How participatory institutions deepen democracy through broadening representation: the case of participatory budgeting in Porto Alegre, Brazil

How Participatory Institutions Deepen Democracy through

Broadening Representation:

The Case of Participatory Budgeting in Porto Alegre, Brazil

Laurence Piper

Abstract: At the same time as democracy has ‘triumphed’ in most

of the world, it leaves many unsatisfied at the disjuncture

between the democratic ideal and its practical expression.

Participatory practices and institutions, as exemplified in

the participatory budgeting process of the local government of

Porto Alegre in Brazil, claim to embody a more substantive

version of democracy that can settle this deficit. This

article interrogates this promise through examining closely

the case of Porto Alegre. In addition to demonstrating clear

democratic outcomes, this examination also reveals that the

meaning of democratic deepening is not cashed out exclusively

in terms of participation but in terms of representation too.

More specifically, participatory budgeting serves to broaden

representation in the budgeting process as a whole, by better

including and amplifying the voice of marginalised groups in

aspects of the budgeting process, albeit through participatory

practices and events. On reflection this should not be

surprising as participatory budgeting introduces new decision-

making procedures that supplement rather than replace existing

representative institutions, and reform rather than transform

expenditure patterns. Thus although termed participatory, at

the level of the municipal system as a whole, participatory

institutions assist in better representing the interests of

marginalised groups in decision-making through participatory

means. Deepening democracy, therefore, at least as far as new

participatory institutions are concerned, is about new forms

of representation and participation, rather than replacing

representation with participation.

Keywords: Brazil, budgeting, democracy, participation,

representation

Introduction

Democracy, we all agree, means ‘rule by the people’, but

unfortunately that is where the consensus ends. Democracy is

an ‘essentially contested’ concept, both in the sense that

there are a variety of meanings attached to it, but also in

that these are disputes that cannot be settled easily (Gallie

1956). Hence the tremendous variety of historical forms of

democratic governance (Dahl 1998) as well as empirical and

normative models (Held 2006). Normatively speaking, the

mainstream understanding of democracy is typically some

version of the idea that when it comes to governing an

association ‘all members are to be treated as if they were all

equally qualified to participate in the process of making

decisions about the policies the association will pursue’

(Dahl 1998: 37). Like Dahl most theorists link the

justification of democracy to the values of liberty and

equality, but there is clearly no consensus on what the

meanings of these values are nor the optimal relationship

between them, including in respect of which democratic process

or what set of institutions. These normative debates are

further complicated by the various invocation of further

democratic values such as ‘plurality’, ‘fraternity’,

‘community’, ‘justice’ and so on (Raphael 1976: 143).

In contemporary times this variety is assumed away in

popular discourse following the hegemony of liberal

representative democracy (Huntington 1993), which is

constituted by multi-party competition for office through

free, fair and frequent elections, under the rule of law which

upholds political and civil liberties including private

property, and which embraces various checks and balances, such

as the separation of powers functionally, spatially and in

time. In large part a set of institutions captured in Robert

Dahl’s (1972) Polyarchy, and inspired by Schumpeter’s (1942)

positivist affirmation of democracy as the electoral method

for choosing leadership, this particular incarnation of

democracy is the one operationalised in the mainstream

political science comparative work on ‘transitions’ from

authoritarian rule, and the subsequent ‘consolidation’ of

democracy. Recently John Keane (2009) has further nuanced this

model by drawing attention to the enormous proliferation of

new institutions that monitor governance since the Second

World War.

Yet liberal democracy is not without its discontents. As

Gaventa (2006) notes, at the same time as liberal-democratic

capitalism has become hegemonic, there is a growing sense of

democratic deficit in both established and new democracies as

the promise of inclusion and access is never quite fulfilled,

and in moments of crisis the rule of elites seems well

exposed. This was most profoundly illustrated in 2011 by the

juxtaposition of the demand for democracy of the ‘Arab spring’

with the ‘occupy’ movement in much of the developed north. On

both versions, the general demand is the same – for some form

of inclusion in political decision-making – albeit that the

specific demands may be for democracy or some version of

‘real’ democracy respectively. Notably, this paradox seems to

fulfil John Dunn’s (1992: vii) claim that ‘today in politics

democracy is the name for what we cannot have – but cannot

cease to want’. By this is meant the idea that democracy

involves some kind of direct inclusion in rule, the

substantive equality that Dahl refers to, but is practically

impossible in modern, complex and large-scale societies.

But is this conclusion correct? The recent wave of

participatory experiments across the global south suggests

that new forms of democratic institution, or at least the

deepening of current forms, are possible (Baiocchi et al.

2011). It is in this context that the call to deepen democracy

through forms of participation in development projects and in

local governance gains greater resonance. For one thing,

issues of scale disappear or are ameliorated at the local

level, making citizen participation more possible and

representation less necessary. The implication of this line of

thought is that the democratic deficit can be settled by

replacing representation with participation, by supplanting

Schumpeter’s elite representative model with participatory or

‘real democracy’, as Bryan (2003) might term it.

This article will argue against this view by

demonstrating that even in the archetypal instance of

‘participatory democracy’, namely participatory budgeting in

Porto Alegre, democratic deepening is achieved at the

municipal or system level through representation –

specifically through broadening representation in the

decision-making process by granting better access and voice to

marginalised groups – rather than through replacing

representation with the direct inclusion of citizens. To make

the point another way, the direct participation of citizens in

decision-making processes and events at the local level has

the effect of broadening representation at the municipal

level. Furthermore, not only is this practice reproduced in

all the subsequent substantive incarnations of this model that

I am aware of, but there are also good grounds for expecting

representation to play a central role in any democratic

process.

Lastly, while new participatory institutions may have

representational and well as participatory impacts, the extent

to which these actually deepen democracy relies on meeting a

number of conditions including the political will of

government to make the ‘invited spaces’ work, the empowerment

of these spaces to have some real impact, and the popular

mobilisation of marginalised groups into these spaces. While

empirical research on the impact of these institutions is

relatively new, the results so far suggest mostly modest

outcomes. The experience of participatory institutions is thus

one of new forms of representation and participation that

supplement the mainstream model, and reform policy outcomes in

a pro-poor direction rather than transform them.

In making this case the article starts by establishing a

working conceptual frame, before exploring the wider

literature on participatory institutions, and then applying

these to participatory budgeting in Porto Alegre, Brazil.

How to Measure ‘Deepening Democracy’?

As already noted, beyond ‘rule by the people’, democracy is an

essentially contested concept to which there can never be only

one definite answer. Thus instead of advancing a grand theory

of democracy a priori and then applying to real-world cases, this

article will move from an emergent theory of democracy as

embodied in ‘participatory institutions’. It is nevertheless

useful to foreground some distinctions and conceptions here

that will be used in the analysis. The first is that it is

possible to assess the democratic credentials of any

institution in terms of both the internal process of decision-

making and the actual outcomes of the decision-making. Thus

where the former refers to process: how members of the

affected community are represented and/or participate in the

institutional decision-making process, the latter speaks to

outcomes: the extent to which the decisions actually advance

the interests of the community, or the majority of the

community, in some tangible way (Christiano 2004). A key claim

of this article is that there is at least some evidence of a

connection between democratic processes and outcomes – at

least in the case under discussion. This is consistent with

the wider, if limited, literature on policy impact of

participatory institutions (Baiocchi et al. 2011).

The second set of distinctions concerns the unpacking of

what is meant by the process of decision-making. Here I follow

Geuss’ (2001: 113–4) characterisation of politics as being

about power in some form. Further, Geuss argues that power is

exercised in three dimensions, that is all decision-making

processes have three phases: deliberation – ‘where possible

options for action are proposed, discussed and ... evaluated’;

decision – ‘the actual decision mechanism by which a group

chooses a course of action’; and implementation – ‘the

execution or implementation of decision’. This basic schema

allows us to interrogate more closely the decision-making

process of participatory institutions by exploring more

closely the forms of representation and participation evident

in each phase. Thus we can explore who participates, in what

way, in whose name and to what end in the deliberative phases,

for example. This facilitates the building of an empirical

description of the nature of the decision-making process.

Indeed, in the case below, we are able to demonstrate how the

inclusion of more representatives from previously marginalised

groups in the deliberative phase of decision-making can lead

to a deepening of democracy in both procedural and outcome

terms.

The third set of distinctions concerns the concepts of

democracy, participation and representation. This article

seeks to undermine the idea that deepening democracy is either

about new forms of participation or new forms of

representation by pointing to the existence of different

combinations of both in the institutions under discussion.

Further, the case is not made by finessing the definitions of

these terms in unfamiliar ways, but rather by affirming the

common meanings. Hence I use the concept of democracy in the

mainstream way Dahl identifies above, as about the assumption

of substantive equality in the decision-making process.

Relatedly I affirm that participation is about the direct

involvement of citizens – or members of the affected community

in cases of smaller groups – in the decision-making process.

Not only is this substantive meaning of participation the one

used by advocates of direct democracy, sometimes termed ‘pure

democracy’ (Budge 2001) or ‘real democracy (Bryan 2003), but I

would suggest that all large-scale models of participatory

democracy in Western political thought (city, workplace and

civil society) invoke the direct participation by citizens in

the decision-making process – especially as regards the

deliberation and decision moments of politics.

Now, however, I want to contrast participation with

representation, which I take in the broadest to mean a person

or small number standing in the place of a larger group.

Perhaps the most famous modern discussion is Pitkin’s (1967)

characterisation of representation as a social relationship of

a principal–agent sort, with two formal elements,

authorisation of the agent to act on behalf of the principal,

and accountability of the agent to the principle. In addition

to this formalistic character of representation, Pitkin notes

that representation can be symbolic in that the representative

‘stands for’ the represented in a meaningful way (e.g.

traditional dress); descriptive, in that the representative

resembles those represented (i.e. gender or race); or

substantive, in that the representative acts in the interests

of those represented. Notably, these analytical forms can take

various substantive shapes in practice. Thus Mansbridge (2003)

surveys the varieties of representational practice in American

politics, arguing that the common assumption that politicians

are selected on the basis of the promises they make, and held

accountable in these terms (which she calls the ‘promissory’

model of representation), is far from how representation

actually works. Instead, she identifies three actual

representational practices, namely ‘anticipatory

representation’, in which a representative tries to please

future voters by imagining what they will think at the next

election, rather than what they thought at the last;

‘gyroscopic representation’, in which representatives act in

ways that the voters like, without external incentives, as

they pursue ‘in-built’ goals like environmental or social

justice; and ‘surrogate representation’, which is the practice

by representatives of advancing the interests of those who did

not elect them but with whom they share an affinity, for

example gender. Mansbridge’s arguments serve as an example of

the kind of analysis below, that is to tease out precisely the

empirical meaning of representation encoded in the practice of

‘participatory’ institutions.

Fourthly, we can interpret the meaning of deepening of

democracy not only in procedural terms but also in terms of

the outcomes of the decision-making process. More

specifically, I follow the view that an increased dividend of

social goods for the majority, in both cases poor and

marginalised groups, can be read as a democratic gain in

contexts like Brazil. Notably, this definition practically

follows Aristotle’s definition of democracy as inherently pro-

poor. Further, while the assessment about whose interests are

served by certain outcomes is obviously vexed, in the case of

budget process, a reasonably compelling metric is the portion

of the budget spent on poor populations, although even this

will have limits depending on one’s underlying model of

economic well-being. Nevertheless, this metric is compelling

in an immediate way, especially when it coincides with the

express preferences of poor communities and their

representatives in the decision-making process around budgets.

Fifthly, we must deal with the question of power. It is

noted that identifying a participatory practice or institution

at local level does not necessarily make for participatory

local government as a whole. As implied above, the question is

better framed in terms of the extent and nature of

participation across a decision-making process. In this way we

can distinguish between practices and institutions. An example

of the former would be a relatively discrete set of actions

such as holding leaders accountable to delivery targets in a

public meeting on the implementation of a decision. The

implementation phase is one aspect of the decision-making

process in participatory budgeting, where the participatory

budgeting process as a whole could be termed an institution

that includes deliberative and decision phases too, as well as

the human and financial resources to operationalise these in

reality.

To this distinction between practices and institutions we

can add systems. By this it is meant a political system at the

macro-level as the totality of practices and institutions that

constitute a particular physical and social space. These are

most readily associated with levels of the state in our cases,

but they are not necessarily the same. Thus the existence of

one participatory developmental institution in one city does

not mean that all governance in the city is now participatory.

Cognisance would have to be taken of the sum or combined

(these may not be the same) effect of all institutions on key

decisions that affect the area. Hence, even in cities of

Brazil where participatory institutions are the most common

and longest-established, the governance system remains

primarily a representative one supplemented by forms of

participation that serve to broaden representation by bringing

poorer and historically marginalised groups into city

decision-making (Baiocchi et al. 2011: 143). Thus,

participatory practices do not necessarily constitute

institutions that, in turn, do not necessarily constitute

systems. Hence we can have representative practices (like

voting for delegates) in a participatory budgeting process in

a representative local government system

Relatedly, the argument so far has referred to

‘citizens’, but of course not all decision-making processes

occur at the state level, and so the question becomes who

replaces the citizens in the case of decision-making at

different scales, or specific sectors, or on particular

issues. It seems reasonable to assume that in such cases the

place of citizens can be taken by ‘members of the relevant

community’. This is probably satisfactory when considering

models of processes and design principles, but as is the case

with representation below, it signals that in actual practice

the claim to participation must be further interrogated with

questions of who participated in each phase, and how they

participated. These are sometimes termed issues of ‘access’

and ‘voice’ (Young 2000), and link to complex contextual

issues of identity politics of race, nationality, ethnicity

and gender, socio-economic inequality and issues of capacity,

historical forms and legacies of marginalisation that link to

political culture (Fung 2004). It follows logically that the

larger social context of state–society relations will have a

profound impact on who participants and how, regardless of

institutional design. In what follows I will frame broadening

representation as improving access and voice for historically

marginalised groups.

In sum then, assessing the democratic nature of new

‘participatory’ institutions requires assessing these

institutions in terms of both their internal processes and the

policy outcomes. Processes are unpacked in terms of

participation and representation in three phases of

deliberation, decision and implementation, and outcomes are

assessed in terms of pro-poor impact. The democratic dividends

of both are established by reflecting on the relationship

between these practices, institutions and the wider system of

governance, including local state–society relations. In what

follows the article will examine the success story of Porto

Alegre that reveals both new forms of direct inclusion in the

decision-making process on budgets and better outcomes for the

poor majority. Before moving on to that, however, let us

consider the existing literature on the conditions required

for successful participatory institutions.

When Do Participatory Institutions Work?

As noted above, participatory institutions are relatively new

innovations in democratic governance, emerging mostly from the

global south. Examples include the most famous case of

participatory city budgeting in Porto Alegre, Brazil, village

governance in Kerala, India, but also innovations like

citizen’s juries in the United Kingdom, and on paper if not in

practice, new forums like ward committees in the local

government in South Africa (Piper and Deacon 2009). These

participatory institutions can be seen as part of a wider

emergent discourse on public participation, especially in

local government, that spread across the globe in the 1990s

and 2000s. Some of this is due to new approaches to

development (World Bank 1996), the rise of deliberative and

discursive theories of democracy (Cohen 2002; Habermas 2002),

as well as new theories of citizenship (Cornwall 2002). Many

have followed Cornwall (2002: 17) in framing these new

participatory institutions as the ‘invited spaces’ of local

governance. These invited spaces include Hendricks’ (2006:

486) ‘micro deliberative structures’ and the ‘empowered

deliberative democratic structures’ of Fung and Wright (2001:

5).

The term ‘invited spaces’ takes its meaning from the fact

that such institutions or opportunities are created by the

state to ‘invite’ local communities into a process of

engagement, whether it be information transfer, consultation,

deliberation or even joint decision-making, to summarise

Arnstein’s (1969) ladder of participation. However, just as

important for state–society relations as ‘invited spaces’

created ‘from above’ is popular mobilisation led ‘from below’

by civil society or citizens. Hence Cornwall (2002: 17)

contrasts the ‘invited spaces’ created ‘from above’ by the

state with ‘organic spaces’ created ‘from below’ by those

outside the state. These include engagements with the state

created by mobilisation by citizens, as well as forums in

which ‘like-minded people join together in common pursuits’.

Holston (2009) describes the emergence of a rights-based or

‘insurgent’ citizenship amongst the urban poor in Brazil that

is mobilised through social movements, and that aims to

disrupt social relations emergent from neoliberalism from the

ground up. Miraftab (2006) paints a picture of ‘invented’

spaces opposing ‘invited’ spaces in South Africa, but also

elsewhere in the world, and also, for the same reason, the

globalisation of neoliberal economic policy.

Importantly, as Cornwall and Coelho (2007: 1) indicate,

the conceptualisation of local state–society relations is not

exhausted by a binary opposition between top-down, state-

driven invited spaces and bottom-up, social movement driven

invented spaces. Hence they talk of a ‘participatory sphere’

that lies at the interface of the public sphere and the state,

composed of hybrid institutions, some of which are extensions

of the state and some of which are claimed from the state.

Lastly, and this is the critical point, the relationship of

these institutions with the state and the general public is

partial: ‘its institutions have a semi-autonomous existence,

outside and apart from the institutions of formal politics and

everyday associational life … They are spaces of contestation,

but also of collaboration and co-operation’.

In addition to this larger literature on citizenship and

state–society relations is the more specific examination of

new participatory institutions led by Fung and Wright (2001)

that identifies among the key criteria of successful

initiatives that the institutions have (i) a practical

orientation to specific issues that concern citizens (e.g.

water); (ii) forms of bottom-up participation where those

directly affected must participate in the deliberation on

solutions; and (iii) deliberative solution generation such

that decisions must be made according to deliberative

principles of equal access, voice and impact on outcomes. In

addition to these principles, they also hold that successful

participatory institutions are empowered to take key

decisions, but also exist in some relation with the state, not

least so that the state can supervise and co-ordinate

activities and make the institution sustainable over time. In

short, participatory institutions need to be well designed and

have state support to be meaningful and sustainable

structures.

While endorsing this view, Gaventa et al. (2007) add that

good institutional design is not enough for effective

participatory institutions, and that political will from the

state’s side is also required, at least to get them going.

This view is endorsed by Baiocchi et al. (2011). Last, but not

least, there is no point building new participatory

institutions with government support if there is not also the

requisite popular mobilisation of citizens into these new

spaces, especially popular mobilisation of historically

marginalised groups. Hence the critical role of civil society

in also making participatory institutions work, and the

significance of the history of state–society relations for

creating the larger context conducive to institutional

innovations (Cornwall and Coelho 2007; Mohanty et al. 2011).

Notably, the model for effective institutions of public

participation requires ‘working both sides of the equation’ of

state–society relations, reflecting a high level of co-

operation between state and society actors, as well as well-

designed institutions (Gaventa et al. 2007: 2). Getting these

several conditions aligned, it follows, is a substantial

achievement, and participatory institutions are thus

challenging to implement – although, as Baiocchi et al.

demonstrate, clearly it is possible.

In what follows then, the article will locate the Porto

Alegre experiment in participatory budgeting in respect of

this literature on participation institutions, largely

redeeming the Gaventa et al. (2007) model, but also unpack the

meaning of democracy it represents, revealing how this

participatory budgeting enhances the representation of

marginalised groups in the city.

Porto Alegre and Deepening Democracy through Participatory

Budgeting

Porto Alegre is the capital city of the Brazilian state of Rio

Grande del Sul and has 1.4 million inhabitants. Participatory

budgeting, in Portuguese Orcamento Participativo or OP, was

introduced in 1989 when the Workers Party (Partido dos

Trabalhadores or PT) won the local government elections, and is

defined by Marquetti et al. (2009: 3) as ‘a form of

participatory democracy in which citizens and civil society

organizations have the right to participate directly in

determining fiscal policy’. The ideological commitment of the

PT to participation (Santos 1998) is key to explaining the

kinds of political will forthcoming from the Brazilian state

to support these new institutions. In what follows this

assessment will be nuanced by pointing to the indirect form of

representation central to the Porto Alegre case, both in

respect of the actual OP process itself and as regards the

significance of the PT process in the overall fiscal

management of the city. The case is made that OP in Porto

Alegre is mostly about broadening representation to improve

access and voice for poor and marginalised groups. In making

this case we begin with the deliberative phase of the OP

process.

Deliberation

Between 1998 and 2004 the participatory budgeting process in

Porto Alegre involved three kinds of forums in a hierarchical

‘chain of sovereignty’ (Baiocchi et al. 2011: 11). At the most

local level were regional assemblies or rodadas, and at central

level were the Forum of Delegates and the Council of the

Orcamento Participativo or COP. The participatory budgeting process

began each year in the lowest of these levels, the Regional

Assemblies, taking place in each of the sixteen regions of the

city. As Aragones and Sanchez-Pages (2008: 2) note, the

assemblies are the principal forum for popular participation.

They are open to any citizens, all of whom have the right to

participate, speak and vote, and indeed explicitly prevent

organisations from claiming rights to representation in them

to prevent forms of capture by organised groups. According to

Marquetti et al. (2009: 8–9), the regions were established to

create economic and social homogeneity among the constituents,

and so their sizes vary significantly. The largest, Downtown

(Centro), had a population of 266,896 in 2000, while the

smallest, Northeast (Nordeste), had a population of 28,518.

All the regions, however, were limited to a size that was felt

could ‘facilitate popular participation’.

Until 2001, the OP process started with a series of

meetings in each region from March to June. There were two

main regional meetings, called first and second rounds. These

were coordinated by the City, but the agenda was set jointly

by the local regional leaderships and the City. In addition,

the communities organised several local meetings called

‘intermediaries’ between rounds. These regional assemblies had

two main tasks. The first was to discuss local issues and then

democratically to decide on local investment priorities.

According to Aragones and Sanchez-Pages (2008: 2) discussions

focused on reaching consensus ‘on a rank of priorities for

each region and a list of hierarchical demands inside each

priority … each region selects as priorities five out of the

13 areas available’. All decisions were taken by majority rule

if consensus was not achieved. The second task of the regional

assemblies was to choose representatives for the later stages

of the OP process. Thus the regional assemblies also elected

members to the Forum of Delegates and the COP.

Parallel to the regional meetings, citywide thematic

assembly meetings were held to discuss themes of general

interest to the city, including: city organisation and urban

and environmental development; health and social assistance;

economic development and taxation; transport and circulation;

culture; and education, sport and leisure. As Marquetti et al.

(2009: 9) note, they were introduced in 1994 as the OP was

reformed and, like the regional assemblies, were open to the

direct participation of citizens but also civil-society

formations and city administrators. At this point the citywide

priorities identified through both the regional and citywide

deliberations were referred to the COP, where they were

refined in light of the investment funds made available from

the City. Hence, the scope of OP deliberations did not extend

to the division of budget between operational and investment

lines, but was restricted to the latter.

Already some important insights into the democratic

nature of the deliberation phase are evident. The first is

that any citizen is entitled to participate in defining and

ranking the important issues of the local region, through a

deliberative process. This is important as it differs sharply

from the Schumpeterian model where democracy is about elites

competing for office, supposedly on the basis of different

policy offerings. In the OP approach, the policy innovation

comes from citizens rather than representatives, and it is

subject to public debate under certain conditions of equality,

thus approaching the standards required for a ‘deliberative

democratic process’ (see Cohen 2002). Furthermore, the

decision to allow participation to individuals rather than

organisations is important to facilitating access to groups

not represented by civil-society formations, and allowing

their views to be heard. In this respect it is notable that in

2002 at least 61 per cent of participants were members of at

least one organisation, but that most of these were

neighbourhood associations, community centres and street

associations. Perhaps even more significantly, however, was

the fact that, in 1998, citizens with a household income not

greater than a third of the city average represented 57 per

cent of the participants in the meetings, whereas they only

accounted for 32 per cent of the population of Porto Alegre as

a whole. By 2003 this gap had closed significantly such that

the OP process represented the city as a whole much more

accurately. According to Aragones and Sanchez-Pages (2008: 11)

this was because the incentive for poor people to participate

in the OP process dropped off once their primary needs were

met.

In many ways this development redeems the points by

Gaventa et al. (2007) and Mohanty et al. (2011) about the

importance of social mobilisation for the effective

functioning of participatory institutions. Indeed, the Porto

Alegre case confirms the claim made by Mohanty et al. (2011:

23) that ‘the historical organised sectors of civil society

have managed to correspond to the technical requirements of

state programmes and to follow the numerous bureaucratic

negotiations taking place in simultaneous arenas of debate’.

Hence the drop-off in participation from non-civil society

mobilised sections. As noted above, it also confirms the

importance of political will in the critical role of the PT in

leading the innovation of participatory budgeting, thus

enabling the conditions set on ‘both sides of the equation’

(Gaventa 2006: 23).

While the article will discuss outcomes in more detail

below, the key point is that OP proved an effective way of

including historically marginalised groups – in this case the

poor – in the deliberative phase of the budgeting process in

an unprecedented way. Notably though, the highest cumulative

turnout in the OP process between 1989 and 2006 was 33,625 in

2002, namely 2.5 per cent of all 1.4 million citizens and

about 5 per cent of voters (ibid.: 3). Thus, while the

regional assemblies were open to all, only a tiny minority of

the eligible population participated. So while ‘ordinary

citizens’ may have participated directly in key aspects of

deliberation, it was not the community as a whole

participating, or even a representative sample for that

matter. I would suggest that the direct participation by

citizens in the OP process is better understood, from a

citywide systemic view, as a way of broadening representation

out to constituencies that have been historically marginalised

rather than an attempt at a city assembly. It is democratic in

enhancing equal influence in the decision-making process.

What we witness in the OP process then is the broadening

of representation in the citywide fiscal management system

through the use of participatory institutions and practices. In this

regard it is notable that Aragones and Sanchez-Pages (2008:

13) claim that the advantage of participatory democracy over

representative ‘is that it constitutes a reliable channel for

the transmission of information from the citizens to the

legislator about their preferences’. This point about the

representative impact of participatory institutions at the

system level is reinforced when one reflects on the influence

of city officials and others on the subsequent phases of the

deliberative and decision-making processes.

Decision

The process of actually deciding the final city budget happens

at the central level of the city where the recommendations

emerging from the OP process are formally endorsed by elected

politicians. Formally then, the decision moment rests with

representatives not a direct mechanism like a popular assembly

or referendum. Furthermore, even in finalising the proposed

budget, representatives play a central role – both those

elected from the regional assemblies and also those

representing key civil society groups and the administration

of the city.

Having determined the citywide investment in accordance

with the priorities identified by the regional assemblies, the

COP then distributes the resources among the regions. As

Marquetti et al. (2009: 10–11) note, the criteria for the

regional allocation include: (i) the lack of public services

and/or infra-structure in the region, (ii) the total

population in the region, and (iii) correspondence of the

priorities chosen by a region with those chosen by the city as

a whole. Each criterion gave a certain number of points for a

region. The resources were then allocated in proportion to the

points. The role of representatives in defining the final

budget proposal is made clear – not least in affirming the

priorities of the city over its component parts in criterion

(iii) above. Indeed, it is hard to see any other way that

budget proposals from sixteen regions of a city of over a

million people, and with insufficient resources to meet all

demands, could be worked through other than by a group of

representatives. Also important to note, though, is that the

COP is not only comprised of representatives from the regional

assemblies, but also has representatives from the Residents

Association Union of Porto Alegre, and the City Hall’s

Attendants Union. In this way the interests of more

established or elite social groupings are included at the

business end of the deliberative process.

The investment budget document was submitted to the City

Council at the end of September. Technically it is the City

Council and not the COP that decides the final investment

budget. As Marquetti et al. (2009: 4) note, while the elected

politicians of the Chamber of Deputies have total autonomy to

amend or defeat the budget as proposed by COP, the political

costs of turning it down are very high as the proposal has

been approved by citizens, assemblies and community

organisations. Nevertheless, there has been a history of

tensions around this issue, and especially since the Workers

Party (PT) lost the local elections in 2004. The process of

allocation among the regions was worked out in October and

November, and by December the COP had prepared the Plan of

Investment and Services (PIS). This booklet listed the entire

configuration of the budget and all public works approved for

all regions.

If the influence of representatives in the deliberation

process becomes more noticeable as the hard choices get made,

the power of elected representatives is profoundly affirmed in

their right to support, to reform or to reject the budget

emerging from the OP process. While all accounts note the

infrequency of wholesale changes, the point remains that the

opportunities for direct citizen participation in the

budgeting process are expressly excluded from the decision

phase. Thus while an important part of the deliberative (and

as we shall see) also the implementation phases, direct

participation is not a feature of the decision moment of OP.

Implementation

According to Marquetti et al. (2009: 11), the implementation

of the budget by the executive branch typically begins in

January of each year. They argue that there are four main

channels of monitoring this implementation. The first is

continued oversight by the COP, which continues until the new

COP is elected after the second round of the local assemblies

in year two of the OP process. The second is the elected Forum

of Delegates, whose responsibilities included monitoring

public works, keeping the community informed during the OP

process and collecting new demands for future work. Notably,

these are both elected forums and thus representative

mechanisms of oversight over executive implementation. In

addition, however, there are some participatory mechanisms

too. The third mechanism for monitoring is the local and

regional assembly meetings that began again in March of year

two of the OP process. These meetings have a monitoring role

as they begin with report-backs on the results from the

previous year’s implementation. The final channel is the PIS

booklet available to all citizens that outlines all the

formally approved public works. This document allowed the

regional assemblies, and indeed citizens more widely, directly

to monitor the city government to see if it was executing the

agreed-upon projects.

Importantly, as with the COP and Forum of Delegates, the

power of oversight does not convey formal powers of censure so

much as apply political pressure that can become profoundly

meaningful at election time. Thus, once again, the power of

citizens in the implementation phase is exercised indirectly

through various opportunities to scrutinise and to evaluate

publicly the performance of professionals overseen by elected

officials. It was nevertheless effective, and I think

ultimately consistent enough with Fung and Wright’s (2001)

requirements that participatory institutions be empowered to

influence decisions significantly. Hence, as Aragones and

Sanchez-Pages (2008: 13) observe, ‘the main advantage of the

system of participatory democracy over direct democracy is

that it includes a plausible policy implementation mechanism

that the latter lacks’.

Outcomes

Above a case has been built for the ways in which

participatory budgeting in Porto Alegre is an institution

designed to include citizens in their own right, and through

their representatives, directly in the deliberation and

implementation phases of the fiscal management process in the

city, and also indirectly in the decision phase by offering a

proposal with great legitimacy to the elected representatives

of the city. Certainly during the reign of the PT, the party

that championed the OP process, the commitment to public

participation meant that the popular influence manifest in the

OP process carried through relatively seamlessly into actual

policy adoption and implementation. With the succession of

parties in 2004, the OP continued but enjoyed less standing

and thus less influence with the formal political rulers of

Porto Alegre. In brief, the political will to support

participatory institutions had been weakened.

In addition to arguing that the OP process in and of

itself deepened democracy through broadening the

representation of historically marginalised groups through

participation institutions and practices, there is clear

evidence that the outcomes of the OP process in Porto Alegre

clearly benefitted the poor, especially in the early years.

Thus, as Aragones and Sanchez-Pages (2008: 11) note:

In 1989 in Porto Alegre only 49% of the population was

covered by basic sanitation; this percentage rose to 85%

in 1996. Over the same period, the proportion of

households with water supply reached the 99.5% and the

number of students enrolled in elementary schools

increased by 190%; five times more housing units were

constructed in the period 1993–1996 than in the period

1986–1988.

Indeed, Aragones and Sanchez-Pages (2008: 13) argue that OP

led to a very high level of redistribution of income within

the city. Given that until 1988 Brazil was an authoritarian

system, one might well expect a new democratic government of

any kind to direct more resources to the poor, and thus the

special character of participatory budgeting as a pro-poor

fiscal instrument might be called into question. However, the

fact that budget priorities were defined through the OP

process and the fact that disproportionate levels of poor

people participated suggest that OP did offer a unique

opportunity for a historically marginalised constituency to

express itself. This counts as significant, if circumstantial,

evidence of the real empowerment of the participatory

budgeting process in Porto Alegre. In this regard it is

notable that, once basic needs had been covered in the first

few years, both the participation rates of poor people dropped

off and the priorities identified through the OP process

shifted to issues that were also attractive to middle-class

citizens, like education, health and social services.

In addition to the above outcomes directly linked to

budget spending, there are two indirect democratic outcomes

that Aragones and Sanchez-Pages (2008: 3) identify. First, the

city witnessed a remarkable improvement of the behaviour of

politicians and community leaders who, as in the rest of

Brazil, were used to clientelism. Confronted with a more

informed population and civil-society formations, the

opportunities for the politics of ‘gift exchange’ were

reduced, and the accountability mechanisms around the budget

implementation have reduced corruption. Similar results have

been found in other places in Brazil where OP has been

implemented (Baiocchi et al. 2011). Another remarkable success

of OP has been the continual increase in public participation

up to 5 per cent of the population and, most importantly, the

substantial participation of the less educated and wealthy

segments of the population – ‘precisely those typically

disengaged from the political process prior to the adoption of

the OP’ (ibid.: 3).

To these indirect outcomes we can add another, the

dramatic spread of the OP practices across Brazil and the

world. Hence, Cabannes (2006) estimated that more than 1,000

of Brazil’s roughly 16,000 municipalities had adopted some

form of participatory budgeting by 2006, and similar

experiments are evident today in countries as diverse as

India, Mozambique, the United Kingdom and the United States.

Further, while there is not yet substantial literature on the

impact of participatory budgeting across these diverse

contexts, the work that does exist on the Brazilian cases

suggests that it is making some, usually modest, contribution

to deepening local democracy (Baiocchi et al. 2011). It is

clearly an institution that can work, although to the extent

of reforming rather than transforming the budgeting process of

local governments.

Conclusion

John Dunn’s claim that ‘today in politics democracy is the

name for what we cannot have – but cannot cease to want’

(1992: vii) conceals as much as it reveals because it

downplays the potential benefits of democratic innovation.

Thus participatory budgeting in Porto Alegre yielded both

procedural and substantive democratic gains for poor citizens.

Notably, these gains seemed to rise and fall in correspondence

with the rise and fall of political will and social

mobilisation, key conditions of the effective functioning of

participatory institutions identified by Gaventa et al.

(2007). Of course, participatory budgeting supplements rather

than replaces mainstream representative institutions, and it

reforms rather than transforms the fiscal management of local

government, but it nevertheless makes a real difference to the

lives of poor people – or at least it can.

Indeed, while the Porto Alegre process experimented with

practices of direct democracy, and is widely termed a

‘participatory’ institution, the systemic effect of these

innovations was to broaden the representation of marginalised

groups in the budgeting process. Further, it was not the case

that the interests of poor people were not championed by

politicians prior to participatory budgeting, but rather than

the innovations of direct inclusion allowed for greater

access, louder voices and thus more impact on decisions that

were ultimately made by politicians. In this systemic sense

then, the participatory budgeting process can be seen as

broadening out representation to reflect more equitably the

interests of the poor majority. Notably, a key finding of the

article not captured in the Gaventa model is how the success

of participatory institutions in delivering outcomes can

negatively affect the ongoing participation of marginalised

groups, thus yielding temporary procedural democratic gains.

Participatory budgeting is one particular institution

used to reform representative democracy, but many others do

exist, and it is clearly possible that many other kinds could

exist too. For those of us who do want democracy, and feel the

pinch of democratic recession, participatory budgeting reminds

us that it is possible to find new ways of representing

marginalised groups in the decision-making processes of our

liberal-democratic capitalist system that work. Paradoxically

these innovations may include forms of direct democratic

practice and new types of participatory institutions, but

perhaps these are just the components of the popular

representation of the marginalised required by the name

democracy.

References

Aragones, E. and S. Sanchez-Pages. 2008. ‘A Theory of

Participatory Democracy Based on the Real Case of Porto

Alegre’, European Economic Review 53(1): 56–72.

doi:10.1016/j.euroecorev.2008.09.006

Arnstein, S. R. 1969. ‘A Ladder of Citizen Participation’,

Journal of the American Planning Association 35(4): 216–24.

Baiocchi, G., P. Heller and M. K. Silva. 2011. Bootstrapping

Democracy: Transforming Local Governance and Civil Society in Brazil.

Stanford: Stanford University Press.

Bryan, F. M. 2003. Real Democracy: The New England Town Meeting and How

It Works. Chicago: University of Chicago Press.

Budge, I. 2001. ‘Direct Democracy’, in P. A. B. Clarke and J.

Foweraker (eds), Encyclopedia of Democratic Thought. London:

Taylor and Francis, 224–7.

Cabannes, Y. 2006. ‘Les budgets participatifs en Amérique

Latine’, Mouvements 47: 128–38.

Christiano, T. 2004. ‘The Authority of Democracy’, Journal of

Political Philosophy 12(3): 266–90.

Cohen, J. 2002. ‘Deliberation and Democratic Legitimacy’, in

D. Estlund (ed.), Democracy, Maldon and Oxford: Blackwell,

87–106.

Cornwall, A. 2002. ‘Making Spaces, Changing Places: Situating

Participation in Development’, IDS Working Paper 170,

Brighton: Institute of Development Studies, University of

Sussex.

Cornwall, A. and V. S. Coelho (eds). 2007. Spaces for Change? The

Politics of Participation in New Democratic Arenas. London: Zed.

Dahl, R. A. 1972. Polyarchy: Participation and opposition. New Haven,

CT: Yale University Press.

Dahl, R. A. 1998. On Democracy. New Haven, CT: Yale University

Press.

Dunn, J. 1992. Western Political Theory in the Face of the Future.

Cambridge: Cambridge University Press.

Fung, A. 2004. Empowered Participation: Reinventing Urban Democracy.

Princeton: Princeton University Press.

Fung, A. and E. O. Wright. 2001. ‘Deepening Democracy:

Innovations in Empowered Participatory Governance’, Politics

and Society 29(1): 5–41.

Gallie, W. B. 1956. ‘Essentially Contested Concepts’,

Proceedings of the Aristotelian Society 56: 167–98.

Gaventa, J. 2006. ‘Triumph, Deficit or Contestation: Deepening

the Deepening Democracy Debate’, IDS Working Paper 264.

Brighton: Institute of Development Studies, University of

Sussex.

Gaventa, J., R. McGee and T. Zipfel. 2007. ‘How Do Local

Governments Support Citizen Engagement? Background Note

for the International Workshop for Champions of

Participation in Local Government’, IDS Workshop,

University of Sussex, 29 May–4 June.

Geuss, R. 2001. History and Illusion in Politics. Cambridge: Cambridge

University Press.

Habermas, J. 2002. ‘Deliberative Politics’, in D. Estlund

(ed.), Democracy. Maldon and Oxford: Blackwell, 107–126.

Held, D. 2006. Models of Democracy. Cambridge: Polity.

Hendricks, C. M. 2006. ‘Integrated Deliberation: Reconciling

Civil Society’s Dual Role in Deliberative Democracy’,

Political Studies 54(3): 468–508.

Holston, J. 2009. Insurgent Citizenship: Disjunctions of Democracy and

Modernity in Brazil. Princeton, NJ: Princeton University

Press.

Huntington, S. 1993. The Third Wave: Democratization in the Late 20th

Century. Norman: University of Oklahoma Press.

Keane, J. 2009. The Life and Death of Democracy. London: Simon and

Schuster.

Mansbridge, J. 2003. ‘Rethinking Representation’, American

Political Science Review 97(4): 515–28.

Marquetti, A., C. E. S. da Silva and A. Campbell. 2009.

‘Participatory Economic Democracy in Action:

Participatory Budgeting in Porto Alegre, 1989–2004’, Paper

for the Eastern Economic Association Meeting, 27 February–1 March.

Miraftab, F. 2006. ‘Feminist Praxis, Citizenship and Informal

politics: Reflections on South Africa’s Anti-Eviction

Campaign’. International Feminist Journal of Politics, 8 (2): 194–

218.

Mohanty, R., L. Thompson and C. Schattan Coelho. 2011.

‘Mobilising the State? Social Mobilisation and State

Interaction in India, Brazil and South Africa’, IDS Working

Paper 359. Brighton: Institute of Development Studies,

University of Sussex.

Piper, L. and R. Deacon. 2009. ‘Too Dependent to Participate:

Ward Committees and Local Democratisation in South

Africa’, Local Government Studies 35(4): 415–33.

Pitkin, H. F. 1967. The Concept of Representation. Berkeley:

University of California Press.

Raphael, D. D. 1976. Problems of Political Philosophy. London: Pall

Mall.

Santos, B. de S. 1998. ‘Participatory Budgeting in Porto

Alegre: Towards a Redistributive Democracy’, Politics and

Society 26: 461–510.

Schumpeter, J. 1942. Capitalism, Socialism and Democracy. London:

Routledge.

World Bank. 1996. The World Bank Participation Sourcebook. Washington,

DC: World Bank.

Young, I. M. 2000. Inclusion and Democracy. Oxford: Oxford

University Press.