishes the fund.
134
Reg. 69 Art. 4/1 Reg. 107 Art. 24/1-7 However a city may not take
Any city administration, municipality and Cities can collect taxes from: cattle any loan without the prior
emerging cities in the Region may, having markets and market stalls and consent of its residents where
determined their tariff as the objective service charges on other services, due to the loan sought, the
situation of locality, without prejudice to the kebele house rental payments, from annual loan repayment size
minimum and maximum ceiling tariff rates taxes that are set aside for would exceed 25% of its annual
indicated in parallel on revenue sourcing municipal centers in laws revenue. Details shall be
titles as indicated in the table, collect provided by law.
revenue of monthly or yearly taxes, dues 6. The Federal and Regional
and service charges from the service seeking Governments, cities and the
society. donor community may together
form a city fund from which
5/2 Cities shall, having assessed their tariff cities may borrow. Details shall
rates, have the power to request service be provided by the instrument,
charges where they provide various services which establishes the fund.
therewith. 7. Where deemed advisable, the
Government may furnish a
guarantee for loans that cities
may take from financial
institutions. Details shall be
determined by Regulations.
Autonomy in setting rates, tariffs Autonomy in setting rates, tariffs Autonomy in setting rates, Autonomy in setting rates, tariffs
Proc. No. 91 Proc. 65/2003 tariffs Proc. No. 107
Art. 8/J Art. 8 Proc. No. 51/2002 Art. 13/2
As regards financial matters that fall under Powers of urban local governments: Art.10/2/c A city administration On matters which
the jurisdiction of the city, levy, determine c) to introduce, adjust and collect A city shall introduce, adjust fall under its jurisdiction shall levy,
and collect taxes, dues and service charges; taxes and service charges under its and collect taxes and service determine and collect taxes, dues and
lift taxes and impose penalties jurisdiction in accordance with the charges subject to regional law, service charges; lift taxes and impose
Art. 50/1 every city administration or law. Art. 32/1 Every city shall have penalties. It will also be involved in own
municipality shall have the power and Art. 38/1 every urban local the power and responsibility to income generating schemes, voluntary
responsibility to organize and manage its government shall have the power organize and manage its contributions and donations.
finance. and responsibility to organize and resources. Art.31. Every city shall have the power
manage its resources. and responsibility to organize and manage
its resources.
135
Annex 3B: Comparison of selected Municipal Revenue clauses in City Proclamations and regulations of Gambela, Somali,
Benishangul Gumuz and Harari states.
Gambela Proclamation 73/2008 for the Somali Regional State Urban Benishangul-Gumuz regional Harari city Proclamation No. 58/2006
Establishment, Organization and Definition Centers Establishment, state Harari people Regional State Harari city
of Powers and Duties of Urban Centers of Organization and Definition of their urban revenue and tariff municipality revenue and tariff regulation
the Gambela Peoples National State Powers and Duties 90/2011 regulation No. 53/2011 No.28/2009 (Revised)
Gambela Regional State Revised Cities’
Revenue Title and Tariff Determination,
Council of the Regional Government
Regulation No. 42/2012
136
City Finance City Finance Revenue Sources City Finance

Proc. No.73 Proc. No 90 7. (1-16) Based on this
47) Revenue Sources of the City 51) Revenue Sources of the City Regulation, city administrations, Proc. 58 Art. 20 Revenue sources of the
a) Charges and taxes collected from service Urban centers shall have the municipalities and emerging city
delivery, following sources of income: towns shall collect revenues Property rental and tax
b) Block grant from Regional Government; a) Charges and taxes collected from from the following sources: 2. Service and usage charges
c) Loan service delivery, Commercial and professional Registration and licensing charges
d) Other sources b) Block grant from Regional service tax; Income generation programs
Reg. 42 part two Government; Urban land holding/land lease Donations and other supports
When cities provide services that are not in c) Various loans and donations and rent/tax; Loans
the tariff bands they have the power to issue d) Other sources Property valuation tax; 7. Government subsidy
new rates. 2. The council of the regional Cattle market duty; And others.
Cities can collect revenue from the government may issue a regulation Urban road use fee (excluding
following: trade and labour tax, land lease on levying urban service charges cross country roads)
and rental property evaluation, from cattle and taxes. Loading and unloading of
trade, road usage, from loading, unloading, agricultural products as well as
and sale of agricultural products, sell of same in towns;
administrative professional service, property Administrative and professional
and contract registration, abattoir services, service fees;
from production of building materials, Property and equipment rent;
advertisements, dividend tax from entities Vehicle registration and annual
that are under the municipalities, for service fees;
Environmental protection and sanitary Property and contracts
services and other city services. registration;
Abattoir service;
Production of construction
materials;
Advertisement tax;
From enterprises/corporations
owned by the city
administration and
municipalities and collected in
the form of dividend;
Environmental protection and
sanitary services; and
Revenue collected from other 137
municipal services.
Autonomy in setting rates, tariffs Autonomy in setting rates, Autonomy in setting rates, tariffs Autonomy in setting rates, tariffs
Proc. No. 73 Art.8/2 A city tariffs 6/a. Any urban center shall have the Proc. No. 58
administration On matters which fall Art. 13/8& 9 A city administration power and duty to levy and collect 7. Powers of the city
under its jurisdiction shall levy, On matters which fall under its tax; request and gather required 2/c to collect and determine taxes service
determine and collect taxes, dues and jurisdiction shall levy, determine information from the tax-payer; charges and other revenues according to
service charges; it can also lift taxes and collect taxes, dues and service 24. Rates to be Determined by the law.
and penalties. It will also be involved charges in accordance with the Market Price: Reg. No.28/2009 (Revised)
in own income generating schemes, law; lift taxes and impose Towns/cities may determine and Art. 6/1 according to this regulation the
voluntary contributions and donations. penalties. It will also be involved collect tariff based on the current municipality has the power to allocate,
Proc. No. 73. 48) Financial control in own income generating market price on the following collect, and inquire information from tax -
and Administration schemes, voluntary contributions revenue sources: payers.
1) Every city and municipal and donations. Rental of their own assets and
administration has the power to Art. 52/1 every city administration property; and
organize and administer its finances. shall have its own system, to Payments to maintain road dug to
2) Every city and municipal organize and manage its finances. install or maintain other
administration will open a bank infrastructures and payable per
account to deposit all revenues square meter.
collected based on this law and others.
It can also close this when it finds it
necessary.
138
Annex 4: Example of City Tariffs as Regulated by a Region

(Benishangul-Gumuz Regional State, Urban Revenue And Tariff Regulation, No. 53/2011)
Revenue Title
Business,
Grade of Grade Grade
Professional and Grade 2 Grade 3 Grade 4
City/town 1 5
License Revenue
Annual Service
Charge
To bars, clubs, pensions, guesthouse, conference halls, garages, fuel stations, abattoirs and related
institutions from their annual sales in percent (Whether profitable or not profitable).
Up to 100, 000 3% 3% 3% 3% 3%
Up to 300,000 /for the 2% 2% 2% 2% 2%
additional amount/
Up to 500,000 /for the 1.5% 1.5% 1.5% 1.5% 1.5%
additional amount/
Up to 1 million /for 1% 1% 1% 1% 1%
the additional
amount/
Above 1 million /for 0.25% 0.25% 0.25% 0.25% 0.25%
the additional
amount/
Minimum service 200 150 150 100 50
charge in Birr
Other businesses, professional services and all industries, service providers, manufacturing enterprises
and related institutions from their annual sales in percent (Whether profitable or not profitable).
Up to 100, 000 3% 3% 3% 3% 3%
Up to 300,000 /for the 2% 2% 2% 2% 2%
additional amount/
Up to 500,000 /for the 1.5% 1.5% 1.5% 1.5% 1.5%
additional amount/
the additional
amount/
the additional
amount/
charge in Birr
Unions, cooperatives, micro and small enterprises from their annual sales in percentage (Whether
profitable or not profitable)
Up to 100, 000 2.5% 2.5% 2.5% 2.5% 2.5%
Up to 300,000 /for the 2% 2% 2% 2% 2%
additional amount/
Up to 500,000 /for the 1.5% 1.5% 1.5% 1.5% 1.5%
additional amount/
139
Revenue Title
Business,
City/town 1 5
License Revenue
the additional
amount/
the additional
amount/
charge in Birr
Ethio-telecom, Ethiopian Electric Power Corporation, governmental and non governmental Banks and
insurance companies from their annual sales in percentage/lump sum (Whether profitable or not
profitable)
Up to 10 million Birr 0.2% 0.2% 0.2% 0.2% 0.2%
Above 10 million Birr 50,000 50,000 50,000 50,000 50,000
Minimum service charge in Birr 2,000 2,000 2,000 2,000 2,000
Contractors who undertake construction works shall pay the following percentages from the amount of
contracts they enter (Whether profitable or not profitable)
Up to 1 million Birr
0.5% 0.5% 0.5% 0.5% 0.5%
From 1 million and 1 Birr up to 10
0.2% 0.2% 0.2% 0.2% 0.2%
million Birr
From 10 million and 1 Birr up to 50
0.1% 0.1% 0.1% 0.1% 0.1%
million Birr
Above 50 million Birr 0.03% 0.03% 0.03% 0.03% 0.03%
House rent income tax for monthly
2% 2% 2% 2% 2%
rent above Birr 100, in percentage
When issuing business license up
on the establishment of a business
125.00 100.00 75.00 50.00 25.00
firm or a new business (for once)
To erect a business title board on permission per year
To attach business title on private
100.00 30.00
holding, building or fence /on one 75.00 50.00 40.00
side per sq. meter/
To attach business title on private

holding, building or fence /on both 150.00 125.00 100.00 75.00 50.00
sides per sq. meter/
To erect a business title board on
private holding, building or fence 75.00 50.00 40.00 30.00 25.00
/one side per sq. Meter/
private holding, building or fence
125.00 100.00 75.00 50.00 25.00
/both sides per sq. meter/
140
Revenue Title
Business,
City/town 1 5
License Revenue
150.00 50.00
municipal holding, /on one side per 125.00 100.00 75.00
sq. meter/
To erect a business title board
municipal holding, /on both sides 200.00 150.00 125.00 100.00 75.00
per sq. meter/
Business title posted on top of a
building /per sq. meter/ 150.00 100.00 75.00 50.00 30.00
Permission for land use for public notice board per sq. meter
Permission to erect public notice
board per sq. meter for a year only 100 75 50 40 25
on one side
Permission to erect public notice
board per sq. meter for a year on 200.00 150.00 125.00 100.00 75.00
both sides
Banner / per sq. meter per day/ 10.00 5.00 3.00 2.00 1.00
Advertisement by megaphone on
100.00 50.00 40.00 30.00 30.00
vehicle per day
Rent (service charge) for market stall on markets or places designated as market place by the
municipality
Monthly tax collected from merchants trading various commodities, agricultural products and other
/meant for profit/ by rank
1 10.00 8.00 6.00 5.00 4.00

2 9.00 7.00 5.00 4.00 3.00
To those engaged in
3 8.00 6.00 4.00 3.00 2.00
any trading activity
4 7.00 5.00 2.00 1.00 1.00
5 5.00 3.00 1.00 1.00 1.00
To those using
15.00 15.00 10.00 7.00 5.00
weighing scale
Registration charge 1 40.00 25.00 20.00 15.00 10.00

when renting a plot of 2 30.00 20.00 15.00 10.00 8.00
market area / to be 3 20.00 15.00 10.00 8.00 5.00
paid upon receiving 4 15.00 10.00 7.00 5.00 4.00
or issuing/ 5 10.00 8.00 6.00 5.00 3.00
1 20.00 15.00 10.00 7.00 5.00

To renew market
2 15.00 10.00 7.00 5.00 4.00
instalment rent
3 10.00 7.00 5.00 4.00 3.00
141
Revenue Title
Business,
City/town 1 5
License Revenue
4 10.00 5.00 4.00 3.00 3.00
5 8.00 - - -
Daily rent for mobile merchants /those who do not have permanent instalment but who sell on market
and off market days/
Merchants who trade potato, onion

3.00 3.00 2.00
and other agricultural products per 4.00 2.00
quintal on market days
Merchants who trade potato, onion
and other agricultural products per 3.00 2.50 2.00 1.50 1.00
quintal off market days
From honey, butter products and oil
seeds / sesame, incense/ per quintal 6.00 5.00 5.00 5.00 5.00
on market days
From honey, butter products and oil
seeds / sesame, incense/ per quintal 5.00 4.00 4.00 3.00 3.00
off market days
Retailers who sell goods on market
8.00 7.00 7.00 7.00 5.00
days
Retailers who sell goods off market
7.00 6.00 6.00 5.00 4.00
days
Sale of goods on a vehicle
(moving/ parked) from other region 40.00 35.00 30.00 30.00 25.00
or town.
Loading and unloading of agricultural products, various drinks and goods that enterer and leave the cities
/towns
Beer and alcohol per vehicle

175.00 150.00 125.00 100.00 75.00
Soft drinks, packed water, and
mineral water per vehicle 150.00 125 100 75 50
Agricultural products/ maize,

millet, sorghum, teff and others/
potato, onion per quintal 1.00 0.70 0.70 0.70 0.70
From honey, butter, coffee, from

oil seeds /sesame, / incense, and
spices per quintal 1.00 1.00 1.00 1.00 1.00
Different factory products and
goods from their purchase price 150.00 125.00 100.00 70.00 50.00
142
Revenue Title
Business,
City/town 1 5
License Revenue
From fuel, oil and grease retailers 0.25

per 5 litre /per kg/ 0.25 0.25 0.25 0.25
From chat per kg 2.00 2.00 2.00 2.00 2.00

Service charge collected from public transport and freight vehicles when staying the night, loading and
unloading
Annual fee for vehicles that operate in the urban centre for staying the night, loading and unloading
For vehicles with carrying capacity 250 225.00 200.00 175.00 150.00
up to 5 tons
For vehicles with carrying capacity

over 5 tons 400.00 375.00 325.00 275.00 225.00
For vehicles with trailers 500.00 475.00 425.00 375.00 325.00

Monthly fee for vehicles assigned
for a month and less for staying the
night, loading and unloading
For freight vehicles and others their
carrying capacity is up to 50
35.00 30.00 30.00 30.00 25.00
quintals
For freight and others vehicles their

carrying capacity is up to 50
75.00 70.00 60.00 50.00 30.00
quintals
For freight and other vehicles that

have trailers 130.00 125.00 100.00 75.00 50.00
Monthly fees for freight vehicles assigned for not more than six months for staying the night, loading
and unloading
For freight vehicles and others their
carrying capacity is up to 50 30.00 25.00 20.00 15.00 15.00
quintals
For freight and others vehicles their
carrying capacity is greater than 50 55.00 50.00 40.00 30.00 25.00
quintals
For freight vehicles that have
80.00 75.00 60.00 50.00 40.00
trailers
Daily fees for freight vehicles that stay temporarily
For freight vehicles and others their carrying capacity is up to 50 quintals
143
Revenue Title
Business,
City/town 1 5
License Revenue
For loading and
17.00 15.00 12.00 7.00 5.00
unloading
For staying the night 10.00 7.00 6.00 5.00 4.00
For freight and other vehicles their carrying capacity is greater than 50 quintals
For loading and

25.00 20.00 15.00 8.00 6.00
unloading
For freight vehicles that have trailers
For loading and
40.00 35.00 30.00 25.00 20.00
unloading
Loading and unloading for public transport vehicles / in urban centers that have bus stations and render
services/
Regularly assigned by
1.25 1.00 1.00 0.50 0.50
seat number
For those that are not assigned for staying the night per day
For buses 7.50 7.00 7.00 7.00 7.00

For medium buses 7.00 6.00 6.00 5.00 4.00
For mini buses 6.00 5.00 5.00 4.00 4.00
For those not regularly assigned coming for one time loading and unloading
For buses 7.50 7.00 7.00 7.00 7.00

For medium buses 7.00 6.00 6.00 5.00 4.00
For mini buses 6.00 5.00 5.00 4.00 4.00
For urban centers that don’t have bus stations
For regular users per month
For buses 125.00 120.00 100.00 75.00 50.00

For medium buses 80.00 75.00 75.00 50.00 40.00
For mini buses 35.00 30.00 25.00 20.00 15.00
For those not regularly assigned
For buses 15.00 13.00 10.00 7.00 5.00

For medium buses 13.00 10.00 6.00 5.00 4.00
For mini buses 10.00 7.00 5.00 4.00 3.00
For those not regularly assigned coming for loading and unloading passengers for one time
For buses 7.50 7.00 7.00 7.00 5.00
For medium buses 5.50 5.00 5.00 5.00 4.00
For mini buses 4.50 4.00 4.00 4.00 3.00
Rent cattle shelter /on market day per night
For cow or oxen 2.50 2.00 1.00 1.00 1.00
144
Revenue Title
Business,
City/town 1 5
License Revenue
For horse, mule or
2.50 2.00 1.00 1.00 1.00
donkey
For sheep or goat 1.25 1.00 0.50 0.40 0.30
Fees for domestic animals staying/sold and leaving through the market gate
For cow or oxen 5.00 5.00 5.00 5.00 5.00
For horse or mule 5.00 5.00 5.00 5.00 5.00
For donkey 3.00 3.00 3.00 3.00 3.00
For sheep or goat 2.00 2.00 2.00 2.00 2.00
Fees for issuing a municipality certificate for sold cattle
For oxen, bull, cow or 4.00 4.00 4.00 4.00 4.00
heifer
For horse, mule or 4.00 4.00 4.00 4.00 4.00
donkey
For sheep or goat 4.00 4.00 4.00 4.00 4.00
Charge for Sanitary Service (Monthly)
Caught red handed for disposing waste after getting advices shall be penalized as per the sanitary and
hygiene regulation of the region
Restaurants, hotels, bars, tea rooms, etc. that lack the necessary sanitary both internally and externally
after given the necessary advise shall be penalized by the sanitary and hygiene regulation of the region
Home brewed tell sellers that lack the necessary sanitary after given the necessary advice shall be
penalized by the sanitary and hygiene regulation of the region
Any resident/ organization found committing activities that could disturb the residents as well as the
environment of the urban centre shall be penalized by the sanitary and hygiene regulation of the region
Fees fixed by considering the expenses to be incurred for the collection of dead animals
For cow or oxen 15.00 14.00 13.00 12.00 10.00
For calf 12.00 10.00 10.00 8.00 8.00
For sheep or goat 6.00 5.00 5.00 5.00 5.00
For lamb or goat kid 5.00 3.00 3.00 3.00 3.00
For horse or mule 15.00 15.00 13.00 12.00 10.00
For donkey 12.00 15.00 13.00 12.00 10.00
For dog 6.00 5.00 5.00 5.00 5.00
For puppy 2.50 2.00 2.00 2.00 2.00
Penal fee for parking animals in prohibited places; for lost and found as well as for keeping and
sanitation
For ruminants 20 - 80 20 - 80 20 - 80 20 -80 20 -80
For non ruminants 30 -100 30 -100 30 -100 30 -100 30 -100
For sheep or goat 10 -50 10 -50 10 -50 10 -50 10 -50
Additional payment per day for feeding the animal up to 15 days (If the owner does not appear within 15
days, the animal shall be sold and the money becomes the property of the town/city)
For ruminants or non
10 10 10 10 10
ruminants
For sheep or goat 5 5 5 5 5
145
Revenue Title
Business,
City/town 1 5
License Revenue
Penalties for unhygienic practices will be effected per the procedures of health department
Identity card for
regulating the health 6.00 5.00 4.00 3.00 2.00
of dogs
For abattoir services & transportation and health check up
On week days
Cattle /steer ox, bull,
30 - 60 25 - 50 20 - 40 15 - 30 15 - 25
cow or heifer/
Sheep or goat 5.00 5.00 4.00 4.00 3.00
On holidays
25.00 20.00 17.00 15.00 12.00
cow or heifer/
Sheep or goat 7.50 7.00 5.00 5.00 4.00
Fees for inspection and permission of slaughtering outside slaughter house
30.00 25.00 24.00 20.00 20.00
cow or heifer/
Sheep/goat 10.00 8.00 6.00 5.00 4.00
Fees for inspection of animals slaughtered without permission in unauthorized places
Week days
40.00 35.00 30.00 20.00 20.00
cow or heifer/
For sheep and goat 7.50 6.00 5.00 4.00 3.00
On holidays
50.00 40.00 35.00 30.00 25.00
cow or heifer/
Goat/sheep 12.00 10.00 6.00 5.00 4.00
Fees for professional and labour service / fees for property valuation
The fee for property valuation by
an engineer or other employee will 1% 1% 1% 1% 1%
be in percentage, it should not
exceed 15,000 Birr and the
minimum will be 50 Birr
When an engineer or other employee is assigned for Land transfer through selling or gift
For the first time 50.00 50.00 45.00 45.00 45.00
For the second time 75.00 75.00 60.00 60.00 60.00
For the third time 100.00 100.00 80.00 80.00 80.00
If requested more than three times, fifty Birr will be added over the last fee
If the municipality is requested to measure private holdings and if an engineer or other relevant employee
is sent
For the first time 50.00 50.00 45.00 45.00 45.00
For the second time 75.00 75.00 60.00 60.00 60.00
146
Revenue Title
Business,
City/town 1 5
License Revenue
For the third time 100.00 100.00 80.00 80.00 80.00
If requested more than three times, fifty Birr will be added over the last fee
For house or fence construction permit fee to be considered from the project cost in percentage
For residential house/from the total

construction/ project cost 1% 1% 1% 1% 1%
estimation in percentage and the
minimum fee will be 100 Birr
For institutional buildings/from the
total construction/ project cost 2% 2% 2% 2% 2%
estimation in percentage and the
minimum fee will be 500 Birr
For fence / guard house building

35.00 30.00 25.00 25.00 20.00
house
Building construction supervision
/per each supervision visit/
For residential from two story to

five story
150.00 100.00 100.00 80.00 50.00
For institutional
From 1 storey – three storey

250.00 200.00 200.00 150.00 100.00
From 4 storey – five storey
350.00 300.00 300.00 250.00 200.00
Above five storey
450.00 400.00 250.00 200.00 150.00
For construction permit registration and renewal
For registration
Frome Grade 1 – 4 350.00 250.00 250.00 200.00 150.00
From Grade 5 – 7 250.00 200.00 200.00 150.00 100.00
From grade 8 - 10 150.00 50.00 50.00 50.00 50.00
For renewal
250.00 200.00 200.00 150.00
Frome Grade 1 – 4
150.00 100.00 100.00 80.00
From Grade 5 – 7
80.00 30.00 30.00 30.00
From grade 8 - 10
Cartaway on prepared
12.00 10.00 8.00 5.00 3.00
area by sq. meter
147
Revenue Title
Business,
City/town 1 5
License Revenue
Land holding
certificate and
ownership book
For first-time request 60 50.00 40.00 35.00 30.00
For second-time
75 70.00 50.00 40.00 35.00
request
For credit and
injunction registration 125 100.00 75.00 50.00 40.00
and cancellation
For the fulfillment of
bid process if a 125 100.00 75.00 50.00 40.00
professional is sent
For document
125 100.00 75.00 50.00 40.00
clarification
Photo copy service
1.00 1.00 0.75 0.50 0.50
per page
To take the standard building plan
For storey building
For three storey
375.00 350.00 350.00 300.00 240.00
building and above
For one and two
325.00 300.00 250.00 200.00 160.00
storey building
For floor
For villa 175.00 150.00 125.00 100.00 80.00
For service type
75.00 70.00 65.00 60.00 50.00
building
For the first time plan
50.00 40.00 35.00 30.00 20.00
copy
For the second time
55.00 45.00 40.00 35.00 25.00
plan copy
For the third time
75.00 70.00 65.00 60.00 35.00
plan copy
If the plan copy is requested for
more than three times, 15 Birr will
be added on the last payment
Service provided for agreement
When a house is sold or transferred 3 3 3 3 3
through gift/will, the payment for
title transfer will be in percentage
of the house value
When bank loan is taken or related 0.1% 0.1% 0.1% 0.1% 0.1%
affairs are presented, the payment
will be in percentage from the
indicated value
148
Revenue Title
Business,
City/town 1 5
License Revenue
Fees for the guarantee for acts, 0.05% 0.05% 0.05% 0.05% 0.05%
protection and registration service
will be based on the percentage of
guarantee
Public registration
To receive municipal and kebele 22.00 22.00 22.00 22.00 22.00
identification card
To register and issue marriage certificate
To register and issue
50 50 50 50 50
marriage certificate
To issue a copy of
marriage certificate 50 50 50 50 50
for the second time
To issue a copy of
marriage certificate 50 50 50 50 50
for the third time
To issue a copy of marriage 50 50 50 50 50
certificate for more than three times
10 Birr will added over the third-
time fee
Divorce registration 20.00 18.00 15.00 12.00 10.00
Issuing birth certificate
For the first time 50 50 50 50 50
For the second time 50 50 50 50 50
For the third time 50 50 50 50 50
If requested more 50 50 50 50 50
than three times
If a tent is erected on a road
For marriage, mourning, and other purposes per day
On asphalt road 30.00 30.00 25.00 20.00 15.00
On gravel road 15.00 15.00 10.00 5.00 4.00
Dumping construction 20.00 20.00 15.00 10.00 6.00
materials on road side
per day
When a road is dug for water, telephone, electricity installation and maintenance, the payment will be
calculated based on the area dug and on the current price for road construction
Asphalt
Terrazzo or cement/
sand screed
Gravel
Soil
Registration on Land book or
6.00 5.00 4.50 4.50 3.00
business permit with public register
If authentication for a copy is requested the service charge /out of a certificate/
149
Revenue Title
Business,
City/town 1 5
License Revenue
For searching a
25.00 15.00 10.00 5.00 5.00
receipt / for one time/
If a photo copy of any 10.00 5.00 5.00 5.00
other document is 10.00
requested, / for one
time per page fee/
For cart/ bicycle/ rickshaw /Bajaj/ plate permit and renewal
For plate and ownership certificate (for one time)
For cart
For government or
commercial owned 50.00 40.00 35.00 30.00 25.00
cars
For Rickshaw /Bajaj/ 150.00 100.00 75.00 50.00 40.00
Bicycle (private or
35.00 30.00 25.00 20.00 15.00
governmental)
Motor bicycle
(private or 55.00 50.00 35.00 30.00 25.00
governmental)
Annual fee
For cart 100.00 70.00 65.00 60.00 40.00
For Rickshaw /Bajaj/ 225.00 200.00 150.00 100.00 90.00
For bicycle 30.00 25.00 20.00 15.00 10.00
For motor cycle 75.00 70.00 65.00 60.00 55.00
Penalty fee when someone do not pay municipal fee
Slaughtering in
50.00 50.00 50.00 50.00 50.00
unauthorized place
Contractors based on 300.00 250.00 200.00 200.00 200.00
the contractual
amount
If tax-payer does not
pay his/her duty on
time, in percent
If one month is 10% 10% 10% 10% 10%
passed
If 2 – 4 months are 20% 20% 20% 20% 20%
passed
If 4 – 6 months are 25 % 25 % 25% 25 % 25%
passed
If above six months, the tax-payer will pay an additional 5 % over the highest fee; moreover, bank
interest will be paid for the additional time
Roof tax /To be calculated based on Annual rent in percent
"To be decided based on national legal framework which is expected to be released sometime in the
future”.
Land holding (in sq.mts)
150
Revenue Title
Business,
City/town 1 5
License Revenue
For residential
For commercial
Entertainment service providers
Individuals or institutions that are
running Digital satellite television 1200.00 1000.00 800.00 600.00 400.00
show services or Bingo games
/whether they have permit or not/
the fee will be annual
Jotoni, Pool, and Karanbula service providers
Jotoni (annual) 65.00 60.00 50.00 40.00 30.00
Pool (monthly) 65.00 60.00 50.00 40.00 30.00
Billiards (monthly) 65.00 60.00 50.00 40.00 30.00
Individual or institutions that 140.00 120.00 100.00 100.00 100.0
provide Music, Theatre and film
show services coming from or out
of the region per show
During a bazaar show, institutions 65.00 60.00 50.00 40.00 30.00
registered and performing
commercial activities will pay fee
per day
During a bazaar and trade fair, 65.00 60.00 50.00 40.00 30.00
institutions coming from or out of
the region and sell their goods will
pay fee per day
Construction material
Vehicle on a trip (8 cu.mt)
Stone 30.00 20.00 15.00 5.00 5.00
Sand/gravel 30.00 20.00 15.00 5.00 5.00
Soil/earth 20.00 15.00 10.00 5.00 5.00
By cart on a trip
Sand/gravel 3.00 3.00 2.00 2.00 1.00
Soil/earth 2.00 2.00 1.50 1.00 1.00
Parking Services /for 30 minutes 0.50 - - - -
151

Component2 LGRevenueStudy Final 1

Uploaded by

Copyright:

Available Formats

Component2 LGRevenueStudy Final 1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Component2 LGRevenueStudy Final 1

Uploaded by

Copyright:

Available Formats

Report No.

Ethiopia Local Government Revenue Study

Urban Development and Services Practice 1 (AFTU1)

Document of the World Bank

Abbreviations and Acronyms

RAP Resettlement action plan

Regional Vice President: Makhtar Diop

8.1 Policies and Guidance for Urban Planning / CIPs................................................................................................... 75

Government Fiscal Year

USD – ETB Exchange Rate

Part I: Institutional Background and Governance Context

Part II: The Fiscal Position of Local Governments

Source: MOFED (IBEX) and IMF.

Part III: City Management Practices

Planning: Two areas could be improved upon:

• Transparency is currently measured with regards to publication of accounts, budgets, etc.

• On site analyses of bank reconciliation,

Revenue enhancement: In order to improve significantly on cities municipal revenue management it

2) Improved revenue collections and management:

Sustainability of the performance based urban grant. Based on an analysis of intergovernmental

1.4 Organization of this Report

The report is divided into four main parts:

Part I: Institutional Background and Governance Context

Part II: The Fiscal Position of Local Governments

Part III: City Management Practices

PART I: INSTITUTIONAL AND GOVERNANCE CONTEXT

2. Overview of the institutional relationships between federal,

2.1 Overview of the intergovernmental institutional structure of the public sector

The Constitution defines the federal-

In accordance with the constitutional

“Interdependent (or collaborative) federalism”. Although Ethiopia’s intergovernmental system

Table 2.1. Regional population and urban population (July 2013)

2.2. An overview of functional assignments and local regulatory powers

Following Article 50 of the Constitution, municipalities or “city administrations” are recognized

2.3 Overview of the intergovernmental administrative structure of the public sector

as opposed to sub-national budget allocations, and sub-national (especially regional) councils

Administrative interdependence in financial management. As already noted, in Ethiopia’s system

Consolidated government accounts focus on expenditures and revenues, not on intergovernmental

3. Relations between regional and city governments

3.1 De jure autonomy of urban local governments versus de facto interdependence

Legal and Institutional Framework for Cities

participation, democratization and enhance decentralized service delivery through institutional

3.2 Regional and City/Municipal legislation on municipal revenues

Table 3.1 Degree of Autonomy for City Management of Revenues

Region Degree of autonomy set by City Tariff Regulations

Tigray Tariffs and rates are set by region – no autonomy

3.3 Regional Oversight and Support for City Financial Management

3.4 Regional Oversight Over Urban Development

3.5 Regional Auditing of Urban Local Governments Accounts

The following audits of local governments were included in the sample:

The scope, comprehensiveness and frequency of audit

Compliance with audit standards

Supervision and review

Reporting standards and reports

Interaction and audit follow-up

Legislative scrutiny of external audit reports

The capacity of the external audit unit

PART II: THE FISCAL POSITION OF LOCAL GOVERNMENTS

4. An overview of the intergovernmental fiscal system and sub-

4.1 Local Public Sector Expenditure Profile