The Municipality of Maputo, with funding from the World Bank, sought to involve the citizens in the allocation of funds to community infrastructure projects.
Problems and Purpose
Following a report by the World Bank, significant reforms were promised to the city of Maputo with hopes to advance the cities development, which was far behind that of similar sized cities. These developments included a total contribution to Maputo’s public infrastructure budget of around $30 Million, with a portion being allocated to the cities struggling public infrastructure budget .
With a significant cash injection, and the pre-identification of a poor bureaucratic system , the mayor of Maputo, Eneas Comiche decided to attempt to mitigate these issues through use of public participation, with the aim to both allocate the funds fairly, and revitalise democratic participation in the city.
Four Main investment concerns were raised and recommended for emergency investment. These 4 areas were: Solid waste collection, Drainage, Road Maintenance and Cemeteries . Whilst these 4 areas were identified as being the most in need of investment, they were ultimately only advisory topics, and therefore there was no entrenched requirement for projects proposed during the initiative to combat any of these aforementioned issues.
Background History and Context
The end of White Minority rule in Southern Africa, and the end of Soviet Assistance in the late 20th century moved Mozambique away from Socialism and towards open markets and decentralisation encouraged by the World Trade Organisations Structural Adjustment Programmes (SAPs) .
Because of these decentralisation measures adopted by the ruling Frelimo party, the countries constitution included reference to municipalities and how the people should have a role in their governance . The Frelimo party went on to dominate Mozambiquan national and local politics, but not without allegations of voter manipulation and fraud .
In response to international condemnation, and internal dissatisfaction with the current state of Mozambiquan politics, reformist factions in the Frelimo party comprising largely of young party members began efforts to shift the party towards a mandate of enfranchising the electorate in more of the countries policy decisions . This ultimately culminated in the former Mayor Artur Canana being ousted by the Frelimo party, where he was replaced by Eneas Comiche who ran for the role of Mayor on a mandate of involving citizens in the decision making process of the city . Comiche was successful elected on the 19th November 2003, securing 76% of the vote, a result that was interpreted to be the beginning of vast democratic reforms across the Frelimo party .
Prior to Comiche’s election, the International Development Association (IDA) of the World Bank had identified Maputo as a “problem city” where development was desperately needed . 70% of Maputo residents lived in informal settlements, and despite the official population expanding by only 1% between 1997 and 2007, surrounding areas such as Matola had grew by as much as 4%, placing a huge burden on under-funded public infrastructure . As such the World Bank announced an initiative named “PROMAPUTO” which would not only inject cash into the city’s public infrastructure projects, but would also aim to restructure the cities bureaucracy, strengthen democratic integrity, root out corruption and bring Maputo to a point where it could compete with other leading African cities.
Whilst there was no legal requirement for the funds received from PROMAPUTO to be assigned via a method of democratic innovation, the World Bank were well aware of Eneas Comiche’s desire to involve the people, having been in communication with him during his campaign, and having hosted his team at the Africa Regional Seminar on Participatory Budgeting organised in Durban, South Africa by the United Nations. Furthermore, Comiche had sent two delegates to Porto Alegre to observe the process with the hopes of implementing his own PB back in Maputo, and therefore the World Bank remained informed of the success of the project throughout .
On the announcement of the PROMAPUTO allocation of funds to the Maputo public infrastructure budget, Comiche partnered with Eduardo Nguenha, widely considered to be the leading voice of PB in Mozambique having organised the successful implementation of PB in Donda, a system which the Maputo initiative would attempt to mimic (Albeit with significant changes). After a planning period, Comiche announced that a Participatory Budgeting initiative would be used to help allocate the funds granted to the city from the PROMAPUTO scheme, and would be scheduled to begin in 2008, with the 2nd round in 2010, although this did not go ahead as planned (See: “Influence, Outcomes and Effects section”).
Organizing, Supporting, and Funding Entities
The Participatory Budgeting observed during this initiative was undertaken entirely under the leadership of the Maputo Municipal Council (MMC). Whilst funds may have been received from the World Bank, they ultimately had no say over the planning of the initiative, and instead the process design was done by Eneas Comiche and Eduardo Nguenha, who had previously planned successful initiatives in the North of the Country.
The design of the process placed heavy emphasis on citizen participation, whereby little moderation was required. However, trained facilitators were brought in to guide discussion and ensure that the initiative moved on smoothly as there was no pre-determined structure. It should be noted however that these “Trained facilitators” were often members of the Frelimo party who were observed to be making partisan statement throughout, and when reported, would often go unpunished .
Participant Recruitment and Selection
The City of Maputo is split into 7 individual administrative districts. The original intention of the Participatory Budgeting initiative was to split it into two cycles: the 2008 iteration focusing on the districts of KaMphulo, KaMaxakeni, KaMavota and KaNyaka, and a 2010 iteration focusing on the remaining 3. As such, participation in the 2008 iteration of the PB was only open to residents of those 4 districts.
The recruitment strategy used was two pronged; firstly, posters were put up across the eligible areas, and secondly “Block chiefs” were asked to convey the information through word of mouth to residents . Block Chiefs were often traditional leaders who held prominent status in the community. In total there were 97 Block Chiefs, all of whom were tasked with holding informal sessions prior to the official PB to discuss possible projects in the hopes it would speed up the real meetings, however it was noted that very few Block Chiefs actually met this obligation, and hesitation from these people is seen as one reason for the projects failure .
In order to incentivise participants, the organisers announced that Districts with attendance higher than 1% of their total population would be given preferential treatment when it came to the later stages of the initiative . The initial neighbourhood meetings were held on a Saturday, and the organisers attempted to keep all relevant meetings to dates whereby all participants were free, such as weekends, as no compensation was available for participants. Despite this, only one district, KaNyaka met the 1% threshold recommended by organisers.
Methods and Tools Used
Multi-channel Face-To-Face Participatory budgeting was adopted in this process to allow for a vast amount of people to be involved whilst also allowing the individual significant power in influencing potential policy. The process attempted to combine elements of two successful implementations of
PB, the Porto Alegre system that has seen widespread success in Brazil, and the Donda initiative which was the first PB of its kind in Mozambique. Emulating the success of Donda was essential for the success of the initiative, as the city was similarly split into districts with smaller neighbourhoods. However, the two cities differed largely in terms of population, with Donda being home to 70,000, whilst Maputo housed over a million.
Within the 4 districts chosen for the 2008 Participatory Budgeting initiative, there were 33 individual neighbourhoods, all of which needed their own meeting before the next stage could commence.
After these 33 individual sessions, there would be 4 district level meeting, followed by one, final meeting at the city level. In each case, those in attendance were asked to elect a representative to follow their decisions to the next level, and ensure it was properly represented.
A key tool used in this initiative are the use of delegates. This ultimately means that the people are not the ones making the final decisions but rather those that they elect, like a representative democracy. Delegates are first seen at the neighbourhood level, where, after reaching a consensus, they are elected to represent the neighbourhood at the district level. Similarly, at the district level meeting, 2 more delegates are elected to represent the district at the city level. Whilst, theoretically, any member of the public could become one of the delegates, they were almost always members of the Frelimo party . This was not helped by the method of voting used at each level for electing delegates.
In order to elect delegates, each meeting used a first past the post system, whereby the 2 candidates with the most votes would be elected and receive the official status of delegate. This method is justified in that factionalism was not meant to be present in the initiative, and therefore vote splitting would not be an issue. It was found, however, that the process had been widely “partisanised” and that this may have been aided by pre-planning from party members to ensure their delegate was elected . This vote only goes ahead if there were more than two nominations for the role, otherwise the two nominees are elected without a vote. A delegate system allowed for the views of 33 neighbourhoods to be represented at a city level, and for a smooth transition up the chain of command. Furthermore, delegates were afforded a certain level of autonomy whereby they could choose not to present projects they felt were too similar to others or would not meet the criteria if they progressed. This ultimately afforded the initiative organisers with a smoother process as delegate meetings were not swamped with duplicate projects.
Voting for the proposed projects was done differently, however. Participants were given pieces of paper where they could rank their preferred projects and would then deposit them in the box of their first choice. Much like a preferential voting system, the lowest supported project would be eliminated and have their votes redistributed until the session had 3 clear projects. It should be noted that whilst every neighbourhood could present 3 projects at the district level, many of those elected to carry the projects to the later stages chose not to present certain projects as they were too similar to others presented, or did not fit the criteria presented.
What Went On: Deliberation, Decisions, and Public Interaction
In an attempt to speed up the 33 neighbourhood meetings that were due to take place, the events organisers encouraged Block Chiefs (Prominent and traditional leaders of the community) to discuss with the population and establish potential projects prior to the official meeting on the Saturday.
The length of a pilot session ran by volunteers was found to be around 3 hours long, with an hour period before starting to allow for latecomers, as is Mozambiquan etiquette. As such, there was concern that the process would consist almost entirely of Ideation rather than anything close to settling on ideas, and therefore organisers wanted to encourage discussion prior to the event itself in the hope it would result in one stage being informally completed ahead of time. This however was unsuccessful as Block Chiefs chose not to consult the populace .
The first formal stage was held at the neighbourhood level, where 33 individual neighbourhoods met on a Saturday to discuss their propositions for projects. At this stage, participants were told that each project must be beneficial for the entire district, and therefore must be justified as such. In total, 3199 people participated in the first stage, with attendance varying between neighbourhoods, for a high of 258 and low of 10. In each meeting, participants were invited to present projects to those in attendance, who were in turn invited to scrutinise and question each project. At the end of proceedings, participants were each given a piece of paper where they could rank their project preferences. They would then place their paper into the box that corresponded to their first choice.
The votes would then be counted, and the 3 most supported projects would receive the neighbourhood’s nominations. There would then be a second vote to elect the two representatives to follow the neighbourhood’s projects to the next level. This vote was done by a show of hands.
At the second formal stage, each neighbourhood representative was invited to attend one of the 4 district level meetings. The meeting would be chaired by the “district administrator” a city employee whose job it was to oversee the district. At this meeting, the representatives in attendance would present their neighbourhoods projects and the other representatives would be invited to scrutinise. Autonomy was afforded to the representatives to decide if a project was worth presenting, or if it was too similar to another. This stage also allowed the district administrator to inform those in attendance what would or wouldn’t be possible, further whittling down the propositions. Once all representatives had presented, the group would reach a consensus on which project would be best to present at the city-wide meeting. Whilst there were no quotas like at the previous stage, the administrators were made aware that focusing on a project would give it a greater chance of success. This decision was made through consensus, and the district administrator held the ultimate decision as to which projects were suitable for proposition. Those in attendance would then hold a second vote, this time to elect 2 more representatives from amongst them, and one secretary, whose role would be to report the ultimate findings back to the district. These 3 would then accompany the district administrator to the city-wide meeting.
In the third and final official stage, the 4 district representatives would attend a city meeting chaired by Eneas Comiche, the Mayor of Maputo. At this stage, each district delegation would present their propositions. Also in attendance was the Strategic Unit of Planning and Finance (UPE), who would assume the role of the accountant, and would inform those present if projects were financially viable. As previously mentioned, districts that contributed over 1% of their population would be given preferential treatment, and as such, 50% of all funding went to KaNyaka, despite having the lowest population of the districts involved . At the end of the session, the mayor would formally sign off on a proposed budget, fully costed by the UPE and with details about the proposed projects. This budget would then be presented to the Assembleia Municipal for formal ratification.
Influence, Outcomes, and Effects
To understand if the 2008 Maputo Participatory Budgeting scheme was a success, we should first restate what the initiatives organisers wanted to accomplish, which was two pronged.
Firstly, the Mayor had been elected on a mandate to increase participatory democracy, and therefore this first initiative was widely seen as being a pilot for participatory democracy in Maputo. Secondly, people expected the projects that had been approved to be acted upon, and failure to do so could undermine any success of the system itself.
Overall, 6 individual projects were authorised at the city level, receiving a total investment of MT8,388,915 ($140,304). However, the success of these projects varied, an issue brought about by a failure to consider maintenance costs when allocating budgets, partly due to the absence of civil engineers at the meetings .
The original plan of the Participatory budgeting in Maputo was to have a 2nd round for the remaining 3 districts that had not participated in 2008, and to then rotate through the districts every two years. However, at the time that the 2nd round was due to take place, 3 of the original approved plans had not started, and a further 2 had only been partially completed. The type of projects approved is also grounds for concern, organisers initially wanted the PB to focus on issues such as waste management and road maintenance, however no hard criterion were in place, and therefore those who attended preferred Quality of Life investment over infrastructure, such as the construction of 2 fountains in KaNyaka.
Throughout the process there was opposition from traditional leaders, and members of the Frelimo party who were against the Mayor and his pro-participation manifesto. From the start, Block Chiefs refused to hold preliminary meetings, and many played no role in raising awareness of the neighbourhood meetings, as organisers had hoped they would. Traditionalists inside the Frelimo party saw Comiche’s efforts to push decision making towards the people as attempts to strip them of power, and therefore they opposed the process throughout.
The world bank, interested in the success of the process hired Brian Wampler to briefly assess the process whilst ongoing, with his report casting the organisers in a positive light . And whilst the process itself went smoothly and demonstrated multichannel PB was possible, the failure to properly ensure the completion of projects shattered public trust and resulted in poor turnout for the 2010 round, to the point where it went through a radical redesign and was postponed by a year. The initiative was also given the label “para o inglês ver”, an allegation that the entire process was only done as a show to encourage further investment from the World Bank and interested parties.
Ultimately, whilst the initiative itself was an organisational success, the failure to properly act on the projects approved by the participants rendered the process as a whole a failure. The initiative was seen as evidence that Frelimo’s push towards participatory democracy was fruitless, and Comiche was symbolically replaced as the party’s nomination for Mayor in late 2008, with his rival David Simango receiving 86% of the vote.
Analysis and Lesson Learned
The Participatory budgeting held in Maputo attempted to combine elements of those seen in Porto Alegre with those in the North of the Country to run a successful multichannel system accommodating half a million citizens. Graham Smith offers an analytical framework to assist in the comparative analysis of democratic innovations using six “Democratic Goods” . In terms of inclusiveness, there were no artificial incentives to encourage certain groups to participate over others, and there is no statistical data on attendance of minority groups, however we do know that gender turnout was relatively equal, with a slight shift towards men at 53% . To mitigate the lack of financial compensation, the organisers scheduled the meetings to take place on Saturdays, and also allowed an hour grace period before the start of sessions to ensure latecomers were able to participate. I would argue that Popular control varied depending on which meeting one attended.
For example, meetings that accommodated 258 people, such as the one held in Polana Canico A were less persuadable by the individual than meetings of 10, such as the one in the “Central B” neighbourhood. Ultimately, every meeting elected 2 representatives for the next level, and therefore those who were able to attend smaller meetings had greater individual power than those in bigger meetings. Much like Inclusiveness, the organisers made no effort to involve any form of capacity building in the process. Citizens were invited to be the form of scrutiny, and as members of the local community, they were the experts on local matters themselves. Transparency was lacking in the 2008 Maputo PB, with no transcripts being kept of the meetings, including those inaccessible to the public. At each stage however, elected representatives were asked to feedback progress to the level below them, however there was no established medium for this to be done, and so word- of-mouth became the most efficient tool. Overall, the process was efficient; it accomplished its goal of allocating budgets to projects that originated at the neighbourhood level. The use of volunteers and existing city employees meant the overall costs of the project remained low, and therefore we can consider the initiative itself to be efficient in terms of cost. Similarly, the initiative was, design wise, a success. The success of the multi-channel system used meant that, theoretically, more districts and neighbourhoods could be accommodated as the system would scale well, and another vertical layer, for example, if the system were to be used nationally, would work as it was ultimately scalable.
We can begin to understand the failings of Participatory Budgeting in Maputo by the successes of PB in Porto Alegre. Graham Smith argues that implementing PB alone would not be sustainable, as it would result in lists of projects that the governing body has no ability to execute, thus destroying trust in the process . Indeed this appears to have been the issue in Maputo, the World Bank had identified fundamental flaws in the cities infrastructure application and therefore it should have been expected that Maputo would not have had the capacity to deal with the sudden influx of projects, regardless of if the money was available. Perhaps, therefore, the Maputo PB was far too ambitious in its allocation of funding. Wampler identifies 4 key principles for Participatory Budgeting: Voice, Vote, Oversight and Social Justice . Whilst we can establish that the PB successful franchised participants in terms of Voice and Vote, the lack of physical results despite conclusion demonstrates a clear failure to accommodate the final two values. Furthermore, PB’s take value from moving beyond representative democracy, however the Maputo initiative was heavily plagued with Frelimo party members who were elected neighbourhood representatives, ultimately rendering the district and city meetings quasi-parliaments rather than groups of regular citizens . Paolo Spada identifies an issue with multi-channel PB in that it can produce duplicate ideas across meetings, and this can generate tension between the organisers and the participants . It was perhaps this concern that steered participants away from the desired road maintenance and towards quality of life projects such as fountains, as those suggesting projects wanted catchy, unique ideas as opposed to “Fixing roads” which they may have assumed would be the norm.
Ultimately, the initiative itself was planned well, and was a good demonstration of a PB system that was both transferable and affordable. The system accomplished its goal of producing a series of costed projects that were originally proposed by the community, however it was the failure of the city to act on these projects that dictated the initiatives failure. Furthermore, the presence of Partisan actors, namely members of the Frelimo party hampered popular control and the initiative could’ve done better by separating party from people. The single greatest lesson to be learnt from the 2008 Maputo Participatory Budgeting would be that regardless of how successfully designed an initiative is, if the results that are promised are not delivered, trust will be harmed, and the entire process will be considered a failure.
Participatory Budgeting: https://participedia.net/method/146
 Líria Quelídio Langa, “Participação dos Cidadãos no Processo de Orçamentação Participativa: o caso do Município da Cidade de Maputo (2008-2011)”: p.34
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.57
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.25
 Pitcher, Anne. 2002. Transforming Mozambique: The Politics of Privatisation, 1975-2000. New York: Cambridge University Press.
 Reaud, Beatrice, The Political Economy of Local Democracy: Revenue Effects on Service Delivery in Four Mozambican Municipalities (August 8, 2011). P.24
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.15
 Harrison, Graham, ““Elections in Mozambique”, Review of African Political Economy”
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.17
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.26
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.27
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.29
 IEG Report: “PROMAPUTO, MAPUTO MUNICIPAL DEVELOPMENT PROGRAM (MMDP I and II)” Retrieved from: https://s3.amazonaws.com/participedia.prod/d3600cad-84cd-4d0a-80c6- 7cba3da2ba95
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.28
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.38
 Cabannes, Yves, “Contribution of Participatory Budgeting to provision and management of basic services: Municipal practices and evidence from the field”. P.7
 Líria Quelídio Langa, “Participação dos Cidadãos no Processo de Orçamentação Participativa: o caso do Município da Cidade de Maputo (2008-2011)”: p.33
 Líria Quelídio Langa, “Participação dos Cidadãos no Processo de Orçamentação Participativa: o caso do Município da Cidade de Maputo (2008-2011)”: p.25
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.34
 Líria Quelídio Langa, “Participação dos Cidadãos no Processo de Orçamentação Participativa: o caso do Município da Cidade de Maputo (2008-2011)”: p.37
 Cabannes, Yves, “Contribution of Participatory Budgeting to provision and management of basic services: Municipal practices and evidence from the field”. P.7
 Líria Quelídio Langa, “Participação dos Cidadãos no Processo de Orçamentação Participativa: o caso do Município da Cidade de Maputo (2008-2011)”: p.43
 William Nylen, “Participatory Budgeting in a Competitive Authoritarian Regime: A Case Study (Maputo, Mozambique)”: p.35
 Smith, Graham, “Democratic Innovations” (2009): p.12
 Líria Quelídio Langa, “Participação dos Cidadãos no Processo de Orçamentação Participativa: o caso do Município da Cidade de Maputo (2008-2011)”: p.24
 Smith, Graham, “Democratic Innovations” (2009): p.37
 Wampler, Brian, “Participatory Budgeting: Core principles and Key Impacts”, (2012): p.2
 Wampler, Brian, “Participatory Budgeting: Core principles and Key Impacts”, (2012): p.10
 Spada, Paulo, “Integrating Multiple Channels of Engagement in Democratic Innovations: Opportunities and Challenges” (2017): p.9