final report plot 10, saddler lane, naguru p.o. box 8699...

STUDY ON THE BUDGET ALLOC ATION FORMUL A

F I N A L R E P O R T

Plot 10, Saddler Lane, Naguru

P.O. Box 8699 Kampala www.ahcul.com

2012

http://www.ahcul.com/

STUDY ON THE BUDGET ALLOCATION FORMULA FINAL REPORT

2

TABLE OF CONTENTS ACRONYMS AND ABBREVIATIONS 4

LIST OF TABLES 5

LIST OF FIGURES 6

EXECUTIVE SUMMARY 7

1. INTRODUCTION 17

1.1 Problem ....................................................................................................................................... 17

1.2 Background ................................................................................................................................. 17

1.3 Scope and Objectives ................................................................................................................... 18

1.4 Document Overview .................................................................................................................... 19

2. STUDY METHODOLOGY 20

2.1. Inception Phase ........................................................................................................................... 20

2.2. Literature Review Phase .............................................................................................................. 21

2.3. Formula Development Phase ....................................................................................................... 21

2.4. Reporting Phase .......................................................................................................................... 22

3. ROAD MAINTENANCE FINANCING REVIEW 23

3.1. Sources of Road Maintenance Financing ..................................................................................... 23

3.2. Historical Assessment of Potential Road Maintenance Financing ................................................ 27

3.3. A review of current and historical Road Financing ....................................................................... 28

4. ROAD MAINTENANCE NEEDS ASSESSMENT 30

4.1. Highway Development and Management (HDM-4) Tool ............................................................. 30

4.2. Available Information for the Uganda Road Network .................................................................. 31

4.3. Uganda Road Network Data ........................................................................................................ 32

4.4. Maintenance Standards and Unit Costs ....................................................................................... 40

4.5. Analysis Results – Road Maintenance Needs and Funding Gap .................................................... 45

5. A REVIEW OF THE CURRENT FUND ALLOCATION APPROACH USED BY THE UGANDA ROAD FUND 47

5.1. Current Approach used in Budget Allocation ............................................................................... 47

5.2. Critique of the Current Allocation Formulae ................................................................................ 51

6. ROAD MAINTENANCE FUND ALLOCATION FORMULA DEVELOPMENT METHODOLOGY 53

6.1. Introduction ................................................................................................................................ 53

6.2. Road User Charging ..................................................................................................................... 53

6.3. Incorrect Objectives and Principles for Allocation ....................................................................... 54

6.4. Factors that Influence Road Maintenance Cost............................................................................ 55

6.5. The Formula Development Methodology .................................................................................... 56

7. THE ROAD MAINTENANCE FUND ALLOCATION FORMULA 63

7.1. Formula Development ................................................................................................................. 63

7.2. Stage 1: Allocation to Road Surface Types ................................................................................... 63

7.3. Stage 2: Allocation to Road Networks ......................................................................................... 66


3

7.4. Stage 3: Allocation to Designated Agencies ................................................................................. 67

7.5. Formula Coefficients and Factors ................................................................................................. 71

7.6. Operationalization of the Allocation Formula .............................................................................. 78

7.7. Formula Coefficients and Factors ................................................................................................. 81

8. APPLICATION OF THE FORMULA 82

8.1. Introduction ................................................................................................................................ 82

8.2. Worked Example ......................................................................................................................... 82

8.3. Sensitivity Analysis ...................................................................................................................... 85

8.4. Discussion of Results ................................................................................................................... 86

8.5. Road Map for Transition from Formula Based Approach to Needs Based Approach .................... 86

9. CONCLUSIONS AND RECOMMENDATIONS 89

REFERENCES 91

APPENDICES 93

APPENDIX 1 – ASSIGNMENT TERMS OF REFERENCE 94

APPENDIX 2 – LIST OF KEY STAKEHOLDER MET 104

APPENDIX 3 – VEHICLE FLEET CHARACTERISTICS 105

APPENDIX 4 – MoWHC (2006) STRATEGIC ANALYSIS USING HDM-4 107

APPENDIX 5 – IT TRANSPORT (2011) 109

APPENDIX 6 – TYPICAL ROAD DETERIORATION RATES 123


4

ACRONYMS AND ABBREVIATIONS

AHC – AH Consulting AREP – Annual Road Expenditure Programme ARMP – Annual Road Maintenance Programme CAR – Community Access Roads DUCAR – District, Urban and Community Access Roads FMC – Fund Management Committee FY – Financial Year HDM-4 - Highway Development and Management Tool KCCA - Kampala City Council Authority MoLG – Ministry of Local Government MOWT – Ministry of Works and Transport MOFPED – Ministry of Finance, Planning and Economic Development MTEF - Medium Term Expenditure Framework NPV - Net Present Value OYRMP – One Year Road Maintenance Plan PFM – Public Financial Management RSDP - Road Sector Development Plan RUCs - Road User Charges SWG – Sector Working Group UNRA – Uganda National Roads Authority URA – Uganda Revenue Authority URF – Uganda Road Fund


5

LIST OF TABLES Table 1: Fuel charges per product 24

Table 2: Annual revenue from Fuel (in billions of Uganda Shillings) 27

Table 3: Percentage total revenue contribution by fuel charge type per year 27

Table 4: Current financing available for road maintenance per year (in billions of UShs) 28

Table 5: URF allocations as a share of total collections (in billions of Ushs) 28

Table 6: Fund allocation to the Road sector from an MTEF perspective (in billions of UShs) 29

Table 7: Uganda Road Network Length (in kilometers) 33

Table 8: Summary of National Road Network Characteristics 34

Table 9: Summary of District Road Network Characteristics 35

Table 10: Summary of Community Access Road Network Characteristics 36

Table 11: Summary of Urban Road Network Characteristics 37

Table 12: Summary of Municipal Road Network Characteristics 38

Table 13: Average 2011 Daily Traffic on the Uganda Road Network (in vehicles / day) 39

Table 14: Traffic Composition, PCU and ESALF 39

Table 15: Road Maintenance Work Types and Unit Costs 41

Table 16: Paved Roads Maintenance Standards 42

Table 17: Gravel Roads Maintenance Standard 43

Table 18: Earth Roads Maintenance Standard 44

Table 19: Annualised Unconstrained Maintenance Needs (in millions of US $) 45

Table 20: Uganda Road Fund Allocations for the 2011/2012 Financial Year (in millions US $) 46

Table 21: Global allocation of Funds by URF for the 2010/11 Financial Year 48

Table 22: 2010/11 Percentage Allocations of funds to the DUCAR Network 49

Table 23: Global allocation of Funds by URF for the 2011/12 Financial Year 49

Table 24: Ride Quality Threshold Values for Paved and Unpaved Roads 59

Table 25: Typical Road Deterioration Rates by Rainfall Amount 75

Table 26: Variable Input Data Sources 75

Table 27: Calibration Approaches to Formulae Coefficients and Factors 76

Table 28: Stage 1 Input Assumptions 82

Table 29: Stage 1 Relative Allocations 83

Table 30: Stage 2 Inputs 84

Table 31: Stage 2 Allocation by Road Network in Billion Ushs 84

Table 32: Road Map for Transition from Formula Based to Needs Based Approach 87


6

LIST OF FIGURES Figure 1: Key URF Institutional Stakeholders (Source: URF OYRMP & AREP 2010/11) 20

Figure 2: Current URF Formula for allocation of funds to District Roads 47

Figure 3: Current URF Formula for allocation of funds to Urban Roads 47

Figure 4: Current URF Formula for allocation of funds to Community Access Roads 48

Figure 5: Flow Chart – The three stage allocation structure 58

Figure 6: The effect of traffic and road length on gravel road maintenance cost 62

Figure 7: Importance of National Objectives 73

Figure 8: Scores of National Objectives by Road Network 73

Figure 9: Annual Rainfall (mm) Map of Uganda 74

Figure 10: Overall Allocation Process 79

Figure 11: Relative Allocation by Road Surface Type (Stage 1) 83

Figure 12: Relative Allocation at Stage 2 by Road Network 85

Figure 13: Sensitivity of Key Input Variables 86


7

EXECUTIVE SUMMARY Introduction Road maintenance needs in Uganda are huge and available resources are not sufficient to allow all desired maintenance activities and interventions to be carried out. In any given planning period it is therefore important to consider the financial resource constraints through prioritization with the goal of achieving maximum impact and value-for-money on road maintenance expenditure. At present, the Uganda Road Fund (URF) allocates proportions of the total available road maintenance budget to road classes based on an approach that has been used rather historically, from the time Ministry of Finance, Planning and Economic Development (MoFPED) that is not objective or satisfactory in the eyes of many stakeholders in the road sector. The Uganda Road Fund requires relevant and up to date information in a variety of forms in order to analyse and to justify allocation of road maintenance funds to the different road classes and road agencies. More objectively, road fund allocation methods can be classified into two broad frameworks as Formula based and Needs based. The formula based method is applied by fixed ratios or indices determined through negotiations, consultations or mathematical formulae. The allocation by consultations and negotiations is subjective and could result in biases. Under the Needs based method, the fund allocation approach is determined by the cost of treatment works for pavement deficiencies. A critical review of the features of these two broad frameworks is provided in the Literature Review Report prepared during Phase 2 of this Study. The primary aim of this study is to develop mathematical formulae for use by URF to allocate road maintenance funds to different road classes and road agencies. This document is the Final Report of the Study. Scope of Services and Study Objectives The overall objective of the Study is to give effect to Section 22(2) of the Uganda Road Fund Act, 2008, which states that:

“Allocations from the Fund to the designated agencies shall be based among other factors, on the conditions of the public roads, maintenance requirements, the length of the road network and the relevant volume of traffic or derived from an approved maintenance management tool”

The specific objectives of the assignment were to:

(a) Examine the current road administration of Uganda and analyse the historic trend of

allocation of road maintenance resources and the relative impacts thereof;


8

(b) Carry out detailed analysis of designated agencies to develop a proposed hierarchy and pattern of distribution among road classes and designated agencies, identifying common influencing factors on the cost of maintenance of roads;

(c) Review the current practices of road funds in the region and elsewhere in allocation of funds to road maintenance, with emphasis on the transferability of such practices;

(d) Develop, recommend and demonstrate robustness of an equitable, transparent, fair and justifiable approach to allocation of road maintenance resources, vertically between road classes and horizontally between designated agencies responsible for each class of road;

(e) Based on the state of reform of the road sub-sector and the likely pace of development of institutional, technical and financial capacity of designated agencies; propose a road map for transition from an indirect formula – based approach to a needs based approach.

Study Methodology The assignment was carried out over the following key phases:

i. Inception Phase ii. Literature review iii. Formula development iv. Reporting (including presentation to FMC, SWG and other key stakeholders).

The Inception Phase set the direction of the Study. Tasks performed in this phase were the following:

Review of available reports and data on institutional arrangements for road administration at national and local levels, governance issues, the legal framework underpinning the sector among others.

Meetings with the URF Executive Director and with the URF Manager Programming, Manager Fund Management and key URF staff associated with the Study;

Formulation of a work plan within the allowable budget and duration of the study; and

Presentation of the results of the Study in the form of a written report and audio-visually to the URF Team

In the literature review phase, the consulting team reviewed a number of available literature on the subject of budget allocation in the roads sub-sector. Allocation approaches in other sectors where direct measurement of need is not practicable were also reviewed for commonality of approach. The key tasks carried out under this phase were the following:

Acquisition of existing literature on fund allocation from all available sources, especially relating to other road funds in Sub-Saharan Africa;


9

Review of literature to identify and critically review the approaches used by other road funds regionally and internationally;

Review of similar indirect allocation approaches used in other sectors, both in Uganda and elsewhere;

Review current allocation formula of URF and methods used by MoFPED, MoWT and key Donor Agencies in similar funds disbursed countrywide;

Visits to other regional road funds to establish the evolution of their allocation approaches and to identify particular successes and failures;

Analysis of data, with particular emphasis on extraction of a set of most commonly used factors and proxies for estimation of road maintenance needs including but not limited to Geographic and environmental, traffic, condition related and demographic factors; and

Preparation and presentation of the Literature Review Report to the Uganda Road Fund team.

In the formula development phase, the Consultant developed an approach to fund allocation that is robust and defensible taking care of the following key principles:

Equitable sharing of funds between road classes and designated agencies;

Since the user pays, the allocation shall be fairly representative of road usage;

On that basis, the maintenance of the national road network shall be fully funded subject to demonstration of value for money and allocated on the basis of a road management system;

Funds shall be allocated to give the optimum effect in terms of road condition;

Since traffic levels on district and community access roads are generally low, the social benefit of these roads shall be recognised by a significant contribution to their upkeep from the consolidated fund or other charges and that shall be a condition of Road Fund support, lest the money allocated be wasted on inadequate maintenance. The need for counterpart funding shall also be given consideration here;

The formula shall as much as possible, reflect the policies, plans and strategies of the Government of Uganda relating to the road sub-sector and also to general development (as reflected in the current National Development Plan and in Millennium Development Goals); and

The validity of the formulae has been tested and verified by application to a sample of agencies.

Key Findings


10

Literature Optimal fund allocation is required both in situations with funding shortfalls and in situations with adequate funding. This is because offering a lot of resources does not necessarily mean fulfilling different points on the risk reward spectrum and misallocation could result in wastage. Limited resources also have to be maximised for optimal returns. For best practice requirements, the Federal Highways Administration of USA, (1999) recommends for the decision analysis for road fund allocation to include; strategic goal, quantitative factors, performance measures, performance prediction capabilities, qualitative issues and link to the budget process using engineering and economic analytical tools. Heggie (1995) defines five principles for road fund allocation as needs based, ensuring economic efficiency, equity, transparency, fairness and simplicity. Howard (1983) suggests for the process to be divided into smaller components for logical analyses to arrive at reasonable conclusions. These issues have been discussed in this report. The review of approaches used in comparator countries shows a wide spectrum of options. The Namibia Road Fund Administration allocates road maintenance funds under the framework of the Namibia Road Master Plan and prioritises the roads to be worked on by agency with the help of data from the utilised the Highway Development and Management tool (HDM-4) and the Road Management System. The allocations are discussed and agreed with stakeholders annually. The Ghana Road Fund also uses a consultative approach in allocating funds to agencies but with the use of data derived from various decision support tools i.e. the Pavement Management Program for trunk roads, the Maintenance Management System for urban roads and the Maintenance Performance Budgeting System for feeder roads. The USA system is based on objective measures of system extent and usage, notably mileage and vehicular travel. Also due to varied geographical and economic characteristics efficiency is put forth as the primary determinant in fund allocation but equity appears to be a much more significant consideration. The New Zealand road fund adopts the needs assessment approach to road fund allocation complemented with economic prioritization. The road network condition is assessed to determine the structural integrity and defective sections through condition surveys for diagnosing appropriate interventions for costing. Although economic growth through efficient transportation remains a long term goal which clearly requires a Needs Based Approach, short term objectives should include the protection of the current road infrastructure in a way that is equitable to the road user and the designated road agencies. It should furthermore be accepted that limited information is available on actual road conditions and needs, but this should not prohibit the formulation of a sensible approach in the short term. Road maintenance financing


11

Uganda Revenue Authority collects road user charges from the key revenue sources of Petroleum Duty, Temporary Road Licences (International Transit Fees) and Fees collected under the Traffic Act. Over the past 5 years, the fees collected have been increasing at an annual growth rate of between 26% and 14% giving an average annual growth rate of 18%. From a share perspective, the Petroleum Duty accounted for an average of 86% of the Total fees collected by URA while Fees under the traffic Act accounted for 10% and International Transit fees 4%. A review of the current and historical funding available for road maintenance reveals an increase in funding from UShs.85.73 billion in 2006/07 to UUhs.277.15 billion in 2010/11, the first financial year that the Uganda Road Fund became operational. Over the past 5 years, Road User Charges far outstripped the Road Maintenance Allocations from the Consolidated Fund by an average of 30% (varying annually between 19% and 36%). This is indicative of potential increased revenues for the purpose of the maintenance of roads should the overall sum be allocated to the Road Fund as is expected (i.e. assuming the full component of the Road User Charges are allocated). From a Medium Term Expenditure (MTEF) point of view, the funding allocated to the Sector is expected to grow from the 2011/12 allocated UShs.280.95 billion by an annual 22% factor to UShs.341.95 billion in 2012/13 and to UShs.416 billion in 2013/14. It is worth noting though that while the ceiling for vote 118: Uganda Road Fund of Shs.280.95bn is what was planned for, this ceiling was not based on the established needs of the designated agencies as generated by the URF Board for the 2011/12 Financial Year which were estimated at Shs.413.94 billion in effect giving a shortfall of Shs.133 billion. This report also provides a comparison of the current funding to the maintenance needs of the network. It is critical that the fund quickly finalizes all legal reforms to allow the collection and direct remittance of Road Users Charges (RUCs) especially fuel levy to a URF account that should be maintained with Bank of Uganda. Under this new proposed framework, the URF Board should be able to independently recommend RUCs to Minister for Finance as required by law based on established needs of agencies. Road maintenance needs HDM-4 was used to investigate the maintenance needs of the Uganda road network assuming no constraints in budget given the standards defined above. This analysis scenario results in an unconstrained road agency costs based on a life cycle analysis that considers the different investment alternatives for each representative road section. The alternative giving the highest NPV is assigned to each section. The road works assigned by HDM-4 represents the optimum maintenance that should be applied to the road network in accordance with the specified standards.


12

The unconstrained maintenance needs have been annualized over three ranges of years 1 to 5, 1 to 10 and 1 to 20. Overall, the required annualized maintenance expenditures (excluding reconstruction or major improvements) are US$ 289 million over years 1-5, US$ 258 million over years 1-10, US$ 245 million over years 1-20. The required unconstrained road maintenance cost was estimated to range between US$ 245 million per year to US$ 289 million per year. This, in comparison to the maintenance funding of US$ 112.66 million based on the road fund allocations for the 2011/2012 financial year shows a significant funding gap that ranges between US$ 132 million to US$ 176 million per year, given the assumptions used in the analysis. Critique of current allocation formula The current road fund allocation formula considers both vertical allocation by road classes and horizontal allocation by road agencies. The vertical allocation process firstly allocates proportions of the total budget to road classes on a basis that seeks to fund maintenance of national roads as fully as possible and allocates remaining funds to Districts, Urban and Community Access Roads on the basis of formulae similar to those adopted by the Ministry of Finance hitherto. The main criticisms of this formula are the following: (i) It assumes that the budget estimate for national roads is properly estimated in an

objective manner based on an analysis that considers needs and reflects efficiency. It is not certain if this is true?

(ii) Seeks to fund maintenance of national roads “as fully as possible”. It is not clear how this level of funding is decided or determined as a proportion of the total available budget.

(iii) The approach does not take into account the effects of funding level on the future road condition and road user effects. Therefore the Road Fund is not certain on whether or not the money is contributing effectively to the improvement of the road network. Thus the allocation is not transparent.

The horizontal allocation is characterised by the following:

(i) It is good that the formula takes into account some element of equity in terms of the

uniform allocation.

(ii) The districts road formula does not explicitly consider the extent of the road network in any particular district. A district can be big in size, but the road length density may be low. This would cause bias in fund allocation.

(iii) The urban road formula may be biased against the new or more rapidly expanding urban areas since the formulae includes “previous asphalt” component.


13

(iv) Although population is used in the formulae as a proxy for road utilization, in Uganda it is important to consider income level of the population in order to assess the affordability of using motorized transport by the local population.

Formula Development The development of the proposed road maintenance allocation formula is based on the following general principles for road user charging:

Principle of full cost recovery;

Principle of economic efficiency;

Principle of equity.

A full explanation of these principles is given by Heggie (1995). A three-stage allocation structure is proposed to consist of the following broad process:

Stage 1 – allocates available funding to surface types;

Stage 2 – allocates the funding per surface type between the road network jurisdictions;

Stage 3 – allocates the funding per road network and surface type to the designated authorities within each district, town councils, municipals, and sub-counties

There are three road surface types considered (i.e. paved, gravel and earth); five road network jurisdictions (i.e. National, District, Town Council, Municipal and Community). There are 111 districts, 174 town councils, and 23 municipalities including Kampala City Authority. The proposed road fund allocation formula is based on the following criteria depending on the allocation stage:

Stage 1 1. Vehicle utilization – to reflect on the principle of user pays. The allocation parameters

considered here are:

traffic volume measured in passenger car unit kilometers (PCU-Km), and

traffic loading in equivalent standard axle load kilometers (ESAL-Km) 2. Road length – to reflect maintenance requirements in terms of extent/size. This also takes

into account the effect of non-traffic related elements on road deterioration (e.g. environmental factors) and routine maintenance needs.

3. Road asset value – to reflect the investment that needs to be preserved in the interest of

national economy and wealth generation.


14

4. Optimum maintenance requirements - to reflect the fact that different road surface types and standards require different levels of funding to maintain it to an acceptable condition. This is expressed in terms of the ratios of optimum maintenance cost for each road surface type. Road surface types that require higher funding for their optimal level of maintenance should receive higher allocations.

Stage 2 1. Vehicle utilization – to reflect on the principle of full cost recovery that maintenance cost of

roads should be recovered from road users. The most trafficked roads generate most revenue for the road fund and conversely should get higher allocations. The parameters considered are:


traffic loading in equivalent standard axle load kilometers (ESAL-Km) 2. Road functional class priority and the key role of supporting the most highly ranked national

objective of promoting economic growth and wealth creation. This is expressed in terms of relative weights for the different road networks (i.e. the designated road agencies). Road classes with higher relative weight should receive higher allocations.

3. Other sources of road maintenance funding – road agencies or networks that possess other

sources of raising funds (e.g. through vehicle parking charges, taxation, etc.) should receive less allocations.

Stage 3 For Urban, Municipal and District networks the parameters considered are the following: 1. Vehicle utilization – the most trafficked roads deteriorate faster thus generating higher

maintenance needs therefore it is essential that these get higher allocations. The parameters considered are:


traffic loading in equivalent standard axle load kilometers (ESAL-Km) 2. Road length – to reflect maintenance requirements in terms of extent of the road network.

Designated authorities with bigger network size should receive higher allocations. 3. Equity – to reflect social concerns and needs of the population regardless of which area they

live in. Every designated authority should receive a minimum fixed amount of funding expressed as a percentage of the available fund for each network (urban, municipal or district).

4. Climatic factors - rainfall is a major climatic factor that influences road deterioration. Roads

in areas with high rainfall will deteriorate faster resulting into more frequent and higher


15

maintenance requirements. These areas should therefore get proportionally higher allocations.

5. Variations in unit costs of road works across the country. There are several factors that lead

to variations in unit costs of works and this influences the total amount of road maintenance fund requirements by designated authorities. Designated authorities with high unit costs of works should get higher allocations, on average.

For Community and Access Roads the parameters considered include all those given above for urban, municipal and district networks but excluding vehicle utilization since all these roads are carry very low traffic volumes with little variations. Another important parameter considered for CAR maintenance fund allocation is population of the area. Population is considered a surrogate for travel demand and needs for social services. Areas with big population should get higher allocations. The allocation procedure requires that the input data, including the available fund to be allocated, should be determined and provided at the start of the process. Stage 1 is a vertical allocation to the different road surface types that is performed using the formula developed. Then Stage 2 allocation to the different road networks (i.e. horizontal allocation) is performed. After this stage, the results obtained will be compared (e.g. with road network maintenance needs) to assess whether or not these are satisfactory. If the outcome appears to be unsatisfactory then some fine-adjustments need to be made on the values of the model coefficient and factors and the allocation process repeated from Stage 1 to Stage 2. If the results obtained after Stage 2 are satisfactory then the allocation process continues to Stage 3. Application of the Formula Developed To demonstrate the application of the formula developed, a worked example has been presented to allocate an assumed available budget of UShs 400 billion. The results indicated that for vertical allocation to road surface types, paved roads should get 54 percent, gravel roads 40 percent and earth roads 6 percent. For horizontal allocation to road network and designated agencies, UNRA should get 56 percent, District 24 percent, Urban 15 percent, and CAR 4 percent. This can be compared with the 2010/2011 allocation as follows: UNRA should get 67 percent, District 14 percent, Urban 16 percent, and CAR 3 percent. Conclusions and Recommendations Based on the road maintenance finance review, the URF is significantly underfunded with a current allocation of 30% of total URA collections from road user charges to road maintenance work annually. The current approach for budget estimates at the designated agencies is constrained by the ceilings given to the URF by the MOFPED which limits the best approach (Bottom-up approach)


16

where the needs are determined each year and funding sought to meet the road maintenance needs. The actual road maintenance needs derived from HDM-4 depicts a gap in relation to the actual available budget from MOFPED creating a funding gap which can only be closed by an increase in the total funding allocated to URF by MOFPED. The current formula in use by URF has a number of weaknesses as outlined above. In order for the URF to operate efficiently and meet the required road maintenance demand the current funding should be boosted by a 58% markup moving it from 30% allocation from MOFPED to 88% allocation of total URA Collections under road user charges which was UShs 885.28 billion in FY 2010/11. Based on the road maintenance needs assessment undertaken using the HDM-4 tool the actual road maintenance needs for the first 5 years annualized across each of these years is US$289 million, for a 10 year band the annuity would reduce to US$ 258 million, and finally an annualized US$ 245 million for a 20 year spread if considered in order to deliver the required road standards, condition targets, and needs. It is recommended that URF adopts the proposed “Three-Stage Allocation Formula” which addresses the weaknesses with the current formula in use. In order to obtain more accurate allocations the URF will have to update their database and obtain latest input data on the factors that impact on the formula. These factors are explained in the body of the report. The rationale followed to derive the study conclusion and recommendation is outlined in the body of the report.


17

1. INTRODUCTION 1.1 Problem

Ideally, road maintenance funds should be allocated as and when maintenance requirements fall due. However this cannot often be achieved due to limited availability of road maintenance funds, data on road condition and maintenance requirements on a detailed and ongoing basis.

At present, the Uganda Road Fund allocates proportions of the total available road maintenance budget to road classes on a basis that seeks to fund maintenance of national roads as fully as possible while allocating the remaining funds to district, urban and community access roads on the basis of formulae. The basis of these formulae is population of the target districts, previous asphalt and surface area.

In the medium and longer terms, it is the intention of the Fund that monies will be allocated to road classes and between competing agencies on the basis of need. That however cannot be achieved until designated agencies have a full and reliable knowledge of the extent and condition of their networks and until there is a uniform capacity to interpret such data, to plan and prioritise road maintenance works and to formulate coherent and properly costed programmes of work. It may be several years before that position is reached.

In the intermediate term, the URF Board must annually consider and approve the method by which funds will be allocated between classes of road and between designated agencies for national, district, urban and community access roads. The Board has expressed the wish to have a formula that is practical and in sync with the URF Act which requires that allocations of funds to designated agencies be based on amongst other factors conditions of public roads, maintenance requirements, length of the road network and the relevant volume of traffic. Following the Literature Review Report which formed a high level review of international and regional practices on the area of road maintenance and road maintenance fund allocation to enable development of a conceptual framework ahead of development of the required Fund Allocation Formula; we now articulate in this Final Report the outcomes of the road maintenance finance review, the road maintenance needs assessment, the criticisms of the current allocation formula and the proposed allocation formula including its application.

1.2 Background

The Uganda Road Fund (the Fund or URF) was established by the Uganda Road Fund Act, 2008 (the Act) to facilitate the delivery of road maintenance services and to create an environment that is conducive to the efficient and effective maintenance of public roads. The Fund is to provide its services in the most economic, efficient and effective


18

manner and to manage its affairs in a business-like and cost-effective manner and in accordance with modern management practices and techniques. The roles of the Fund are defined by the Act to include funding routine and periodic maintenance of public roads; ensuring that public roads are maintained at all times; and advising Ministers on the preparation and efficient and effective implementation of the Annual Road Maintenance Programme. The monies of the Fund principally comprise road user charges including fuel levies; international transit fees collected from foreign vehicles entering the country; road licence fees; axle load fines; bridges tolls and road tolls; and weight distance charges.

The Act provides for designation of agencies that are responsible and accountable for the development, rehabilitation and maintenance of public roads in a manner consistent with the economy and set standards. Allocations from the Fund to the designated agencies are to be based, amongst other factors, on the conditions of the public roads, maintenance requirements, the length of the road network and the relevant volume of traffic or derived from an approved maintenance management tool.

1.3 Scope and Objectives

The overall objective of the Study is to give effect to Section 22(2) of the Uganda Road Fund Act, 2008, which states that: “Allocations from the Fund to the designated agencies shall be based among other factors, on the conditions of the public roads, maintenance requirements, the length of the road network and the relevant volume of traffic or derived from an approved maintenance management tool”

The specific objectives of the assignment were to: (f) Examine the current road administration of Uganda and analyse the historic trend

of allocation of road maintenance resources and the relative impacts thereof;

(g) Carry out detailed analysis of designated agencies to develop a proposed hierarchy and pattern of distribution among road classes and designated agencies, identifying common influencing factors on the cost of maintenance of roads;

(h) Review the current practices of road funds in the region and elsewhere in allocation of funds to road maintenance, with emphasis on the transferability of such practices;

(i) Develop, recommend and demonstrate robustness of an equitable, transparent, fair and justifiable approach to allocation of road maintenance resources,


19

vertically between road classes and horizontally between designated agencies responsible for each class of road;

(j) Based on the state of reform of the road sub-sector and the likely pace of

development of institutional, technical and financial capacity of designated agencies; propose a road map for transition from an indirect formula – based approach to a needs based approach.

1.4 Document Overview

The Final Report has been prepared to provide the following:

(i) A review of present and forecast revenue potential based on likely levels of road user charge and comparison with provisions in national medium term expenditure framework;

(ii) A review of estimated needs for road maintenance based on available information and consultation with Uganda National Road Authority and Ministry of Works & Transport;

(iii) An identification of potential assessment criteria for indirect allocation of funds; (iv) Optional approaches based on international best practice and employing the

assessment criteria; (v) A methodology and formula for the proposed allocation of Road Maintenance

Funds to designated agencies; (vi) A simulation/testing of the formula with sample data from designated agencies;

and (vii) A transition road map from formula-based to needs-based allocation of funds to

road maintenance.


20

2. STUDY METHODOLOGY

The assignment was carried out over the following key phases:

i. Inception Phase ii. Literature review iii. Formula development iv. Reporting (including presentation to FMC, SWG and other key stakeholders).

2.1. Inception Phase

The Inception Phase set the direction of the Study. Tasks in this phase were the following:

(a) Review of available reports and data on institutional arrangements for road

administration at national and local levels, governance issues, the legal framework underpinning the sector among others. The key institutional stakeholders depicted in Figure 1.

Figure 1: Key URF Institutional Stakeholders (Source: URF OYRMP & AREP 2010/11)

(b) Meetings with the URF Executive Director and with the URF Manager Programming, Manager Fund Management and key URF staff associated with the Study;

POLICY

FINANCING

IMPLEMENTATION

STAKEHOLDERS

MOWT MLG MFPED

UGANDA ROAD FUND

FUND MANAGEMENT

BOARD

SECRETARIAT

Managing the Financing of

the Network

District

Roads

DESIGNATED AGENCIES

District Councils

Community

Access Roads

Urban Councils/

Municipalities

Urban Roads

UNRA

National Roads

ROAD USERS

Freight

Transport

Association

Passenger

Transporters

Association

Accountants

Profession

Engineering

Profession

Owners and Users of the Network represented by:

UGANDA

REVENUE

AUTHORITY

OTHERS

USE OF THE NETWORK

ROAD USER CHARGES

Flows of Finance

Representation

COLLECTION OF

ROAD USER

CHARGES

Representing the

Interests of Road Users

DISTRICT ROAD COMMITTEES


21

(c) Formulation of a work plan within the allowable budget and duration of the study; and

(d) Presentation of the results of this phase of the Study in the form of a written report and presentation to the URF Team.

2.2. Literature Review Phase

In this phase, the consulting team collected and reviewed all available technical literature on the subject of budget allocation in the roads sub-sector. Allocation approaches in other sectors where direct measurement of need is not practicable were also reviewed for commonality of approach. The key tasks carried out under this phase were the following:

i) Acquisition of existing literature on fund allocation from all available sources,

especially relating to other road funds in sub-saharan Africa;

ii) Review of literature to identify and critically review the approaches used by other road funds regionally and internationally;

iii) Review of similar indirect allocation approaches used in other sectors, both in Uganda and elsewhere;

iv) Review current allocation formula of URF and methods used by MoFPED, MoWT and key Donor Agencies in similar funds disbursed countrywide;

v) Visits to other regional road funds to establish the evolution of their allocation

approaches and to identify particular successes and failures;

vi) Analysis of data, with particular emphasis on extraction of a set of most commonly used factors and proxies for estimation of road maintenance needs including but not limited to Geographic and environmental, traffic, condition related and demographic factors; and

vii) Preparation and presentation of the Literature Review Report (dated June 2011) to the Uganda Road Fund team. This report provides a detailed explanation of the previous studies that informed the formulae study.

2.3. Formula Development Phase

In this phase of the Study, we have developed an approach to fund allocation that we believe is robust and defensible taking care of the following key principles:

(i) Equitable sharing of funds between road classes and designated agencies;


22

(ii) Since the user pays, the allocation shall be fairly representative of road usage;

(iii) On that basis, the maintenance of the national road network shall be fully funded subject to demonstration of value for money and allocated on the basis of a road management system;

(iv) Funds shall be allocated to give the optimum effect in terms of road condition;

(v) Since traffic levels on district and community access roads are generally low, the

social benefit of these roads shall be recognised by a significant contribution to their upkeep from the consolidated fund or other charges and that shall be a condition of Road Fund support, lest the money allocated be wasted on inadequate maintenance. The need for counterpart funding shall also be given consideration here;

(vi) The formula shall as much as possible, reflect the policies, plans and strategies of

the Government of Uganda relating to the road sub-sector and also to general development (as reflected in the current National Development Plan and in Millennium Development Goals); and

(vii) The validity of the formulae has been tested and verified by application to a sample

of agencies. 2.4. Reporting Phase

In this phase of the Study, this Report was produced and shall be supported by presentations that will be made to the Board and other stakeholders.


23

3. ROAD MAINTENANCE FINANCING REVIEW

3.1. Sources of Road Maintenance Financing

Section 21 of the Uganda Road Fund Act (2008) indicates that the monies of the Fund are expected to consist of the following:

(i) road user charges as the Minister may, on the recommendation of the Board

determine by statutory order, including: (i) fuel levies; (ii) international transit fees, collected from foreign vehicles entering the country; (iii) road license fees; (iv) axle load fines; (v) bridges tolls and road tolls; and (vi) weight distance charges;

(ii) fines under the Traffic and Road Safety Act; (iii) monies appropriated by Parliament for the purposes of the Fund; (iv) revenue or assets received by the Fund in the performance of its functions under

the Act; (v) grants, monies or assets donated to the Fund by Government with the approval of

the Minister or assets donated to the Fund by any foreign government, international agency or other external body of persons, corporate or unincorporated;

(vi) any revenue derived from the sale of any property, movable or immovable, by or on behalf of the Fund;

(vii) money received by the Fund by way of voluntary contributions; (viii) fines payable under this Act; and (ix) earnings from investments made and interest accruing from monies of the funds

deposited in Banks.

While the URF Act highlights road user charges and fines under the Traffic and Safety Act as revenues to accrue to the fund, The URA Act (1991) indicates that URA is the agency to administer and give effect to the laws indicated in the First Schedule of the URA Act and for that purpose assess, collect and account for all revenue to which those laws apply. The Schedule indicates the Traffic and Road Safety Act and all other taxes and non tax revenue as the Treasury may prescribe which effectively include the road user charges highlighted above. In addition to this, Section 14 of the URA Act states that all revenue collected by or due and payable to Uganda Revenue Authority under the URA Act shall be credited to the Consolidated Fund except as may from time to time be authorized by the minister to


24

retain a percentage of the revenue so collected for the purpose of URA meeting its expenditure. The key sources of Road Maintenance Funding to Government at the moment are the following:

3.1.1 Excise duty on Fuel

Presently, Government of Uganda through Uganda Revenue Authority collects duty for every litre of fuel imported under the category of taxes on international trade as given in Table 1.

Table 1: Fuel charges per product

Fuel Product Exemptions Charge per Litre in Shs

1 Kerosene (BIK) Sales to Diplomats, Embassies and Bicycle Manufacturers

200

2 Diesel (AGO) Sales to Diplomats, Embassies and Bicycle Manufacturers and Power Generators

530

3 Petrol (PMS) Sales to Diplomats and Embassies 850

Source: URA Annual Revenue Bulletin Reports and the OYRMP 2011/121

3.1.2 International transit fees, collected from foreign vehicles entering the country Government currently through Uganda Revenue Authority collects a temporary road license fee of USD 0.22 for every foreign registered vehicle entering Uganda.

3.1.3 Road License Fees Previously, Uganda Revenue Authority used to collect a Road License fee but this was replaced by a Fuel Tax which is effectively included in the price of fuel starting with the 2007/08 financial year. The main reason fronted for its abolishment was it being too problematic to collect due to many forgeries.

3.1.4 Axle Load Fines This is a punitive fine for the abuse of axle load. There are currently many problems involved in axle load control mainly because of a weak law, errant drivers and ignorance of transporters. Axle load refers to the weight allowed per tyre – The maximum gross axle load is 18 tonnes for six-tyre trucks, 24 tonnes for those with 10 tyres and 48 tonnes for trailers. Note that maximum axle load may now be 54 tonnes as a result of

1 One-Year Road Maintenance Plan and Annual Road Expenditure Programme (2011/12)

2 URA Annual Revenue Bulletin Reports


25

EAC harmonization. The idea behind these fines is that instead of taking the culprits to the courts (which do not operate on weekends), the excess weight is computed and the offender made to pay there and then.

3.1.5 Bridge and Road Tolls These refer to charges paid per road and/or bridge user to have access to the facility. Currently, a very small component of revenue reverts to Government on account of these tolls with the bulk of the revenue being collected at the Local Government level.

3.1.6 Weight Distance Charges These charges are based on how much a vehicle is driven hence the more it is driven, the more is paid for it. The fee is generally noted to be more economically efficient and fair than many other charges. It however is not currently in use in Uganda.

3.1.7 Fines under the Traffic and Road Safety Act (a) Fine not less than ten currency points and not exceeding forty currency points

for the use or permitting to be on any road a motor vehicle, trailer or engineering plant: (i) which is not registered in accordance with the Act; (ii) without the prescribed registration plates issued and affixed in the prescribed

manner; (iii) without a license to use that motor vehicle, trailer or engineering plant issued

for the licensing period in question and affixed in the prescribed manner; (iv) without any insurance prescribed by law for that use; or (v) while any registration plate or license affixed to it in the prescribed manner is

in any way obscured or is rendered or allowed to become not easily distinguishable by night or by day.

(b) Fine not less than five currency points and not exceeding fifteen currency points

for the use or permitting to be used on any road a motor vehicle, trailer or engineering plant: (i) of which the person is the owner unless he or she or his or her authorized

agent has possession of the registration book issued in relation to it; or (ii) for which a license fee has been paid less than the fee payable for a vehicle of

that class, (iii) commits an offence and is liable on conviction to a fine of not less than five (iv) currency points and not exceeding fifteen currency points or to imprisonment (v) not exceeding six months or both.


26

(c) Fine not exceeding thirty currency points for affixing or causing to be affixed to any motor vehicle, trailer or engineering plant: (i) any registration plate not authorized by the Act or by regulations and which is

likely to be mistaken for any authorized registration plate; or (ii) any license not being the license issued for that motor vehicle, trailer or

engineering plant for the licensing period in question or authorized to be used on that motor vehicle, trailer or engineering plant for that licensing period,

(iii) commits an offence and is liable on conviction to a fine not exceeding thirty (iv) Currency points or imprisonment not exceeding twelve months or both.

(d) Fine of not less than ten currency points and not exceeding thirty currency points for giving false information or making an incorrect statement: (i) when effecting or changing the registration of a motor vehicle, trailer or

engineering plant; (ii) when applying for a license or a change of license for a motor vehicle, trailer

or engineering plant; (iii) as to the mileage recorded on the mileage recorder (if any) of a motor vehicle,

trailer or engineering plant; (iv) when effecting the change of ownership of a motor vehicle, trailer or

engineering plant; or (v) when applying for a dealer’s vehicle license,

(e) Fine of not less than five currency points and not exceeding fifteen currency points

for any person that: (i) alters the mileage recorder (if any) on a motor vehicle, trailer or engineering

plant so as to reduce the mileage recorded; (ii) obtains, uses or permits to be used dealer’s plates otherwise than in

accordance with the Act; (iii) fails to surrender his or her registration plates or his or her registration book

for alteration or cancellation as required so to do under the Act; or (iv) fails to surrender his or her license and registration plates for alteration or

cancellation as required so to do under the Act.

3.1.8 Monies appropriated by Parliament for the purposes of the Fund Ever since the formation of the Uganda Road Fund, all funds obtained have been appropriations by Parliament to enable the achievement of the purposes of the fund. The trend of appropriations for road maintenance activity before and after formation of the Uganda Road Fund is analysed in a Section lower below.


27

3.2. Historical Assessment of Potential Road Maintenance Financing

A historical perspective of Road User Charges as collected by Uganda Revenue Authority for the key revenue sources of Petroleum Duty, Temporary Road Licenses (International Transit Fees) and Fees collected under the Traffic Act is provided in Table 2.

Table 2: Annual revenue from Fuel (in billions of Uganda Shillings)

No. Source Financial Year

2006/07 2007/08 2008/09 2009/10 2010/11

1 Petroleum Duty 383.97 579.11 632.32 716.23 821.18

2 Less: Refunds on Diesel (8.41) (68.99) (66.11) (78.00) (92.30)

Net Petroleum Duty 375.56 510.12 566.21 638.23 728.88

3 Temporary Road Licenses 13.31 19.20 23.55 25.43 45.14

4 Fees under the Traffic Act 63.21 40.00 47.83 64.10 111.26

Total 452.08 569.32 637.59 727.76 885.28

Annual Growth (Shs.bn) 117.24 68.27 90.17 157.52

Annual Growth (%) 26% 12% 14% 22%

Source: Uganda Revenue Authority Annual Revenue Bulletin Reports (Shs Billion)

As shown in the Table 3, these fees as collected by URA over the 5 years have been increasing at an annual growth rate of between 26% and 14% giving an average annual growth rate of 18%.

Table 3: Percentage total revenue contribution by fuel charge type per year

No. Source Financial Year Average

2006/07 2007/08 2008/09 2009/10 2010/11

1 Petroleum Duty

83% 91% 90% 89% 84%

87%

2 Temporary Road Licences

3% 3% 3% 3% 5%

3%

3 Fees per the Traffic Act

14% 6% 7% 8% 11%

9%

Total 100% 100% 100% 100% 100% 100%

Source: URA Annual Revenue Bulletin Reports as analyzed From a share perspective, the Petroleum Duty accounted for an average of 87% of the Total fees collected by URA while Fees under the traffic Act accounted for 9% and International Transit fees 3%.


28

3.3. A review of current and historical Road Financing

A review of the current and historical funding available for road maintenance reveals an increase in funding from Ushs.85.73 billion in 2006/07 to Ushs.277.15 billion in 2010/11, the first financial year that the Uganda Road Fund became operational. Table 4 shows the financing that has been made available for road maintenance over the most recent 5 year period.

Table 4: Current financing available for road maintenance per year (in billions of UShs)

No. Network Financial Year

2006/07 2007/08 2008/09 2009/10 2010/11

1 National Road Network and Bridges

68.47 141.61 165.94 96.56 184.49

2 District Road Network 13.06 29.40 40.00 72.42 38.35

3 Urban Road Network 4.20 26.50 26.50 40.55 47.29

4 Community Access Road Network

- - - 8.00 7.01

Total Allocations 85.73 197.51 232.44 217.53 277.15

Annual Growth (Shs.bn) 111.78 34.93 (14.91) 59.63

Annual Growth (%) 130% 18% -6% 27%

Source: Ministry of Works and Transport Ministerial Budget Policy Statements

As shown in Table 5, over the past 5 years, Road User Charges far outstripped the Road Maintenance Allocations from the Consolidated Fund by an average of 30% over the period (between 19% and 36%). This is indicative of potential increased revenues for the purpose of the maintenance of roads should the overall sum be allocated to the Road Fund as is expected (assuming the full component of the Road User Charges are allocated).

Table 5: URF allocations as a share of total collections (in billions of Ushs)

No. Item Financial Year

2006/07 2007/08 2008/09 2009/10 2010/11

1 Road User Charge Collections

452.08 569.32 637.59 727.76 885.28

2 Road Maintenance Allocations

85.73 197.51 232.44 217.53 277.15

3 Variance (1-2) 366.35 371.81 405.15 510.23 608.13

4 Allocations as a share of collections

19% 35% 36% 30% 31%


29

From an Medium Term Expenditure (MTEF) point of view, the funding allocated to the Sector is expected to grow from the 2011/12 allocated Shs.280.95 billion by an annual 22% factor to Shs.341.95 billion in 2012/13 and to Shs.416 billion in 2013/14, see Table 6.

Table 6: Fund allocation to the Road sector from an MTEF perspective (in billions of UShs)

Budget MTEF Budget Projections

2010/11 2011/12 2012/13 2013/14

Vote: 118 Road Fund 283.88 280.95 341.95 416.96

Source: 2011/12 Ministry of Works and Transport Ministerial Budget Policy Statement

It is worth noting though that while the ceiling for vote 118: Uganda Road Fund of Shs.280.95bn is what was planned for, this ceiling was not based on the established needs of the designated agencies as generated by the URF Board for the 2011/12 Financial Year which were estimated at Shs.413.94 billion in effect giving a shortfall of Shs.133 billion. A comparison of the current funding to the maintenance needs of the network is provided in the next Section. It is critical that the fund quickly finalizes all legal reforms to allow the collection and direct remittance of Road Users Charges (RUCs) especially fuel levy to a URF account that should be maintained with Bank of Uganda. Under this new proposed framework, the URF Board should be able to independently recommend RUCs to Minister for Finance as required by law based on established needs of agencies.


30

4. ROAD MAINTENANCE NEEDS ASSESSMENT 4.1. Highway Development and Management (HDM-4) Tool

4.1.1 Overview

Unconstrained maintenance needs for Uganda Road Network was assessed using the Highway Development and Management Tool (HDM-4). The HDM-4 analytical framework (Odoki and Kerali, 2000) is based on the concept of pavement life cycle analysis, which is typically 15 to 40 years. This is applied to predict road deterioration, road works effects, road user effects and socio-economic and environmental effects. The HDM-4 model was used to simulate future changes to the Ugandan road system from current conditions. The reliability of the results is dependent upon two primary considerations:

How well the data provided to the model represent the reality of current conditions and influencing factors, in the terms understood by the model; and,

How well the predictions of the model fit the real behavior and the interactions between various factors for the variety of conditions to which it is applied.

Application of the HDM-4 model thus involves two important initial steps:

Data input: a correct interpretation of the data input requirements, and achieving a quality of input data that is appropriate to the desired reliability of the results.

Calibration of output: adjusting the model parameters to enhance the convergence of the computed road behavior with that observed in the field for the various interventions. Calibration of the HDM model focuses on the components that determine the physical quantities, costs and benefits predicted by the analyses.

Since HDM-4 is designed to be used in a wide range of environments, it needed to be configured to reflect the characteristics of the road network in Uganda. The data for this relates to traffic flows, climate zones and road types. Calibration is intended to improve the accuracy of predicted pavement performance and vehicle resource consumption. A fundamental assumption made prior to using HDM-4 is that the pavement performance models will be calibrated to reflect the observed rates of pavement deterioration on the roads where the models are to be applied. The extent of calibration may be defined as follows:

Level 1: Application: based on a desk study of available data and engineering experience of pavement performance.


31

Level 2: Verification: based on measured pavement condition data collected from a large number of road sections.

Level 3: Adaptation: experimental data collection required to monitor the long-term performance of pavements within the study area.

This study has benefited from the results of previous studies related to the implementation of HDM-4 in Uganda. Nevertheless, it was necessary to update the configuration and calibration data to suit the study as reported under the data requirements for HDM-4.

4.1.2 HDM-4 Data Requirements The main data sets required as inputs for HDM-4 analyses are categorized as follows:

Road network data comprising: inventory, geometry, pavement type, pavement strength, road condition, average annual daily traffic on each road section and traffic composition;

Vehicle fleet data including vehicle physical characteristics, tyres, utilisation, loading and performance;

Road works data comprising a range of road maintenance activities practised in Uganda and their associated unit costs.

Economic analysis parameters comprising discount rate of 12% and base year 2011. The 12% discount rate is recommended for road investment appraisal in Uganda and is used by development partners of Uganda.

Data required for HDM-4 analysis were largely sourced from URF, UNRA, Ministry of Works, reports of previous HDM-4 studies conducted in Uganda, and internet literature review. Logical assumptions were made to fill in the gaps in cases where data were either lacking or missing.

4.2. Available Information for the Uganda Road Network

4.2.1. Available Sources of Information The following source information was used in the development of the allocation procedure that includes the calibration and use of HDM-4.

2010 Statistical Abstract, Uganda Bureau of Statistics, June 2010 Procedural Guide to Economic Aspects of Conducting Road Feasibility Studies in Uganda,

MWHC, Louis Berger International, October 1999 Procedural Guide to Economic Road feasibility Studies, MWHC, March 2006 Road Sector Development Programme Phase 2 Project (RSDP) – Consultancy Services for

the Preparation of the National Transport Master Plan Including a Master Plan for the Greater Kampala Metropolitan Area, MWHC, TAHAL Consulting Engineers, March 2005


32

Road Sector Institutional Support Technical Assistance Project (RSISTAP) – Technical Assistance Consulting Services of Setting Up a Road Fund in Uganda, MWHC, BKS Consulting Engineers, September 2005

Field Data Collected for current study in terms of road lengths and surface type per designated agency, June-August 2011

4.2.2. Key Assumptions

1. Traffic Demand Data – The 2003 base data (from the RSDP Phase 2) was increased with

a growth rate of 3% per annum to estimate the 2011 traffic demand for all the road networks. This annual growth rate is based on historical traffic counts by road network and road type.

2. National Road Network – The national road network was expanded with the recent 10,000km from the districts in terms of both ADT and Surface type by using the available national road characteristics on a regional basis. Updated road condition data for a sample of the national road network was obtained from the Stations during May 2011.

3. District Roads, Community Access Roads, Urban Roads – The 2003 base data (from the RSDP Phase 2) was adjusted and applied to Districts on a Regional Basis

4. Municipal Roads, Kampala Roads – The 2003 base data (from the RSDP Phase 2) was adjusted and applied to Municipalities on a Regional Basis.

5. At this stage distinction is made between road condition and traffic levels only in terms of road network and surface type and not on a more disaggregate level for designated agencies and road links. This data is simply not available and should be obtained / collected by the designated agencies for future allocations.

4.3. Uganda Road Network Data

4.3.1. Inventory and Condition Data

The total Uganda Road Network was determined to be about 133,000 km in extent. The current extent and distribution of the Uganda Road Network, that was analyzed, is shown in Tables 8 to 12. About 10,000km of roads were recently transferred from the districts to the UNRA (only the length and location in terms of an UNRA Station is known for these roads. The relative distribution of the road network in terms of jurisdiction and surface type is shown in Table 7.


33

Table 7: Uganda Road Network Length (in kilometers)

Road Network Paved Roads Unpaved Roads Total

National Roads 3,252 17,310 20,562

District Roads 0 22,500 22,500

CAR - 30,000 30,000

Urban Roads 1100 4,500 5,600

Total 4,352 74,310 78,662

Source: One-Year Road Maintenance Plan and Annual Road Expenditure Programme (2011/12) The above table has been updated with more current figures from the MoWT and UNRA however MoWT was not able to furnish us with a breakdown of the road length by road class i.e. Paved, Gravel and Earth.

Road Network Paved Roads

Unpaved Roads

Total Total (Not given by road type)

MoWT Total

National Roads 3,252 17,310 20,562 471 21,033

District Roads - 29,383 29,383 4,443 33,826

CAR - 75,750 75,750 7,027 82,777

Urban Roads 263 2,741 3,004 5,723 8,727

Municipal Roads 996 3,219 4,215

2,737

Total 4,511 128,403 132,914 17,664 149,100

Source: MoWT DUCAR Database and UNRA Database 2012 NOTE: For purposes of the calibration of the data in the HDM-4 tool, we have considered only the available road network data disintegrated by road type. i.e. highlighted above totalling 132,914 Kilometres and have used this in the formula spread sheet. URF will be expected in discussion with the agencies to update this information on an annual basis. The road network data (or attributes) were used to create road network matrix for the Uganda Road Network for purposes of HDM-4 modelling. A road network matrix is defined according to the key attributes that most influence pavement performance and road user costs. The matrix formulated comprises cells of equivalent two-lane road lengths defined by unique characteristics including road class (National, Urban, Municipal, District, Community Access), traffic class (Very High Traffic, High Traffic, Medium Traffic, Low Traffic, and Very Low Traffic), pavement surface type (Asphalt Mix, Surface Treatment, Gravel and Earth), and condition (Very Good, Good, Fair Poor and Very Poor). Sub-sequent sub-sections summarise the key characteristics of the National, District, Community Access, Urban and Municipal Road Networks.


34

4.3.1.1. National Road Network The estimated lengths of the National Road Network are summarized in Table 8 by traffic category, road condition, and surface type.

Table 8: Summary of National Road Network Characteristics

Asphalt Mix

Condition (IRI)

Very Good Good Fair Poor

Very Poor Total

Traffic (AADT) 2 3 4.5 8 12

Very Low <1000

4 17.3 8.4 6.4 4 40.1

Low 1000-3000 0 0 0 0 0 0

Medium 3000-10000 7 30.1 14.7 11.2 7 70

High 10000-30000 3.8 16.3 8 6.1 3.8 37.9

Very High >30000

0 0 0 0 0 0

Total 14.8 63.6 31.1 23.7 14.8 148

Surface Treatment

Condition (IRI)


Very Poor Total

Traffic (AADT) 3 4 5.5 9 13

Very Low <300

5.8 20.8 27.4 19.1 10 83

Low 300-1000 38.5 137.4 181.4 126.4 66 549.6

Medium 1000-3000 92.6 330.8 436.6 304.3 158.8 1,323.10

High 3000-10000 37.6 134.3 177.2 123.5 64.5 537.1

Very High >10000

6.4 22.7 30 20.9 10.9 90.8

Total 180.9 645.9 852.6 594.2 310 2,583.60

Gravel

Condition (IRI)


Very Poor Total

Traffic (AADT) 5 7 11 16 20

Very Low <10

63.9 228.3 301.4 210.1 109.6 913.4

Low Oct-30 190.7 681.2 899.2 626.7 327 2,724.90

Medium 30-100 313 1,118.00 1,475.70 1,028.50 536.6 4,471.90

High 100-300 646.2 2,307.80 3,046.30 2,123.20 1,107.70 9,231.20

Very High >300

34.2 122.3 161.4 112.5 58.7 489.1

Total 1,248.10 4,457.60 5,884.10 4,101.00 2,139.70 17,830.50

Earth


35

Condition (IRI)


Very Poor Total

Traffic (AADT) 7 9 13 18 22

Very Low <10

0

Low Oct-30

0

Medium 30-100

0

High 100-300

0 Very High >300 0

Total 0 0 0 0 0 0

Source: UNRA Database, 2010

4.3.1.2. District Road Network The estimated characteristics of the District Road Network are summarized in Table 9 by traffic category, road condition, and surface type.

Table 9: Summary of District Road Network Characteristics

Asphalt Mix

Condition (IRI)

Very Good Good Fair Poor Very Poor

Traffic (AADT) 2 3 4.5 8 12 Total

Very Low <1000 0.0

Low 1000-3000 0.0

Medium 3000-10000 0.0

High 10000-30000 0.0

Very High >30000 0.0

Total 0.0 0.0 0.0 0.0 0.0 0.0

Surface Treatment

Condition (IRI)



Very Low <300 37.4 240.7 149.8 64.2 42.8 534.8

Low 300-1000 52.7 339.1 211.0 90.4 60.3 753.5

Medium 1000-3000 30.5 195.8 121.8 52.2 34.8 435.0

High 3000-10000 2.2 14.3 8.9 3.8 2.5 31.7

Very High >10000 0.6 3.6 2.2 1.0 0.6 8.0

Total 123.4 793.3 493.6 211.6 141.0 1,763.0

Gravel

Condition (IRI)


Traffic (AADT) 5 7 11 16 20 Total

Very Low <10 89.4 715.1 536.3 268.2 178.8 1,787.8

Low 10-30 142.9 1,142.8 857.1 428.6 285.7 2,857.0


36

Medium 30-100 423.9 3,391.5 2,543.6 1,271.8 847.9 8,478.8

High 100-300 666.1 5,328.4 3,996.3 1,998.2 1,332.1 13,321.1

Very High >300 0.0 0.0 0.0 0.0 0.0 0.0

Total 1,322.2 10,577.9 7,933.4 3,966.7 2,644.5 26,444.8

Earth

Condition (IRI)



Very Low <10 28.7 229.4 172.0 86.0 57.3 573.4

Low 10-30 30.1 240.8 180.6 90.3 60.2 601.9

Medium 30-100 0.0 0.0 0.0 0.0 0.0 0.0

High 100-300 0.0 0.0 0.0 0.0 0.0 0.0

Very High >300 0.0 0.0 0.0 0.0 0.0 0.0

Total 58.8 470.1 352.6 176.3 117.5 1,175.3

Source: Ministry of Works DUCAR Database, 2011

4.3.1.3. Community Access Roads The estimated characteristics of the Community Access Roads are summarized in Table 10 by traffic category, road condition, and surface type.

Table 10: Summary of Community Access Road Network Characteristics

Asphalt Mix

Condition (IRI) Very Good Good Fair Poor Very Poor


Very Low <1000 0.0

Low 1000-3000 0.0

Medium 3000-10000 0.0

High 10000-30000 0.0


Total 0.0 0.0 0.0 0.0 0.0 0.0

Surface Treatment



Very Low <300 0.0

Low 300-1000 0.0

Medium 1000-3000 0.0

High 3000-10000 0.0


Total 0.0 0.0 0.0 0.0 0.0 0.0

Gravel



Very Low <10 0.0 216.0 575.9 431.9 216.0 1,439.8


37

Low 10-30 0.0 297.5 793.4 595.1 297.5 1,983.6

Medium 30-100 0.0 259.1 691.0 518.3 259.1 1,727.6

High 100-300 0.0 0.0 0.0 0.0 0.0 0.0

Very High >300 0.0 0.0 0.0 0.0 0.0 0.0

Total 0.0 772.6 2,060.4 1,545.3 772.6 5,151.0

Earth



Very Low <10 0.0 7,059.8 19,767.6 20,473.6 23,297.5 70,598.5

Low 10-30 0.0 0.0 0.0 0.0 0.0 0.0

Medium 30-100 0.0 0.0 0.0 0.0 0.0 0.0

High 100-300 0.0 0.0 0.0 0.0 0.0 0.0

Very High >300 0.0 0.0 0.0 0.0 0.0 0.0

Total 0.0 7,059.8 19,767.6 20,473.6 23,297.5 70,598.5

Source: Ministry of Works DUCAR Database, 2011

4.3.1.4. Town Council Road Network The estimated characteristics of the Urban Road Network are summarized in Table11 by traffic category, road condition, and surface type.

Table 11: Summary of Urban Road Network Characteristics

Surface Treatment



Very Low <300 6.2 35.2 20.3 17.7 12.7 92.1

Low 300-1000 5.3 30.2 17.4 15.2 10.9 78.9

Medium 1000-3000 3.6 20.1 11.6 10.1 7.2 52.6

High 3000-10000 1.8 10.1 5.8 5.1 3.6 26.3

Very High >10000 0.9 5.0 2.9 2.5 1.8 13.2

Total 17.8 100.7 57.9 50.7 36.2 263.1

Gravel



Very Low <10 3.8 21.6 12.4 10.9 7.8 56.5

Low 10-30 19.7 111.5 64.1 56.1 40.1 291.4

Medium 30-100 31.6 179.2 103.0 90.2 64.4 468.4

High 100-300 14.8 83.6 48.1 42.1 30.1 218.6

Very High >300 0.7 4.0 2.3 2.0 1.4 10.4

Total 70.6 399.8 230.0 201.2 143.7 1,045.3

Earth


38



Very Low <10 23.1 130.7 75.2 65.8 47.0 341.6

Low 10-30 56.4 319.3 183.7 160.7 114.8 834.8

Medium 30-100 35.0 198.5 114.2 99.9 71.3 518.9

High 100-300 0.0 0.0 0.0 0.0 0.0 0.0

Very High >300 0.0 0.0 0.0 0.0 0.0 0.0

Total 114.4 648.5 373.0 326.4 233.1 1,695.3

Source: National Transport Master Plan, Greater Kampala Study, TAHAL (2005)

4.3.1.5. Municipal Road Network The estimated characteristics of the Municipal Road Network are summarized in Table 12 by traffic category, road condition, and surface type.

Table 12: Summary of Municipal Road Network Characteristics

Asphalt Mix



Very Low <1000 0.0 0.0 0.0 0.0 0.0 0.0

Low 1000-3000 0.1 0.2 0.2 0.1 0.0 0.5

Medium 3000-10000 3.0 12.1 10.0 3.0 2.1 30.3

High 10000-30000 2.6 10.2 8.5 2.6 1.8 25.6

Very High >30000 0.0 0.1 0.1 0.0 0.0 0.2

Total 5.7 22.7 18.7 5.7 4.0 56.7

Surface Treatment



Very Low <300 27.8 111.4 91.9 27.8 19.5 278.5

Low 300-1000 17.8 71.2 58.7 17.8 12.5 178.0

Medium 1000-3000 15.9 63.4 52.3 15.9 11.1 158.6

High 3000-10000 17.1 68.3 56.3 17.1 11.9 170.7

Very High >10000 15.4 61.6 50.8 15.4 10.8 153.9

Total 94.0 375.8 310.1 94.0 65.8 939.6

Gravel



Very Low <10 6.7 40.4 43.1 33.6 10.8 134.6

Low 10-30 21.7 130.5 139.2 108.7 34.8 434.9

Medium 30-100 57.8 346.9 370.0 289.1 92.5 1,156.2

High 100-300 34.5 207.1 220.9 172.6 55.2 690.2

Very High >300 2.9 17.3 18.5 14.5 4.6 57.8

Total 123.7 742.1 791.6 618.4 197.9 2,473.7


39

Earth



Very Low <10 4.9 29.5 31.5 24.6 7.9 98.5

Low 10-30 11.7 70.4 75.1 58.7 18.8 234.7

Medium 30-100 18.6 111.6 119.0 93.0 29.8 371.9

High 100-300 2.0 11.9 12.7 9.9 3.2 39.8

Very High >300 0.0 0.0 0.0 0.0 0.0 0.0

Total 37.2 223.5 238.3 186.2 59.6 744.8

Source: National Transport Master Plan, Greater Kampala Study, TAHAL (2005)

4.3.2. Traffic Data Assumptions of Average traffic levels (2011) based on historical traffic counts and an average traffic growth rate of 3% per annum by road network and road type (Paved, Gravel, and Earth) are summarized in Table 13. The average daily traffic on the Uganda Road Network was estimated at 160 vehicles per day including figures for CAR network and 360 vehicles per day excluding figures for the CAR network, as given in Table 13. The traffic composition by vehicle type, the passenger car unit (PCU) and the equivalent standard axle load factor (ESALF) used in the Study are given in Table 14.

Table 13: Average 2011 Daily Traffic on the Uganda Road Network (in vehicles / day)

Road Network Paved Roads Gravel Roads Earth Roads Total

National Roads 3,044 151 - 536

District Roads 884 99 10 142

CAR - 23 6 7

Urban Roads 1,810 81 31 204

Municipal Roads 4,519 96 39 1,131

Total 2,581 109 7 159

Table 14: Traffic Composition, PCU and ESALF

Vehicle type Composition1 (in %)

Passenger car units (PCU)2

Equivalent standard axle load factor (ESALF)1

Motorcycles 21 0.5 0

Cars 22 1 0

Pick Up (LGV) 14 1 0

Four Wheel Drive 7 1 0.01

Minibus 20 1.5 0

Big Buses 2 2.5 2.39

Light Trucks 4 1.9 0.02

Medium Trucks 3 1.9 3.5


40

Heavy Trucks 4 1.9 8.27

Articulated Trucks 3 2.9 17.63

Source: MoWHC Strategic Analysis of Uganda Road Network using HDM-4 (2006) Notes to Table 14:

1. Derived from Uganda Ministry of Works Housing and Communications, Strategic Analysis of Uganda Road Network using HDM-4 (2006). Executive summary of the report is provided in Appendix 4.

2. Obtained from UK Department for Transport, TAG Unit 3.9 (2011).

4.4. Maintenance Standards and Unit Costs Road maintenance works and associated unit costs used in the analysis are summarized in Table 15. The road works were logically grouped into four investment alternatives for paved roads and three investment alternatives for unpaved road sections. These maintenance standards range from provision of routine pavement maintenance on (Base Maintenance) to a Desired (in the case of unpaved roads) or Ideal maintenance (in the case of paved roads). For each maintenance standard, road works standards were pre-defined as shown in Tables 16, 17 and 18.

The maintenance standards were assigned by traffic classes, since with condition responsive maintenance intervention, the interval between maintenance applications will clearly be influenced by the traffic volume.


41

Table 15: Road Maintenance Work Types and Unit Costs

Surface Type

Work Type Description Units Financial

Costs Economic

Costs

Paved

Patching Potholes Repair of surface distresses such as potholing, wide structural cracking and ravelling

m2 11.55 10.07

Edge Break Repair Patching edge failures on paved roads m2 11.69 9.94

Crack Sealing Treatment of transverse thermal cracking and wide structural cracking on paved roads

m2 9.5 8.25

Surface Dressing (Resealing)

Single sealing of the carriageway with shape correction in order to delay major intervention and to renew the skid resistance.

m2 9.55 8.3

Overlay 40mm overlay to existing paved roads. m2 25.99 22.6

Unpaved

Regravelling Regravelling existing unpaved road to a final gravel thickness of 150mm

m2 19.2 16

Heavy Grading Heavy motorised grading of unpaved roads with water and light roller compaction

Per km 1260 1050

Light Grading Light motorised grading with little or no water and no roller compaction

Per km 695 591

Spot Regravelling Spot regravelling to unpaved roads to replace 80% of annual material loss

m3 8.53 7.11

Source: Adapted from Strategic Analysis of Uganda Road Network, Ministry of Works Housing and Communications, 2006.


42

Table 16: Paved Roads Maintenance Standards

Investment Alternative For High and Very High Traffic For Medium Traffic For Low and Very Low Traffic

Base Maintenance

Drainage Interval >= 1 Year Interval >= 1 Year Interval >= 1 Year

Edge Repair Interval >= 1 Year Interval >= 1 Year Interval >= 1 Year

Patching Potholing >= 10 No/km Potholing >= 25 No/km Potholing >= 50 No/km

Crack sealing Wide Structural Cracking >=5% Wide Structural Cracking >=10%

Wide Structural Cracking >=15%

Basic Maintenance




Reseal Total Damaged Area >= 20% Total Damaged Area >= 30% Total Damaged Area >= 40%

Desired Maintenance




Overlay Roughness >= 6 IRI and Total Damage Area >= 15%

Roughness >= 7 IRI and Total Damage Area >= 20%


Ideal Maintenance




Reseal Total Damaged Area >= 15% Total Damaged Area >= 25% Total Damaged Area >= 35%

Overlay Roughness >= 5 IRI and Total Damage Area >= 10%



Source: Adapted from Strategic Analysis of Uganda Road Network, Ministry of Works Housing and Communications, 2006.

Note: The symbol >= means greater or equal to


43

Example:

Table 16 gives treatment intervention levels for treatment types under each investment alternative. For example, under the desired

maintenance standard and High or Very High traffic levels, overlays are scheduled to be triggered when roughness is greater or

equal to 6 IRI and Total Damaged Area is greater or equal to 15%.

Table 17: Gravel Roads Maintenance Standard

Investment Alternative For Medium, High, and Very High Traffic For Low and Very Low Traffic

Base Maintenance

Heavy Grading Interval >= 365 days Interval >= 365 days

Spot Regravel Gravel Thickness <= 100mm -

Basic Maintenance


Spot Regravel Gravel Thickness <= 100mm Gravel Thickness <= 100mm

Regravel Gravel Thickness <= 50mm -

Desired Maintenance


Spot Regravel Gravel Thickness <= 100mm Gravel Thickness <= 100mm

Regravel Gravel Thickness <= 75mm -

Source: Adapted from Strategic Analysis of Uganda Road Network, Ministry of Works Housing and Communications, 2006

Note: The symbol >= means greater or equal to; the symbol <= means less or equal to


44

Table 18: Earth Roads Maintenance Standard

Investment Alternative For Medium, High, and Very High Traffic For Low and Very Low Traffic

Base Maintenance

Spot Regravel Roughness >= 11 m/km Roughness >= 13 m/km

Basic Maintenance

Light Grading Roughness >= 16 m/km Roughness >= 20 m/km


Desired Maintenance

Grade Roughness >= 13 m/km Roughness >= 16 m/km


Source: Adapted from Strategic Analysis of Uganda Road Network, Ministry of Works Housing and Communications, 2006. Note: The symbol >= means greater or equal to


45

4.5. Analysis Results – Road Maintenance Needs and Funding Gap

4.5.1. Unconstrained Maintenance Needs HDM-4 was used to investigate the maintenance needs of the Uganda road network assuming no constraints in budget given the standards defined above. This analysis scenario results in an unconstrained road agency costs based on a life cycle analysis that considers the different investment alternatives for each representative road section. The alternative giving the highest NPV is assigned to each section. The road works assigned by HDM-4 represents the optimum maintenance that should be applied to the road network in accordance with the specified standards. Table 19 summarizes the unconstrained maintenance needs annualized over three ranges of years 1 to 5, 1 to 10 and 1 to 20. Overall, the required annualized maintenance expenditures (excluding reconstruction or major improvements) are US$ 289 million over years 1-5, US$ 258 million over years 1-10, US$ 245 million over years 1-20.

Table 19: Annualised Unconstrained Maintenance Needs (in millions of US $)

Works Road Class Annualised

Costs (Years 1 - 5)

Annualised Costs (Years 1-10)

Annualised Costs

(Years 1-20)

Periodic Maintenance (overlay, resealing, and regravelling)

National 97.71 80.49 71.23

District 64.93 51.33 48.15

Urban 3.24 3.04 2.59

Municipal 17.64 16.19 14.59

CAR 0.00 0.00 0.00

Sub-Total 183.51 151.05 136.56

Recurrent Maintenance (pothole patching, edge repair, drainage cleaning, spot regravelling and grading)

National 37.21 37.31 38.55

District 52.30 53.91 53.76

Urban 2.05 2.11 2.25

CAR 13.80 13.70 13.99

Sub-Total 105.37 107.04 108.55

Overall Annualised Needs 288.88 258.08 245.12

4.5.2. Road Fund Maintenance Gap

The required unconstrained road maintenance cost was estimated to range between US$ 245 million per year to US$ 289 million per year. This is in comparison to the maintenance funding of US$ 112.66 million based on the road fund allocations for the 2011/2012 financial year (see Table 20).


46

Table 20: Uganda Road Fund Allocations for the 2011/2012 Financial Year (in millions US $)

Agency Road Maintenance

Mechanical Workshops

Administration Total

URF 3.36 3.36

UNRA 75.65 1.74 77.39

District Roads 16.06 16.06

CAR 2.94 2.94

Urban Roads 18.01 18.01

Mechanical 1.80 1.80

Total 112.66 1.80 5.10 119.55

Source: Budget Framework Paper 2011 Vote 118 Based on this information it is concluded that the maintenance gap ranges between US$ 132 million per year to US$ 176 million per year given the assumptions used in the analysis.


47

Where:

5. A REVIEW OF THE CURRENT FUND ALLOCATION APPROACH USED BY THE UGANDA ROAD FUND

5.1. Current Approach used in Budget Allocation

As an interim measure, for Financial Year 2010/2011 only, the Fund has allocated monies to designated agencies on a similar basis to the previous year. This process firstly allocates proportions of the total budget to road classes on an arbitrary basis that seeks to fund maintenance of national roads as fully as possible and allocates remaining funds to district, urban and community access Roads on the basis of formulae similar to those adopted by the Ministry of Finance hitherto. These formulae are of the form shown in Figure 2, Figure 3 and Figure 4.

5.1.1 District Roads

Figure 2: Current URF Formula for allocation of funds to District Roads

5.1.2 Urban Roads

Figure 3: Current URF Formula for allocation of funds to Urban Roads

Where:


48

Where:

5.1.3 Community Access Roads

Figure 4: Current URF Formula for allocation of funds to Community Access Roads

The formulae above follow from the Global allocation of funds by URF which for the 2010/2011 financial year is understood to have been as shown in Table 21.

Table 21: Global allocation of Funds by URF for the 2010/11 Financial Year

No. Programme Item Amount (UShs bn)

% of Total Remarks/Purpose of Release

1. Maintenance of National roads (by Uganda National Road Authority)

177.794 62.63% Routine and periodic maintenance of the National Road Network.

6.501 2.29% Operational expenses of UNRA.

Total for maintenance of National roads

184.295 64.92% To maintain an expanded network of approximately 21,000km.

2. Maintenance of District, Urban and Community Access Roads (DUCAR)

38.352 13.51% Routine and periodic maintenance of District Roads. (111No. Designated Local Government Districts).

7.0146 2.47% Removal of bottlenecks on Community Access Roads. (111No. Designated Local Government Districts) and their sub-counties.

17.447 6.15% 169 Town Councils.

25.549 9.00% Routine and periodic maintenance of Municipality roads (including for Kampala City Council divisions). 27No. Municipalities including 5KCC divisions.


49


% of Total Remarks/Purpose of Release

Regional Mechanical workshops

4.298

1.51% For repair and maintenance of equipment and plant for Districts and Municipalities.

Total for maintenance of DUCAR network

92.658 32.64% Maintenance of DUCAR network and Regional Mechanical Workshops.

3. Items administered by the Secretariat

4.769 1.68% Administrative expenses of the URF Secretariat.

0.937 0.33% Research.

1.221 0.43% Such activities relevant to maintenance of public roads as may be determined by the Board.

Total for URF Secretariat 6.926 2.44%

Grand Total 283.88 100.00%

(Source: URF OYRMP and AREP 2010/11)

As a result, the distribution of funds allocated to routine and periodic maintenance of District, Urban and Community Access Roads between categories for the financial year 2010/11 was as given in Table 22.

Table 22: 2010/11 Percentage Allocations of funds to the DUCAR Network

Class of Road / Facility Allocation [% total funds]

District, Urban & Community Access Roads 32.64%

Allocated as follows:

District Roads 19.6%

Urban Roads 9%

Community Access Roads 2.47%

Regional Mechanical workshops 1.51%

Source: Uganda Road Fund OYRMP & AREP 2010/11

The Global allocations for the 2011/12 Financial Year are as given in Table 23.

Table 23: Global allocation of Funds by URF for the 2011/12 Financial Year


% of Total

Remarks/Purpose of Release

1 Maintenance of National roads (by Uganda National Road Authority)

177.79 63.28% Routine and periodic maintenance of the National Road Network.

4.08

1.45% Operational expenses of UNRA.


50


% of Total

Remarks/Purpose of Release

Total for maintenance of National roads

181.87

64.73% To maintain an expanded network of approximately 21,000km.

2 Maintenance of District, Urban and Community Access Roads (DUCAR)

37.74 13.43% Routine and periodic maintenance of District Roads. (112No. Designated Local Government Districts).

6.9 2.46% Removal of bottlenecks on Community Access Roads. (112No. Designated Local Government Districts) and their sub-counties.

17.17 6.11% 174 Town Councils.

25.15 8.95% Routine and periodic maintenance of Municipality roads (including KCCA). 22No. Municipalities

Mechanical imprest* 4.23 1.51% For repair and maintenance of equipment and plant for Districts and Municipalities.

Total for maintenance of DUCAR network

91.19 32.46% Maintenance of DUCAR network and Mechanical Imprest

3 Items administered by the Secretariat

6.090 2.17% Administrative expenses of the URF Secretariat.

0.9 0.32% Research.

0.9 0.32% Such activities relevant to maintenance of public roads as may be determined by the Board.

Total for URF Secretariat 7.890 2.81%

Grand Total 280.95 100%

Source: URF OYRMP and AREP 2011/12 The main advantage of the current allocation methodologies are attributed to their simplicity, comparability with allocation approaches used by other Government agencies such as Ministry of Water in the allocation of District Water and Sanitation Conditional Grant. The formulae though are not in sync with Section 22 (2) of the URF Act which requires that fund allocation be on the basis of conditions of public roads,


51

maintenance requirements, the length of the road network and the relevant traffic volume or an approved maintenance management tool.

5.2. Critique of the Current Allocation Formulae

The current road fund allocation formula considers both vertical allocation by road classes and horizontal allocation by road agencies.

5.2.1 Vertical allocation

This process firstly allocates proportions of the total budget to road classes on an arbitrary basis that seeks to fund maintenance of national roads as fully as possible and allocates remaining funds to Districts, Urban and community access roads on the basis of formulae similar to those adopted by the Ministry of Finance hitherto.

The main criticisms of this formula are the following:

(i) It assumes that the budget estimate for national roads is properly estimated in an objective manner based on an analysis that considers needs and reflects efficiency. It is not certain if this is true?

(ii) Seeks to fund maintenance of national roads “as fully as possible”. It is not clear how

this level of funding is decided or determined as a proportion of the total available budget.

(iii) The approach does not take into account the effects of funding level on the future road condition and road user effects. Therefore the Road Fund is not certain on whether or not the money is contributing effectively to the improvement of the road network. Thus the allocation is not transparent.

5.2.2 Horizontal Allocation

A critique of the horizontal allocation is provided below: (i) It is good that the formula takes into account some element of equity in terms of the

uniform allocation.

(ii) The districts road formula does not explicitly consider the extent of the road network in any particular district. A district can be big in size, but the road length density may be low. This would cause bias in fund allocation.

(iii) The urban road formula may be biased against the new or more rapidly expanding

urban areas since the formulae includes “previous asphalt” component.


52

(iv) Although population is used in the formulae as a proxy for road utilization, in Uganda it is important to consider income level of the population in order to assess the affordability of using motorized transport by the local population.


53

6. ROAD MAINTENANCE FUND ALLOCATION FORMULA DEVELOPMENT METHODOLOGY

6.1. Introduction The design of a strategy or methodology for the allocation of available funding for road maintenance cannot and should not be done without due consideration of the source of this funding. This connection between source and application is extremely important since it answers many of the questions and it simplifies the allocation problem in many respects.

6.2. Road User Charging The system of road user charges was recently implemented in Uganda which forms part of an international drive to reform and commercialize the road sectors by treating road sectors as a commercial business. The specific aim of road user charging is to shift the burden of road expenditure from the general taxpayers to the road user in a way that is economically efficient, equitable, and recovers the full cost of providing and maintaining roads. The system has the potential to promote efficiency in expenditure on roads and to maintain expenditure levels at lower and more stable aggregate levels than is presently the case. This can improve efficiencies in the engineering and construction industries with lower prices resulting. The general principles for road user charging are the following:

Principle of full cost recovery;

Principle of economic efficiency;

Principle of equity. A full explanation of these principles is given by Heggie (1995). What is important though is that road user charging are based on the above principles which must also be reflected in the allocation methodology for these principles to be meaningful; the charging in itself is only one step in the process that ends in the allocation and actual maintenance of roads. One should furthermore never forget the basic rationale behind road user charges; i.e. having the road user pay for road maintenance: The World Bank has proven that every U$1.0 of required road maintenance cost that has not been invested for this purpose will cost the road user U$3.0 in additional vehicle operating cost for the same period. This is an astounding figure that shows why road users should be in control of the payment for road maintenance – since they will suffer the consequences of poor maintenance directly. It also resulted in the saying that you pay for road maintenance whether you


54

have it or not. Some economists go that far to state that no country should have more roads (in length) than the road user can actually afford to maintain through road user charges. In laymen terms the meaning of the road user charges can be summarized by stating that the road user is paying for the full cost of road maintenance which must be expended on the same roads for which the charges are levied to ensure economic efficiency in terms of minimum (optimum) vehicle operation cost for the road user and minimum (optimum) road maintenance cost for the road agency while the process of allocation and expenditure must be equitable between different types of road users and between different road agencies. One can thus conclude that the allocation of available road maintenance funding is simply the reverse of the road user charging methodology that was used to determine the road user charges in the first instance! Although this is true on a very theoretical level, in practice it is unfortunately not quite that simple.

6.3. Incorrect Objectives and Principles for Allocation As soon as one recognizes the direct connection between the road user charging methodology and the allocation methodology, the formulation of an optimum allocation methodology becomes less complex. It is perhaps easier to firstly state the allocation objectives or principles that are NOT valid in the allocation of available funding: (1) This funding should not be used to pursue any land use or economic planning

objectives that are forward looking; i.e. to stimulate or encourage future growth and development in any particular area or region;

(2) This funding should not be used to rectify any past inequalities and shortcomings in terms of the provision and maintenance of roads;

(3) This funding should not be used to benefit any one road network above the other. This may be thrown upon since clearly the national road network is more important than the district road network? We respectfully suggest that this should not be the case for maintenance when the road user is paying for the maintenance. Think about it in the following way: What is difference in the importance of a national road section and a district road section (same standard) that both serve 300 vehicles per day from the perspective of the road user. We suggest that there is no difference and for long-term efficiencies in the sector both should be treated exactly the same.

(4) This funding should not be used to rectify the backlog in road maintenance; i.e. for new construction, partial reconstruction, or even rehabilitation simply because this was not included in the road user charges in the first instance.


55

Using the road user charges for this purpose will thus result in a shortfall in other areas, i.e. eliminating the backlog in one area will create a backlog in another area. For this reason it is suggested that alternative funding should be sourced to eliminate the backlog.

(5) Very little funding – probably insufficient to cover the actual maintenance requirements - should be allocated to community access roads and Class 5 streets in urban areas since these roads serve very little traffic. The charges collected on these roads are thus insufficient to maintain these roads. The allocation of sufficient funds from the road fund for their maintenance will thus result in a shortfall (backlog) in other areas. For this reason and based on the actual function of these roads they should be considered a social service which require additional maintenance funding from the consolidated budget.

The available road user charges (road fund) will always be limited. Expenditure on roads for which no charges was collected will result in a shortfall (backlog) in other areas. Inappropriate allocations (i.e. not linked to actual charging) will thus result in an endless process of trying to eliminate backlogs in one area and simultaneously creating backlogs in another area. The only way to move towards long-term sustainability and efficiency in the road sector is to allocate the funding for the purposes that it was collected in the first instance (which has a sound rationale) and then to cover the gaps with additional funding from development partners and the government.

6.4. Factors that Influence Road Maintenance Cost The typical factors that determine road maintenance cost are accurately known from the disciplines road engineering and pavement design technology and are hence divided into primary or key determinants and secondary or contributing factors: (A) PRIMARY OR KEY DETERMINANTS

Length of road;

Surface type namely Paved, Gravel, and Earth;

Prevailing traffic demand on the road as well as the percentage heavy vehicles;

Condition of roads typically divided into Good, Fair, Poor, and Very Poor;

Design standard of road; and

Maintenance strategy and approach; (B) SECONDARY OR CONTRIBUTING FACTORS

Climate of area – i.e. wet or dry;

Typical topography of area;

Geology and availability of suitable gravel and other road building materials;


56

Location in relation to nearest construction materials and equipment;

Availability of suitable road maintenance contractors;

Existing demand versus supply situation in the market and road sector;

International fuel prices that influence the cost of bitumen products and transportation cost;

The prevailing escalation rates; A much longer list can be formulated of all the secondary and contributing factors. However it suffices to say that the first six factors describes and determines more than 80% - and easily up to 90% - of the typical variability in road maintenance cost. It should be noted that a measure of aggregation was assumed, i.e. variability is reduced when all the gravel roads in one district is considered because outliers tend to cancel each other out. The actual road maintenance needs can thus be determined accurately in practice by using a number of pavement management models or methodologies. The allocation of funding can subsequently de done in proportion with the actual needs. However, this approach assumes the use of a Pavement Model on a network basis, which is in effect nothing more than a range of formulas integrated into a systematic process.

6.5. The Formula Development Methodology

6.5.1 Introduction If one assumes the availability and use of a Pavement Model for the total road network, the allocation method is straight forward and simply entails subdividing the available funding in proportion to actual needs. In the development of an allocation formula, the only real practical approach is to use sound technical input/parameters to allocate funding on an aggregate basis to the various road networks subdivided into the broad surface types. Five (5) road networks (based on jurisdiction) and the three (3) surface types (paved, gravel, and earth) were used in the development of the allocation methodology.

6.5.2 Proposed Allocation Structure A three-stage allocation structure is proposed as illustrated in Figure 5. The first step in the allocation process involves determination of the roads in maintainable state and defining the input data required. The three-stage allocation structure that follows the first step consists of the following:


57

Stage 1 – allocates available funding to surface types;

Stage 2 – allocates the funding per surface type between the road network jurisdictions;

Stage 3 – allocates the funding per road network and surface type to the designated authorities within each district, town councils, municipals, and sub-counties

There are three road surface types considered (i.e. paved, gravel and earth); five road network jurisdictions (i.e. National, District, Town Council, Municipal and Community). There are 111 districts, 174 town councils, and 23 municipalities including Kampala City Authority.


58

Figure 5: Flow Chart – The three stage allocation structure


59

6.5.3 Allocation Criteria It is recommended that funds should be allocated to roads in maintainable condition or state. Threshold values for road condition based on roughness given in Table 24 may be used to screen roads in maintainable state. The appropriate definition is that roads in poor and bad condition threshold form non-maintainable part of the network and the rest form part of the maintainable network.

Table 24: Ride Quality Threshold Values for Paved and Unpaved Roads

Condition Band

Description

Paved Roads IRI in m/km Unpaved Roads IRI in m/km

Range Representative

Value Range

Representative Value

Good

Ride very smooth and very comfortable; No/slight unevenness of the profile; No rutting or potholes

<4 3.5 <5 4

Fair Moderate unevenness of the profile; Moderate rutting but not potholes

4 – 5 4.5 5 – 9 7

Poor Frequent unevenness of the profile and significant rutting

5 - 7 6 9 - 16 12.5

Bad Severe unevenness on paved roads. Unpaved roads impassable except by 4WD

>7 9 >16 18

Source: UNRA visual condition assessment guidelines (2011). It is proposed that road fund allocation should be based on the following criteria depending on the allocation stage. Further details of this are provided in Sections 7.2, 7.3 and 7.4. Stage 1 1. Vehicle utilization – to reflect on the principle of user pays. The allocation

parameters considered here are: traffic volume measured in passenger car unit kilometers (PCU-Km), and traffic loading in equivalent standard axle load kilometers (ESAL-Km)

2. Road length – to reflect maintenance requirements in terms of extent/size. This also

takes into account the effect of non-traffic related elements on road deterioration (e.g. environmental factors) and routine maintenance needs.


60

3. Road asset value – to reflect the investment that needs to be preserved in the interest of national economy and wealth generation.

4. Optimum maintenance requirements - to reflect the fact that different road surface

types and standards require different levels of funding to maintain it to an acceptable condition. This is expressed in terms of the ratios of optimum maintenance cost for each road surface type. Road surface types that require higher funding for their optimal level of maintenance should receive higher allocations.

Stage 2 1. Vehicle utilization – to reflect on the principle of full cost recovery that maintenance

cost of roads should be recovered from road users. The most trafficked roads generate most revenue for the road fund and conversely should get higher allocations. The parameters considered are:


traffic loading in equivalent standard axle load kilometers (ESAL-Km) 2. Road functional class priority and the key role of supporting the most highly ranked

national objective of promoting economic growth and wealth creation. This is expressed in terms of relative weights for the different road networks (i.e. the designated road agencies). Road classes with higher relative weights should receive higher allocations.

3. Other sources of road maintenance funding – road agencies or networks that

possess other sources of raising funds (e.g. through vehicle parking charges, taxation, etc.) should receive less allocations.


61

Stage 3 For Town Council, Municipal and District networks the parameters considered are the following: 1. Vehicle utilization – the most trafficked roads deteriorate faster thus generating

higher maintenance needs therefore it is essential that these get higher allocations. The parameters considered are:


traffic loading in equivalent standard axle load kilometers (ESAL-Km) 2. Road length – to reflect maintenance requirements in terms of extent of the road

network. Designated authorities with bigger network size should receive higher allocations.

3. Equity – to reflect social concerns and needs of the population regardless of which

area they live in. Every designated authority should receive a minimum fixed amount of funding expressed as a percentage of the available fund for each network (urban, municipal or district).

4. Climatic factors - rainfall is a major climatic factor that influences road

deterioration. Roads in areas with high rainfall will deteriorate faster resulting into more frequent and higher maintenance requirements. These areas should therefore get proportionally higher allocations.

5. Variations in unit costs of road works across the country. There are several factors

that lead to variations in unit costs of works and this influences the total amount of road maintenance fund requirements by designated authorities. Designated authorities with high unit costs of works should get higher allocations, on average.

For Community and Access Roads the parameters considered include all those given above for urban, municipal and district networks but excluding vehicle utilization since all these roads are carry very low traffic volumes with little variations. Another important parameter considered for CAR maintenance fund allocation is population of the area. Population is considered a surrogate for travel demand and needs for social services. Areas with big population should get higher allocations.

6.5.4 Length and Traffic The two most significant parameters for the allocation of road maintenance funds are traffic and road length. Traffic need to be considered in terms of both volume and loading. Figure 6, determined by means of the RONET Model for Uganda assuming all other things to be equal, clearly shows the influence of increasing traffic on the


62

maintenance costs. The different gradients of the various ADT graphs are an indication that increasing length does not explain the maintenance cost fully. The data used was obtained from this study through secondary sources and include the data given in Tables 8 to 14, Chapter 4.

Figure 6: The effect of traffic and road length on gravel road maintenance cost

Maintenance Cost for Gravel Roads

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

200 400 600 800 1000

Length (km)

To

tal M

ain

ten

an

ce C

ost

(US

$/a

nn

um

)

5 ADT 20 ADT 65 ADT 200 ADT 650 ADT


63

7. THE ROAD MAINTENANCE FUND ALLOCATION FORMULA

7.1. Formula Development The allocation formulae proposed are deterministic, where funds are allocated as a precise value on the basis of mathematical functions of known relevant quantified variables. It is important to appreciate that models are simplified representations of reality used to study effects of policies, strategies, programmes and designs. The two general classes of deterministic models used for road management are mechanistic and empirical. Mechanistic models are based on first principles and knowledge of the physical laws governing the fundamental theories of behaviour of a system. They are usually very data intensive and rely on parameters which are difficult to quantify in the field. Empirical models are usually based on statistical analyses of locally observed behavior or trends, and may not be applicable outside the specific conditions upon which they are based. To develop the allocation formulae, the study team adopted a structured mechanistic-empirical approach, which is similar to that used by Paterson (1987) for developing road performance models. This is based on identifying the functional form and primary variables affecting each stage of fund allocation from both first principles and empirical information and then using various statistical techniques to quantify their impacts. This has the advantage that the resulting formulae combine both the theoretical bases of mechanistic models with the behaviour observed in empirical studies. The allocation criteria defined in Section 6.5.3 of this report were used as the basis for developing the road maintenance fund allocation formulae. The formulae developed for the allocation of road maintenance funds can be used for both vertical and horizontal allocation in the three-stage process defined in Section 6.5.2.

7.2. Stage 1: Allocation to Road Surface Types The assumptions made in developing Stage 1 allocation formula are the following: 1. Roads of different surface types deteriorate at different rates, and the rate of

deterioration depends on the level of vehicle utilization among other factors such as the climate. Also different categories of road users cause different levels of road deterioration which should be offset by maintenance interventions. The cost of maintenance intervention should be shared equitably among users. That is to reflect on the principle of user pays. The higher the level of utilization the higher the amount paid by the user. The allocation parameters considered there are:


64

traffic volume measured in passenger car unit kilometers (PCU_Km), and

traffic loading in equivalent standard axle load kilometers (ESAL-Km)

2. Maintenance requirements depend partly on the extent and size of the road

network. The larger the network the bigger is its maintenance requirements. 3. To reflect the investment that needs to be preserved in the interest of national

economy and wealth generation, it is important to take into account the asset value of the road type. The asset value will depend on the road surface type. Paved roads have higher asset value per kilometer length than gravel and earth roads.

4. In the absence of stringent budget constraints, the roads should receive both

technically and economically optimum and timely maintenance interventions. The cost associated with optimum maintenance is considered in the allocation formula to reflect the fact that different road surface types and standards require different levels of funding to maintain them to an acceptable condition. This is expressed in terms of the ratios of optimum maintenance cost for each road surface type. Road surface types that require higher funding for their optimal level of maintenance should receive higher allocations.

5. Traffic and Length are the two most important factors in the formula. Therefore the

percentage contribution of impact, measured in terms of the amount of funds allocated to each road surface type, will be higher for the parameters associated with these two factors. Determining costs of road use requires sub-dividing the road management costs into: Fixed costs associated with the management of the road network, which do not vary with road use, and Variable costs which are the additional costs incurred due to traffic flow and loading, which change with road use in terms of volume (vehicle-kilometers) and axle loading (cumulative equivalent standard axles).

From the above assumptions, a conceptual formula for Stage 1 allocation is proposed as follows:

Conceptual formula for allocation to road surface types – Equation 1

Ms = Ps*B Where Ms = allocation to road surface type s B = total available budget to URF


65

B

LCopt

LCoptK

AV

AVK

KMESAL

KMESALa

KMPCU

KMPCUaKM

S

Ss

Ss

S

S

S

S

S

S

S

S

SS *

*

*

_

_

_

_3

1

133

1

123

1

123

1

1111

Ps = a parameter defined as function ƒ(traffic, road length, asset value, maintenance cost) for road surface type s

Equation 1 can be taken as a simplified or an abridged version of the formula for Stage 1 allocation, which in full is given by the following expression:

Formula for allocation to road surface types – Equation 2

Where Ms = allocation to road surface type s B = total available budget to URF PCU_KMs = total traffic volume in passenger car units kilometers for each road

surface type s (s = paved, gravel or earth) ESAL_KMs = total traffic loading in equivalent standard axle loads kilometers for

each road surface type s AVs = asset value of road surface type s Copts = relative weight for optimal maintenance cost for road surface type s Ls = total length of road surface types a11 = coefficient for the impact of traffic volume (default value = 0.7) a12 = coefficient for the impact of traffic loading (default value = 0.3) K11 = traffic impact factor (default value = 0.5) K12 = asset value factor (default value = 0.15) K13 = maintenance cost and road length factor (default value = 0.35) The formula coefficients a11 and a12 are explained in Section 7.5 of this report. The K factors relate to Assumption 5 above. The values of K factors should add up to 1.0 (i.e. 100%) in the above formula. According to research done by Paterson and Archondo-Callao (1991), the share of variable costs rises from 24 per cent to 74 per cent in developing countries. This means that the value of traffic impact factor (K11) varies from 0.24 to 0.74 depending on the construction quality and pavement strength. In this study, it is assumed that the value of traffic impact factor should range from 0.45 to 0.6 (i.e. the middle part of the range of the share for variable costs) depending on the other competing factors included within each formula. In Stage 1 allocation it is assumed that the influence of Assumption 4 is greater than that of Assumption 3. Hence, this is reflected in the magnitude of K13 being greater than K12.


66

7.3. Stage 2: Allocation to Road Networks

The assumptions made in developing Stage 2 allocation formula are the following: 1. The most trafficked roads generate most revenue for the road fund and conversely

should get higher allocations. The parameters considered are for vehicle utilization:


traffic loading in equivalent standard axle load kilometers (ESAL-Km) This reflects on the principle of full cost recovery that maintenance cost of roads should be recovered from road users.

2. The formula should include road functional class priority defined in terms of the key

role of supporting the most highly ranked national objective of promoting economic growth and wealth creation. This is expressed in terms of relative weights for the different road networks (i.e. the designated road agencies) defined as the perceived proportional contribution of the road functional class to the overall national objective of promoting economic growth and wealth creation. Road classes with higher relative weights should receive higher allocations.

3. Although road agencies or networks that possess other sources of raising funds (e.g.

through vehicle parking charges, taxation, etc.) should receive less allocations, the assumption is made that this should not be considered in the conceptual formula. Instead this issue should be addressed within the overall allocation process at the stage of consultation with the Agencies as discussed in Section 7.6 and illustrated in Figure 10. Therefore no percentage contribution for agencies that have significant local revenue will be considered in Stage 2 formula.

4. Road length is included implicitly in Stage 2 formula through the parameters PCU-

Km and ESAL_Km. Hence, there is no need to include length again explicitly as a separate parameter in the formula to avoid the issue of redundancy or double counting of effects.

From the above assumptions, a conceptual formula for Stage 2 allocation is proposed as follows:

Conceptual formula for allocation to road surface types – Equation 3

Msj = Wj*Qj*Ms Where Msj = allocation to road network j (j = national, district, town council,

municipal or CAR)


67

s

S

Sj

Sj

S

Sj

Sj

jSj M

KMESAL

KMESALa

KMPCU

KMPCUaWM *

_

_

_

_5

1

225

1

21

Ms = allocation to road surface type s (obtained from Stage 1) Wj = relative weight based on perceived proportional contribution of

each road hierarchy and functional class to the overall national objective of promoting economic efficiency for road network j

Qj = a parameter defined as function ƒ(traffic, road length) Equation 3 can be taken as a simplified or an abridged version of the formula for Stage 2 allocation, which in full is given by the following expression:

Formula for allocation to road networks for various surface types – Equation 4

Where

Msj = allocation to road network j (j = national, district, urban, municipal or CAR)

Ms = allocation to road surface type s (obtained from Stage 1) PCU_KMs = total traffic volume in passenger car units kilometers for each road

surface type s (s = paved, gravel or earth) ESAL_Kms = total traffic loading in equivalent standard axle loads kilometers for

each road surface type s a21 = coefficient for the impact of traffic volume (default value = 0.7) a22 = coefficient for the impact of traffic loading (default value = 0.3) Wj = relative weight based on perceived proportional contribution of

each road hierarchy and functional class priority for promoting economic efficiency for road network j (default values are 0.28 for National, 0.20 for Town Council, 0.20 for Municipal, 0.20 for District, and 0.12 for CAR)

The formula coefficients a21 and a22 are explained in Section 7.5.1. The relative weight W relates to Assumption 2 above, and is explained in Section 7.5.3 of this report.

7.4. Stage 3: Allocation to Designated Agencies Allocation to designated agencies at Stage 3 uses formulae developed separately for town councils, municipals and district roads and for community and access roads. The key assumptions made in developing Stage 3 allocation formulae are given below.


68

For Town Councils, Municipal and District networks: 1. The most trafficked roads deteriorate faster thus generating higher maintenance

needs therefore it is essential that these get higher allocations. The parameters considered are:


traffic loading in equivalent standard axle load kilometers (ESAL-Km) 2. Road length should be included in the formula to reflect maintenance requirements

in terms of extent of the road network. Designated agencies with bigger network size should receive higher allocations.

3. Traffic and Length are the two most important factors in the formula. Therefore the

percentage contribution of impact, measured in terms of the amount of funds allocated to each designated agency, will be higher for the parameters associated with these two factors.

4. Social concerns and needs of the population should be considered regardless of

which area they live in. Every designated agency should therefore receive a minimum fixed amount of funding expressed as a percentage of the available fund for each network (town council, municipal or district).

5. Rainfall is a major climatic factor that influences road deterioration. Roads in areas

with high rainfall will deteriorate faster resulting into more frequent and higher maintenance requirements. These areas should therefore get proportionally higher allocations.

6. Variations in unit costs of road works across the country should be included in the

allocation formula. There are several factors such as road density, topography, geology, terrain that lead to variations in unit costs of works and this influences the total amount of road maintenance fund requirements by designated agencies. Designated agencies with high unit costs of works should get higher allocations, on average.

7. For Community and Access Roads the parameters considered include all those given

above for town council, municipal and district networks but excluding vehicle utilization since all these roads carry very low traffic volumes with little variations. Another important parameter considered for CAR maintenance fund allocation is population of the area. Population is considered a surrogate for travel demand, attractiveness of an area and needs for social services. Areas with big population should get higher allocations.


69

SjZZN

Z

SZ

SZ

N

Z

SZ

SZ

N

Z

SZ

SZ

SZ MKN

UFCF

L

LK

KMESAL

KMESALa

KMPCU

KMPCUaKM *

1**

_

_

_

_33

1

32

1

32

1

3131

Conceptual formula for allocation to town councils, municipalities and districts –

Equation 5

Msz = Qz*Rz*E*Msj Where Msz = allocation to designated agency z for road surface type s Msj = allocation to road network jurisdiction j and road surface s

(obtained from Stage 2) Qz = a parameter defined as function ƒ(traffic, road length) Rz = a parameter defined as function ƒ(climate, unit cost), see

Assumption 5 above E = a parameter defined as function ƒ(equity, social concerns), see

Assumption 4 above Equation 5 can be taken as a simplified or an abridged version of the formula for Stage 3 allocation to town councils, municipals and districts, which is given in full by the following expression:

Formula for allocation to designated agencies – Equation 6

Where Msz = allocation to designated authority z for road surface type s Msj = allocation to road network jurisdiction j and road surface s

(obtained from Stage 2) N = number of designated authorities (N=111 for Districts, N=23 for

Municipalities, N=174 for Town councils but the data is presently aggregated by district for the 111 districts)

PCU-KMs = total traffic volume in passenger car units kilometers for each road surface type s and designated authority z

ESAL-Kms = total traffic loading in equivalent standard axle loads kilometers for each road surface type s and designated authority z

Lsz = total length of road surface type s under designated authority z CFz = climatic factor for area z UFz = unit cost factor for area z a31 = coefficient for the impact of traffic volume (default value = 0.7)


70

SjZZN

Z

SZ

SZ

N

z

z

zSZ MK

NUFCF

L

LK

POP

POPKM *

1** 43

1

42

1

41

a32 = coefficient for the impact of traffic loading (default value = 0.3) K31 = traffic impact factor (default value = 0.6) K32 = length impact factor (default value = 0.20) K33 = Equity factor (default value = 0.20) The formula coefficients a31 and a32 are explained in Section 7.5. The K factors relate to Assumption 3 above. The values of K factors should add up to 1.0 (i.e. 100%) in the above formula. See also explanation of K-factors in Section 7.2. In Stage 3 allocation for town councils, municipals and districts it is assumed that the influence of Assumption 4 is equal to that of Assumption 2. Hence, this is reflected in the magnitude of K33 being equal to that of K32.

Conceptual formula for allocation to Community Access Roads – Equation 7

Msz = Dz*Rz*E*Msj Where Msz = allocation to designated agency z for road surface type s Msj = allocation to road network jurisdiction j and road surface s

(obtained from Stage 2) Dz = a parameter defined as function ƒ(population, road length) Rz = a parameter defined as function ƒ(climate, unit cost) see

Assumptions 5 and 6 above E = a parameter defined as function ƒ(equity, social concerns) Equation 7 can be taken as a simplified or an abridged version of the formula for Stage 3 allocation to community access roads, which in full is given by the following expression:

Formula for allocation to Community Access Roads – Equation 8

Where Msz = allocation to designated authority z for road surface type s Msj = allocation to road network jurisdiction j (j=CAR) and road surface s

(obtained from Stage 2) N = number of designated authorities (N=111 districts) POPz = population of area z


71

Lsz = total length of road surface type s under designated authority z CFz = climatic factor for area z, see default values in Table 25 UFz = unit cost factor for area z, default value = 1.0 K41 = population impact factor (default value = 0.45) K42 = length impact factor (default value = 0.35) K43 = Equity factor (default value = 0.20) The K factors relate to Assumption 7 above. The values of K factors should add up to 1.0 (i.e. 100%) in the above formula. In Stage 3 allocation for community access roads it is assumed that the influence of population is greater than that of road length and the influence of road length is greater than that of equity. Hence, this is reflected in the value of K41 being greater than that of K42 and K42 being greater than K43.

7.5. Formula Coefficients and Factors Determining costs of road use requires sub-dividing the road management costs into:

Fixed costs associated with the management of the road network, which do not vary with road use

Variable costs, which are the additional costs incurred due to traffic flow and loading, which change with road use in terms of volume (vehicle-kilometers) and axle loading (cumulative equivalent standard axles)

One of the key findings from Paterson and Archondo-Callao (1991) is that approximately 13 per cent of expenditures for maintenance and renewal are due to load-related deterioration, 33 per cent are due to vehicle-related deterioration, and the remaining 54 per cent are fixed costs. On the basis of this information, the traffic related expenditures can be split into two portions:

i) loading-related portion accounts for 28.3% (i.e. 13*100/[13+33])

ii) volume-related portion accounts for 71.7% (i.e. 33*100/[13+33])

The formula coefficients a12, a22, a32 are traffic load related and for this study their value is rounded up to 0.3, that is accounting for 30% of expenditures for maintenance and renewal. The formula coefficients a11, a21, a31 are traffic volume related and their value is rounded up 0.7, that is accounting for 70% of expenditures for maintenance and renewal.

The allocation of maintenance funding between the surface types is dependent on the average maintenance cost per surface type expressed in terms of a relative weight. Although typical industry standards can be used for this purpose, the actual appropriate average maintenance costs for the network in question should rather be used. The


72

advantage is that the actual weights take into account the many local factors that influence maintenance costs in addition to length and traffic. These additional factors are:

Condition of roads;

Design standard of roads; and

Maintenance strategy and approach; The average annual maintenance cost per km for the Uganda Road Network was assumed to be the following – based on typical unit rates for the African Continent:

Paved Roads = US$ 5,000 / km;

Gravel Roads = US$ 1,500 / km;

Earth Roads = US$ 150 / km; By normalizing the above values, that is by dividing each of the three values by 5,000 (the average annual maintenance cost paved roads) the relative weights of 1.00 : 0.30 : 0.03 on average road maintenance costs are determined for paved, gravel, and earth roads, respectively. These are the values to be used for parameter Copts in Stage 1 allocation formula given in Section 7.2. Road hierarchy and functional class priority in promoting economic efficiency is considered in Stage 2 allocation formula using scores for each road network. The score S is defined as the proportional contribution of each road class (or network) to the overall national objective of promoting economic growth and wealth creation. The scores (S) were determined based on the findings of consultation with Uganda road sector stakeholders reported by I.T. Transport (2011). The executive summary to the report is provided in Appendix 5. Economic growth was identified as the most important national objective of the road network as shown in Figure 7 and the relative weights of the four road classes associated to this objective are shown in Figure 8. The scores S from Figure 8 (0.35 for National roads, 0.25 for Town Council and Municipal roads, 0.25 for District roads, and 0.15 for CAR) were used to derive relative weights W for stage 2 allocation formula. The derived relative weight values are 0.28 for National roads, 0.20 for Town Councils, 0.20 for Municipal roads, 0.20 for District roads, and 0.12 for CAR.


73

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Economic growth

Growth of the Agricultural Sector

Poverty reduction

Exports promotion

National Cohesion

Environmental sustainability

Re-distribution of poverty/ Social equity

Regional (EAC) Cohesion

Importance

Figure 7: Importance of National Objectives Source: Adapted from IT Transport, 2011.

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Econ

omic

Gro

wth

Pove

rty

Red

ucti

on

Re-

dist

ribu

tion

of

pove

rty/

Soc

ial

equi

ty

Expo

rts

prom

otio

n

Envi

ronm

enta

l

Sust

aina

bilit

y

Nat

iona

l Coh

esio

n

Reg

iona

l ( E

AC

) C

ohes

ion

Gro

wth

of

the

Agr

icul

tura

l sec

tor

Rel

ativ

e W

eigh

ts

National District Urban Community Access

Figure 8: Scores of National Objectives by Road Network Source: Adapted from IT Transport, 2011.


74

Climate factors impact significantly on road deterioration. The value of climate factor CF depends on the amount of mean annual rainfall in each area under a designated authority. For this study the values of CF have been determined using HDM-4 analysis of road deterioration as follows:

(a) Five climate zones (given in Table 25) based on typical average annual rainfall in mm were obtained from the rainfall map of Uganda given in Figure 9.

Figure 9: Annual Rainfall (mm) Map of Uganda

Source: Department of Geography at Makerere University, 2003.

(b) HDM-4 was used to determine the rate of road deterioration on a representative road section for each typical annual rainfall. The deterioration rates are summarized in Table 25 and detailed in Appendix 5. The data used was obtained from a previous study of unsealed road deterioration in Uganda carried out by TRL and reported in International Labour Organization

(2006). (c) The deterioration rates were normalized to derive the relative Climate

Factors (CF) given in Table 25.


75

Table 25: Typical Road Deterioration Rates by Rainfall Amount

Typical Annual Rainfall (mm) Deterioration Rate

(∆IRI/Year) Relative Climate Factor

(CF)

1526 0.2996 1.00

1465 0.2974 0.99

1288 0.2911 0.97

1246 0.2891 0.96

650 0.1491 0.50

Source: This Study (2011)

Data on unit cost variations across the country were not available at the time of writing this report. These data are necessary for determining the values of unit rate factor UF. The default value of this parameter is 1.0. This will be updated when the data on unit rate variations become available. To date no reliable data is available to the study team. The formulae input data can be considered in two groups:

Variable data; and

Fixed coefficients and factors (temporarily fixed) Table 26 provides sources of the variable input data required for the proposed allocation formula.

Table 26: Variable Input Data Sources

Data Type Units Possible Sources of Data

Road Length Km Designated road agencies

Asset Value US$ or Ushs

Estimated from replacement costs of roads and the extent (length) of roads by surface type.

Traffic Volume PCU_km Traffic volume should be requested from designated agencies.

Traffic Loading ESAL_km Traffic loading should be elicited from reports on axle load surveys available from the designated agencies.

Population No. Population data may be obtained from Uganda Bureau of Statistics.


76

Table 27 provides guides to calibration approaches for the fixed formula coefficients and factors.

Table 27: Calibration Approaches to Formulae Coefficients and Factors

Formula Coefficients and Factors Calibration Approaches

Stage 1 Equation 2

Traffic volume coefficient a11 Conduct a study on road use cost in which variable costs and fixed costs are examined using historic data available supplemented by additional data to be collected. Determine the load-related and volume related portions of total variable cost using the methodology described in Paterson and Archondo-Callao (1991). The volume related portion provides the value of a11 and the load-related portion provides the value of a12.

Traffic loading coefficient a12

Traffic impact factor K11 Conduct a study on road management cost in which variable costs and fixed costs are examined using historic data available supplemented by additional data to be collected. The portion of variable cost in relation to total road use cost will give the value of K11. An appropriate methodology for this is described Paterson and Archondo-Callao (1991).

Asset value factor K12 This factor should be determined based on consultations with Uganda road sector stakeholders using methods of questionnaires, interviews and workshops.

Maintenance cost and road length factor K13

This factor should be determined based on a comparative study on expenditures on routine and periodic maintenance. The results will provide useful information to determine the magnitude of K13 in relation to K11.

Stage 2 Equation 4

Traffic volume coefficient a21 As described for Stage 1 above.


Relative weight W The relative weights for each functional road class should be determined based on consultations with stakeholders as described in Section 7.5.3.

Stage 3 Equation 6 and

Traffic volume coefficient a31 As described for Stage 1 above.


Traffic impact factor K31 First K31 should be determined as for K11


77

Equation 8 Length impact factor K32 described in Stage 1. Then the value adjusted based on the relative importance of the three factors to be determined through consultations with road sector stakeholders in Uganda.

Equity factor K33

Population impact factor K41 The values of these factors should be determined based on consultations with Uganda road sector stakeholders using methods of questionnaires, interviews and workshops. The objective will be to determine the relative importance of the three factors.

Length impact factor K42

Equity factor K43

Climatic factor CF Conduct a study on road deterioration for the different climate zones defined based on rainfall distribution in Uganda. Apply the method used in this study as described in Section 7.5.4 to determine CF for the climate zones defined.

Unit cost factor UF Carry out a study to collect data on unit costs of road works from different parts of Uganda. In particular a matrix should be defined to encompass the key variables which likely do influence unit rates (e.g. terrain, road density, topography, geology, socio-economic activities). The results of data analysis should yield the unit cost factors UF for various parts of Uganda.


78

7.6. Operationalization of the Allocation Formula

7.6.1 Overall Allocation Process

The overall allocation procedure is illustrated in Figure 10. This procedure can be summarized into the following steps:

Step 1: Determine the roads in maintainable state and define the required input data, including the available fund to be allocated.

Step 2: Apply Stage 1 allocation, which is a vertical allocation to the different road surface types using the formula given in Section 7.2.

Step 3: Apply Stage 2 allocation to the different road networks, which is a horizontal allocation using the formula provided in Section 7.3.

Step 4: Compare results obtained to the estimated need of each road network and assess the impact of these allocations on the network performance.

Step 5: Consult with the agencies and address issues such as percentage contribution of for agencies that have significant local revenue, counterpart funding for CARs, etc.

Step 6: Assess whether or not these allocations are satisfactory. If the outcome appears to be unsatisfactory then some fine-adjustments need to be made on the values of the model coefficient and factors and the allocation process repeated from Step 2 onwards. If the allocations are satisfactory then the allocation process continues.

Step 7: Apply Stage 3 allocation using the formulae provided in Section 7.4.

End for this Financial Year, and continue assessment and monitoring.


79

Figure 10: Overall Allocation Process

Yes

Determine roads in maintainable state and

define input data and

Available Budget

Start

Stage 1 Allocation

Stage 2 Allocation

Compare Allocations and Impacts

Adjust Formula Coefficients

Are Allocations

Satisfactory?

Stage 3 Allocation

END

Consult with Agencies

No

See Tables 24 and 26


80

7.6.2 Steps Within the Three Stage Allocation Formula Stage 1:

STEP 1 Determine / update the total kilometers of paved roads, gravel roads, earth roads for the Uganda road network.

STEP 2 Determine the average annual maintenance cost per km for paved roads, gravel roads, earth roads for local conditions. This has been determined from output from HDM-4, and expressed in relative weight ratios.

STEP 3 Determine the relative allocation to surface types using the Stage 1 formula.

Stage 2:

STEP 4 Determine / update the total kilometers of roads for the various road networks divided between paved roads, gravel roads, and earth roads for the Uganda road network. These should be determined for paved national roads, gravel national roads, earth national roads, and similarly for the gravel and earth roads.

STEP 5 Determine / update the total traffic volume (in PCU-km) and total traffic loading (in ESAL-km) for the various road networks divided between paved, gravel, and earth roads. These should be determined for paved national roads, gravel national roads, earth national roads, and similarly for the gravel and earth roads.

STEP 6 Determine the relative allocation to road networks using the Stage 2 formula.

Stage 3:

STEP 7 Determine / update the relative distribution of length (km) per road link for each designated authority for the various surface types.

STEP 8 Determine / update the total traffic volume (in PCU_km) and total traffic loading (in ESAL_km) for the various road networks divided between paved, gravel, and earth roads. These should be determined for each designated authority.

STEP 9 Determine the relative allocation to designated authorities using the Stage 3 formula.


81

7.7. Formula Coefficients and Factors It is important to note the following issues relating to the use of the proposed Allocation Procedure:

1. The quality of input data into the formula should be improved and available information – received from designated agencies – should be updated annually.

2. Two important network characteristics are the existing traffic levels and existing road conditions. Data in this regard was obtained from several available sources but are by no means complete. It was generally possible to distinguish between road networks and surface types, but not necessarily between designated agencies. Aggregated data was thus used in this regard when disaggregated data was not available.

3. Accurate traffic demand data is not a prerequisite, but only in terms of traffic levels which can easily be estimated. A roving traffic counting program is nevertheless recommended for all the designated agencies.

4. A lack in any available data is not a fatal flaw for the implementation of the proposed three-stage allocation procedure. Realistic assumptions will still outweigh any factor-based allocation procedure.

5. It is important to update the unit maintenance costs per work item with the actual unit costs that are experienced in different parts of Uganda.

6. Finally, the connection between Maintenance Scenarios & Costs and Road User Charges should be actively pursued with the view to match and maintain a balance between available road maintenance funding and required road maintenance costs; i.e. introducing sustainability into the road sector.


82

8. APPLICATION OF THE FORMULA

8.1. Introduction

The application of the formula described in the previous chapter is demonstrated at allocation Stages 1 and 2 assuming an available budget of UShs 400 billion. A meaningful demonstration of allocations at stage 3 requires input data on population for each designated district, urban area and municipality.

The worked example presented here utilizes formula coefficients described in Chapter 7.

8.2. Worked Example

8.2.1 Stage 1 Allocation

Inputs:

Table 28: Stage 1 Input Assumptions

Road Category

PCU-km ESAL-km

(in millions) AV

(in million US$) Copts L

Paved 18,816,415 6.91 1,519 1.00 5,754

Gravel 8,510,238 3.12 197 0.30 52,946

Earth 631,585 0.23 142 0.03 74,214

Notes to Table 28: 1. PCU-km is the total traffic volume in passenger car unit kilometers for each road

surface type (paved, gravel or earth). Values given in Table 28 were derived from average traffic volume given in Chapter 4 and PCU assumptions provided in Table 14.

2. ESAL-Km is the total traffic loading in equivalent standard axle loads kilometers for each road surface type. Values shown in Table 28 were derived from traffic data given in Chapter 4.

3. AV is the asset value of road surface type. AV values in Table 28 are based on findings of a study undertaken to apply the ROAD Network Evaluation Tools model (RONET) in Uganda.

4. Copts are the relative weights for optimal maintenance cost for road surface type s (s = paved, gravel or earth)

5. Ls denotes the Uganda Road Network length by surface type s


83

Vertical Allocation: The relative allocations at stage 1 are given in Table 29 and Figure 11.

Table 29: Stage 1 Relative Allocations

Road Type (S)

Traffic Component Asset Value Proportion

Maintenance Costs and Length Stage 1

Allocation in Percentages

Allocations in Billion

Ushs Volume

Proportion (V)

Loading Proportion

(E) (V+E) (Copts*Ls)

Relative (Copts*Ls)

Paved 0.47 0.20 0.67 0.82 5,754 0.24 54% 217

Gravel 0.21 0.09 0.30 0.11 15,884 0.67 40% 160

Earth 0.02 0.01 0.02 0.08 2,226 0.09 6% 22

Total 0.70 0.30 1.00 1.00 23,864.34 1.00 54% 400.00

217,440,684

160,404,724

22,154,592

Paved Gravel Earth

Allocation to Surface Class

Figure 11: Relative Allocation by Road Surface Type (Stage 1)

8.2.2 Stage 2 Allocation Inputs: The inputs to the allocation formula at Stage 2 are provided in Table 30 by surface type and road network. These inputs must be consistent with those used at Stage 1.


84

Table 30: Stage 2 Inputs

Road Category

Notation National District Municipality Town

Council CAR Total

Paved PCU-km 11,441,433.1 0 1,671,014.8 3,364,491.0 0.0 16,476,938.9

ESAL-Km 4.2 0 0.6 1.8 0.0 6.7

Gravel PCU-km 3,302,825.1 4,531,642.4 143,518.9 312,444.5 219,806.9 8,510,237.8

ESAL-Km 1.2 1.7 0.1 0.2 0.1 3.2

Earth PCU-km 0.0 20,583.8 61,759.4 62,128.0 487,482.6 631,953.8

ESAL-Km 0.00 0.01 0.02 0.05 0.18 0.25

Relative Allocation: Relative allocations derived using Stage 2 formula is summarized in Table 31 and Figure 12.

Table 31: Stage 2 Allocation by Road Network in Billion Ushs

Road Type National District Municipality Town

Council CAR Total

Paved 162 0 17 39 0 217

Gravel 76 74 2 6 2 160

Earth 0 1 3 4 14 22

Total 238 75 22 48 16 400

% Allocations 59% 19% 6% 12% 4% 100.00%


85

59%

19%

6%12%

4%

National District Municipalities Town Councils Community Access

Roads

Allocation to Road Class

Figure 12: Relative Allocation at Stage 2 by Road Network

8.3. Sensitivity Analysis

Sensitivity of the proposed model to changes in key input variables of traffic volume, traffic loading, asset value and road length was assessed and the results are illustrated in Figure 13. Considering Figure 13, the horizontal axis represents the percentage change in input parameter (i.e. traffic volume, traffic loading, asset value or road length) while the vertical axis denotes the percentage change in paved roads allocation. Traffic volume was found to be the most sensitive parameter with a 50% change in input value resulting in a 6% change in allocation. The results suggest that the proposed formula is generally robust to changes in key input variables.


86

-6.0%

-5.0%

-4.0%

-3.0%

-2.0%

-1.0%

0.0%

1.0%

2.0%

3.0%

-50.0% -40.0% -30.0% -20.0% -10.0% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0%

Pe

rce

nta

ge

Ch

an

ge

in P

ave

d R

oa

d A

llo

cati

on

Percentage Change in Input Parameter Values

Traffic Loading Traffic Volume Asset Value Length

Figure 13: Sensitivity of Key Input Variables

8.4. Discussion of Results

This formula-based approach is relatively simple and although it does not require any prior knowledge of the actual maintenance cost, knowledge of the relative maintenance cost per km for the various surface types is required. This can be based on typical industry standards for the African Continent but actual local cost will ensure an improved allocation. The largest disadvantage of this methodology however is that a direct linkage is not created and maintained over time between the actual modeled maintenance costs and the allocated funding.

8.5. Road Map for Transition from Formula Based Approach to Needs Based Approach

It is proposed that the road map for transition from an indirect formula based approach to a needs based approach should comprise, at least, the following five activities:

1. Review of current state of affairs and consultations with road sector stakeholders in Uganda

2. Assessment of data needs and data collection

3. Adaptation and calibration of the analytical tools to be used

4. Conducting a Pilot Study to demonstrate how the system will work

5. Training of staff at URF and other stakeholders

The outline of the road map is provided in Table 32.


87

Table 32: Road Map for Transition from Formula Based to Needs Based Approach

No. Activities Description Outputs Typical

Duration

1

Review current

State Review and

consultation with

Stakeholders in

the Ugandan road

sector

Assessment of appropriate tools for

estimating maintenance needs and

allocation of funds. The tools to be

assessed may typically included: the

Highway Development and

Management System (HDM-4),

Deighton Transport Infrastructure

Management System (dTIMS), and

Road Networks Evaluation Tools

(RONET)

This activities includes consultation

of key stakeholders

Appropriate

framework and

tools for

implementation of a

needs based

approach.

2 to 4 Weeks

2

Assessment of

data needs and

data collection

Types of data required to

implement a needs based approach

will largely depend on the tools

selected, therefore:

Data necessary to calibrate and adapt the selected tools will need to be determined.

Data currently held by designated agencies should be reviewed to determine their appropriateness and reliability.

URF should request designated agencies to collect historical (time-series) data in a consistent format.

Data for use in

calibration and

adaptation of

selected tools

4 to 5 Years,

depending on

current

progress with

data

acquisition by

the Agencies

3

Adaptation and

Calibration of the

analytical tools

selected

Adaptation refers to configuration

and customisation of key default

data used in road management

tools to local conditions, standards

and practices. The primary objective

of adaptation is to make the analysis

Calibrated and

configured tools

Default data sets.

1 to 2 Months


88

results from the model relevant and

compatible to the particular

environment it is being used.

Calibration is aimed at improving

the accuracy of predictions by the

models embedded in selected tools.

4 Pilot Study

A pilot study is recommended to

validated and refine the tools and

methodologies proposed.

Refined tools and

acceptance testing. 2 to 4 Weeks

5 Training

Training of URF staff (primary users)

on the concepts and

implementation of the needs based

approach. There should also be

training of the stakeholders (or

secondary users of the system)

Trained staff 1 to 2 Weeks


89

9. CONCLUSIONS AND RECOMMENDATIONS

9.1. Key Study Conclusions

Based on the road maintenance finance review, the URF is significantly underfunded with a current allocation of 30% of total URA collections from road user charges to road maintenance work annually. The current approach for budget estimates at the designated agencies is constrained by the ceilings given to the URF by the MOFPED which limits the best approach (Bottom-up approach) where the needs are determined each year and funding sought to meet the road maintenance needs. The actual road maintenance needs derived from HDM-4 depicts a gap in relation to the actual available budget from MOFPED creating a funding gap which can only be closed by an increase in the total funding allocated to URF by MOFPED. The current formula in use by URF has a number of weaknesses as outlined above. The current formula in use by URF has a number of weaknesses which include under the vertical allocation; the assumption that the budget estimates for national roads are properly and objectively derived using an analysis that considers needs and reflect efficiency which is not accurate, the formula seeks to fund maintenance of roads fully yet it is not clear how the level of funding is decided as a proportion of the total available budget and the formula does not take into consideration the effects of funding levels on the future road condition and road users. Horizontally though the formula takes into account equitable allocation of funds across designated agencies it does not precisely consider the extent of the road network in any particular district making it subjective in terms of district size and road length density. To demonstrate the application of the formula developed, a worked example has been presented to allocate an assumed available budget of UShs 400 billion. The results indicated that for vertical allocation to road surface types, paved roads should get 54 percent, gravel roads 40 percent and earth roads 6 percent. For horizontal allocation to road network and designated agencies, UNRA should get 56 percent, District 24 percent, Urban 15 percent, and CAR 4 percent. This can be compared with the 2010/2011 allocation as follows: UNRA should get 67 percent, District 14 percent, Urban 16 percent, and CAR 3 percent.

9.2. Key Study Recommendations

In order for the URF to operate efficiently and meet the required road maintenance demand the current funding should be boosted by a 58% markup moving it from 30% allocation from MOFPED to 88% allocation of total URA Collections under road user charges which was UShs 885.28 billion in FY 2010/11.


90

Based on the road maintenance needs assessment undertaken using the HDM-4 tool the actual road maintenance needs for the first 5 years annualized across each of these years is US$289 million, for a 10 year band the annuity would reduce to US$ 258 million, and finally an annualized US$ 245 million for a 20 year spread if considered in order to deliver the required road standards, condition targets, and needs. It is recommended that URF considers for adoption the proposed “Three-Stage Allocation Formula” which addresses the weaknesses with the current formula in use. To obtain more accurate allocations the URF will have to update their database and obtain input data on the factors that impact on the formula that were not readily available during the assignment e.g. data on unit cost factor variations across the country. The following points should be noted:

1. The quality of input data into the formula should be improved and available information received from designated agencies should be updated annually.

2. Accurate traffic demand data is not a prerequisite, but only in terms of traffic levels which can easily be estimated. A roving traffic counting program is nevertheless recommended for all the designated agencies.

3. A lack in any available data is not a fatal flaw for the implementation of the proposed three-stage allocation procedure, since the formula is based on realistic assumptions.

4. The connection between Maintenance Scenarios & Costs and Road User Charges should be actively pursued with the view to match and maintain a balance between available road maintenance funding and required road maintenance costs; i.e. introducing sustainability into the road sector.

5. A road map for transition from a formula based approach to a needs based approach has been proposed.

The conclusion and recommendation are based on the following rationale, in sequential order:

That a critical review of available literature on this subject has been undertaken to inform the study

Useful lessons have been learned from other countries that have used formula based allocation methods before

The formula development methodology is logical and practical, yet based on sound fundamental theories

The underlying assumptions used for the formula development are realistic

The application of the formula to allocate a given budget using the spreadsheet developed appears to yield credible results

Sensitivity analysis conducted on the key input variables has produced results which prove the allocation formula to be robust.


91

REFERENCES

Department of Geography Makerere University Uganda et al. (2003). Assessing the Impacts of Climate Change and Variability on Water Resources in Uganda: Developing an Integrated Approach at Sub-regional Scale. START Project, Final Report.

FHWA, (1999), Asset Management Primer, Office of Asset Management, United States Department of Transportation.

Heggie, I. (1995) Management and Financing of Roads: An Agenda for Reform. World Bank Technical Paper 275. World Bank. Washington D .C.

Howard, Ronald A. (1983), “The Evolution of Decision Analysis,” in The Principles and Applications of Decision Analysis, Volume 1. Ed. Ronald Howard and James Matheson. Strategic Decisions Group,: pp 6-15.

IT Transport (2011), Consulting Services for Reviewing and Updating UNRA’S Investment Appraisal Tools and Procedures. Interim Report.

Ministry of Works and Transport Ministerial Budget Policy Statements 2010/2011

Ministry of Works Housing and Communications, Strategic Analysis of Uganda Road Network using HDM-4, 2006.

Odoki, J.B. and Kerali, H. (2000) Highway Development and Management (HDM-4) Technical Reference Manual Volume 4. PIARC, Paris.

Paterson, W D O and R Archondo-Callao, 1991. Estimating road use costs. World Bank Infrastructure and Urban Development Discussion Paper. Washington DC: unpublished.

Procedural Guide to Economic Road feasibility Studies, MWHC, March 2006

TAHAL Consulting Engineers, (March 2005) Road Sector Development Programme Phase 2 Project (RSDP) – Consultancy Services for the Preparation of the National Transport Master Plan Including a Master Plan for the Greater Kampala Metropolitan Area, MWHC.

Uganda Bureau of Statistics (June 2010), 2010 Statistical Abstract

Uganda Revenue Authority Annual Revenue Bulletin Report 2006/07

Uganda Revenue Authority Annual Revenue Bulletin Report 2007/08

Uganda Revenue Authority Revenue and Trade Performance report July to June FY 2008/09

Uganda Road Fund Act (2008)

Uganda Road Fund OYRMP and AREP 2010/11

Uganda Road Revenue Act (1991)

UK Department for Transport, TAG unit 3.9 (2011). Detailed Guidance on Major Scheme Appraisal in Local Transport Plans


92

International Labour Organisation (2006). Increased Application of Labour based methods through appropriate Engineering Standards.


93

APPENDICES


94

APPENDIX 1 – ASSIGNMENT TERMS OF REFERENCE

1. PURPOSE

The Uganda Road Fund (URF) intends to award a contract for a provider to undertake research on an interim formula for allocation of the monies of the Fund to competing agencies. The provider may be a consulting firm or a research institute or academic institution. These Terms of Reference set out the requirements of this commission.

2. BACKGROUND

2.1 General Background

The Uganda Road Fund (the Fund or URF) was established by the Uganda Road Fund Act, 2008 (the Act) to facilitate the delivery of road maintenance services and to create an environment that is conducive to the efficient and effective maintenance of public roads. The Fund is to provide its services in the most economic, efficient and effective manner and to manage its affairs in a business-like and cost-effective manner and in accordance with modern management practices and techniques. The roles of the Fund are defined by the Act to include the following:

funding routine and periodic maintenance of public roads

ensuring that public roads are maintained at all times; and

advising Ministers on the preparation and efficient and effective implementation of the Annual Road Maintenance Programme.

The monies of the Fund principally comprise road user charges including:

fuel levies;

international transit fees, collected from foreign vehicles entering the country;

road licence fees;

axle load fines;

bridges tolls and road tolls; and

weight distance charges.

The Act provides for designation of agencies that are responsible and accountable for the development, rehabilitation and maintenance of public roads in a manner consistent with the economy and set standards. Allocations from the Fund to the designated agencies are to be based, among other factors, on the conditions of the public roads, maintenance requirements, the length of the road network and the relevant volume of traffic or derived from an approved maintenance management tool. The maintenance of national roads, totalling some 20,000 kilometres in length, is the responsibility of the Uganda National Road Authority (UNRA). District, Urban and


95

Community Access Roads (DUCAR), of which district roads total around 23,000 kilometres and municipality roads some 4,500 kilometres, are maintained by around 100 designated agencies (the number of districts is growing). Roads in urban areas and in rural communities (approximately 30,000 kilometres) are maintained by Town Councils and Sub-counties respectively acting as sub-agencies to districts. The thrust of works on these minor roads is on the removal of bottlenecks to enable accessibility. It is the responsibility of each designated agency to formulate an annual maintenance programme for its jurisdiction and to submit that programme to the Fund in a timely manner. The Fund then consolidates all agencies’ programmes into an Annual Road Maintenance Plan and prepares an Annual Expenditure Programme for approval of the Ministers responsible for Finance and Roads and subsequently Parliament. The Fund is also required to prepare three-year and five-year Road Maintenance Plans. UNRA has been active in measuring the extent and condition of the national road network, although information is not yet collected on some 10,000 kilometres of district roads that were reclassified as national roads on 1st July 2009. The extent and condition of DUCAR roads is not known with any reliability.

2.2 Fund allocation

In the medium and longer terms, it is the intention of the Fund that monies will be allocated to road classes and between competing agencies on the basis of need. That cannot be achieved until designated agencies have a full and reliable knowledge of the extent and condition of their networks and until there is a uniform capacity to interpret such data, to plan and prioritise road maintenance works and to formulate coherent and properly costed programmes of work. It may be several years before that position is reached.

In the meantime, the Board must annually consider and approve the method by which funds will be allocated between classes of road and between designated agencies for national, district, urban and community access roads. The Board has expressed the wish that the method of allocation of funds should be clear, transparent and readily understood. It should be clearly related to the strategic vision, mission and objectives of the Fund, which need to be seen in the wider context of the National vision, insofar as it embraces the road sub-sector and in terms of the Act, which sets out the purpose, powers and operations of the Road Fund.

Mechanisms need to be put in place for allocating funds between road classes and between the different administrations. They need to be simple, transparent and encourage consistency of standards between the different administrations. Other than


96

the needs-based approach there are two basic methods that have been commonly used for this purpose:

Simple allocation formula

Indirect assessment of needs

2.2.1 Simple formulae

Simple formulae can be used to assign funds on the basis of predefined percentages to different parts of the network. For example: Japan allocates a fixed percentage of road related revenues to different local government areas; in the past, Ghana spent 52 per cent of revenues on trunk roads, 28 per cent on rural roads and 20 per cent on urban roads. Until recently, Zambia spent 20 per cent on urban roads, 40 per cent on feeder roads and 40 per cent on trunk and main roads. Such allocation mechanisms are simple and transparent but, over time, are related only weakly to current need or use of the network.

2.2.2 Indirect assessment

Indirect assessment of need is used where there are no reliable data for measuring needs directly, or where the cost of doing so would be disproportionate to the size of the budget being allocated. The method is used mostly for the allocation of budget to low cost/low volume roads. Criteria that have been used in the indirect assessment include:

Land area covered by the administration

Road density

Population

Agricultural production or potential

This method often divides the funds initially between the different types of road administration, main, urban and rural, and then subdivides each allocation between the individual administrations within each type. Most countries use fairly crude formulae for doing this to ensure that each administration gets an equitable share of the revenues available. The factors are weighted according to their perceived importance. This approach provides a pragmatic basis for allocating funds in appropriate cases that is cost effective, and acknowledges the socio-political aspects of the decision making. Ideally, the road fund should be designed to fund fully the maintenance of all roads. However, in many countries, there are low levels of motor vehicle usage and there are


97

real constraints on governments to raise fuel prices to a level that can include an adequate utility charge for the road fund. In such cases, the balance of the cost of maintaining the social road network must normally be financed by a government subsidy that recognises the social nature of rural access roads. The subsidy can come from central or local government revenues. The road fund model does not really address the maintenance of the social network of tracks, trails and paths, where usage is predominantly non-motorised.

2.2.3 Present approach

As an interim measure, for Financial Year 2010/2011 only, the Fund has allocated monies to designated agencies on a similar basis to the previous year. This process firstly allocates proportions of the total budget to road classes on an arbitrary basis that seeks to fund maintenance of national roads as fully as possible and allocates remaining funds to district, urban and community access Roads on the basis of formulae similar to those adopted by the Ministry of Finance hitherto.

3. STUDY OBJECTIVES

3.1 Overall Objective

The overall objective of the Study is to give effect to Section 22(2) of the Uganda Road Fund Act, 2008, which states that: “Allocations from the Fund to the designated agencies shall be based among other factors, on the conditions of the public roads, maintenance requirements, the length of the road network and the relevant volume of traffic or derived from an approved maintenance management tool” This will be accomplished in the short term by development of fund allocation criteria that ensure equity, transparency and accountability and fairness of funding across road classes and between designated agencies in the sharing of available resources.

3.2 Specific Objectives

The specific objectives are to: (k) Examine the current road administration of Uganda and analyse the historic trend

of allocation of road maintenance resources and the relative impacts thereof;

(l) Carry out detailed analysis of designated agencies to develop a proposed hierarchy and pattern of distribution among road classes and designated agencies, identifying common influencing factors on the cost of maintenance of roads;


98

(m) Review the current practices of road funds in the region and elsewhere in allocation of funds to road maintenance, with emphasis on the transferability of such practices;

(n) Develop, recommend and demonstrate robustness of an equitable, transparent, fair and justifiable approach to allocation of road maintenance resources, vertically between road classes and horizontally between designated agencies responsible for each class of road;

(o) Based on the state of reform of the road sub-sector and the likely pace of

development of institutional, technical and financial capacity of designated agencies; propose a road map for transition from an indirect formula – based approach to a needs based approach.

3.3 Study Outcomes

This study will be fundamental to the business planning of the Fund and to the political acceptability of the annual road maintenance plans and expenditure programmes going forward until such time that robust needs based allocation process can be implemented. The desired outcomes of the Study are as follows: (a) A clear understanding of international best practice in countries and road

administrations that use, or have recently used a formula based or indirect approach to allocation of monies to road maintenance;

(b) A well reasoned and defensible allocation approach for the Uganda Road Fund that can be demonstrated to produce fair and equitable allocation of funds;

(c) A transition road map from formula-based to needs-based allocation of funds to road maintenance.

4. SCOPE OF SERVICES

The Study is limited to the allocation of monies of the Fund to routine and periodic maintenance of the public road network, estimated at a total length of some 80,000 kilometres and categorised as national, district, urban and community access roads. The Services will be delivered in three phases, which may overlap:

Inception Phase

Literature review

Formula development

Reporting


99

4.1 Inception Phase The Inception Phase will quickly set the direction of the Study. Tasks in this phase will include, but not be limited to: (a) Review of all available reports and data, including: institutional arrangements for

road administration at national and local levels; governance issues; sector reform initiatives; legal framework underpinning the sector etc;

(b) Meetings with sector stakeholders;

(c) Formulation of a work plan within the allowable budget and duration of the study that will give the greatest level of confidence in the results; and

(d) Presentation of the results of this phase of the Study in the form of a written report supported by a presentation to the Fund Management Board, members of the Secretariat and other selected stakeholders

4.2 Literature review

In this phase, the provider will collect and review all available technical literature on the subject of budget allocation in the roads sub-sector. Allocation approaches in other sectors where direct measurement of need is not practicable should also be reviewed for commonality of approach. The tasks in this phase will include, but not be limited to: (a) Acquisition of existing literature on fund allocation from all available sources,

especially relating to other road funds but also including state budgetary allocation processes in the road sub-sector;

(b) Review of literature to identify and critically review the approaches used by other road funds regionally and internationally;

(c) Review of similar indirect allocation approaches used in other sectors, both in Uganda and elsewhere;

(d) Formulation of a questionnaire to be issued to relevant agencies and authorities to establish the evolution of their allocation approaches and to identify particular successes and failures;

(e) Visits to other regional road funds to follow up on specific aspects;


100

(f) Analysis of data, with particular emphasis on extraction of a set of most commonly used factors and proxies for estimation of road maintenance needs including but not limited to:

Geographic and environmental factors

Condition related factors

Demographic factors

(g) Preparation of a literature review report.

4.3 Formula Development In this phase of the Study, we will develop an approach to fund allocation that is robust and defensible and which recognises the following principles: (a) Funds should be equitably shared between road classes and designated agencies;

(b) Since the user pays, the allocation should fairly represent road usage;

(c) On that basis, the maintenance of the national road network should be fully

funded subject to demonstration of value for money and allocated on the basis of a road management system

(d) Funds should be allocated to give the optimum effect in terms of road condition;

(e) Since traffic levels on district and community access roads are generally low, the social benefit of these roads should be recognised by a significant contribution to their upkeep from the consolidated fund or other charges and that should be a condition of Road Fund support, lest the money allocated be wasted on inadequate maintenance;

(f) The formula should, as much as possible, reflect the policies, plans and strategies of the Government of Uganda relating to the road sub-sector and also to general development (as reflected in the current National Development Plan and in Millennium Development Goals)

(g) In this task, the provider will identify assessment criteria and develop formulae combining the effects of the various criteria. The validity of the formulae will be verified by application to a sample of agencies representing the range of climate, terrain, population, road density and traffic volume that is encountered in Uganda. In the sample agencies, data on the length and condition of the road network will be collected by agency staff trained and supervised by the provider.


101

The tasks in this phase will include, but not be limited to:

Review of present and forecast revenue potential based on likely levels of road user charge and comparison with provisions in national medium term expenditure framework;

Review of estimated needs for road maintenance based on available information and consultation with Uganda National Road Authority and Ministry of Works & Transport;

Identification of potential assessment criteria for indirect allocation of funds;

Development of optional approaches based on international best practice and employing the assessment criteria;

Identification of not more than ten sample agencies representing a range of environments and traffic conditions;

Collection of data on road length and condition to establish true needs;

Collection of values of indirect assessment criteria for sample agencies;

Testing of approaches with sample data from agencies where the needs can be directly measured ;

Review of approaches in workshops with stakeholders;

Presentation of approaches to Secretariat and Board;

Recommendation of preferred approach.

4.4 Reporting In this phase of the Study a detailed report will be produced on all activities undertaken during the Study. The report will be supported by presentations that will be given to the Board and to other stakeholders, including the Transport Sector Working Group and the Parliamentary Infrastructure Committee.

5. The Team

It is for the Provider to determine the appropriate resource level and team composition. It is however important that the following positions/skills are represented in the team and these positions will be specifically evaluated.

Position Term Qualifications/Experience Resource

Team leader Full time

A graduate and professionally qualified economist or engineer with a post graduate degree in transport economics or a related discipline. At least 15 years experience of transport sector studies and research in developing countries, preferably in Africa. Significant


102

Position Term Qualifications/Experience Resource

experience should have been gained as a team leader engaged in similar services.

Public administration specialist

Full time

A graduate social scientist with a post graduate qualification in public administration and at least 10 years experience in developing country governance and resource allocation principles and practice. She/he should possess deep knowledge of the national and local government structures and modalities in Uganda, with emphasis on the provisions of the Local Government Act and the policies of decentralisation in the country.

Transport planner

Short term

A graduate civil/highway engineer with a post graduate qualification in transport planning or related studies and at least 10 years experience or road maintenance in developing countries.

Transport Economist

Short term

A graduate in economics or engineering with a post graduate degree in transport economics and at least 10 years experience of the road sub-sector in developing countries.

6. Capacity Building & Technology Transfer

It is an important objective of the Services that the capacity of the Secretariat in the approaches to allocation of funds should be enhanced. For this to occur it is vital that the provider should work closely with the relevant officers of the Secretariat and involve them in every stage of the Study. In particular the provider shall facilitate members of the Secretariat to participate in visits to other road funds at the expense of the Fund.


103

7. Commencement and Duration

It is anticipated that the Services will commence during the fourth quarter of 2010. The duration of the Services shall be six calendar months at the most. It is essential that the new allocation method is available for use in allocation of the monies of the Fund for FY 2011/12

8. Deliverables

The Provider shall produce the following deliverables during the execution of the Services:

No Deliverable Description Due (months after Commencement)

1. Inception Report

A brief report describing the understanding of the tasks insofar as it has developed during the Inception Phase, identifying issues that have arisen and how they will be addressed and including a detailed and resourced programme for delivery of the Services.

In draft 2 weeks Final within five days of receipt of comments

2. Literature Review Report

A report that describes the findings of the literature review with findings analysed and summarised in a clear and readily understandable manner.

In draft 2 months Final within five days of receipt of comments

3. Final Report This is the major deliverable of the Services, which will describe all work undertaken during the Study and present clear recommendations as a basis for Board decision.

In draft 5 months Final within five days of receipt of comments

4. Progress Reports

A brief report, submitted electronically by email, including administrative information and setting out the achievements in the reporting period, issues encountered and how they have been addressed.

Every month commencing 1 months after commencement and submitted within 5 days of the end of the reported month.


104

APPENDIX 2 – LIST OF KEY STAKEHOLDER MET

Stakeholder Met Designation

1. Jean Nsengiyumva Manager Programming, Planning & Policy, Namibia Road Fund Administration

2. Harris Ntema

Manager Internal Audit, Namibia Road Fund Administration

3. Fenni M. Angukku Head Corporate Services, Namibia Road Fund Administration

4. Irene Viene Head Information Technology Department, Namibia Road Fund Administration

5. Jeremiah Kitongo Manager Fund Management, Namibia Road Fund Administration

6. Vincent Sasele Divisional Manager of Road Traffic and Transport Inspectorate of Namibia

7. Alberto Z Toruncha Chief Engineer, Namibia Ministry of Works and Transport

8. Rainus A !Gonteb Windhoek Regional Engineer, Namibia

9. Sophia Belet Tekie Namibia Roads Authority Acting ED

10. B. Ssebbugga Kimeze Director of Operations, Uganda National Roads Authority

11. Justine Ongom Odongo Roads Maintenance Manager, Uganda National Roads Authority

12. Dr Mbadhwe John Infrastructure Advisor, Community Agricultural Infrastructure Improvement Programme

13. Stephen Kitonsa Principal Engineer in charge of District Roads, Ministry of Works and Transport

14. Eng. Dr Francis Bazirake Board Chairman, Uganda Road Fund

15. Eng. Dr Michael Odongo Executive Director, Uganda Road Fund

16. Andrew Naimanye Manager, Programming, Uganda Road Fund

17. John Ocitti Manager Fund Management, Uganda Road Fund

18. Joseph Etiang Internal Audit Manager, Uganda Road Fund

19. Ronald Namugera Road Maintenance Engineer, Uganda Road Fund

20. John Digber Acting Executive Director, Ghana Road Fund


105

APPENDIX 3 – VEHICLE FLEET CHARACTERISTICS

Veh

icle

Nam

e

Pas

sen

ger

Car

Sp

ace

Eq

uiv

alen

ce

Nu

mb

er o

f W

hee

ls

Nu

mb

er o

f ax

les

Bas

e N

um

ber

of

Rec

aps

Re-

trea

d C

ost

s as

a

Per

cen

tage

of

Ne

w

Tyre

Co

sts

Kilo

met

ers

Dri

ven

P

er Y

ear

Wo

rkin

g H

ou

rs

Pri

vate

Use

(%

)

Nu

mb

er o

f P

asse

nge

rs

Wo

rk R

elat

ed

Tri

ps

(%)

Op

erat

ing

Wei

ght

(To

ns)

Nu

mb

er o

f Eq

uiv

alen

t St

and

ard

Axl

es

Articulated Trucks 1.8 22.0 6.0 3.6 15.0 60000.0 1700.0 60.0 3.0 40.0 40.3 17.7

Big Buses 1.6 6.0 2.0 2.4 15.0 60000.0 2500.0 80.0 58.0 20.0 12.2 2.4

Four Wheel Drive 1.0 4.0 2.0 1.3 15.0 25000.0 1200.0 70.0 3.0 45.0 1.5 0.0

Heavy Trucks 1.6 10.0 3.0 2.4 15.0 55000.0 1500.0 60.0 3.0 40.0 22.6 8.3

Light Trucks 1.3 4.0 2.0 2.4 15.0 40000.0 1200.0 65.0 3.0 35.0 4.0 0.0

Medium Trucks 1.4 6.0 2.0 2.4 15.0 50000.0 1500.0 60.0 2.0 40.0 10.6 3.5

Motorcycle 0.5 2.0 2.0 1.3 15.0 10000.0 600.0 80.0 1.0 20.0 0.2 0.0

Pick Up (LGV) 1.0 4.0 2.0 1.3 15.0 20000.0 600.0 40.0 3.0 60.0 1.5 0.0

Taxi Bus (Mini bus) 1.4 4.0 2.0 1.3 15.0 40000.0 2500.0 75.0 13.0 25.0 2.6 0.0

Toyota Corona (Cars) 1.0 4.0 2.0 1.3 15.0 20000.0 600.0 40.0 3.0 60.0 1.5 0.0


106

V

ehic

le N

ame

Economic Unit Costs of Vehicle Resources

Ne

w V

ehic

le C

ost

s

Tyre

Co

sts

Fuel

Co

sts

(Per

Lit

re)

Oil

Co

sts

(pe

r lit

re)

Mai

nte

nan

ce L

abo

ur

(pe

r h

ou

r)

Cre

w W

ages

An

nu

al O

verh

ead

Co

sts

Inte

rest

(%

)

Co

st o

f P

asse

nge

r

Wo

rkin

g Ti

me

Co

sts

of

No

n-

Wo

rkin

g Ti

me

Car

go H

old

ing

Co

sts

per

ho

ur

Articulated Trucks

109700.0 269.6 0.5 1.8 2.2 1.3 3000.0 12.0 0.0 0.0 0.1

Big Buses 77412.0 182.5 0.5 1.8 2.2 1.8 3000.0 12.0 0.3 0.0 0.0

Four Wheel Drive

29500.0 107.8 0.5 1.8 1.7 0.5 200.0 12.0 0.9 0.0 0.0

Heavy Trucks

71250.0 269.6 0.5 1.8 2.2 1.2 2500.0 12.0 0.0 0.0 0.0

Light Trucks 27692.0 80.9 0.5 1.8 2.0 0.7 2000.0 12.0 0.0 0.0 0.0

Medium Trucks

34000.0 161.8 0.5 1.8 2.2 1.0 2000.0 12.0 0.0 0.0 0.0

Motorcycle 1700.0 10.4 0.5 1.8 1.7 0.2 20.0 12.0 0.3 0.0 0.0

Pick Up (LGV)

18300.0 39.4 0.5 1.8 1.7 0.7 200.0 12.0 0.3 0.0 0.0

Taxi Bus (Mini bus)

26300.0 37.3 0.5 1.8 1.7 1.2 200.0 12.0 0.3 0.0 0.0

Toyota Corona (Cars)

17667.0 31.9 0.5 1.8 1.7 1.1 150.0 12.0 0.9 0.0 0.0


107

APPENDIX 4 – MoWHC (2006) STRATEGIC ANALYSIS USING HDM-4

Executive Summary

Introduction

Senior policy makers and road managers within the Ministry of Works (MoWHC), Uganda require relevant and up to date information in a variety of forms in order to analyse and to justify longer-term strategies as well as day-to-day decisions. This study was formulated to improve decision-making on expenditures in the road sector by enabling effective and sustainable utilization of the latest Highway Development and Management (HDM) knowledge. It is expected that this will bring about a significant change in the quality (i.e. effectiveness and efficiency) of planning at strategic level and determining the effects of different levels of funding to be provided to MoWHC for the capital expenditure of road maintenance on roads in Uganda.

HDM-4 is a computer-based system that simulates the change in road condition and maintenance requirements by representing the total life cycle conditions and costs for an analysis period under a user-specific scenario of circumstances. Interacting sets of costs, related to those incurred by the highway authority and those incurred by the road user, are added together over time in discounted present values. Costs are determined by first predicting physical quantities of resource consumption and then multiplying these resource quantities by their unit costs. Economic benefits are then determined by comparing the total cost streams for various maintenance and construction alternatives with a base case (Do Minimum) usually representing minimal routine maintenance.

The Ministry of Works, Housing and Communication (MoWHC), Uganda intends to use HDM-4 to carry out a strategic analysis of the road system under their jurisdiction to assist in determining:

The long-term, or strategic, planning estimates of expenditure for road preservation under various budgetary and economic scenarios

The effects of different levels of funding to be provided to MoWHC

Road Network Matrix

The MoWHC is responsible for the National (Truck) Road Network which totals to about 10,000 km and carries about 80% of Uganda’s traffic. The National Road Network is classified into primary, secondary and tertiary roads. Of the 10,000 km of National roads, about 2,200 km are bituminous and the rest are gravel. For a Strategy Analysis purposes, the road network was represented by a multi-dimensional matrix, with each cell representing a significant part of the overall network being analysed. Key variables used to formulate the matrix were: Region (Central, Eastern, Northern, Southern), Pavement Type (Asphalt Mix on Stabilised Base, Surface Treatment on Stabilised Base, Gravel), Road Class (Principal and Classified), Surface Condition (Good, Fair, Poor and Bad), Traffic Volume (Low, Medium, High).

Analysis Scenarios

The effect of the following capital budget scenarios were examined examined:

Scenario 1: an unconstrained budget;

Scenario 2: a 15% cut to the unconstrained budget;

Scenario 3: a 30% cut to the unconstrained budget; and

Scenario 4: a 50% cut to the unconstrained budget.


108

Analysis Results

The estimated unconstrained budget for strategy 1 is presented in Table1. The effects of budget constraints on network performance under scenarios 1 to 4 is depicted in Figure 1 and Table 2.

Table 1: Unconstrained budget for strategy 1

Unconstrained Capital Budget (US $ Million)

Unconstrained Recurrent Budget (US $ Million)

Unconstrained Total Budget (US $ Million)

359.744 1182.2 1541.944

2

3

4

5

6

7

8

9

10

11

12

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

Years

Av

era

ge

IRI

…

.

Scenario 1 (US $ 359.7m) Scenario 3 (US$ 214.0m)

Scenario 4 (US$ 107.0m) Scenario 2 (US$ 305.8.0m)

Figure 1: Predicted Road Network Performance Trends

Table 2: Impact on performance of all budget scenarios

Surface Type Road Class Weighted Average Roughness (m/km)

Scenario 1 Scenario 2 Scenario 3 Scenario 4

Paved

Class I 4.0 4.5 5.0 8.6

Class II 3.8 3.8 8.8 10.3

Class II 4.4 5.4 9.0 9.0

Unsealed

Class A 12.0 13.3 13.9 15.3

Class B 11.9 12.6 13.8 15.5

Class C 11.7 12.3 13.2 15.1

All Network 7.9 8.6 10.6 12.3


109

APPENDIX 5 – IT TRANSPORT (2011)

Executive Summary to Interim Report

Introduction

Senior policy makers and road managers within Uganda National Roads Authority (UNRA) require relevant and up to date information in a variety of forms in order to analyse and to justify longer-term strategies, programmes, projects as well as day-to-day decisions. This study has been formulated to standardise and improve decision-making on expenditures in the road sector by enabling effective and sustainable utilization of the latest highway development and management knowledge. The project aims to produce a comprehensive and standardised set of tools and procedures to perform economic, financial, social and engineering assessment of road transport investment strategies, programmes and projects. The assessment of transport investment impacts can be considered in two parts as Appraisal and Evaluation.

This Study is required to achieve the following six specific objectives:

1. An assessment of requirements in order to align the UNRA investment appraisal and evaluation methodology and procedural guide with the international best practices;

2. Based on the assessment of (1), the development of a standardised framework and procedures to support the project, programme, and strategic level decisions on economic, financial, social and multi criteria issues;

3. An assessment of the input requirements in order to customise and calibrate the investment appraisal tools, and the development of standardised framework and procedures for the collection and updating of the same; and

4. The collection of the data defined in (3) and updating of the appraisal instruments;

5. The preparation of UNRA Manual on road investment appraisal; and

6. Training and monitoring of UNRA staff in order to ensure the sustainable use of the tools and procedures developed under the consultancy services.

The overall output of the study will be: UNRA’s investment appraisal tools and procedures, which are streamlined resulting in the strengthening of the investment appraisal and evaluation function. The output will also be relevant for the overall road sector in Uganda. This will be achieved through the development of standardised framework and procedural guide (manual) for the evaluation of project, programme, strategic level and transport policy decisions and through providing training to UNRA staff on the appropriate tools and procedures.

This interim report is intended to provide an outline of the study methodology, describe the progress of the various project tasks and activities and identify activities that are planned to be undertaken prior to the completion of the study. To that end, Chapter 2 of the report provides a summary of the study methodology that was detailed in the project inception report. Chapter 3 reviews the current investment appraisal tools and procedures used in Uganda and in other countries and recommend amendments. Chapter 4 presents an analysis of the responses from stakeholders’ consultations, the key findings, comments from the client and the responses to some of these comments. The proposed


110

investment appraisal framework and systems are discussed in Chapter 5, while Chapter 6 deals with collection of data necessary for configuring and calibrating the appraisal tool. Issues pertinent to the calibration of the appraisal tools are discussed in Chapter 7. An outline of the new investment appraisal and evaluation manual is given in Chapter 8. Finally, Chapter 9 sets out the outstanding tasks.

Study Methodology

The overall approach to the study has been divided into the following four distinct phases:

Phase 1 – Review of existing systems and documents used in Uganda and other countries; and definition of future systems/framework for the appraisal/evaluation of transport investment strategies, programmes, and projects.

Phase 2 – Development of standardised data collection framework/procedures and strategy, data collection and data analysis.

Phase 3 – The production of UNRA investment appraisal/evaluation manual, and development of policy on the implementation of investment appraisal system and monitoring and feedback procedures.

Phase 4 – Training of primary users, acceptance testing and hand over.

These phases including the relationships between activities are explained in Chapter 2 of this report.

Review of Existing Appraisal Tools and Procedures

The Study Team has undertaken a preliminary review of the road investment appraisal methods used in

the following countries and institutions: Uganda, Tanzania, Ghana, United Kingdom, Japan, Germany,

European Union and the World Bank. Approaches used in a few other countries will also be reviewed.

The findings, issues and constraints identified have been used to define the updated UNRA’s framework

for investment appraisal and evaluation.

The current road investment appraisal guideline used by UNRA is known as the “Procedural Guide to

Economic Road Feasibility Studies”. The guide was prepared for the Road Agency Formation Unit (RAFU)

and was last updated in March 2006. The guide covers in appropriate detail most essential aspects and

theoretical principles that should be included or addressed in the economic analysis of road project

alternatives.

The main contents headings are: Introduction, Approach, Appraisal Methodology – Modeling, Traffic

Studies, Other Project Benefits, Vehicle Operating Costs, Physical Road Characteristics, Road

Maintenance and Construction Costs, Economic Appraisal, Sensitivity and Risk Analysis, and Checklist.

The basic focus of the guide is the ‘Feasibility Study’ phase of the project cycle, which is a combination

of technical (i.e. engineering), economic appraisal aspects. The guide is also mainly concerned with

‘project’ level analysis. It does not cover the programming and strategic planning level of analyses.

Furthermore, the guide is built around Cost-benefit analysis (CBA) as the preferred analysis

methodology. CBA is a method for identifying costs and benefits, quantifying, costing them and then

comparing them over time. The method can be manual but the Guide stresses the use of computer


111

modeling using highway development and management tool (HDM-4). HDM-4 is the most widely used

road investment analysis tool in developing country contexts and is preferred by the majority of multi-

lateral and bi-lateral donors. The model requires considerable input data on road user costs and

suggested values for all these are presented in the report. The primary data often requires analysis

before input into the model and some of these requirements are also presented in the current UNRA’s

Guide. Guidance is also given on traffic data collection and forecasting.

It is however important to note that a comprehensive appraisal framework is much wider in scope,

often with a need for explicit consideration of environmental and social impacts. Road schemes often

require land acquisition and sometimes resettlement, so both environmental and social impact

mitigation costs need to be considered. Principles and techniques for arriving at reasonable estimates

for these costs need to be included in future guidelines.

The appraisal and evaluation of road investments may sometimes need to go beyond CBA. This may be

necessary, for example, at the programming and strategy analysis levels where precision in forecasting

costs and benefits is not so much important as making broader non-quantifiable assessments of

alternative strategies. In such cases approaches incorporating Multi-Criteria Analysis, for example, might

be more suitable. These issues are not addressed within the current UNRA’s Appraisal Guide.

The current UNRA’s Guide provides detailed road user cost data for HDM-4. More information is

however needed on data collection methods (there is coverage for traffic data) and data quality

assurance procedures.

Furthermore, CBA methodology makes use of many economic concepts that require specific analytical

steps. They in turn require specific data - if they are to contribute to the overall analysis. These include

data required for estimating growth rates, converting financial to economic values, estimating accident

costs, agricultural producer surplus, and traffic generation. To implement these and many other

concepts several important questions must be addressed:

Does the data exist?

Can it be collected?

Is it possible to do so within the cost and time frame of a study?

A useful feature of a Guide would be to provide worked examples to demonstrate the links between any

set of primary data and the desired values. This useful feature is lacking in the existing UNRA’s Guide.

The preferred method for carrying out evaluations of policy is to procure, as is currently the case, the

services of outside consultants. Since such is the case the key to any ‘Guide’ is that it should translate

successfully into a Feasibility Study or a Programme/Strategy analysis document whose methods and

assumptions can be easily compared to each other, thus eliminating uncertainties in the commitment of

scarce funds.

Following a review of previous feasibility study reports it is apparent that the objective, through the

current appraisal Guide, for consistency and standardization has not been entirely successful. For


112

example, inconsistencies have been identified in vehicle types recommended by the current UNRA

guideline with those used in the economic analysis at the feasibility stage on five projects; and

inconsistencies in the vehicle economic costs between feasibility studies.

Findings from Stakeholders’ Consultations

As a part of the stakeholders’ consultation exercise, the study team prepared a set of questionnaires

and distributed this to stakeholders in the Uganda road sector. The questionnaire was targeted at the

wider road sector stakeholders in Uganda and was distributed to Uganda National Roads Authority,

Uganda Government Ministries, International Organisations, Road Agencies, consultants and to those

listed in Appendix 1 of this report.

The questionnaire was aimed at eliciting current appraisal practices, specific needs as well as more

general opinions regarding the road investment appraisal process in Uganda. The outputs will be used to

guide the development of a revised set of investment appraisal and evaluation tools and procedures for

UNRA that should reflect international best practice subject to their relevance to the development

needs of Uganda. The questionnaire comprised the following four major parts:

Assessment of framework and systems;

Impact assessment and valuation;

Analysis method and processes; and

Additional comments.

Ten responses have been received so far, and summaries of the analysis of the responses are presented

in this report. The outputs of the analysis will be updated when additional responses from the

stakeholders are received.

Some of the key findings are as follows:

i) Respondents related the national objectives to each road class (National, District, Urban,

Community Access) as shown in Table E.1.

Table E.0.1: Relating National Objectives to Road Class

Road Class Main National Objectives

1st

2nd

3rd

National Economic growth National cohesion Exports promotion

District Growth of the

agricultural sector Economic growth Poverty reduction

Urban Economic growth and

Environmental Re-distribution of poverty / Growth of the


113

sustainability Social Equity agricultural sector

Community Access Growth of the

agricultural sector

Poverty reduction and re-

distribution of poverty/

social equity

Economic growth /

environmental

sustainability

Table E.1 shows that for National roads investment appraisal and evaluation should emphasize

economic growth, national cohesion and exports promotion. For District roads the assessment emphasis

should be on growth of the agricultural sector, economic growth and poverty reduction. Appraisal and

evaluation of Urban roads should focus on economic growth, environmental sustainability, re-

distribution of poverty and social equity and growth of the agricultural sector. While for Community and

Access roads, growth of the agricultural sector, poverty reduction, social equity and economic growth as

well environmental sustainability are the main issues to be considered in investment appraisal and

evaluation.

ii) Organisations that did not have their own formalised transport appraisal guidelines mostly used

UNRA and World Bank guidelines. African Development Bank and European Union guidelines have also

been used.

iii) Stakeholders were asked to indicate whether investment appraisals categorised under project,

programme, strategy and master-plan decision making levels were performed in their organisations in-

house or by outside consultants. The findings indicate that appraisals categorised under the projects and

programme levels were the most common. Appraisals at the project level are largely performed by

outside consultants while at the programme level equal proportions of appraisals are undertaken in-

house and by outside consultants. Strategic level appraisals are predominantly undertaken in-house.

Financial and Economic assessment criteria are the most common while Environmental and Social

criteria at the least common.

iv) Most organisations who responded to the questionnaire are to a larger extent directly involved

in the Problem Identification, Pre-feasibility, Detailed Design, Implementation and Monitoring and

Evaluation Phases. A bigger proportion of the respondents play a supporting role as opposed to direct

role during the procurement phase

v) Based on the premise that a particular transport problem may have many solutions some of

which may be optimal, not from an economic point of view, but from an engineering point of view, the

stakeholders were asked to suggest important aspects of alternative design they consider in investment

appraisal. The suggested aspects are grouped in Table E.2 under Economic, Engineering, Sustainability,

Environmental/Social, and Safety sub-headings.


114

Table E.2: Summary of Important aspects of Alternative Designs in Investment Appraisal

Economic Engineering Sustainability Environmental and

Social Issues Safety

Construction costs

Cost benefit

Cost efficiency

Economic returns

Internal Rate of Return

Low cost roads

Net Present Value

Operation and maintenance costs

Resettlement Costs

Feasibility of constructing the design solution

Alternative design routes

Pavement and Material Types

Pavement surface material

Road pavement strength

Staged construction, starting with options that would maximise the use of natural materials (design based on layer coefficients and final structural number)

Structures Type

Traffic flow patterns and characteristics

Weather /climate characteristics

Recycling in the case of rehabilitation of existing paved roads

Environmental degradations

Environmental and social impact

Environmental Impact

Impact on the environment

Public acceptability

Land requirement

Resettlement Action Plan

Social Impact

Trading centres and the need to preserve major towns

Safety of motorist and pedestrians

Safety and Accessibility

vi) Stakeholders were asked what cost items should be included in investment appraisal and

evaluation in addition to the following pre-defined cost items: Construction costs, Road maintenance

costs, Vehicle operating costs, Passenger travel time costs, Cargo delay costs, Accident costs, Land

acquisition costs. The following additional impacts were proposed: access to transport systems, accident

reduction costs, capital gain in land arising from road investments, economic efficiency for road users,


115

environmental impact/mitigation, exogenous benefits (e.g. improvement in agricultural farm gate

prices), land-use policy, providers and consumer user reliability, risks, security, social impact, survey and

design costs, and transport interchange

vii) More than half of the responses stated that passenger travel time should be determined

separately for the following situations: Time period; Travel time component; Network characteristics;

and Mode of transport. Wage rate studies, stated preferences/ contingent valuation, function of GDP

per capita and international comparisons are generally considered appropriate for valuation of travel

time during working hours. Revealed preferences and to a lesser extent international comparisons were

selected for valuation of travel time during non-working hours.

viii) Of the stakeholders who responded, 80% agree and 20% disagree that fatal, serious injury, slight

injury and damage only accident types should be considered in transport investment appraisal and

evaluation in Uganda. The respondents who disagreed suggested the following alternative list of

accident types that should be considered in transport investment appraisal in Uganda:

Fatal

Serious

Slight injury

Some argued that data on damage only is likely to be very sparse and/or unreliable.

Life insurance approach was selected by all respondents as the method that should be used for costing

fatal injury, however, gross output approach and implicit public sector valuation approach were also

selected by at least halve of the respondents. Net output, willingness to pay for the reduced risk and

Court award were the preferred methods for costing injury types of accidents. Halve of the responses

indicated that the implicit public sector valuation approach can also be used.

ix) The following methods were suggested to be used for valuing land taken up for road

infrastructure in Uganda in urban areas:

Competitive market rate (replacement costs)

Use the open market value and not CGV

UNRA has land acquisition management system established for land valuation

Market Value

x) The following methods were suggested to be used for valuing land taken up for road

infrastructure in Uganda in rural areas:

Use of open market value and not District Land Board

UNRA has Land Acquisition Management System established for Land Valuation

Market value for undeveloped land and cost method for land with developments


116

xi) The following valuation techniques were proposed to be used for valuing costs/impacts where

there is no readily available market data such as environmental, social or health impacts:

Estimation of impact and valuing it in monetary term

Replacement cost technique

Substitute cost

'Benefits' or 'value' transfer

Multi-Criteia Analysis (MCA)

NATA

Analytical hierarchy process (MCA method)

xii) Stakeholders were asked to identify the appropriate analysis methods that should be used for

each study level of project, programme and strategy. The findings are summarised in Table E.3.

Table E.2: Relating Study Levels to Analysis Method

Study Level Appropriate Analysis Method

Primary Secondary

Project Cost-Benefit Analysis Cost Effectiveness Analysis

Programme Multi-Criteria Analysis Cost-Benefit Analysis

Strategy Multi-Criteria Analysis Cost-Benefit Analysis

At the project level, CBA is the preferred analysis method for high traffic volume roads and CEA should

be used for low traffic volume roads. Typical road projects include rehabilitation, widening, upgrading

and new link construction. MCA is more appropriate for use at the programme and strategy levels.

xiii) CBA was the preferred method for paved roads rehabilitation, upgrading and new link. While

for unpaved roads, CBA is by far the most suited method for interventions that require upgrading to

paved standard, improvement to higher standard gravel, and upgrading to gravel. Both DA and CBA are

equally suited to new unpaved links.

xiv) Considering that Cost Benefit Analysis can use either Consumer Surplus or Producer Surplus

method for valuing benefits arising from transport investment depending in part on the traffic level

using the road, stakeholders were asked to indicate the level of traffic that would apply to the consumer

surplus method and to the producer surplus method. The responses suggest that the Consumer Surplus

method is generally preferred over the producer surplus method at all traffic levels. Considering the

trends of the responses with traffic, the Consumer Surplus approach is more applicable at higher traffic

levels (say greater than 200 AADT), while the Producer Surplus approach seems to be more appropriate

for use at lower traffic levels say less that 200 AADT.


117

xv) The stakeholders were asked to indicate the computer-based models (or systems) that their

organisation use for road transport investment appraisal and evaluation. Most organisations that

responded to the survey use HDM-4, Red and Excel based models.

Additional comments from Stakeholders are also given in Chapter 4 of this report.

Proposed Framework for Investment Appraisal

Definitions

The proposed framework for the new investment assessment tools and procedures will perform

economic, financial, social and engineering assessment of road transport investment strategies,

programmes and projects. The assessment of transport investment impacts can be considered in two

parts as Appraisal and Evaluation:

Appraisal aims at ex ante assessment of whether action is worthwhile. It is always carried out prior to implementation using forecasted data. The outputs are normally used for project procurement, policy, strategy and programme design. The decision criteria will include economic and engineering indicators.

Evaluation aims at ex post assessment of whether action was worthwhile, and can be carried out during implementation (formative) and after implementation (summative). The output can be used to provide feedback for future procurement, project management, wider policy and strategy debate, and future programme management. The decision criteria consider whether correct criteria were used prior to implementation.

Evaluation is similar in technique to appraisal and often the terminologies “economic appraisal” and

“economic evaluation” are used interchangeably. To avoid confusion, in this report a clear distinction in

terminology is made between economic appraisal, which refers to economic assessment prior to

implementation, and economic evaluation which refers to economic assessment as used in monitoring

after implementation. Evaluation uses historic (actual or estimated) rather than forecast data and takes

place after implementation.

Principles

The new appraisal and evaluation framework is based on the principles determined from consultations

with stakeholders in the Uganda road sector, literature review of existing systems and international best

practice. These principles are summarised as follows:

(i) Providing value for money (VFM) – defined as a criterion that judges the quality of provision, processes or outcomes against the monetary cost of making the provision, undertaking the process or achieving the outcomes.

(ii) Enabling the development of Uganda’s road network in a more consistent, objective and transparent manner


118

(iii) Relating national objectives closely to the road classes in Uganda

(iv) Relating the assessment criteria (economic, engineering, financial, social, environmental and sustainability) to different study levels: project, programme and strategy

(v) Considering the life cycle phases of proposed investments: problem identification, feasibility, detailed design, procurement, implementation, operation and monitoring and evaluation

(vi) Considering the impacts of different investment alternatives and their assessment and valuation methods

(vii) Considering the analysis methods and processes (CBA, CEA, MCA, QMA and DA) of quantifying life cycle costs and benefits arising from different types road investments

(viii) Considering the use of existing computer based models/systems (e.g. HDM-4, RED, dTIMS, BMS) and their data requirements as well as availability within UNRA

Analytical Tools

It is proposed that HDM-4 will be used for Projects with high traffic roads of more than 200 vehicles per

day, where economic impacts are considered to be dominant. It will also be used for Strategy Analysis

and Programme Analysis. The RED model will be used for Projects with traffic between 50 and 200

vehicles per day where benefits are largely of socio- economic nature. This is similar to guidelines given

in the Tanzanian investment appraisal manual.

The proposal is that low-traffic roads (carrying less than 50 vehicles per day) will be prioritised using the

Cost Effectiveness Analysis method. The Cost Effectiveness approach has been proposed for use on low-

volume roads because the influence of the environment on road deterioration is considered to be higher

on such roads than the influence of traffic. A suitable tool has to be identified and procured for this

purpose. For example, a form of CEA such as in the basic access approach which used the amount of

money per population served as an indicator of the relative value of competing investments.

Other analytical tools proposed for the use within the new framework will be based on locally adapted

versions of multi-criteria analysis, based on some agreed relevant combination of observables/criteria

such as traffic and economic activity and chosen social weights. Also the producer surplus method will

be considered for CBA of low volume roads.

Data Requirements

For purposes of investment appraisal and evaluation, the data requirements of the selected systems

have been assessed. Most of the required data are available within UNRA. However, the quality of the

data needs to be assessed and structured within the IQL concept in ways that suit the needs of the

different levels of decision making in road management.

Data transformation techniques will be applied to assemble the data required for different levels of

application (strategy, programming and project) using the new investment appraisal tools. The data


119

required for application of HDM-4, for example, can be grouped under road network inventory, road

condition, vehicle fleet characteristics, traffic characteristics, road work standards (work types,

intervention levels, unit costs and effects).

Appraisal Process and Presentation of Results

Appraisals are often iterated a number of times before their proposals are implemented in full. Therefore the stages set out below, extracted from the Green Book (HM Treasury, 2003) may be repeated, and they may not always be followed sequentially. In particular, as options are developed, it will usually be important to review more than once the impact of risks, uncertainties and inherent biases. This helps to avoid spurious accuracy, and to provide a reasonable understanding of whether, in the light of changing circumstances, the proposal is likely to remain good value for money.

As the stages of an assessment progress, data must be refined to become more specific and accurate. The effort applied at each step should be proportionate to the funds involved, outcomes at stake, and the time available. Accordingly, in the early steps of identifying and appraising options, summary data (IQL3 or IQL4) only is normally required. Later on, before significant funds are committed, the confidence required must increase (IQL2).

The ultimate outcome of any appraisal is a decision whether or not to proceed with a proposal or a

particular option. As these decisions will often have far reaching consequences, the presentation of the

conclusions and recommendations to decision makers and key stakeholders can be as important as the

analysis itself. In all cases, transparency is vital. Presentations and reports should be clear, logical, well

founded, and geared towards helping the decision at hand. Summary reports in particular should be

drafted in non-technical language wherever possible, but, if it is necessary to use technical terms, they

should be explained.

The results of risk analysis, sensitivity and scenario analyses should also generally be included in

presentations and summary reports to decision makers, rather than just single point estimates of

expected values. Decision makers need to understand that there are ranges of potential outcomes, and

hence to judge the capacity of proposals to withstand future uncertainty.

One of the on-going tasks of this project is the development of customised outputs from HDM-4, RED

and other systems to be selected.

Nesting within Transport Planning System

The new framework proposed will standardise the appraisal and evaluation process throughout UNRA

thus providing more consistency and transparency. The appraisal and evaluation tools and procedures

will be compatible with and link to the Road Asset Management System (RAMS) being developed by

UNRA.


120

The RAMS will comprise separate or integrated sub-systems for:

Data collection

Database management

Decision support

Management information

Data Collection Strategies

The following data categories are necessary to facilitate the configuration and calibration of the

proposed appraisal and evaluation framework and tools:

Vehicle fleet data including: representative vehicle types, physical characteristics and unit costs of vehicle resources amongst others;

Road inventory and condition data;

Traffic data including: traffic volume, speed-flow data, traffic flow pattern and traffic growth;

Climate data for the whole country;

Economic data including: standard conversion factors, and discount rates.

This report sets out the proposed standardised procedures for collecting data that are relevant to road

investment appraisal and introduces proposed strategies for data collection. A summary of the data

types that have been collected and those that are still being collected is also provided. Drafts of the

proposed procedures and strategies are provided in Appendices 2 and 3, respectively.

The availability and accuracy of data provides the basis for road investment appraisal and evaluation on

such aspects as:

Assessing current levels of road and bridge condition;

Determining appropriate levels of investment in the road sector;

Prioritising capital improvements and investments in road maintenance;

Estimating the cost of improvements;

Simulating the effects of any improvements on the future condition and performance of the road system;

Determining optimum road condition, and the maintenance strategies and expenditures as well as controlling on-going expenditures.

Model Calibration


121

One of the main analytical tools proposed for use in this updated framework for investment appraisal

and evaluation is the Highway Development and Management tools (HDM-4). The HDM-4 analytical

framework is based on the concept of pavement life cycle analysis. This is applied to predict the

following over the life cycle of a road pavement, which is typically 15 to 40 years:

Road deterioration (RD);

Road works Effects (WE); and,

Road user effects (RUE). Since the HDM-4 model simulates future changes to the road system from current conditions, the

reliability of the results is dependent upon two primary considerations:

1. How well the data provided to the model represent the reality of current conditions and influencing factors, in the terms understood by the model; and,

2. How well the predictions of the model fit the real behaviour and the interactions between various factors for the variety of conditions to which it is applied

Application of the model thus involves two important steps:

(i) Data input: a correct interpretation of the data input requirements, and achieving a quality of input data that is appropriate to the desired reliability of the results.

(ii) Calibration of outputs: adjusting the model parameters to enhance how well the forecast and outputs represent the changes and influences over time and under various interventions. Calibration of the HDM model focuses on the components that determine the physical quantities, costs and benefits predicted for the RD, WE, RUE and Socio-Economic Effects (SEE) analysis

This chapter describes the HDM-4 model adaptation and calibration work to be carried out by the

Consultant.

Proposed Outline of Investment Appraisal Manual

The proposed outline of the revised Investment Appraisal and Evaluation manual include the following

topics:

The appraisal process

Financial versus economic analysis

Consideration of alternatives

Time and discounting

Valuation of costs and benefits

Economic decision criteria

Traffic characteristics

Appraisal tools

HDM-4 calibration and adaptation to Ugandan conditions

Road economic decision model (RED)

Programme analysis

Strategy analysis


122

Other appraisal frameworks

Monitoring and evaluation

Case studies

The outline will be reviewed following consultations with UNRA and other stakeholders.

Future Work

To achieve the project objectives, the study has been divided into four distinct phases of activity as follows: Phase 1 – Mobilisation, review of existing systems/framework and development of future

systems/framework for the appraisal and evaluation of transport investment strategies, programmes, and projects. This phase will be completed by the end of Month 2 of the project.

Phase 2 – Development of standardised data collection framework/ procedures and strategy, data collection and data analysis. This period will run from beginning of Month 3 to end of Month 5 of the project.

Phase 3 – The production of UNRA investment appraisal and evaluation manual, and development of policy on the implementation of investment appraisal system and monitoring and feedback procedures. This phase will cover the period from beginning of Month 6 to end of Month 7 of the project.

Phase 4 – Training of primary and secondary users and acceptance testing. This phase will start from the beginning of Month 8 and end with handover of at the end of Month 8.

Future work in Phase 2 includes:

Collection of additional data required, as identified in Chapter 6 and analysis of the data collected;

Testing and fine-tuning the framework including the development of standardised reporting formats.

All activities listed under Phases 3 and 4 in the schedule in Figure 9.1 are either on-going or planned.


123

APPENDIX 6 – TYPICAL ROAD DETERIORATION RATES

y = -0.0005x2 + 0.1591x + 3.8523R² = 0.994

3.50

4.00

4.50

5.00

5.50

6.00

6.50

7.00

7.50

0 5 10 15 20

IRI (m

/km

)

Years

Annual Rainfall (650mm)

y = 0.0039x2 + 0.2131x + 3.8293R² = 0.9983

3.50

4.50

5.50

6.50

7.50

8.50

9.50

10.50

0 5 10 15 20

IRI (m

/km

)

Years

Average Annual Rainfall (1288mm)


124

y = 0.0038x2 + 0.2131x + 3.8287R² = 0.9983

3.50

4.50

5.50

6.50

7.50

8.50

9.50

10.50

0 5 10 15 20

IRI (m

/km

)

Years


y = 0.0043x2 + 0.2114x + 3.8308R² = 0.9984

3.50

4.50

5.50

6.50

7.50

8.50

9.50

10.50

0 5 10 15 20

IRI (m

/km

)

Years



125

y = 0.0044x2 + 0.2116x + 3.8289R² = 0.9985

3.50

4.50

5.50

6.50

7.50

8.50

9.50

10.50

0 5 10 15 20

IRI (m

/km

)

Years

Annual Rainfall (1526mm)

final report plot 10, saddler lane, naguru p.o. box 8699...

Documents