The Power of Ambiguity: How Participatory Budgeting Travels the Globe

From its inception in Brazil in the late 1980s, Participatory Budgeting has now been instituted in over 1500 cities worldwide. This paper discusses what actually travels under the name of Participatory Budgeting. We rely on science studies for a fundamental insight: it is not enough to simply speak of “diffusion” while forgetting the way that the circulation and translation of an idea fundamentally transform it (Latour 1987). In this case, the travel itself has made PB into an attractive and politically malleable device by reducing and simplifying it to a set of procedures for the democratization of demand-making. The relationship of those procedures to the administrative machinery is ambiguous, but fundamentally important for the eventual impact of Participatory Budgeting in any one context.


Introduction
To speak of Participatory Budgeting today is to speak of a seemingly infinitely malleable set of institutions that, as the articles in this issue reflect, continues to attract attention from actors on all ends of the political spectrum.It is to speak about an institution present in more than 1500 cities spread over five continents, and with training manuals in dozens of languages (Sintomer et al., 2010).Observers have noted that the literature on PB has progressed in phases, first looking at successful cases, to then developing a comparative framework, asking whether it has achieved the same outcomes and how this varied by context (Avritzer, 2006;Baiocchi, 2005;Marquetti, 2003;Santos, 1998;Wampler, 2007).
A new critical discourse about Participatory Budgeting has also appeared to temper the generally more positive earlier literature.For example, a number of recent studies on participatory budgets have begun to point to the weak impact they have had on the dynamics of the municipalities where they have been implemented beyond Brazil (He, 2011;Sintomer et al., 2008), even though the experiences in question were rhetorically justified by the "Porto Alegre Story."In those cases, in fact, Participatory Budgeting was peripheral to city administrations, outcomes were technically over-determined, and the decision-making of participants was far removed from any the locus of local power.
In this essay we wish to shift the discussion away from the impact of PB in different contexts to try to answer what we think is a more important, and more fundamental question: What is it is that actually travels under the name of Participatory Budgeting?As we detail elsewhere, Participatory Budgeting has been the subject of dozens of international exchange programs, literally hundreds of conferences, and has been the primary reason for the existence (and funding) of several NGOs that promote and help implement it.We rely on science studies for a fundamental insight: it is not enough to simply speak of "diffusion" while forgetting the way that the circulation and translation of an idea fundamentally transform it (Latour, 1987).In this case, the travel itself has made PB into an attractive and politically malleable device by reducing and simplifying it to a set of procedures for the democratization of demand-making.The Participatory Budgeting that has traveled promises to solve one of the problems of democracy, namely its unruliness and unpredictability, substituting this with rational, more inclusive, and more transparent demand-making.This makes it compatible with both "good governance" and New Public Management discourses as well as with some social justice projects.Whether it ultimately leans one way or another depends on a whole host of factors, the most important of which is its relationship to government structures and procedures.And on this, the prescription is ambiguous, which helps account for its polyvalent appeal.The separation from administrative reforms that made PB malleable and adaptable also made its relationship to an imaginary of social justice ambiguous.
Our argument in this essay is straightforward.There are two phases in the travel, with a dividing line in the late 1990s.In the first phase, PB traveled as part of a set of comprehensive administrative reforms.PB emerged out of the cauldron of leftist experimentalism in Brazil in the early 1990s, as a particularly successful instrument, one that seemed to render compatible social justice, good governance, and electoral fortunes for the left.It traveled, largely in Brazil, but to some extent in Latin America in the 1990s as a centerpiece of a political strategy.At that point, administrations implemented it wholesale, often adapting and innovating parts, but also implementing administrative reforms.It was a centerpiece of a political strategy, and, as has been documented, traveled as a representative of a leftism that could work because it broke with clientelism as well as with movements or the idea that leftists could only oppose capitalist institutions instead of transforming them.In the late 1990s, however, things changed.PB attracted international attention, becoming a best practice that was taken up by a number of international networks.Now, it traveled as a politically neutral device, one that could improve governance and generate trust in government.In this article we insist that this latter transformation was premised on the decoupling of Participatory Budgeting from a broader set of institutional reforms of which it had been part in 1990s.
Scholars influenced by science studies make a distinction between policy instruments and devices (Lascoumes and Le Gales, 2007).A policy instrument "is a means of orienting relations between political society (via the administrative executive) and civil society (via its administered subjects), through intermediaries in the form of devices."(Lascoumes and Le Gales, 2007: 5).Devices, more humbly, "mix technical components (measuring, calculating, the rule of law, procedure) and social components (representation, symbol)."Participatory Budgeting in its earlier stages would have been closer to a policy instrument, for it implied a very specific way of orienting the relationship between political society, civil society, and the state, while later versions are closer to an isolated device that has less to say on those relationships.Social justice became less important in the latter version as the mechanisms that linked participation to redistribution disappeared from the prescription.The story very much confirms other accounts of "fast policy transfer" in the current moment.Today policies are more likely to "move in bits and pieces-as selective discourses, inchoate ideas, and synthesized models" (Peck and Theodore, 2010), rather than as complete and coherent blueprints.
Before making these points, however, we first turn to the origins of Participatory Budgeting, in Porto Alegre, in the mid 1980s, before discussing its transformation into an international policy device as a tool for good governance and the ambiguities it generates.Here we focus on Porto Alegre, as its importance and influence on the global travel of PB is well accepted.But as has been well discussed, the implementation of PB in Brazil in other cities at the time generally shares the characteristic of being a policy instrument.Though widely varied, and often innovative in specific contexts, PBs were implemented as part of a whole set of administrative reforms.

The beginnings of a new policy instrument
In October of 1985, when community activists met with mayoral candidates in what was to be the first free election in the city of Porto Alegre since the military coup in 1964, they must have had little idea of how the discussion that evening would resonate around the world over the next quarter century.The minutes from the meeting describe how activists who had come together under the umbrella of UAMPA, the Union of Neighborhood Associations, had prepared a slate of questions to each of the candidates.These covered many of the concerns of urban social movements at the time: for example, "How would the candidate, as Mayor, improve public housing or transportation?"But one of the questions also was whether, and how, the Mayor would implement community control over municipal finances, an idea endorsed earlier that year by the neighborhood associations in its yearly congress. 1he mayor elected later that year, of the Labor Party (PDT), gestured toward participation but in the end did not create the imagined institutions.The mayor elected in the following election (1988), Olívio Dutra of the Workers' Party (PT), eventually set in motion a process so that by 1990 and 1991 something called "Participatory Budgeting" was implemented, took hold, and by 1992 could be said to have worked.Porto Alegre´s administrators managed to do something that had eluded leftists up to that point: combine good governance, social justice and redistribution (Utzig, 1996).However, the way to put a new policy instrument in motion wasn't easy.
The story, so far, is well known.The PowerPoint presentation on the history of Participatory Budgeting that precedes its introduction anywhere usually has a slide that makes the same point: social movements introduced an idea, the government of the Workers' Party took it up, it worked, and it traveled the world.There are two elements often missing from presentations that we discuss here.First, is that there is one crucial difference between the proposal that came from social movements and the thing that worked.The translation of the idea, the realignment of meanings and interests as new allies were brought on board, changed it in one fundamental respect: it de-emphasized the role of existing associations and their leaders in favor of the individual citizen.2It was perhaps its most visible part, but was not its only, nor even perhaps its most important, part.Second is that Participatory Budgeting was an integral part of a whole administrative project that worked.

Breaking with Associational Democracy
The participatory proposal made by civil society in Porto Alegre was based on associations.If we return to the original document from the neighborhood associations, it called for a system "where the investment priorities of each district would be discussed with popular leaders of each district," where there would be "Popular Councils throughout with proportional representation of the community movement to discuss the municipal budget."(UAMPA, 1985).The UAMPA proposal was one in which representation of associations was central.In it, representatives of clubs, churches, associations and "others" would come together in a forum to debate proposals.
The Participatory Budget that was implemented in 1990, in contrast, had as one of its key principles that "meetings would be open to anyone" without privileging of existing associations or movements.Any citizen -associated or not, could come to the meeting and have an equal voice.The distance between the two ideas was immense and profoundly consequential.In terms of political theory, it is the difference between associational democracy (or corporatism for that matter) and direct democracy.It is the difference between placing special value in autonomous spaces in civil society and their capacity to generate demands, and placing special value in spaces where all citizens can participate on equal footing. 3  Collectives, like unions, had been previously understood as privileged interlocutors to speak on behalf of the whole, became symbolically reduced to a segment smaller than the whole 4 .In our view, this was a very important turning point in the transformation of the PT from a party of movements into a party that seeks to govern well.The administration's discourse on the autonomy of civic organizations and the government responsibility for the whole city were rooted in a new imaginary of how interactions between state and society should be.
This process was something of a novelty: these were to be procedures based on the participation of "ordinary" citizens allowed to debate the general interest, as a new form of management of public affairs.It did not go against associations per se -as many imagined this process would empower civil society, as much as it was a challenge to their monopoly of representation of the people, or the idea that they represented all citizens.But it required new administrative habits to handle this new political subject (all citizens), which implied a big change for an administrative apparatus developed around privileged interlocutors.

Administrative Reforms
As the administration began to change its discourse in terms of a city for all, it also began to reorganize its administrative machinery.In order for a "new public sphere" to emerge, it was necessary to reach all citizens and develop transparent spaces of interactions with the administration (Genro, 1999).Former Mayor Raul Pont (2003) argued that Brazil did not have traditions to draw on like the referendum in Switzerland or participatory traditions from the Middle Ages in Europe, so it was necessary to invent something new.Although an exaggeration, it gives a sense that the PB was not only understood as an invitation for people to participate, but as part of a thoroughgoing set of reforms to devise a wholly new way of governing.The reform consisted of four lines of action.
First, all social demands were channeled through the PB, and all other channels were essentially closed.Contact with the administration on the part of the population was to be made almost exclusively through the PB.The government transformed the whole administration in such a way that it was impossible to receive any funds, investments, or projects outside of the participatory process (Baiocchi, 2005).This was a rupture with old traditions on the political right and the left.It was meant as a way to avoid cronyism, circumventing politicians, but it also circumvented social movements.One of the founders of UAMPA recalled that, though the federation had played a role in the early stages, the process went on without it, and its role as interlocutor was marginalized because "when any demand reached the city councilors, their message was 3 The discussion of associational democracy is beyond the scope of this essay.See Cohen and Arato (1992), for example, for a key text. 4And in response to the question of whether this represented a rupture with union allies, Genro responded in the same interview that ¨this was not a severing, because there would have had to be an alliance (…) The administration does not establish alliances with unions.That is a completely distorted vision." (1990:58) to say to the people that they had to go to the PB to get anything done about it."However, in time, "everybody was for the PB; cronyism was finished."Second, in order for "participation to come into the administration" it was necessary to create new patterns and practices within the administration.This was a combination of "political centralization with administrative decentralization" (Navarro, 1996).Centralization was achieved by the creation of a new cabinet-level department that centralized all participatory inputs and coordinated these efforts and stood above municipal departments.The idea was to "ring-fence" the capital budget from other sources of pressure, and to ensure impartiality in implementation.Control from above helped prevent policy implementation from becoming currency in the jockeying for political power.Finally, a centralized planning department above municipal departments also helped have a vision of the whole, preventing superfluous projects.This was combined with de-centralization efforts and with a series of administrative reforms to prepare the administrative machinery to receive the inputs from the participatory process.For example, all municipal departments were required to create positions of community facilitators.Community facilitators were to be the "face" of each municipal department in each of the city's districts, and required to attend PB meetings with the express purpose of helping participants prepare technically viable projects and to be accountable for the ongoing projects.Community facilitators attended a weekly forum to keep participatory processes coherent.Finally, there was an effort to subsume technical expertise, what has been well documented and described as "techno-democracy" (Santos, 2002).As much as possible, "technical expertise was to be made subservient to the popular mandate, and not the other way around," as one of the facilitators described it in an interview in 19995 .
Third, the public decision-making process was organized on the basis of a multi-stage process in which decisions were adopted progressively.If associations lost their role as privileged connectors between state and civil society, it was necessary to supply other devices.Rather than bargaining, deliberative and preference-averaged procedures were combined to achieve a measured decision within a universal scenario where all citizens had a voice.This included a deliberative procedure in which citizens had to evaluate the distribution of scarce resources applying criteria of social justice.The criteria will be subsequently used to prioritize citizens' proposals.It was a procedure close in spirit to Rawls's "veil of ignorance" mechanism.This process could solve the problem of uncertainty, transforming the usual logic of protest into a rationalized process of demand-making.This could help participants from making snap decisions; instead they were encouraged to take the general space (the municipality) into account and prioritize their proposals accordingly (Abers, 2000).
Fourth, there was a significant tributary and fiscal reform to increase revenues.The reforms introduced tax progressivity in the two most important Brazilian municipal taxes, the taxes on real estate, and on services; different utility rates were updated and indexed to inflation; control was also increased over tax fraud.Real estate taxes became increasingly important, going from 5.8% of the volume of municipal revenues to nearly 18% of total revenues, while the services tax went on to account for 20% of municipal revenues (Santos, 2002: 68).Tax evasion was also significantly reduced over ten years later (Baeirle, 2003).These reforms essentially doubled the city's income over the ten years.To put it in context, while roughly two thirds of municipal budgets in Brazil, overall, come from transfers from higher levels of government during the time period, Porto Alegre´s own revenues accounted for almost 60% of its budget.This was almost the level of local fiscal autonomy of European cities (Carvalho and Felgueiras, 2000;Santos, 2002;Bremaeker, 2004).
PB thus became a specific device that allowed the state to rule and imagine other relations between state and civil society.These relations, as Genro showed, meant a new process of democratization, but from above.It wasn't based only on participation but also on new administrative routines.But as we'll see in the next section, while PB got to translate the wishes of change made by civil society organization into administrative language, as an isolated device this change provoked confusion and ambiguities.

Rationalized Participation
During the 1990s PB was a success.Porto Alegre's poorer citizens participated in droves in a relatively simple system that promised, and delivered, results.Many poor urban denizens, otherwise having little voice in government or political affairs, became apt and loyal participants, diligently coming to meetings week after week to debate the arcana of municipal finances and regulations as they decided on investment priorities for their neighborhoods, boroughs, and the city itself.The administrative reform improved the administrative machinery, improved the conditions of poor people, and established a new way of administering that would eventually cause admiration elsewhere.And it proved politically efficacious: the PT administration was comfortably re-elected in 1992, 1996, and 2000, each time advertising the PB as the centerpiece of a mode of governance that benefited the "whole city."But as a form of democratization from above, based on transparent devices and supported by citizens' participation, the process was not without problems.As it was supposed to transform collective action from protest to proposition, it gave civil society a new role in public affairs.If we imagine, for a moment, an idealized Habermassian sequence (cf.Habermas, 1996) in which informal debate in the public sphere (by citizens) is followed by structured deliberation of formal positions (by civil society organizations) that are then passed along to authorities, PB radically changed the usual stages of this political process.Instead of an informal debate, PB set up a structured debate among people and this formal deliberation was translated directly to administration.This would take place in a horizontal public space where citizens gained influence over public decisions.If earlier, utopian versions of civil society theory (cf.Cohen and Arato, 1992) imagined social movements as connectors between public opinion and public policy, with PB it was the administration that established and regulated those communication channels.But it did so on its own terms.Participatory Budgeting translated the wishes that emerged in grassroots democracy into a technical and rational language, and into sensible projects that could be weighed against each other in a transparent way, thus helping citizens present their needs.
But this idea is premised on accepting certain limits of the public debate, mainly based on administrative limits and schedule.Within Participatory Budgeting meetings, demands do not exceed the boundaries of the process (they can exceed but they are then disqualified), individual participants are not unfairly swayed or overwhelmed by organized groups, and there is a value-neutral way to adjudicate between competing demands.That is, through procedure, PB promises to make democracy more bounded, more fair, and more objective.But this, by definition, excludes demands, projects, and ways of making claims.This was not lost on Porto Alegre's civil society.For neighborhood leaders accustomed to the "dance" of protest and favor-trading (and privileged access to clientelist politicians) the new relations between state and civil society structured by universal procedures come to be seen as a way to neutralize the countervailing powers of social movements.For some radical activists within UAMPA, the PB meant imbuing participation with a different logic, focused on resolving the specific issues of life in the city, "which prevented us from attacking the core problems: how the city was financed, what state model was desired."The UAMPA decided to remain outside "because it seemed a contradiction to be discussing how to share out a scant municipal resource, fighting for a piece of the budget, instead of debating the financing of the cities."6For many of those activists it was seen as forcing social movements to accept the power of the administration to shape the terms of the debate.In most other contexts, the introduction of Participatory Budgeting as an isolated device constitutes a kind of "democratization from above," in which limits may come to be experienced strongly as external impositions by civil society organizations.

From a new policy instrument to an isolated device
The ideas of the rationalization of demand-making, transparency, and sensitivity to public opinion fit well with new ideas taking root in both international agencies and policy networks in the Global North at the time.
Renewed attention to good governance, participation, and institutional reforms was part of the mood in development circles, when Participatory Budgeting started to attract international attention in the mid-1990s.Similarly, ideas of civic engagement as a tool for administrative modernization were becoming more and more important in Europe at the time (OECD, 2001).
PB was actually recognized as a best practice at the UN Habitat Istanbul meeting of 1996.The description of the best practice at that moment, and at subsequent iterations, was a simplification: PB was defined as sequence of meetings premised on universal participation and a fair and transparent decision-making.Ideas about state reforms as necessary conditions for establishing PB all but disappeared, and the close connection between participation and administration was severed.In fact, the logic was turned on its head: Participatory Budgeting was now understood as a device that itself could help improve administration rather than device within a set of reforms to administration.Instead of pointing to fiscal reforms as a pre-condition to PB, increased revenues were now sometimes framed as an outcome of PB.
Europe was the first continent beyond Latin America to adopt PB, and progressive European politicians saw possibilities for PB within the context of ongoing discussions.Public administrations throughout the continent were experimenting with ways to bring the citizenry closer.Legislative changes carried out in several countries actually sought to facilitate citizen involvement.This was the case, for example, in the United Kingdom (Local Government Act 2000), France (Proximity Democracy Law, 2002), Spain (Local Government Modernisation Law 2003), Holland (Local Democracy Law 2002).All of these reforms were seeking to increase the transparency of public management in order to enhance democratic legitimacy, by: 1) improving accountability, which seeks to make executive responsibility more transparent; and 2) increasing the amount and volume of participation in government.The effect of this has been the proliferation of new participatory instruments in France (Sintomer and Maillard, 2007), in the United Kingdom (Birch, 2002), Spain (Ganuza and Francés, 2011) and many other countries (Smith, 2009).Between 2000 and 2010, European experiences increased from a handful to more than 200.
The European experiences are very different from one another, with some clear national patterns (Sintomer et al., 2008).The British experiences, for example, sometimes have to do with smaller amounts of money and with social service projects, while Spanish experiences have more of a focus on urban infrastructure.But Spanish experiences are unique in their mixed model of representation, allowing for a set-aside number of citywide seats for neighborhood associations, while French experiences have taken place at the sub-municipal level.The diversity speaks to the inventiveness of implementers and the role that different kinds of local actors have played in cobbling their experiences together.But it also speaks to the plasticity of PB as an institution, a plasticity that comes from its apparent political neutrality and low institutional profile.In the majority of cases in Europe, PB required no institutional reforms or changes.
One clear consequence of the transformation of PB into a best practice has been the marginalization of social justice principles that inspired the initiative in the first place.It has instead joined the loose toolkit of ideas for innovative good governance, part of the "fast policy transfer" that has been described as characteristic of our era (Peck, 2011).As such it can be reassembled and rendered compatible with the most diverse projects.In the 2000s, for example, in the Andean region, PB was promoted by both the USAID and internationalist activists of the Chavez government (which also adopted it as a national policy).The PB Unit, a promoting organization in England, in one of its how-to pages, describes how to make a pitch for PB to your local city councilor, instructing advocates to choose from a menu of arguments to make for it based on the councilor's political leanings.Greens and progressives find resonance in PB's local empowerment, but centrists and conservatives do so as well: as a "sensible step in decentralising and localising responsibility," PB is promoted as fostering "community cohesion," "innovation," "social entrepreneurship" and "restoring trust" in government. 7y and large, PB is justified in terms of its results in terms of good governance, those "things that enable a government to deliver services to its people efficiently" (Wolfowitz 2006: 3)."

Ambiguities and Political Openings
Like many other tools for good governance, PB today is prized for its value-neutrality, its ease of implementation, and its ability to attract many different kinds of institutional stakeholders.But this has not been without ambiguities, however.
There are potential conflicts provoked by a new idea of participation (of a universal political subject) and an administration with routines and objectives designed for economic efficiency.Even when PB is a marginal to an administration, there is also an unpredictable element.Once public space is open to all citizens, it allows people to have a voice, even when its influence is diffused.Competing logics of participation, representation, and what counts as expert knowledge can come to clash within even the most well-organized PB process.
A second ambiguity concerns the perceived need for PB.It is not surprising that if PB is called for when there is a deficit in good governance, then PB can be also treated as a redundant process.At an interview at the World Bank, an expert told us that his concern was where there is the rule of law, the PB can be a costly repetition of institutions of representative democracy.Only in absence of democratic participation, "then one has to have some sort of participatory process to hear the voices that have not been heard."This is a common perspective among politicians.
There are also profound ambiguities in what PB is and what it is supposed to do.In the practitioner and scholarly literatures there are often disputes about how to define PB, or if a particular experience "is or isn't really" a case of Participatory Budgeting.In Europe, for example, most experiences are advisory with the exception of Spain.If we take into consideration these procedural characteristics, PB in Brazil (Avritzer, 2006) would have nothing to do with what would happen in Europe (Sintomer et al., 2008) and none of these with what happens in China (He, 2011).For many practitioners, the bright line has to do with decision-making.Is decision-making within a particular experience of PB "binding" or merely "advisory"?For others, the dividing line has to do with a minimum quality of participation.Vague ideas about "participation" and even about Participatory Budgeting, are like other things that travel in "ideoscapes," those constant flows of "ideas, terms and images, including 'freedom', 'welfare', 'rights', 'sovereignty'" (Appadurai, 1993: 224) The sources of these ideas can be manifold, their coherence is loose, and multiple local interpretations are always possible.Particularly because the proponents of Participatory Budgeting can be so different, it is possible, at any one site, for experts and implementers to run up against quite divergent local interpretations of what PB can be.How central social justice is for a particular experience, for example, can be a profound source of ambiguity.
But the most salient concern about PB is the ambiguous relationship to the administration.In most cases, the implementation of PB has been outside of the administrative machinery, not as a way to transform governing, but a new way to link administration and civil society.The logic of a participatory experience anchored in a process of decision-making on public affairs can come to collide with institutional structures set up for something else.Many administrations promoted PB as an alternative to the existing connectors between civil society and administration, but without transforming the latter.Participatory Budgeting was then expected to achieve desired outcomes (to improve the administration, for example, or to increase citizen trust), regardless of changes in administrative organization.This has also been a source of tremendous confusion.
It is for this reason that we think that Participatory Budgeting is not at an ideological end-game and that emancipatory possibilities exist within even experiences completely divorced from the administrative apparatus and organized around the rhetoric of new public management.We agree with critics that it appears unlikely that the conditions that produced a political project that linked participation to both real administrative reforms and an imaginary of social justice will be repeated.This does not mean that particular experiences cannot occasion moves toward transformation.Rather it suggests different critical tasks, and indeed a renewed importance of critique.The study of these new mechanisms of participation ought not to forget this political dimension of any policy instrument, which means thinking about the consequences and impact of the new participatory logic from the viewpoint of the exercise of power and not only from the viewpoint of those who participate.Indeed, can the citizenry use PB to mount a practical opposition to the state?If citizens cannot debate and change the rules, if there is no plural inclusion of citizenry, or if decision-making procedures are not transparent, then PB may conceal a new form of domination that has nothing to do with a new process of democratization.In this respect, as long as we recognize the importance of society's self-organizing capacity as a means of controlling public powers, PB can offer a good environment as much as it can be a black box for other purposes.