ANGOLA ROAD SECTOR PUBLIC EXPENDITURE REVIEW Mustapha Benmaamar, Fatima Arroyo Arroyo and Nelson Tisso Eduardo June 2020 CONTENTS Executive Summary....................................................................................................................................... 3 1 Introduction .......................................................................................................................................... 8 2 Angola road network ............................................................................................................................ 9 2.1 Road density.................................................................................................................................. 9 2.2 Road density by province .............................................................................................................. 9 2.3 Road characteristics .................................................................................................................... 11 2.4 Road conditions .......................................................................................................................... 12 2.5 Traffic volumes and composition ................................................................................................ 12 2.6 Demand for road services ........................................................................................................... 12 2.7 Road sector Institutional set up .................................................................................................. 13 3 Public expenditure .............................................................................................................................. 15 3.1 Public expenditure on infrastructure .......................................................................................... 15 3.2 Public expenditure on the transport sector................................................................................ 16 3.3 Public expenditure on the road sector ....................................................................................... 16 3.4 Public expenditure at subnational level...................................................................................... 18 4 How efficient is public spending in the road sector? .......................................................................... 21 4.1 National roads ............................................................................................................................. 21 4.2 Unclassified/Municipal roads...................................................................................................... 22 5 Road sector financing needs ............................................................................................................... 24 5.1 Road sector financing needs assessment approach. .................................................................. 24 5.2 Input Data Assumptions.............................................................................................................. 24 5.3 Estimation of road rehabilitation and maintenance needs ........................................................ 28 6 Recommendations .............................................................................................................................. 31 7 Appendixes.......................................................................................................................................... 34 7.1 Appendix 1: Road accessibility analysis to markets, education and health................................ 34 7.2 Appendix 2: Angola RONET Evaluation ....................................................................................... 39 1 LIST OF FIGURES FIGURE 1: ROAD DENSITY AND INCOME PER CAPITA 9 FIGURE 2: SPATIAL ROAD DENSITY BY PROVINCE 10 FIGURE 3:TRAVEL TIME TO REACH THE CLOSEST HOSPITAL AND SCHOOL (MIN) 10 FIGURE 4: ANGOLA ROAD NETWORK 11 FIGURE 5: ANGOLA VEHICLE FLEET 13 FIGURE 6: INFRASTRUCTURE PUBLIC EXPENDITURE AS % OF GDP (2008-2018) 15 FIGURE 7: PUBLIC EXPENDITURE IN THE TRANSPORT AND THE ROAD SECTORS AS % OF GDP (2008-2018) 16 FIGURE 8: PUBLIC EXPENDITURE IN THE TRANSPORT SECTOR BY MODE OF TRANSPORT 16 FIGURE 9: ROAD SECTOR EXPENDITURE AS % OF GDP (2008-2018) 17 FIGURE 10: PUBLIC EXPENDITURE ON ROAD MAINTENANCE (2008-2018) 18 FIGURE 11: SHARE (%) OF ROAD MAINTENANCE IN TOTAL ROAD INFRASTRUCTURE EXPENDITURE (2008-2018) 18 FIGURE 12: PUBLIC EXPENDITURE TRENDS BY LEVEL OF GOVERNMENT IN KZ (% GDP), 2008–2016 19 FIGURE 13: EXPENDITURE TRENDS BY LEVEL OF GOVERNMENT, 2008-2016 (% TOTAL SPENDING) 19 FIGURE 14: SHARE OF ROAD SECTOR EXPENDITURE AT CENTRAL AND SUB-NATIONAL LEVELS (2008-2018) 20 FIGURE 15: ROAD SECTOR EXPENDITURE AT CENTRAL AND SUB-NATIONAL LEVELS (IN BILLION KZ), (2008-2018) 20 FIGURE 16: SPENDING ON SUB-NATIONAL ROADS (IN BILLION KZ), (2008-2018) 22 FIGURE 17: QUALITY OF ROAD INFRASTRUCTURE 23 FIGURE 18: LPI SCORE IN MAJOR ECONOMIES IN SSA 23 LIST OF TABLES TABLE 1: ROAD AGENCIES* STAFF EFFICIENCY 13 TABLE 2: INFRASTRUCTURE INVESTMENT AS % OF GDP, BY SCENARIO AND REGION, 2011 15 TABLE 3: LENGTH OF NATIONAL ROADS PAVED AND REHABILITATED AND EXPENDITURE (2008-2017) 21 TABLE 4: CENTRAL GOVERNMENT LEVEL SPENDING ON ROADS AND POTENTIAL PHYSICAL ROAD OUTPUTS 22 TABLE 5: SUB-NATIONAL GOVERNMENT LEVEL SPENDING ON ROADS AND POTENTIAL PHYSICAL ROAD OUTPUTS 23 TABLE 6: NETWORK LENGTH BY TYPE AND ROAD CONDITION (KM) 25 TABLE 7 :TRAFFIC COMPOSITION AND VOLUME BY ROAD NETWORK TYPE 25 TABLE 8: UNIT COSTS FOR ROAD WORKS IN AFRICA (US$ THOUSAND PER TWO-LANE KM) 26 TABLE 9: AVERAGE UNIT COSTS FOR ROAD WORKS IN SUB-SAHARAN AFRICA (1,000US$ PER TWO-LANE KM) 26 TABLE 10: TWO-LANE UNIT COSTS (US$/KM) OF ROAD WORKS BY SURFACE TYPE AND CONDITION 27 TABLE 11: TWO-LANE UNIT COSTS OF ROAD WORKS BY SURFACE TYPE AND CONDITION 28 TABLE 12: COMPARISON OF ACTUAL EXPENDITURES AND FUNDING REQUIREMENTS (IN US$ MILLION/YEAR) 29 TABLE 13: ANNUALIZED COSTS YEARS 1 TO 20 30 TABLE 14: ROAD CONDITION PER TRANSPORT COSTS SCENARIO 30 2 EXECUTIVE SUMMARY Angola has made an unprecedented effort to salvage its infrastructure that was devasted during the 27-year civil war (1975-2002). By the end of the civil war in 2002, the Government of Angola has launched a large infrastructure reconstruction program by allocating on average US$4.4 billion per year between 2002-2009. This effort was sustained over the last decade (2008 -2018) and Angola spent on average an additional 4.1% of its GDP per year. Angola’s infrastructure public expenditure is on a par with the average expenditure of Lower and Medium Income Countries (4.0% of GDP for 2011) and is nearly twice more than Sub-Saharan African countries (2.5% of GDP). Most of the public expenditure on infrastructure was in the transport sector of which more than two- thirds was spent on roads. Angola has made significant effort to rebuild its dilapidated road network during the first post-civil war years and has spent on average US$2.8 billion/year in road re-opening programs (2002-2009). This effort was pursued over the 2008-2018 period and around US$2.1 billion/year or 2.1% of Angola’s GDP was allocated to the road sector. The quasi-totality (97%) of the public expenditure in the road sector was allocated to capital expenditure, consisting mainly of rehabilitation and repaving works. The share of road maintenance expenditure was, on average, only 3.5% of the total expenditure on roads, or $US28.0 million/year. Had the resources been spent efficiently, Angola could have built three times more kilometers of national roads and doubled the length of its current municipal road network. Real increases in spending have not been accompanied by commensurate increases in physical road output and quality. During the period (2008-2017), central government has spent a total of around US$20.64 billion on the national road network (fundamental and complementary roads), or around US$2.52 million/km. This is a very high unit cost compared to the unit cost to build a new two-lane road which, on the high range, is estimated to be around 0.8-1.0 million/km. Considering the total amount of road sector expenditure during this same period, Angola could have added a total of around 25,795 km of new roads to its national network, three times more than the actual road length of 8200km reported by INAE and MINCON. During the 2008-2017 period, around 17.5 billion Kwanzas/year, on average, were spent on provincial/tertiary roads (or an equivalent of US$171 million/year). However, the sub-national road network is still in a critical condition and shrinking in size. Angola could have added a total of around 34,000 km of new sub-national roads or doubled the current size of its municipal road network. The shift from road development to road maintenance has not occurred during the last decade. The quasi-totality (97%) of the public expenditure on the road sector was allocated to capital expenditure during the 2008-2018 period. During this period, the share of road maintenance expenditure was on average 3.5% of the total expenditure on roads, or $US28.0 million on average per year over the same period (Figure 10 and 11). There was a sharp increase in the share of road maintenance expenditure (25%) in 2015 which could be explained by the creation of the road fund (2015) but hardly any road maintenance expenditure was undertaken during the last three years (2015-2018). Oil revenues have taken a sharp decline during this same period and road maintenance has subsequently not become a priority. The impact of recent investment levels is minor and can be seen in the poor-quality of infrastructure in Angola. Around two-thirds (64%) of the total rod network is in poor or critical condition. Angola’s quality of road infrastructure is ranked 136th out of 141 countries, based on the 2019 Global Competitiveness 3 Report. Angola’s score is 2.2 out of 7.0 and is one of the lowest in the Africa region. Angola is lagging behind countries with the same GDP/capita along with many countries with lower GDP/cap (i.e. Ghana, Senegal, Cote d’Ivoire). Angola ranks 159 out of 160 countries in the 2018 Logistics Performance Index (LPI) and 153 out of 160 in the Infrastructure Performance as part of the LPI. This low quality of infrastructure is one factor constraining further improvements in Angola’s competitiveness ranking. The current road network asset value is around US$11.250 billion or around 11.0% of Angola’s GDP and every US dollar spent on road maintenance will generate US$3.4 in road user cost saving. The current asset value as a share of maximum road asset value is 74.5%. This indicates that more than 25% of the road network asset value is lost due to lack of maintenance or deferred road rehabilitation work. Over the next five years, a total of US$ 924.8 million per year is needed for rehabilitation (US$613.2m), periodic maintenance (US$241.2m) and routine maintenance (US$70.4m) of the total network, of which US$771.4 m or 83.4% should go to the national roads (Fundamental and complementary roads) and the remaining US$153.4m or 16.6% to the sub-national roads (unclassified/municipal roads). In years six to twenty, once this initial rehabilitation backlog program is completed, US$320.4 million per year will be needed for preservation works of the entire network of which around 90% should be spent on routine and periodic maintenance work. The annual net present value of road agency costs over the 20-year evaluation period, at 6% discount rate, is US$660.8 million/year. This also indicates that every US dollar spent on road maintenance will generate US$3.4 in road user cost saving. Angola’s current average annual road sector expenditure (US$2,231 billion) is nearly two and a half times (2.4 time) more than road rehabilitation and maintenance requirements (US$0.925 billion). The estimated annual cost to properly rehabilitate and maintain the national roads (US$771.4 million) is 2.67 less than the current average annual expenditure (US$2.06 billion) on national roads (2008-2017). As for the sub-national roads, an estimate of 153.4 million per year is needed to rehabilitate and maintain the municipal road network (33000km). This is less than the current average of US$171.0 million spent each year through sub-national government on municipal roads. The average annual expenditure exceeds by 11% the road rehabilitation and maintenance requirements. There is a need to invest in rural road development programs. Half of the population is located further than 2km from any road. Access to social services (health centers and schools) is poor overall, only 37% of the total population can access hospitals and schools in less than two hours. Access of agriculture production to main markets is limited, almost three quarter of the total agriculture production value cannot reach markets, although this varies across provinces. This indicates that there is a need to invest in road development programs to fill the rural road accessibility gap in the eastern, northern and southern provinces of the country. There is a need to improve the governance and accountability of the road sector. The level of spending on national roads (fundamental and complementary roads) is nearly two and a half times (2.4 time) more than the estimated road rehabilitation and maintenance requirements. This increase in spending has not been accompanied by commensurate increases in physical road output and condition. This has resulted in much higher average costs for both preservation and development works. This cannot be explained only by a lack of technical and management capacity but is clearly due to poor road sector governance. The current institutional set up of the road sector allows for the separation of policy, management and financing (Road Fund, 2015) functions however this does not seem to be implemented effectively. Addressing these issues will be necessary to gain efficiencies and cost reductions that can be 4 passed on to road users. There is a need to introduce more efficient and business-oriented road management practices and to improve financial and managerial accountability. A set of measures could be taken including (i) improving the road fund effectiveness to fully play its role of mobilizing adequate revenues to cover road maintenance needs and to make sure the resources are used in efficient way. (ii) Consider the creation of road management agencies at provincial level to increase sub-national public expenditure effectiveness and efficiency and to address the low technical and managerial capacity issue. (iii) Develop and sustain a road data base and asset management system and review the road network classification to assert the ownership of the complementary roads (the missing middle) and the unclassified roads (Municipal roads); and (iv) support the development of a competitive road contractors’ industry. Chapter 6 discusses the recommendations of this public expenditure review and a summary is presented in the table below. Rationale/challenge Remedial measure/action Responsible Institutions Road Financing: -Issue implementing regulations enabling the Road -Current spending is general Fund to fully play its role of mobilizing sustainable RF budget based, erratic and half road maintenance financing MINCON of the road maintenance MINFIN needs -Road Fund to validate road maintenance programs submitted by INAE on technical and - Road maintenance programs economic sound basis; are not developed on technical and economic sound -Road Fund to carry out independent financial and basis; technical audits of road maintenance work programs - Quality of road maintenance work is below standard -Issue implementing regulations to mobilize and transfer road user charge-based revenue to the - Current Road Fund is Road Fund (fuel levy, vehicle license fee, transit established on strong legal fee…); basis and its revenue based on road user charges but there -The procurement of road maintenance work are no clear implementing function is no longer within the RF mandate – RF regulations; should operate based on a lean management organization consisting mainly of technical staff. Current Road Fund staff size is higher than the recommended average number of twelve staff. Road Management: -Create a road data base entity within the INAE INAE -Description of the road road planning/maintenance division and allocate MINCON network is scarce and adequate budget to carry out regular road unreliable inventory, condition and traffic surveys 5 -INAE has no website, has no -Develop a road database website and Geographic road inventory database and Information System does not carry out road condition and traffic surveys; -Hire the services of a consultant to carry out the above and develop a road asset management -No budget allocations to system. cover the cost of road inventory and traffic surveys; -No technical capacity to develop and maintain a road asset management system. Road classification: MINCON -The functions and status of -Review the current road network classification road networks in Angola are regulated by Decree 21/92 -Revise the current road classification based on and 46/92 issued in 1992 the new road database to assert the ownership of the complementary roads (the missing middle) -The current classification along with the unclassified roads. does not adequately reflect the country’s administrative and organization structure (decentralization levels) -The main focus is placed on the fundamental road network and little attention is given to the complementary roads and to the municipal road network Sub-national roads: MINCON Provinces -The level of spending on road -Develop a municipality road development maintenance and program based on accessibility gap analysis rehabilitation is slightly higher than the estimated needs, but the municipal road network -Establish a road management agency at provincial remains in critical condition level on a pilot basis -Almost three fourths of the total agriculture production value cannot reach markets -Half of the population is located further than 2km from 6 a road. Only 37 % can access hospitals in less than 2 hours -There is no coordinated approach to complete the rehabilitation backlog -There is a lack of management and technical capacity at subnational level. Road contractor’s industry MINCON -Review the current unit costs of road works -The unit cost of Angola’s road works is higher than SSA -Carry out a study of the road construction average industry to support the development of a competitive local road contractors’ industry -There is a need to have a better understanding of the -Use of competitive bidding methods road construction industry -Develop and mainstream the use of e- procurement. Currently only used in the heath - 54% of all public sector procurement activities used “Simplified Procurement” or sole source method 7 1 INTRODUCTION Considering its regional economic position, natural endowment, and its socio-economic development aspirations, it is becoming imperative to Angola to develop a safe, clean and efficient transport sector. The bulk of transport needs are provided by the road transport sub-sector, with rail playing a less important role. Air and sea transport provide the principle modes of international passenger and freight movement respectively. Angola has a relatively large road network (around 76,000km). However, only 18 percent is paved, and a good share of the secondary and tertiary road network is impassable during rainy seasons. When passable, average speed is less than 20km/h, rendering the commercialization of agriculture products a challenging task1. It takes 49–531 days to export or import goods, among the longest times in Africa. Road access is particularly problematic for firms outside Luanda. Regional road corridors are underdeveloped, and the poor road condition increases transport costs and hampers Angola’s value proposition as the regional transport gateway for Southern DRC and land locked Zambia and Botswana. Angola’s 27-year civil war (1975-2002) devastated its infrastructure. For instance, most of the road network was in a very bad condition and its size shrank as a result of years of underinvestment and neglect during a civil war that ended in 2002. Poor road infrastructure and transport logistics slow Angola’s overall economic development. The Government has recognized that the Angola’s economy needs the support of a well-integrated and efficient transport sector and made significant efforts to reconstruct its dilapidated road infrastructure and it has established a road maintenance fund (2015). This study has made good use of the Angola’s Ministry of Finance BOOST data base to review and analyse the volume and structure of public spending in the road sector and identify any trends. The Road Network Evaluation Tools Model (RONET), developed by the World Bank, is used to evaluate the preservation (maintenance and rehabilitation) requirements of the Angola road network. A geospatial analysis is also carried out to measure the accessibility by road to agriculture markets, health centers and schools. This road sector Public Expenditure Review (PER) examines the size and composition, allocation, and implementation of public expenditure on roads and assesses the efficiency and effectiveness of these expenditures both at the national and sub-national levels. It also reviews the overall policy and institutional framework and makes relevant recommendations. The objectives of this road sector PER are fully aligned with the objectives of the World Bank’s Country Partnership Strategy for Angola, which places a strong emphasis on supporting Angolan institutions. By working in partnership with government institutions, this review aims to assist in building capacity to analyze the efficiency of spending on roads, and to allocate funds in a way that supports effectively Angola’s development. 1INCATEMA (2018) Diagnostic of Supporting Infrastructure Current Status in the Project area. Commercial Agriculture Development Project. 8 2 ANGOLA ROAD NETWORK 2.1 ROAD DENSITY Angola’s road density is relatively low, compared to countries with the same level of GDP/capita. Angola’s road density is lower than many Sub-Saharan Africa countries with lower GDP/capita (Figure 1). Angola has a road network of around 76,000 km or a spatial density of 6 km per 100 km2. This density varies across regions as the main road links are in the western half of the country along the coast which are in reasonable condition, while roads further inland are scarce and more dilapidated. Figure 1: Road Density and Income per Capita Source: Infrastructure Development in Sub-Saharan Africa. A Scorecard. The World Bank 2.2 ROAD DENSITY BY PROVINCE The average spatial density of the classified roads is low (3.3 km per 100 km2), compared to many countries in Sub-Saharan Africa and varies across provinces (Figure 2). The spatial density is above average in the provinces along the coast (i.e. Cabinda) and below average in provinces in the central and eastern part of the country. Data on the distribution of the unclassified roads (32,345 km) by province is not available and it is not clear whether this classified road density gap is filled by the unclassified/tertiary roads which are important to provide access to basic social services. 9 Figure 2: Spatial road density by province 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 Source: INAE, 2018 Half of the population is located further than 2km from any road. A geospatial analysis to measure the accessibility by road to agriculture markets, health centers and schools was carried out using data from open street mapWorldPop 2015 estimates, points of interest, and the International Food Policy Research Institute’s (IFPRI) SPAM model which identifies the value of crop production The results show that access of agriculture production to main markets is poor overall, almost three quarter of the total agriculture production value cannot reach markets, and this varies across provinces. Access to social services (health centers and schools) is also poor, around half of the country’s population is located further than 2km from any road and only 37% can access hospitals and schools in less than 2 hours. This travel time is much longer in provinces in the east, north and south of the country (Figure 3). This could indicate that there is a need to build new municipal roads and develop rural development programs. Detailed results of the geospatial analysis are presented in Appendix 1. Figure 3:Travel time to reach the closest hospital and school (min) 10 2.3 ROAD CHARACTERISTICS Nearly half of the total road network is unclassified roads and there is no data on their characteristics and condition. The total size of the Angola road network is around 76,000 km. The classified network is around 43,655 km in length or 58% of the total road network. This classified road network includes around 26,000 km of fundamental roads which connect the capital to the 18 provinces and the main cities to each other and to around 17,500 km of complementary roads. The fundamental road network is under the responsibility of INAE, of which around 13,600km, or 52% are paved roads. The remaining classified road network (17,500 km) are complementary roads and are also under the responsibility of INAE of which only 210 km are paved. Most of the classified roads are 6 meters wide but there are also a few road sections which are 4.5 meters wide. The remaining road network consists of around 32,345 km of unclassified local roads which provide connectivity within the 164 municipalities areas (Figure 4) and are under the responsibility of the provinces. Data on the characteristics and on the conditions of this unclassified road network is not available. Figure 4: Angola road network Paved Unpaved 35,000 30,000 32,345 25,000 20,000 15,000 17,445 10,000 13,600 12,400 5,000 210 0 - Fundamental Complementary Municipal/Unclassified Source: INAE, 2018 Part of the fundamental road network forms the Angolan sections of the Sothern African Development Community, SADC corridors. These corridors represent a major asset in the road network. SADC identifies road corridors that are of regional importance in facilitating trade and movement between member states. These are to be developed to a common standard to support the future potential traffic demands. There are five SADC corridors within Angola: i. Corridor 1: North – South; ii. Corridor 2: Luanda - Soyo - Cabinda (Angola, Congo, and DRC); iii. Corridor 3: Lobito (Angola, DRC and Zambia); iv. Corridor 4: Malange (Angola, DRC); v. Corridor 5: Namibe (Angola, Namibia, Botswana, Zambia). The corridors total length is 6216 km, of which 4,628 km are paved (representing 74% of the total length). Works have already been undertaken to upgrade these corridors. The Angola National Road Institute, INEA states that upgrading involves widening and paving of the roads to bring to a SADC standard. 11 2.4 ROAD CONDITIONS Angola does not carry out regular surveys about road inventory and condition and information on road conditions are either sketchy or not available. Traffic surveys are not conducted and there is no road asset management database. The National Transport Strategy and Master Plan (NTSMP) report (2018) prepared by AfDB shows that the fundamental road network is on average 45% in critical or bad condition, 18% in reasonable or good condition, and the rest is unknown. There is no available data on the conditions of the provincial and the unclassified road network. 2.5 TRAFFIC VOLUMES AND COMPOSITION There are no national roads with traffic volumes above a threshold to sustain financial viability and attract private sector financing. For the preparation of the NTSMP, traffic surveys were conducted on 31 surveys stations on the fundamental road network in 2018. The results show that the average daily traffic is around 900 vehicles. The highest traffic volume was observed on the Dande – Ambriz road section, with a total two- of almost 4,000 vehicles, the second highest traffic volume is around 2500 vehicles. The lowest traffic volumes observed are below 100 vehicles. Most of the highest traffic volumes are along the coast while traffic volumes in the central and east part of the country are low. There are no national roads with traffic volumes above a threshold to sustain financial viability and attract private sector financing. Light vehicles represent around 75 percent of the traffic, public transport (buses) and trucks represent 11 and 13 percent respectively. 2.6 DEMAND FOR ROAD SERVICES Demand for road services is increasing and vehicle ownership is concentrated in Luanda. The demand for road transport is increasing rapidly as the vehicle fleet has grown by 2.6 times over the 2005-2015 period or at 10 percent a year. The total vehicle fleet reached 880,000 million in 2015 (Figure 5) and will be around 1.4 million vehicles by 2020 using the same annual growth rate. The motorization is only 45 vehicles per 1,000 people, but with an income of US$3,600 per capita, growth in the vehicle fleet is likely to continue at a rapid pace unless there are policy interventions on vehicle ownership and use. The concentration of around 36 percent of the vehicle fleet in Luanda will exasperate congestion problems. Authorities face trade-offs between allowing market forces to prevail for vehicle ownership and use based on current taxation levels, or to intervene by reducing demand through higher taxation, a reduction in fuel subsidies and stricter regulation. International experience shows that urban congestion needs a combination of measures to optimize city travel including fiscal policy, transport demand management and substantial investments in public transportation. Special regulations are also frequently needed to control two-wheeled traffic, especially motorcycles. 12 Figure 5: Angola vehicle fleet 1000000 900000 800000 700000 600000 500000 400000 300000 200000 100000 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: International Organization of Vehicle Manufacturers, 2018 2.7 ROAD SECTOR INSTITUTIONAL SET UP The road sector is under the responsibility of the Ministry of Construction and Public Works. The National Road Institute (Institute de Estradas de Angola, NAE), which is part of the Ministry of Construction and Public Works (MPW), has responsibility for the management and development of the fundamental and complementary road networks (26,000 km and 17,500 km respectively. INAE has also a supporting advisory role to the unclassified roads, managed by the provinces/municipalities. INAE has regional offices in each of the 18 provinces and employs a total of around 700 staff. The INAE staff/network level of 1.6 staff/100km is close to the internationally recommended efficiency level of two staff/100km and compares well with many good practice agencies in Sub-Saharan Africa (Table 1) Table 1: Road agencies* staff efficiency INAE Comparison to Length of national Total Number Staff/100 km Roads Agencies* network (km) of Staff Angola (INAE) 43,000 700** 1.6 Uganda 20,800 916 4.2 Tanzania 33,012 719 2.2 Namibia 15,819 278 1.8 New Zealand 10,906 189 1.7 South Africa 16,170 195 1.2 Source: World Bank, INAE *Road Authorities with their construction and maintenance works contracted out to the private sector. **Excluding support and contractual staff, but including INAE regional offices who work on fundamental roads In addition to INAE, DNIP (Direcção Nacional de Infra-Estruturas Públicas) is responsible for the coordination and technical control of the construction of public infrastructure and DNOE (Direcção Nacional de Obras Engenharia) is responsible for the management of all projects related to bridges with a total length of more than 10m. Road Fund: The Angola road fund was established by a Presidential decree in 2015. The road fund is dedicated to financing maintenance works of the fundamental and complementary road networks. The road fund employs around thirty staff and has a board of directors, an advisory council and a supervisory board. A decree in 2019 made two main amendments to the legislation (i) the road fund to report to the 13 Ministry of Construction and Public Works instead of the Ministry of Finance (MINFIN) and (ii) the transfer of the function of procurement of road maintenance activities from the road fund to INAE. As per the presidential decree, the road fund revenue includes a fuel levy, vehicle license fee, vehicle and spare parts import taxes, and road tolls. The Ministry of Finance also allocates a yearly budget for emergency road works. However, revenues are not channeled directly to the road fund as per the presidential decree. MoF allocates an annual budget mainly to cover the cost of emergency road works. Unclassified roads (Municipal roads). The Direcção Provincial Dos Transportes, Telecomunicações, E Tecnologia De Informação in each of the 18 provinces manages the unclassified roads (municipal roads). Provincial governments also identify priorities for road improvements/maintenance, select contractors for the execution of services and supervise unclassified road works. INAE regional offices provide support and equipment to municipalities for emergency works. There is a lack of technical capacity at provincial level. For instance, the Directorate of transport of Huila, second largest province after Luanda, has 29 staff including two architects and four engineers but has no road engineers. Revenue from local taxes is insignificant, sub-national government rely on central government transfers. The so-called ordinary central government transfers are essentially financial quotas defined by the MINFIN2 for specific recurrent and capital purposes. The size of the financial quotas is ad hoc, with the allocation negotiated each year, mainly between provinces and central government, and approved by the State annual budget law. Based on 2015-2017 municipal revenue data, approximately one-third of the municipalities have ordinary transfers as their only source of revenue. At present time, municipalities’ revenue autonomy is largely confined to the management of central sources. Central government transfers including shared taxes represent the quasi-totalities of municipalities’ revenue. Local taxes represent less than 0.5 percent of the total revenues. 2MINFIN allocates the transfers. However, the Ministry of Planning and line ministries participate in the definition of the financial quotas. 14 3 PUBLIC EXPENDITURE 3.1 PUBLIC EXPENDITURE ON INFRASTRUCTURE Angola has spent a total of around US$US45 billion on its infrastructure over the last decade (2008- 2018), or around US$4.5 billion/year. By the end of the civil war in 2002, the Government of Angola has launched a large infrastructure reconstruction program spending on average US$4.4 billion per year on infrastructure between 2002-2009. This effort was sustained over the last decade (2008 -2018) and Angola spent on average 4.1% of its GDP per year on its infrastructure sector (Figure 6). Angola’s infrastructure public expenditure is on a par with the average expenditure of Lower and Medium Income Countries (4% of GDP for 2011) and is nearly twice more than Sub-Saharan African countries (2.54% of GDP)3, Table 2. The variation in public expenditure on infrastructure4 has followed the variation in oil revenues over the same period. Public expenditure has decreased following the 2014-15 oil-price dive. Figure 6: Infrastructure public expenditure as % of GDP (2008-2018) 9% 120.0 Infrastructure Expenditure 8% 100.0 7% Oil price (USD) 6% 80.0 (% of GDP) 5% 60.0 4% 3% 40.0 2% 20.0 1% 0% 0.0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 % capital invesment in infra vs GDP Price of oil Source: MINFIN Table 2: Infrastructure investment as % of GDP, by scenario and region, 2011 Lower-bound Central Upper-bound Region estimate estimate estimate East Asia and Pacific 5.36 5.72 6.71 Europe and Central Asia 1.51 2.73 4.36 Latin America and Caribbean 2.02 2.39 3.22 Middle East and North Africa 1.67 4.79 4.73 South Asia 3.59 4.42 4.25 Sub-Saharan Africa 1.87 2.54 3.47 LMIC average 3.38 4.11 4.99 Source: Beyond the Gap. How Countries Can Afford the Infrastructure They Need while Protecting the Planet. Policy Note 1/6. World Bank (2018) 3 How much is needed? Infrastructure investments for sustainable development, World bank, 2019 4 Infrastructure sectors include transport, energy, water and sanitation, irrigation urban infrastructure 15 3.2 PUBLIC EXPENDITURE ON THE TRANSPORT SECTOR Most of the public expenditure on infrastructure was in the transport sector of which more than two- thirds were spent on roads. The lion share of Angola’s infrastructure public expenditure was spent on the transport sector and around three-quarters of the public expenditure in the transport sector was on the road sub-sector (Figure 7), followed by the railway, port and the air transport sub-sectors (Figure 8). Figure 7: Public expenditure in the transport and the road sectors as % of GDP (2008-2018) 7.00% 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Public expenditure in transport sector as % of GDP Public expenditure road sector as % of GDP Source: MINFIN Figure 8: Public expenditure in the transport sector by mode of transport 450 400 Public Expenditure in transport - by 350 300 sub-sector (Billions of 250 200 150 Kz) 100 50 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Road Sector Railway Sector Air Transport Sector Water Transport Sector Source: MINFIN 3.3 PUBLIC EXPENDITURE ON THE ROAD SECTOR Angola has spent an unprecedented amount to salvage its road network but the share of public expenditure on road maintenance is negligible. Angola’s 27-year civil war (1975-2002) devastated its infrastructure. For instance, most of the road network was in a very bad condition and its size shrank by around 20,000 km as a result of years of underinvestment and neglect during the civil war that ended in 2002. In addition, more than 300 bridges of varying lengths and capacities have been destroyed and 16 needed urgent rehabilitation. Angola has therefore made unprecedented effort to rebuild its dilapidated road network during the first post-civil war years and has spent on average US$2.8 billion/year5 in road re-opening programs (2002-2009). This effort was pursued over the 2008-2018 period and around US$2.1 billion/year or 2.1% of Angola’s GDP was allocated each year to the road sector (Figure 9). Figure 9: Road sector expenditure as % of GDP (2008-2018) 7.0% 6.0% 5.0% 4.0% % GDP 3.0% 2.0% 1.0% 0.0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Source: MINFIN With the decline in oil revenue since the 2014-15 oil-price shock, the public investment program has been increasingly financed by external project loans. The Government recognizes that the extent of the road sector rehabilitation backlog is significant and has launched road salvation programs using external financing sources. The implementation of the first road rehabilitation program (2013-2017) worth US$176m has achieved only around half of its objectives. The Government is currently implementing a second “road salvation program” for the 2018-2022 period. This program is mostly (65%) financed externally by China credit line, German commercial Bank, Eurobonds and other bilateral (Portugal) and multilateral financing partners (AfDB). The shift from road development to road maintenance has not occurred during the last decade. The quasi-totality (97%) of the public expenditure on the road sector was allocated to capital expenditure during the 2008-2018 period. During this period, the share of road maintenance expenditure was on average 3.5% of the total expenditure on roads, or $US28.0 million on average per year over the same period (Figure 10 and 11). There was a sharp increase in the share of road maintenance expenditure (25%) in 2015 which could be explained by the creation of the road fund (2015) but hardly any road maintenance expenditure was undertaken during the last three years (2015-2018). Oil revenues have taken a sharp decline during this same period and road maintenance has subsequently not become a priority. 5 Pushak and Foster (2011), Angola’s Infrastructure: A Continental Perspective , p. 18. 17 Figure 10: Public expenditure on road maintenance (2008-2018) 8,000 25% Millions Share of road maintenance in road Expenditure in roads maintenance 7,000 infrastructure expenditure (%) 20% 6,000 5,000 15% 4,000 (Kz) 3,000 10% 2,000 5% 1,000 0 0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Expenditure in roads maintenance (Kz) Share of road maintenance in road infrastrcuture Expenditure (%) Source: MINFIN Figure 11: Share (%) of road maintenance in total road infrastructure expenditure (2008-2018) 25% 20% 15% 10% 5% 0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Source : MINFIN 3.4 PUBLIC EXPENDITURE AT SUBNATIONAL LEVEL Public expenditure is highly concentrated at the central government level, with only few public resources allocated to subnational entities. As a share of Angola’s GDP, central, provincial and municipal spending6 represent 34%, 5% and 1% respectively over the 2008-2016 period (Figure 12). Central, provincial and municipal expenditures represent respectively on average 86%, 12% and 2% of the total public expenditure on all the sectors of the economy (Figure 13). The decrease in the central 6 Central government spending includes the ministries and their reporting entities, public institutions, and expenditures from the legislative and judicial branches. 18 government spending since 2014 (from 47.7% to 28.7 of GDP) is largely attributable to the decline of oil prices. Figure 12: Public Expenditure trends by level of government in Kz (% GDP), 2008–2016 60 50 40 as a share of GDP 30 20 10 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 Central Province Municipality Figure 13: Expenditure trends by level of government, 2008-2016 (% total spending) 100% 90% 80% % total spending 70% 60% 50% 40% 30% 20% 10% 0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 Central Province Municipality Source: MINFIN. Only a small fraction of the road sector expenditure is allocated to tertiary/municipal roads. Road allocations to provincial level represent on average 13% of the total expenditure on the road sector over the 2008-2018 period. Most (87%) of Angola road sector expenditure is allocated to national roads (Figure 14). 19 Figure 14: Share of road sector expenditure at central and sub-national levels (2008-2018) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Central Government/National roads Provincial Government/Teriary roads Source: MINFIN. This trend was reversed for the 2015 year where subnational roads were allocated around 70% of the total expenditure on road sector but in absolute terms, this represents only around 24 billion Kwanzas, or around $US19.0 million (Figure 15). Figure 15: Road sector expenditure at central and sub-national levels (in Billion Kz), (2008-2018) 450 400 350 300 250 200 150 100 50 - 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Central Government/National roads Provincial Government/Teriary roads Source: MINFIN. 20 4 HOW EFFICIENT IS PUBLIC SPENDING IN THE ROAD SECTOR? 4.1 NATIONAL ROADS Assessing the efficiency of road sector expenditure in Angola is a challenge. The expenditure on road sector based on the MINFI BOOST database (2008-2018) is well described in chapter 2. However, data on outputs, including length of roads paved, rehabilitated and maintained is not? available. INAE has no road database or road asset management system and data at subnational level is scarce. An estimate of the road sector outputs over the 2008-2018 period is based on an analysis of aggregate data described in the National Development Plan (NDP 2013-17), the National Transport Sector Master Plan (2019) and the limited data from the Ministry of Construction and Public Works (MINCON) for the 2001-2013 period and from INAE for 2013-2017. Angola’s efforts to rebuild its road network were below the ambitious targets set in its National Development Plans. According to the Ministry of Construction and Public Works, as a result of the effort made to rebuild its road network, Angola has rehabilitated/paved 6402 km7 over the post-civil war period 2002-2010, or around 800km/year with a further 3330 km rehabilitated between 2011-2012. During 2013-2017, INAE data as reported in the National Transport Strategy Master Plan report (2018), shows that 3259.8 km were rehabilitated. It is therefore estimated that a total of around 8200 km of roads were paved/rehabilitated over the whole 2008-2017 period (Table 3). This is below the ambitious target of 15,500 km of primary roads and 6,000 km of secondary roads set by the National Development Plan, NDP (2013-2017). Table 3: Length of national roads paved and rehabilitated and expenditure (2008-2017) Year/period Paved/rehabilitated (Km) 2008-2010 1601.0 2011-2012 3330.0 2013 1171.1 2014 1335.7 2015 321.1 2016 111.1 2017 320.4 Total (2008-2017) (in km) 8190.4 Total spending (2008-2017) (in US$ billion) US$20.64 Unit cost (US$ million/km) 2.52 Source: INAE/NTSMP, MINCON, MINFI, World Bank estimates Real increases in spending have not been accompanied by commensurate increases in physical road output. During the period (2008-2017), central government has spent a total of around US$20.64 billion on national roads, or around US$2.52 million/km. This is a very high unit cost compared to an average unit cost to build a new two-lane road which, on the high range, is estimated to be around 0.8-1.0 million/km. Considering the total amount of road sector expenditure during this same period of around US$20.64 billion, Angola could have added a total of around 25795 km of new roads to its national network (Table 4) or three times more than the actual road length (8200km)reported by INAE and 7 Angola’s Infrastructure Ambitions Through Booms and Busts Policy, Governance and Reform, Research Paper Søren Kirk Jensen Africa Program, September 2018 21 MINCON. The potential increase in physical outputs would be much higher (31266 km) in the case the total spending was used to undertake reconstruction works. Table 4: Central government level spending on roads and potential physical road outputs Average road expenditure at central Government Total (US$ level (US$ Billion/year) Billion) Time period (2008-2017) 2.06 20.64 Current Condition Asphalt Mix Road Work Two-Lane Unit Costs of Total (km) Type Road Works (US$/km) Critical condition Reconstruction 660,000 31266 No Road New Construction 800,000 25795 Source: MINFI, World Bank estimates 4.2 UNCLASSIFIED/MUNICIPAL ROADS Angola could have added a total of around 34,000 km of new sub-national roads or two-thirds of the current length of its municipal road network. During the 2008-2017 period, around 17.5 billion Kwanzas/year, on average, were spent on provincial/tertiary roads (or an equivalent of US$171 million/year), Figure 15. These allocations/expenditures were erratic and varied from 1.0 to 25 billion Kz over this period. With the assumption that the entire allocations was spent on municipal/tertiary/unclassified road network of 33,000 km, the average unit cost would be around US$51,815/km. This is more than a unit cost of a full reconstruction of a two-lane earth road on the high range. Considering the total amount of the sub-national road sector expenditure during this same period was around US$1.71 billion, Angola could have added a total of around 21350 km of new municipal roads or two-thirds of the current length of its municipal road network. The potential increase in physical outputs would be much higher (34198km) if the total spending was used to undertake reconstruction works (Table 16). Figure 16: Spending on sub-national roads (in Billion Kz), (2008-2018) 30 25 20 15 10 5 - 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Source: MINFIN. 22 Table 5: Sub-national government level spending on roads and potential physical road outputs Average road expenditure at sub-national Total (US$ Government level (US$ Million/year) Billion) Time period (2008-2017) 171.0 0.171 Current Condition Earth Road Work Type Two-Lane Unit Costs of Total (km) Road Works (US$/km) Critical condition Reconstruction 50,000 34198 No Road New Construction 80,000 21374 Source: MINFI, World Bank estimates The impact of recent investment levels is minor and can be seen in the poor-quality of infrastructure in Angola. The quality of Angola’s road infrastructure is ranked 136th out of 141 countries, based on the 2019 Global Competitiveness Report. Angola’s score is 2.2 out of 7 and is one of the lowest in the Africa region. Angola is lagging behind countries with the same GDP/capita (US$3,600/cap) (Figure 17 as well as many countries with lower GDP/cap (i.e. Ghana, Senegal, Cote d’Ivoire…). Angola ranks 159 out of 160 countries in the 2018 Logistics Performance Index (LPI)8 and 153 out of 160 in the Infrastructure Performance component of the LPI9 (Figure 18). This low quality of infrastructure is one factor constraining further improvements in Angola’s competitiveness ranking. Figure 17: Quality of road infrastructure Source: Connecting to Compete 2018, Trade Logistics in the Global Economy, The World Bank Figure 18: LPI Score in major economies in SSA 8 Connecting to Compete 2018, Trade Logistics in the Global Economy, The World Bank 9 The logistics performance (LPI) is the weighted average of the country scores on the six key dimensions: i) Efficiency of the clearance process (i.e., speed, simplicity and predictability of formalities) by border control agencies, including customs; ii) Quality of trade and transport related infrastructure (e.g., ports, railroads, roads, information technology); iii) Ease of arranging competitively priced shipments; iv) Competence and quality of logistics services (e.g., transport operators, customs brokers); v) Ability to track and trace consignments; 6) Timeliness of shipments in reaching destination within the scheduled or expected delivery time. 23 5 ROAD SECTOR FINANCING NEEDS 5.1 ROAD SECTOR FINANCING NEEDS ASSESSMENT APPROACH. Most of the road network in Angola is in poor condition and the opportunity cost of neglecting road maintenance is high. The Road Network Evaluation Tools Model (RONET) developed by the World Bank, was used to evaluate the preservation (maintenance and rehabilitation) requirements of the Angolan road network. RONET is a tool for evaluating the performance of road maintenance and rehabilitation strategies and the importance of the road sector to the economy. It assesses the current network condition and traffic and computes the asset value of the network and road network monitoring indicators. It uses country-specific relationships between maintenance spending and road condition and between road condition and road-user costs, to assess the performance over time of the network under different road works standards. It determines, for example, maintenance and rehabilitation road works that minimize total transport costs or the cost for sustaining the network in its current condition. It also estimates the savings or the costs to the economy to be obtained from maintaining the network at different levels of service. In addition, it determines the allocation of expenditures between recurrent maintenance, periodic maintenance, and rehabilitation road works. Finally, it can be used to determine the “funding gap,” defined as the difference between current maintenance spending and required maintenance spending (to maintain the network at a given level of road condition), and the effect of under-spending on increased transport costs. 5.2 INPUT DATA ASSUMPTIONS The available information allows only for a macro evaluation of the network. RONET performs a macro evaluation of the network for monitoring and strategic planning purposes by characterizing the road network using road categories of functional classification, surface type, condition and traffic. Thus, one of the main inputs to RONET is the distribution of the network length by different road categories, together with the average unit costs of road works and vehicle fleet characteristics. The data gaps were filled by making assumptions based on available information including unit costs of road works in Africa region and by revising the default data used by RONET to take into consideration the country context (i.e. Road work unit costs). Road network size and condition The assessment of road financing needs is undertaken for the whole road network (76000km) as presented in Figure 2. Separate road financing needs estimated for (i) the fundamental road network (26000 km); (ii) the complementary road network (17000 km) and (iii) the unclassified/municipal road network (33000 km). Road network condition estimated for each road network class based on the results of road inventory and condition carried out on around 19000 km of paved and unpaved roads under the Transport National Strategy, NTSMP, prepared by AfDB (2018). The estimates show that the share of the road network in poor and critical condition for the paved, unpaved and unclassified roads is 25%, 60% and 80%. The results show that around two-third (64%) of the total rod network is in poor or critical condition. This result is well aligned with the road conditions reported in the National Transport Strategy and Master Plan (NTSMP, 2018). The road condition data used to estimate the financing needs of the total road network of 76,000 km is presented in Table 6. 24 Table 6: Network length by type and road condition (km) Fundamental Fundamental and Unclassified Paved complementary Municipal Total Network Unpaved Average Road Condition % Very Good 2% 0% 0% 0% Good 41% 15% 5% 13% Fair 36% 25% 15% 20% Poor 12% 35% 45% 38% Very Poor 8% 25% 35% 28% Road traffic Traffic composition and average traffic volume using each class of road was derived from the traffic counted on 30 survey stations conducted for the preparation of the NTSMP in 2018. Traffic composition and the share of traffic using each type of road class is presented in Table 7. Table 7 :Traffic composition and volume by road network type Vehicle type Fundamental & Unclassified % Fundamental Complementary Municipal Daily Traffic (AADT) Paved Unpaved From To (%) Private Vehicle 34% 0 10 0% 0% 0% Delivery Vehicle 7% 10 30 0% 0% 25% Truck Light 4% 30 100 3% 25% 50% Truck Medium 2% 100 300 27% 50% 25% Truck Heavy 2% 300 1000 23% 25% 0% Truck Articulated 5% 1000 3000 37% 0% 0% Bus Light 4% 3000 10000 10% 0% 0% Motorcycles 42% 10000 30000 0% 0% 0% Road works unit costs Data on unit costs of road works in Angola was not made available to carry out a country specific analysis. Unit cost estimates were made based on road projects undertaken in Africa. Table 8 presents a summary of a study of road unit costs commissioned by the African Development Bank (AfDB). Road projects financed by the World Bank and AfDB were considered from Benin, Botswana, Burkina Faso, Cameroon, Chad, Ethiopia, Ghana, Morocco, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, São Tomé and Principe, Swaziland, Tanzania, Tunisia, and Uganda. The projects were contracted through 2000-2010. The unit costs in the report are in terms of lane-kilometers. The values are double the report values to relate to two-lane roads and have been updated from 2006 values to 2018 using the USA CPI. One set of statistics is provided for projects smaller than 50 (two-lane) km (typically subject to a large variance) and those larger than 50 (two-lane) km (typically demonstrating small variance). 25 Table 8: Unit costs for road works in Africa (US$ thousand per two-lane km) Project length and Re-graveling / periodic Periodic maintenance Rehabilitation of Construction and statistic maintenance of of paved roads paved roads upgrading of paved unpaved roads roads < 50 km Upper quartile 24.8 - 684 1,004 Median 22.6 - 426 538 Lower quartile 19.2 - 260 392 ≥ 50 km Upper quartile 30.2 170 308 382 Median 26.6 152 200 348 Lower quartile 22.6 134 112 274 Source: Study on Road Infrastructure Costs: Analysis of Unit Costs and Cost Overruns of Road Infrastructure Projects in Africa, African Development Bank, May 2014. Of the projects analyzed, local contractors were generally used for maintenance of unpaved roads. Contractors from the UK and France were used for a third of the periodic maintenance projects on paved roads, with local contractors used for the other periodic maintenance projects. Seventy-five rehabilitation projects were analyzed. Of these, seventeen projects used local contractors, thirty-three projects used contractors from a range of countries (China, France, Italy, Netherlands, Portugal and South Africa), and the contractor’s nationality was not reported for the remaining projects. Fifty-four construction/ upgrading projects were included in the analysis. Local contractors were used on eight of these projects, international contractors (Brazil, China, France, Germany, Italy, Kuwait, Netherlands and South Africa) were used on twenty-eight of the projects, and the contractor’s nationality was not reported for the remaining eighteen projects. Table 9 presents unit costs for road works in Sub-Saharan Africa financed by the World Bank. The values are for a 2-lane, 7-metre wide road equivalent, and have been updated from 2007 values to 2018 using the USA CPI. Table 9: Average unit costs for road works in Sub-Saharan Africa (1,000US$ per two-lane km) Country Re-graveling of Periodic maintenance Rehabilitation and Upgrading of gravel unpaved roads of paved roads reconstruction roads to paved Congo 77.6 Congo (DRC) 75.0 260 Ethiopia 443 368 Ghana 298 Kenya 1,090 Madagascar 62.2 120 Malawi 480 Mozambique 225 318 Tanzania 397 Uganda 479 Zambia 28.5 100 Source: Monitoring Road Works Contracts and Unit Costs for Enhanced Governance in Sub-Saharan Africa. Transport Paper TP- 21, World Bank, September 2008. The average costs for rehabilitation and reconstruction works is particularly high in Kenya in comparison to other countries in the sample. The data for Kenya relates to four contracts with an asphalt mix pavement course and bituminous base. Of the projects analyzed, local African firms executed 43%, Chinese firms 27%, European firms 17%, firms from other African countries 8%, and firms from outside Africa and Europe 5%. Review of costs of road construction projects in Ethiopia10 found that the average 10 Review of Costs of Road Construction Projects in Ethiopia, Report No. ACS14210, World Bank Group, June 2015. 26 cost for upgrading a gravel road to asphaltic concrete (AC) standard was US$ 710,000 per kilometer, with maximum and minimum values of US$ 1.04 million and US$ 504,000 respectively at January 2018 prices. Unit costs used to estimate road rehabilitation and maintenance needs in Angola are based on a 2010 study of unit costs of road works in Africa. This analysis has led to using conservative unit costs twice higher than the unit costs used by default by RONET. Table 8 presents the average unit costs per km per road roughness level for a two-lane road for capital and recurrent maintenance works. Table 10: Two-lane Unit Costs (US$/km) of Road Works by surface type and condition Current Surface Type Condition Road Work Class Road Work Type (US$/km) Asphalt Mix Good Condition Periodic Maintenance Preventive Treatment 24,000 Fair Condition Resurfacing (Overlay) 200,000 Poor Condition Rehabilitation Strengthening (Overlay) 400,000 Very Poor Condition Reconstruction 660,000 No Road New Construction New Construction 800,000 Surface Good Condition Periodic Maintenance Preventive Treatment 24,000 Treatment Fair Condition Resurfacing (Reseal) 54,000 Poor Condition Rehabilitation Strengthening (Overlay) 320,000 Very Poor Condition Reconstruction 520,000 No Road New Construction New Construction 660,000 Gravel Good Condition Periodic Maintenance Spot Re-gravelling 6,000 Fair Condition Re-gravelling 34,000 Poor Condition Rehabilitation Partial Reconstruction 80,000 Very Poor Condition Full Reconstruction 120,000 No Road New Construction New Construction 160,000 Earth Good Condition Periodic Maintenance Spot Repairs 400 Fair Condition Heavy Grading 1,600 Poor Condition Rehabilitation Partial Reconstruction 16,000 Very Poor Condition Full Reconstruction 50,000 No Road New Construction New Construction 80,000 Recurrent Maintenance Works Unit Costs (US$/km-year) Two-Lane Unit Costs of Road Works (US$/km-year) Road Surface Type Condition Paved Unpaved Tertiary Urban Unclassified Cement Concrete Very Good 4,000 4,000 4,000 2,000 2,000 Good 5,000 5,000 5,000 2,500 2,500 Asphalt Mix Fair 6,000 6,000 6,000 3,000 3,000 Poor 3,000 3,000 3,000 1,500 1,500 Surface Treatment Very Poor 3,000 3,000 3,000 1,500 1,500 Gravel Very Good 2,000 2,000 2,000 1,000 1,000 Good 2,500 2,500 2,500 1,252 1,252 Fair 3,000 3,000 3,000 1,500 1,500 Poor 1,500 1,500 1,500 750 750 Very Poor 1,500 1,500 1,500 750 750 Earth Very Good 600 600 600 300 300 Good 900 900 900 450 450 Fair 1,200 1,200 1,200 600 600 Poor 600 600 600 300 300 Very Poor 600 600 600 300 300 27 5.3 ESTIMATION OF ROAD REHABILITATION AND MAINTENANCE NEEDS RONET evaluated the network performance under different preservation standards in order to determine the optimal requirements for recurrent maintenance, periodic maintenance and rehabilitation. For each road class, RONET identified the standard that minimizes the present value of total transport costs (road agency costs plus road user costs), at a discount rate, thus maximizing society net benefits (NPV). The RONET evaluation adopted the following assumptions: (i) traffic growth rate of 3%/year (ii) discount rate of 6% and (iii) 20 years evaluation period; (iv) unit costs of road works based on the highest road work costs in Africa; and (v) average unit road user costs based on current average vehicle fleet characteristics in Angola. RONET results show that the current road network asset value is around US$11.250 billion or around 11.0% of Angola’s GDP. The current asset value as a share of maximum road asset value is 74.5%. This indicates that more than 25% of the road network asset value is lost due to lack of maintenance or deferred road rehabilitation work. RONET estimates the total road network preservation annual requirements, considering the scenario that minimizes total transport costs11 for each road class, for rehabilitation, periodic maintenance and routine maintenance in years 1 to 5 and in years 6 to 20. In this analysis, the following definitions are used: • Periodic maintenance costs are the costs of road works applied to roads in good and fair condition; • Pavement rehabilitation costs are the costs of road works necessary for roads in poor or very poor condition; • Routine maintenance costs are applied to all the roads. Table 11 below presents the required expenditures without budget constraints over the next five years and in years 6 to 20, in US$ million per year. Over the next five years, a total of US$ 924.8 million per year is needed for rehabilitation (US$613.2m), periodic maintenance (US$241.2m) and routine maintenance (US$70.4m) of the total network, of which US$771.4 m or 83.4% should go to the national roads (Fundamental and complementary roads) and the remaining US$153.4m or 16.6% to the sub- national roads (unclassified/municipal roads). Once this rehabilitation backlog program is implemented during the first five years, in years 6 to 20, US$320.4 million per year is needed for preservation works on the entire network of which around 90% will be spent on routine and periodic maintenance work. The annualized net present value of road agency costs over the 20-year evaluation period, at 6% discount rate, is US$660.8 million/year. Table 11: Two-lane Unit Costs of Road Works by surface type and condition Years 1 to 5 Annual Road Works Costs (US$ million/year) Routine Periodic Network Maintenance Maintenance Rehabilitation Total Percent Fundamental 52.0 220.4 386.0 658.4 71.2 Complementary 9.1 8.7 95.2 113.0 12.2 Sub-Total/National 61.1 229.1 481.2 771.4 83.4 11RONET computes, for each road class and for different preservation standards, the present value of road agency costs, road user costs, and total transport costs (sum of road agency and road user costs); thus, the optimal standard per road class is the one that minimizes the present value of total transport costs. 28 Unclassified/Municipal 9.3 12.1 132.0 153.4 16.6 TOTAL 70.4 241.2 613.2 924.8 100 Percent 7.6 26.0 66.4 100 Years 6 to 20 Annual Road Works Costs (US$ million/year) Routine Periodic Network maintenance maintenance Rehabilitation Total Percent Fundamental 49.6 195.2 32.5 277.3 86.5 Complementary 9.1 10.3 0.0 19.4 6.1 Sub-Total/National 58.7 205.5 32.5 296.7 92.6 Unclassified/Municipal 9.3 14.4 0.0 23.7 7.4 TOTAL 68.0 219.9 32.5 320.4 100 Percent 21.2 68.6 10.1 100 Source: RONET Angola’s average annual road sector expenditure (US$2,231 billion) is nearly two and a half times (2.4 time) more than road rehabilitation and maintenance requirements (US$0.925 billion). The estimated annual cost needed to properly rehabilitate and maintain the national roads (US$771.4 million) is 2.67 less than the average annual budget allocations (US$2.06 billion) to national roads (2008-2017). The results also show that US$70.4 million/year are required to undertake routine maintenance interventions on the whole road network of which US$61.1 million will cover national roads and US$9.3 million the sub-national roads. The analysis of the BOOST database shows that only US$28.0 million, on average, are allocated each year to road maintenance work for the whole road network (2008-2018). This indicates that the budget allocations for road maintenance cover only 40% of the routine maintenance needs. As for the sub-national roads, an estimate of 153.4 million per year is needed to rehabilitate and maintain the municipal road network (33000km). This is less than the current average annual budget of US$171.0 million allocated each year through sub-national government to municipal roads. The average budget allocations exceed by 11% the road rehabilitation and maintenance requirements (Table 12). Table 12: Comparison of actual expenditures and funding requirements (in US$ million/year)12 Actual Expenditure Estimated Needs National road network (Fundamental & Complementary 2,060.0 771.4 roads) Sub-national (Unclassified/Municipal roads) 171.0 153.4 Total road network 2,231.0 924.8 Source: MINFI, RONET 12 The RONET results (Appendix 1) compare actual revenue and estimated needs for funding gaps analysis. However, considering the weak relationship between revenues from road user charges and expenditure on road preservation in Angola, actual expenditure is used 29 Table 13 presents the annualized present value of road-agency, road-user and total transport costs over the next 20 years, at a 6% discount rate, for the “minimize transport costs” scenario and for a “do minimum” scenario. For the minimized total transport costs scenario, road user costs represent 89% of total transport costs. The increase in annualized present value of total transport costs of “do minimum “corresponds to US$5,462 million per year. Compared with “minimize total transport costs” scenario, for every dollar saved by the road agency under the” do minimum” scenario, road user costs increase by 3.4 times. This indicates that every US dollar spent on road maintenance will generate US$3.4 in road user cost saving. Table 13: Annualized costs years 1 to 20 Minimize Transport Costs Scenario (Million US$/year) Road Agency Road Users Society 4,464 70,865 75,329 Do Minimum Scenario (M US$/year) Road Agency Road Users Society 3,226 76,547 79,772 Reduction in Costs Compared with Do Minimum (M US$/year) Road Agency Road Users Society -2,260 7,723 5,462 Increase in Road User Costs per Decrease in Agency Costs 3.4 times Source: RONET In the ‘minimize transport costs’ scenario, the percentage of roads in good and fair condition will improve and the road roughness index will decrease from the current 15.3 to 8.4. However, under the “do minimum” scenario the share of roads in good and fair condition will decline as the roughness index will increase to 20.4. A situation where the road network will be barely passable (Table 14). Table 14: Road condition per transport costs scenario Minimize Transport Costs Scenario (Roughness Index) Current 2030 15.3 8.4 Do Minimum Scenario (Roughness Index) Current 2030 15.3 20.4 Source: RONET 30 6 RECOMMENDATIONS The Government of Angola recognizes the important role of road transport and, since the end of the civil war in 2002, has made transport infrastructure a key priority with ambitious targets to be achieved. In the last decade (2008-2018), budget expenditures have increased significantly and a total of around US$23.0 billion was allocated to the development of the road sector, but there is evidence that the efficiency of such spending could be substantially improved. The road expenditure level is three times higher than the outputs. The share of public expenditure on road maintenance is negligible and two- thirds of the road network is still in critical condition. Effective and efficient functioning of road transport infrastructure is central to accelerating Angola’s economic development. The recommendations that follow are based on the findings of the public expenditure review of the sector and on the best practice principles governing successful commercialization of the road sector. Such change should be undertaken in a holistic and integrated manner with a clear vision of the objectives of the reform program and the strategies for its implementation. Improve the road sector governance. The level of spending on national roads (Fundamental and complementary roads) is nearly two and a half times (2.4 time) more than the estimated road rehabilitation and maintenance requirements. This increase in spending has not been accompanied by commensurate increases in physical road output and condition. This has resulted in much higher average costs for both preservation and development works. This inefficiency cannot be explained by a lack of technical and management capacity alone but is likely due to issues with the road sector governance. While the current road sector institutional set up may be sound; with a separation of the policy (MINCON), management (INAE) and financing (Road Fund, 2015) functions it does not, however, seem to be implemented effectively. Recent changes in the structure of the organization that made both INAE and the Road Fund report to the same Ministry (MINCON) could be the first step to improve coordination. There is clearly a need to introduce more efficient and business-oriented road management practices and to improve financial and managerial accountability. Addressing these constraints through improvements in the organization of road management services will be necessary to gain efficiencies and cost reductions that can be passed on to road users. The Road Fund should fully play its role of mobilizing adequate revenues for road maintenance activities and to make sure the resources are used in efficient way. Improve the effectiveness of the road fund. The road fund of Angola is established on a strong legal basis and on paper it meets all the second-generation road fund criteria. However, there are no implementing regulations for the road fund to effectively operate. The road fund revenues are still based on erratic budget allocations of only $US28.0m/year on average. This is half the required road maintenance needs of fundamental and complementary road network estimated at around $US59.0/year. The Road Fund should also validate the road maintenance programs and demand value for money by carrying out financial and comprehensive independent technical audits. The technical and financial auditing of maintenance projects is undertaken by most Road Funds, often on a sample basis, and has proved valuable in improving accountability by revealing the quality of the work done, the quantity of work completed and the timeliness of the work. Auditing should be undertaken for each stage of the project cycle from design to completion of works in order to ensure that Government and other stakeholders are receiving value for money. 31 Develop and sustain a road data base and asset management system. Description of the road network is scarce and unreliable. INAE has no road inventory database and does not carry out road condition and traffic surveys. Managing a road system in an optimal manner is a complex task that is likely to be influenced by a variety of factors including political. In such an environment, INAE will have to routinely face important policy questions from stakeholders as well as increasing demands upon the monies allocated by financing institutions (MoF, Road Fund). To address policy issues in a rationale manner, INAE would have to adopt an asset management approach using an appropriate road asset management system (RAMS). Data collection is expensive, and it is essential that the road agency only collects the data which are required for its management purposes. This data should be collected at a frequency and a level which is appropriate to make informed decisions on road planning and maintenance programs. INAE should consider contracting out the data collection, operation and upkeep of the RAMS to competent local consultants who would also be responsible for providing technical back- up. With such an arrangement, INAE would retain strategic responsibility for directing the evaluation of alternative strategies and developing programs for maintaining the road network in an optimum manner. Review the road network classification to assert the ownership of the complementary roads (the missing middle) and of the unclassified roads. The functions and status of road networks in Angola are regulated by Decree 21/92 and 46/92 issued in 1992. The status classification is the most important since it determines responsibility for financing of works, but the mapping between functional and status classifications is undermined by flexible interpretation of the term ‘strategic road’. The current classification does not adequately reflect the country’s administrative and organization structure (decentralization levels). The focus is put on the fundamental road network (26000km) and little attention is given to the complementary roads (17000km) and to the municipal road network (33000km). The latter provides access for most of the population but is still an unclassified network. The ownership of the complementary road network, which provides the link between the arterial and the tertiary roads, is not clear, and the whole road network continuity and integration are becoming a challenge (the missing middle). There is a need to review the current road classification. The main objective is to identify the road links, their characteristics and the role each link plays within the entire road network. The criteria for classifying the various roads will include population density and land use and traffic volumes as the main critical variable. The functional classification system will provide a foundation for proper highway planning and fiscal policy. Develop rural road development programs. Half of the population is located further than 2km from any road. Access of agriculture production to main markets is poor overall. almost three quarter of the total agriculture production value cannot reach markets, although this varies across provinces. Half of the population is located further than 2km from any road. Access to social services (health centers and schools) is overall poor, only 37% of the total population can access hospitals and schools in less than 2 hours. This indicates that there is a need to rehabilitate or build new municipal roads. Establishment of road management agencies at provincial level. The level of spending on maintenance and rehabilitation is slightly higher than the estimated needs, but the municipal road network remains in critical condition. This is partly due to lack of management and technical capacity at subnational level. There are currently more than 182 entities responsible for sub-national road maintenance (18 provinces and 164 municipalities), making it very difficult for a coordinated approach to address the maintenance backlog. A road maintenance demonstration project could be considered by establishing a semi- 32 autonomous provincial road agency to strategically plan and manage the extended sub-national road network. In addition to the provincial roads under the responsibility of the respective province based on the results of a road work reclassification, municipalities could delegate the management of all or part of their road networks using a contract management approach. The implementation could be tested through a demonstration project. The objectives of such a new institutional arrangement would be to increase regional coordination, take advantage of economies of scale and improve the effectiveness and efficiency of public expenditure. The proposal also addresses the technical and managerial capacity issue without undermining local decision-making authority. This institutional set up has been widely used with success in several countries in Latin America (Box 1). The new provincial road agency should not be perceived as a parallel institution to the existing arrangements. The personnel of the provincial road agency would be established from the existing Public Works staff who are currently managing province and local government roads. Representatives of the provinces and municipalities would form an oversight board for the agency. The board would advise on the planning, budget allocation and management of the road network. Box 1: Peru Provincial Road Institutes Peru has experienced a decentralization process in 2002. As a result, a three-tier structure is now in place with 24 elected regional governments and 2,006 municipalities (194 provinces and 1,812 districts). The World Bank and the Inter-American Development Bank have supported a substantial change in the decentralized management of sub- national roads. The first rural road project in Peru has helped establish the first decentralized “provincial road institute” which are now established in every province. The Provincial Road Institute mandate is to design and implement province-wide road strategies that provide an effective means to coordinate between districts and local government, private sector providers and local communities. The Provincial Road Institute helped overcome weak financial and administrative capacity in districts to maintain roads. The Project has resulted in the rehabilitation of 15,000 km of district roads and 2,700 km of provincial roads. The proportion of the current sub-national roads in good condition has nearly doubled. Various studies show substantial positive impact of the improved roads, including reduction in average travel time by 50 percent, a 78 percent reduction in passenger travel fares, 18 percent in freight cost and an increased access to education. The project concept is being currently extended to cover other infrastructure sectors (water and sanitation, rural electricity) and health. Source: World Bank Support the development of a competitive road contractors’ industry. It will be useful to carry out a study pf the current road construction industry to have a better understanding of the road contracting market, and to identify priority measures to address any constraints to its development. The performance of local contractors is viewed by many roads agencies as source of concern. However, there are often mitigating reasons for this which have been recognised in most countries where serious efforts are being made to improve the local contracting industry. Roads agencies and other stakeholders, such as national construction industry councils, should be encouraged to promote measures for improving local contracting capacity, including plant and equipment leasing and/or financing, use of multi-year rather than single year contracts and appropriate packaging of contracts into manageable sizes. 33 7 APPENDIXES 7.1 APPENDIX 1: ROAD ACCESSIBILITY ANALYSIS TO MARKETS, EDUCATION AND HEALTH This Appendix presents the results of the accessibility analysis to markets, education and health facilities in Angola. The analysis aims to build an understanding about the role of roads in the rural economic and the human capital development. This Appendix includes the following sections. (i) Access to agriculture markets: This section assesses the ability of agriculture production by value to access markets and identifies regions in Angola where access to markets for agriculture production is limited (ii) Access to social services: This section assesses the ability of people to reach schools and hospital and identifies people with different level of access. The analysis is rooted on geospatial modelling and the concept of accessibility to point of interest to capture indicators linked to two dimensions: (i) economic growth: analyzing data on agriculture production value and location of regional markets to assess the accessibility of rural farmers to regional markets, as a key indicator to foster economic growth in rural areas; and (ii) human capital development: analyzing the accessibility of population to critical social services such as schools and hospitals. The method uses a geospatial database that includes road network and location of specific points of interest (agriculture, markets, schools, health centers). The data utilized is mainly Figure 1: Road network, Source: OSM open data available from different data sources. It worth indicating the limitation to verify this open data. The data sources are: ▪ Population: WorldPop, 2015 estimates of people per grid square, Resolution 0.000833333 decimal degrees (approx.. 100m at the equator) ▪ Points of Interest (POI): Hospitals, Schools location from Open Street Map. Markets locations: considered cities with more than 30,000 people (Africapolis), where main markets are located Road Network: Open Street Map (Figure 1: Road network, Source: OSM ▪ ) Figure 2: Value of crop production (2010), Source: SAPM ▪ Agriculture: Value of Crop Production (2010) at 10km resolution – International Food Policy Research Institute (IFPRI) SPAM model (Figure 2: Value of crop production (2010), Source: SAPM 34 ▪ ) National roads density Fundamental road density Complementary road density (fundamental + (km/100km2) (km/100km2) complementary) (km/100km2) National roads (fundamental + Fundamental roads Complementary roads complementary) Km per 1000 people (km/1000 Km per 1000 people Km per 1000 people (km/1000 inhabitant) (km/1000 inhabitant) inhabitant) Figure 3: Road density by province 35 Agriculture products access to markets Agriculture production value is predominant in the north and east of Angola. There are five provinces - Moxico Uige, Luanda Sul, Malanje and Luanda Norte – that account for two thirds of the total agriculture value of Angola. Figure 4: % of agriculture value per province and Figure 5: % of agriculture value per province present the distribution of the agriculture production value per province. Figure 4: % of agriculture value per province Figure 5: % of agriculture value per province Uige Malanje Bie Huila Benguela Cuanza Norte Zaire Luanda Huambo 0% 5% 10% 15% % Agriculture Value The access of agriculture production to main markets is in overall poor, almost three fourths of the total agriculture production value cannot reach markets. Of the overall agriculture production value, 71 percent cannot reach markets13. Out of the 29 percent of the agriculture production value that can reach markets, only 37 percent of value can reach markets in less than 2 hours. Figure 6: % of agriculture production value that reach the main Figure 7: Accessibility to markets –Minutes to the closest markets under time thresholds market 13 The analysis considers the agriculture production within a buffer of 2 km around all classified and unclassified roads can reach markets 36 15% 22% 29% 34% 0% 20% 40% 60% 80% 100% <1h <2 h <4h >4h A significant part of the agriculture value in the top five provinces that account for two thirds of the national value does not have access to markets. The distribution of access to markets varies per province (Figure 8). Although Moxico has the largest agriculture production value of around 16 percent of the national value, it does not have the highest value accessible to markets because 83 percent of its value is not accessible. Uige and Malanje have the highest agriculture value accessible to markets, that represents around 4 percent of Angola agriculture value per province. Uige and Cuanza Sul are the provinces where the highest agriculture production can reach markets in less than 2 hours. Figure 8 presents the distribution of the agriculture production value, as a percentage of the national agriculture production value Figure 8: Distribution of the agriculture production value, as a percentage of the national agriculture production value by time thresholds and non-accessible Uige Malanje Bie Huila Benguela Cuanza Norte Zaire Luanda Huambo 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% <1h <2h <4h >4h Value Non-Accessible Access to social services Access to social services is overall poor, both in the access to hospitals and schools. Around half of the country’s population does not have hospitals14 and only 37 percent can access hospitals in less than 2 14 The analysis considers population within a buffer of 2 km around all classified and unclassified roads can access facilities 37 hours. The access to schools follows similar trends. These values offer a general understanding of the access issues to social services; however, limitations should be acknowledged of the data of schools and hospital location (from Open Street Map). It is therefore recommended that further analysis should be carried out using more accurate data. Figure 9: Accessibility to hospitals – time to reach closest Figure 10: Distribution of the population within thresholds of hospital (min) time to access hospitals 0% 20% 40% 60% 80% 100% < 30 min <1 h <2 h <4 h >4h Population without access Figure 11: Accessibility to schools – time to reach closest Figure 12: Distribution of the population within thresholds of school (min) time to access schools 0% 20% 40% 60% 80% 100% < 30 min <1 h <2 h <4 h >4h Population without access 38 39 7.2 APPENDIX 2: ANGOLA RONET EVALUATION Input Data Assumptions Unit Cost of road Works Used twice the default unit costs given on RONET resulting on the following unit costs per km for a two- lane road for capital works and recurrent maintenance works: Capital Road Works Unit Costs Two-lane Unit Costs of Road Works ($/km) Surface Type Current Condition Road Work Class Road Work Type ($/km) Asphalt Mix Good Condition Periodic Maintenance Preventive Treatment 24,000 Fair Condition Resurfacing (Overlay) 200,000 Poor Condition Rehabilitation Strengthening (Overlay) 400,000 Very Poor Condition Reconstruction 660,000 No Road New Construction New Construction 800,000 Surface t Good Condition Periodic Maintenance Preventive Treatment 24,000 Treatment Fair Condition Resurfacing (Reseal) 54,000 Poor Condition Rehabilitation Strengthening (Overlay) 320,000 Very Poor Condition Reconstruction 520,000 No Road New Construction New Construction 660,000 Gravel Good Condition Periodic Maintenance Spot Regravelling 6,000 Fair Condition Regravelling 34,000 Poor Condition Rehabilitation Partial Reconstruction 80,000 Very Poor Condition Full Reconstruction 120,000 No Road New Construction New Construction 160,000 Earth Good Condition Periodic Maintenance Spot Repairs 400 Fair Condition Heavy Grading 1,600 Poor Condition Rehabilitation Partial Reconstruction 16,000 Very Poor Condition Full Reconstruction 50,000 No Road New Construction New Construction 80,000 Recurrent Maintenance Works Unit Costs Two-Lane Unit Costs of Road Works ($/km-year) Road F. F. Surface Type Condition Paved Unpaved Tertiary Urban Unclassified Cement Concrete Very Good 4,000 4,000 4,000 2,000 2,000 Good 5,000 5,000 5,000 2,500 2,500 Fair 6,000 6,000 6,000 3,000 3,000 Poor 3,000 3,000 3,000 1,500 1,500 Very Poor 3,000 3,000 3,000 1,500 1,500 Asphalt Mix Very Good 4,000 4,000 4,000 2,000 2,000 Good 5,000 5,000 5,000 2,500 2,500 Fair 6,000 6,000 6,000 3,000 3,000 Poor 3,000 3,000 3,000 1,500 1,500 40 Very Poor 3,000 3,000 3,000 1,500 1,500 Surface Treatment Very Good 4,000 4,000 4,000 2,000 2,000 Good 5,000 5,000 5,000 2,500 2,500 Fair 6,000 6,000 6,000 3,000 3,000 Poor 3,000 3,000 3,000 1,500 1,500 Very Poor 3,000 3,000 3,000 1,500 1,500 Gravel Very Good 2,000 2,000 2,000 1,000 1,000 Good 2,500 2,500 2,500 1,252 1,252 Fair 3,000 3,000 3,000 1,500 1,500 Poor 1,500 1,500 1,500 750 750 Very Poor 1,500 1,500 1,500 750 750 Earth Very Good 600 600 600 300 300 Good 900 900 900 450 450 Fair 1,200 1,200 1,200 600 600 Poor 600 600 600 300 300 Very Poor 600 600 600 300 300 Vehicle Fleet Basic Characteristics and Unit Costs Used the default RONET vehicle fleet characteristics and unit costs Traffic Composition Based on the traffic composition given on the AfDB report. Motorcycle 42% Car Small 0% Car Medium 34% Delivery Vehicle 7% Four-Wheel Drive 0% Truck Light 4% Truck Medium 2% Truck Heavy 2% Truck Articulated 5% Bus Light 1% Bus Medium 2% Bus Heavy 1% Traffic Growth Rate Based on the employment forecast given on the AfDB report (3% per year). For all vehicles, 3 % traffic growth per year was adopted. Road Network Length The AfDB indicates that the fundamental network totals 26,000 km. The CIA factbook indicates that in 2018 in Angola there are 13,600 km of paved roads and 12,400 of unpaved roads (totaling 26,000 km). There is a total of around 76,000 km in the country, of which around 17,000 km are complementary roads. Thus, we evaluated the following road networks in RONET: 41 Fundamental Paved Roads (13,600 km) Fundamental Unpaved Roads (12,400 km) Complementary Roads (17,000 km) Unclassified Roads (33,000 km) Total (76,000 km) We assumed that the fundamental unpaved roads are gravel roads and the complementary and unclassified roads are earth roads. We assumed that the paved roads with less than 1,000 vehicles per day are surface treatment roads, while roads with more than 1,000 per day are asphalt concrete roads. Road Network Traffic The AfDB report presents the traffic counted on 30 points on the main roads. There is no traffic data per road section. I computed the histogram of the traffic given on the 30 traffic counting points and got the following histogram. We used the results of the histogram to subdivide the paved roads by traffic ranges. Fundamental Unclassified and Municipal complementary % Fundamental Unpaved Daily Traffic (AADT) Paved % From To (%) 0 10 0% 0% 0% 10 30 0% 0% 25% 30 100 3% 25% 50% 100 300 27% 50% 25% 300 1000 23% 25% 0% 1000 3000 37% 0% 0% 3000 10000 10% 0% 0% 10000 30000 0% 0% 0% 30000 100000 0% 0% 0% For gravel roads (fundamental unpaved network), I use the following percentages to subdivide the gravel roads by traffic ranges Traffic (AADT) Percent Range (%) <30 0% 30-100 25% 100-300 50% 300-1000 25% >1000 0% For earth roads (tertiary network), We use the following percentages to subdivide the earth roads by traffic ranges 42 Traffic (AADT) Percent Range (%) <10 0% 10-30 25% 30-100 50% 100-300 25% >300 0% Road Network Condition The Excel file received with the road network data has data for 19,037 km summarized below. Road Number Paved Unpaved Total 100 1,301 301 1,602 105 492 492 110 423 468 891 120 926 519 1,445 140 1,166 268 1,434 160 238 1,464 1,702 170 1,471 1,471 180 1,041 269 1,310 190 795 795 195 247 247 210 291 183 474 220 277 159 436 225 1,009 1,009 230 1,170 1,170 240 849 849 245 224 224 250 1,305 1,305 260 912 912 280 1,269 1,269 Total 12,893 6,144 19,037 The Excel file has road condition data for 9,885 km of paved roads (there are 3,008 km of paved roads with no condition data). It is not clear the year of the road condition data. The road condition data available is given on the table below. Paved Roads Unpaved Very Very No Province Good Good Fair Poor Poor data Total Roads Total Bengo 253 253 159 412 Benguela 403 136 262 801 252 1,053 Bíe 284 213 214 112 823 118 941 Cabinda 0 116 116 Cuando Cubango 77 536 613 1,451 2,064 Cuanza Norte 90 263 134 487 487 Cuanza Sul 241 705 123 0 1,069 103 1,172 Cunene 37 117 154 332 486 Huambo 183 332 99 614 0 614 43 Huíla 789 208 997 195 1,192 Luanda 184 155 339 339 Luanda Norte 247 247 247 Lunda Norte 165 165 289 454 Lunda Sul 281 292 584 1,157 442 1,599 Malange 314 95 444 853 638 1,491 Moxico 975 80 1,233 2,288 1,576 3,864 Namibe 386 386 185 571 Uíge 247 515 146 908 288 1,196 Zaire 226 291 222 739 739 Total 226 4,026 3,530 1,274 829 3,008 12,893 6,144 19,037 We used this knowledge to subdivide the paved road network per road condition classes assuming the same percentages as the ones for the 9,885 km with road condition data. Thus, we have for the 13,600 km paved roads Very Good 2% Good 41% Fair 36% Poor 13% Very Poor 8% We don’t have road condition data for the gravel roads, Thus, I assumed the following for the 12,400 km of gravel roads. Very Good 0% Good 15% Fair 25% Poor 35% Very Poor 25% We don’t have road condition data for the earth roads, Thus, I assumed the following for the 50,000 km of complementary and unclassified earth roads. Very Good 0% Good 5% Fair 15% Poor 45% Very Poor 35% These assumption yield for the fundamental network (paved and gravel) a total of 40% in poor or very poor condition. The AfDB report indicates that 45% of the national roads are in critical or bad condition (I assume that national roads refer to the fundamental network). We have the following for the fundamental network of 26,000 km: Very Good 1% Good 29% 44 Fair 31% Poor 23% Very Poor 16% These assumption yield for the total network a total of 60% in poor or very poor condition. We have the following for the total road network of 76,000 km: Very Good 0% Good 13% Fair 20% Poor 38% Very Poor 28% Road Network Condition and Traffic RONET requires to know the distribution of the road network length per road condition and traffic level combined. For Angola, we have the assumed distribution of the road network length per road condition separate from per traffic level. Thus, to use RONET in Angola, an important simplification was done that assumes the same traffic distribution applies for each road condition class. Thus, we have the following network evaluated with RONET. Fundamental Paved Asphalt Mix Condition Very Very (IRI) Good Good Fair Poor Poor Traffic (AADT) 2 3 4.5 8 12 Total Traffic I <300 0.0 0.0 0.0 0.0 0.0 0.0 Traffic II 300-1000 0.0 0.0 0.0 0.0 0.0 0.0 Traffic III 1000-3000 100.6 2,063.1 1,811.5 654.2 402.6 5,032.0 Traffic IV 3000-10000 27.2 557.6 489.6 176.8 108.8 1,360.0 Traffic V >10000 0.0 0.0 0.0 0.0 0.0 0.0 Total 127.8 2,620.7 2,301.1 831.0 511.4 6,392.0 Fundamental Paved Surface Treatment Condition Very Very (IRI) Good Good Fair Poor Poor Traffic (AADT) 3 4 5.5 9 13 Total Traffic I <300 81.6 1,672.8 1,468.8 530.4 326.4 4,080.0 Traffic II 300-1000 62.6 1,282.5 1,126.1 406.6 250.2 3,128.0 Traffic III 1000-3000 0.0 0.0 0.0 0.0 0.0 0.0 Traffic IV 3000-10000 0.0 0.0 0.0 0.0 0.0 0.0 Traffic V >10000 0.0 0.0 0.0 0.0 0.0 0.0 Total 144.2 2,955.3 2,594.9 937.0 576.6 7,208.0 Fundamental Unpaved 45 Gravel Condition Very Very (IRI) Good Good Fair Poor Poor Traffic (AADT) 5 7 11 16 20 Total Traffic I <30 0.0 0.0 0.0 0.0 0.0 0.0 Traffic II 30-100 0.0 465.0 775.0 1,085.0 775.0 3,100.0 Traffic III 100-300 0.0 930.0 1,550.0 2,170.0 1,550.0 6,200.0 Traffic IV 300-1000 0.0 465.0 775.0 1,085.0 775.0 3,100.0 Traffic V >1000 0.0 0.0 0.0 0.0 0.0 0.0 Total 0.0 1,860.0 3,100.0 4,340.0 3,100.0 12,400.0 Complementary and Unclassified Earth Condition Very Very (IRI) Good Good Fair Poor Poor Traffic (AADT) 7 9 13 18 22 Total Traffic I <10 0.0 0.0 0.0 0.0 0.0 0.0 Traffic II 10-30 0.0 625.0 1,875.0 5,625.0 4,375.0 12,500.0 Traffic III 30-100 0.0 1,250.0 3,750.0 11,250.0 8,750.0 25,000.0 Traffic IV 100-300 0.0 625.0 1,875.0 5,625.0 4,375.0 12,500.0 Traffic V >300 0.0 0.0 0.0 0.0 0.0 0.0 Total 0.0 2,500.0 7,500.0 22,500.0 17,500.0 50,000.0 46 RONET Results Network Monitoring Indicators The resulting network monitoring indicators are given below. Network Monitoring Indicators Monitoring Indicator F. Paved F. Unpaved Complementary Unclassified Overall Network Length Road network length km 13,600 12,400 17,000 33,000 76,000 Road network length that is unpaved km 12,400 17,000 33,000 62,400 Road network length that is paved km 13,600 13,600 Road network length that is paved % 100.0% 17.9% Network Density Road network per thousand land area km/1000 sq km 10.91 9.95 13.64 26.47 60.96 Road network per thousand total population km/1000 persons 0.45 0.41 0.56 1.09 2.50 Road network per thousand rural population km/1000 persons 0.81 0.74 1.01 1.97 4.54 Road network per thousand vehicles km/1000 vehicles 17.00 15.50 21.25 41.25 95.00 Road network per $ million GDP km/million $ 0.13 0.12 0.16 0.31 0.72 Paved road network per thousand land area km/1000 sq km 10.91 10.91 Paved road network per thousand total population km/1000 persons 0.45 0.45 Paved road network per thousand rural population km/1000 persons 0.81 0.81 Paved road network per thousand vehicles km/1000 vehicles 17.00 17.00 Paved road network per $ million GDP km/million $ 0.13 0.13 Network Condition Percentage of road network in good and fair condition % 79.0% 40.0% 20.0% 20.0% 33.8% Percentage of unpaved road network in good and fair condition % 40.0% 20.0% 20.0% 24.0% Percentage of paved road network in good and fair condition % 79.0% 79.0% Percentage of paved road network with roughness 4 m/km IRI or less % 43.0% 43.0% Paved roads average roughness weighted by km IRI, m/km 5.42 5.42 Paved roads average roughness weighted by vehicle-km IRI, m/km 5.46 5.46 Network Access Percentage of unpaved roads that are all-weather roads % 40.0% 7.9% All-weather roads area of influence (4 km wide) as a share of per land area % 1.6% 6.0% Network Standards Percentage of unpaved roads with 30 AADT or less % 25.0% 6.8% Percentage of unpaved roads with 300 AADT or more % 25.0% 25.0% 18.2% Percentage of paved roads with 300 AADT or less % 30.0% 30.0% Percentage of paved roads with 10,000 AADT or more % 0.8% 0.8% Network Utilization Annual motorized vehicle utilization million vehicle-km 8,476 1,262 543 2,951 13,232 Annual freight carried over road network million ton-km 15,939 2,372 1,021 5,549 24,882 Annual passengers carried over road network million pass-km 30,259 4,504 1,938 10,535 47,237 Average network annual average daily traffic vehicles/day 1,708 279 88 245 477 Network Safety Annual number of fatalities persons 848 126 54 295 1,323 Annual number of serious injuries persons 8,476 1,262 543 2,951 13,232 Annual number of casualties persons 9,324 1,388 597 3,246 14,555 Annual casualties cost million $ 723 107.68 46.34 251.87 1,129.32 Annual casualties cost as a share of GDP % 0 0.1% 0.0% 0.2% 1.1% Annual number of fatalities per total population #/100,000 persons 3 0.42 0.18 0.97 4.36 Network Asset Current road asset value million $ 7,867.1 1,148.2 935.7 1,297.4 11,248.4 Current road asset value as a share of maximum road asset value % 79.7% 57.9% 68.8% 68.8% 74.5% Current road asset value as a share of GPD % 7.4% 1.1% 0.9% 1.2% 10.6% Evaluation of Paved Fundamental Road Network (13,600 km) Paved Network Required Expenditures Without Budget Constraints (Optimal Scenario) The table below presents the required expenditures without budget constraints over the next five years and in years 6 to 20, in US$ million, for the paved fundamental network of 13,600km. 47 Optimal Scenario Years 1-5 Years 6-20 Road Agency Costs (M$) Road Agency Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 2,267.0 0.0 2,267.0 100% F. Paved 2,373.8 0.0 2,373.8 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 2,267.0 0.0 2,267.0 100% Total 2,373.8 0.0 2,373.8 100% Percent 100% 0% 100% Percent 100% 0% 100% Rehabilitation Costs (M$) Rehabilitation Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 1,185.8 0.0 1,185.8 100% F. Paved 487.3 0.0 487.3 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 1,185.8 0.0 1,185.8 100% Total 487.3 0.0 487.3 100% Percent 100% 0% 100% Percent 100% 0% 100% Periodic Maintenance Costs (M$) Periodic Maintenance Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 933.7 0.0 933.7 100% F. Paved 1,479.7 0.0 1,479.7 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 933.7 0.0 933.7 100% Total 1,479.7 0.0 1,479.7 100% Percent 100% 0% 100% Percent 100% 0% 100% Recurrent Maintenance Costs (M$) Recurrent Maintenance Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 147.5 0.0 147.5 100% F. Paved 406.9 0.0 406.9 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 147.5 0.0 147.5 100% Total 406.9 0.0 406.9 100% Percent 100% 0% 100% Percent 100% 0% 100% The table below presents the required expenditures without budget constraints over the next five years and in years 6 to 20, in US$ million per year. Over the next five years, US$ 453 million per years is needed for rehabilitation, periodic maintenance and routine maintenance of the paved network. For rehabilitation works US$ 237.2 million per year is needed. For periodic maintenance works US$ 186.7 million per year is needed. For recurrent maintenance works US$ 29.5 million per year is needed. 48 Optimal Scenario Years 1-5 Years 6-20 Road Agency Costs (M$/year) Road Agency Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 453.4 0.0 453.4 100% F. Paved 158.3 0.0 158.3 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 453.4 0.0 453.4 100% Total 158.3 0.0 158.3 100% Percent 100% 0% 100% Percent 100% 0% 100% Rehabilitation Costs (M$/year) Rehabilitation Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 237.2 0.0 237.2 100% F. Paved 32.5 0.0 32.5 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 237.2 0.0 237.2 100% Total 32.5 0.0 32.5 100% Percent 100% 0% 100% Percent 100% 0% 100% Periodic Maintenance Costs (M$/year) Periodic Maintenance Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 186.7 0.0 186.7 100% F. Paved 98.6 0.0 98.6 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 186.7 0.0 186.7 100% Total 98.6 0.0 98.6 100% Percent 100% 0% 100% Percent 100% 0% 100% Recurrent Maintenance Costs (M$/year) Recurrent Maintenance Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 29.5 0.0 29.5 100% F. Paved 27.1 0.0 27.1 100% F. Unpaved 0.0 0.0 0.0 0% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 29.5 0.0 29.5 100% Total 27.1 0.0 27.1 100% Percent 100% 0% 100% Percent 100% 0% 100% Paved Network Comparison of Budget Scenarios The table below presents the rehabilitation, periodic maintenance and routine maintenance needs over the next five years, in US$ million per year, of the different budget scenarios evaluated. Scenarios Optimal +1 and Optimal +2 require more expenditures than the optimal scenario, which maximizes the network NPV. The other scenarios represent different levels of budget constraints. 49 Annual Road Agency Costs Years 1-5 (Annual Costs Years 1-5) Annual Costs Years 1-5, M$/year Network Scenario Rehabilitation Periodic Maint. Recurrent Maint. Road Agency Total Optimal +2 334.6 244.5 52.4 631.5 Network Optimal +1 279.4 244.5 48.1 572.0 Optimal 237.2 186.7 29.5 453.4 Optimal -1 237.2 19.6 25.7 282.4 Optimal -2 150.8 19.6 18.2 188.5 Optimal -3 150.8 0.0 17.2 168.1 Do Minimum 127.5 0.0 16.4 143.9 Do Nothing 0.0 0.0 0.0 0.0 Custom 334.6 243.0 60.3 637.9 The table below show the comparison of the budget scenarios in terms of present value of agency costs at 6% discount rate. Present Value of Society Costs Present Value Years 1 to 20 at 6 percent (M$) Network Scenario Road Agency Road Users Society Total Optimal +2 5,560 37,908 43,468 Network Optimal +1 4,375 38,235 42,610 Optimal 3,236 38,788 42,024 Optimal -1 2,190 41,208 43,397 Optimal -2 1,490 41,819 43,309 Optimal -3 1,265 42,935 44,200 Do Minimum 1,202 43,354 44,556 Do Nothing 0 45,995 45,995 Custom 5,451 38,127 43,578 The table below show the resulting NPV of the different budget scenarios. Present Value Society Net Benefits Compared to Do Minimum Standard (NPV) Society Costs Net Benefit Net Benefit Network Scenario (M$) (M$) (M$/year) Total Optimal +2 43,468 1,087 54.4 Network Optimal +1 42,610 1,946 97.3 Optimal 42,024 2,532 126.6 Optimal -1 43,397 1,159 57.9 Optimal -2 43,309 1,247 62.3 Optimal -3 44,200 355 17.8 Do Minimum 44,556 0 0.0 Do Nothing 45,995 -1,439 -71.9 Custom 43,578 978 48.9 A chart of the network NPV compared to the present value of agency costs of the different budget scenarios is given below. 50 Network NPV X Present Value of Agency Cost (US$ Million) 3000 Optimal 2500 Optimal +1 2000 1500 Optimal -2 Optimal +2 Optimal -1 1000 Optimal -3 500 0 0 1000 2000 3000 4000 5000 6000 The table below show the average roughness of the network under the different budget scenario. The Optimal scenario will keep the network at around the same condition as today at around 5.2 IRI for the next 10 years. The budget constraints scenarios will increase the average network roughness. Roughness Weighted by Km Roughness by Km (IRI, mm/km) Network Scenario Current Year 5 Year 10 Year 20 Total Optimal +2 5.4 3.9 4.4 2.8 Network Optimal +1 5.4 4.2 4.5 5.2 Optimal 5.4 4.6 5.2 6.8 Optimal -1 5.4 5.1 6.5 7.8 Optimal -2 5.4 5.5 6.5 9.2 Optimal -3 5.4 5.6 7.1 9.5 Do Minimum 5.4 5.7 7.2 9.5 Do Nothing 5.4 6.7 8.2 12.0 Custom 5.4 3.9 3.9 3.0 51 Evaluation of All Fundamental Road Network (26,000 km) Fundamental Network Required Expenditures Without Budget Constraints (Optimal Scenario) The table below presents the required expenditures without budget constraints over the next five years and in years 6 to 20, in US$ million, for the fundamental network of 26,000km. Optimal Scenario Years 1-5 Years 6-20 Road Agency Costs (M$) Road Agency Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 2,267.0 0.0 2,267.0 69% F. Paved 2,373.8 0.0 2,373.8 57% F. Unpaved 0.0 1,025.0 1,025.0 31% F. Unpaved 0.0 1,786.4 1,786.4 43% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 2,267.0 1,025.0 3,292.0 100% Total 2,373.8 1,786.4 4,160.2 100% Percent 69% 31% 100% Percent 57% 43% 100% Rehabilitation Costs (M$) Rehabilitation Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 1,185.8 0.0 1,185.8 61% F. Paved 487.3 0.0 487.3 100% F. Unpaved 0.0 744.0 744.0 39% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 1,185.8 744.0 1,929.8 100% Total 487.3 0.0 487.3 100% Percent 61% 39% 100% Percent 100% 0% 100% Periodic Maintenance Costs (M$) Periodic Maintenance Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 933.7 0.0 933.7 85% F. Paved 1,479.7 0.0 1,479.7 51% F. Unpaved 0.0 168.6 168.6 15% F. Unpaved 0.0 1,449.3 1,449.3 49% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 933.7 168.6 1,102.3 100% Total 1,479.7 1,449.3 2,928.9 100% Percent 85% 15% 100% Percent 51% 49% 100% Recurrent Maintenance Costs (M$) Recurrent Maintenance Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 147.5 0.0 147.5 57% F. Paved 406.9 0.0 406.9 55% F. Unpaved 0.0 112.4 112.4 43% F. Unpaved 0.0 337.1 337.1 45% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 147.5 112.4 259.9 100% Total 406.9 337.1 744.0 100% Percent 57% 43% 100% Percent 55% 45% 100% 52 The table below presents the required expenditures without budget constraints over the next five years and in years 6 to 20, in US$ million per year. Over the next five years, US$ 658 million per years is needed for rehabilitation, periodic maintenance and routine maintenance of the total network, of which 69% should go to the paved roads. Optimal Scenario Years 1-5 Years 6-20 Road Agency Costs (M$/year) Road Agency Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 453.4 0.0 453.4 69% F. Paved 158.3 0.0 158.3 57% F. Unpaved 0.0 205.0 205.0 31% F. Unpaved 0.0 119.1 119.1 43% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 453.4 205.0 658.4 100% Total 158.3 119.1 277.3 100% Percent 69% 31% 100% Percent 57% 43% 100% Rehabilitation Costs (M$/year) Rehabilitation Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 237.2 0.0 237.2 61% F. Paved 32.5 0.0 32.5 100% F. Unpaved 0.0 148.8 148.8 39% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 237.2 148.8 386.0 100% Total 32.5 0.0 32.5 100% Percent 61% 39% 100% Percent 100% 0% 100% Periodic Maintenance Costs (M$/year) Periodic Maintenance Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 186.7 0.0 186.7 85% F. Paved 98.6 0.0 98.6 51% F. Unpaved 0.0 33.7 33.7 15% F. Unpaved 0.0 96.6 96.6 49% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 186.7 33.7 220.5 100% Total 98.6 96.6 195.3 100% Percent 85% 15% 100% Percent 51% 49% 100% Recurrent Maintenance Costs (M$/year) Recurrent Maintenance Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 29.5 0.0 29.5 57% F. Paved 27.1 0.0 27.1 55% F. Unpaved 0.0 22.5 22.5 43% F. Unpaved 0.0 22.5 22.5 45% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 29.5 22.5 52.0 100% Total 27.1 22.5 49.6 100% Percent 57% 43% 100% Percent 55% 45% 100% 53 Fundamental Network Comparison of Budget Scenarios The table below presents the rehabilitation, periodic maintenance and routine maintenance needs over the next five years, in US$ million per year, of the different budget scenarios evaluated. Scenarios Optimal +1 and Optimal +2 require more expenditures than the optimal scenario, which maximizes the network NPV. The other scenarios represent different levels of budget constraints. Annual Road Agency Costs Years 1-5 (Annual Costs Years 1-5) Annual Costs Years 1-5, M$/year Network Scenario Rehabilitation Periodic Maint. Recurrent Maint. Road Agency Total Optimal +2 520.0 265.6 78.6 864.2 Network Optimal +1 461.3 270.8 73.0 805.1 Optimal 386.0 220.5 52.0 658.4 Optimal -1 369.9 48.0 53.2 471.1 Optimal -2 250.0 82.8 50.5 383.3 Optimal -3 206.6 71.7 25.8 304.1 Do Minimum 127.5 0.0 21.1 148.6 Do Nothing 0.0 0.0 0.0 0.0 Custom 461.7 327.3 89.0 878.0 The table below show the comparison of the budget scenarios in terms of present value of agency costs at 6% discount rate. Present Value of Society Costs Present Value Years 1 to 20 at 6 percent (M$) Network Scenario Road Agency Road Users Society Total Optimal +2 7,707 44,270 51,977 Network Optimal +1 6,506 44,655 51,161 Optimal 5,178 45,272 50,450 Optimal -1 4,033 48,255 52,288 Optimal -2 3,164 50,245 53,409 Optimal -3 2,408 53,345 55,753 Do Minimum 1,798 55,904 57,702 Do Nothing 0 59,295 59,295 Custom 7,340 45,694 53,034 The table below show the resulting NPV of the different budget scenarios Present Value Society Net Benefits Compared to Do Minimum Standard (NPV) Society Costs Net Benefit Net Benefit Network Scenario (M$) (M$) (M$/year) Total Optimal +2 51,977 5,725 286.3 Network Optimal +1 51,161 6,541 327.1 Optimal 50,450 7,252 362.6 Optimal -1 52,288 5,415 270.7 Optimal -2 53,409 4,293 214.7 Optimal -3 55,753 1,949 97.5 Do Minimum 57,702 0 0.0 Do Nothing 59,295 -1,593 -79.7 Custom 53,034 4,668 233.4 54 A chart of the network NPV compared to the present value of agency costs of the different budget scenarios is given below. Network NPV X Present Value of Agency Cost (US$ Million) 8000 Optimal 7000 Optimal +1 Optimal +2 6000 Optimal -1 5000 Optimal -2 4000 3000 Optimal -3 2000 1000 0 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 The table below show the average roughness of the network under the different budget scenario. The optimal scenario will decrease the average roughness from the current 9.7 IRI to 5.0 IRI in ten years. The Optimal -2 scenario will keep the network at around the same condition as today. Roughness Weighted by Km Roughness by Km (IRI, mm/km) Network Scenario Current Year 5 Year 10 Year 20 Total Optimal +2 9.7 4.7 5.0 4.1 Network Optimal +1 9.7 5.2 5.3 5.7 Optimal 9.7 5.7 6.0 6.8 Optimal -1 9.7 7.0 7.7 8.4 Optimal -2 9.7 8.9 9.4 10.9 Optimal -3 9.7 11.0 11.8 13.0 Do Minimum 9.7 13.5 14.2 15.4 Do Nothing 9.7 15.4 16.2 18.2 Custom 9.7 7.4 7.4 6.9 Fundamental Network Required Expenditures to Keep Current Road Condition (Optimal -2 Scenario) The table below presents the required expenditures, to keep the current network condition (Optimal -2 scenario), over the next five years and in years 6 to 20, in US$ million, for the fundamental network of 26,000km. 55 Optimal -2 Scenario Years 1-5 Years 6-20 Road Agency Costs (M$) Road Agency Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 942.7 0.0 942.7 49% F. Paved 1,313.8 0.0 1,313.8 46% F. Unpaved 0.0 973.8 973.8 51% F. Unpaved 0.0 1,538.8 1,538.8 54% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 942.7 973.8 1,916.5 100% Total 1,313.8 1,538.8 2,852.6 100% Percent 49% 51% 100% Percent 46% 54% 100% Rehabilitation Costs (M$) Rehabilitation Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 754.0 0.0 754.0 60% F. Paved 919.0 0.0 919.0 100% F. Unpaved 0.0 496.0 496.0 40% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 754.0 496.0 1,250.0 100% Total 919.0 0.0 919.0 100% Percent 60% 40% 100% Percent 100% 0% 100% Periodic Maintenance Costs (M$) Periodic Maintenance Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 97.9 0.0 97.9 24% F. Paved 168.6 0.0 168.6 14% F. Unpaved 0.0 316.2 316.2 76% F. Unpaved 0.0 1,054.0 1,054.0 86% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 97.9 316.2 414.1 100% Total 168.6 1,054.0 1,222.6 100% Percent 24% 76% 100% Percent 14% 86% 100% Recurrent Maintenance Costs (M$) Recurrent Maintenance Costs (M$) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 90.8 0.0 90.8 36% F. Paved 226.2 0.0 226.2 32% F. Unpaved 0.0 161.6 161.6 64% F. Unpaved 0.0 484.8 484.8 68% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 90.8 161.6 252.3 100% Total 226.2 484.8 710.9 100% Percent 36% 64% 100% Percent 32% 68% 100% 56 Optimal -2 Scenario Years 1-5 Years 6-20 Road Agency Costs (M$/year) Road Agency Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 188.5 0.0 188.5 49% F. Paved 87.6 0.0 87.6 46% F. Unpaved 0.0 194.8 194.8 51% F. Unpaved 0.0 102.6 102.6 54% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 188.5 194.8 383.3 100% Total 87.6 102.6 190.2 100% Percent 49% 51% 100% Percent 46% 54% 100% Rehabilitation Costs (M$/year) Rehabilitation Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 150.8 0.0 150.8 60% F. Paved 61.3 0.0 61.3 100% F. Unpaved 0.0 99.2 99.2 40% F. Unpaved 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 150.8 99.2 250.0 100% Total 61.3 0.0 61.3 100% Percent 60% 40% 100% Percent 100% 0% 100% Periodic Maintenance Costs (M$/year) Periodic Maintenance Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 19.6 0.0 19.6 24% F. Paved 11.2 0.0 11.2 14% F. Unpaved 0.0 63.2 63.2 76% F. Unpaved 0.0 70.3 70.3 86% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 19.6 63.2 82.8 100% Total 11.2 70.3 81.5 100% Percent 24% 76% 100% Percent 14% 86% 100% Recurrent Maintenance Costs (M$/year) Recurrent Maintenance Costs (M$/year) Network Paved Unpaved Total Percent Network Paved Unpaved Total Percent F. Paved 18.2 0.0 18.2 36% F. Paved 15.1 0.0 15.1 32% F. Unpaved 0.0 32.3 32.3 64% F. Unpaved 0.0 32.3 32.3 68% Complementary 0.0 0.0 0.0 0% Complementary 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Urban 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Unclassified 0.0 0.0 0.0 0% Total 18.2 32.3 50.5 100% Total 15.1 32.3 47.4 100% Percent 36% 64% 100% Percent 32% 68% 100% 57