1 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA MOBILITY AND TRANSPORT CONNECTIVITY SERIES SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA 2 MOBILITY AND TRANSPORT CONNECTIVITY SERIES © 2021 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissem- ination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Disclaimer—This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they repre- sent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202- 522-2625; e-mail: pubrights@worldbank.org. Foreword Africa point out that in some Countries RFs could gen- erate resources based on users’ charges that could be dedicated to cover maintenance needs combined with rehabilitation investments under PBCs approach in a more efficient way. When revenue generation is sta- ble and ringfenced, RFs can also potentially be used to leverage commercial financing for new investments, or to backstop government obligations under PPP schemes (e.g. Availability Payments, Construction Milestone Pay- ments). This, in turn, can enable a greater participation of private entities, not only as contractors, but also as long-term investors and managers of both brownfield The Maximizing Finance for Development (MFD) frame- and greenfield road assets. The modernization of RFs work is part of a global agenda to crowd-in private and the increase of private sector participation in road sector resources and solutions. In this regard, the asset management, e.g. through road PPPs, are distinct World Bank Group is promoting a stronger collabora- but intersecting agendas each raising specific issues. tion across its Institutions and beyond, to help catalyze The availability and reliability of a robust public funding finance for development. This is the rationale why the is a fundamental factor of feasibility of PPPs (particularly World Bank, together with PPIAF, IFC and African Devel- those structured as Gov.-Pays). RFs could play a role in opment Bank, has been developing a study that seeks to that regard in the development of road PPPs. However, (i) analyze the performance of second-generation road it would require a clarification of their mandate, a com- funds (RFs) in Africa; (ii) examine how to develop and prehensive legal and institutional framework, a strong transform RFs into a new (third) generation of instru- governance system, and sustainable and diverse sources ments and (iii) explore how to scale up private participa- of financing. Some RFs were able to raise funds on finan- tion in the financing of road assets, through public-pri- cial markets, but pledging future revenues to raise long- vate participations (PPPs) including performance-based term finance, if it may increase the funding available for contracts (PBCs). For instance, the World Bank has sup- road assets at a given time, is not equivalent to a PPP ported the creation of second-generation RFs and road which involves risk transfer to the private sector. PPPs in agencies to increase the financial and technical capac- the roads sector, on the other hand, could be looked at ity and ensure the sustainability of road investments. from the point of view of the “private sector’s appetite”. The key drivers behind the successful experiences are From this perspective, RF’s participation is simply a fac- effective regulatory frameworks, increased levels of tor, among others. The rationale of this study is to imag- autonomy and greater capacity for planning, budgeting, ine how RFs can play a useful role in the development financial management and auditing. The recent analysis and management of road assets in LICs, though bank- carried out by the World Bank covering most of RFs in able road PPPs or PBCs, as part of an MFD approach. Ibou Diouf Former SSATP manager Practice Manager Central Africa transport unit Disclaimer This report is based on Road Fund (RF) data collected between October 2018 and May 2019 – these data have been compiled in a digital folder available separately. The data collection process has been difficult and time-consuming, except for the information regarding the few RFs that timely publish an annual report on their website. The authors are confident, however, that the data obtained are representative and support the conclu- sions and recommendations of the report. This report addresses two distinct but intersecting agendas: (1) how to strengthen RFs, and (2) how to scale up private sector financing in the road sector in Sub-Saharan Africa. Acknowledgments This study was conducted by a team of consultants led by Alain Labeau and comprised of Julien Morel, Gilles Veuillot, Pablo Goulemot, and Youssouf Sakho. The Team benefited from the input of Eunice Wahome, World Bank intern. The team worked under the guidance of a WBG steering committee led by Federico Antoniazzi (WB). The Team would like to acknowledge the assistance of, and comments and suggestions made by Pierre Pozzo di Borgo, Daniel Pulido, John Graham, Elsa Le Borgne and Elisabeth Tedros (IFC); Nicolas Peltier, Ben Eijbergen, Fatouma Toure Ibrahima Wane, Daniel Benitez, Pankaj Gupta, Jean-François Marteau, Ben Gericke, Mustapha Benmaamar, Danye Aboki, Marc Navelet, Kulwinder Rao and James Markland (WB); Patrick Rugumire and Jean Kizito Kabanguka (AfDB); Ibou Diouf (SSATP) and Philippe Neves (PPIAF). The report was edited by Adam Jankowski (WB). This work would not have been possible without the collaboration of the African Road Maintenance Funds Association (ARMFA) whose President, Souleymane Traore, kindly invited the Team to their 17th General Assembly held in Namibia. The team wishes to express its gratitude to the many heads of Road Funds who shared information at this venue. Finally, the Team would like to acknowledge Abiy Woretaw, Deputy General Manager of the Ethiopian Toll Roads Enterprise, and Allan Munyua, East Africa Director of Meridiam for their time and sharing of insightful information. The views expressed are those of the authors and not necessarily those of the WBG. Contents Foreword.................................................................................................................................................................................................................................3 Disclaimer..............................................................................................................................................................................................................................4 Acknowledgments.............................................................................................................................................................................................................4 List of abbreviations..........................................................................................................................................................................................................6 Glossary of terms...............................................................................................................................................................................................................7 Executive Summary...........................................................................................................................................................................................................9 I. Introduction............................................................................................................................................................................................. 12 II. Road Funds in Sub-Saharan Africa: background and recent evolutions................................................................................14 1. Background on Road Maintenance Initiative in Sub-Saharan Africa.......................................................................15 2. Main conclusions from the 2006 Performance Survey.............................................................................................. 16 3. 2013-2017 Performance Survey ................................................................................................................................... 16 4. Current RF grouping with reference to the 2nd Generation status........................................................................19 III. Public-Private Partnerships in the Roads Sector......................................................................................................................... 21 1. Road maintenance funding in Sub-Saharan Africa: progressively reducing the dependence on fuel levies and shifting toward more distance-based charges...........................................................................22 a. Overview of possible road funding instruments and criteria for suitability in the context of Sub-Saharan Africa.......................................................................................................................... 22 b. Key lessons from the qualitative assessment of possible instruments to diversify Road Funds revenues...... 25 2. How to make road PPPs more attractive for the private sector in Sub-Saharan Africa – lessons from the electricity generation sector........................................................................................................... 26 3. Lessons learned from road PPPs in Sub-Saharan Africa and Latin America.......................................................... 27 4. Key lessons to better prepare and structure road PPPs and scale-up private sector investment......................29 IV. Restoration Concept: a novel road PPP promotion instrument................................................................................................ 31 1. Restoration Concept......................................................................................................................................................32 2. Restoration Concept financial modelling impact.......................................................................................................35 3. A Road Restoration PPP model to scale-up private sector participation in Sub-Saharan Africa.........................38 a. Typical features of a Road Restoration PPP.............................................................................................................. 38 b. Typical commercial structure of a Road Restoration PPP........................................................................................ 39 c. Key risks allocation for the proposed Road Restoration PPP.................................................................................. 41 V. SSATP and World Bank Group assistance to implement the Restoration Concept............................................................ 43 1. A checklist of activities to implement the Restoration Concept..............................................................................44 2. Next steps........................................................................................................................................................................ 46 6 MOBILITY AND TRANSPORT CONNECTIVITY SERIES List of abbreviations AADT Average Annual Daily Traffic BNDES Banco Nacional de Desenvolvimento Econômico e Social DBFOMT Design Build Finance Operate Maintain and Transfer DFI Development Finance Institution EPC Engineering Procurement and Construction GDP Gross Domestic Product HGV Heavy Goods Vehicle LIC Low-Income Countries IBRD International Bank for Reconstruction and Development IDA International Development Association IFC International Finance Corporation IMF International Monetary Fund MIGA Multilateral Investment Guarantee Agency MDB Multilateral Development Bank MFD Maximizing Finance for Development NPV Net Present Value OPRC Output and Performance-Based Road Contracts PPA Power Purchase Agreement PPP Public-Private Partnership RA Road Agency RF Road Fund RMI Road Maintenance Initiative RRW Road Restoration Window RUC Road User Charge SOE State-Owned Enterprise SSA Sub-Saharan Africa SSATP Sub-Saharan Africa Transport Policy Program VOC Vehicle Operating Cost WBG World Bank Group 7 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Glossary of terms Maintenance (periodic): The planned resurfacing financed Design-Build-Maintain-Operate-Transfer for- of a paved road or the scheduled placement of more mat. The Contractor in an OPRC may, however, pre-fi- gravel on the wearing course of an unpaved road to nance a very limited portion of the capital expendi- account for that lost due to vehicle usage. tures (usually on its balance sheet). Maintenance (routine): The work that is planned Performance-Based Contract: Where payment is and performed on a routine basis to maintain and made based on the quality of the asset provided (e.g. preserve the condition of the road. US$ for having the road within a specific roughness limit). The Contractor takes the risks related to the Milestone Payments: In this report, Milestone Pay- resources, quality and quantity of work. ments are understood as payments made by a con- tracting authority during the construction period of a PPP: A long term contract between a public party PPP or an OPRC, to (partially) compensate for the cost and a private party for the development (or signifi- of works. These payments are usually made as works cant upgrade or renovation) and management of a are progressing based on the achievement of pre- public asset (including potentially the management defined milestones. Milestone Payments (also known of a related public service). Under such contract the as Investment Grant or Subsidy) are a form of Viabil- private party bears significant risk and management ity Gap Funding in a PPP1. responsibility throughout the life of the contract. It must provide a portion of the financing at its own Off-taker: Designates the entity that is committed to risk while its remuneration is linked to performance purchase the output of an electricity generation proj- and/or the demand for the asset and/or services it ect. By analogy with the electricity generation sector, provides1. this term is used to designate the entity responsible for making Annuity Payments in a Government-Pays Government-Pays or Gov.-Pays (PPP): A sub-type road PPP. of PPP in which the private party derives its revenues from payments made by the public party. When these Output- and Performance-Based Road Contract: payments are not linked to usage (i.e. the number of An arrangement whereas the Contractor is respon- users of the public asset) but rather to the availability sible for the design of the rehabilitation, improve- of the asset at a certain level of service, they are also ment and emergency works required to reach and known as Annuity Payments1. maintain specified service levels over the contract period (employer may provide design for improve- User-Pays (PPP): A sub-type of PPP in which the pri- ment works). Rehabilitation and improvement works vate party derives its revenues from payments made are executed upfront, followed by the operation by the users of the public asset. User-Pays PPPs are and maintenance periods. It is part of the perfor- also known as “concessions” in many jurisdictions1. mance-based contracts and usually follows a publicly 1 APMG PPP certification guide. Chapter 1: Public-Private Partnership – Introduction and Overview. The APMG certification program is an innovation of most multilateral development banks (ADB, EBRD, IsDB, IADB, WBG), which was funded by PPIAF. 8 MOBILITY AND TRANSPORT CONNECTIVITY SERIES Priority Alignment: In this report, priority align- Road financing: How funds are raised at the outset ments are understood as the main roads connecting of a project/program to meet investments and/or national and international business/urban centers. maintenance needs. The financing may flow from the public sector (either from cash reserve if any or debt Project Company: In this report, a Project Com- raised by a public entity such as a Road Fund), from pany is understood as the contractual counterparty the private sector (either from equity or debt raised of the Contracting Authority in a PPP arrangement. by a private entity such a Project Company in a PPP A Project Company is constituted specifically for the arrangement) or a combination of both2. purpose of signing and executing the PPP contract. It may also be referred to as Special Purpose Vehicle Road funding: Who ultimately pays in the long term (SPV) or private party1. for the investments and/or maintenance of the roads. The funds may come from taxpayers and/or road Restoration Contract/Road Restoration PPP: A typ- users2. ical restoration contract starts with roadway repairs, stabilization (resurfacing) and reconstruction works Road Agency: An agency created via enabling leg- immediately followed by structural, safety, traffic, islation or ministerial decrees. Road Agencies are and climate change-related improvement works. As intended to be leaner, more consumer-oriented and soon as rehabilitation and improvement works are market-responsive than traditional Road Adminis- completed (2-3 years), operation and maintenance trations. As independent legal entities, they can sign activities can start for at least one cycle of periodic contracts. maintenance (>7 years). In the context of this study, a Restoration Contract is interchangeably used with Road Fund: Special account into which the proceeds Road Restoration PPP. of the collection of road users’ charges (e.g. vehicle license fees, heavy vehicles license fees, interna- tional transit fees, fuel levy, bridge and ferry tolls) are deposited to pay for road maintenance expenditures. 2 CEDR (Conference of European Director of Roads). Funding formulas of roads: inventory and assessment. March 2017. 9 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Executive Summary The purpose of this study is to evaluate non-tradi- or never reached it, and only a few countries have tional means to raise additional private financing for been able to raise enough resources through RFs to the upgrade and maintenance of developing coun- fully cover their road maintenance needs. The same tries’ road networks. To achieve this goal, it combines survey highlights the pervasive issues that currently an in-depth review of Sub-Saharan Africa (SSA) coun- perpetuate the Build-Neglect-Rebuild vicious cycle tries’ Road Funds’ (RF) performance and Road PPPs (see Figure A). to evaluate the potential for RFs to fund road PPPs when specific conditions are met. 2. There is a need to stress the importance of sustainable advancement of SSA country RFs towards 3rd Generation status, by first ensuring Upgrading Sub-Saharan Africa countries’ that the criteria for 2nd Generation status are met Road Funds to a 3rd Generation status and upheld in a sustainable manner. The 3rd Gen- eration status is currently only achievable by some 1. Road Funds (RFs) were created as part of the RFs and focus should be placed on the foundational Road Maintenance Initiative (RMI) in the late 80s. importance of helping those funds that have yet to A performance assessment conducted in 2006 achieve 2nd Generation status to do so, and to assist showed mixed results, although it recognized the country RFs that have lost their 2nd Generation that the so-called 2nd Generation RFs were able to status to regain it. It also needs to be stated that this secure more resources for road maintenance. A study proposes a country-specific approach rather survey of RF performance over the 2013-2017 period than an abstract set of criteria, which led to the mixed shows that some RFs lost their 2nd Generation status, success of the 2nd Generation RF initiative. Figure A: The pervasive Build-Neglect-Rebuild vicious circle. Source: Authors’ analysis. 10 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 3. Even though many of SSA countries’ RFs may funding schemes over others. Tolling requires a currently not be ready for the transition, there is sensitive approach with consideration for country already a potential to upgrade a few RFs to 3rd Gen- specifics. Among all potential funding instruments, eration status. This could allow them to leverage few present the most desired characteristics of a their resources to mobilize private financing. The sustainable funding system: equity, yield potential, success achieved by some RFs could pave the way to administrative simplicity, and consistency with user- improving RFs’ ability to generate substantial, stable pays principle. Fuel levies meet these characteristics. and regular ring-fenced revenues and to use them However, raising them presents some challenges to underwrite credit worthy road PPPs. Taking into (e.g. political resistance to de-funding fiscal resources account the lessons learned from the 2nd Generation any further, and the increasing share of fuel-efficient exercise and the RF performance surveys, a selec- or electric vehicles). Distance-based charges, such as tive pilot approach has been identified as optimal for classic tolling or Heavy Goods Vehicle (HGV) charges, the transition to 3rd Generation status This is set to also possess the desired features. They have proved tailor the approach to particular country conditions successful when adequately implemented (e.g. in the in order to ensure that the 3rd Generation status is Polish experience). Access-based charges such as reached and upheld by those countries in a sustain- vehicle registration and licensing fees also meet most able manner. Based on a number of factors discussed sought-after characteristics. However, their yield in this report, the countries with the RFs with the potential is lower, and they require a high level of highest potential to achieve 3rd Generation status administrative enforcement. have been identified as Kenya and Ivory Coast. 5. The second step is to rethink how road PPPs are selected, structured and prepared in SSA. Attracting the private sector to road Private sector participation in the road sector is low PPPs in Sub-Saharan Africa compared to the electricity generation sector in the region. This can be explained by factors inherent to 4. The necessary first step is to design and the nature of road projects (e.g. the relatively long secure a sustainable funding system by raising, construction period, or the multiplication of technical, and adjusting for inflation, fuel levies to at least social and environmental issues), as well as the low the recommended value of US$15c/l equivalent. level that user tariffs (i.e. tolls) must be set at to gain Afterwards, reliance on complementing sources of social acceptability. These issues can be somewhat revenue, such as inflation-adjusted tolls and other mitigated if the appropriate projects are selected (i.e. forms of distance-based charges (e.g. Heavy Goods those that present the least technical and E&S issues, Vehicles charges) should be investigated as well. This which is naturally associated with brownfield/pre-ex- study intends to present available solutions, without isting roadways) and if the challenges linked to user taking a stance on their appropriateness under coun- charges are partially or fully eliminated (i.e. indirect try circumstances. The listing of the different funding user payments are replaced by a contracting authori- systems’ advantages and disadvantages does not ty’s direct payments). signify the endorsement of some of the enumerated 11 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA A Restoration Concept to lastingly improve high- Supporting client countries in demand roads and scale-up private participation implementing the Restoration Concept in the Sub-Saharan African road sector 7. Implementing the Restoration Concept is 6. The Restoration Concept proposes to anchor expected to require a multidimensional sup- the benefits of the RMI by upgrading qualified port from the World Bank Group (WBG) to client RFs to a 3rd Generation status and using them as countries. Designing a WBG advisory and financial a creditworthy counterpart in a series of bank- support for client countries, from a turnkey solution able Road Restoration PPPs meant to lastingly to a flexible one-stop-shop window, is beyond the improve high-demand roads. This approach could scope of this study. However, the report proposes a be based on the establishment of a ring-fenced Road checklist of appropriate and practical activities relat- Restoration Window (RRW) within RFs, which would ing to the various phases of implementation, from be used to fund privately financed Restoration Con- upstream to downstream. These activities range from tracts. These contracts would primarily target existing increasing the knowledge on road conditions and road network Priority Alignments. Over time, the pos- drafting standardized tender documents for the Road sibility to combine increased fuel levies with toll reve- Restoration PPPs, to introducing sector reforms. The nues from some sections of these alignments would proposed activities would mobilize a large array of promote the expansion and sustainability of the Road WBG non-financial (e.g. technical assistance, PPP Restoration PPP program. This model is designed to advisory services) and financial instruments (e.g. address the impediments to attracting more private credits/loans, guarantees). Collaboration with other participation in the SSA road sector. It draws from donors (AfDB, UE, SSATP) is also essential to achieve a lessons learned from the road PPP projects and broad consensus and scale-up this approach. programs analyzed as part of this study. The model comprises (1) a typical commercial structure where the Road Agency (RA) is the contracting authority and the RF is the authority responsible for making the Annuity Payments to the Project Company, and (2) an allocation of the key risks with an outline of their mitigation mechanisms. A simulation of the Resto- ration Concept over a period of 30 years highlights the growing impact of RF’s increase in revenues over time. As the total length of restored high-demand roads expands, more funds will become available to carry out roadway maintenance financing as well as mobilize private sector financing. 12 MOBILITY AND TRANSPORT CONNECTIVITY SERIES I. Introduction 13 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Until now, private sector financing in SSA has been mostly limited to a mix of green- and brownfield road projects, mainly executed in urban areas. The scale of these projects has been limited by: i) the ability of host nations to monetize road user demand (i.e. toll rates based on socially/politically acceptable prices rather than ability to pay), and/or ii) their fiscal cred- ibility in providing long term, off-take, such as annu- ity payments to private operators/investors. Conse- quently, it is estimated that private financing in the road sector accounts for less than 10% of global road financing needs in emerging markets and even less in Sub Saharan Africa (SSA). This lack of private funding for roads has left local Governments with the task of mobilizing enough money to carry out road networks preservation and expansion. In the former case, this has resulted in the establishment of intermediary public payment agents known as Road Funds (RFs), mostly in SSA. These RFs, which have been supported by Road Agencies (RAs), haven taken on the dedicated role to provide financ- ing mostly for road maintenance activities. Some of them have evolved over time from structures located within the Ministries of Public Works or Transport to fully-fledged, separate public agencies. In this role, these RFs have so far been unable to play a credible role as an off-taker of public annuity payment obliga- tions towards private operators/investors or to lever- age their future funding streams into larger invest- ment programs through the raising of long term, local or international, debt. This report proposes to explore how, in few selected cases, SSA RFs could be reformed to substantially increase the amount of public and private monies flowing towards the maintenance and/or upgrade of the core road networks of SSA countries. 14 MOBILITY AND TRANSPORT CONNECTIVITY SERIES II. Road Funds in Sub-Saharan Africa: background and recent evolutions 15 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA 1. Background on Road Maintenance Initiative in Sub-Saharan Africa3 8. A ruinous road asset management legacy. SSA 10. The failure of 1st Generation RFs led to a rec- ommendation for establishing 2nd Generation countries expanded their road networks considerably RFs. To address 1st Generation RFs’ weaknesses, the from the moment they gained their independence RMI suggested that road assets should be, to the until the 1980s. They failed to keep them in good extent possible, commercialized. This required under- condition, however, with too little spending allocated taking reforms in four areas (referred to as the four to both routine and periodic maintenance. By the building blocks): early 1990s, nearly one third of the investment made in road assets had been lost and SSA road networks a) Ownership: effectively involve road users in the were mostly in poor condition. They had accumulated management of roads to win support for increase US$43 billion in deferred maintenance backlog. in taxation, control potential monopoly power, and limit road spending to what is affordable; 9. The Road Maintenance Initiative (RMI) was designed to remedy poor road network condi- b) Financing: secure an adequate and stable flow of tions through the creation of Road Funds (RFs funds; of the 1St Generation). The United Nations Eco- nomic Commission for Africa and the World Bank c) Responsibility: clarify responsibility for fund col- launched the Africa RMI in 1989 to identify the lection, network to be maintained, size of annual underlying causes of and remedies for poor road work program, personnel hiring and firing; and network maintenance policies in SSA. It concluded that road assets were managed within a disabling d) Management: as part of a stand-alone RF agency, institutional framework whereby funding of mainte- strengthen financial management by using effec- nance activities depended exclusively on scarce and tive programming and performance monitoring erratic general tax revenues. To correct this issue, it systems, procurement and payment procedures was recommended that SSA countries establish RF and checking compliance through independent accounts which would be funded directly by road audits. user charges to specifically pay for road maintenance activities. These RF accounts would be off-budget line 11. RFs were complemented by the creation of items managed by the relevant line ministry. They did Road Agencies (RAs) to execute road maintenance not involve the creation of dedicated entities. Sadly, programs more effectively. To reduce governance they performed poorly due to a host of issues (e.g. interference in the execution of road work programs, absence of independent audits, use of funds for non- RMI suggested the creation of autonomous RAs under road related expenditures and weak oversight/finan- the oversight of the Ministry of Transport or Public cial management). Works. The role of RAs was going to: i) collect traffic data and monitor the condition of the road network, ii) prepare road work programs and execution plans underpinning the strategy adopted by the Govern- 3 Heggie I.G. Management and Financing of Roads - An Agenda for ment, and iii) use RF funds to implement and manage Reform. World Bank Technical Paper Number 275, Africa Technical Series, 1995. road maintenance work programs and contracts. 16 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 2. Main conclusions from the 2006 Performance Survey 12. A review of the financing of road maintenance cases, funds earmarked for road maintenance in SSA covering 27 active RFs in SSA was carried remained insufficient with only 11 out of the 27 out in 20064 to assess the progress towards 2nd RFs surveyed meeting their routine maintenance Generation RFs. It concluded that: expenditure needs. Only 13 countries reported that direct funding for RF budgets was is in place; – 2nd Generation RFs had become a significant and feature of road sector reform programs in SSA but enabling reforms supporting them had led to dditional efforts were required to: i) diversify RFs’ – A mixed results; revenue sources (e.g. expand road user charges) and channel them directly to their accounts, ii) – 2nd Generation RFs had secured a more stable make better use of available financial resources flow of funds for road maintenance. In most by improving road management practices. 3. 2013-2017 Performance Survey 13. RFs’ levels of resources vary widely across inflated by commercial loans, which conceal a low SSA but remain insufficient to cover the cost of fuel levy. Ethiopia comes last because it has elected routine and periodic maintenance. Total revenues to invest massively in the expansion and develop- from Road User Charges (RUCs) ranged in 2016 from ment of its road network rather than its maintenance. US$5M equiv. in Burundi to about US$600M equiv. in Kenya. Relative to GDP, RUCs represent anywhere 15. The chronic maintenance funding gap forces from 0.1% to 1.2%. The range among countries is RFs to fund investment works. Constant mainte- narrower when considering the percentage of classi- nance backlog forces RAs to undertake spot recon- fied road network whose maintenance is covered by structions when sections of the road network have RFs mainly because of inherent differences in the size reached a condition when maintenance is no longer of each country’s classified network (see Figure 1). a viable technical solution. Conversely, RFs end up funding these activities rather than financing routine 14. The 2017 ranking of RF Revenues/GDP can maintenance. This explains why a growing number of be misleading. Sierra Leone’s ratio is overstated them have requested amendments to their articles of because of its low GDP. That of the Ivory Coast is incorporation and by-laws to include these activities. overstated because its RF’s revenues are artificially 4 Benmaamar, M. 2006. Financing of Road Maintenance in Sub-Saharan Africa, Discussion Paper No 6, Road Management and Financing Series, The Sub-Saharan Africa Transport Program. 17 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Figure 1: RF revenues in a sample of SSA countries. Source: Authors’ analysis. 18 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 16. Since most RFs do not fully spend their annual components of RUC revenues are vehicle license and revenues, some can end up with large surpluses. registration fees and tolls. While tolls have the poten- Unspent funds are either funds that were not tial to generate substantial revenues, they are sel- released to the RF or funds released to but not spent dom used mostly because of the public’s resentment by RAs. The former case is generally due to a delay against them. in the collection and transfer of funds to the RF. The latter is a direct consequence of low financial execu- 18. While fuel levies vary widely among coun- tion rate by RAs. The percentage of unreleased funds tries, they remain below levels necessary to fully is more difficult to track. In Kenya, it is estimated to fund road network maintenance needs. Studies have reached 21% of the total earmarked funding carried out, inter alia by RMI (PAM, 2004), suggest between 2015 and 2017. Meanwhile, the financial that the fuel levy should exceed US$10c/l to meet all execution rates5 across RAs ranges from 60% to 90%, road maintenance needs. In SSA the actual average which is mainly linked to a cumbersome procurement is probably much higher at about US$13c to 15c/l, process and road works delays tied to unskilled local or equiv. to US$17.4c/l in 2017 when adjusted for road contractors. inflation. These numbers are much higher than the median of about US$7.5c and the average of about 17. RFs gather most of their resources from US$9c/l recorded in 2017 in SSA (see Figure 2). RUCs, and the highest share still comes from fuel levy revenues. On average, RFs derive 80% of their 19. Some RAs have gained experience in using resources from RUCs except for Senegal and Cam- long-term performance-based contracts (PBC) eroon which receive a relatively low level of RUCs in lieu of traditional contracts which lessen the because of complementary Treasury financing. RUCs inherent drawbacks of classic contracting (i.e. are comprised of fuel levies (a fraction of the excise underutilization of funds, better long-term plan- duty on road fuels), vehicle license and registration ning and execution of road maintenance strate- fees, axle load tax, driver’s license fees, load-distance gies, reduction in cost overruns, etc.). Countries charges for HGVs, international transit tolls, overload like Chad, Kenya, Tanzania, and Zambia have gained fines6, road tolls and ferry/bridge tolls. The fuel levy experience with the use of PBC. This welcome move is a funding instrument common to all RFs, contrary is still modest and uneven in SSA countries because to the other abovementioned instruments. On aver- of the capacity limitations of the local contracting age, the fuel levy collection makes up 76% of RUC industry and the resistance of the finance ministries revenues, ranging from 100% in Senegal and Guinea and central tender boards to long-term contracts. to about 30% in Niger. The second and third largest 5 A low execution rate means that the funds made available, possibly including funds carried over, cannot be spent entirely and leave an unused balance at the end of the fiscal year. These balances are added to possible unreleased funds and rolled over. Over time, if the financial execution rate does not improve, they can grow into significant surpluses. In countries where there is partial or no carry over, it is a net loss for road maintenance. 6 These resources are expected to decrease over time thanks to a more effective enforcement of axle load regulation and increasing truck compliance. 19 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Figure 2: Evolution of diesel fuel levy in a sample of SSA countries between 2006 and 2017. Source: Authors’ analysis. 20. Finally, most RFs’ annual reports lack clarity funding. Likewise, the production of standardized and simplicity when, and if, they are published. annual reports would enable government regulators The RFs’ oversight function is not facilitated by not only to benchmark their own RFs and RAs’ per- reporting requirements whose format varies from formance but also identify financial and performance one year to another. Introducing yearly independent issues early on. Detailed analyses of the RFs of Sen- road condition surveys combined with road traffic egal, Tanzania, Kenya, Chad, Cameroon, and Ivory surveys would allow RFs and RAs to determine sci- Coast (including an analysis of the legal and institu- entifically, and not politically, which sections of the tional framework applicable to the Ivory Coast RF) road network should receive priority maintenance have been prepared as part of this study. 4. Current RF grouping with reference to the 2nd Generation status 21. RFs can be grouped according to their abil- interaction with other institutions, etc. For these rea- ity to reach, sustain and upgrade from the 2nd sons, a ranking based purely on performance might Generation principles. The ability of RFs to play an not be relevant. What could be more telling is how efficient role is generally the result of a multitude of RFs have evolved with respect to the 2nd Generation factors, often beyond their control: clarity of the legal principles. This approach allows for the ranking of and institutional framework and its proper applica- RFs into three broad categories: tion, operational autonomy and political interference, 20 MOBILITY AND TRANSPORT CONNECTIVITY SERIES a) 1st Generation RFs that have never reached 2nd – Fund road investment works, to administer dis- Generation status; tance-based charges (see Box 1), collect 100% of distance-based charges (net of collection costs) b) RFs that have reverted from 2nd to 1st Generation as part of the RUC resources, to invest its funds status; and responsibly on the financial market; c) RFs that have consolidated their 2nd Generation – Raise long-term debt on favorable term from Gov- status and have the potential to upgrade to what ernments or Multilateral Development Banks, and could become a 3rd Generation status. in the long-term raise commercial debt possibly without sovereign guarantee; 22. This grouping is provided for comparison purposes only to highlight how similar the oper- – Abide by stricter regulations and reporting and ational performance of similar institutions – oversight functions7. The condition of the roads namely Road Funds - varies and delivers different under their purview would be surveyed regularly results. The RFs in the first group are: Cameroon and by an independent party; Senegal. The Gabon RF even disappeared completely recently. The RFs belonging to the second group are: – Operate in a country where: there is a positive Benin, Chad, Burundi and Mozambique. The RFs in track record for implementation of RF legisla- the third group are Ivory Coast, Kenya, Zambia, Libe- tion, a mature PPP Law and an experienced PPP ria, Malawi and Tanzania. Unit; where the Procurement Authority autho- rizes long-term contracting and has a good track 23. The 3rd Generation status would be an upgrade/ record for procurement processing; where the strengthening of the 2nd Generation status mainly convertibility restrictions on national currency are in terms of funding sources and amounts, scope limited and where the national road construction of works to be carried out (e.g. road rehabilitation industry comprises a reasonable number of expe- works), governance and overall autonomy. It would rienced large contractors; be designed specifically to address all the known shortcomings of current 2nd Generation RFs. As such – Work in a team and under a clear separation of it would: duties with a capable RA; and – Meet all the 2nd Generation requirements, includ- – Fund, among others, Restoration Contracts (see ing collecting a fuel levy regularly adjusted and Chapter IV). equivalent to at least US$15c/l in 2018; 7 Oversight and regulatory functions of the government are often weak because of a lack of qualified staff and/or interest but can become over- prescriptive when financial transactions are ruled by a public accounting officer. Both situations are unhelpful. Adding to the problem, RF reporting can be complicated or obscure and lack simple score cards, and financial audits are often conducted mechanically with little understanding of the true mission of the RF. It looks like RFs, and their Board, self-evaluate and do not seem to be challenged often by their supervising authority. Conversely, there are cases where the supervising authority interferes in the RF’s mission, e.g. by diverting funds earmarked for maintenance into new road investments (Cameroon, Sierra Leone). 21 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA III. Public-Private Partnerships in the Roads Sector 22 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 1. Road maintenance funding in Sub-Saharan Africa: progressively reducing the dependence on fuel levies and shifting toward more distance-based charges 24. Before designing sophisticated road-financ- i. Affordability refers to the financial capacity of ing schemes involving the private sector, the first users. It varies widely and can be limited; necessary step is to design a sustainable road funding system. The public road network has been ii. The yield potential is linked to the revenue’s identified as the largest public infrastructure asset, generation potential of the asset under manage- and SSA countries particularly have a higher value ment. It needs to be predictable and stable over of road asset value to GDP ratio compared to world the long-term. It is a key feature of any project if average8. SSA countries will need to spend more and private sector participation is sought after; better on their road networks to realize their eco- nomic potential and achieve the Sustainable Develop- iii. Users’ shared interest is predicated upon the ment Goals. To allow private investment to support implementation of the user-pays principle. It is the SSA road asset program, the funding structure important to ensure that different categories of will determine the financing options and the bank- road users pay their fair share of the wear and ability of the program. tear they cause to the road (particularly HGVs) in order to avoid socially unfair cross subsidies; and a. Overview of possible road funding iv. Administrative simplicity advocates for trans- instruments and criteria for suitability parent, easy to understand and audit, rules to in the context of Sub-Saharan Africa enforce assets operations and maintenance obligations in the context of limited institutional 25. Countries around the world usually rely on a capacities. mixture of instruments9 to fund their road sec- tor as depicted in Box 1. The suitability of these instruments in the context of SSA countries may be assessed against a set of qualitative criteria/most desired features such as: i) affordability, ii) yield potential, iii) users’ shared interests, and iv) adminis- trative simplicity: 8 Brushett S. Management and financing of road transport infrastructure in Africa. Sub-Saharan Africa Transport Policy Program (SSATP). Discussion Paper 4, March 2005. 9 E.g. Acosta, L. National Funding of Road Infrastructure – comparative summary. The Law Library of Congress, March 2014. 23 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Box 1: Eight categories of road funding instruments General all-purpose taxes2,10,11. They do not have a direct link to the road network use or development. They may consist in direct taxes (e.g. income tax) or indirect taxes (e.g. Value-Added Tax). Non-recurrent access-based charges12,13. They are related to the purchase of a motor vehicle. They are one-off charges paid to allow the use of a motor vehicle on the road network. They include a variety of charges e.g. Vehicle Registration Fee, Custom Duties for imported vehicles, Luxury Car Taxes. Recurrent access-based charges11,13. They are paid regularly by vehicle owners and users. These charges include for example: Vehicle Licensing Fee, Axle Tax, Insurance Contract Tax, Driver’s License Fee. Distance-based charges2,14. They are paid directly by the user in exchange for the use of the road. They are usually based on a unitary tariff (e.g. US$ct/mil traveled). The tariff can be modified according to sev- eral parameters such as maximum authorized weight, number of axles, period of the day or even Green House Gas (GHG) emission class. These charges can be paid by the user either using cash or electronic fund transfer and consist in: tolls (on specific stretches and linked to the repayment of road investment, e.g. in a User-Pays PPP scheme31), international transit fees, HGV charges (on a network basis and not necessarily earmarked to repay a specific investment). Time-based charges2, 10, 14. They are paid directly by the user in exchange for the right to use the road (or network) during a specified period. These charges are usually based on a unitary tariff (e.g. US$/ day, week, month or year). The tariff can be modified according to several parameters such as maxi- mum authorized weight, number of axles, period of the day or even GHG emission class. These charges include for example: vignettes and urban charges. (Box.1 continues on the next page) 10 Ministry of Transport of New Zealand. Future Funding - Revenue tools for transport, November 2014 11 Chen, C., and Bartle J.R. Infrastructure Financing: a guide for local governments managers. Prepared for International City/County Management Associa- tion and Government Finance Officers Association, 2017 12 Deloitte. Road Pricing and transport Infrastructure funding: reform pathways for Australia – Discussion paper. 2013. 13 Association Mondiale de la Route. Financement, dévolution et gestion des investissements routiers – Comité Technique A.2 de l’AIPCR, rapport 2012R08FR. 2012. 14 Schwarz-Herda, F. Road pricing for heavy vehicles: a key for financing road infrastructure? A successful example in Austria. Route – Roads 2013, volume 358. www.piarc.org. 15 Committee for a study of the future Interstate Highway System. Renewing the National Commitment to the Interstate Highway system: a Foundation for the Future. Transportation Research Board. 2018. 24 MOBILITY AND TRANSPORT CONNECTIVITY SERIES Fuel-consumption based charges12, 15. They are excise taxes, meaning that they are included in the price paid at the pump by consumers and the taxpayer is either the fuel producer or distributor. They consist in a unitary tariff (e.g. US$/gal), which can be modified depending on the fuel type. For example, less polluting fuels like Liquefied Petroleum Gas (LPG) or Compressed Natural Gas (CNG) may have a lower rate. The “fuel levy” which represents in average 80% of the RFs’ resources (see para. 17) is usually carved out of the fuel excise tax. Value Capture charges16. Value capture is defined as the public recovery of all or a portion of increased property value created because of public infrastructure investment, or that takes benefits from it. Sub- ject to enabling conditions (e.g. real estate market vitality, zoning and land use entitlements), new road capacity and new road accesses can create business opportunities and value in the surrounding land and real estate. They include for example: Impact fees, Special Assessment Districts or Tax Increment Financing. Toward a Universal Road User Charge? 10, 12, 15. Recognizing the limits of the current road funding sys- tem, the road maintenance and investment gaps, and the future shortcomings of fuel taxes as one of the main funding instruments, some countries have started considering a new road pricing scheme that would first complement and then replace current funding instruments. The related charging arrange- ments would apply to all motor vehicles and the entire road network. The concept is basically to charge users for the distance they travel on roads. In addition to distance the pricing could include: Vehicle weight class, Time of day, Location, Type of fuel (or GHG class). Source: Authors’ analysis. 15 Committee for a study of the future Interstate Highway System. Renewing the National Commitment to the Interstate Highway system: a Foundation for the Future. Transportation Research Board. 2018. 16 Page, S., Bishop W.L., Wong W. Guide to value capture financing for public transportation projects. TCRP (Transit Cooperative Research Program Research) Report 190. National Press Academies. 2016. 25 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA b. Key lessons from the qualitative assessment of possible instruments to diversify Road Funds revenues 26. Road funding systems are usually complex 27. In the short-term, fuel-consumption based and opaque with associated public acceptance charges (e.g. fuel levies) remain a pragmatic issues. Countries generally rely on a combination of road funding instrument. A shift toward more funding instruments which are by design earmarked distance-based charges might be necessary in the for the road sector. The path of returning collected mid- to long-term as fuel consumption growth may funds to investment, maintenance and operation of temper off or turn into a decrease with the mass roads is complex, however. In some countries, reve- introduction of electric vehicle fleets. Table 1 sum- nues from road sectors subsidize other policies (e.g. marizes the qualitative assessment of possible road some European countries) while in others it is the funding instruments conducted in the context of SSA opposite that happens (e.g. USA)17. countries against the most desired features outlined in para. 28. Table 1: Assessment of the most desired features of road funding instruments in SSA countries Users’ Shared Administrative Affordability Yield Potential Interest Simplicity General taxes Low Medium Low High Non-recurring access-based charges Low Low Low medium Recurring access-based charges Low Medium Medium Medium Distance-based usage charges Medium High High Medium Time-based usage charges Low Medium Medium Low Fuel consumption-based user charges Low Medium Medium High Value Capture Medium Medium Low Low Universal Road User charges Medium High High Low Source: Authors’ analysis. 17 Gomez, J., and J. Vassallo. 2014. Comparative Analysis of Road Financing Arrangements in Europe and the United States. Journal of Infrastructure Systems, Vol. 20, No. 3, Sept. 26 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 28. To anticipate fuel levy shortcomings, Govern- implementation of a Universal Road Charging ments in SSA countries need to shift toward more system in the long term as the sole road funding distance-based charges as these charges present instrument. In countries where enabling conditions a significant yield potential. Even if implementa- are fulfilled (e.g. sophisticated fiscal administration, tion and operation costs are higher than fuel con- adequate institutional capacity, dynamic real estate sumption-based charges and can prove challenging, market), value capture mechanisms could be intro- distance-based charges are generally accepted by duced to fund targeted capital expenditures in the road users when they translate into better road con- road network. These instruments require a high dition and increased road safety. However, ensuring degree of sophistication in terms of fiscal manage- willingness to pay requires a strong political will and ment (e.g. proper land titles, and efficient property a relevant strategy to publicize benefits to overcome tax assessment systems) combined with the ability initial opposition. of public authorities to engage in transparent coop- eration with private real estate developers. For these 29. Preparing now the implementation of reasons, these mechanisms are probably out of reach distance-based charges on a network basis for now for most SSA countries. (e.g. HGV charges) could pave the way to the 2. How to make road PPPs more attractive for the private sector in Sub-Saharan Africa – lessons from the electricity generation sector 30. In SSA countries, private sector participation 31. This failure to attract private sector participa- in infrastructure is concentrated in electricity tion can be explained by factors inherent to the generation projects. The World Bank’s Private Par- nature of road projects. As linear infrastructures, ticipation in Infrastructure18 database shows that the roads cross a variety of land, multiplying technical, electricity sector accounts for 51% of private sector environmental and social issues compared to the participation whereas the road sector accounts for limited land-related challenges faced by electricity only 3%. This discrepancy is even more pronounced generation projects. Among these issues, right-of- in LICs. Only one road PPP reached financial close way acquisition and clearing, and the associated during the last decade whereas 46 electricity gen- resettlement of communities are probably the most eration projects did so. Moreover, during the 2011- prominent. Construction periods can also be much 2017 period, 41 projects (21 located in SSA), received longer (as long as 5 years20), which is a challenge in private participation from institutional investors, but project-financed projects such as road PPPs. only one of them was a road project19. 18 https://ppi.worldbank.org/ 19 World Bank. Contribution of Institutional Investors – Private Investment In infrastructure 2011-H12017. 2017. 20 The 4G road PPP in Colombia have for example construction phase lasting around 5 years. 27 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA 32. Project preparation and structuring decisions most countries, it tends to make financially weak local made by contracting authorities also contribute contractors’ participation all but impossible. Most to making road PPPs less attractive than electric- importantly in LICs, when 99% of the Project Compa- ity generation projects. Road projects presented to nies derive their revenues from payments from the the market in SSA countries are about 35% more cap- Government or a public utility (under Power Purchase ital-intensive than electricity generation projects (75% Agreements - PPAs), 75% of Project Companies oper- if only focusing on LIC in SSA). The capital-intensity ating road projects generate their revenues from tolls can have adverse impacts on: i) affordability of the collected from road users. This implies that these road PPP for users and the contracting authority, and companies are exposed to traffic risks that equity ii) attractiveness for lenders and equity providers. In providers and lenders are averse to. 3. Lessons learned from road PPPs in Sub- Saharan Africa and Latin America 33. The following key lessons are drawn from to downstream phases, potential investors need to case-studies of road PPP projects and programs in be convinced, inter alia, that bidders’ competition Latin America (i.e. Brazil and Colombia) and SSA will be fair and transparent, that the government’s (i.e. Ivory Coast, Liberia, Kenya and Senegal). Three objectives can be realistically achieved, and that the of these case-studies cover PPP projects (Henri Konan public funding mechanism that supports the PPP Bedie toll bridge in Ivory Coast, Ganta-Zwedru road framework is creditworthy. Public sector entities corridor rehabilitation in Liberia, and Dakar-Diamnia- need to be appropriately staffed and funded to face dio toll highway in Senegal), while the others cover the high workload generated during the structur- PPP programs (4G road PPPs in Colombia, Federal ing and tendering phases as well as during contract highway concessions in Brazil, toll roads and roads management. annuity programs in Kenya). These projects and pro- grams cover a wide range of features (e.g. brownfield 35. Lesson 2: A programmatic approach with vs. greenfield, User-Pays vs. Gov. Pays). Most of them standardized documents generates many bene- benefitted from the support of the WBG. fits, among which the reduction in transaction costs for both public and private stakeholders is 34. Lesson 1: Successful road PPP projects or pro- the most notable. Due diligence of different types grams require high-level government champion- (legal, technical or financial) and other transaction ing. High-level government championing is a neces- costs are significant and may not necessarily be fully sary condition at all phases of the project/program recovered by losing bidders. Standardizing tender implementation. In the upstream phases, significant documents and preparing a pipeline of similar proj- sector (in transport and finance) and cross-sector ects creates economies of scale for potential bidders. reforms (creating an adequate PPP framework) usu- It can incentivize them to participate in several bids. ally need to be implemented to enable the develop- For the public sector, even if preparation of standard- ment of roads PPP project/programs. In midstream ized documents may take longer, a more streamlined 28 MOBILITY AND TRANSPORT CONNECTIVITY SERIES tender phase can yield substantial gains (i.e. 4G road portion of the investment financing needs (e.g. 42% PPPs in Colombia). Moreover, ensuring maximum of CAPEX for Dakar’s DTH). In more mature invest- bidder participation increases competition, which in ment markets, like Colombia, scaling-up private turn can help Governments achieve greater Value for sector investment in the road sector required the Money for these projects. adoption of a hybrid User-Pays and Government-Pays PPP model. The Brazilian exception to this lesson, 36. Lesson 3: Public authorities need to take a where the “pure” User-Pays model is still widely used, realistic look at the local road contracting indus- has more to do with, among other factors, the subsi- try before scaling their road PPP projects or pro- dized long-term financing that was widely provided grams. When structuring tender documents (par- by Banco Nacional de Desenvolvimento Econômico ticularly Request for Qualifications and Request for e Social (BNDES, the State-Owned National Develop- Proposals), contracting authorities will need to define ment Bank) until 2016. qualification criteria. These criteria will need to be balanced to enable a truly competitive environment 38. Lesson 5: Annuity Payments funded by a while weeding out bidders with weak balance-sheet clear ring-fenced mechanism, independent from performance or technical skills, making it impossible annual public budgeting cycle, can strongly for them to secure performance-based or concession enhance a project’s bankability. Some recent type contracts. In many SSA countries, there is no examples show Project Companies either derive proven track-record of road PPPs or other forms of their revenues from publicly funded Annuity Pay- performance-based contracts. This inevitably leads to ments only (e.g. roads annuity program in Kenya, looking to foreign contractors, at least in the short- Ganta-Zwedru road corridor rehabilitation in Liberia) term. If public authorities want to secure local con- or a mix of tolls revenues and Annuity Payments (4G tractors’ participation to make road PPPs more politi- program in Colombia). In all these cases a dedicated cally appealing/acceptable, they can proactively favor ring-fenced mechanism, independent from public local contractors through various means, including annual budgeting cycles, was established to fund by setting aside percentages of road works to be each Government’s Annuity Payments obligations. executed by local contractors under each contract or This approach gave extra comfort and visibility to directly providing technical and/or financial support potential bidders and played a central role in each to these indigenous contractors. project’s marketability and bankability. 37. Lesson 4: The User-Pays PPP model is not 39. Lesson 6: Tolls or other forms of dis- easily replicable when scaling up private par- tance-based charges can be successfully imple- ticipation in the road sector in SSA. The two SSA mented. They require strong political support, a User-Pays PPPs reviewed (Dakar Diamniadio Toll transparent tolling policy and visible improvements Highway – DTH - and Henri Konan Bedie Toll Bridge for road users. Four of the cases reviewed involved - HKBTB) required considerable time to prepare. tolling as a revenue stream for Project Companies. Reaching financial close required either substantial All raised user acceptance issues, triggering the need public upfront payments or the establishment of for public authorities, at some point of the project’s tailor-made demand-risk mitigation mechanisms. In cycle, to temporarily lower the tariffs or suspend both projects the private sector funded only a small tolling altogether. Users’ willingness to pay did rise, 29 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA however, as they experienced sustained improve- NIBs solution for the majority of SSA countries seems ments in road safety, travel time and comfort. This implausible because most SSA countries are non-in- experience underscores the need for this type of vestment grade, the opportunity to use this approach projects to adopt and widely communicate a trans- can be examined, acknowledging the associated ben- parent tolling policy to all stakeholders21. Such a pol- efits and challenges22. icy should, inter alia, govern toll tariffs (e.g. affordabil- ity) and their periodic adjustment (e.g. to reflect local 41. Lesson 8: Coordinated MDB support can be inflation). instrumental to the success of road PPP projects or programs. Except for the Roads Annuity Program 40. Lesson 7: National Infrastructure Banks (NIBs), in Kenya, all projects or programs reviewed benefit- like BNDES in Brazil or Financiera de Desarollo ted from some coordinated support from WBG enti- Nacional (FDN) in Colombia can play a key role in ties. This support was predicated upon WBG availing assisting in the implementation of road PPP proj- a large array of non-financial (e.g. technical assis- ects or programs. BNDES in Brazil or FDN in Colom- tance to conduct reforms, implement E&S issues sur- bia acted as the financial arm of Governments during rounding Right of Way issues, etc.) and financial (e.g. the roll out of each country’s road PPP Program. credit/loans, political risk insurances, and guarantees) These NIBs were either able to reduce projects’ cost instruments to each project. Similar MDB support will of capital, therefore increasing affordability for users be necessary in SSA countries where there is little, if and taxpayers, or helped crowding-in private invest- any, proven track-record of private participation in ment by simply participating in each project. While a the road sector. 4. Key lessons to better prepare and structure road PPPs and scale-up private sector investment 42. To enhance the appeal of road PPPs to debt 43. SSA Governments and contracting authorities and equity providers, right-of-way and resettle- should prepare moderate-sized road PPPs focus- ment issues should be settled before financial ing on brownfield projects. Carefully designed strat- close, and related risks should be retained by the egies covering pipeline building, standardization of public sector. In countries where there is no proven tender documents, bundling of several projects into a track-record of successful road PPPs, it is therefore single contract, etc. are needed to lower transaction recommended to initially focus on brownfield proj- costs. It should make them more attractive to both ects23 for which right-of-way, resettlement and engi- local and international contractors. neering issues are limited. 21 The case-study of the Polish Road Fund delivers the same lesson. Poland has introduced HGV charges on selected sections of the national network in 2011. Despite initial opposition, the trucking industry accepted the principle of road charging. The revenues from these charges are steadily growing. 22 Global Infrastructure Hub. Guidance note on National Infrastructure Banks and similar financing facilities (consultative draft). April 2019. 23 In SSA countries, almost ¾ of the active road PPPs are brownfield according World Bank’s PPI database. 30 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 44. SSA Governments and contracting authorities should retain demand risk when structuring road PPPs as Government-Pays. The funding mechanism of Annuity Payments needs to be carefully designed. Backstop payment guarantees from the Govern- ments or MDBs are worthy safety nets, yet a funding mechanism with predictable and sustainable cash- flows is key to attracting the private sector. 45. Annuity Payments should be linked to a cred- itworthy off-taker, a role 3rd Generation RF could play. In addition to traditional RF resources (e.g. fuel levy), distance-based user charges (e.g. tolls) collected on the roads after works completion could contribute to funding the public authority’s Annuity Payments obligations. RFs’ creditworthiness could be enhanced with Government and WBG support (e.g. payment obligation guarantees). 46. In the long term, other private sector partici- pation models like road asset recycling could be introduced in SSA countries. Road asset recycling requires transferring the demand risk to the private sector. The feasibility of private sector participation in road projects needs nevertheless to first be demon- strated through the implementation of successful Gov.-Pays road PPPs. 31 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA IV. Restoration Concept: a novel road PPP promotion instrument 32 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 1. Restoration Concept 47. The Restoration Concept combines the fol- whereas publicly financed contracts (i.e. OPRC lowing features: i) consolidating multiple road and performance-based contracts - PBCs) would improvement projects under a single contractual remain financed from the RF’s general fund. The umbrella, ii) increasing RF revenues by increas- RRW would be carved into the funds earmarked for ing traditional RUCs (namely the fuel levy), and c) national roads maintenance which represent about introducing tolls or other forms of distance-based 80% of RF resources in SSA. A notional percentage of charges on sections of restored road networks. about 20% of the funds allocated for the maintenance Restoration Contracts would bring the entire length of national roads could be transferred into the RRW of selected national Priority Alignments to good con- (or about 16% of RF revenues25). As a result, funding dition and improve and maintain them as necessary24. assigned to the maintenance of national roads would Tolls or other forms of distance-based charges could be reduced from 80% to 64% of RF revenues while be raised on suitable sections of the restored Priority the portion of its resources allocated to the mainte- Alignments when and if traffic volumes are sufficient nance of local roads would remain unchanged (see and a sound toll policy is in place. These users’ reve- Figure 3 below). nues would be earmarked to the RF and used to pay, partially or fully, for the cost of Restoration Contracts, 49. During the ramp-up phase of the Restoration including the associated future routine and periodic Concept, RFs’ existing financing Window (EW) maintenance contracts. Over time, any cash surpluses would gradually move away from funding short- would be used to fund additional Restoration Con- term input-based maintenance contracts to fund- tracts and, possibly, cross-subsidize the maintenance ing long-term PBCs. This change would require that: of other national roads and/or lower category roads. i) RFs’ statutes be amended to enable them to fund spot reconstruction works26, ii) training programs for 48. Privately financed Restoration Contracts local road contractors be carried out and, iii) procure- (implemented using the PPP model outlined ment legislation be amended as needed. previously) would be funded by a Road Resto- ration Window (RRW) carved into a qualified RF 24 Improve the service level to address safety and climate change requirements as well as increased traffic volume. These improvements should be designed with the intention of staying as close as possible to the maximum marginal benefit, i.e. avoiding overdesign. Improvement works can comprise localized structural strengthening; installation of guard rails, proper marking and signaling, pedestrian protection, bus lay-bys, separate parking and loading areas, removal of black spots; construction of overtaking lanes, crawling lanes on steep slopes, strengthening of the wearing course in hairpin turns; and increased drainage capacity, protection against flooding, reinforcement of slope stability, etc. 25 On average 80% of the RF resources are earmarked for the maintenance of national roads and about 10% to 40% of these resources are unused every year. The unused resources are generally carried over to the next year but in some countries, the carry-over is not authorized, and funds are lost to the RF. In this respect, sizing up the RRW to 20% of the funds earmarked for the maintenance of national roads is reasonable. Another point of comparison is the Kenya Roads Annuity Fund which represents about 16% of the RF resources. 26 As opposed to the entire reconstruction of long road segments funded under the government budget. 33 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Figure 3: Development of the Restoration Concept Source: Authors’ analysis. 34 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 50. Fuel levy as well as other user charges would 52. The Toll Policy adopted should aim at set- need to be raised and adjusted for to increase RFs’ ting affordable, consistent, transparent and fair resources during the ramp-up phase. On average, pricing of road usage. It should also regulate the fuel levy represents 76% of RF revenues in SSA. tariff revisions. Since tolls would be collected by the Raising the fuel levy to at least US$15c/l equivalent public sector, their pricing system would need to be would lift annual RF revenues significantly in many designed to be welfare-oriented, contrary to a pri- countries: they would reach US$600M in Ivory Coast vate concession scheme where it is profit-oriented27. and about US$300M in Namibia, Ghana and Cam- The Toll Policy should address, inter alia: the pricing eroon. Vehicle registration fees contribute to about formula that should consider the differentiated wear 10% to 20% of RF revenues but are not common in and tear caused by different types of road users (e.g. all SSA countries. Raising them or introducing them HGVs vs. light vehicles); the type of roads targeted for would also increase RFs’ revenues significantly. tolling, the method to adjust tariff rates to account for price escalation, congestion pricing, etc. Currently 51. During the cruising phase of the Restoration in SSA, HGVs rarely pay their fair share of road user Concept, EW would only fund long-term OPRCs charges, although some countries started experi- as local road contractors improved capabilities menting with HGV charges28. would enable them to successfully implement this type of contracts. Meanwhile, RRW could collect 53. For obvious economic reasons, a selection sys- additional revenues from tolls (or other forms of dis- tem based on road condition and demand must tance-based charges) derived from suitable restored be set to prioritize where restoration financing is Priority Alignments. As shown by the qualitative implemented in SSA. The road sections targeted for assessment of road funding instruments (see Section restoration should be part of the national paved Pri- III.1), road tolling represents a promising revenue ority Alignments and regional trade corridors where generation tool for RFs. The Brazilian, Colombian and higher traffic volumes occur. Most of these Priority Polish case studies show that initial resistance against Alignments are bituminous two-lane roads connect- new tolls can soften if road users associate tolls with ing larger urban centers. Their traffic volume varies the direct benefits they accrue from using better between 4,000 and 12,000 vehicles per day (vpd), while roads. While tolling systems should be administered their condition ranges between 20% to 60% in good by RFs/RRWs, it is recommended that their opera- condition, 30% to 50% in fair condition, and 10% to tions be outsourced. Toll revenues (net of collection 30% in poor condition. They often lack adequate safety costs) should be retained in full by the RRW and infrastructures, have not been adapted to climate potentially be tax-exempt. Over time, as the com- change, and their original design has not incorporated bined traditional toll revenues exceed the costs borne the need to accommodate increasing traffic volumes. by the RRW, excess revenues could overflow to the RFs must commission regular independent road con- EW and cross-subsidize the maintenance program of dition and traffic volume surveys to be knowledgeable local roads. about evolving road demand and conditions. This will allow them to target the right Priority Alignments. 27 Bonnafous, A. The economic regulation of French highways: just how private did they become? Transport Policy, 41. 2015. 28 Namibia for example recently started experimenting with the satellite-based tracking systems for HGVs, already in use in Europe; with the collaboration of ICT firms, this development could be within the reach of many African countries sooner than later. 35 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA 54. A pipeline of Restoration Contracts should be particularly when they have no prior PPPs expertise. prepared and then implemented at a pace com- Substantial pre-tender work (e.g. Priority Alignment mensurate with the level of RRW resources so selection, robust feasibility studies including traf- that the cumulated RRW’s cash-flow balance stays fic forecasts, standardization of tender documents, positive at all times. The financial simulation model appropriate bundling of Priority Alignments, etc.) shows that there is a significant difference between will be needed from RAs to reduce transaction costs, the number of restoration contracts that can be secure bidders’ interest and foster maximum com- funded solely with traditional resources versus tradi- petition among them. Private Sponsors will be more tional resources magnified by toll revenues. likely to participate in a bidding process if the tender- ing system enables scalability (enough investment to 55. Preparation, awarding and management of justify the sunk costs) and offers a prospect for repli- Restoration Contracts will require significant cability (bidding for similar projects). efforts from the RAs with support from a PPP Unit. The RAs should be tasked with the preparation 56. Accordingly, Restoration Contracts should first of the pipeline of Restoration Contracts. It should be piloted with RFs that present the potential to team up with a PPP Unit since these contracts would transition to a 3rd Generation status and gener- be implemented as Road Restoration PPPs. Pre- ate at least the equivalent of US$100m in annual paring and implementing PPPs can prove consid- revenues. erably time-consuming for contracting authorities, 2. Restoration Concept financial modelling impact 57. A customizable Excel financial model was equity and associated costs. The model is adjusted developed to test the impact and limitations of for a customizable inflation and denominated in the Restoration Concept for various plausible sce- USD. The depreciation of RF and toll revenues is narios. The user can test a large set of financial and incorporated and covered by customizable catch-up technical assumptions against a Restoration program mechanisms. and the follow-up maintenance contracts over a 30y period and by extension simulate the total length 58. The cost of Restoration Contracts has been of roads restored and fully maintained thereafter. assessed for scenarios associated with the typical Among the key assumptions are: i) the RF’s level of condition of national priority roads in SSA coun- resources; ii) whether tolls would be raised and under tries and other logical and realistic assumptions. what terms (e.g. tariffs, and their coverage of the Based on road conditions described previously, unit restored Priority Alignments); iii) the road condition; rehabilitation costs derived from known studies, and iv) the contract duration; and v) the mix of debt and financing assumptions drawn for recent PPP schemes 36 MOBILITY AND TRANSPORT CONNECTIVITY SERIES in SSA, ten scenarios were established for a nominal US$125M per year would be able to restore 10,000 100km stretch of roads29. Four of these scenarios km of Priority Alignments over 30 years. This were computed in the financial model30: two high- figure is to be compared with the mere 400 km that case and two low-case scenarios based on the vol- could be restored in the absence of tolls. It shows ume of works implied, respectively labeled A1, B1, C3 how the outcome of the Restoration Concept (total and C4. A1 and B1 restoration contracts were struc- length of restored and thereafter fully maintained tured as 15y Government-Pays PPPs and C3 and C4 Priority Alignments) increases linearly with traditional as 10y OPRCs owing to their lower estimated cost31. resources and exponentially if tolls are collected on The respective nominal non-discounted costs from some sections of the restored Priority Alignments. In the public sector’s point of view amounted to about the case of the Kenya, whose RF collected US$625M US$117M for A1, US$113M for B1, US$68M for C3 in 2017, an increase of the toll coverage from 20% to and US$65M for C4. 30% of the Priority Alignment network would induce a leap from 5,000 km to 10,000 km of restored and 59. Assuming no tolls are raised, a RF collecting maintained priority roads over 30 years. Despite the about US$125M equivalent in annual revenues32 challenges associated with any tolling program, these would only be able to fund 4 Road Restoration numbers illustrate the need for SSA Governments to PPPs over a period of 30 years (i.e. would be able seriously consider the tolling option. to restore and fully maintain 400 km of Priority Alignments). An RF collecting US$625M of annual 61. If SSA countries agree to raise the fuel levy to revenues would be able to fund 23 of these contracts the minimum recommended level of US$15c/l and or restore and fully maintain 2,300 km of Priority adjust it regularly to account for inflation, Ivory Alignments over the same period. RFs collecting Coast and Mozambique would join Kenya among annual amounts between these two limits would the RFs collecting more than US$500M annually be able to restore a proportional length of roads as while Namibia and Ghana would collect about shown by the linear relationship between RF reve- US$300M annually. Consequently, Mozambique, nues and km restored. Kenya, and Ivory Coast could restore 2,800, 2,300, and 2,200 km of Priority Alignments over 30 years 60. Assuming that (1) tolls can be raised on 50% without raising any toll. If tolls were raised using the of the length of all restored Priority Alignments, assumptions previously described, these countries (2) the average toll is equivalent to US$5c/km would be able to restore and maintain nearly all of and (3) the AADT is 8,000 vpd, then a RF collecting their national road networks. 29 Other notable assumptions: direct O&M costs increased by 15% to consider the Project Company’s own costs (staff, headquarters, taxes and insurance) in case of privately financed Restoration contract, and reserve accounts waived for government loan to the RRW. Moreover, the cost of installation and operation of the tolling system is included in the restoration contract. As a conservative estimation, it is assumed that the toll operator will be hired during the last year of the construction period. Another year is then allowed to train the work force and test the equipment. Hence, assuming three years for the construction period, tolling could become effective 5 years into a restoration contract on the entire or selected sections of the restored road. 30 The financial model built in an Excel spreadsheet is provided with the report. 31 The simulation model allows testing a restoration program composed of a mix of Government-Pays PPPs and OPRCs. Conversely, lower cost scenarios can be bundled and treated as PPPs. 32 In 2017, eight SSA countries had a RF collecting that amount or more. However, none of these RF had automatically adjusted resources. 37 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA 62. Figure 4 presents the output of the simulation average of 1 per year during the years 1-10, to 3 per model using the average revenues of US$115m year during the years 11-20 and then to 9 per year currently collected by SSA’s RFs. It also assumes the during years 21-2734. This snowball effect type of that tolls would be collected. If tolls were raised on growth would give RFs, RAs, and their associated PPP 50% of the restored Priority Alignments, a typical RF Units time to upgrade their skills and management would be able to implement over 30 years up to 100 capacity. This simulation also quantifies the magni- restoration contracts (52 PPPs and 48 OPRCs)33, 36 tude of the tolling effect: after 30 years, the cumu- follow-up maintenance contracts, and restore 10,000 lated traditional adjusted RF revenues would reach km of Priority Alignments. The number of Resto- about US$750M annually, whereas the cumulated toll ration Contracts would gradually ramp up from an revenues would top US$8bn. Figure 4: Cumulated Restored roads. Contract costs and resources. 33 Ideally, all contracts should be structured as PPPs. The simulation can be computed accordingly but we chose to use a combination of both that looks more realistic in our view. 34 The restoration program is suspended in year 27 because the restored network nearly covers the common length of the national networks 38 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 3. A Road Restoration PPP model to scale-up private sector participation in Sub-Saharan Africa a. Typical features of a Road Restoration PPP maintenance. The whole duration of the PPP would be no less than 15 years, including a 3-year construc- 63. The proposed model of Road Restoration PPP tion period, but could be longer (e.g. 20 to 25 years) addresses the identified impediments to more to optimize life-cycle management or increase afford- private sector participation and leverages les- ability (for example, if the project includes signifi- sons learned from the analyzed road PPPs. It is cant capacity increase or upgrading) among other intended to be adapted and replicated in various factors35. SSA countries. 66. A typical Road Restoration PPP would not 64. A Road Restoration PPP would be structured as necessarily be CAPEX-intensive. A Road Restoration a Government-Pays PPP with demand risk retained PPP would entail minimal upgrade works as it would by the public sector. Annuity Payments would flow target brownfield sections of the road network’s from the RRW carved into the RF. Milestone Payments Priority. Despite this relatively low risk profile, these could be made during the construction period cover- projects would still need to benefit from scalability ing a portion of CAPEX. The Project Company would and replicability to lower their transactions costs, finance the remaining CAPEX with a mix of debt and both for the RA/RF and the sought-after private spon- equity. The share of CAPEX covered by private finance sors. Additional strategies designed to make these and Milestone Payments would need to be adjusted to contracts more attractive to private sponsors could consider an RF’s RRW capacity to pay. entail bundling several Priority Alignments into a sin- gle Road Restoration PPP. 65. A Road Restoration PPP would focus on the reconstruction and improvement of exist- Table 2 provides a summary of Road Restoration ing Priority Alignments with a possible capac- Contracts’ main features. ity increase. It would include operations and 35 ADB, EBRD, IDB, IsDB, MIF, PPIAF and WBG. The APMG Public-Private Partnership certification guide. Chapter 5: Structuring and drafting the tender documents and contract. 2016. 39 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Table 2: Proposed typical features of a Road Restoration PPP. Feature Description Type of PPP project Brownfield DBFOMT linked to Annuity Payments. Demand risk retained by the public sector. Private Sector to take works completion risks and/or maintenance risks (i.e. Annuity Payments can be reduced if road condition falls below a certain threshold). Scope Rehabilitation/reconstruction of existing paved roads, including bridges, culverts, roads intersection, etc. Localized upgrading (from gravel to pavement) and capacity increase. Operation and maintenance, including periodic maintenance and renewal. Length Length to be determined, but should cover a significant portion of a Priority Alignment CAPEX To be determined regarding affordability for the public sector. Possibility to bundle several Priority Alignments under the same contract to make contract sizeable and if practical from a technical point of view. Milestone Payments To be determined regarding affordability for the public sector. Contract Duration Not less than 15 years and up to 20-25 years, including a construction period of about 3 years. Contracting Authority Road Authority. Project Company Revenues Annuity Payments (twice a year or quarterly) made by the Road Fund. Gov. Support As owner of both the RA and the RF, the Government would be expected to step-in in case of revenue shortfall regarding Annuity Payments or early termination payments. Potential WBG Support Entities of the WBG can provide a range of financial instruments (loans/credits and guarantees) to either the Project Company or the Government (and its agencies). Other Support Application for Global Infrastructure Facility funding to fund advisory services (either project definition or project preparation and structuration activities). Source: Authors’ analysis. b. Typical commercial structure of Milestone Payments during the construction period. a Road Restoration PPP These payments would be made from the RRW using a mix of revenue sources as described previously. 67. The proposed commercial structure is derived RRW’s capacity to pay for these PPPs could be further from a typical PPA with the RA acting as Contract- buttressed by the RF’s ability to contract long-term ing Authority while the RF is the designated “off- commercial loans from local and/or international taker” tasked with making annuity payments to banks (i.e. using its future revenue flows as collateral the private Project Company. The RF would shoul- as shown in Figure 5). der the payment risks in this structure, including 40 MOBILITY AND TRANSPORT CONNECTIVITY SERIES Figure 5: Proposed typical commercial structure of a Road Restoration PPP. Source: Authors’ analysis. 68. Alternative commercial structures could be 69. Other contracts could be specific to the pro- adopted without substantially changing the posed structure: envisioned PPP structure. These could be tailored to countries with different institutional frameworks – The funding agreement. This agreement would (e.g. no RA, or PPP Unit as the Contracting Authority), be signed between the Government, the RA and alternative toll operation schemes, or even alter- the RF. It would be part of the tender documents native funding mechanism for Milestone Payments to give visibility to bidders on how payments during the construction period. to the Project Company would be governed. Its 41 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA purpose would be to give enough confidence c. Key risks allocation for the proposed to private sponsors that RF’s payments would Road Restoration PPP carry acceptable credit risks. This contract would address, inter alia: the responsibilities for making 70. The key risks allocation proposed below is Milestones Payments for works, Annuity Pay- intended to guide the structuring of balanced and ments, Early Termination Payments and Payment bankable Road Restoration PPP program. Some Procedures including, but not limited to, Pay- of these risks should be mitigated during project ments’ approval process. preparation. For the risks that will be dealt with after financial closing, mitigation mechanisms would be – Toll agreement. Some of the Priority Alignments reflected in the project agreement. could be tolled once restored. Whether there will be only one or several toll operators depends on 71. Key risks include36,37: the Government’s decisions. The toll agreement would be signed between the toll operator and – Land availability (risk of delay and cost over- the RF. It would govern, inter alia: the technical runs in the acquisition of the land necessary specificities to ensure interoperability in case sev- to develop the project.). This risk is mitigated by eral toll operators co-exist on the tolled network, the fact that Road Restoration PPPs focus mainly performance objectives in terms of toll collection on the restoration of existing Priority Alignments. and organization of flows from the users to the However, Right-of-Way should ideally be fully RF. The toll system could be developed and man- cleared before financial close. Any remaining aged under various schemes (from traditional Right-of-Way (e.g. for service areas proposed public procurement to PPP schemes). by bidders) should be cleared within a specified timeline. In case of WBG support (e.g. IDA/IBRD – Interface agreement. Based on the assump- credit/loan), Right-of-Way clearance could be a tion that the tolling system would be developed condition precedent for drawdown; and managed by a third-party (i.e. not the Project Company itself). This interface agreement, would – Environmental and social risks (i.e. delays and/ govern, inter alia: technical issues like access to the or cost overruns in obtaining environmental site to install the system and/or to operate/man- clearance and conducting resettlement/com- age the system, mutual liquidated damages, etc. pensation of impacted population). The Reset- tlement Action Plan defined and approved by the Government as part of the full ESIA (Environmental and Social Impact Assessment) study should be implemented before commercial close by the Con- tracting Authority. However, the Project Company would be responsible to implement the agreed environmental action plan during the entire con- tract duration; 36 https://ppp-risk.gihub.org/risk_category/road/ 37 PPP in Infrastructure Resource Center for Contracts, Laws and Regulations (PPPIRC). Matrix of risk distribution for roads. March 200 42 MOBILITY AND TRANSPORT CONNECTIVITY SERIES – Demand and “off-taker” risks (i.e. when Liquidated Damages that will be deducted from its resources collected from project users are Annuity Payments. RA/RF will seek to limit this risk below forecasted revenues and/or RF is not since a well-maintained roadway will increase and able to meet its payment obligations). Annuity preserve users’ willingness to pay a toll whenever Payments will be made by the RF irrespective of the road is tolled. The selection of adequate and whether the forecasted revenues from the tolled reasonable Key Road Performance Indicator tar- sections are met or not. The funds available in the gets will be the basis of a balanced risk allocation RRW will be mobilized to that aim. To help bidders between the Project Company and the RA/RF; assess the off-taker risk, the funding agreement, as well as the last financial and annual reports of – Foreign exchange risk (i.e. currency mismatch the RF should be part of the tender documents. between revenues and debt/equity). Depending Since the concept of carving out an RRW will be on country context, there are different options to a novelty, a payment guarantee from the Gov- mitigate this risk. The Contracting Authority may ernment, possibly counter-guaranteed by IBRD/ wish to propose in the tender documents that IDA, could be included in the tender documents; a percentage of Annuity Payments (capped and at least until the RF has gained sufficient market to be proposed by bidders) be denominated in a credibility to free itself from this requirement; hard currency. If that percentage does not fully cover debt service and equity distributions, then – Maintenance and operation costs (i.e. project the Project Company can investigate securing maintenance - routine and periodic - and oper- financial hedging products (i.e. foreign exchange ation costs are higher than expected). As this swaps) for the unsecured portion of its revenues/ refers to life-cycle management of the road, this payment obligations; risk should be transferred when possible to the Project Company. Bidders will make their own traf- – Currency convertibility and transferability (i.e. fic studies that will be reviewed by their lenders to inability to legally convert local currency into determine the cost of maintenance (particularly hard currency and/or transfer converted cur- the share of HGVs). To face the periodic mainte- rency - cross-border investments only). Depend- nance costs, Lenders will require the Project Co. ing on country context, Lenders and Equity Inves- to use a portion of the Annuity Payments to flow tors in the Project Company may wish to contract into a specific reserve account. Since axle load a Political Risk Insurance (PRI) that can provide regulation enforcement in SSA often represents coverage against this risk. an unmitigated risk; specific contractual clauses might be inserted in the PPP documentation to 72. The so-called “acceptability” clauses are address this risk; related to the public sector’s rights and aim at increasing political acceptance of the PPP – Availability and performance risk (i.e. road scheme. These clauses include, inter alia: refinancing condition becoming sub-par). From the Proj- gain sharing mechanisms, social inclusion targets, ect Company’s point of view, this risk can mate- control and auditing of the contracting authority over rialize through Performance Deductions and/or the Project Company‘s activities. 43 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA V. SSATP and World Bank Group assistance to implement the Restoration Concept 44 MOBILITY AND TRANSPORT CONNECTIVITY SERIES 1. A checklist of activities to implement the Restoration Concept 73. The following recommendations aim at guid- the Project Company can mitigate the off-taker and ing the implementation of the Restoration Con- early termination payment risks while helping attract cept in SSA countries where RF exhibits a poten- Lenders and Equity Investors. MIGA Political Risk tial for 3rd Generation status. Since many RFs in SSA Insurance products can insure cross-borders inves- have slid back from 2nd Generation Status or have tors (either lenders or equity providers) against some never reached it, this study recommends technical political risks (e.g. currency convertibility) and thus assistance to invert this trend in such cases. help mobilize private finance. 74. Depending on country context, the implemen- 76. The timing and mobilization of WBG financial tation of the Restoration Concept is expected to and non-financial instruments to implement the require a multidimensional WBG support to client Restoration Concept and Road Restoration PPPs countries. This support would range from increasing should be optimized and coordinated to ensure knowledge (on national network condition and traffic; client countries buy-in. These instruments are either on local contractors and financiers’ skills and capac- tailored to the Public Sector or future Project Com- ities, etc.) to Road Restoration PPP programming, panies (as well as its Lenders and Equity Providers). structuring, tendering and further management as They are likely to be mobilized at different phases of well as passing potential reforms (national procure- the Concept implementation timeline. For example, ment framework, RF’s mandate and/or legal nature, IBRD/IDA Technical Assistance would likely be needed tolling policy, PPP framework, restructuring of an during upstream phases to help client countries existing National Infrastructure Bank, etc.). assessing Priority Alignments or identifying gaps in their institutional/legal frameworks that may prevent 75. Regarding Road Restoration PPPs only, enti- or hamper the implementation of Road Restoration ties of the WBG can provide a range of financial PPPs. IDA/IBRD credits/loans and guarantees would (loans/credits, guarantees and insurance prod- likely be discussed with client countries during the ucts) and non-financial (technical assistance, advi- Project’s appraisal phase so that guarantees can be sory services) aiming at increasing affordability proposed as part of the tender documents40. Finally, and attractiveness of Road Restoration PPPs38,39 during the tendering phase, bidders may request to either the Contracting Authority or the Proj- MIGA Political Insurance Products and IFC loans. ect Company. IBRD/IDA loans/credits provided at favorable or concessional terms to the Government 77. Designing WBG support to client countries for can co-finance Milestone Payments, thus lowering implementing the Restoration Concept, whether the cost of capital and increasing affordability for the a replicable and standardized turnkey solution41 public sector. IFC loans to the Project Company can or a flexible one-stop-shop window42, is beyond help mobilize the private financing needed, increase the scope of this study. Nonetheless, Table 3 pro- affordability through longer tenors and favorably vides a checklist of activities that could use the impact the PPP structure. IBRD/IDA guarantees to support of the WBG. 38 World Bank Group. Maximizing Finance for Development in Transport. Getting from concept to investments. Report 2: Operational Guidance. 2019. 39 World Bank Group. World Bank Group Guarantee Products, Guidance note. April 2016. 40 As was the case for the Ganta-Zwedru road corridor rehabilitation PPP in Liberia. 41 See for example the scaling solar initiative (https://www.scalingsolar.org/). 42 See for example InterAmerican Development Bank’s PPP framework (https://blogs.iadb.org/bidinvest/en/support-structuring-public-private-partnerships/). 45 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Table 3: Checklist of activities to perform to implement the Restoration Concept. Surveys and data analyses Upstream phases of implementation. – Independent survey and assessment of the national road network to (1) improve knowledge of road condition and traffic levels and (2) select Priority Alignments to be restored and, further on, potentially tolled; – Appraisal of the road contracting industry to assess local contractors’ ability to participate in PBCs, OPRCs and Road Restoration PPPs; – Appraisal of the financial industry to identify (1) local commercial banks and local institutional investors (e.g. pen- sion funds) and assess their capacity to provide long term-financing and (2) any existing National Infrastructure Bank and assess its capacity to address long-term financing gaps; – Legal and institutional gap analysis to identify (1) areas for improvement in the road asset management ecosys- tem, (2) necessary improvements to the RF legal mandate, structure and legal status (e.g. moving from an admin- istration to an SOE), (3) areas for improvement in the PPP framework, and (4) areas for improvement in the public procurement framework. Institutional and legal reforms as appropriate Upstream phases following the outcomes of data analyses. – Road Fund reform e.g. changing RF legal status, creating an RRW, enabling the funding of spot reconstruction bun- dled with maintenance in long-term contracts; – Financial sector reform e.g. enabling local institutional investors to provide long-term financing for public infra- structure projects, reforming/restructuring an existing National Infrastructure Bank; – Toll policy elaboration; – PPP framework adjustment as required; Capacity building Upstream and midstream phases of implementation. – Road contractors: Advertise and explain the Restoration Concept, organize classroom and on-the-job training pro- grams on PBCs, OPRCs and Road Restoration PPPs to improve their capacity to qualify and bid successfully; – Financiers: Advertise and explain the Restoration Concept, improve their project finance skills and capacity to pro- vide long-term financing; – Public sector institutions: Assistance to RFs (e.g. management of resources, disbursement processes, reporting and auditing) and RAs (e.g. additional skills and staff needed to manage the PPP contracts); 46 MOBILITY AND TRANSPORT CONNECTIVITY SERIES Road Restoration PPP programming and implementation Midstream phases of implementation. – Fundraising for advisory services: prepare and manage application for Global Infrastructure Facility funding43. Funding may support programming as well as transaction implementation activities and is compatible with advi- sory mandates executed by either an independent advisor or by IFC; – Preparation of a pipeline of Priority Alignments to be restored using Road Restoration PPPs based on surveys and data analyses; – Drafting of standardized tender documents (Request for Qualifications, Request for Proposal, project agreement, direct agreement, funding agreement, toll agreement and interface agreement); – Market sounding in order to (1) collect private sector stakeholders’ feedback on standardized tender documents and improve them as appropriate, and (2) identify market gaps requiring WBG financial instruments (e.g. payment guarantees, Political Risk Insurances); – Tendering the Road Restoration PPPs. If the market sounding reveals that WBG financial instruments will be neces- sary to support bankability, these instruments should be made available as early as possible in the tender process. 2. Next steps 78. Activities that could be undertaken promptly by – Prepare pitchbooks and business cases to can- the WBG are proposed below: vass potential pilot countries (e.g. Ivory Coast and Kenya); and – Collect private stakeholders’ feedback on the pro- posed Road Restoration PPP model; – Organize a workshop during the next SSATP annual meeting to discuss and elicit feedback on – Initiate the design of WBG support to be pre- the Restoration Concept. sented to client countries; 43 https://www.globalinfrafacility.org/ 47 SCALING UP PRIVATE SECTOR PARTICIPATION IN ROAD ASSET MANAGEMENT IN SUB-SAHARAN AFRICA Photo Credits Cover Page: Trevor Samson, World Bank Page 4: Romel Simon. World Bank Page 6: Simone Mc Courtie. World Bank Page 12: James Martone. World Bank Page 13: James Martone. World Bank Page 14: Romel Simon. World Bank Page 21: Trevor Samson, World Bank Page 30: Wenxin Qiao. World Bank Page 31: Trevor Samson, World Bank Page 43: Curt Carnemark. World Bank 48 MOBILITY AND TRANSPORT CONNECTIVITY SERIES