Bamako THE TIME TO ACT IS NOW! An Engine of Growth and Service Delivery URBAN SECTOR REVIEW B Bamako THE TIME TO ACT IS NOW! An Engine of Growth and Service Delivery URBAN SECTOR REVIEW 5 Table of contents Acknowledgements 7 Introduction and Summary 9 Summary of recommendations 13 Bamako is not an engine of growth or service delivery  17 Bamako is not very competitive 18 … and much improvement is needed to make it livable 22 Bamako’s urban development has been fragmented 27 People are not close to jobs or services 31 And low connectivity exacerbates lack of access…  34 …to public services 37 …..and to jobs 38 What factors underlie Bamako’s challenges? 43 Inefficient land markets deter productive investments 44 Urban accessibility is hampered by inefficient land use and transport 50 Institutional fragmentation and lack of investments hinder service delivery 57 Special Spotlight on Solid Waste Management  63 How to unleash Bamako’s potential? 71 Coordinate land-use with connective infrastructure 73 Finance and manage better public service delivery 82 Invest in Urban Institutions – Think outside the Box 90 Conclusion 101 Acknowledgements The Bamako Urban Sector Review was prepared by a team led by Megha Mukim (Se- nior Economist), and consisting of Christian Eghoff (Urban Specialist), Paolo Avner (Economist), Harris Selod (Senior Economist), Alex Chunet (Consultant), Caroline Plançon (Consultant on Land policy), Tristan Reed (Economist), Alexandr Trubetskoy (consultant) and Zie Ibrahima Coulibaly (Senior Infrastructure Specialist). This report was conceptualized, researched and drafted in collaboration with offi- cials and experts from the Ministry of Housing and Urban Planning, the Ministry of Local Governments, the District of Bamako and Communes within in the geo- graphic area of the District of Bamako. The team is incredibly grateful to the Na- tional Institute of Statistics (INSTAT) in Mali for the provision of data, including census and geographical, to the Cellule de Préfiguration de l’Agence Urbaine de Bamako (CPAUB) for sharing studies insights, and, together with La Direction de la Régulation de la Circulation et des Transports Urbains (DRCTU) for providing the Sotrama network. The team also greatly appreciates the continued guidance and feedback received from representatives of sector agencies and organizations who contributed as members of the Review Technical Committee. The team gratefully acknowledges the peer reviews and inputs from these World Bank Group colleagues: Andrea Liverani (Program Leader), Shomik Raj Mehndiratta (Practice Manager) and Ellen Hamilton (Lead Urban Specialist), Sylvie Debomy (Lead Urban Specialist), Peter Ellis (Lead Urban Economist), Johannes Hoogeveen (Lead Economist), Boubacar Sidiki Walbani (Senior Op- erations Officer), Andre Marie Taptue (Economist), Alice Duhaut (Economist), Alexandre Laure (Senior Private Sector Specialist), Cheick Omar Tidiane Diallo (Transport Specialist), Yu Lu (Takashi Riku) (Research Assistant) and Olivier Beguy (Senior Economist). Senior management of the Urban Global Practice and the Country Unit provid- ed guidance and strong support throughout the research, including Meskerem Brhane (Practice Manager), Somik Lall (Lead Economist), Soukeyna Kane (Coun- try Director) and Michel Rogy (Program Leader). The team would like to acknowledge gratefully the Department for International Development/UK AID for helping finance this study through their program on urbanization and spatial development. Connie Kok Shun (Senior Program Assistant), Aoua Sow Toure (Program Assis- tant) and Aissata Diop Diallo (Program Assistant) provided administrative assis- tance. The design of the report is credited to Francis Gagnon. 8 9 Introduction and Summary Cities have many advantages. They allow workers to be closer to jobs, increas- ing opportunities and fueling productivity. They bring people together physically, facilitating the exchange of ideas and bringing about innovations. High densities make it cheaper to provide services efficiently and equitably. As such, many of the benefits of urban life – productivity and livability - are associated with proximity within the city. Fragmented urban development, thus, makes cities less productive and less livable. Many of Mali’s development challenges have a spatial dimension – with Bamako at its core. Bamako and the neighboring commune of Koulikoro were the only net recipients of migrants (as reported by the 2009 Household Census1), while the regions of Ségou, Mopti, Tombouctou and Gao were the largest net contributors of migrants2. While pull factors (such as access to jobs and ser- vices) matter for movement from rural to urban areas, push factors such as poor living conditions, weather, limited economic opportunities and insecurity were also particularly important in Mali. Bamako also has potential to be a regional hub in West Africa - recent analysis conducted for a regional report on urbanization in the Sahel region3 found that inland capitals (Bamako, Ouagadougou and Niamey) would be the biggest gainers from investments to relieve present transportation bottlenecks across West Africa. In fact, of all West African cities, Bamako shows the highest potential for a hub for market access both regionally and globally. While Mali’s rate of urbanization has risen above 40% in the last years, its GDP per capita (US$1971 PPP, constant 2011 USD) remains at a far lower level compared to other developing countries with similar urbanization levels. Mali’s annual urbanization rate at 4.9% per annum ex- ceeds by far the overall population growth rate of 2.9% per annum. Currently 41% urbanized, the country is expected to cross the 48% urbanization mark by 20304. However, urbanization has not been associated with commensurate increases in GDP, especially compared to Asian developing countries – see Figure 1. Besides, in the last decade, urbanization has been accompanied by deindustrialization - man- 1 RGPH -Recensement Général de la Population et de l’Habitat (du Mali) 2 World Bank (2015) The Geography of Poverty in Mali. 3 World Bank (2018) forthcoming Unlocking Productivity and Livability – A Tale of 3 West African Cities. 4 World Resources Institute Data 10 ufacturing is now only 18% of national GDP, down from 28% in 20065. This im- plies that the country has low levels of wealth, fiscal resource, service delivery, and that institutions and systems remain weak – see Greening Africa’s Cities (2017). Regarding Bamako, the population of the district capital city has more than dou- bled over the 2000-2015 period, and the pressure on the city is expected to grow in the future. Figure 1. Change in GDP depending on urbanization, per country 80% Korea, Rep. 60% Urban share of population China 40% Mali Guinea Vietnam 20% Niger 0% 500 10,000 50,000 GDP per capita (PPP, 2011 US) Source: Henderson, V. and Nigmatulina, D. (2016). The Fabric of African Cities: How to Think about Density and Land Use. Draft, April 20, London School of Economics. 5 World Bank National Accounts Data 11 This study focuses on Bamako, the capital of Mali, that dominates the country’s urban landscape. A central premise of policy-making in cities is that the flexibility, practicality, and focus of local governments make them ideal play- ers to understand and respond to the needs of their citizens. Indeed, cities mostly aim their problem-solving at local conditions. In Mali, the economic importance of the capital city cannot be understated – it is the nerve center of the national economy6. If the capital, Bamako, was to be removed, Mali would lose 36% of GDP - see Figure 2. Thus, reforms and investments aimed at tackling urban develop- ment challenges in the capital will have knock-on effects on national economic development. Figure 2. Reduction in gross GDP if the capital city was removed (% of GDP) -47% Burkina Faso without Ouagadougou -42% Côte d'Ivoire without Abidjan -39% Tanzania without Dar es Salaam -37% Madagascar without Antanarivo -37% Guinea without Conakry -36% Mali without Bamako -34% DRC without Kinshasa -33% Zimbabwe without Harare -28% Sierra Leone without Freetown -22% Ghana without Accra -22% Botswana without Gabarone -18% Chad wit hout N'Djaména -17% Niger wit hout Niamey -16% Mozambique without Maputo -15% Nigeria without Lagos Source: Oxford Economics Yet Bamako is neither an engine of growth, nor of service delivery. Despite Bamako’s importance to the national economy, Bamako is failing to make progress on increasing its competitiveness over time, or delivering on urban services for its citizens. Labor productivity, calculated by gross value added per capita, is low and has remained stagnant in the last 15 or so years, compared to the average for 15 cities in Sub-Saharan Africa (see Figure 3). Even more troubling is that urban service delivery in Mali (indicated by an index combining quality of access to water, electricity and sanitation) also continues to lag the Sub-Saharan average, and is not showing signs of catching up - see Figure 4. Thus, Bamako, and by extension, Mali, is failing to reap the benefits usually associated with urban growth. 6 In fact, the primacy of Bamako in Mali’s urban hierarchy has remained fairly stable over the last 30 years. 12 Figure 3. Trend in Competitiveness (GVA per Figure 4. Trend in access to public services in worker) urban areas 4000 100% South Asia 80% Sub-Saharan Africa 3000 15 Sub-Saharan capital cities 60% Mali 2000 Bamako 40% 1000 20% 0 0% 2000 2003 2006 2009 2012 2015 1999 2002 2005 2008 2011 2014 Source: Oxford Economics. Comparator cities: Abidjan, Source: World Development Indicators Accra, Antananarivo, Bamako, Conakry, Dar Es Salaam, Freetown, Gabarone, Harare, Kinshasa, Lagos, Maputo, N’Djaména, Niamey, Ouagadougou Urban development in Bamako has been fragmented, providing an important explanation of the failure to realize the advantages associ- ated with city growth. Much of new urban construction in Bamako has taken place far from existing urban concentrations, exacerbating the challenges of urban accessibility and access to services. The high level of urban fragmentation is fet- tering both - productivity, by preventing opportunities for matching people and jobs - and livability, by driving up the costs of urban infrastructure and service delivery. 13 Summary of recommendations To become an engine of growth and services delivery the focus should be on how Bamako is built and organized spatially. Bamako is grappling with inefficient land markets which have resulted in haphazard urban develop- ment, leading to investments in buildings and infrastructure away from the urban centers. Insufficient investments in cost-effective network infrastructures, includ- ing transport and public services, has further exacerbated urban accessibility. Underlying these drivers, urban planning institutions at the local level are also weak, and urban governments struggle with small mandates and limited control over revenues for development expenditures. To unleash Bamako’s potential, a balanced approach to reforming institutions, putting the right policies in place and investing in infra- structures and attention to implementation will be needed. The main recommendations from the study are summarized below: • Coordinate land use and connective infrastructure: Public land and properties could be used to promote investments, which would help encourage denser economic infrastructure investments within the city. An inventory of public holdings – land, administrative buildings and infrastructure, would be a first step to help set the tone to enable the development of efficient land and property markets. In parallel, efforts should also be made to implement the new legal frameworks put in place to ensure better land tenure and manage- ment, for instance, through the installation of local land commissions to help tackle uncontrolled land conversion taking place in peri-urban areas. To better connect people to opportunities, investments should be made in upgrading road quality and better managing public spaces. And finally, digital technolo- gies should be leveraged to work with communities to better monitor and track accessibility changes. • Finance and manage better public service delivery: To meet the growing needs for investments in public infrastructure, systems should be developed to increase the sources of own-source fiscal and other revenues at the local level, through mechanisms such as increasing tariffs and taxes and national government transfers. At the same time, there should be a focus on reducing inefficiencies in the system – for instance, in the case of solid waste management, better placement of landfills and more opportunities for com- posting would significantly impact public revenues. And lastly, contractual ar- rangements can be organized better to maximize the incentives for public and private service delivery providers to increase access while recovering costs. • Invest in urban institutions: Since much of current and future urban growth will spill-over to neighboring communes, systems and incentives for inter-jurisdictional coordination on planning for investments in infrastructure and service provision will be paramount. At the same time, institutional re- sponsibilities across jurisdictions and between national and local levels need to be better clarified, and capabilities to implement the local administrative remit be reinforced. Citizen engagement and private sector can and should be leveraged as a crucial element of addressing urban infrastructure financing and governance, including using data and technology. 14 Reform Institutions Modify Policies Invest & Implement Coordinate Land Inventory public land, Manage public spaces bet- Implement new legal Use and Connec- buildings and infrastruc- ter, including road rehabil- frameworks, for instance, tive Infrastructure ture itation and maintenance through the installation of local land commissions Finance and man- Develop systems to Reduce inefficiencies in Organize better contrac- age better public increase its own-source the system, for instance in tual arrangements, to service delivery fiscal and other revenues, the case of SWM building maximize incentives for such as through increasing a landfill closer to the city increasing access and cost tariffs, taxes, and national and composting 50% of recovery government transfers the organic waste within the city Invest in Urban Develop systems and Clarify institutional re- Leverage citizen engage- Institutions – Think incentives for inter-juris- sponsibilities and reinforce ment and competitive outside the Box dictional coordination on capabilities at the local markets as a crucial ele- planning for infrastructure level ment of addressing urban investments and service infrastructure financing provision and governance, including using digital technologies What to Prioritize? NOW SOON LATER Inventory public land, build- Assess investment needs and Implement new legal frame- ings and infrastructure identify of potential sources works for land, for instance, of investment funds, including through the installation of local Review past performance for own-source revenue and land commissions of structuring PPPs in solid private capital mobilization waste and identify future Operationalize a metropolitan investments, including fea- Finance catalytic activities, management and coordinating sibility of constructing a new including those to crowd-in entity landfill, nearer to the city private sector ideas Invest in building local gov- Make commuting data pub- Invest in urban mobility sys- ernment capabilities for urban licly available to inform public tems and road construction & planning and inter-jurisdic- and private decision makers, rehabilitation tional coordination including passengers 15 Given the small public purse and several competing and unmet de- mands, smart urban innovations and solutions could also arise from bottom-up actions. New technologies could be leveraged to give citizens more opportunities to have a say in the functioning of their communities. Beyond in- vestments in heavy infrastructure, if the right platforms can be developed, citizens can be empowered to transform the city space. This report also demonstrates how a variety of data could be used for urban innovations: opportunistic data, which is collected for one purpose and then used for another (such as data owned by cellphone companies and then used to understand urban mobility); purpose- ly-sensed data, which is collected using cheap and ubiquitous sensors that can be deployed in public spaces (for instance, to better understand land and building use); and user-generated data, which comes from engaging people through social media platforms or crowdsourcing (for instance, through Open Street Map com- munities to track urban infrastructure investments and use). The Government can play a fundamental role in fostering innovation, using open platforms but also through establishing synergies with the various actors that can make a city smart, such as businesses, innovation hubs and research centers. Bamako already supports accelerators, and can scale up investments in a bottom-up, innovative ecosystem to engage citizens in finding solutions to its urban challenges. The window for coordinated investments in urban infrastructure is narrow. Owing to its post-colonial legacy, Bamako is less chaotic in its ur- ban planning, compared to its East African counterparts. This provides it with an opportunity to make early investments in connective infrastructure, closely synchronized with land-use planning. Bamako’s District Government will need to build coalitions with its neighboring communes, regional development agencies and with the national government, while finding ways to expand its sources of revenues. Urban investments are long lived and path dependent. The time to act is now. 16 te r h a p c 1 17 Bamako is not an engine of growth or service delivery Bamako dominates the urban and economic landscape in Mali, but it is not an engine of growth or of service delivery. Bamako’s economy remains dominated by small firms, that mainly operate in non-tradable sectors, and which lack scale and specialization. Informal employment is prominent. Even though the city attracts the most qualified migrants from the country, the lack of skilled labor remains a constraint to firms. Bamako is also not a very livable city. Although ac- cess to public services has improved somewhat, access to important services, such as water and waste collection, lags Sub-Saharan African cities. Traffic congestion also poses a big burden on livability. Much of these outcomes are related to Bamako’s fragmented spatial form. Historically most urban development took place on the left bank of the city, although new growth has been on the right bank – and adequate investments in infrastructure and service delivery have failed to keep up. While jobs are concen- trated in or next to the city center, residential density is distributed across the city. Given the lack of urban accessibility, this has resulted in a spatial mismatch. In- deed, accessibility in Bamako has been doubly penalized – making it more expen- sive and inefficient to provide urban services equitably across the city, and placing a greater burden on peripheral households in terms of urban mobility. Accessi- bility is limited owing to a combination of a fragmented urban area where jobs and people are far from one another, but also exacerbated by poor quality of the connective infrastructure, including an insufficient coverage of public transport. 18 Bamako is not very competitive Jobs in Bamako are concentrated in non-tradable sector – repair and retail account for around 1/3rd of total employment. According to the 2009 cen- sus, only 16% of workers are employed in manufacturing and construction. Public administration and other non-tradable services (transport, education, domestic work) account for the remainder. Indeed, location quotients7 illustrate that Ba- mako is specializing in mostly non-tradeable service sectors such as real estate, domestic workers, hotels and restaurants and retail. This is worrying since global8 and Africa-specific regional9 research reiterates the importance of tradeable sec- tors in driving urban economic competitiveness. In Bamako, it seems like a large proportion of the workforce are employed in local activities, which have limited scope for growth and job creation as they are held back by their production struc- tures and by the size of the local market. The two business registries considered in this study, 2012 and 2015, include respectively 5,786 and 8,808 firms for the Bamako urban area. The registries include all formal firms. Compared to the census, mining and banking services are x bo 1 Business Registry (2012, 2015) The two business registries considered in this First, there is considerable increase in the number study, 2012 and 2015, include respectively 5,786 of entries between 2012 and 2015 which might and 8,808 firms for the Bamako urban area. The be owing to differences in methodologies used registries include all formal firms. Compared and the geographical coverage, affecting any to the census, mining and banking services are inter-temporal analysis. Second, the wage bill somewhat overrepresented within the registries. is reported at the level of the firm (and not the However, if the top 5% of firms are excluded, the plant). The top 5% of firms with the biggest wage decomposition is much more similar. Since the bills were dropped from the analysis, as these data was geo-referenced at the neighborhood would run the greatest risk of reporting firm (and level, it feeds into the spatial analysis of economic not plant) level data. In addition, large mining, activity. The data is somewhat limited, and only extractive and security firms were dropped from allows computation of the total number of firms the analysis, since it can be safely assumed that and the total wage bill per neighborhood. How- a large proportion of their operations take place ever, two caveats are worth mentioning upfront. outside of the city. 7 A location quotient (LQ) is an analytical statistic that measures a region’s industrial specialization relative to a larger geographic unit (usually the nation). A LQ is computed as an industry’s share of a regional total for some economic statistic (earnings, GDP by metropolitan area, employment, etc.) divided by the industry’s share of the national total for the same statistic. 8 Competitive Cities for Jobs and Growth (2015), World Bank. 9 Opening Doors to the World (2017), World Bank 19 somewhat overrepresented within the registries. However, if the top 5% of firms are excluded, the decomposition is much more similar. Since the data was geo-ref- erenced at the neighborhood level, it feeds into the spatial analysis of economic activity. The data is somewhat limited, and only allows computation of the total number of firms and the total wage bill per neighborhood. However, two caveats are worth mentioning upfront. First, there is considerable increase in the number of entries between 2012 and 2015 which might be owing to differences in method- ologies used and the geographical coverage, affecting any inter-temporal analysis. Second, the wage bill is reported at the level of the firm (and not the plant). The top 5% of firms with the biggest wage bills were dropped from the analysis, as these would run the greatest risk of reporting firm (and not plant) level data. In addition, large mining, extractive and security firms were dropped from the anal- ysis, since it can be safely assumed that a large proportion of their operations take place outside of the city. A caveat to this analysis is the prominence of informal employment in Bamako. About 68% of the employed population works within the informal sector both in Bamako and other urban areas. Besides, there is a strong effect of income and education as only 34.6% of workers with secondary education and 55.7% of workers of the fifth quintile work for an informal company at national level10. Considering that the business register datasets analyzed only include for- mal firms, there is an inherent bias to the analysis. However, considering the im- portance of the formal sector and the higher productivity of formal companies, the analysis remains highly relevant to the assessment of Bamako’s economic activity. Bamako’s economy consists of predominantly small firms. For instance, sectors such as business support services and retail in non-specialized shops, make up more than half of the total number of firms, and yet account for less than 20% of the total aggregated payroll – see Figure 5. Industrial sectors firms - plas- tic, steel and chemicals production – also represent a non-trivial proportion (7%) of total payroll while only around 1% of total firms. Research from other urban, developing country, contexts11 suggest that sustained job creation over time is cre- ated by medium and large enterprises. In fact, since much of job creation by small firms is usually volatile (with job destruction often matching job creation) and of much lower quality – the predominance of small firms in Bamako is worrying. 10 EMOP (2014), Enquête Modulaire et Permanente auprès des ménages 11 See Rijkers, B et al (2014) “Which firms create the most jobs in developing countries? Evidence from Tunisia,” Labour Economics, Elsevier, vol. 31(C), pages 84-102; and see Mukim, M (2016) “How do Cities in Ethiopia create Jobs?”, World Bank Research Working Paper #7866. 20 Figure 5. Share in total number of firms and total wage bill per sector Sector's share in total wage bill (%) 14% Business services 12% 10% Banking Construction 8% Non-specialized retail Industry 6% Electricity Transport 4% ICT Mining 2% 0% 0% 5% 10% 15% 20% 25% 30% 35% 40% Sector's share in the total number of firms (%) Source: Business Registry 2015 Bamako attracts some of the most qualified migrants, and yet the lack of skilled labor is one of the most important constraints faced by firms in the city. According to the 2009 Census, 79 percent of internal migrants with a tertiary education end up in Bamako; by contrast only 55 percent of migrants with no education go to Bamako. And yet, according to WBG enterprise surveys, while only 11.3% of firms considered the lack of skilled labor to be a major constraint in 2010, 37% designated it as a major or very severe constraint in 2016. Employers in the formal and the informal sector complain of difficulties in finding individuals with the rights skills, including basic technical skills. One of the main determi- nants of skilled labour scarcity is low enrolment rate in primary and secondary education. More than 70% of individuals between 15 and 25 were out of school in 2010, with very limited qualifications and skills for employability which leads them to generally low productivity jobs in the agricultural or service sector (over 90% of all jobs)12. Access and usage of information and communications technology in Bamako also remains limited – in terms of extent and quality of cov- erage – limiting opportunities for digital innovation. While there has been a significant in access to internet over the last years reaching 11% in 2016, a huge gap remains with the rest of Sub-Saharan Africa for which average access is at 20%. Internet penetration in Bamako has historically been one of the lowest in West Africa as only 9.5% of households had access to internet in 2013 which can be linked to high cost and poor infrastructure. While 9 in 10 adults in the city had access to a mobile phone in their household, usage rarely goes beyond texting and phone calls13. It is also worth noting that personal mobile phone ownership is sys- tematically higher among men (88.6%) compared to women (70.7%), and higher 12 World Bank (2014), Skills development and youth employment project, Education Central and West Africa Department 13 Gallup (2013), Media use in Bamako, Mali 21 among individuals with secondary education (91.6%) compared to non-educated people (74.3%). Economic activity is highly concentrated in a few neighborhoods. Figure 6 presents the distribution of wages and the distribution of firms across neighborhoods in Bamako. Hamdallaye ACI, which is not displayed, accounts for about 12% of firms and 18% of total wages paid. In general places that host more firms also account for a larger share of the wages paid. The Centre Commercial is an exception, which clearly seems to host far more, possibly smaller firms, which account for a small share of the wage bill. The mirror exception is the industrial zone, which hosts far fewer firms (mostly in chemical, steel and plastic produc- tion), but accounts for more than its fair share of wages. Bamako needs far more firms with scale and the right specialization. Although political instability and inadequate access to finance have been high- lighted as big impediments by firms, almost half of firms identify transportation as an issue (twice the proportion compared to other Sub-Saharan countries) and around 41% identify access to labor as a constraint. This requires the right ‘habi- tat’ for firms, and goes far beyond the national regulatory environment. Figure 6. Distribution of Firms and Wages (without Hamdallaye-ACI) Neighborhoods's share 16% in total wage bill Hamdallaye-ACI 14% 12% 10% 8% Faladie 6% Badalabougou Hippodrome Zone Indu. Baco-djikoroni 4% Korofina Niamakoro Centre commercial Torokorobougou Kalaban-coura Sotuba 2% Sogoniko Magnambougou Banankabougou Kalabancoro Lafiabougou Djelibougou 0% -2% 0% 2% 4% 6% 8% 10% 12% Neighborhood's share in total number of firms -2% Note: Bubble Size: Neighborhood Population (Global Human Settlement Layer, GHSL, 2015) Source: Business Registry 2015, GHSL 2015 22 … and much improvement is needed to make it livable Access to public services has improved between 1996 and 2009, how- ever some services, such as access to tap water, remain limited – see Figure 7. While only 36% of households were connected to the electrical grid in 1996, almost 75% benefitted of the service in 2009. Similarly, access to tap water jumped from 25% to 45% on the same period, which however remains relatively low. Finally, while waste management was not considered in the 1996 census, the 2009 figures display a relatively high collection rate with 70% of households served either by public or private companies14. Although collection rates are high, proper disposal lags – this will be discussed in detail in later chapters. Figure 7. Services decomposition Sources of water Sources of electricity Waste collection 1996 2009 1996 2009 Public Public grid 60% collection 70% 70% 50% 60% 60% 50% 50% 40% Tap water 40% 40% 30% 30% Well 30% 20% Fountain 20% 20% Oil lamp Private Street / collection Nature Dumpsters 10% Pile 10% 10% Drilling Flashlight Gutter 0% Surface water 0% Other 0% 2009 Source: RGPH 2009, 1996 14 Data on solid waste collection used here for consistency. More recent estimates put the level of household collection at 40 to 70 percent depending on neighborhood, see chapter 3. 23 x bo 2 A special focus on solid waste management This study will place a special spotlight on solid collection rates compared to regional benchmarks waste in Bamako. This is for many reasons. Solid (Ouagadougou at 47 percent, Greater Monrovia waste is an important environmental and health at 41 percent, Senegal (country) at 55 percent), concern and studies have shown that inadequate the absence of a sanitary landfill and the very low municipal solid waste management can impact collection rates in the Greater Bamako area (20 negatively on GDP. For example, economic losses percent in the surrounding Region of Koulikoro) resulting from inadequate provision of solid waste is an additional reason to a focus on this service services have been estimated at 0.83 percent of in the present report. Solid waste management GDP in Burkina Faso and 0.5 percent of GDP in also exemplifies some of the coordination failures Morocco. Focusing on solid waste management is related to urban planning and service delivery in thus justified from an economic growth perspec- the Greater Bamako area and is one of the few tive. Furthermore, while according to the 2009 municipal services essentially delivered by local census Bamako fares relatively well in terms of government. 15 World Bank (2017). Burkina Faso. BF-Improving Solid Waste Service Delivery. Report No: ACS22298. Washington DC: World Bank. 16 World Bank (2009), ‘Municipal Solid Waste Sector Development Policy Loan’, Program Docu- ment, P104937, Washington DC: World Bank. 17 World Bank (2017). ‘Cheesemanburg Landfill and Urban Sanitation Project’, Project Appraisal Document, P159961, Washington DC: World Bank. 18 World Bank (2017). ‘Senegal Municipal Solid Waste Management Project’, Project Information Document/Integrated Safeguards Data Sheet, P161477, Washington DC: World Bank. 24 Access to urban services in Bamako lags average access in urban ar- eas in Sub-Saharan African countries. While, 87% of households benefit from improved access to water in Sub-Saharan Africa, it is the case of only 71% of households in Bamako. Similarly, only 28% of households have access to im- proved sanitation compared to 40% in SSA – see Figure 8. Access to electricity compares reasonably well, although this also lags somewhat. These gaps can also be embodied by the higher share of slums in Bamako (63%) compared to the aver- age urban areas of the region (55%)19. Figure 8. Access to services in Bamako, Malian urban areas and Sub-Saharan urban areas (% of population) Bamako SSA - Urban areas Mali - Urban areas 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Water Electricity Sanitat ion Slums Source: UN Habitat and RGPH 2009 19 While the estimates for the Sub-Saharan African regions are based on the UN Habitat definition, the estimate for Bamako is based only on housing quality 25 Traffic congestion is also a huge burden on liveability. The physical form of Bamako is prone to congestion - there has been exponential growth on Rive Droite, and yet, most of the administrations, the higher education estab- lishments, the industrial zone and the main markets remain concentrated on the Rive Gauche, near the city centre. Much of this disconnect is owing to the legacy of urban development, wherein new growth has been on the right bank. The in- creased congestion in Bamako seems to be associated with lower overall mobility rather than longer commutes. While the average daily travel time in Bamako is 62 minutes, which aligns with international norms20, the number of daily trips per person is one of the lowest in the region with only 3 trips a day – see Figure 9. Besides, 53% of residents surveyed had either not travelled at all or only by walk- ing on the previous day compared to only 38% in Douala, 43% in Ouagadougou21. The spatial range of trips in Bamako also seems somewhat geographically limited. In Bamako, more than half of trips (53%) are either intra-district or restricted to close neighbouring districts. Considering the prominence of walking in transport modes, short trips are symptomatic of poverty and traffic congestion in African cities. For example, trips within the same district or neighbouring districts repre- sent 57% of the total in Niamey, 66% in Conakry and 52% in Douala. This points to spatial fragmentation within the city, which is associated with low mobility but higher levels of congestion. Figure 9. Time spent per day and number of trips per person Average travel time (min) Number of daily trips 88 88 66 63 62 58 3,7 4,6 3,8 4,4 3 3,2 Conakry Douala Ouaga- Niamey Bamako Dakar Conakry Douala Ouaga- Niamey Bamako Dakar dougou dougou Source: Lourdes Diaz, O., Plat, D., Pochet, P. (2013). The puzzle of mobility and access to the city in Sub-Saharan Africa. Journal of Transport Geography, Elsevier 20 Schafer, A. & Victor, D.G. (2000). The future mobility of the world population. Transportation Research A 34 (3), 171-205. 21 Lourdes Diaz, O., Plat, D., Pochet, P. (2013). The puzzle of mobility and access to the city in Sub-Saharan Africa. Journal of Transport Geography, Elsevier. 26 Access to electricity in Bamako constitutes an important obstacle to firms’ activity – see Figure 10. Approximately 68% of firms in Mali complain about electricity being a major constraint to their activity, compared to a Sub-Sa- haran average around 40%. This is an even greater constraint for firms in Bamako (72% of firms designate it as major constraint) than in Mopti, Ségou and Sikasso (49% of firms). In fact, the percentage of firms identifying energy as a major con- straint in the capital has doubled between 2010 and 2016. To cope, almost 70% of firms rely on an either a shared or private generator22. The main drivers of this situation are the important urban expansion of Bamako, which would need large investments to connect new areas to the grid, exacerbated by the weakness of the electrical network and production, fueled by the financial state of Energie du Mali. The company suffers from high levels of indebtedness due to low tariffs, a shift towards thermal generation and a deficit in installed capacity23. Figure 10. Percentage of firms designating electricity as a major constraint, ex- periencing electrical outages and owning/sharing a generator Bamako Mopti, Ségou, Sikasso SSA 70% 60% 50% 40% 30% 20% 10% 0% % of firms designating it as a % of firms owning/sharing major constraint generator Source : Enterprise Survey 2016 22 Enterprise Survey 2010 and 2016 23 African Development Bank (2015). Renewable Energy in Africa: Mali Country Profile 27 Bamako’s urban development has been fragmented The population of Bamako has grown exponentially in the last 40 years and is almost equitably distributed between the Rive Gauche and the Rive Droite – see Figure 11. Jumping from under 500 thousand in- habitants in 1990, to more than a million in 2000, the population is estimated at approximately 3.3 million today24. This translated in the urban expansion of the city which grew on both sides of the river Niger and in every direction (except for the south, limited by the exclusion airport area). The population is divided almost equitably between the Rive Gauche and the Rive Droite which represent respec- tively 54% and 46% of the population. Figure 11. Urban expansion of Bamako in the last 30 years 1990 2000 2013 Urban built-up Suburban built-up Source: Atlas of urban expansion, UNHABITAT-NYU, 2017 24 Global Human Settlement Layer 1990, 2000 and 2015 28 The nature of the urban expansion differs between the two city banks – see Figure 12. In the last two decades or so, the urban expansion on the Rive Gauche consisted mainly in extension of the already existing urban built-up areas. However, the urban expansion occurring on the Rive Droite has consisted mainly in leapfrog development, i.e. spatially discontinued urban expansion. Expansion refers to new construction at the edge of the consolidated urban area, leapfrog refers to parcels of newly built land that do not border on existing development, and infill refers to construction on unbuilt parcels surrounded by existing devel- opment. Inclusion refers to urban built-up areas that were previously leapfrog development, but are now within the urban extent. Figure 12. Nature of added areas since 2000 Built-up area 2000 Infill Extension Leapfrog Inclusion Source: Atlas of urban expansion, UNHABITAT-NYU, 2017 29 Population density in Bamako averages 8,200 inhabitants per square kilometer – but this masks huge differences and fragmentation across neighborhoods. For one thing, population density on the Rive Droite is around 9,100 inhabitants per square kilometers which is significantly higher than the Rive Gauche with approximately 7,400 inhabitants per square kilometer. However, pop- ulation density is very unequally distributed across the two banks – see Figure 13. Population density also declines, slowly but steadily, with distance from the center of the city – denoted by the location of the Grand Marché – see Figure 14. Con- cerning the Grand Bamako area, two urban centers stand out, Koulikoro and Kati25, which densities are respectively around 5,750 and 2,900 inhabitants per square kilometer. Figure 13. Surveyed population density per 250 meters squares Figure 14. Population density gradient People per km2 5000 4000 3000 2000 1000 0 1 2 3 4 5 6 7 8 9 10 11 Distance from city center (km) Source: Landscan 2012 and GHSL 2015 25 Kati, being located in the close north-west suburb of Bamako, is considered as being part of the Bamako urban area although it constitutes a separated administrative division. 30 x bo 3 Geographical coverage of the analysis The study uses two different geographical levels ibabougou, N’Gabacoro, Sangarebougou, Diago, for analysis – for the District of Bamako, and Dialakorodji, Dio-Gare, Doubabougou, Kambila, for the commune. The first, i.e. the District-level Kati Commune, Safo, Ouelessebougou, Bougoula, includes the urban area covered by the District Dialakoroba, Sanakoroba, Bancoumana, Siby, (shown by yellow in Figure A) and the suburban Koulikoro Commune, Meguetan, Tienfala areas (show in green). This neighborhood levels allows an analysis of the build-up area of Bama- The communes of Kati and Koulikoro, part ko, which spills into neighboring communes. The of the Greater Bamako area, do not seem analysis at commune level was focused on 25 to have become strong relays to Bamako’s communes that are part of the Greater Bamako development. Kati and Koulikoro are respec- area (see list below) and located in the circles of tively around 35 minutes and 100 minutes from Kati, Koulikoro and the district of Bamako (the Bamako’s city centre. Both have experienced a latter circled in black in Figure B). While the population growth between 1.5% to 2% per year “Greater Bamako” area is often considered to be on the 2000-2015 period which is relatively low the area associated with the administrative circle compared to the district of Bamako (6.2%). Be- of Kati thereby including relative isolated areas sides, housing quality remains significantly worse and excluding closer areas from the Koulikoro in Kati and Koulikoro with, for example, under circle, the definition used here focused on areas 20% of households having good quality roofs that are functionally linked to the Bamako district compared to more than 33% in Bamako (results through economic corridors (primary roads). being similar for floor and wall quality). Regard- ing tenure, the proportion of households with Communes included: Baguineda, Mountougoula, land titles is approximately the same across the Tiele, Dogodouman, Kalabancoro, Mande, Mor- three areas. Figure A. Urban area of Bamako Figure B. Greater or Grand Bamako Ü Ü Koulikoro Circle Kati Circle Koulikoro commune Kati commune Commune 1 Com. 3 Commune 2 Commune 4 Commune 6 Commune 5 District of Bamako (inner-Bamako) Grand Bamako Communes of Bamako 0 12.5 25 50 Kilometers Suburban areas 0 2 4 8 Kilometers 31 People are not close to jobs or services Firms in Bamako are concentrated in or next to the city center. Of the 7,511 companies registered in the 2015 business registry for the urban area of Ba- mako, 66% are located on the Rive Gauche (down from 73% in the 2012 business registry). This divide is due to a very high concentration of firms in the center of the city, around the neighborhood Centre Commercial which displays a density of around 344 companies per square kilometer. The spatial analysis at neighborhood level confirms this pattern (see Figure 15), highlighting 12 neighborhoods with a density higher than 100 companies per square kilometers. Figure 15. Number of HQ per neighborhood Ü Number of HQ below 5 6 to 10 11 to 20 0 2 4 8 Kilometers 21 to 40 above 40 Source: 2015 Business Registry Comparing gradients of firm density and population density confirms the divergence between a very high concentration of firms in the cen- ter of the city and a more equally distributed population, which pro- duces a spatial mismatch – see Figure 16. The “Grand Marché” (main mar- ket) of Bamako was used as the central point for this analysis. While population density decreases progressively when distance to the center increases, job density decreases dramatically in the first four kilometers. The calculation of company headquarters26 in the city also seems to indicate concentration in a few neighbor- hoods, almost all of which are on the Rive Gauche. 26 Proxied by firms accounting for the top 5% of the wage bill. 32 Figure 16. Gradient of firm and population density People per km2 Gradient population density Gradient firms density Firms per km2 5000 250 4500 4000 200 3500 3000 150 2500 2000 100 1500 1000 50 500 0 0 1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11 Distance from city center (km) Distance from city center (km) Source: 2015 Business registry and 2009 Census Bamako is an unequal city – access to urban services, such as water, electricity and waste varies widely. Only 42% of the urban population has access to water, and access is also distributed unequally across the city. Neigh- borhoods with good demographic characteristics – such as better education levels and better housing quality - also benefit from good access to water (i.e. Ham- dallaye, Badalabougou, Hippodrome, Korofina Nord). In addition, only 20% of the population has access to drinkable water in the suburban areas cities of the close suburb and 16% in the communes of the Greater Bamako. While 70% of the urban metropolitan area have direct access to electricity, the access to this service remains very much concentrated in the center of Bamako on the Rive Gauche. Access to electricity diminishes drastically when analyzing the suburban areas (39%) and the communes of the Grand Bamako (27%). Similarly, for waste, while 70% of households are served by either public or private waste collection in the inner-Bamako area, only 31% of households in the suburban areas and 19% in the communes of the Greater Bamako are served. It is important to note that even in the central neighborhoods, for instance those close to the Grand Marché, waste collection remains patchy. Peripheral areas are doubly penalized – they seem to be more highly exposed to tenure insecurity and have lower quality housing. This is particularly true for areas in the North of the city, where the topography is steep- er. On average 26% of households in the urban metropolitan area are affected by high tenure insecurity27. Mapping tenure insecurity at neighborhood level reveals that levels rise as one moves from inner Bamako (21%) to close suburbs (39%) to the Grand Bamako area (43%). The spatial patterns of housing quality are like the ones of tenure insecurity – low housing quality28 is concentrated in peripheral areas, particularly on the Rive Gauche. Again, the share of households with low housing quality increases as one moves further away from the city center. Com- bining information on land tenure and housing quality points towards some of the most advantaged areas (Badalibougou, Hamdallaye, and Magnambougou – high- lighted in blue) and most disadvantaged areas (Banconi, Sébénikoro, and Sabali- bougou – highlighted in red) areas within the cities – see Figure 17. 27 Tenure insecurity is calculated as the share of households which are not tenants and who live in a property without owning land rights. 28 Housing quality is calculated as an average of two variables: roof and floor quality. 33 Figure 17. Most advantaged (green) and least advantaged (yellow) neighborhoods Source: 2009 Census The socio-economic divide is directly related to fragmented form of the city. Contrary to common perfection, the socio-economic divide in Bamako is more complicated than a simple Rive Gauche – Rive Droite divide. Indeed, the Rive Gauche does score lower than the Rive Droite in terms of housing quality and land tenure, but it also benefits from somewhat better access to basic public services – water, electricity and waste collection. In fact, the low scores of the Rive Gauche are very much due to the peripheral neighborhoods in the north (a product of urban extension along the edge of the city) which score especially low in terms of land and housing. The low scores for accessibility of public services for the Rive Droite, on the other hand, seem to be related to the prominence of leap- frog urban development. In addition, there is a strong divide between the urban area of Bamako and the surrounding communes (except for Kati and Koulikoro urban centers), with the latter lagging on all socio-economic outcomes. 34 And low connectivity exacerbates lack of access… Road infrastructure in Bamako is relatively well spread but lacks qual- ity due to insufficient paving. The coverage of arterial roads in Bamako is relatively good, with 80% of the population within walking distance of an arterial road (compared to 68% in Accra, 90% in Addis Ababa and 86% in Kampala)6 – see Figure 18. The average road width in Bamako is 6.5 meters, which slightly above average compared to the main Sub-Saharan African cities (6.1 meters)29. However, paved road density in Bamako lags other cities in Africa – see Table 1 below. Table 1. Road density in Bamako and 6 other comparator cities Length of Share of Paved road density City road network paved roads (km per km2) Accra 1.899 50% 2.8 Douala 1.800 25% 2.4 Conakry 815 32% 2.3 Abidjan 2.042 59% 2.1 Bamako 836 24% 0.8 Ouagadougou 1.827 11% 0.4 Kigali 984 12% 0.2 Source: Kumar and Barrett (2008) Figure 18. Road network of Bamako Road network Primary Secondary Tertiary paved Tertiary unpaved Source: Open Street Map Data 29 Atlas of Urban Expansion (2016), Bamako Mali, UN-Habitat, NYU, Lincoln Institute of Land Policy 35 What constrains accessibility is how the road infrastructure is being used – see Figure 19. In Bamako, roughly 9% of trips are made using a car and 17% using buses. More than half the trips in the city (57%) are made by walking – suggesting a somewhat limited geographical range of movement and thus limited access to the opportunities and services that the city can offer. In many cases this is because they cannot afford any means of motorized transport on a regular basis. While walking to work is good way to commute it also has consequences on how many jobs and services can be reached in each time frame when mode choices are constrained. Assuming a person walks for one hour, they would be able to reach opportunities that lie within a 28-square kilometer radius (assuming a generous 3km/hour speed along a straight line). For instance, in 2015, this would cover only 10% of the urban area of Bamako – see Box 4. x bo 4 Computing accessibility Accessibility across Bamako is computed by day and corresponding speeds. The analysis covers accounting for the road infrastructure (network the inner and outer Bamako road network, i.e. and road type), different modes of transportation beyond the administrative boundaries of the city. (pedestrian, car and public transport), times of the Speeds for network analysis (km/h) Car Car SOTRAMA SOTRAMA Pedestrian rush hour rush hour Primary 60 30 42 21 3 Secondary 40 16 28 11.2 3 Tertiary paved 35 14 24.5 9.8 3 Tertiary unpaved 25 10 17.5 7 3 36 Figure 19. Decomposition of transport modes as share of total trips per city (%) Dakar Conakry Niamey 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Walking Minibus Motorized two-wheeler Car Shared taxi Bicycle Metered taxi Institutional transport Douala Bamako Ouagadougou 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Walking Minibus Motorized two-wheeler Car Shared taxi Bicycle Metered taxi Institutional transport Sources: Olvera et al (2012), Godard (2011)30 Public transport services are lacking - in many ways. The SOTRAMA net- work, which is composed of privately-owned buses traveling on different routes, covers most of the city. However, according to customers the quality of the service is poor in terms of frequency, speed and comfort. There are no designated stops, and frequency of the service is unpredictable with often lengthy wait times (up to an hour for certain lines). Reflecting the density of economic activity within the city, SOTRAMA densities are higher on the Rive Gauche (more than 10 kms per square kilometer) – see Figure 20. Figure 20. SOTRAMA network and associated Kernel density Ü Km of line per km2 below 6 6 to 12 12 to 18 18 to 24 0 2 4 8 Kilometers above 24 Source: Urban agency of Bamako 30 Olvera, L., Plat, D. and Pochet. P. (2012). Mobilité et accès à la ville en Afrique Subsaharienne. Godard, X. (2011b). Sustain able urban mobility in “Francophone” Sub-Saharan Africa’, regional study prepared for the Global Report on Human Settlements 2013 37 …to public services The three existing hospitals in Bamako are concentrated on the Rive Gauche which results in relatively low accessibility for many parts of the city – Figure 21. Only 16% of the urban population can access hospitals with- in a 30-minute trip using public transport during rush hours – 53% during nor- mal hours. Comparing this to accessibility using cars provides some sense of the inequality of access within Bamako – 76% of car users can reach hospitals within 30 minutes during rush hours (99% during normal hours). Peripheral areas have lower accessibility to social infrastructures and, to a lesser extent, to markets – see Figure 22. Other health infrastructures (which include clinics, doctors, attention points but exclude traditional medicine and drugstores) can be assessed by 81% of the urban population within a 30-min- ute walk; 84% considering public transport during rush hours. The accessibility of primary schooling and secondary education is very high with 89% and 84% of the urban population living within a 30 minutes’ walk respectively. 67% of the urban population lives within a 30-minute walking distance of a market; 83% by public transport even during rush hours. What is noticeable is that peripheral areas to the north of the city and the urban centers in suburban areas suffer from the low- est levels of accessibility. Figure 21. Hospitals accessibility across transport modes Pedestrian Sotrama Car Ü Ü Ü 0 1.5 3 6 Kilometers 0 1.5 3 6 Kilometers 0 1.5 3 6 Kilometers Figure 22. Primary and secondary education infrastructure accessibility for pedestrians Other health infrastructure Secondary education Markets Ü Ü Ü 0 1.5 3 6 Kilometers 0 1.5 3 6 Kilometers 0 1.5 3 6 Kilometers Source: Education infrastructure GIS data (Juliette Coulibaly) 38 …..and to jobs Public transport services do not help ameliorate the spatial mismatch between people and jobs. Labor market accessibility is assessed at the neigh- borhood level by examining the share of jobs which can be accessed within a given time frame using various transport modes (Sotrama and cars31). More central neighborhoods have the best accessibility - they are located closer to jobs and they benefit from a denser road and Sotrama network – see Figure 23. Converse- ly, since population density extends far beyond the central city, neighborhoods located further away fare worse on all counts – they have lower road and Sotrama densities. Exacerbating inequality, public transport users suffer from much lower accessibility than private transport users (cars32). Figure 23. Share of accessible jobs using Sotramas (left) and cars (right) for var- ious time thresholds (30 and 60 minutes) during peak hour Sotrama (30 minutes travel) Cars (30 minutes travel) 0 2.5 5 7.5 10 km Public Transport Sotrama Network Road Network primary secondary tertiary Accessibility levels 0% - 10% 10% - 20% Sotrama (60 minutes travel) Cars (60 minutes travel) 20% - 30% 30% - 40% 40% - 50% 50% - 60% 60% - 70% 70% - 80% 80% - 90% 90% - 100% 31 Accessibility metrics will also be computed for pedestrian travel. 32 On average accessibility levels will resemble those provided for Sotramas as equipment rates in Mali are low: 12 vehicles per 1000 inhabitants which can be compared to the world average of 182 and an average for Africa of 42. 39 Congestion on the network seems to be a large determinant of how quickly people can reach potential employment opportunities. Table 2 summarizes the share of jobs accessible on average within the urban area for various travel modes, time thresholds and time of day. Within 30 minutes, with- out congestion, Sotrama users can reach 41% of all jobs on average while the figure drops to 11% during peak hour. Thus, while road infrastructure quality in Bamako has room for improvement, big effects on labor market accessibility could be attained simply by finding ways to increase average public transport network speeds, and reducing congestion during peak hours. Table 2. Average accessibility levels to employment opportunities at the urban area’s level Sotrama Cars Time peak hour off-peak peak hour off-peak 30 mins 6.4% 26.5% 49.5% 99.2% 45 mins 22.9% 60.9% 82.8% 100% 60 mins 45.1% 83.8% 98.6% 100% The costs of congestion are borne unequally, with those using public transport bearing a much heavier burden. While car users can access most infrastructure and jobs within the city no matter where they live, location matters hugely for Sotrama users as they depend entirely on public transport for accessi- bility (see Figure 24). The costs of mobility, and of congestion on the network, are thus borne by poorer residents, especially those living in more peripheral areas. Besides, when accounting for congestion, the GINI coefficient jumps from 0.50 to 0.66 – implying a huge increase in inequitable access. This increase is mainly driven by the fact that congestion lowers the accessibility levels of the people that live far from jobs (or broadly, the city center) while people living close to jobs are less affected. Figure 24 displays the Lorenz curves and the associated Gini coeffi- cient for cars (left) and Sotramas (right) for peak and off-peak travel. Figure 24. Lorenz curve for accessibility for car users (left) and Sotrama users (right) for peak and off-peak travel within 30 min travel Lorenz curves for accessibility using Lorenz curves for accessibility using 100% cars Sotramas Share of 90% accessible 100% Share of opportunities 90% accessible 80% opportunities 70% no 80% congestion 70% 60% Gini=0.01 60% 45 degree 50% line 40% congestion 50% 45 degree Gini=0.39 40% 30% line no congestion 20% 30% Gini=0.50 20% congestion 10% Gini=0.66 0% 10% 0% 20% 40% 60% 80% 100% 0% Share of population 0% 20% 40% 60% 80% 100% Share of population 40 Congestion has spatially differential effects on accessibility between car and Sotrama users. Congestion reduces accessibility for car users who live away from the center – this is logical, since they have longer distances to travel. However, the loss in accessibility owing to congestion is higher for Sotrama users who live closer to the center of the city. This is because Sotrama users who live on the outskirts of the city already have very low accessibility to the city center – and so congestion on the network doesn’t lead to very much more loss of accessibili- ty33. However, if the time threshold of travel is increased to an hour or more, then Sotrama users in all parts of the city are equally affected by congestion on the network. 33 Calculations and associated maps for different time thresholds, available on request. 41 42 te r h a p c 2 43 What factors underlie Bamako’s challenges? Bamako isn’t reaching its potential owing to inefficient working of land markets, low accessibility even with a reasonable road network, and lack of resources for much-needed investments in service deliv- ery. This chapter examines the underlying drivers of the trends identified in the first chapter. Inefficiencies relating to land – access, markets, rights and planning – are a significant deterrent to productive investments within the city, leading to less opportunities for infill development and exacerbating urban fragmentation. At the same time, urban accessibility within Bamako is limited - not because of the lack of road infrastructure - but rather owing to its fragmented urban form, poor quality of the road infrastructure, an insufficient coverage of public transport, the natural bottleneck that the Niger river poses and competition for public space – including road and sidewalk. And lastly, service delivery in Bamako is hindered not only by lack of resources, but also by institutional fragmentation and lack of metropolitan coordination. Local governments lack capabilities, including fiscal, to develop and implement urban plans and invest in urban infrastructure, which in turn, has contributed to ineffective spatial management of the Bamako metro- politan area. 44 Inefficient land markets deter productive investments Dysfunction in land markets is a significant deterrent to productive investment in cities, affecting not just the patterns of land-use and building volumes, but also limiting the ability of local governments to capture the value of land (for instance, through taxation) for economic development investments. In many developing countries, weak property right systems and poor land gover- nance contribute to making access to land costly and insecure. This in turn under- mines the willingness of other actors to invest in the city’s future growth leading to cities that lack urban planning and suffer the consequences of urban sprawl. Because the urban land market in Mali operates under such conditions, it is failing to provide the enabling environment to sustain productive urbanization, likely exerting a significant toll on the economy. These conditions also prevent local gov- ernments to potentially capture the value of land (for instance through taxation) to fund economic development investments. The land tenure system in the Greater Bamako area34 is characterized by several layers of complexity, which deter efficient and equitable urban development. Under the current system, different land tenure regimes (referred to as legal pluralism35) co-exist, beset by poor governance in the man- agement of land at all levels of government, leading to a largely uncoordinated pattern of land-use and land-tenure transformation that has accompanied urban population growth and urban spatial expansion. Since the population of Bama- ko is expected to expand steadily in the coming decades (at a rate of 5.4 percent annually reaching 6 million residents in 2030 for the Bamako District alone, see Ville de Bamako, 2012)36, the pressure on urban land will continue to grow. Currently, agricultural land at the outskirts of the city are being transformed into residential plots at a massive scale (Durand-Lasserve et al., 2015)37, within the unequitable context of weak property rights and unplanned development. The most visible implication of these multiple layers in the legal framework is the coexistence of various land tenure situations. The typology of tenure situations includes the following categories: customary posses- sion of land, non-customary possession of land (often for plots purchased from a customary possessor and held with a certificate of sale), occupation of a plot with 34 We define the Greater Bamako area as the area comprising the District of Bamako and the peri-ur- ban area and rural hinterland in surrounding communes, up to about 80 km away from the city center. 35 Legal pluralism is a typical feature of West African countries which experienced a succession of land regimes that were established through the pre-colonial area, the colonial period, the post-in- dependence socialist transition, and following the market liberalization phase over the past two or three decades. The changes in both political and economic orientations that took place over time added cumulative layers to the legal and regulatory framework that governs land while never completely abolishing the previous system in place. 36 Ville de Bamako. 2012. Bamako 2030: Croissance et Développement – Imaginer des Stratégies Urbaines pour un Avenir Maîtrisé et Partagé. 37 Durand-Lasserve, A., M. Durand-Lasserve and H. Selod. 2015. Land Delivery Systems in West African Cities. The Example of Bamako, Mali, Washington D.C.: The World Bank and Agence Française de Développement, 133 pages 45 an administrative document, occupation of a plot with a use right (concession), and ownership with a title (titre foncier/title deed). These different tenure situa- tions are encountered throughout the Greater Bamako Area, with concessions and ownership titles mostly in the urban core, and customary land or land purchased from customary owners mostly in the peri-urban area or the rural hinterland (a large part of the Greater Bamako area, beyond the 6 communes of the District is agricultural land). The classification and the conversion of rural land to urban land relies not only on different legal provisions more or less clear and effectively implemented regarding land use planning and decentralization legal framework38 but also on actual stakeholder practices. For the latter, the conversion of rural land to urban land involves a variety of mechanisms, including direct purchase of private parties to customary holders, followed or not by formalization, or subdivi- sion schemes at the initiative of prefects or mayors. A wide variety of situations is encountered during these processes which involve different degrees of legality and are often informal even when at the initiative of official stakeholders. For instance, subdivision by a mayor is not legal if the land was not previously registered in the name of the state and transferred to the commune. Service provision is almost never considered. All things else equal, these different tenure situations command different levels of tenure security and transferability and ease of formalization and thus command different prices. The ownership title is generally the most secure, but is the most expensive way of holding land39 and perhaps may have proven not be the most suitable way of securing the land rights for a clear majority of house- holds residing in the Greater Bamako Area, especially beyond the boundaries of the District40. Plots of land can be obtained through allocation processes outside land markets or through market transactions. The specific tenure on a plot of land reflects the way the plot was obtained, but it can also be transformed over time. Allocation processes outside land markets can take the form of customary allocations by village chiefs without involving any monetary payment in accor- dance with customary rules (which still occurs in the peri-urban area and the rural hinterland) or public attributions of use rights (concessions) by local or central au- thorities at non-market prices. As for market transactions, they may be formal or informal, depending on the tenure of the plot, the legality of the transaction, and whether the transaction is registered41. Tenure can also be transformed by follow- ing a sequence of land tenure formalization steps whereby it is possible to move up the ladder of formality and tenure security towards a use right or an ownership title42. Very often, however, the formalization process is never completed. The ten- ure situation of a plot, thus reflects where the current owner or occupant decided 38 Land use planning at the Cercle and Region levels and distinction between Communes rurales and communes Urbaines. The definition of these categories and the delimitation of each Commune is not identified from a legal point of view. 39 According to observers of the land market in the Bamako area and a study of land prices (Du- rand-Lasserve and Selod, 2015), a plot with an ownership title could cost three to six times more than a plot with similar characteristics but no title (see Durand-Lasserve et al., 2015). 40 Only 150,000 ownership titles have been issued in Bamako since 1932. 41 At the city fringe, where the city expands, there is a growing market for land sold by customary possessors and for plots held with an administrative document. Those markets can be viewed as informal in the sense that such transactions are not authorized in the law, although, to some extent, they are tolerated or even encouraged. The formal market strictly speaking is that of land held with an ownership title. 42 For a detailed description of these sequences, see the description of land delivery channels in Durand-Lasserve et al. (2015). 46 to stop along the chain of formalization. For instance, a very common form of land tenure in the peri-urban area is the bulletin, an administrative document which is obtained before a concession can be delivered. Although bulletins do not provide a legal property right per se,43 households keep them as if they were property rights and trade them subsequently on the informal segment of the market44. In practice, land in the peri-urban area and in the hinterland of Ba- mako is often obtained through the purchase of land directly from customary possessors without any legal document. However, this does allow for the possibility to directly sell on the informal market or, in a minority of cases, to formalize it and then to trade it on the formal market. These purchases of customary land accompany the massive ongoing process of transformation of agricultural land into residential plots that occurs as the city expands. In the zones adjacent to the District of Bamako, land can also be obtained through land subdi- vision schemes or lotissements initiated by various stakeholders (see below, more details on the lack of urban planning and the practice of lotissements) with or without a use right. In the core urban area, which corresponds to zones that were developed several decades ago, an overwhelming majority of plots were obtained under concessions (i.e. through public allocation from the State), or for a minority of them, through formal real estate development (and are then usually held with an ownership title). Possessors of title deeds may have purchased the land directly from a real estate developer, or on the secondary market, or may have upgraded a use right to a title deed. The market for formal land is limited, and serves mainly high-income households. One of the most striking features of the land system in Bamako is the small size of the formal land market with title deeds. It is also the result of sev- eral factors that limit the supply of formal land, driving up its price. The growth of the formal land market is indeed hindered by the lengthy and costly formalization processes that involve many stakeholders who capture rents at each step, which ends up deterring users from fully formalizing all the way to the ownership title.45 The formalization of land is also complicated by the difficulty of land adjudication given the diversity and complexity land tenure situations and the risk of conflict- ing claims (see the below paragraph on conflicts). Finally, the small size of the formal land market also reflects the scarcity of developments by private proper- ty development companies and housing cooperatives (who are confronted to a scarcity of land that can be formally transferred to them by authorities) and the 43 Bulletins are not property rights but could be used as proof of legitimacy. 44 For a household, not pursuing the formalization process beyond the bulletin stage has the advan- tage of not having to pay the fees that would be required to obtain a concession. Although this is not legal, bulletins are often issued with the recipient’s name left blank to facilitate transactions: the buyer can then later continue the formalization process by writing his or her own name on the bulletin. 45 Measuring the costs of formalization is difficult given the many procedures and direct and indirect costs involved. For instance, Durand-Lasserve et al. (2015) describe a process of formalization of customary land involving payments in kind to surveyors and civil servants whereby the original customary possessors are left with a titled property (title deed) on ten percent of the land they originally owned under the customary regime. This is the example where the largest costs are borne. For the buyer of customary land who establishes an ownership title, it is not unrealistic to assume that the total costs that will be paid will be in the same range of magnitude as the value initially paid to the customary owner. It should be noted however, that the value of the land con- siderably increases with formalization in a factor of two to six according to Durand-Lasserve et al. (2015). For other, but other examples of formalization costs, see Djiré (2007). 47 narrowness of home ownership programs (known as logements sociaux; see the below section on housing)46. For most households, given the limited supply, access to land under formal tenure is either unaffordable or unattainable (due to the lack of social connections required to navigate the land administration). Most house- hold lack the social connections needed to navigate the land administration, or lack information on property rights47. Low and middle-income households, thus, access land for housing, invest in savings, or seek capital gains almost exclusively through informal land markets. This occurs at the cost of tenure insecurity and at high prices both in the formal segment and the informal segment (given that the majority of low and middle-income households compete with a minority of wealthy buyers who have the economic means and social connections to purchase informal plots and the option to later formalize them for resale on the formal seg- ment of the market) 48. Instead of coordinated urban planning, Bamako is built through a pro- cess uncoordinated and often unauthorized lotissements – leading to urban fragmentation. By far, the most common type of land development in- tervention in the Greater Bamako Area is the lotissement. In theory, a lotissement is “the subdivision of a single piece of bare land into plots with appropriate pro- vision of infrastructure and collective facilities to host the buildings to be erected by the future occupants49”. Since 2002, the law stipulates that lotissements are to be carried out by communes and require the prior allocation of land from the state’s private domain to the commune. In such “authorized lotissements”, plots are allocated to beneficiaries with a concession that can later be transformed into an ownership title. In a large majority of cases, however, the practice is very dif- ferent from the theory. For ownership of the land title, the commune is required to have a planning scheme approved. Since most communes do not have such a planning scheme in place, they engage in lotissements without a land title trans- ferred to them by the State, or on customary land. As a matter of fact, lotissements continue to be initiated by a variety of stakeholders (not only the communes but other levels of central government despite the devolution of lotissements to com- munes). The word has also become a catchall term that refers to a wide range of land subdivisions, including at the initiative of customary landholders themselves, with the help of surveyors. Some forms are legal/authorized, but most are illegal/ unauthorized50 in some way. Because the majority of lotissements is believed to 46 Employees of public organizations and private companies may gain access to land through mem- bership in cooperatives. Cooperatives were originally set up by trade unions and by professional associations but are now also popular with private companies and Malians abroad, who may join together to buy land. They obtain can obtain land from the State and then subdivide if for its mem- bers or purchase customary land without title and then obtain an ownership title. 47 Many households are poorly informed about their own tenure situations and the rights conferred by tenure upgrading. There is also a common misunderstanding among the population that ad- ministrative documents are property rights or that use rights are ownership rights. 48 Other factors contributing to high prices include demographic pressure and speculation (there is a strong demand to purchase land plots at the periphery of the city in anticipation of urban expan- sion). 49 See Decree No. 05-115 of March 9, 2005. Infrastructure and collective facilities are specified as “roads, drains/sewers, water supply, electricity and telephone.” 50 Illegal private constructions occur on the banks of the Niger river in spite of the land being unsuit- able for construction and in the public domain of the State (and thus not transferable to private parties). 48 be unauthorized, this also adds to the confusion surrounding land tenure in the Greater Bamako Area. The system of Lotissements hinders land-use planning, leading to un- planned and spatially-fragmented urban expansion. Lotissements occur in all rural communes in the peri-urban areas of the Greater Bamako, and can be quite lucrative. For authorities that engage in lotissements, the underlying motiva- tion is often to generate revenues from these lucrative interventions, rather than to use lotissements to implement predetermined urban plans51. In fact, there is no coordinated planning of lotissements at the scale of the urban area, which makes their implementation inconsistent with larger schemes of development of serviced land and with access to public services. Because of this, there is no control over the way the city expands and infrastructure and services are often very limited or not provided at all. This generates immediate environmental costs and results in haphazard patters of land use development that can potentially hinder the func- tioning of the city and make the future provision of services very costly. Unautho- rized lotissements further threaten the poor’s access to public services as many informal neighborhoods face difficulties to be connected to public services due to unplanned development and impediments to formalizing informally subdivided land. Informal tenure also prevents the possibility of implementing any form land taxation, which would be needed to fund infrastructure for urban expansion. The pervasiveness of land conflicts in Mali are symptomatic of the dys- functions in the land market, placing a major constraint on the urban development of the Bamako area. Newspaper articles about land conflicts are very common in the Malian press. It is estimated that around 80 percent of court cases in Mali involve land tenure issues52. In the Greater Bamako Area, land disputes are also very common and can result in the destruction of housing and evictions. Conflicts can involve many different aspects, which makes establishing a typology difficult (see Durand-Lasserve et al., 2015). Although they can overlap, three broad categories can be distinguished: (1) conflicts in the peri-urban area stemming from the vagueness regarding the scope and recognition of customary rights, which creates problems when the land is transferred and/or other claims emerge for the same piece of land; (2) conflicts in the Greater Bamako area stem- ming from the implementation of lotissements when ownership of the underlying land is unclear; and (3) conflicts related to the allocation and transfer of land by the State and local authorities to individuals, property development companies and housing cooperatives. What all three categories have in common is the lack of clarity on property rights and enforcement of these rights, which often stems from the weakness of the State to establish and guarantee rights53. They also have social consequences in that the weakest stakeholders (customary possessors and low-in- come households) have limited access to information and weak social and political networks and limited financial means through which they can access secure land. And economic consequences, through efficiency losses as insecurity deters invest- ments and prevents markets from functioning properly and having land put to its best use. 51 For some mayors, lotissements are the only way they can generate revenues to fund investments. 52 République du Mali, 2009. « Actes de la rentrée judiciaire. » Bulletin d’information de la Cour suprême du Mali, No. 10, Bamako. 53 This leads to problems of multiple claims, and sometimes even the issuance of multiples rights on the same plot. 49 Bamako has had past success in the provision of serviced and titled land, and the development of land and housing development by the State – but demand now far outstrips supply. Formal land and housing development in the Greater Bamako Area dates back to 1992 with the creation of a parastatal land development agency, the ACI (Agence de Cesssion Immobilière). The ACI channels serviced and titled land to the market and has been instrumen- tal in developing central parts of Bamako, including the high-standing residen- tial and business neighborhood ACI 2000. A small number of land and property development companies exist with an opportunity to carry out operations in public-private partnerships for social housing and the provision of formal hous- ing. However, supply of housing remains very limited in the face of the enormous needs in the Greater Bamako Area. Most real estate developers face difficulties se- curing titled land from authorities. Their activities are also limited by the narrow- ness of the middle-class, the low and irregular income of households, the complex procedures and restrictive standards constraining norms for planning, construc- tion, and development, and the weakness of the housing finance system. Attempts by the State to provide social housing suffer from lack of adequate urban planning. Responding to the difficulties faced by the middle class to access affordable and secure housing (i.e., those who have relatively high incomes but cannot directly afford to obtain titled land), the Malian government has engaged over the past two decades in a policy of “social housing”, which aims at helping households to access property ownership. The social housing pro- gram is usually carried out in connection with a public-private partnership, with the state providing a property development company with land and subsidizing household purchases. So far, it has provided a relatively small number of dwelling units (a few thousand units) at a subsidized price on lease-to-purchase arrange- ments or bank loans. The schemes seem to have failed to integrate properly new developments into the urban fabric: the new social housing developments, which are located on the edges of the city’s built area, frequently lack collective facilities, forcing residents to travel long distances to go to school or to the market or fetch water from the nearest water point. As part of the “social housing” policy, the gov- ernment of Mali has also experimented with a pilot for the construction of vertical buildings and denser neighborhoods (known as the “80 logements” in the Faladié neighborhood of the N’Tjiba commune, constructed in 1990). This is a form of collective ownership as the ten buildings were built on a single land title. The ex- perience, however, has led to difficulties of cohabitation with the deterioration of common parts over time. 50 Urban accessibility is hampered by inefficient land use and transport Cities are often referred to as centers of opportunity because they re- duce the economic distance between workers, employers, buyers and sellers. This density reduces economic distance, decreases the unit cost of sup- plying infrastructure such as transport networks, allows ideas to circulate, grow and mature, and promotes economic specialization whereby people can focus on what they do best or enjoy the most (Glaeser and Kahn 2004; Ciccone and Hall 1996). These advantages of cities explain why urban residents are generally more productive than workers outside urban areas, what economists commonly call agglomeration economies which stem from proximity and exchanges (Duranton and Puga 2004; Rosenthal and Strange 2004). Integrated labor markets also produce more inclusive cities, avoid- ing the disconnection of entire neighborhoods of urban areas from the rest of the local economy. When accessibility is good, cities can become integrated labor markets, that provide opportunities to residents allowing them to choose jobs from larger pools, hence leading to increased welfare (Bertaud 2014). Integrated labor markets exist when it is possible for an individual to reach a large share of the employment opportunities within a city at a reasonable cost or within a reasonable time-frame. Large and integrated labor markets, by increasing the number and the diversity of employers and job seekers, allow for an appropriate match that makes the best of their respective skills and aspirations. Empirical evidence suggests that the benefits of large and integrated labor markets are not simply a theoretical argument but that accessibility, measured as the number of opportunities an individual can access within a given amount of time, matters for productivity and well-being in cities. But cities are not always associated with high levels of employment accessibility. Although cities concentrate on a restricted area of land large numbers of households and employment opportunities, this simple geographic proximity does not imply that all households have an equal or even good access to jobs. In Mexico for example, urban sprawl, massive population decentralization into single use residential and peripheral settlements together with uncoordinated urban growth has widened the distance between jobs and housing, undermining Mexican cities’ ability to match skills to jobs. In Mexico City low-income house- holds living in the peri-urban areas can spend an additional four hours commut- ing per week compared to low-income families residing in more central areas (Kim and Zangerling 2016). Low accessibility levels, carry negative consequences for productivity and well-being in cities. Low accessibility leads to fragmented labor markets where matching takes place at a local level only. Jobs located outside high-density economic clusters tend be scarcer, more informal and lower paying. Low accessi- bility levels also force households to locate closer to jobs. The excess housing costs in these locations can be unaffordable for some households, which can be forced to compromise on living conditions by settling in informal areas, thus fueling the phenomenon of central slums found for example in many African cities (Antos, 51 Lozano-Gracia, and Lall 2016). Such an outcomes has been documented in Nairo- bi where most slum dwellers have jobs and comparatively high levels of education relative to those living in formal housing, yet their living conditions remain basic (Gulyani, Talukdar, and Jack 2010). Interventions aiming at coordinating the land use and transport si- multaneously reduce the disconnect between residential areas and employment opportunities through efficient planning while enabling for affordable and time scarce connections. Reductions in transportation costs have been found to have positive impacts on the employment prospects of the youth in Ethiopia (Franklin 2015) and in France (Le Gallo, L’Horty, and Petit 2017) while land use interventions also carry the potential to increase employment accessibil- ity (Avner and Lall 2016; Quirós and Mehndiratta 2015). Conversely, inadequate urban policies can foster disconnection within urban areas. The massive sprawl and low-density housing development in Mexican cities was partly the result of the housing policy reform of 2000. Bamako’s accessibility is driven by two main drivers – the use of land and the efficiency of transport. The intensity of land-use underlies the urban layout of the city and the level of density, which provide more opportunities for interaction and for better access to services. At the same time, the efficiency of the transportation network connects people with jobs and services, including by reducing congestion. This section will thus outline how these two factors affect accessibility within Bamako. Given lack of micro and geo-coded data on past land use and transportation, the analysis will showcase impact through simulations for future growth. Changing Land Use The objective of the model is to understand the consequences of the spatial distribution of people and jobs on levels of accessibility. Bama- ko is a fast growing, sprawling urban area with low population densities. Thus, on average people must travel longer distances to reach the same number of jobs than compared to in denser cities. Achieving more compact urban forms cannot take place overnight, as cities display inertia and path dependence through the considerable amount of sunk capital already invested in structures and infrastruc- tures. To account for this inertia, the model assumes that Bamako’s city structure can only change at the rate of its population growth, i.e. current city dwellers will remain where they live today but newcomers can decide where to settle in the urban area. Counterfactual scenarios are run for the year 2030 with the following demographic and spatial scenarios: 52 • 3 Demographic Scenarios: The rate of urban population growth could be low, normal or high. The normal growth scenario assumes a similar rate of growth as the one Bamako experienced from 1990 to 2015 – i.e. 5.26% annu- ally. The high and low growth scenarios assume annual demographic growth rates of 6.3% and 4.2% respectively (20% higher and lower than the past growth rate respectively) – see Table 3 for details. Table 3. Demographic scenarios for Bamako from 2015 to 2030 High popula- Normal popula- Low popula- tion scenario tion scenario tion scenario Annual population 6.3% 5.26% 4.2% growth rate (%) Population in 2030 8,007,000 6,898,000 5,934,000 • 3 Spatial Scenarios: Different scenarios for the spatial distribution of peo- ple and jobs are also considered. These include: a baseline scenario, a periph- eral scenario and a densification scenario. The baseline scenario assumes that jobs will continue to be distributed spatially as they are in 2015 and that pop- ulation will be added to each neighborhood in proportion to its share of total population in 2015. The peripheral scenario assumes that jobs and people will be drawn to peripheral neighborhoods in priority. Jobs in the periphery will be 1.5 times higher than they were in 2015 and those in the central neighbor- hoods 50% lower. And new urban dwellers will settle in the periphery in larg- er numbers than in the center. Finally, the densification scenario supposes the opposite trends to the peripheral scenarios with jobs and people densification in the central neighborhoods of the city. The central and peripheral neighbor- hoods considered for these exercises are depicted in Figure 25. Figure 25. Central and peripheral scenarios considered Ü Periphery Centre 0 2 4 8 Kilometers 53 Changing density patterns in Bamako has a big impact on accessibility. The consequences of these demographic and spatial scenarios are depicted be- low. When averaged over the entire urban area of Bamako, and assuming normal population growth rates, the densification scenario provides significant gains in terms of accessibility – approximately 50% of jobs will be accessible compared to 42.8% in the baseline and 36.2% under the peripheral scenario – see Figure 26 and 27. Put differently, without densification employment accessibility is lower by 7.2 percentage points and further growth in the peripheral areas would further re- duce accessibility by 6.6%. Thus, fragmentation and low densities in Bamako have probably exerted a large cost in terms of matching people with jobs and conse- quently for opportunities for increased productivity. And higher future population growth rates would exacerbate these costs. Figure 26. Difference in share of accessible jobs using the Sotrama network at peak-hour between the densification and peripheral scenarios. 0 2.5 5 7.5 10 km Change in share of accessible jobs with densification vs peripheral scenario -10% - -5% 5% - 10% -5% - 0% 10% - 15% 0% - 5% 15% - 20% 20% - 25% 54 Figure 27. Change in average share of accessible jobs within a 60-minute peak- hour commute by Sotrama and foot depending on the land use scenarios in comparison to the baseline Densification scenario Peripheral sc enario 8% 6% Sotrama 4% Pedestrians 2% 0% Pedestrians -2% -4% -6% Sotrama -8% Changing Transport Changes in the speed of the transport network is another important driver of accessibility in Bamako. Congestion in Bamako is high, in part, because of the natural bottlenecks created by the river Niger, owing to which only three bridges can channel the flow of commuters mainly coming from the Rive Droite of the river to the Rive Gauche in the morning. Freight transport also con- tributes to the congestion in Bamako and to the rapid depreciation of roads. Ba- mako is not only a main destination for products from Dakar, Conakry or Lomé, it is also a transit point. This means that because of the lack of ring roads, Bama- ko’s limited urban transport network not only supports the freight destined to its markets but also the goods travel destined to other West African cities. Finally, whereas the density of roads in Bamako is within or above regional averages, lack of road paving considerably reduces travel speeds, forcing people off the second- ary and tertiary networks and on to the main arteries and adding to congestion levels in the process. If the congestion in Bamako could be eliminated completely and travelers could commute at speed limits instead of congested speeds54 , an additional 38% of jobs within the city would be accessible within one hour by the Sotrama network – see Figure 28. While ensuring free flow speeds at all times and on all road links is not a realistic scenario, it never the less demonstrates the importance of travel speeds as a driver of accessibility. 54 This results hinges on the assumptions that are made about speeds on the transport network and are provided as a means of showing the costs of congestion only. 55 Figure 28. Loss of accessibility using the Sotrama network within one hour of travel because of congestion. 0 2.5 5 7.5 10 km Loss of accessibility due to congestion in 60 minutes by Sotrama Sotrama network 40% - 50% 0% - 10% 50% - 60% 10% - 20% 60% - 70% 20% - 30% 70% - 80% 30% - 40% 80% - 90% 90% - 100% Given the spatial layout of the Sotrama network, individual segments matter differently for overall accessibility. Simulations that increase speeds by 30% on six Sotrama corridors (see Figure 29) provide orders of magnitude of the impact on accessibility across different corridors. As expected, an increase in speed across all corridors simultaneously would have the largest impact on acces- sibility, with a 12% difference compared to no changes – see Figure 30. Individu- ally, the Sogoniko corridor seems to have the largest impact with a 4.3% difference in accessibility levels compared to the counterfactual - see Figure 29. The impor- tance of transport speeds across corridors is also different spatially – for instance, higher speeds on the Kati corridor would have limited impacts for the accessibility level in Bamako overall, but high impacts for certain parts of the city. 56 Figure 29. Sotrama corridor priorities for intervention (left) and accessibility gains assuming a 30% speed increase on all six corridors simultaneously (right). 0 2.5 5 7.5 10 km 0 2.5 5 7.5 10 km Accessibility gains associated with transport interventions (%) Sotrama corridors: All_interventions 25.0 - 30.0 65.0 - 70.0 priorities for interventions 0.0 - 0.0 30.0 - 35.0 70.0 - 75.0 Sogoniko corridor: 1st priority 0.0 - 5.0 35.0 - 40.0 75.0 - 80.0 Kalabancoro corridor: 2nd priority 5.0 - 10.0 40.0 - 45.0 80.0 - 85.0 Western corridor: 3rd priority 10.0 - 15.0 45.0 - 50.0 85.0 - 90.0 East corridor: 4th priority 15.0 - 20.0 50.0 - 55.0 90.0 - 95.0 Banconi corridor: 5th priority 20.0 - 25.0 55.0 - 60.0 95.0 - 100.0 Kati corridor: 6th priority 60.0 - 65.0 Figure 30. Gains in average share of jobs accessible within a 60-commute by Sotrama with interventions to increase speeds on select Sotrama corridors 12% 10% 8% 6% 4% 2% 0% 6 corridors Sogoniko Kalabancoro Ouest Est Banconi Kati 57 Institutional fragmentation and lack of investments hinder service delivery This section analyses aspects of institutional structure and metropol- itan coordination that have repercussions on service delivery in the greater Bamako region. The District of Bamako is part of the Bamako metro- politan region, and is becoming integrated with the cities of Kati and Koulikoro into a multi-city agglomeration marked by fragmented governance structures. This entails specific governance and coordination challenges at the metropolitan level for provision of infrastructure and basic services. Solid waste services, which are an important environmental and health concern (identified by stakeholders in the public and private sector and civil society representatives) is discussed as a special case. Solid waste management (SWM) also exemplifies some of the coordi- nation failures related to urban planning in the Bamako metropolitan area. The administrative landscape in Mali is characterized by institutional fragmentation and a complex relationship among deconcentrated and decentralized units of government. Mali has 771 local government units, consisting of 703 municipalities, 49 circles, 18 regions (of which 10 effectively established following a 2012 reorganization), and the District of Bamako. The regions and circles are deconcentrated administrative units of central government (circonscriptions administratives), while also being local government units (col- lectivités territoriales) with elected governing bodies. The District of Bamako has a governing council elected by direct suffrage and is under oversight by the Minis- try of Local Government55, while the District is also composed of six local govern- ments (commues) and an additional 29 local governments compose the greater Bamako metropolitan region. Overlaps in competencies bestowed upon local governments by legis- lation makes responsibility for delivery of services unclear. The local government law56 provides limited clarity on horizontal and vertical repartition of competencies. According to the local government law and the law on specific ar- rangements for Bamako, the delivery of a large array of services, e.g. water, ener- gy, health, education, and environmental protection are attributed to regions, cir- cles, the District of Bamako and communes. In the case of roads and cleanliness, this is also the case, but only “if management has been transferred”. The transfer is to be approved by the Council of Ministers and this has not yet happened for these services, which is much more advanced in the education and health sectors, with a clearer hierarchy of services. The absence of formal transfer creates a legal limbo in which de jure and de facto responsibilities are not clearly attributed. Since no local government can exercise control over another local government according to law this creates a situation which obstructs coordination of planning within the administrative boundary of the District of Bamako for services such as drainage and solid waste management. 55 Loi N°2017-051 du 2 oct. 2017 portant code des collectivités territoriales (local government law). 56 Loi N°053 du 2 oct. 2017 portant statut particulier du district de Bamako (the Bamako law). 58 Lack of clarity from legislation is further compounded by sector and provider overlaps. National, parastatal and city-level agencies are involved in infrastructure investments and services provision, with overlapping and unclear regulatory scope and responsibilities. Central government ministries plan and im- plement by sector, but there is no coordinated plan for development of the urban space and no planning of services in urban extension areas. Also, utility compa- nies deliver services formally under the jurisdiction of local governments by law (water and electricity), but as local authorities do not have a seat on the governing bodies or play any formal role in planning for services the lines of accountability are unclear and complicate all aspects of service delivery from planning over im- plementation to enforcement. The superposition of layers of decentralized authorities (the District and six communes) and support structures without clearly defined and implemented functional mandates constrains coordination and integrated urban planning. According to the Bamako law the District is re- sponsible for its schéma d’aménagement, while the local government law ascribes to the communes responsibility for elaborating a schema d’aménagement territo- rial, consistent with that of the cercle. On the other hand, the urban planning law bestows upon the communes the responsibility for elaborating urban reference plans. The regional development agencies (Agences de Développement Régional) are partly independent support structures under the MCT that provide reimburs- able technical assistance to local governments in the areas of planning and service delivery. In addition, the District of Bamako also has its own Cellule de Préconfig- uration de l’Agence d’Urbanisme de Bamako, which launches planning initiatives for the District and has made some progress in getting surrounding communes to sign up to more formalized arrangements for collaboration to plan the Bamako metropolitan area. Parallel attempts are being made by the surrounding com- munes to organize in an inter-communal collaboration and develop a vision of proximity to Bamako, focused on solid waste management and spatial manage- ment. However, none of the coordination attempts have so far come to fruition in terms of actual coordination of service delivery. Decentralization has not so far resulted in ability to develop urban plans by local governments. Mali’s decentralization process, including for planning, takes its impetus form the 1992 constitution. Although -as described above- the responsibility for elaborating and implementing urban plans has in principle been decentralized, it remains largely the function of the central govern- ment. This situation reflects international trends, with gradual decentralization of planning responsibility since the mid-1990s.57 As in many other developing coun- tries in the process of decentralization, devolution of responsibilities is subject to interpretation and in the case of Mali neither the fiscal or human capacity is pres- ent to support implementation of legislators’ intentions. While the government yearly funds on average two urban plans for major cities, no local government has the financial resources to elaborate urban plans. Urban plans and planning institutions are ineffective and are not managing the spatial development of the Bamako metropolitan area. 57 Somik Vinay Lall, J. Vernon Henderson and Anthony J. Venables: Africa’s Cities. Opening Doors to the World. World Bank Group, Washington DC, 2017 59 The total number of accredited urban planners in Mali is about 2058, or 0.13 per 100,000 capita. This is very low, even compared to an average ratio of 0.97 plan- ners for every 100,000 people in selected African countries.59 The last approved urban structure plan for Bamako dates back to 1995, and although it contains some considerations of trunk infrastructure in the metropolitan area, it is com- pletely surpassed by development and expansion of the city in the meantime. A 2005 revision was elaborated to integrate parts of the surrounding area into the District but was never approved, since the institutional stakeholders could not agree on the boundaries of the District. A further attempt was made by the District in 2014 to recruit a planning agency (on external funding) to update the structure plan but floundered due to disagreement between the government and the District of Bamako over planning authority. The District of Bamako and six communes of the District have urban planning departments, and the District has developed a plan to ad- dress some identified shortcomings. The ‘Vision 2030’ includes investments to, among others, densify the commercial center of the city, minimize heavy trans- port and unloading in the city through construction of dry ports, and rebalance the development of the city through moving some institutions and commercial activities to the right bank of the Niger River. The ‘Vision 2030’ constitutes a good starting point for developing an urban structure plan, but the legal and regulatory foundation for any investments to reshape the city are not in place in the absence of an approved and enforceable plan. In the absence of such a plan and institu- tional capacity to assure its implementation, the city is peri-urbanizing through unauthorized lotissements to which the local government authorities and techni- cal services participate (see section on land markets in Chapter 2) and densities remain lower than optimal in the absence of a land market and taxation of unde- veloped land in the city center. The cost of fragmentation is not directly quantifiable but is likely high. The literature is not unanimous on the effects of fragmentation, there is some evidence that fragmentation has a negative effect on productivity in OECD coun- tries while countries in Latin America and the Caribbean could conversely benefit from fragmentation.60 In the case of Bamako, with 7 administrative units in the 58 According to Diarra Sissoko, the President of the Malian Association of Urban Planners, inter- viewed November 22, 2017. 59 Authors’ calculations based on the number of planners in 12 African countries in: The State of Planning in Africa. African Planning Association and UN-Habitat, 2014. 60 See e.g. Ferreyra, M., Roberts, M., Gracia, N., Cadavid, P. and Selod, H. (2017). Falling Short. Cities and Productivity in Latin America and the Caribbean. They find that LAC metropolitan areas are likely being affected negatively by their levels of fragmentation. With on average 9.39 administra- tive units per metropolitan area, this level is detrimental to productivity. The study finds no evi- dence that presence of a governance body at the metropolitan level mitigates the negative effects of fragmentation (but the study does not include measures of efficacy of such governance structure). Ahrend et al (2017) also find that cities with fragmented governance structures tend to have lower levels of productivity (10 percent increase in the number of municipalities leads to 0.6 percent lower productivity). But to the contrary they find that the negative effects are mitigated by almost half by the existence of a governance body at the metropolitan. See Ahrend, R., Farchy, E., Kaplanis, I. and Lembcke, A. (2017). What makes cities more productive? Evidence from five OECD countries on the role of urban governance. Journal of Regional Science, Vol. 00, No 0, 2017 pp 1-26. Bartolini (2015) also finds that municipal fragmentation within a given country has a negative impact on per capita GDP growth and that for a given level of municipal fragmentation overlapping of function is more severe in urban regions (where people are likely to commute over municipal boundaries) than rural regions. See Bartolini, D. (2015). Municipal Fragmentation and Economic Performance of OECD TL2 Regions. OECD Regional Development Working Papers, 2015/02, OECD Publishing, Paris. 60 District of Bamako and 31 local governments making up the Greater Bamako area, the qualitative analysis supports that fragmentation causes problems of coordina- tion between the metropolitan center and the surrounding local governments (and with public utility companies) regarding service delivery. There is also no capacity to invest by Bamako local governments and hence the government is in principle the main funder of infrastructure in the capital. The combined identifiable infrastructure expenditure (operating and capital expenditure) by the seven local governments in Bamako (the District and six communes) was $0.72 per capita in 2016.61 Figure 31 provides a com- parison of this figure to other capitals and Figure 32 provides a comparison by income tier. Budgets typically contain high proportion of emoluments (62 percent on average) and only 4.6 percent of budgets go towards investments. Indeed, the recent increase in overall local government revenues in Bamako appears to have been driven by dramatic hikes in the size of earmarked conditional transfers, largely intended to cover the costs of running devolved services (such as education and health). These functions (or “competences”) have been devolved as part of a wider national policy of shifting sector responsibilities from central to sub-nation- al government. As such, these conditional transfers do not represent a significant increase in the quantum of resources available for local investments; recipient lo- cal governments do not have much latitude to decide on the use of these transfers. The District of Bamako has benefited from a substantial increase in transfers over the past years, up from $10,7 million in 2015 to $40,2 million in 2016, bringing total revenue to $50.4 million but this is however also not reflected in increased investments. Own-source revenues remain largely unexploited and e.g. the local and regional development tax that is dedicated to service provision and infrastruc- ture is, at $166,000 collected, very far from the potential of about $13 million. Thus, the government funds service delivery directly, although according to law this falls under the purview of local governments. Figure 31. Infrastructure investments by Bamako and comparators (2016) USD per capita 0 50 100 150 200 250 300 Washington, D.C. Malta Buenos Aires Addis Ababa Johannesburg Infra CAPEX Infra OPEX Amman Medellin Santa Cruz de la Sierra Kiev Mbarara Yangon Nairobi Casablanca Kathmandu Accra Karachi Dar es Salaam Freetown City BAMAKO Dodoma Source: From Yu Lu: City-level capital and operational expenditures for infrastructure, benchmark- ing for Bamako (2017). 61 Analysis of latest available budget data for this report: Budget execution reports for 2016 for the District and communes, except Commune I for which budget execution for 2015 was used. 61 Figure 32. Infrastructure investments by Bamako compared to income levels (2016) 0 50 100 150 USD Upper middle income Lower middle income Infra CAPEX Infra OPEX Low income Bamako Source: From Yu Lu: City-level capital and operational expenditures for infrastructure, benchmark- ing for Bamako (2017). Bamako’s scope and capability seem somewhat mismatched – with neither being very high. Urban governance structures are the outcome of spe- cific historical and political circumstances. Not only does the administrative scope or remit of Bamako matter, but also its capabilities to implement its remit, given access to resources – fiscal and technical. It seems that Bamako lacks financial resources and technical capabilities, to varying degrees, to tackle urban planning for the delivery of urban infrastructure and services (see Figure 33 and accompa- nying box). The total number of urban planners in Mali is 20, or 0.13 per 100,000 people (compared to 0.97 across Africa)62. The city scope and capability radar graph shows that Bamako’s fiscal scope (above the line) and its capability (below the line) is reasonably well matched – neither is very high. However, the city has no control over its waste management function63 - this will be discussed in detail below. There is also a big disconnect in terms of business environment, wherein the Doing Business measure regarding construc- tion permits (mainly a de jure measure) lags far behind actual time to get a con- struction permit (i.e. de facto) measured by Enterprise Survey data from Bamako. 62 Authors’ calculations based on: The State of Planning in Africa. African Planning Association and UN-Habitat, 2014. 63 The absence of revenue in the local governments’ budgets reflect the funding structure of SWM ser- vices: households pay directly to small collection companies (groupements d’intérêt économique) and the government funds the collection contract with a private company (Ozone) directly. The District of Bamako stopped budgeting expenditure for the Roads and sanitation agency separately as of the 2016 budget, after signing the Ozone contract in on September 29, 2015. 62 Figure 33. City Scope and Capabilities (Bamako) The City Scope & Capabilities radar graph outlines how policy reforms and inter- ventions at the city level are channeled through a city government’s operational scope and ca- pability – i.e., the powers at the government’s disposal, and, the ability to effectively and ef- ficiently carry out those powers. It includes 6 variables, which consider city scope and capa- bilities in terms of 3 different dimensions: fis- cal, infrastructure, business environment. The limits of the graph on each variable are given by the lowest and highest scores among 8 African cities – Bamako, Niamey, Conakry, Lagos, Ka- mala, Addis and Johannesburg. The scope for Bamako – as a sub-national “city” government – to shape, finance and determine urban outcomes is limited by several factors. Firstly, decision-making in Bamako is highly fragmented. The central government is a major player in what happens in the national capital; central government may not always share the same vision as local government. In addi- tion, Bamako includes two legally autonomous local government tiers (District and communes), neither of which can impose decisions on the other; and there is clearly a good deal of overlap in the functional mandates of each tier, in practice if not in theory. Moreover, different local governments are governed by different political parties, making it more difficult to make common decisions. Lastly, Ba- mako (for better or worse) is the focus of Mali’s voice-empowered socio-political elites, a further complication in any already fragmented decision-making frame- work. Secondly, Bamako’s local governments (like all other local governments in the country) are faced with a growing set of functional mandates as the central government devolves sector functions to the sub-national level. Rapid sector devo- lution, partly driven by symmetric regionalization and secessionist “safeguards”, is probably swamping sub-national capacities in a variety of ways and, in Bama- ko, may well be constraining local officials’ ability to focus on and address urban development challenges. 63 Special Spotlight on Solid Waste Management Reflecting the coordination failures of planning and service provision, there is substantial lack of clarity of responsibilities for provision of solid waste management services, and a substantial funding gap. The District of Bamako and communes are responsible for provision of solid waste management services, but in reality, the government carries the majority of fund- ing responsibility, and the six communes of the District play a limited role as regu- lators and supervisors of primary collection agents. The identified investments in solid waste management in the seven local government’s budgets are $0.03 per captia. The absence of revenue in the local governments’ budgets reflect the fund- ing structure of SWM services: households pay directly to small collection compa- nies (groupements d’intérêt économique – GIEs) and the government funds the collection contract with a private company (Ozone) directly. The District of Bama- ko stopped budgeting expenditure for the Roads and sanitation agency (Direction des Services Urbains de Voirie et d’Assainissement – DSUVA) separately as of the 2016 budget, after signing the Ozone contract in on September 29, 2015. Unreg- ulated consumption of land and means that there are no land reserves within the District boundaries to construct SWM infrastructure, including transfer stations, leading to inefficiencies in the collection chain. The problem largely overflows the boundaries of the District, as until recently there was no engineered landfill to dump waste. Hence, the majority of waste is dumped in surrounding local gov- ernment circumscriptions (and the District of Bamako is in fact benefiting from free ridership for disposal while negative spillovers are borne by the surrounding communes), or in an unsanitary landfill located centrally in the city of Bamako. About 71% of households in Bamako subscribe to private waste col- lection services by some estimates. As shown in Figure 34, this share is substantially larger than other urban areas outside the Koulikoro region, where this percentage is 25%. Urban areas of the Koulikoro region have slightly lower subscription rates, at 20%, indicating room for improvement within the region. While limited data is available for cross-country benchmarking, Bhada-Tata and Hoornweg (2012) report collection rates (by weight) of 77% for Zinder, Niger and 76% for Conakry, Guinea, both measured in 2007. If one assumes a constant weight generation per household, and that “dépotoire public” counts as collection, then Bamako preforms fairly well relative to these benchmark, with 88% of waste collected. Although the collection frequency might be high, in reality the majority of waste gets dumped on open land in the surrounding communes. The collection rates in the Koulikoro Region, within which the District of Bamako is located are very low, at 20%, and would benefit from a coordinated approach to collection and disposal. 64 Figure 34. Bamako has substantially more developed waste collection services than the rest of the country Household collection Transfer station Burned or buried by Unauthorized dumping household or other 70% 70% 70% 70% 60% 60% 60% 60% 50% 50% 50% 50% 40% 40% 40% 40% 30% 30% 30% 30% 20% 20% 20% 20% 10% 10% 10% 10% 0% 0% 0% 0% Bamako Koulikoro Other Bamako Koulikoro Other Bamako Koulikoro Other Bamako Koulikoro Other District Region urban District Region urban District Region urban District Region urban areas areas areas areas SOURCE: World Bank staff calculations using the Enquête Agricole de conjoncture intégrée aux con- ditions de vie des ménages (LSMS), 2014. The scope of the Maire du District du Bamako to improve the quality of solid waste management services is primarily defined by contracts between the city and private firms. As is common globally and across Africa, the district of Bamako has fully outsourced solid waste collection, with pre-col- lection being completed by approximately 180 GIEs that collect waste and fees directly from households under 3 to 5 year contracts with the communes that give them monopolies over a set of land plots (referred to as concessions, that include groups of on average 2.1 households) within a specific neighborhood (quarti- er). The transport of waste from the GIE’s transfer depots, collection from large producers, disposal, as well as street sweeping, has been contracted to Ozone, an international firm that signed a convention with the city in 2014 giving it this responsibility (see Maire du District de Bamako et Ozone, 2014). The city can therefore only affect quality of services by using sanctions available within existing contracts with the private sector, for instance penalties for underperformance.64 In the longer term, it has the opportunity to revise contracts, as it is doing currently with Ozone and the GIEs, through in this case, it appears that the national govern- ment, via the Prime Minister’s office and also the Ministre de la Décentralisation et de la Fiscalité Locale exerts substantial de facto control over the negotiation process. For more detail see ARTELIA (2015). 64 Penalties available under the current convention are difficult to enforce, though this is recognized by the authorities and the convention is currently being renegotiated. For instance, the 2014 con- vention allows the city to apply a fine, as described in Article 56, for « secteur non collecté, non balayé (4 heurs après signalement par le délégant) » at the rate FCFA/jour 300,000 to the share of sectors plausibly not cleaned by Ozone. The phrase « 4 heurs après signalement par le délégant » in the convention indicates that sanctions can only be enforced if the Mayor of Bamako notifies Ozone immediately when a sector is not cleaned. This implies that for the sanctions in a contract to work, the city must (i) have a daily flow of information on whether Ozone is completing its tasks, and (ii) a way to transmit that information readily to Ozone. Future contract negotiations should consider methods of establishing these flows of information. In principal, the GIE could play an important role in supporting such a system, providing DSUVA with regular information on whether Ozone has picked up waste from specified areas, such as transfer depots. If poor per- formance is reported, Ozone could be sanctioned. In principal, such sanctions could be enforced under the existing contract. 65 The only areas of direct formal government participation in the sys- tem are in the function of environmental regulation, specifically with regards the control of unauthorized dumpsites (dépôts anarchiques). These activities are carried out simultaneously by the Brigade Urbaine de Protec- tion de l’Environnement (BUPE), a commune funded agency, and two parallel de- centralized agencies of the MAEDD, the Direction Régionale de l’Assainissement et du Contrôle des Pollutions et Nuisances du District de Bamako (DRACPN-BD), at the district level, and Service de l’Assainissement et du Contrôle des Pollutions et des Nuisances (SACPN), at the commune level. The scope of the Maire du District du Bamako to improve the quali- ty of disposal is limited by its ability to plan and develop projects in collaboration with the surrounding communes outside the district. Though local governments are de jure responsible for environmental protection and cleanliness, the key activity and often most substantial investment in dispos- al, construction and operation of landfills, is carried out by the national agency Direction Nationale de l’Assainissement et du Contrôle des Pollutions et Nui- sances (DNACPN) within the Ministère de l’assainissement, de l’environnement et du développement durable (MAEDD). To exemplify this, the Noumoubougou sanitary landfill for the District of Bamako, was constructed by the government without prior consultation with the District authorities. If the city is to assert more control over the quality of disposal in the future, it will have to find ways to plan, finance and build projects on its own. In doing so, there are potential economies of scale from working with surrounding communes to develop projects requiring waste at scale, such as for instance, biogas or landfill gas electricity generation facilities, though the technical requirements of these facilities are such that they would need to be developed as greenfield projects. The capabilities to tackle the challenges associated with SWM seem limited and are constrained by institutional overlaps and land issues. The District does have some planning and service delivery capacity through the Road and Sanitation Unit but the multitude of stakeholders involved in solid waste management exemplifies the limited intra-jurisdictional coordination, central government infringement on local government prerogatives, and lack of control over land use. The landfill was constructed by the central government, based on a 2003 solid waste strategy, but ownership has not been transferred to the District. Some minor problems to render the landfill operational are being sought solved by the MAEDD, but so far unsuccessfully, and in the meantime the District does not dispose of the landfill that is an integral part of the contract with the private collection operator. In addition, the landfill is located 35 kilometers from the city, which increases the transport of solid waste While the 2003 solid waste strategy identified a second site for a landfill in Dialakorobougou outside the District limits, to serve communes V and VI of the city (Rive Droite), issues of land ownership were left unresolved. No land has been reserved for a second landfill in the vicinity of the city on the right bank of the Niger River, which could contribute to reducing transport cost and through traffic in the center of the city. In several cases land reserved for transfer stations in the solid waste sector plan has been attributed to private owners or is being slowly but steadily encroached upon. This means that the window of opportunity to invest in infrastructure to improve efficiency of the collection chain is quickly closing. 66 Under current contractual arrangements and revenue mobilization plans, the District of Bamako faces an estimated $16.7 million annual operating deficit, shown in Figure 35.65 This estimate of total cost of the sys- tem includes $9.4 million for pre-collection, borne by GIEs, largely for labor and fuel costs, $17 milllion for the Ozone contract (this was the amount invoiced for 2016), and $200,000 for a fully funded environmental control program involving 48 agents with motorcycles, enough for 8 per commune. The estimated revenues rely on the current proposal of a FCFA 3,000/concession/month tariff, and an 80% collection rate. This graph makes clear that even with very high rates of col- lection, the proposed tariff will not come close to covering the costs of the system. Indeed, it will only cover those costs of the GIEs activity, which they currently cover by collecting tariffs ranging between FCFA 1,000-5,000/concession/month themselves. This deficit is expected to accumulate over time, making revenue mobilization a top priority. The district currently has FCFA 11.7 billion in arrears to Ozone, and communes are also understood to have arrears of an unknown size to the GIEs. Figure 35. Under current plans, Bamako’s solid waste management system faces a US$ 16.6 annual operating deficit USD Environmental Millions GIEs Ozone control Revenue Defic it 0,0 -5,0 -9,4 -16,7 -10,0 -15,0 -17,0 -20,0 9,8 -25,0 -0,2 -30,0 SOURCE: World Bank staff calculations. Estimated total cost includes labor, fuel, materials and depreciation of vehicles used in waste collection and transport but does not include costs associated with management of the landfill or transfer depots. The $9.4 million GIE bill is extrapolated from the business model of a single GIE, provided by Le Collectif des Groupements Intervenant dans l’Assainissement au Mali (COGIAM), assuming a 90% subscription rate. This calculation there- fore assumes a fully functioning system, given that subscription rates, or the share of concessions procuring service from the GIEs, currently range from just 40-70% depending on quartier. The $17 65 Expenses associated with private contracting of waste management services can run out of con- trol, and lead to failure of contracts. For instance, in 2014 Nouakchott, Mauritania cancelled a 7-year contract with the international Pizzorno Environment Group, after the cost exploded. In 2013, the bill had reached FCFA 10 billion annually, similar to that of Bamako with Ozone, despite the fact that Nouakchott has approximately 1/3 the population. Under an arrangement with the national government, the Urban Community of Nouakchott (CUN) had to cover FCFA 2 billion (20%) from its own resources, with 80% coming from the government. Given the small revenue base available, the city and national government cancelled the contract. 67 million Ozone bill is equivalent to the FCFA 9 billion in invoices submitted by Ozone during 2016. The $100,000 BUPE bill is estimated assuming 24 officers, each with a motorcycle, and 2 managers. The proposed revenue stream assumes a 3,000 FCFA/concession/month fee, an 80% collection rate and 182,789 concessions, as reported by the 2014 Annuaire Statistique du District de Bamako. Performance of private contractors under the current system has been far from perfect. Subscription rates by GIEs range between 40-70% per com- mune, suggesting household collection rates within that range. Ozone is also not meeting its obligation: estimates by DSUVA indicate that in 2016, just 76.62% of the waste quantities targeted in 2014 Ozone convention had been collected. This number was lower for street sweeping, where just 68.82% of kilometers of street have been swept. The primary concern from the perspective of environmental regula- tion is the control of “dépôts anarchique,” or unauthorized dumpsites, of which there are approximately 5-7 per commune at any given time. The agencies responsible for this task, namely BUPE, DRACPN-BD, and SACPN lack staff and key resources, such as vehicles and GPS equipment that would be helpful in more tightly controlling these sites. Evacuation of unauthorized dump- sites not effectively controlled by these agencies cost the Maire du District de Bamako at least FCFA 222.5 million (~$416,000) in 2016, at a rate of FCFA 5,000 per cubic meter (DSUVA, 2016).66 An investment of 48 motorcycles (at FCFA 2 million each) and GPS devices (at $50 each) to support 8 control agents across all 6 districts, would cost only ~$180,000, which would be easily justified by the savings generated if it could reduce the volume of unauthorized dumping by 50%. Annual operating costs of an upgraded control system are estimated to be approx- imately $200,000 a year, less than 1% of total operating costs. Intangible benefits from a cleaner city, depending on how they are valued, are likely to be substantial- ly greater. Solid waste management needs are substantially greater in the Kou- likoro region than in the District of Bamako, especially when recent urban growth is taken in to account. Figure 36 provides an estimate of daily waste generation in the Koulikoro region. 36% of waste generation occurs outside of the district, and so there is therefore potential to achieve economies of scale by combining solid waste management operations of the regional government and the District du Bamako. These include primarily economies in: (i) procurement of equipment (i.e. trucks, transfer station containers), (ii) management of contrac- tors that deliver transportation services. 66 This number includes the evacuation of 44,503 m3 from 21 unauthorized dumpsites in communes I, II, VI and V reported by the Direction Des Services Urbains De Voirie et D’assainissement (DSUVA) for September 5, 2016. 68 Figure 36. 36% of waste produced in the Koulikoro region is produced outside of Bamako district SOURCE: World Bank staff calculations based on the 1998 and 2009 censuses. The forecast assumes that urban waste production is 0.58kg/day/capita (equal to the daily generation implied by waste generation targets and populations listed in Table 1 of the Convention with Ozone); that rural pro- duction is 0.1 kg/day/capita; and that population growth since 2009 in urban areas has been 5.08% per annum and in rural areas 1.75%, consistent with United Nations forecasts. The forecast classifies as urban all communes with population growth rates higher than 5% p.a. between 1998-2009 in the census, or with population more than 50,000 in 2009. This results in a 33% higher forecast than projections made using the current urban/rural classification used by the Ministre de l’Administra- tion Territoriale de la Décentralisation et de la Réforme de l’Etat. 69 70 te r h a p c 3 71 How to unleash Bamako’s potential? To unleash Bamako’s potential, a balanced approach to reforming institutions, putting the right policies in place and investing in infra- structures and attention to implementation will be needed. This section highlights steps taken by policy makers in Mali to address Bamako’s challenges, but also brings in practical examples from other resource-constrained and often low-capability urban environments Given the competing demands on the public purse, the chapter also demonstrates how technologies, community-based and private-sector based approaches can be used effectively to provide many solutions to urban issues in the city. The main recommendations are summarized below: • Coordinate land use and connective infrastructure: Public land and properties could be used to promote investments, which would help encourage denser economic infrastructure investments within the city. An inventory of public holdings – land, administrative buildings and infrastructure, would be a first step to help set the tone to enable the development of efficient land and property markets. In parallel, efforts should also be made to implement the new legal frameworks put in place to ensure better land tenure and manage- ment, for instance, through the installation of local land commissions to help tackle uncontrolled land conversion taking place in peri-urban areas. To better connect people to opportunities, investments should be made in upgrading road quality and better managing public spaces. And finally, digital technolo- gies should be leveraged to work with communities to better monitor and track accessibility changes. • Finance and manage better public service delivery: To meet the growing needs for investments in public infrastructure, systems should be developed to increase the sources of own-source fiscal and other revenues at 72 the local level, through mechanisms such as increasing tariffs and taxes and national government transfers. At the same time, there should be a focus on reducing inefficiencies in the system – for instance, in the case of solid waste management, better placement of landfills and more opportunities for com- posting would significantly impact public revenues. And lastly, contractual ar- rangements can be organized better to maximize the incentives for public and private service delivery providers to increase access while recovering costs. • Invest in urban institutions: Since much of current and future urban growth will spill-over to neighboring communes, systems and incentives for inter-jurisdictional coordination on planning for investments in infrastructure and service provision will be paramount. At the same time, institutional re- sponsibilities across jurisdictions and between national and local levels need to be better clarified, and capabilities to implement the local administrative remit be reinforced. Citizen engagement and private sector can and should be leveraged as a crucial element of addressing urban infrastructure financing and governance, including using data and technology. Reform Institutions Modify Policies Invest & Implement Coordinate Land Inventory public land, Manage public spaces Implement new legal Use and Connective buildings and infrastruc- better, including road frameworks, for instance, Infrastructure ture rehabilitation and main- through the installation of tenance local land commissions Finance and man- Develop systems to Reduce inefficiencies in Organize better contractual age better public increase its own-source the system, for instance arrangements, to maximize service delivery fiscal and other reve- in the case of SWM incentives for increasing nues, such as through building a landfill closer access and cost recovery increasing tariffs, taxes, to the city and compost- and national government ing 50% of the organic transfers waste within the city Invest in Urban Develop systems and Clarify institutional Leverage citizen engage- Institutions – Think incentives for inter-juris- responsibilities and rein- ment and competitive outside the Box dictional coordination on force capabilities at the markets as a crucial ele- planning for infrastruc- local level ment of addressing urban ture investments and infrastructure financing service provision and governance, including using digital technologies 73 Coordinate land-use with connective infrastructure To help defragment Bamako, and to lay the foundations for a more compact city in the future, coordination between land use and con- nectivity will be critical. Accessibility between people and jobs, and for more efficient service delivery, is a function not just of where people and employment opportunities locate, but also the ease of connectivity between them. Thus, policy makers must use both levers to influence the spatial form of Bamako in the future. This will require not only well-functioning land markets, on the one hand, but also planning for connective reforms and investments on the other. Public land and properties can and should be used to promote invest- ments, including in economic infrastructure, which would also help generate revenues to government in a transparent way. Real estate and asset management is a powerful tool in the city’s arsenal, which can also set the tone to enable the development of efficient land and property markets. The first step would be to do an inventory of public property holdings – land, administra- tive buildings and infrastructure – which usually constitute the lion’s share of local public wealth67. Government property holdings are often very large, and can directly impact the urban economy and the urban spatial form – for instance, an inventory of public property holdings revealed that the city of Warsaw, Poland (with a population of 1.7 million, similar to Bamako) held 11,312 government buildings and facilities and 10,000 shops of more than 1 million m2. An invento- ry of public land would also help identify pockets of land within the city where it would be possible to crowd in commercially-viable infill development by the pri- vate sector. And proper asset management can also result in huge public revenues and savings opportunities. An example from Kuwait City illustrates the magnitude of forgone revenues. Kuwait City leased nearly 600 hectares of land to citizens at low fixed rates – land which was immediately sub-leased informally to small entrepreneurs, with the rental revenues foregone by the city equivalent to about 1.5% of GDP, annually68. Such savings, could in turn, be used for much-needed in- vestments in infrastructure assets for service delivery. A recent Public Investment Management69 report by the IMF inventories assets for the central government, and could be used as a basis for a sub-national, Bamako-specific, assessment. The current focus on undertaking land reforms, aimed at better land tenure security and land management, is laudable and should be scaled. Until recently, a main principle in Mali has been that land without a title is presumed to belong to the State. This remains a key principle - both the recent Rural Land Act (enacted in April 2017) and the State Land and Land Act under reform include provisions that could strengthen the recognition of land rights (in- cluding land use rights, particularly for agricultural lands), the process to improve 67 Capital assets can account for 66% (Los Angeles, USA), to 69% (Cape Town, South Africa) to 94% (Warsaw, Poland) of total assets (i.e. capital and financial). See Peterson, G. and Kaganova, O. (2010) ‘Integrating Land Financing into Subnational Fiscal Management’, World Bank Working Paper #5409. 68 World Bank (2003) Land and Real Estate in Kuwait: Market Analysis and Policy Implications 69 IMF (2017) Evaluation de la gestion des Investiseements publics, Rapport d’Assistance Technique. 74 land tenure security, and land management. The reform of the legal and institu- tional framework that governs rural and urban land tenure aims to secure land rights while ensuring that land administrations are improved, and that revenues from land ownership and State property are mobilized. Three ongoing processes need to be kept in mind in the attempt to improve the land sector in the Bamako area: (i) the Rural Land Act which will apply to part of the greater Bamako, (ii) the State land and Land law reform (the State has a central role in urban areas to register and manage land), and (iii) the process of cadaster building which started in six communes in the Bamako District70. With these ongoing land reform pro- cesses, an important challenge for the Greater Bamako Area will be to clarify the scope of each law, i.e. identify where each will be implementable and how they will be coordinated. However, building a land cadaster, with the accompanying geo- graphical information and legal frameworks, is a longer-term objective. The use of advances in digital technologies could be used to make progress on registration processes (see Box 5 below). 70 The enactment of the Rural Land Act, the adoption of the application decree (including the opera- tionalization of rural local land commission), the amendment of the State land and Land Act, and the cadaster building process are included in a budget support of the World Bank. 75 x bo 5 Using Blockchain to administer and manage land Blockchain technology has the potential to revo- could be used as a pilot to increase trust and lutionize the way records or value, such as a land transparency. The simplest application would plot, are stored and transferred. Blockchain, as be virtual notarization on a public blockchain the name suggests, is a “chain of blocks”, where accessible to all on the network. Instead of requir- each block represents a record. This record could ing a notary to certify previous ownership while represent an asset like a land plot, an identity, or transferring an asset, the blockchain can process even a cryptocurrency. Blockchain is useful since the virtual notarization at a lower cost. While it is – decentralized (processing takes place on the public blockchain is more transparent and several “nodes” or computers connected to the tamper-proof, a private blockchain pilot can also blockchain network decreases the transaction be useful if it comes with a stamp of approval of processing time and possibly, cost71), distributed the government that would uphold the legality of (the data is spread across different nodes, in- transactions on the blockchain platform. creasing transparency and reliability and improv- ing disaster recovery) and an immutable ledger However, blockchain requires certain “off- (blocks are connected through a complex mathe- chain” conditions to function well. For a matical formula that is cryptographically secure, blockchain-based solution for land administration making it almost impossible to change a record in to work effectively, accurate, digitized records the past). are required74. This is what is behind the success of BenBen, a private company, in Ghana, that Evidence from pilots indicates that block- uses surveying and mapping techniques to obtain chain can help to register land titles and accurate field information before digitizing it on transactions in a tamper-proof way. This their platform. Once accurate, digital information can be seen in the cases of the Bitfury pilot in is available, it is used for transactions75, which Georgia72 and the Consensys pilot with the Dubai in turn has helped galvanize the use of land in Land Department73 which register land titles and commercial markets. It should be noted, however, transactions respectively on a private blockchain. that leapfrogging technology has much potential Blockchain can be very useful in low governance but only if major underlying legal problems are environments as time-stamped, tamper-proof properly addressed. transactions can be stored on a blockchain. This 71 The Internet of Value-Exchange, Deloitte Report: https://www2.deloitte.com/content/dam/De- loitte/uk/Documents/Innovation/deloitte-uk-internet-of-value-exchange.pdf 72 The First Government To Secure Land Titles On The Bitcoin Blockchain Expands Project: https:// www.forbes.com/sites/laurashin/2017/02/07/the-first-government-to-secure-land-titles-on- the-bitcoin-blockchain-expands-project/#4feddb424dcd 73 Blockchain Virtual GovHack video: https://www.youtube.com/watch?v=-y0WGwzKaxI 74 Other constraints to be considered include the privacy of data, capacity related to understanding blockchain advantages and disadvantages, data storage, the recognition of customary ownership, and the legal recognition of transactions done on the blockchain. 75 World Bank Interview with BenBen team (April 2017). 76 To ensure more sustainable urban development, certain additional land reforms should be undertaken in Bamako. An important first step is the implementation of the new legal frameworks, already put in place. The instal- lation of local land commissions in the Kati Cercle and the Koulikoro region, for instance, would help tackle uncontrolled land conversion taking place in peri-ur- ban areas, which are still managed under rural rules for the most part. These local bodies could also help monitor and prevent unauthorized lotissements. A system- atic inventory of public land should also include the validation of titles established in the name of the State. Improvements can be made in the methodology for the validation and conversion of existing precarious titles as well as for the preven- tion/resolution of land conflicts in urban areas – see an example of how this was done in Benin. For the Greater Bamako Area, the process will need to assess and make efficient use of the body of land information that already exists in the form of registries, plans and inventories at the local level (communes). Additionally, the new provisions will need to include a method and timeline for dealing with the accumulated inconsistencies in these records. x bo 6 The conversion of use rights to ownership rights (Benin) In Benin, permits to occupy (permis d’habiter), into title deeds. Although the program (supported albeit insecure, had become the key document by the Millennium Challenge Corporation) did to have for households wishing to access urban not achieve the desired results and was much property. In the 2000s, 20 communes began pi- slower than anticipated in moving towards its loting a large program of conversion to ownership target76, its implementation provided the follow- titles based on the idea that the numerous urban ing lessons: (i) the original idea of strengthening permits to occupy that were initially considered the rights of permit holders was sound; and (ii) transitional and temporary had become de facto this was an extremely ambitious exercise given permanent. Many of these permits to occupy are the legal situation at the outset77, resulting in a old and constitute the only documentation that lengthy and expensive process. Although mea- bona fide occupants hold as evidence of their sures have been taken to improve the timeframe rights. Thus, in 2001, the Government in Benin and correct errors, many cases are still waiting to initiated an operation to convert these permits be processed. 76 In 2009, the initial target of 30,000 title deeds was revised to 15,000 in view of the number of ap- plications that were collected. In February 2016, 6,810 title deeds had been signed since the start of operations, and 4,016 title deeds collected by beneficiaries. 77 It is not possible to describe here all the legal and procedural difficulties encountered under the Benin program. Note however that the main issues revolve around the legal and technical irreg- ularities relating to parcel description; insufficient funding from the national budget, difficulties mobilizing support and senior staff in the administration; lengthy consultations on technical as- pects of the procedure among stakeholders; basic errors in certain technical parceling documents; 77 The link between land, taxation and urban infrastructure should also be strengthened. Confusion over land rights confuses management of local land, making it difficult to use land to generate local resources for local govern- ments. As in Box 6, Benin has tried to avoid the difficulties of setting up a legal cadaster, by relying on the presumption of ownership and occupant behavior. With the help of various donors including the World Bank, Benin also put in place urban land registers – a system for collecting and managing information about land parcels. It generates a map showing the location and addresses of parcels in the city and can be used to establish the tax base for urban areas. The database in- cludes the identity of owners or tenants who are liable to pay taxes (land and non- land) and the characteristics of registered parcels. Although its primary objective was fiscal, the database is also supposed to facilitate land and urban management and provide a planning tool for local authorities (communes). The approach, how- ever, faced various difficulties at the implementation stage given the lack of own- ership and interest from the main stakeholders, especially the commune mayors. Better urban planning will also help remove barriers to private re- al-estate development. Improving the land tenure system will be a prerequisite to ensure a more balanced development of the Greater Bamako Area. This, how- ever, will not be sufficient and it will be important to support the coordination of the different authorities involved in the preparation and validation of urban plans. Given the large fraction of the city that will continue to grow informally, it will be important to have plans that will lay out the future grid of public infrastructure and ensure that urban expansion occurs in way that is consistent with these plans and that funding to support these investments is available (possibly through land taxation). Under an improved land tenure system and improved planning, an important focus will be to identify and have a due process to make land available for formal development consistently with plans for infrastructure provision and spatial urban development, and consistently with good principles of compensation of prior use of the land. The role of the private sector in the provision of affordable housing could also be expanded by reducing the barriers that hinder its develop- ment, especially through the stimulation of long-term finance. Better management of land will need to be integrated with reforms and investments relating to the transport system to ensure accessibili- ty. These can take many forms and do not necessarily require costly investments. Better managing public space can go a long way in increasing speeds as congestion stems in a large part from competing uses of constrained space for circulation. This encompasses several interventions aimed at preserving dedicated space for specific uses: i) sidewalks for pedestrians, ii) space for street vendors so they do not encroach on roads, iii) a parking management strategy to avoid parked ve- hicles from occupying street space, iv) segregated lanes for two-wheelers and as far as possible v) segregated lanes for collective transport buses couple with the creation of bus stops for Sotramas and other collective transport vehicles. In addi- tion, road maintenance to fix potholes can also reduce localized bottlenecks, and rehabilitation or upgrading of the secondary roads network, through paving, could the lack of a reliable information system at the communal level and failure to change names in registers held by certain town halls; the fact that some of the plans and registers sent by certain town halls were unusable; at the end of the process, constraints associated with the formalities that beneficiaries had to complete to collect title deeds; and despite efforts to verify the data, the fact that some of the physical characteristics of parcels (boundaries, location) were still incorrect when the deeds were collected. 78 help reduce the traffic on the primary network and increase average speeds. While the road network coverage is within or above the regional average, lack of paving on the secondary and tertiary road links considerably reduces travel speeds, adds to overall congestion and isolates neighborhoods that are distant from main arter- ies from access to opportunities. Finally, setting-up and enforcing a functioning traffic management system would also greatly improve urban mobility. A focus for policymakers in Bamako, should be on improving the walk- ing experience of people. With such a large share of the urban population travelling on foot to reach services and opportunities (57% of the trips are made by foot in Bamako), it is indispensable to make sure they are safe. This implies a focus on better managing public space, including making sure that sidewalks exist and are not used for parked vehicles, thereby forcing pedestrians to slalom be- tween obstacles. Bamako needs a strategy to deal with the travel of goods which com- petes for the limited road space and increases congestion in the city. Recent decisions have limited the circulation of lorries to evening hours to limit the congestion in the city center and around the main markets. This has provided some relief in terms of daytime congestion but is no panacea as the immobiliza- tion of trucks and lorries remains costly. A more permanent solution might in- volve a combination of a ring road to bypass the city center and of logistic hubs or dry ports in the periphery of the urban area so that only goods destined for Bama- ko’s markets need to be transported through the city. Understanding the failure of large bus companies in Bamako and the operation model of the Sotramas as a pre-condition for potential col- lective transport supply reforms. Large buses have the potential to reduce congestion and be more efficient on main arteries of the urban area where de- mand is concentrated compared to Sotramas. That is because larger buses, when full, occupy less road space and consume less fuel per passenger. Sotramas could complement the large buses by acting as feeder buses or by serving areas that have lower demand density. However, Bamako has a long history of introducing large buses going back to 1960 (Tababus, Bamabus, Diarra Transport) but all these experiences have failed. It is important to understand the reason for these failures and in parallel to understand the reasons for the resilience of the Sotramas in Bamako’s mobility landscape. This stocktaking would shed light on whether there exists a potential for larger buses to circulate on the main arteries of the urban area and what the conditions for success are. In fact, using digital platforms, mo- torcycle taxis have the potential to transform the ‘last mile’ experience for com- muters in Bamako – see Box 7. 79 x bo 7 Motorcycle taxis - a safe, convenient and affordable alternative to Bamako’s existing transport options As Bamako’s population grows and its sub- bring together this multinational corporation urbs swell into the surrounding hinterland, the and Mali’s nascent tech sector and startup eco- city’s road network is coming under increased system, by organizing a hackathon. The result pressure, especially when it comes to the pub- was Teliman, an on-demand transport mobile lic transport and the estimated 600,000 daily app startup, which has raised $1 million from commuters. Regular taxis are too expensive for CFAO and local angel investors to date. By com- the average commuter and the main alternative, bining old and new business models, Teliman SOTRAMA or vans for public transport, are and CFAO are hoping to both secure a supply of uncomfortable and slow. Considering the rising trained and professional drivers and stimulate popularity of motorcycle taxis in many African demand via a convenient app, high-quality ser- cities facing the same urban mobility problem, vice and easy mobile payments. In April 2018, CFAO Motors, Toyota’s African subsidiary, Teliman will launch its first 20 drivers, using decided to conduct a market analysis to see if a them as a prototype to show that the model is motorcycle taxi service was feasible in Bamako. profitable and safe, not only to customers but After months of field studies, the company con- also to potential drivers. By December, a fleet cluded that there was large market potential for of as many as 250 drivers could be on the road, a safe, convenient and affordable alternative to with another 1,250 planned by the end of 2019. the existing transport options. The WBG helped Value for money using Teliman in Bamako: 5 times faster than Sotrama and 4 times cheaper than taxis Moussa lives in the Badalabougou area. Every day, he makes a six kilometers trip, usually at rush hours to get to work at Orabank HQ, in ACI 2000. Option 1: SOTRAMA 75 minutes FCFA 325 Home Work Walk SOTRAMA Walk SOTRAMA Walk Option 2: Taxi 22 minutes FCFA 2000 Home Work Option 3: Teliman 13 minutes FCFA 500 Home Work 80 Bamako also has an opportunity to utilize digital technologies to work with its communities to better monitor and track accessibility chal- lenges. The absence of reliable statistics on urban transport is often a big con- straint in low-income cities, and Bamako is no different. Without such monitoring and tracking exercises, the formulation of urban planning and transportation policies and considerably hindered. Travel surveys78 are the traditional way of understanding mobility patterns in urban areas – but they can be expensive and time consuming to conduct frequently as they require the hiring of large teams of surveyors. Recent technological advances are beginning to make some head- way on collecting useful information at a fraction of the cost. For instance, using anonymized information associated with the usage of mobile phones (referred to as Call Detail Records) sheds light on the location of jobs and people in the urban area, and even travel times. Other initiatives have focused on understanding the supply of collective transport for example for para-transit which is often not well understood79. Mapping these routes and collecting travel times and fares in the form of a publicly available and standard format80 can provide critical information on those urban areas which are under-serviced. Finally, new smartphone appli- cations can register the quality of roads with precise coordinates that can help identify the most pressing needs in terms of road maintenance. 78 Travel surveys usually record origins and destinations of the travelers, frequencies of trips, trans- port modes, travel times and spending. 79 See The Digital Matatus project for example: http://www.digitalmatatus.com/intro_lite.html 80 Over recent years many formal transit agencies (and in some cases informal operators) have ad- opted the General Transit Feed Specification that defines a common format for public transpor- tation schedules. This common format allows for rapid assessment and comparison of transit networks across cities. 81 x bo 8 Using mobile phone data to understand mobility patterns – learning from Haiti Working in data scarce environments locations close to each other) repeatedly over a often requires more innovative sources of 3-month period, this location is deemed to be data and information. To understand acces- “meaningful”. The second step of the analysis is to sibility challenges information on the location determine whether these locations correspond to of jobs in relation to that of people is needed. In “home” or work “locations”, the two places where Haiti, however, the latest census was carried out people are likely to spend most of their time. To in 2003 and there was no business registry. To label the locations as “home” or “work”, a scoring breach this data gap a World Bank team turned to criterion is employed based on the time of day cell phones. It partnered with Digicel – the largest and the day of the week at which calls are placed. cellphone provider in Haiti, and with Flowmind- The underlying assumption is that most people er,– an NGO with vast experience on cell phone spend much of their morning and evening hours data analysis for development purposes, thus with at home and some time over the weekend. Con- multiple stakeholders, alongside government, versely, they are likely to spend a larger part of to handle highly sensitive data. The rich set of their weekday daytime at work. information provided by individuals as they use their cell phones, combined with machine learn- Understanding flows within the city al- ing techniques, allowed the team to get a sense lowed for better urban planning, including of where people live and where they work, key for natural disasters. By getting a sense of information for city planning. where people live and where they work, the World Bank team could understand flows of people Cell phone data allowed extracting infor- within the city’s network. This was combined with mation on where people lived and where information on natural hazards, such as flooding, they worked. Using the Call Details Records to identify the most critical links in the transport (CDR) over a 3-month period, the first step is to network – those that if affected by a natural di- understand people’s “meaningful locations”, that saster would impede access to jobs. Thus, starting is locations that structure the ordinary day of the from a situation of data scarcity, mobile phone cell phone user. The approximate location of a data shed light on the location of the main job user at the time of a call can be ascertained based centers in the city, on the most common trips, and on the cell phone tower that she is connected to. the most critical links of the transport network, When calls are placed from a given location (or when flooding risks are taken into account. 82 Finance and manage better public service delivery Capital and operational investments in Bamako’s urban infrastructure lag far behind those in other low-income cities and countries. Mali will need to contend with 400,000 people migrating each year to urban areas. Beyond land and housing, these people will require access to basic urban services, nota- ble among these is waste collection and disposal. As highlighted earlier, given the rates of urban population growth in Bamako and its neighboring areas, the gener- ation of solid waste in the Greater Bamako area is expected to grow exponentially. The contractual arrangements for managing solid waste are somewhat muddled and provide weak incentives for efficiency. The contracts with Ozone and the GIEs are currently under renegotiation, with the process led by the Government and the District of Bamako. Two key issues are at the center of the discussion: (i) how should annual tariff revenue be collected, and ii) what should be the formal business relationship between Ozone and the GIEs? Under the cur- rent system, all revenue is collected by GIEs directly from households, and GIEs operate independently of Ozone. Under the new system, it is proposed that a FCFA 3,000/concession/month tariff will be collected by 165 collection workers who currently collect the FCFA 3,000/year/adult local development tax for the district. These funds will then be remitted to Ozone, which will in turn will pay the GIEs, either as employees or sub-contractors. However, under this system, the proposed tariff will only cover the operational costs of the GIE, and very little of the costs required to pay Ozone – leaving a US$ 16.6 million deficit. The new system would not address the existing operational deficit. It would also add multiple intermedi- aries between the GIEs and their customers, potentially threatening the economic viability of the GIEs themselves, and thus the entire pre-collection system.81 This section will outline strategies to simultaneously mobilize resourc- es and reduce inefficiencies in the sector, and for better contractual management, aimed at improving service delivery. Although the focus is on the solid waste management sector, many of the recommendations and learn- ings also pertain to other sectors such as water, sanitation and others. How to mobilize more revenues for investments? Bamako must develop systems to increase its own-source fiscal and other revenues. These could be to better finance recurrent costs to serve a grow- ing city, but also to be able to pay for development outlays to invest in the future. In the case of solid waste management, there are three possible ways for the city to mobilize additional revenues – some involve tinkering at the margins, while others will require large scale reforms. • Increase tariffs: The tariff for waste collection should be increased from 3,000 FCFA/plot/month to 3,000/household/month. Since there are an average of 81 This is for two reasons: (i) Ozone may delay payments to GIEs, to cover missed payments from the city, and (ii) GIEs will no longer be paid by those who directly observe quality of service, and so have reduced incentives to perform. 83 2.1 households per plot (see DRPSIAP/DB, 2014) 82, this would more than halve the operating deficit, generating an additional US$ 10.5 million in revenue, and reducing the deficit to just US$ 6.2 million annually. There is evidence (Arter- lia 2015)83 that willingness and ability to pay among many households may be even higher - some households currently pay 5,000 monthly, and anecdotally some GIEs report collecting as much as FCFA 10,000 monthly per plot from larger waste producing households. Figure 37 shows current tariffs and re- covery rates reported by GIEs84 for five neighborhoods in Commune IV. If all households had the same willingness to pay, one would expect this curve to slope downwards. Instead, it slopes upwards, as households in Sébénikoro and Hamdallaye are both willing and likely to pay more. Thus, to maximize reve- nues, any new tariff regime should continue the differential-pricing business model employed by the GIEs. In addition, a system in which GIEs are empow- ered to charge based on weight (for instance by requiring that disposal only take place with authorized bags) would also help standardize such a business model across households. Figure 37. Higher tariffs do not necessarily imply lower recovery rates Recovery rate for waste fee 90% Sébénikoro 80% Djicoroni Para 70% Hamdallaye Taliko 60% Lassa 50% 40% 30% 20% 10% 0% 1200 1400 1600 1800 2000 2200 2400 Monthly fee (FCFA) Source: Coordination des Partenaires intervenant dans l’Assainissement (CPAC) en Commune IV • Tax Plastic: There is evidence indicating that taxes on importers and produc- ers of plastic to cover waste management costs can be an effective policy. Un- der conservative assumptions, a value tax at 5 FCFA/plastic bag could yield an additional US$ 1.7 million in revenue, reducing the deficit to US$ 4.5 million85. Such initiatives have been successful in Morocco (under the label of ecotax), 82 Direction Régionale de la planification de la statistique et de l’informatique de l’aménagement du territoire et de la population du district de Bamako (DRPSIAP/DB), (2014) « Annuaire Sta- tistique du District de Bamako : Année 2014 ». The increase in revenue would come from billing per household rather than per plot. As there are more than two households per plot, the revenue would potentially increase correspondingly. 83 ARTELIA (2015), « Étude des aspects organisationnel, institutionnel et tarifaire liés à l’assainisse- ment urbain : Diagnostic Institutionnel et Organisationnel Volet Stratégie De Gestion Des Déchets Solides de Bamako » 84 Small collection companies - groupements d’intérêt économique. 85 This calculation assumes 500,000 bags/day consumption. According to UNCTAD data, Mali im- ported $74.5 million in plastic and articles thereof in 2014. A 5% tax on this would yield a greater 84 where a special account was established to ring-fence revenue after collection by customs. Other West African countries have instituted or are establishing similar measures (e.g. Senegal, Benin) and the cost-benefit and social accept- ability should be explored further for Mali. • National Government Transfers: Another option would be for the national government to provide an annual transfer to the city to cover the remaining deficit. Such a scheme has been established successfully in Tunisia, where the government has established a dedicated fund which pays cities based on the quality of performance. Establishing such as system would require further ex- pansion of the city’s capabilities to monitor waste collection quality (i.e. collec- tion rates) and the extent of enforcement of unauthorized dumpsites. Given po- litical economy concerns associated with raising tariffs at the household level, a standing government transfer may be required to cover costs in the near term. How to reduce inefficiencies in the system? Bamako should also focus on the efficiency and opportunity costs of local government expenditures. Although a focus on increasing revenues is important, local governments can also focus on better management of their expenditures. Thus, an alternative approach to reduce the operating deficit is to reduce costs. While the best options available achieve only small savings relative to the overall deficit, the low investment costs associated with them suggest that they are likely still positive net-present value investments and should be consid- ered. Two options are most promising: • Study feasibility of constructing a new landfill, nearer to the city: A substan- tial share of the total costs (US$ 3.1 million, or 18%) borne by Ozone – the firm currently contracted to collect waste in Bamako - are estimated to cov- er the fuel, labor, maintenance, depreciation and overhead associated with transporting waste from transfer depots to the final dumpsite, currently the Hôtel Olympe site within the city. These costs are high in part because of a lack economy of scale, as Ozone uses smaller capacity trucks (i.e. 7 m3). Given that the alternative site - the Noumoubougou landfill - is quite far (42 km) from the city, and even assuming larger trucks can be used (say, 24 m3) to achieve more economies of scale, it is expected that transport costs will increase to US$ 4 million once the site is in use. This difference is observed when comparing Options #1 and #2 (see Figure 38) which shows a breakdown of transport costs under different scenarios. Building a landfill nearer to the city in the identified site in Dialakorobougou, shown under Option #3, at just 21 km away, or half the distance, could reduce costs relative to Option #2 by US$ 1.9 million an- nually. With the costs of a sanitary landfill approaching US$ 8-9 million, this implies a pay-back period on such an investment of just 5 years when valuing the savings in transport costs alone. Such a payback period means such an investment would have positive net-present value, even at a 20% discount rate. From an environmental perspective, opening an additional landfill would also allow to definitively close the active (Hôtel Olympe) and closed (Doumanza- na) non-sanitary landfills within the District and this option should be studied more closely. amount, $3.7 million, but would also be difficult to justify since not all articles of plastic are dis- posed of as consumer waste. 85 Figure 38. Annual costs are expected to increase by $900,000 if Noumoubougou is used (Scenario 2), whereas the use of a nearer landfill and a composting pro- gram to reduce waste volume could reduce costs relative to scenario 2 by $2.4 million Anual cost of transportation USD Millions 4,0 Overhead cost 3,5 Labor Depreciation and truck maintenance 3,0 Fuel and oil 2,5 Operating surplus 2,0 1,5 1,0 0,5 0,0 1. Hotel Olympe 2. Noumou-bougou 3. Closer landfill (21 km) 4. Closer landfill (21 km) + Landfill (10 km) Landfill (42 km) compost 50% of organic waste SOURCE: World Bank staff calculations. Overhead is assumed to be 20% of depreciation, fuel and labor costs, and the profit margin is assumed to be 30%. Labor costs are higher under Scenario 1 because smaller capacity trucks are assumed to be used for disposal within the city. Ozone currently has fewer than 5 working trucks with capacity of 24 m3 that are assumed to be used for transport in scenarios 2-4. • Compost 50% of organic waste within the city: While recent waste composi- tion data is not available for Bamako, according to Bhada-Tata and Hoornweg (2012), typically 50% of waste in Sahelian countries, and indeed across Africa, is biodegradable. A scheme to compost 50% of organic waste within the city would reduce the volume of waste shipped to landfills by 25%. As shown in Op- tion #4, this would reduce annual transportation costs to $1.6 million, saving $0.5 million annual relative to Option #3. Given these savings, investment in a city-wide compositing program should be explored. Commune VI currently operates a small-scale composting operation led by a GIE that sources organic waste from ~100 plots, mixes it with animal manure at a 50/50 ratio, and then sells on the compost to horticulturalists in the city. Initial reports suggest that to fully equip the operation would include investing in a crusher to grind the compost, rehabilitating storage space and a covered work area, and adding piped water, a crucial input into the process. This would probably amount to an investment of approximately FCFA 10,000,000 (~US$ 19,000) per operation, which is not large given the savings from reduced transportation costs. It is likely, however, that some operational subsidy will be required on an on-going basis to sustain the business. Managers in Commune VI reported that they currently rely on the unpaid labor of approximately 20 women for operation. To pay them wages (which @ FCFA 70,000/month would equal ~US$ 31,000 annually) does not appear possible under current margins, which are very thin. Such a subsidy, however, would still be small relative to the savings from 86 reduced transportation costs. Further benefits from increased female employ- ment will also increase returns. How to better organize contractual arrangements? Bamako should consider a contractual arrangement in which op- erators are paid based on performance, by those who observe their performance directly. The current contractual arrangement in which GIEs are independent and collect fees directly from households satisfies the theory of contracts86. As contractors add more customers, thus increasing the collection rate, they capture any new fees associated with that effort. And households can directly verify the performance of GIEs. Under the new, proposed arrangement, GIEs would be paid directly by Ozone and it is not clear how their performance would be tied to pay, since Ozone has no direct way of measuring whether a GIE collects waste from a given household. Further, under such a system, willingness to pay by households might also fall since the link between payment and service would be less clear. Figure 39. Under the ideal contract, operators are paid based on performance, by those who observe their performance directly LEGEND Collection chain Households GIEs Funds Info on quality of services Transfer station Institutions Ozone Landfill Bamako District SOURCE: World Bank staff 86 The theory of contracts underlies two basic principles of an optimal contract: (i) The contractor (i.e. the person doing the work) should be a “residual claimant” on his or her marginal effort. Specifically, this means he or she should receive 100% of the value from any additional effort, so that he or she has a complete incentive to work hard (ii) Performance under the contract should be verifiable by the payer, so that the payer can indeed pay the contractor extra if he or she does works hard. For further detail see Bolton and Dewatripont (2005). 87 Figure 39 illustrates an alternative contractual arrangement, which would maximize incentives for collection and payment. Under such a system, GIEs would continue to collect revenues directly from households. Ozone would collect some revenues from large institutional producers, and be paid di- rectly by the city based on its performance in evacuating transfer depots (which is currently being monitored by DSUVA). Under this system, there would be a clear link between performance and pay for each party. Further investments in per- formance monitoring capabilities would make such a system more viable. Imple- menting such a system, however, would require sourcing additional revenue for the city sto pay Ozone. To collect higher tariffs to cover the operating deficit a license fee model for the GIEs could be considered. To increase revenue and to allow GIEs to continue to collect fees directly from households so that they are fully incentivized to deliver services, the following scheme should be considered. In ex- change for the right to operate and collect fees, GIEs would pay an annual license fee to the district equivalent to 50% of 3,000/plot/month for all households in their quartier87. To keep their existing profits, GIEs would need to double collec- tion, raising an estimated US$ 4.9 million. After paying the fee to the district, all profit from marginal effort would go to the GIE, and households (those who ob- serve the quality of service) would pay for the service. Further, the city would not need to invest additional resources in collecting revenues from households. Such a scheme has been successful in Liberia in raising additional revenue for the city. Designing such a scheme would require careful design of the license fee structure so that GIEs remain economically viable and are able to generate jobs for youth (see Box 9 on the experience in Conakry). To further incentivize GIE participation, in exchange for the license, the government could also develop funding mechanisms for GIEs seeking to improve their equipment. Estimates by COGIAM suggest that 112 GIEs need tractors to upgrade their businesses and collect more waste. At FCFA 11.75 million per tractor, this implies a total cost of FCFA 1.3 billion (US$ 2.5 million), a relatively small amount given the potential for improvement in service delivery. The capital cost associated with guaranteeing loans for this amount from banks would be lower. A guarantee scheme may be preferable to current proposals to capitalize the GIE, which involve Ozone purchasing these tractors and lending or giving them to the GIEs. If the GIEs are full own the tractors, and, through fee collection capture all the marginal value created, they will be best incentivized to maintain the equipment. Currently, however given outstanding arrears currently, Ozone appears unwilling to capitalize the GIEs. Any guarantee (or directly lend- ing) scheme to finance tractors for GIEs would have to be negotiated carefully with the GIEs, to ensure that the license fee charged did not hinder their ability to cover loans to finance tractor purchases. An experimental approach could be used to identify which contractu- al arrangements work best to deliver quality of service and increase revenue collection. The arguments above rely on economic theory, and the performance of different contractual arrangements may vary in the real world. As the city considers how to move forward with the GIEs and Ozone, it could take an 87 This scheme is equivalent to a franchise fee in industrial organization: In order for the right to op- erate, the franchisee pays a fee to the franchiser, but keeps any additional profit from the business beyond that fee. 88 experimental approach in which it pilots certain contracts with subsets of GIEs. For instance, 60 GIEs could be selected to continue operations as is, 60 could be selected to operate as employees of Ozone and 60 to operate under a license fee arrangement, as described above. If the GIEs were selected randomly, as in a randomized control trial, performance and revenue collection outcomes could be compared across groups to identify which contract performs the best. Once this is known, the optimal contract could be scaled to the city. 89 x bo 9 Building and maintaining a solid waste system – ups and downs in Conakry Good SWM results were achieved through Creation of a new sector of the urban econ- long-term engagement and strengthening omy and job generation. The SWM industry SMEs. Guinea has a long history of improving generated about 3,300 jobs and the total annual waste collection. The World Bank supported this turnover of the SMEs was about US$415,000, or process through a series of three Urban Devel- about US$1,000 per SME, at an affordable sub- opment Projects, which laid the foundation for a scription rate of US$ 0.375-US$ 1.15 per month professional collection service through compre- per household. The construction of transfer hensive support to urban management and SWM stations helped to improve profitability for SMEs in particular, which included capacity support to since it reduced the average distance between SMEs carrying out primary collection and estab- households and transfer stations. lishment of an SME equipment financing fund, institutional strengthening of the municipal SWM Political crisis and departure from estab- department, construction of transfer stations, lished principles led to decrease in system upgrading of secondary roads to facilitate access performance. The advent of the socio-political for waste collectors, a large communication cam- crisis experienced by the country between 2009 paign, and implementation of a street addressing and 2011 had as a side effect a reduction in the system which helped keep track of the customer city’s capacity to ensure transport of waste from base and collection zones. transit stations to the landfill and broke the trust between the SMEs and the customers, since the High collection rates through a two-tiered waste collected was no longer removed regularly system. The system allowed a full coverage of from transfer stations. A complete privatization of the city and 85% of solid waste was disposed at the collection chain decided by the Government the landfill, a very high rate by regional standards. in the aftermath of the crisis was not based on a The overall sanitary conditions and habits of res- sound preliminary analysis of the technical and idents improved. Household waste was collected organizational prerequisites, the financing needs, by 35 SMEs (14 led by women) in the 44 collec- and the incentives for results. Due to the lack of tion zones of Conakry that had been identified by clarity on the organization of the collection chain using the street-addressing database. The SMEs in the new format the lack of oversight and regu- collected payments from about 115,000 house- lation of the private sector provider, the collection holds (a 73% subscription rate). The municipal rate decreased to 27 percent by the end of the waste department transferred waste to the landfill project. by container trucks and the transfer stations were Sources: World Bank (2007), ‘Third Urban Development monitored to keep track of performance in each Project’, Project Appraisal Document, P091297, Washington link in the collection chain. DC: World Bank. World Bank (2013), ‘Third Urban Development Project’, Implementation Completion and Results Report, P091297, Washington DC: World Bank. 90 Invest in Urban Institutions – Think outside the Box Adequate capabilities for planning are required for effective provision of services, not least in the expansion areas of the District. As noted by Lall et al (2017), “urban planning and land use regulations are central for en- hancing urban connectivity, productivity, and livability because of externalities and coordination failures. […] Well-functioning cities require that economies of scope and complementarities be leveraged in the provision of physical infra- structure” 88. While urban local governments in the Greater Bamako Area have (also partly outdated) spatial plans, in the absence of an integrated plan for the metropolitan area, the only plan to guide the development of the District is the Social, Cultural and Economic Development Plan that is mandated for all local governments. This document does however not contain any spatial vision or pro- visions for land use. No detailed study has been done for land consumption on the fringes of the Bamako urban footprint, but evaluation of larger secondary cities in Mali displaying population trends like Bamako (Kayes and Sikasso) have shown that land set aside for urban extension over the 20-year duration of the urban plan was consumed after 5 years89. The challenges associated with land use and urban expansion (outlined earlier) will only be exacerbated by the lack of capac- ity to plan urban development and provide services in the District and its exten- sion areas in a coordinated manner. The Greater Bamako Area needs an urban structure plan that clearly delineates priority development areas and can serve to align investments to these objectives, e.g. densification in the center and along transport corridors through road paving and specification of areas for provision of water and electricity. Institutional responsibilities should be clarified and capabilities at the local level should be reinforced. In Bamako, there is much room for the mayors of the District and the neighboring communes to better utilize their mandates effectively, given that sub-national government is developing serious teeth90. At the same time, institutional mechanisms are needed to clarify respon- sibilities and to help coordination91 – between the local governments, and with the national government. There is very little capacity at the local level, which lack resources and staff - the total number of urban planners in Mali is 20, or 0.13 per 100,000 people (compared to 0.97 across Africa)92. The District of Bamako and neighboring communes must invest in their capability to address regulatory and investments needs associated with larger development challenges. 88 Somik Vinay Lall, J. Vernon Henderson and Anthony J. Venables: Africa’s Cities. Opening Doors to the World. World Bank Group, Washington DC, 2017, p.131. 89 Projet d’Appui aux Communes Urbaines du Mali: ‘Evaluation de schéma d’urbanisme de la ville de Kayes et environs’, 2014 and Evaluation de schéma d’urbanisme de la ville de Sikasso et envi- rons’, 2014. 90 This is owing to regionalization and the drive to devolve public resources to sub-national govern- ment, both being part of the political deal made to gain peace in the north. 91 For instance, in the land sector in Mali central authorities continue to initiate lotissements when the law stipulates the mandate falls to the communes. 92 Authors’ calculations based on: The State of Planning in Africa. African Planning Association and UN-Habitat, 2014. 91 x bo 10 Less is More – strengthening municipal capabilities in La Paz In the early 1980s, the capital city of Bo- with the personnel restructuring, permitted the livia, La Paz, was in shambles—the national city government to cover the new payroll from fiscal subsidy to the city had collapsed, the city its own revenues, without further external assis- payroll represented 120 percent of local revenues tance. and services were in disarray. The city systems for public works, tax collection, permits and licenses, The reforms in staffing and remunera- payments, and procurements were overstaffed, tion of key positions were accompanied hugely inefficient, and a goldmine of corruption. by sweeping changes in systems and pro- cedures affecting public works (for example, La Paz managed to attract and retain privatization of the city construction company), high-quality professionals for better ser- procurement (cutting down the 26 steps pre- vice delivery. In 1985, with a new administra- viously required), tax collection (reducing the tion, the city embarked on a project to modernize number of municipal taxes from 126 to 7), and and rationalize the city administration, stream- public information (for example, advertising the line staffing, and reform procurement practices, requirements to obtain a license). For the first including by outsourcing public works. Besides time, citizens were asked to be part of an effort to providing a timely infusion of investment funding improve the quality of municipal services. These and technical assistance, the project included a measures had the salutary effect of dramatically novel feature that proved essential to the suc- drying-up much of the worst corruption affecting cess of the reforms: a grant component of US$ the city government. 1 million that was used to top-up the salaries Sources: World Bank (2006), Infrastructure at the Cross- of high-quality staff recruited for key positions, roads: lessons from 20 years of World Bank experience, while some 40 percent of the existing municipal Washington DC: World Bank, pp. 94-95; using “Municipality personnel were let go. The grant permitted this of La Paz” Case Study for the Kennedy School of Govern- critical step to be initiated, but, after the first year, ment, Harvard University, by WBI. improvements in fiscal administration, coupled Viewed from any perspective, the six communes and the District of Ba- mako need more financial resources to fund both their regular service delivery mandates as well as some of the growing infrastructure needs of the city. There are two main sources of revenue available to Bamako’s local governments: Increasing own-source revenues (OSRs): OSRs are either collected by the national tax directorate (on behalf of local governments) and then “shared” or wholly allocated to the District and communes or collected directly by local rev- enue administrations. In the short term, the scope for improvements in revenue administration is probably greatest with respect to the latter, where the District and communes have more room for maneuver. Here a range of options exist (many of which have been piloted in Mali or elsewhere) to improve revenue ad- 92 ministration and thus to increase revenues: modernizing tax assessments/census- es by using new technology, introducing better and more comprehensive billing systems, making payments easier. Where revenues are collected by the national tax directorate (or its local offices), the District and communes still have a role to play by assisting in the identification of tax-payers and in tax collection, and – most importantly – in working actively to promote “willingness to pay” (by being as transparent as possible, delivering services, and the like). In the medium term and strategically, own-source revenue arrangements in Mali (and thus Bamako) would clearly benefit from serious reform, as highlighted by numerous technical reviews and assessments. Much of this is well known in Malian policy circles; but to date, reforms have been slow in taking off. x bo 11 Improving local government own-source revenues in Mali – what the IMF says and recommends Local government own resources are lim- financial resources in the assessment and collec- ited. Their own resources are broken down into tion units, contradictory role of elected officials in extremely diverse categories of revenue (tax and the collection of local taxes, insufficient motiva- nontax). However, the corollary of that diversity tion of officials, and tax avoidance. is that revenue levels are very low. On the tax rev- enue side, there are no less than 21 different taxes There is significant room for improvement and levies … of which 3 (the business license tax, in local taxation. The Malian government is the local development tax, and the tax on wages considering various areas for improvement [such and salaries) represent 77.3 percent of the total as]: revenues collected in 2012, while 12 others repre- Streamlining taxation: sent just 3.4 percent of this total. The findings are similar for nontax revenues (revenues from land • Review numerous existing taxes against their services and miscellaneous sales). More than 80 revenues and costs, and eliminate unprofitable percent of nontax revenues collected by the local taxes in order to concentrate on profitable ones governments related to 13 categories of revenue (taxes on: business license, local development, in 2012, while 30 other categories represented 1 and salaries); percent or less of total nontax revenues. This frag- • Expand the tax base of the local governments; mentation is a source of complexity and bureau- cracy in the collection of revenues and helps to • Identify and list new taxpayers; explain the low productivity of local governments’ own resources. • Reconsider the base for the regional tax for local development (move from a per capita tax The local taxation system is cumbersome. to a property tax in urban areas only); There is some disorganization in the process for the assessment and collection of local taxes and • Subject to the establishment of reliable proper- levies. The weaknesses [include] poor manage- ty registries, create a property tax. ment of the tax rolls, insufficient coordination Source: IMF (2015): Mali – Implementing Fiscal De- between units, insufficient material, human and centralization 93 Transfers from the central government: Grants and transfers from the cen- tral government to the District and communes of Bamako are already a significant source of total revenue. However, a good proportion of those transfers appears to be earmarked for specific sectors; the use of those funds is not left open to local discretion. The central government also provides capital grants (in the form of “drawing rights”93) to local governments through ANICT (Agence Nationale d’In- vestissement des Collectivités Territoriales), funding for which is largely provided by development partners. In 2016, ANICT made CFA 22.7 billion (roughly US$ 42.5 million) worth of investment drawing rights available to all local govern- ments in Mali. ANICT, however, is intended to be a nation-wide system, with a mandate to ensure that investment grants are allocated on an equitable (or equal- izing) basis across the country. As a result, the combined 2016 drawing rights for the District and communes of Bamako amounted to around US$ 4.4 million, representing a little over 10% of the total ANICT allocation for investments, even though their combined population makes up around 14% of the country’s total population. ANICT, understandably given its equalization mandate, does not factor in the high cost of urban infrastructure investments nor the wider economic spillovers from infrastructure investments in Bamako. Nonetheless, an adapted “urban” form of ANICT (or a special urban funding window within ANICT) might provide Bamako’s local governments with access to greater levels of investment capital. A final issue that merits consideration is the need to encourage more efficient public spending by the District and communes of Bamako. As mentioned earlier, the District and communes spend most of their budgets on recurrent items, and especially on payroll costs. Although well beyond the scope of this review, it is more than likely that some of this recurrent spending is either wasteful or un-warranted. The District and communes could be incentivized to review and reduce their recurrent expenditure, by making additional or future investment grants conditional on such actions. There is an urgent need to develop systems and incentives for inter-ju- risdictional coordination on planning for infrastructure investments and service provision. Bamako is becoming integrated with Kati and Koulikoro into a multi-city agglomeration, which entails specific governance and coordina- tion challenges at the regional level. Municipal waste, an important by product of an urban lifestyle, is an excellent illustration of the need for inter-jurisdictional coordination. Solid waste management in the Greater Bamako area requires coordination between the district of Bamako and the surrounding communes, which would help generate economies of scale in collection and disposal, includ- ing the development of projects such as biogas or landfill gas electricity generation facilities. The lack of clarity of responsibilities has been reflected in the failure to coordinate planning for the provision of this crucial service. The multitude of stakeholders make collaboration difficult, but still possible. There is a history of failure to reduce the institutional overlaps in Mali. The Bank-funded Urban Local Government Support Project was restructured when the expected administrative integration of the District and the six com- munes, for which legislation was being prepared in 2009, did not take place and 93 ANICT’s drawing rights are not, strictly speaking, grants. In order to use their drawing rights, local governments must submit investment proposals, which may or may not be accepted by AN- ICT. In practice, then, Malian local governments often do not use their full drawing rights. 94 support to the process was no longer relevant. However, both the local govern- ment law and the urban planning law allow for horizontal and vertical collabora- tion across local and national governments. Disparate coordination attempts have been made and the lessons learnt from these experiments could be the precursor to single-purpose districts. To exemplify these problems and show the potential benefits of coordination, the delivery of solid waste management services taken as a case in point. x bo 12 Box 12. Metropolitan Governance Arrangements – from voluntary to consolidated Various institutional approaches have Metropolitan or regional authorities. These been identified to assure governance of are voluntary organizations where a regional metropolitan regions. Each have advantages authority is an independent legal entity. Two or and disadvantages and the historic trajectories more local governments may associate in this way are often major determinants of success. When to achieve economies of scale, for example for a local governments join forces on opportunistic transport network or to jointly operate a waste basis this can result in “asymmetrical” arrange- disposal facility. Regional authorities, sometimes ments, where service provision is provided or established as utility companies, can usually levy coordinated differently in different sectors, to user charges for services provided, and/or collect make jurisdictional spillovers work for a given from the member local governments. Metropoli- service. Most forms of collaborative governance tan or regional authorities can also take the form arrangements are horizontal, but more vertical of a metropolitan council of governments and arrangements also exist. they can be established for planning or advisory purposes only, or for planning as well as service Inter-municipal cooperation mechanisms. delivery in one or more sectors. The New York These arrangements may take various temporary Regional Plan Association, established in 1929, forms (committees, working groups) or more is an example of an NGO having had substantial permanent networks, associations, and consor- influence as an urban research and advocacy or- tiums. Cooperation can concern a specific issue or ganization. The successes were achieved through investment project but more broad-based collab- a pragmatic and cooperative strategy with other oration is also common. For example, Brazil has a civic groups, governme nt, business and commu- separate legal framework that encourages the for- nity organizations. mation of local government consortiums, which in some cases can become entities somewhat Second level metropolitan local govern- similar to regional agencies. In the metropolitan ment. These arrangements may take the form of area of Los Angeles, USA, the Los Angeles County a higher-level government, by far the largest, provides a variety of services to the metropolitan area’s numerous metropolitan local governments or regional small local governments on a contract basis. governments. Such local governments would not necessarily be hierarchically above the other local governments in the area in terms of reporting 95 relationships, but possibly of equal rank and legal ii. Build metropolitan ownership among key status. For example, the Dar es Salaam metro stakeholders. government has no authority over other local governments, while the mayor of London, elected iii. Ensure national government support. directly, has substantial authority over the area’s iv. Tailor reliable sources of metropolitan financ- lower-level local governments. Experience shows ing. that such institutions will often be weak unless they are supported by the lower-level local gov- v. Design incentives and compensations for met- ernments. ropolitan compromises. Consolidated local government. Annexation vi. Implement a long-term process of metropoli- or amalgamation can under the right circum- tan monitoring and evaluation. stances be effective to achieve efficiency and equi- It is also important under the various arrange- ty in public service delivery, reducing institutional ments to clarify and communicate who is respon- complexity. A jurisdiction that covers a large por- sible for what and to make authority coincide with tion (or all) of the metropolitan area can facilitate representation. Finally, finance should follow equalization in the area since it would have one function, which means that any entity established tax base. The municipal reforms of South Africa to coordinate or provide services to a metropol- are a case in point. For example, the present Cape itan area should ideally be represented by, and Town one-tier municipal government resulted accountable to, the corresponding entire jurisdic- from amalgamation six local governments over- tion and receive corresponding resources. laid by a metropolitan administration and incor- poration of other local governments in the metro- Substantially based on: GIZ and UN-Habitat (2015), Un- politan region to form a single local government packing Metropolitan Governance for Sustainable Develop- ment. Eschborn: GIZ. covering 2,461 square kilometers. This has facili- tated equalization of access to services and spatial Other sources: OECD (2015), The Metropolitan Century: and integrated planning at a metropolitan scale to Understanding Urbanisation and its Consequences, OECD- Publishing, Paris. address in a comprehensive manner the planning challenges and inequalit ies that are the legacy of World Bank (2014): Municipal Finances. A Handbook for Lo- the apartheid area. cal Governments. Catherine Farvacque-Vitkovic and Mihaly Kopanyi, Editors Washington DC: World Bank. Factors of successful metropolitan gover- Regional Plan Association (2017): http://www.rpa.org/ nance: A number of steps for successful met- Accessed November 4, 2017. ropolitan governance arrangements have been identified as follows: Interview with Andrew Boraine, CEO of the South Africa Western Cape Economic Development Partnership. February i. Motivate collaboration by identifying concrete 26, 2018. metropolitan projects. City of Cape Town (2018): http://www.capetown.gov.za. Accessed March 5, 2018. 96 Stakeholder engagement must be a crucial element of addressing ur- ban infrastructure financing and governance. Major infrastructure proj- ects are complex and are often disruptive in the construction stage, and often local communities do not wholly understand the nature of infrastructure planning and development. For an example of how to effectively leverage local communities and stakeholders into the process of urban governance – see Box 13. x bo 13 Bringing citizens into governance – participatory budgeting in Brazil In 1989 the city government of Porto There were several innovations in the Alegre—a Brazilian metropolis with 1.4 actual practice of holding meetings. For million in habitants—initiated a project of example, to minimize inequalities resulting participatory decision-making on the city from differences in speaking ability, short inter- budget, with the aim of strengthening participa- ventions were encouraged rather than formal tion through the creation of a public space for the speech-making. Authority within meetings did expression of citizens’ demands. Citizen assem- not come from education or class, but involved blies in each city district began determining pri- other sorts of social status such as respect within orities for the use of a part of the city’s revenues: the community, often linked with membership ‘ordinary citizens’ made binding decisions on sev- (or even leadership) in various local groups. The eral areas of governmental action, most notably concept spread rather cautiously at first, but then those affecting the city’s new capital investments. expanded rapidly since 1996 and is now adopted by about 180 Brazilian municipalities, as well as The planning process depended on citi- some municipalities in Argentina, Uruguay, Peru, zen participation at each stage. Once a year, Ecuador, Colombia, Bolivia, Mexico, and Chile. in March, a series of preparatory meetings are held in which municipal officials report on what Several clear benefits were observed. In has been accomplished with the previous year’s Belo Horizonte—which has a special participatory budget. Then, in April-June, large city assemblies process for its housing programs and land use and decentralized neighborhood ones discuss and changes—land invasions have declined precipi- vote on spending priorities, electing delegates to tously. Holding meetings at community level also the Council of the Participatory Budgeting as well allowed more women to participate since they as to thematic committees. In July-August, city tend to avoid meetings away from their communi- experts, together with the assemblies’ delegates ties. Likewise, the cost of attending such meetings help in translating the demands into projects. The is significant for lower income citizens, in terms delegates meet and prepare a General Proposal on of transportation costs and the opportunity cost the Budget and a draft of the Investment Plan that of time spent. Concomitantly, the location of will be discussed with the City Council and then projects is related to participation, which in turn approved. Some decisions are made in conjunc- is related to household income. In all municipali- tion with neighboring communities: for example, ties, the proportion of investments serving lower when two communities realize that if they both income communities has increased. rank a clinic first, and thus gain more funding, Based on: della Porta, Donatella & Mario Diani (2006), they can get a better clinic. Overall, participation Social Movements: an introduction, 2nd ed., Malden, MA: increased from less than 1,000 people in 1990 to Blackwell, pp. 223-226. more than 30,000 in 2002. IADB (2005), ‘Assessment of Participatory Budgeting in Bra- zil’, Washington DC: Inter-American Development Bank. 97 Bamako has an opportunity to leverage digital technologies to give its citizens access to decision-making processes in local governance. An exciting development in e-participation tools is the ability to crowdsource, through an online open call for data or proposals for an inquiry or a problem. Planners and citizens can crowdsource information from citizens about plans, policies, the conditions of their neighborhoods and their ideas to address urban problems94. Cell phones allow crowdsourcing through SMS or through mobile-ac- cessible social networks. See Box 14 for an example of how simple data platforms, that leverage local communities, can have an impact on better availability and understanding of geospatial data for decision-making. x bo 14 Use of Geospatial data in Niger Niger has already implemented a Geospa- have been developed. First, users can update tial Information System leveraging com- their data in an excel file which is saved in a munity involvement (www.risques-niger. shared folder. This data is synchronized instant- org). The system includes the following features: ly when the computer is connected to internet. Second, data is collected through smartphones. A geocatalog referencing and disseminating Users can enter any type of data (i.e. points, available data (dissemination is protected by polylines, polygons), which is synchronized and group of users). All data can be displayed on a integrated automatically through internet. The map. capacity of national actors in the collection, A geoportal allowing to display several layers integration and dissemination of information of geographical information from national and is also being reinforced. Complementary tools international sources such as daily rainfall, land are being implemented such as SMS messaging use etc. in case of special event, alerts for rainfall above certain thresholds by distance to points of inter- An offline version also allows users to pre-down- est,3D visualization of areas affected by a flood load data for a specific geographical area to run and so on. the analysis offline. A beta version of the platform for Mali is avail- Concurrently, to ensure with a constant able at the following address: http://ml-risk. flow of up to date information, different pigeosolutions.fr tools leveraging community involvement 94 For examples see Silva, Carlos Nunes (2013) Citizen E-Participation in Urban Governance – Crowdsourcing and Collaborative Creativity. 98 Bamako can also leverage competitive markets alongside regulation to expand basic services. To promote entrepreneurship, policy makers need to remove constraints to business development and support light manufacturing sectors in urban centers. Local governments are often in the lead when experi- menting with ‘smart solutions’ to crowd in private sector players to help tackle urban challenges. In all three countries, there is a large, untapped potential to pilot projects that could demonstrate intermediate success in using digital tech- nologies to address urban challenges, and which would have potential for scale-up in the future. In cases, where urban service provision may be limited in the short- term by resources and capacity, local governments could facilitate crowding in of private sector firms to provide ideas and solutions – see Box 15.95 x bo 15 Creating Markets for Service Delivery – example of sewerage Social enterprises in many, low-income, The business model usually involves a mix cities and countries can serve poor house- of private companies, NGOs, local entre- holds, sometimes also through off-grid preneurs, and, often local government solutions. Millions of people at the bottom of agencies. Under a typical serviced toilet model, the pyramid live in communities that are not the enterprise supplies and installs the toilet and connected to the sewerage system. Most of them evacuates and often processes the waste. For use pit latrines or engage in open defecation. To community toilets, local franchisees or opera- address the problem, dozens of social enterprises tors usually service the toilets, keep them clean, have devised innovative toilets that require no collect fees, and sell other services at the toilet water. These low-cost in-home or community site. Ecotact in Kenya, for example, operates units represent a huge improvement over unim- “toilet malls”, which also serve as retail outlets for proved sanitation and create jobs for people who necessities such as prepaid mobile cards, snacks, install and service them. Serviced toilets are com- and shoe cleaning services. Sanergy (a firm in pact toilet units that can be used in homes and Kenya serving low-income communities) supplies communities without access to centralized sew- prefabricated toilets and evacuates and processes erage systems. Clients usually pay for using the waste. An NGO operates the toilet network, pro- toilets but do not own them. The company that vides training, and supports more than 300 local owns the toilets empties them, treats the waste entrepreneurs, who earn steady incomes from and converts it into fertilizer or fuel. their business. Local municipalities are usually involved in waste management to dispose of the waste at local plants. 95 Esper, Heather, Ted London, and Yaquta Kanchwala. 2013 Improved Sanitation and Its Impact on Children: An Exploration of Sanergy. William Davidson Institute, University of Michigan, Ann Arbor, MI. 99 100 101 Conclusion Given its importance to Mali, and its potential to become an inland hub in the West Africa region, Bamako has much to gain from becom- ing a productive and livable city. But currently it is far from that potential. If the urban form of the city continues to grow in an unplanned, spatially frag- mented way, Bamako and its citizens will be locked into economically and socially unproductive urbanization. The Government has a rare opportunity for smart new policies to transform Bamako into a platform for productivity and livability, and into a city that works for its people. The window for coordinated investments in urban infrastructure is narrow. Urbanization is critically dependent upon good public policies. Owing to its post-colonial legacy, Bamako is less chaotic in its urban planning, compared to its East African counterparts. This provides it with an opportunity to make early investments in connective infrastructure, closely synchronized with land- use planning. Bamako’s District Government will need to build coalitions with its neighboring communes, regional development agencies and with the national government, while finding ways to expand its sources of revenues. Urban invest- ments are long lived and path dependent. The time to act is now. 104 BAMAKO An Engine of Growth and Service Delivery Bamako, the capital of Mali, dominates the country’s urban and economic land- scape – it is the nerve center of the national economy. Bamako has much to gain from becoming a productive and livable city. But currently it is far from that po- tential. The city is neither an engine of growth, nor of service delivery. Its urban development has been fragmented, fettering both - productivity, by preventing opportunities for matching people and jobs - and livability, by driving up the costs of urban infrastructure and service delivery. If the urban form of the city continues to grow in an unplanned, spatially fragmented way, Bamako and its citizens will be locked into economically and socially unproductive urbanization. Tackling urban development challenges in the capital will have knock-on effects on Mali’s economic development. This report argues for a focus on how Bamako is built and organized spatially. A balanced approach will be required – to reforming institutions, putting the right policies in place and investing in infrastructure – along with attention to imple- mentation. The window for coordinated investments in urban infrastructure is narrow. Bamako has an opportunity to use urban planning to make early invest- ments in connective infrastructure, closely synchronized with land-use plan- ning. Local and national government leaders will need to build coalitions with Bamako’s neighboring communes, regional development agencies and with the national stakeholders, while finding ways to expand its sources of revenues. Ur- ban investments are long lived and path dependent. With projected population growth, many of the current 2.5 million residents will live to see the population surpass 13 million by 2050. The time to act is now.