%% LyX 2.0.6 created this file.  For more info, see http://www.lyx.org/.
%% Do not edit unless you really know what you are doing.
\documentclass[english]{article}
\usepackage[T1]{fontenc}
\usepackage[latin9]{inputenc}
\usepackage{babel}
\usepackage{float}
\usepackage{booktabs}
\usepackage{textcomp}
\usepackage{amsthm}
\usepackage{amsmath}
\usepackage{graphicx}
\usepackage{setspace}
\usepackage[unicode=true,pdfusetitle,
 bookmarks=true,bookmarksnumbered=false,bookmarksopen=false,
 breaklinks=false,pdfborder={0 0 1},backref=false,colorlinks=false]
 {hyperref}
\usepackage{breakurl}

\makeatletter

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% LyX specific LaTeX commands.
%% Because html converters don't know tabularnewline
\providecommand{\tabularnewline}{\\}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Textclass specific LaTeX commands.
\numberwithin{equation}{section}
\numberwithin{figure}{section}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% User specified LaTeX commands.
\usepackage{url}
\usepackage{hyperref}
\usepackage{indentfirst}
\usepackage[super,sort&compress]{natbib}

\makeatother

\begin{document}
\begin{doublespace}

\title{An Overview of Participatory Budgeting}
\end{doublespace}

\begin{doublespace}

\author{Adrien Fabre}
\end{doublespace}

\maketitle
\begin{doublespace}

\section{Introduction}
\end{doublespace}

\begin{doublespace}
Participatory Budgeting (PB) is a form of direct democracy at the
local level, where all citizens state annually prioritized sectors
to be financed before allocating investment resources in projects
of their choice in those sectors. It has four key moments: diagnosis,
deliberation, decision-making and follow-up\cite{Human}. Researchers
agree to say that PB has emerged in some Brazilian municipalities
after the end of the military junta in the late 80's, notably in those
run by the Workers' Party. The epitome of PB is Porto Alegre, the
capital of the Brazilian state Rio Grande do Sul: observing its success,
many towns in Brazil engaged also a PB and now the process is spreading
worldly: more than 1500 towns implement diverse variants of PB, including
Buenos Aires, a district of Chicago, Cologne, Paris, etc. Indeed,
Porto Alegre's PB helped to allocate the municipal resources equitably,
to democratize the town, making the decisions more transparent, accountable
and efficient, nurtured solidarity among groups, reinforced social
ties and promoted the collective pursuit of the common good\cite{Zamboni2007}.
As lots of studies have been conducted on PB, this paper will summarize
the main observations. More precisely, we will examine to what extent
PB can be an amelioration of local political life, choices, and social
conditions.
\end{doublespace}

\begin{doublespace}

\section{What is Participatory Budgeting?}
\end{doublespace}

\begin{doublespace}
PB takes many forms, depending on local specificities as well as on
the degree of devolution of power to citizens. To clarify the subject,
Sintomer\cite{SintomerEtAl2014} has retained five criteria required
to qualify a PB : all citizens participate to the choice of the allocation
of resources, this occurs at an administrative level of power (for
example a town, but not a quarter), this process has to be repeated
over years, a deliberation must take place in a specific institution,
and finally citizens should be informed $ex$ $post$ of the status
of the budget and the projects.

We will present here the case of Porto Alegre, which is the most studied,
mainly because it was the first full implementation of PB in a big
city (1.3 million of inhabitants) hence a template that all other
municipalities in Brazil and elsewhere are following. Although the
very organization of the mechanisms vary across towns, they all share
in common frequent physical meetings where everyone is invited to
participate while delegates are elected to express the choices of
citizens in a dedicated council.\cite{WorldBank2008}
\end{doublespace}

\begin{doublespace}

\subsection{A typical PB cycle}
\end{doublespace}

\begin{doublespace}
PB is the central element in the participatory governance architecture
of Porto Alegre, which consists in a complex system of civil society-government
arrangements, including municipal policy councils, participatory planning
fora, and a citywide strategic planning conference. The city government\textquoteright{}s
\textquotedblleft{}Presentation of Accounts\textquotedblright{} from
the previous year marks the beginning of the PB process in March-April.
From April until May, regional and thematic assemblies take place
with the objectives of establishing thematic priorities by voting,
electing councilors for each region, defining the number of delegates,
and repeating the budget review for the preceding year at the local
level. 
\begin{figure}[H]
\begin{centering}
\includegraphics[scale=0.4]{Fig1}
\par\end{centering}

\caption{PB cycle in Porto Alegre}
\end{figure}
These meetings are open to all citizens and constitute the central
and most inclusive component of the process. Porto Alegre is divided
into 16 regions, which are again subdivided into several micro-regions
where regional assemblies are held to ensure close proximity to the
grassroots level. The five thematic areas discussed in the PB at city-level
assemblies are: transport and traffic; education, leisure, and culture;
health and social welfare; economic development and taxation; and
organization of the city, urban and environmental development. From
May until July, regional and thematic assemblies elect delegates whose
main responsibility is the ranking of funding requests under each
theme and sub-theme for each region, from the propositions of citizens
and assessments of projects by the city administration, using a well
defined formula based on objective criteria: number of votes, population
size of the region and level of deficiency of infrastructures. To
encourage participation in district-level elections, the number of
delegates is kept proportional to the participants attending the meeting
in which the elections take place. 
\begin{table}[H]
\centering{}\caption{PB demands per sector, 1990-2007\cite{WorldBank2008}}
\begin{tabular}{ccc}
\toprule 
Sector of demands & Frequency & \% of total PB\tabularnewline
\midrule
\midrule 
road works & 1566 & 26,2\%\tabularnewline
treatment of water and sewage & 1465 & 24,6\%\tabularnewline
housing & 760 & 12,7\%\tabularnewline
transportation & 548 & 9,2\%\tabularnewline
\bottomrule
\end{tabular}
\end{table}
 In the summer, councilors prepare the draft investment plan in coordination
with the municipal administration, which undertakes a technical feasibility
and financial analysis of the proposed projects. In August and September
the investment plan is incorporated into the city\textquoteright{}s
budget and submitted to the local legislature for revision and approval.
Although technically the mayor can refuse the budget proposed by this
procedure, not any veto has ever occurred. The final step, in November
and December, is a revision of PB's statute. Amendments to and modifications
of its rules can be proposed and voted on before the new cycle begins
the following March. After many tries, Porto Alegre has chosen a subtle
combination of direct democracy and representation to build its budget.
Some PB however, like the one of Rio Grande do Sul, are more direct:
there are still delegates but citizens vote directly for the projects
(and not only for thematics).

\begin{table}[H]
\caption{State Rio Grande do Sul \textemdash{} Development voted by Regional
assemblies\cite{IADB2005}}


\centering{}%
\begin{tabular}{cccc}
\toprule 
Year\textbackslash{}Priority & 1st & 2nd & 3rd\tabularnewline
\midrule
\midrule 
2000 & Agriculture & Jobs \& Wages & Transportation\tabularnewline
2001 & Agriculture & Education & Transportation\tabularnewline
2002 & Education & Agriculture & Jobs \& Wages\tabularnewline
2003 & Agriculture & Jobs \& Wages & Education\tabularnewline
\bottomrule
\end{tabular}
\end{table}
 
\begin{table}[H]
\caption{Porto Alegre: Criteria for Allocation of Capital Investments Resources
among Regions and Municipalities\cite{IADB2005}}


\centering{}%
\begin{tabular}{cc}
\toprule 
Category & Grade\tabularnewline
\midrule
\midrule 
\multicolumn{2}{c}{Regional Thematic Priority (Relative Weight 5)}\tabularnewline
\midrule 
First Priority  & 4\tabularnewline
Second Priority  & 3\tabularnewline
Second Priority  & 2\tabularnewline
Forth Priority & 1\tabularnewline
\midrule
\midrule 
\multicolumn{2}{c}{Population Size (Relative Weight 2)}\tabularnewline
\midrule 
Up to 25.00 inhabitants  & 1\tabularnewline
From 25.001 to 45.000 inhabitants  & 2\tabularnewline
From 45.001 to 90.000 inhabitants  & 3\tabularnewline
Above 90.001 inhabitants  & 4\tabularnewline
\midrule
\midrule 
\multicolumn{2}{c}{Degree of Deficiency in Infrastructure or Services (Relative Weight
4)}\tabularnewline
\midrule
From 0,01 to 14,99 \%  & 1\tabularnewline
From 15 to 50,99 \%  & 2\tabularnewline
From 51 to 75,99 \%  & 3\tabularnewline
From 76,00 to 100\%  & 4\tabularnewline
\bottomrule
\end{tabular}
\end{table}

\end{doublespace}

\begin{doublespace}

\subsection{An enhanced civic participation}
\end{doublespace}

\begin{doublespace}
Surveys and attendance to local meetings show that PB is a success
in terms of civic participation and awareness. In Porto Alegre, 10
years after its introduction, around one-fifth of the population reported
having participated in the PB at some point in their lives. Usually,
around 4\% of the population attend the meetings in a PB. This is
not so low that one can first think, because families, people from
a given organization or from a same building often send only one representative
at a time to share their concerns. That said, the number of participants
can be increased using ICT: Belo Horizonte has reached 10\% of citizens
in the biggest e-Participatory Budgeting\cite{Peixoto2008}, 7 times
more than for its traditional PB, while\textbf{ }voting by SMS in
La Plata (Argentina) doubled the number of voters\cite{Um2013}.\textbf{ }

One could also think that having official meetings to discuss local
policies would lower the activity of neighborhood associations, it
is exactly the opposite that happens: people gather more than before
to prepare their demands, knowing that they will be listened. Indeed,
a majority of the delegates in Porto Alegre are also involved in housing
associations or active in a community. On average, activists (like
councilors) participate in 2 or 3 meetings a week, and are regular
attendees in 3 to 4 different forums\cite{Baiocchi1999}.
\end{doublespace}

\begin{doublespace}

\subsection{The issue of representativeness}
\end{doublespace}

\begin{doublespace}
Representativeness is not perfect, although social backgrounds of
participants and councilors are much more diverse than (regularly)
elected representatives. Depending on the design of the PB, poorer
or richer people are incited to participate: in rich countries, high
income and white people are often overrepresented, whereas it is the
opposite in Brazil\cite{LernerEtAl2012,Nieuwland2006}. However, we
observe that men and high income speak more in public meetings than
uneducated and women, and they are more likely to be elected as delegate
or councilor\cite{Baiocchi1999}. This is why, when organizing direct
democracy, one should really make effort to insure representativeness.
First, one has to understand why some categories of population do
not show up: young or entrepreneurs often lack of time, rich lack
of motivation to participate, knowing that the funds are directed
towards poor neighborhood, for paving or basic sanitation, for example.
Or, on the contrary, in Canada or in New York, where basic needs are
already fulfilled, poor people are underrepresented, notably because
they are less prepared to speak in public. Knowing this, the problem
can be addressed by an appropriate plan. In Brazil, we have seen with
the case of Belo Horizonte that high income people can be reached
by letting them vote for the projects on the Internet, even though
they still participate too less in public deliberation. In New York,
teams of volunteers and canvassers dedicated hundreds of hours in
phone calls, door knocking, flyering, meeting with local groups, and
using other tactics to engage low-income New Yorkers, public housing
residents, and immigrant populations in each district (notably by
engaging meetings in Spanish). This work proved invaluable, as targeted
populations turned out to vote in PB at higher rates than in the local
elections. For example, in District 8, where targeted outreach focused
on public housing tenants, 22\% of PB voters had household income
less than \$10,000, compared to 4\% of the district\textquoteright{}s
voters in the 2009 City Council election\cite{LernerEtAl2012}.
\end{doublespace}

\begin{doublespace}

\subsection{Impact on the budget}
\end{doublespace}

\begin{doublespace}
A municipal budget is divided in two parts: running costs (wages of
teachers, doctors, municipal agents, etc.) and investment budget.
PB concerns only the investment part, which rarely outreach 20\% of
the total budget, and often concerns only a fraction of it. For example
in Rio Grande do Sul, which has a PB at the regional level from 1998
(involving more than 350 000 participants and 210 councilors), the
PB covers all capital investments, which accounted for 11.3\% of the
state budget 2002 (\$ 450 millions), whereas in Paris, the nascent
PB is limited to only 5\% of the investment budget (70 millions of
euros). Thus, it is unlikely that the mayor be impeded to run all
her desired policy because of the PB, especially since a common criticism
of PB is that they are largely influenced by the mayor's propositions,
the usual citizens lacking technical knowledge to deepen their projects.
Still, the influence of broadly based local groups on the city councils
seem to be reducing the hold of a few powerful people that generally
prevails in public administration. Indeed, the PB has opted for transparency,
objectivity and relevance in its quest to engage citizens in local
governance. The resource allocation process has made it a rule that
only quantifiable criteria and indicators are used\cite{IADB2005}.

Finally, a reasonable estimate of personnel working full-time on the
PB (the $direct$ impact of PB on the budget) is about 35 municipal
employees in Porto Alegre. 
\end{doublespace}

\begin{doublespace}

\section{History}
\end{doublespace}

\begin{doublespace}

\subsection{80s: the South American origins}
\end{doublespace}

\begin{doublespace}
The extraordinary explosion of urban social movements and urban self-help
organizations in Latin America during the 1960s and 1970s were central
to the emergence of a strong civil society which, in turn, became
an important element in the process which led to the dismantling of
authoritarian government in the 1980s\cite{Stren2000}. In Santiago,
Chile, for example, a survey found that in the 1980s, 20\% of the
marginal urban population participated in popular organizations where
most of the members, as well as the leaders and managers, were women.
In 1994, UNITAID estimated that soup kitchens were feeding one Peruvians
over three. In Brazil in the 70's, in both poor neighborhoods and
middle-class areas, the population organized to demand the right to
basic services \textemdash{} water supply, sewerage, school facilities,
health facilities, roads \textemdash{} and protested against ecological
dangers, development plans which ignored residents' interests like
housing evictions, in local organizations and networks, such as neighborhood
associations, as well as national movements, such as the Housing Movement
and the Collective Transports Movement. Experimentation with participatory
schemes to achieve greater local control over resource allocation
decisions began in a number of municipalities governed by opposition
parties while the military regime was still in power\cite{WorldBank2008}.
\end{doublespace}

\begin{doublespace}

\subsection{90s: the Brazilian case}
\end{doublespace}

\begin{doublespace}
One of the main reasons why Brazilian reforms have moved so quickly
after the dictatorship has been the passing of a new Constitution
in 1988. This Constitution strengthened municipal autonomy, validated
the participation of community groups in municipal decision-making,
and attributed important social and economic policy functions to municipal
authorities. Under Brazil's Constitution, municipalities are given
the opportunity to establish ``organic'' laws, by which they may
structure their own operations and set up what are called ``municipal
boards''. These boards, in turn, have the formal function of mediating
between the local government and organized civil society. The most
important functions of these boards are health and education (defined
in the new Constitution as municipal powers). Aside from the promotion
of municipal boards, eighteen of the fifty largest Brazilian cities
instituted PB. The Workers' Party (WP) has been determinant in this
expansion: from 1989 to 1992, half of PB where taken place in WP cities,
although they used to run only 1\% of municipalities\cite{WorldBank2008}.
The proximity with the mass (the Workers' Party has 1.4 millions of
members) as well as the concern for inequality has contributed for
sure to both the reduction of poverty in Brazil and the sound success
of the WP, in charge of the country from 2002 even if it was practically
unknown in the early 80's. It is worth noting that the PB succeeded
partly because everyone was welcomed to participate, not only supporters
of the WP.
\end{doublespace}

\begin{doublespace}

\subsection{00s: the global spread}
\end{doublespace}

\begin{doublespace}
The World Social Forum held in Porto Alegre of 2001 (and the subsequent
editions) has given a tremendous inspiration to organizations all
over the World to reproduce direct democracy initiatives. Lots of
PB have emerged in Europe in the beginning of the century, supported
by various networks of the civil society such as URB-AL (which established
links with Latin America) or Cities of Tomorrow (echoing the model
of Christchurch in New Zealand), or governmental organizations such
as PB Unit in UK (which however had to shut down in 2012 because of
budgetary restrictions), CGLUA in Africa or Civic Network in Poland.
Poland is by far the European country with the highest number of PB,
thanks to a 2009 law which gives funds to municipalities which implement
PB. In Africa, international institutions helping for the development,
especially the World Bank, have played an important role. In Asia,
some initiatives are conducted by the authorities, in South Korea
and in the Indian state of Kerala notably. Depending on taking a restrictive
or a wide definition of PB, the number of PB vary from 1269 to 2778,
the three quarters being in Poland, Brazil, Peru and Dominican Republic.
Apart from Germany, South Korea and Madagascar, the remaining 48 countries
experiencing some PB have at most 30 of them\cite{SintomerEtAl2014}.
\end{doublespace}

\begin{doublespace}

\subsection{Local PB but national support}
\end{doublespace}

\begin{doublespace}
Wherever PB are largely spread, there has been a strong support or
even an obligation from the national government. In Peru, in the aftermath
of the authoritarian and corrupted government of Fujimori, the process
of democratization has been stimulated by a 2002 law instituting participatory
budgeting in all the country, at the regional level as well as municipal.
However, many of these experiences do not meet the criteria of Sintomer,
mainly because they involve representatives of organizations and not
all citizens, but also because when local governments do not support
PB, they remain factitious. This is why Peru has at the very most
300 real PB (one sixth of all districts). Still, one third of the
infra-national budgets (\$400 millions) were debated in a participative
design in 2007\cite{SintomerEtAl2014}. In Dominican Republic, PB
has also become mandatory in 2007, satisfying a demand of the Federation
of Municipalities aimed to force the central government to honor its
legal obligation to transfer 10\% of the national budget to local
authorities (which has still not been done). In South Korea, some
laws between 2005 and 2007 passed to foster democracy, promoting transparency
and defining a model of PB, in the context of a centralized country,
despite its tradition of participation. Finally, the success of all
PB (measured as the level of participation and the devolution of power)
vary a lot across municipalities of a given country, depending on
the motivation of the city council as well as on the implication of
citizens.
\end{doublespace}

\begin{doublespace}

\section{Effects}
\end{doublespace}

\begin{doublespace}

\subsection{The case of Porto Alegre}
\end{doublespace}

\begin{doublespace}
We will continue to take Porto Alegre as an example, as it is a typical
case of successful PB in a poor country. The main findings are: a
better allocation of resources (more equitably and appropriately distributed
among sectors and communities), an increase in the financing of basic
needs, and a more inclusive, less corrupted democracy. We have similar
results in other cities in Brazil, in Kerala, with its 2 millions
participants PB\cite{HellerEtAl2003} or in La Plata, where there
is widespread evidence that the initiative has been positively impacting
on the neediest sectors of society. For instance, since the project
was implemented, based on citizens' decisions, health care services
provided by the municipality have doubled, which is associated to
a decrease of 1\% in child mortality in the city\textbf{\cite{Um2013}.}

The overall outcomes of PB investments are an impressive testament
to how redistributive this process has been. Between 1983 and 1987
an average of 5296 meters of sewage systems were built per year in
Porto Alegre. The average in 1988 when Olívio Dutra (WP) was elected
mayor was of 8.091 meters. Between 1989 and 1992 the annual average
rose to 21674 and between 1993 and 1996 the average reached 61334
meters\cite{Avritzer2002}. Sewer and water connections went up from
75\% of total households in 1988 to 98\% in 1997\cite{WorldBank}.
In the years between 1992-1995, the housing department offered housing
assistance to 28,862 families, against 1,714 for the comparable period
of 1986-1988\cite{Baiocchi1999}. The number of functioning public
municipal schools reached 86 in 1999 against 29 in 1988. Porto Alegre\textquoteright{}s
health and education budget increased from 13\% in 1985 to almost
40\% in 1996\cite{Baiocchi1999}. 

Before PB, poor neighborhoods were not well organized to claim to
the city council, more prone to listen to the demands of rich ones
for a cleaner and more secure city. Not surprisingly, the new attention
given to poor people raised their standard of living, even without
any fiscal redistribution.

Furthermore, local solidarity and concern for the environment have
resulted in some bold decisions in Porto Alegre. In one instance,
despite the influence and employment assurance of a large motor vehicle
company, the city turned down a proposal for a new automotive plant,
believing that the required subsidies could be better used for other
city requirements. Similarly, a proposal for a five-star hotel at
a decommissioned site was turned down, and it was decided to use the
well-situated site to develop a public park, a convention hall, and
a public symbol of the city\cite{WorldBank}. 

Overall, the level of satisfaction of citizens with the PB, with 58.5\%
of participants claiming having received benefits in their region
(or thematic meeting), and 57.2\% asserting that the population always
or almost always ``really decides'' upon public works, is consistent
with an efficient deliberative system\cite{Baiocchi1999}.

An important effect of the PB process is a burgeoning of civic activity.
As participatory budgeting developed, the numbers of political, cultural,
and neighborhood groups has doubled, especially in poorer districts
where results of self-generated new city expenditures are remarkable.
People in wealthier districts also like what is going on. The value
of their properties in poorer districts is rising. A new city \textquotedblleft{}energy
of accomplishment\textquotedblright{} spawned a campaign to get property
owners to pay their taxes, and it worked\cite{Lewit2002}. The urban
violence decreased also as a consequence of better social inclusion.
\end{doublespace}

\begin{doublespace}

\subsection{An alleged increase in tax revenues and generosity}
\end{doublespace}

\begin{doublespace}
The unexpected by-product of PB over tax revenues can be hard to believe.
Still, the mayor of Porto Alegre claimed in 2000 that the popularity
of the PB has contributed to a tripling of the tax revenues for the
city and an outside study of the city demonstrates that from 1992
to 1995, the city increased its total tax receipts by 34\%\cite{Stren2000}.

Interrogated on the question, Allegretti\cite{Allegretti2013} speaks
of a 
\end{doublespace}
\begin{quotation}
\begin{doublespace}
\textit{virtuous circle. It mostly occurs in Africa, but elsewhere
too. Mayors from three different countries \textendash{} Benin, Zambia
and DR Congo \textendash{} told me that their revenues rose thanks
to PB. Since people now know what the money is used for, they are
more willing to pay taxes. In some mining cities in Madagascar, moreover,
PB allowed citizens and authorities to join forces and press the mining
companies to actually pay the money they owe the local authorities
by law. The companies did not use to pay, but now they do. In Peru
some cities have invented a rights-and-duties mechanism. The lower
a neighborhood\textquoteright{}s tax evasion is, the more money is
allocated.}\end{doublespace}

\end{quotation}


\begin{doublespace}
In the same interview, the researcher is asked if people are not too
egoistic and selfish to decide responsibly. Here is his answer:
\end{doublespace}
\begin{quotation}
\begin{doublespace}
\textit{In the 15 years I\textquoteright{}ve been involved in PB matters,
I\textquoteright{}ve had very many positive surprises. In the Dominican
Republic, people donated some of their land to the municipality for
free, so a park could be built. People are more generous and their
sense of solidarity is stronger than you would think. But authorities
have to stimulate the exchange. If people stay in privacy, they tend
to be more egoistic. Some PB approaches, for instance in Germany,
merely ask people to indicate their individual preferences via the
Internet, so they don\textquoteright{}t actually discuss anything.
And they don\textquoteright{}t learn to listen to each other\textquoteright{}s
perspectives. Other PB approaches put much more emphasis on bringing
people together and building consensus. In those cases, voting is
only a last resort in case people can\textquoteright{}t agree after
long discussions.}\end{doublespace}

\end{quotation}


\begin{doublespace}
Thus, Allegretti reminds us that PB is not only a matter of reducing
the inequality or the corruption, but that it is mainly an inclusive,
insightful and convivial way of taking political decisions.
\end{doublespace}

\begin{doublespace}

\subsection{A comparative study}
\end{doublespace}

\begin{doublespace}
One ought to be skeptical in front of the previous results. Indeed,
an increase in basic needs in one town that implemented PB is not
a proof that PB is efficient to struggle poverty. On the contrary,
one has $a$$priori$ all the reasons to believe that the above figures
are driven by independent, although simultaneous (and even correlated)
factors, such as growth, wider democratization of the country, tenure
of a Workers' Party mayor, or simply a singularity from Porto Alegre.
Fortunately, the World Bank conducted a comparative study in Brazil
which gives credit to the previous results\cite{WorldBank2008}. However,
this is the only study of that kind, and one would like to know if
the results are robust to the country. Yet, PB in Poland or in South
Korea for example, were enacted too recently for studies to have had
the time to assess their consequences.

To assess the impact of PB, the World Bank has compared cities which
implemented PB in Brazil with a random sample of other comparable
cities in Brazil, controlling for GDP per capita and for the number
of votes for the Workers' Party (they preferred to control for this
rather than for the party of the mayor because PB were much more likely
to occur where the WP made good score, even if another party \textemdash{}
often also from the left \textemdash{} eventually won the election).
The econometric analysis suggests that as a mechanism for improving
pro-poor capital investments, participatory budgeting has contributed
to a better access to water and sanitation in all PB municipalities.
It also led to a reduction of poverty rates in municipalities where
it was adopted prior to 1996. However, the results for poverty are
not conclusive for municipalities where PB has been adopted after
1996. It is worth noting that the poverty impact occurred despite
a reduction in GDP per capita in these municipalities, suggesting
that the PB can contribute to a redistributive impact in the long
run. Finally PB seems also to have positive impacts on reducing corruption.

\begin{table}[H]
\begin{centering}
\caption{Impacts using only municipalities that implemented PB before 1996\cite{WorldBank2008} }
\makebox[\textwidth][c]{%
\begin{tabular}{cccc}
\cmidrule{2-4} 
 & \multicolumn{3}{c}{Matching without votes for WP}\tabularnewline
\cmidrule{2-4} 
 & Poverty rates & Access to piped water & Sewerage\tabularnewline
\midrule
\midrule 
PB & -4,82{*}{*} & 0,06{*}{*}{*} & 0,29{*}{*}{*}\tabularnewline
GDP per capita & -0,85{*}{*}{*} & 0,00{*}{*} & 0,00\tabularnewline
R\texttwosuperior{} & 0,19 & 0,51 & 0,28\tabularnewline
\bottomrule
\end{tabular} %
\begin{tabular}{ccc}
\toprule 
\multicolumn{3}{c}{Matching with votes for WP}\tabularnewline
\midrule
Poverty rates & Access to piped water & Sewerage\tabularnewline
\midrule
\midrule 
-11,06{*}{*}{*} & 0,11{*}{*}{*} & 0,34{*}{*}{*}\tabularnewline
-2,93{*}{*}{*} & 0,02{*}{*}{*} & 0,00\tabularnewline
0,52 & 0,39 & 0,41\tabularnewline
\bottomrule
\end{tabular}}
\par\end{centering}

\begin{centering}
Note: {*} p<0,1 ; {*}{*} p<0,05 ; {*}{*}{*} p<0,01
\par\end{centering}

\centering{}The magnitudes of the coefficients are not meaningful
here. 
\end{table}


Nevertheless, findings of the World Bank do not indicate consistent
positive impacts of PB in general on fiscal performance, contrarily
to what was observed in Porto Alegre, where a special campaign happened.

Reinforcing these results, another evaluation\cite{Zamboni2007} concludes
that the PBs in Brazil have improved the spatial distribution of public
goods.
\end{doublespace}

\begin{doublespace}

\section{Conditions of success}
\end{doublespace}

\begin{doublespace}
According to researchers\cite{WamplerEtAl2007}, there are four keys
for a PB to be successful: a strong mayoral support, an active civil
society, self-ruling given to citizens for the mechanisms of the PB,
and consequent financial resources to fund the projects selected by
citizens. Recent research shows in particular that PB often miss to
attire a representative population, whether in\textbf{ }Brazil\cite{Marquetti2005},
in China\cite{He2011} or in Europe\cite{GanuzaEtAl2011}\textbf{.
}For example in the Spanish PBs Talpin analyzed\cite{Talpin2011},
a high presence of politicized participants, ideologically akin to
the government parties and belonging to middle and upper classes,
impeded the PB from embracing marginalized people. This may be due
to a traditional lack of concern from these people, as well as to
the \textemdash{} probably justified \textemdash{} sentiment that
only minor subjects would be discussed. The poor attendance in European
PB due to a lack of stake, as almost all citizens have already to
basic needs, witnesses a more general fact: participation occurs only
if a substantial power of decision is devolved to the citizens through
the PB. For example, a number of municipal governments around Porto
Alegre, as of the cities of Viamão and Alvorada, have elected WP governments
based on the well-publicized success of the Porto Alegre experiment.
However, with per-capita revenues at a fraction of Porto Alegre levels
(and consequently low budgets), participatory budgeting in those cities
has not succeeded in drawing sustained attendances\cite{WamplerEtAl2007}.
Now we will focus on the two other factors: mayoral support and true
empowerment.
\end{doublespace}

\begin{doublespace}

\subsection{A real empowerment of citizens}
\end{doublespace}

\begin{doublespace}
Baiocchi \& Ganuza\cite{BaiocchiEtAl2012} explain how Porto Alegre
included an administrative reform to both subordinate the local bureaucracy
to citizen demands and to protect the \textquotedblleft{}chain of
popular sovereignty\textquotedblright{} from outside influence. This
latter step involved the creation of a cabinet-level special department
above fiefdom-like municipal departments, as well as a decision to
make the PB the only point of contact between citizenry and the local
state. It also included the creation of a citizen council to oversee
the whole process (giving citizens the right to self regulate) as
well as to oversee municipal finances as a whole. These administrative
reforms made PB participation come closer and closer to effective
control of the local state, to the point that a recurrent problem
with those early PB experiments was that elected city councilors often
opposed the process for feeling sidelined. As Baiocchi \& Ganuza say:
\end{doublespace}
\begin{quotation}
\begin{doublespace}
\textit{Without a concomitant transformation of the administrative
machinery, PB becomes reduced to a set of procedures detached from
the inner workings of the administration or its central decision making
points.}\end{doublespace}

\end{quotation}


\begin{doublespace}
There were four guiding principles to the PB processes: openness (meetings
open to all, regardless of belonging to an association or not), transparency
(all proceedings were public), self-rule (participants themselves
decided on the parameters of the process), and actual decision-making
(processes were directly tied to administrative outcomes). That said,
reduction of Participatory Budgeting to an isolated technical device
takes out the crucial emancipatory dimension that allows citizens
to have their say on decisions that matter the most and leads to focusing
on small budgets. For PB to really have an impact on the civic life,
it must allow citizens to state the rule of the process and to modify
the institutions, which is possible only in a country that allows
for decentralization, like Brazil.
\end{doublespace}

\begin{doublespace}

\subsection{The support of the city council}
\end{doublespace}

\begin{doublespace}
Abers\cite{Abers2000} argues that Porto Alegre's PB worked because
of the autonomy of the town in terms of resources and because of its
exceptional WP administration, which enacted a policy ought to include
the excluded in the decision process and to favor them in the spendings,
despite a usually conservative and corrupted country.\cite{Covin2000}

According to Langelier\cite{Langelier2011}, the incensed experience
Porto Alegre became perverted when a mayor from another party was
elected in 2004. The socialist Fogaça promised that he'd maintain
the PB, which in addition to a bad financial situation of city (not
related to the PB) helped him to win. The PB was effectively pursued,
but integrated to a new process, the Local Solidary Governance (LSG),
which aimed to stimulate individual capacities and to build ties between
the components of the community (such as enterprises, neighborhoods,
associations or public services). It seemed to have the same purpose
of civic empowerment as the PB, but turned out to shift the focus
from citizens to enterprises. Enterprises were invited to invest into
the LSG while the PB budget shrunk from 10\% to 4\% of the municipal
budget. This caused a dramatic fall in the share of executed investments
voted in PB, from 95\% in 2001 to 10\% in 2006\cite{WorldBank2008},
destroying the confidence in the PB and the civic willingness to participate.
In 2008, the majority of PB councilors thought that the main function
of the LSG was to weaken the PB. One of them gave an example: \textit{you
own an enterprise and discuss with the mayor a project of kindergarten.
The negotiation remains discrete: private partnerships do not need
a plenary meeting, nor to gather the citizens. Finally, the population
knows nothing.} The new mayor has ended a practice of participation,
for example he decided unilaterally to displace a popular quarter
of 10 000 families to a new place out of the center, far away from
the jobs, in order to give a good image of the city for the World
Cup. A blog animated by these inhabitants denounces a ``social cleaning''.
For a dozen of years, the PB of Porto Alegre allowed to go beyond
particular interests and class antagonisms. Now, it suffers from clientelist
practices. In 2008, a majority of the PB councilors has voted in favor
of the suppression of the obligation to discuss with the delegates
the modifications of the intern rules of the PB, limiting access to
information and power of delegates, while creating a concentration
of power in the councilors, who by the way renew less and less, enacting
the dispossession of the power from popular mass and favoring political
manipulations. This story proves that there can not be an efficient
PB without a sincere engagement of the city council.
\end{doublespace}

\begin{doublespace}

\section{Perspectives}
\end{doublespace}

\begin{doublespace}
A properly set PB seems a beneficial way of governing a town. This
is why PB is supported by diverse networks, such as Real Utopias or
the World Bank. The World Bank has found some features that would
ameliorate the PB processes\cite{WorldBank2008}:
\end{doublespace}
\begin{quotation}
\begin{doublespace}
\textit{Creating incentives and removing barriers to make the PB more
inclusive is a critical challenge. Measures to address these challenges
could include expanding the PB\textquoteright{}s focus to issues of
strategic importance for the city as a whole (including budgetary
oversight, municipal revenues and current expenditures) that affect
the space for PB investments and would attract interest from the middle
class and professional groups, such as entrepreneurs. Furthermore,
participation could be made more convenient to attract those who currently
do not participate, for instance, by making available on-line voting
on investment priorities or reimbursing the financial cost of participation
borne by the poorest groups (e.g., the cost of bus tickets for participants
below a certain income level). Improving the effectiveness of PB council
meetings by reforming meeting procedures would save participants\textquoteright{}
time. Finally, improving the training and education of participants
is essential for fully realizing the potential benefits of the PB
to enhance citizen understanding of the municipal budget proce}ss.\end{doublespace}

\end{quotation}


\begin{doublespace}
Finally, we could extend PB to a larger scale. One could imagine to
use PB in order to decide investment for a country, or a continent.
For example, we could implement a giant PB in the EU to allocate democratically
the resources given nowadays by bureaucratic processes. How would
500 millions of persons choose whether to finance a highway or to
isolate houses? We could allow people to allocate freely on the Internet
a given amount of money (the same for everyone) in the projects of
their choice (including their own projects). For example, one could
give all his amount, let's say 1000¤, to research on HIV, or give
500¤ to her startup's project and 500¤ to build a new stadium in her
town. Each project would have a price, and would be financed at that
price if this threshold has been reached by the votes. The overflow
would be given to almost-financed projects. In order to avoid misappropriation
and bad projects, one could ``oppose'' any projects, and if a projects
receives more ``oppositions'' than financiers, it would not been
funded. Assessment of projects, feedback and accounting would be openly
accessible on the website. If it seems too ambitious, why not start
at the regional scale?

\bibliographystyle{unsrtnat}
\bibliography{articles}
\end{doublespace}

\end{document}