ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL

Page created by Dana Wilson
 
CONTINUE READING
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
ENERGY
SAFETY
NETS
BRAZIL
CASE
STUDY
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
ACKNOWLEDGEMENTS

The Energy Safety Nets: Brazil Case Study was      The research team acknowledges the contri-
researched and written by partners at the Cen-     butions to this work of the following workshop
tre for Energy and Environmental Economics         attendees and key interviewees: COPPE/UFRJ
(Cenergia Lab), COPPE, Universidade Federal        (Carlos Henrique Duarte), Cenergia/UFRJ (Talita
do Rio de Janeiro (www.ppe.ufrj.br), with the      Borges Cruz, Paula Borges, Mariane Zotin, Gerd
help of Antonella Mazzone, from the Oxford         Abrantes Angelkorte, Leticia Magalar de Souza
University Centre for the Environment, Univer-     and Rafael Garaffa), Institute of Applied Eco-
sity of Oxford. The lead researcher from COPPE     nomic Research – IPEA (Elaine Christina Licio),
was Roberto Schaeffer (roberto@ppe.ufrj.br),       Brazilian Electricity Regulatory Agency – ANEEL
supported by a team that included Alexandre        (Mauricio Lopes Tavares), Electric Energy Re-
Salem Szklo, André Frossard Pereira de Lucena      search Center – CEPEL (Márcio Gianini), Energy
and Régis Rathmann.                                Research Company – EPE (Thiago Pastorelli Rod-
                                                   rigues), Ministry of Social Development – MDS
We acknowledge with gratitude the financial        (Denise Direito), Frente por uma Nova Política
support provided by the Wallace Global Fund.       Energética (Joilson Costa).

   This report is based on research jointly implemented by partners at the Cenergia Lab, the
   Overseas Development Institute (ODI) and Catholic Agency for Overseas Development (CA-
   FOD). The research in Brazil is part of a broader program of energy safety nets research also
   carried out in Ghana, India, Indonesia, Kenya and Mexico funded by Sustainable Energy for All
   (SEforALL) as part of its People-Centered Accelerator work program.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                               2
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
TABLE OF CONTENTS

ACKNOWLEDGEMENTS                                                                    2

LIST OF FIGURES                                                                     4

LIST OF TABLES                                                                      5

ABBREVIATIONS                                                                       6

MAP OF BRAZIL                                                                       8

EXECUTIVE SUMMARY                                                                   9

INTRODUCTION                                                                       11

   BACKGROUND                                                                      12

		    The Brazil case study                                                        12

ENERGY ACCESS IN BRAZIL – A CONTEXTUAL OVERVIEW                                    14

THE DESIGN, IMPLEMENTATION AND EVOLUTION OF BOLSA FAMÍLIA,
LUZ PARA TODOS AND TARIFA SOCIAL                                                   17

   BOLSA FAMÍLIA                                                                   18

		    Separate ministries provide separate safety nets (mid-1990s to 2003/4)       18

		Integration into Bolsa Família                                                   19

   LUZ PARA TODOS                                                                  22

		    Electricity access prior to Luz para Todos                                   22

		    Electricity sector reform and the launch of Luz para Todos                   23

		    The limitations of grid extension for providing electricity access           24

		    Shifting focus to prioritize those left furthest behind                      26

   TARIFA SOCIAL                                                                   26

		    The history of subsidized electricity consumption                            26

		    A volume-differentiated tariff that varies by region and beneficiary group   27

BOLSA FAMÍLIA, LUZ PARA TODOS AND TARIFA SOCIAL: IMPACTS AND INSIGHTS              29

   BOLSA FAMÍLIA AND THE IMPACTS OF THE LPG ALLOWANCE                              30

		Poverty reduction                                                                30

		    Challenges of subsidizing energy access within Bolsa Família                 30

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                  3
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
The value of Bolsa Família has not kept pace with the cost of LPG                                 31

		    Impact of the increase in LPG price on energy access                                              33

		    A lack of awareness of social protection programs by beneficiaries may hamper the programs’ impact 34

   THE IMPACTS OF LUZ PARA TODOS AND TARIFA SOCIAL ON PROMOTING ACCESS TO
   ELECTRICITY                                                                                          35

		Luz para Todos: delivery of near-universal electricity access                                         35

		Tarifa Social has supported consumption for the poorest, but not for all poor groups                  35

THE GENDERED IMPACTS OF ENERGY SAFETY NETS IN BRAZIL                                                    36

CONCLUSIONS                                                                                             37

   SUGGESTIONS FOR FURTHER RESEARCH                                                                     38

   POLICY RECOMMENDATIONS                                                                               39

REFERENCES                                                                                              40

   Endnotes                                                                                             43

COPYRIGHT AND DISCLAIMER                                                                                44

LIST OF FIGURES

Figure 1: Brazil: HDI values by region (2010)                                                           13

Figure 2: Brazil’s progress towards achieving SDG7                                                      16

Figure 3: Timeline: Introduction of Social Policies in Brazil, 1995-2011                                20

Figure 4: Population with access to electricity, 1960-2000                                              22

Figure 5: National Grid Coverage                                                                        24

Figure 6: Access to Electricity by State, 2010                                                          25

Figure 7: Evolution of poverty in Brazil                                                                30

Figure 8: Real price variation of LPG to final user by region in Brazil, 2007-2016                      32

Figure 9: Residential consumption of firewood and LPG in Brazil, 2008-2017                              33

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                        4
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
LIST OF TABLES

Table 1: Relevant information about Brazil                     15

Table 2: The composition of Bolsa Família                      19

Table 3: Tarifa Social for low-income households               27

Table 4: Tarifa Social for specific groups                     28

Table 5: Nominal price of LPG and value of Bolsa Família       31

Table 6: Regulated power consumption by end use sector (GWh)   35

ENERGY SAFETY NETS | BRAZIL CASE STUDY                              5
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
ABBREVIATIONS

ANEEL		         Agência Nacional de Energia Elétrica (Brazilian Electricity Regulatory Agency)

BF 		           Programa Bolsa Família (Family Allowance Program)

BPC 		 Benefício Assistencial ao Idoso e à Pessoa com Deficiência (Benefit for Elderly and
		Disabled People)

CAD 		          Cadastro Único para Programas Sociais (Single Registry for Social Programs)

CAFOD           Catholic Agency for Overseas Development

CCC		           Conta de Consumo Combustível (Account for Cross-Subsidization)

CCC-ISOL        Fossil Fuels Consumption Account of Isolated Systems

CDE 		          Conta de Desenvolvimento Energético (Account for Energy Development)

CEPEL		         Centro de Pesquisa em Energia Elétrica (Electric Energy Research Center)

COPPE/UFRJ Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia da
		         Universidade Federal do Rio de Janeiro

CRAS		          Social Assistance Office

EPE		           Empresa de Pesquisa Energética (Energy Research Company)

HDI		           Human Development Index

IBGE		 Instituto Brasileiro de Geografia e Estatística (Brazilian Institute of Geography and
		Statistics)

IEA		           International Energy Agency

IPEA		          Instituto de Pesquisa Econômica Aplicada (Institute of Applied Economic Research)

LPG		           Liquefied Petroleum Gas

LpT		           Programa Luz para Todos (Light for All Program)

MCTIC		         Ministério da Ciência, Tecnologia, Inovações e Comunicações (Ministry of Science,
		              Technology, Innovations and Communications)

MDS		           Ministério do Desenvolvimento Social (Ministry of Social Development)

MMA		           Ministério do Meio Ambiente (Ministry of Environment)

MME		           Ministério de Minas e Energia (Ministry of Mines and Energy)

NGO		           Non-governmental Organization

ONS		           Operador Nacional do Sistema Elétrico (Brazilian´s Energy System Operator)

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                              6
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
PRODEEM          Brazilian Program for Rural Electrification Using Photovoltaics

RF		             Person Responsible for a Family Unit within CAD

SIN		            Sistema Interligado Nacional de Energia Elétrica (National Interconnected System)

TS		             Programa Tarifa Social (Social Tariff Program)

VG		             Vale Gás (Gas Assistance Program)

A note on currency

The Brazilian real (Portuguese: real, pl. reais; sign: R$; code: BRL) is the official currency of Brazil. Cur-
rency conversions were carried out using year-averaged values available from the World Bank/IMF
(https://data.worldbank.org/indicator/PA.NUS.FCRF?locations=BR). If a specific year was available
from the context, this was used, otherwise the most recent data (i.e. 2018) were used.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                           7
ENERGY SAFETY NETS BRAZIL CASE STUDY - SEforALL
MAP OF BRAZIL

                         Roraima                         Amapá

                 Amazonas
                                                       Pará                    Maranhão
                                                                                                         Ceará
                                                                                                                    Rio Grande
                                                                                                                    do Norte

                                                                                                                     Paraíba
                                                                                          Piauí
                                                                                                                 Pernambuco
Acre
                                                                                                                     Alagoas
                                                                   Tocantins
                   Rondônia                                                                                        Sergipe

                                                                                              Bahia
                                         Mato Grosso

                                                                         Distrito Federal
                                                                 Goiás

                                                                               Minas Gerais
                                                                                                      Espírito
                                           Mato Grosso                                                Santo
                                           do Sul

                                                                 São Paulo                Rio de
                                                                                          Janeiro

                                                        Paraná

                                                              Santa
                                                              Catarina

                                               Rio Grande
                                               do Sul

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                           8
EXECUTIVE SUMMARY

This research aims to provide guidance for policy-        fordability for poor and vulnerable households. It
and decision-makers, by identifying measures in           has expanded rapidly and more than one in ten of
Brazil that have been successful in enabling very         all households currently benefit from Tarifa Social
poor people to access modern energy services              to some degree. The volume-differentiated tariff
and exploring the reasons for their success and           applies discounts to the energy bills of households
challenges encountered. Its findings are based on         enrolled in the Cadastro Único (CAD), the unified
quantitative and qualitative analyses of three pro-       gateway registry for social programs. An analysis of
grams that directly and indirectly impact the ac-         consumption levels by poor households suggests
cess and consumption of modern energy sources:            that Tarifa Social has largely insulated electricity
Luz para Todos (Light for All) and Tarifa Social (So-     consumption in low-consuming poor households
cial Tariff) for electricity, and Bolsa Família (Family   from the recent economic downturn. However,
Allowance) for LPG for cleaner cooking.                   the same analysis shows that the scheme would
                                                          need to be modified to adequately support poor
Luz para Todos was launched in 2003 to extend ac-         households to consume more than very basic lev-
cess to electricity connections in rural and remote       els of electricity.
areas of Brazil. Like Tarifa Social, it is supported
by cross-subsidies funded via levies on consumer          In 2003, the Brazilian Government launched the
electricity bills. Through successive iterations, the     Bolsa Família integrated cash transfer to bring
program has focused on connecting households              together support previously provided by 12 indi-
and communities using a prioritization scheme             vidual social assistance programs. This included
that attempts to ensure the most disadvantaged            support for LPG for cooking under the Vale Gás
communities are connected first. Although the             and Auxílio-Gás programs. Bolsa Família also
vast majority of the unconnected households are           uses the Cadastro Único to identify recipients
located in regions of the country scoring low on          and represents an important resource for families
the Human Development Index (HDI), the scheme             that has successfully lifted many millions out of
appears to have been broadly effective. Since its         poverty. However, the program is poorly set up
inception, it has invested more than USD 7.1 billion      to ensure access to modern cooking energy such
(BRL 26 billion, 2018 prices) to reach over 16 mil-       as LPG and data show a recent uptick in the num-
lion people through more than half a million proj-        ber of households reverting to using traditional
ects, putting Brazil on track to achieve universal        biomass for cooking. This is, in part, because the
access to electricity well before 2030.                   program’s energy component is non-earmarked,
                                                          which allows households to use the benefit to
Tarifa Social was launched in 2010 to complement          pay for other services, especially during times of
universal access to electricity connections. The          economic hardship. The scheme’s effectiveness
program was built on previous experience with             in supporting LPG consumption is also ham-
subsidizing electricity consumption to increase af-       pered by the cost of LPG rising much faster than

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                           9
the amount the cash transfer provides to poor         support low-income groups to access affordable,
households. In 2015, the cost of a 13kg cylinder      reliable, sustainable and modern energy services.
of LPG represented 58 percent of the monthly          These include adapting programs to reflect sub-
Bolsa Família transfer, but by 2019 this figure had   national variations, increasing awareness among
increased to 79 percent.                              eligible beneficiaries, and changes to the way that
                                                      support is targeted and delivered to enhance the
The final section of this report details opportuni-   programs’ ability to support energy access by
ties to improve the extent that current programs      poor and vulnerable households.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                      10
INTRODUCTION

ENERGY SAFETY NETS | BRAZIL CASE STUDY   11
BACKGROUND                                                Kenya and Mexico— seeks to answer four research
                                                          questions:
Social assistance measures can enable access to
affordable energy. The use of targeted support            • What policy measures have been used to en-
measures may face inclusivity challenges. Different         able very poor and marginalized people to ac-
mechanisms (e.g. cash transfers, vouchers and price         cess and use modern energy services?
controls) may yield different results, and they may re-   • How effective have these measures been in en-
quire levels of administrative and institutional capac-     abling the poorest social groups to access and
ity that are not found in every country. Country case       use modern energy services?
study research is meant to build an understanding         • What links have there been/are there between
of the degree to which different mechanisms have            these measures and wider/other social assis-
been successful in different contexts, and identifies       tance programs?
areas for further innovation, piloting or research.       • What changes could be made to enhance the
The overall objective of the research is to provide         effectiveness of existing policy measures in en-
guidance for policymakers and decision-makers, by           abling very poor people to access modern en-
identifying measures that have been successful in           ergy services?
enabling very poor people to access modern ener-
gy services and exploring both the reasons for their      Using these general research questions, the case
success and challenges encountered.                       study began by developing questions more directly
                                                          relevant to the context of access to modern energy
The Brazil case study                                     services and social assistance programs in Brazil. A
                                                          literature review was then carried out to identify the
The Brazil case study aims to identify programs and       main government policies and programs comprising
public policies that have been successful in enabling     Brazil's Energy Safety Nets (ESNs).
access to modern energy services in regions with
low HDI values (Figure 1). The programs analyzed          Two stakeholder workshops were also carried out.
are Luz para Todos, Tarifa Social and other govern-       The first reviewed the specific research questions
mental programs, such as the Bolsa Família, which         and identified data sources. The second validated
can indirectly impact the access and consumption of       the preliminary research findings, addressed remain-
modern energy sources.                                    ing information gaps, and agreed preliminary con-
                                                          clusions and policy recommendations. Twenty key
This research used quantitative and qualitative           informant interviews based on the research ques-
methodologies to build a comprehensive picture of         tions were also carried out with stakeholders directly
different experiences of the use of social assistance     involved in the design, implementation, or evalua-
mechanisms for energy access. The case study—             tion of the government programs and/or research
like the other five, covering Ghana, India, Indonesia,    related to them.

   Energy Safety Net (ESN) is an umbrella term            ESNs can make physical access (i.e. connec-
   for government-led approaches to support               tions) to electricity or clean fuels affordable for
   very poor and vulnerable people to access              poor and vulnerable people, or they can make
   essential modern energy services, defined as           the unit price of electricity or fuel affordable
   electricity and clean fuels and technologies           to consume. ESNs include some form of tar-
   for cooking, by closing the affordability gap          geting or eligibility criteria to direct benefits
   between market prices and what poor cus-               to those who need them.
   tomers can afford to pay.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                             12
Figure 1
Brazil: HDI values by region (2010)

                              Roraima                           Amapá

                      Amazonas
                                                              Pará                     Maranhão
                                                                                                                Ceará
                                                                                                                            Rio Grande
                                                                                                                            do Norte

                                                                                                                             Paraíba
                                                                                                  Piauí
                                                                                                                         Pernambuco
 Acre
                                                                                                                             Alagoas
                                                                           Tocantins
                        Rondônia                                                                                           Sergipe

                                                                                                      Bahia
                                                Mato Grosso

                                                                                 Distrito Federal
                                                                        Goiás

                                                                                       Minas Gerais

                                                  Mato Grosso                                                 Espírito
        0.77 – 0.83                               do Sul                                                      Santo

        0.74 – 0.76                                                      São Paulo
                                                                                                      Rio de
                                                                                                      Janeiro

        0.72 – 0.74                                            Paraná

        0.69 – 0.71                                                  Santa
                                                                     Catarina

        0.61 – 0.68
                                                       Rio Grande
                                                       do Sul

Source: Authors' elaboration based on UNDP Atlas 2012 and IBGE 2019.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                               13
ENERGY ACCESS
IN BRAZIL –
A CONTEXTUAL
OVERVIEW

ENERGY SAFETY NETS | BRAZIL CASE STUDY   14
Brazil is an urban-industrial country with an ag-          energy’ (Pachauri et al. 2011:36). In this study,
ricultural sector that plays a leading role in the         energy poverty is interpreted as the inability of
global economy. One of the country’s most dis-             individuals or groups to consume energy services
tinctive characteristics is its energy mix, with a         that are affordable and reliable (UNDP 2000; IE/
significant share of its renewable sources, partic-        UFRJ 2005), yet it should be noted that it can
ularly in the generation of electric energy, afford-       be difficult to separate the definition of energy
ing the sector a low-carbon profile.                       poverty from the general concept of poverty.

Poverty is often the result of a complex economic          Despite the wider definition of poverty concep-
and sociocultural system that favors inequality,           tualized in the Brazilian Constitution, the Instituto
and energy access can play an important role in            Brasileiro de Geografia e Estatística (IBGE), which
eliminating poverty, isolation and deprivation.            carries out the national census, measures poverty
Current definitions of energy poverty focus                according to the following values: “extreme pov-
on a lack of modern energy services and the                erty” indicates an individual earning below USD
reliance on what is available in nature (IEA 2016).        0.75 per day (BRL 89 per month); while “poverty”
However, such definitions often do not specify             is defined as individual income between BRL 89
the reasons for the ‘lack’ of modern energy                and BRL 179 per month (USD 1.5 per day) (IBGE
services, which can be linked to a family’s financial      2015). In 2017, the numbers of people living in
constraints and the unaffordability of energy              poverty and extreme poverty were 23 million
services, whether energy poverty is the result             and 12 million, respectively (IBGE 2019). In Bra-
of geographical isolation (Bouzarovski 2014),              zil, energy access is commonly reported using
or a ‘complex combination of factors, including            the metric of the percentage of households with
lack of physical availability of certain energy            access to electricity and clean cooking fuels–100
types and high costs associated with using                 percent and 96 percent, respectively, in 2017–see

Table 1
Relevant information about Brazil

                                     Total area of 8,515,759,090 km2; divided in five political-administrative
   Territory                         regions – North, Northeast, Midwest, South and Southeast; composed of
                                     26 states and the Federal District

   Population (2018)                 209.5 million

   Gross Domestic Product – GDP
                                     1,689
   (in US$ billions/2018)

   GNI per capita
                                     9,140
   (in US$/2018)

                                     In 2017, renewables accounted for 43.2% of the Brazilian energy mix,
                                     a significantly higher share than world (13.8%) and OECD (10%) averages.
   Energy Mix                        Biofuels and sugarcane biomass supplied 40.3% of total energy.
                                     Renewables supplied 80.4% of electric energy. Hydraulic sources generated
                                     59.4% of total electricity, natural gas 10.5% and wind 6.8%

Source: IBGE 2018; MME 2018; World Bank 2019

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                             15
Figure 2 (IEA et al. 2019). It is not entirely clear      (Ibid 2018:45). It is worth mentioning that while
how well these figures reflect the actual num-            firewood is largely used for cooking (Coelho et
ber of households using electricity and modern            al. 2018), it also has other uses. In the south it is
cooking fuels. For example, one study noted that          mostly used for space heating (EPE 2013), while
in 2004 more than 98 percent of households had            in the warmer north, it is also used in livelihood
access to liquefied petroleum gas (LPG) (Jannuzzi         activities such as brick and pottery making (Maz-
et al. 2004), while another in 2015 showed that           zone 2019b).
54 percent of the population in the northern re-
gions was still exclusively using firewood versus         In this context, the next chapter aims to pres-
46 percent relying on LPG (Coelho et al. 2018).           ent the design, implementation and evolution of
Similar proportions were observed for the most            government programs such as Bolsa Família, Luz
developed regions such as the south and south-            para Todos and Tarifa Social, which can indirect-
east, in which firewood is still used by 42 percent       ly impact access to and consumption of modern
and 45 percent of the population, respectively            energy sources.

Figure 2
Brazil’s progress towards achieving SDG7

                  Brazil

         Region Latin America
                and the Caribbean

     Subregion South America                   100                        96                       46
 Income Group Upper middle income

GDP per capita 9,850
        (USD)                            Access to Electricity    Access to Clean Cooking    Renewable Energy
                                          (% of population)          (% of population)        (% of Total Final
                                                                                            Energy Consumption)
    Population 207,847,520

Source: IEA et al. 2019

                                                                                                                  16
THE DESIGN,
IMPLEMENTATION
AND EVOLUTION
OF BOLSA FAMÍLIA,
LUZ PARA TODOS
AND TARIFA SOCIAL

ENERGY SAFETY NETS | BRAZIL CASE STUDY   17
BOLSA FAMÍLIA                                            transformed into the Bolsa Escola; the Ministry of
                                                         Health launched Bolsa Alimentação; and the Min-
Separate ministries provide separate                     istry of Mines and Energy launched Auxílio-Gás in
safety nets (mid-1990s to 2003/4)                        2001 (more commonly known as Vale Gás after in-
                                                         corporation into Bolsa Família in 2003/2004) (Zim-
Modern social assistance programs were developed         merman 2006). Auxílio-Gás provided families with
in the mid-1990s with a series of policies emerging      BRL 7.50 (USD 1.99) each month to assist with the
to provide food security and eradicate child labor. In   purchase of a refill for a 13kg LPG cylinder.
1995, President Fernando Henrique Cardoso initiat-
ed the Social Safety Net Project (Rede de Proteção       The introduction of Auxílio-Gás took place in the
Social) aimed at redistributing wealth and protect-      context of the liberalization of the cost of oil prod-
ing the most vulnerable people in Brazilian society.     ucts in Brazil that started in the mid-1990s and
Attention to the poorest and the most marginalized       ended a policy of universally subsidizing LPG that
in Brazil increased in the mid-1990s because of the      had lasted from 1954 to 1995 (ANP 2017). This
social consequences of hyperinflation (1985-1994),       long-term policy of LPG subsidization was aimed
which increased the divide between rich and poor         at helping families buy and refill LPG cylinders
and the number of the extreme poor.                      and at encouraging a switch in cooking fuels from
                                                         wood to modern cooking gas. The liberalization
From 1995 to 2002, 12 programs were progressively        of LPG prices meant that policymakers needed to
included in the Social Safety Net Project, including     introduce a monthly allowance to support low-in-
Bolsa Escola (School Allowance), Vale Gás (VG) and       come families’ continued consumption of LPG
Bolsa Alimentação (Food Stamps). Individual minis-       (ANP 2017). According to Coelho et al. (2018),
tries were responsible for specific programs. For ex-    generous subsidisation programs aimed at re-
ample, the Ministry of Education implemented the         placing firewood with modern fuels such as LPG
Programa Agente Jovem (Youth Program) in 2001,           were made available all over the country to make
a cash transfer measure for education that was later     LPG affordable for poor people.

Box 1: Support for energy as support for nutrition

It can be argued that supporting a transition to         ongoing specific measures aimed at fighting ex-
modern energy sources was not the only rea-              treme poverty and hunger went hand-in-hand
son why the Brazilian Government started the             with LPG programs to guarantee food security
LPG-support programs. As Zimmerman (2006)                and nutrition for the poorest. The joint coordi-
argues that Brazil’s first social protection pro-        nation of social safety nets and energy subsi-
gram targeted food poverty and malnutrition.             dies for low-income families and marginalized
Cooking meals guarantees nutrient absorption             groups can be an effective way to target specific
and increases energy intake. For this reason,            aspects of poverty such as malnutrition.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                            18
Integration into Bolsa Família                             municipality received a fixed annual budget that
                                                           was based on its current number of low-income
A desire to address bureaucratic inefficiencies            households. This meant that if any additional
and weak communication between different min-              families were identified, conditions of vulnerabil-
istries appears to have driven the unification of          ity in the following year would not be included in
different social programs into the Bolsa Família           the program (Ibid. 149). For these reasons, the
program, beginning in 2003/2004, as shown in               Cadastro Único para Programas Sociais (CAD),
Table 3. For Zimmermann (2006:149), the lack of            a unified system that could more easily identify
effective coordination in implementing existing            families eligible for more than program was cre-
social policies resulted in leaving newly impov-           ated in 2003. This allowed the government to
erished people behind. As an example of these              create a unified social safety net program in the
systemic problems, Zimmermann notes that each              form of Bolsa Família.

Table 2
The composition of Bolsa Família (values updated to 2019)

        COMPOSITION OF
        BOLSA FAMÍLIA                  FOR WHOM?                           VALUE (IN BRL)

                                       For families living in
                                       extreme poverty
        Basic benefit                                                      89
                                       (income up to BRL 89
                                       per family per month)

                                       Conditional benefits (additional)

                                       For families with
                                       children and
        Children                                                           41
                                       adolescents
                                       (0-15 years old)

                                                                           41
        Pregnancy                      For pregnant women
                                                                           (x 9 months)

                                       For families with
                                                                           41
        Nutrition                      children
                                                                           (x 6 months)
                                       (0-6 months)

                                       For families in                     48
                                       extreme poverty                     (each family can
        Adolescents
                                       with adolescents                    accumulate up to two
                                       (16-17 years old)                   benefits = BRL 96)

                                                                           Additional support
        Benefit against                For families living                 on top of basic benefit
        extreme poverty                in extreme poverty.                 (amount varies depending
                                                                           on circumstances)

                                                   Observation

              Families living in extreme poverty can accumulate the basic benefit and all the
     conditional benefits (pregnancy, nutrition, adolescents) up to a maximum of BRL 372 per month.
             In addition, families can accumulate up to one benefit against extreme poverty.

Source: Authors' elaboration based on CAIXA 2019

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                           19
According to the official government website                            cartão to pay for designated goods and services.
CAIXA i, the main objectives of Bolsa Família are:                      In 2019, the value of the basic benefit of the allow-
                                                                        ance was raised to BRL 89 per month (USD 23.59).
1.   to fight hunger and promote food and nutri-                        Details of the basic benefit and the conditional
     tional security                                                    benefits, i.e. those granted in addition to the basic
2.   to fight poverty and other forms of deprivation                    benefit based on family composition, are shown
3.   to promote access to public services, espe-                        in Table 2.
     cially health, education and social assistance.
                                                                        In addition to Bolsa Família, the government also
Bolsa Família supports families with children liv-                      provides support to households through the Zero
ing below the poverty line to different degrees                         Hunger (Fome Zero), My Home My Life (Minha Casa
depending on their situation. Each family receives                      Minha Vida), and the Green Grant (Bolsa Verde)
a cash card (cartão) to which the benefit amount                        programs. For more details of the evolution of so-
is transferred monthly. Cardholders then use their                      cial assistance policies, see Figure 3.

Figure 3
Timeline: Introduction of Social Policies in Brazil (1995-2011)
                                                                                   Bolsa Verde

                                                                                   Ministry of Social Development
                                                                                   Credits to rural low-income families categorized
                                                                 2011
       Tarifa Social                                                               as traditional groups to conserve and develop
                                                                                   sustainable management of the land
       Ministry of Mines and Energy
       Social tariff for electricity, targeted at poor
                                                                 2010
       and vulnerable households
       9 million households benefit
                                                                 2009
                                                                                   Minha Casa Minha Vida

                                                                 2008              Ministry of Social Development
                                                                                   Safe shelter and housing (tax reduction and
                                                                                   credits for low-income families)
                                                                 2007

                                                                 2006
                                                                                  Bolsa Família

                                                                 2005             Ministry of Social Development
       Auxílio-Gás/Vale Gás                                                       Unified cash transfer for low-income families
                                                                                  (includes Auxílio-Gás/Vale Gás )
       Ministry of Mines and Energy                              2004             Benefited 36 million people between 2006 and 2015
       Low-income families receive BRL 15 every two months
       to support LPG consumption
                                                                 2003             Luz para Todos
       Programa Agente Jovem
                                                                                  Ministry of Mines and Energy
       Ministry of Health                                        2002             Provides electricity connections to unelectrified
       Cash transfer to young people between                                      households
       15 and 17 years old living in extreme poverty and                          3.4 million new connections provided by 2018
       social risk (BRL 65 per month) ()                         2001             Extended four times – now targeting universal
                                                                                  access by 2022
       Bolsa Escola
                                                                 2000
       Ministry of Education
       Schooling allowance (cash transfer) to low-income
       families with young children to boost school attendance   1999

       Bolsa Alimentação
                                                                 1998
       Ministry of Health
       Food stamp                                                                 Program Renda Mínima
                                                                 1997
                                                                                  Municipality distributed cash transfer
                                                                                  Targeted low-income families with young children

       Programa de Erradicação do Trabalho Infantil              1996

       Ministry of Health                                        1995
       Cash transfer to children aged 7 – 14 to prevent
                                                                                    FHC
       their employment
       25 BRL (rural)/40 BRL (urban)
                                                                                    LULA

Source: Authors' elaboration based on Zimmermann 2006 and MDS 2019

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                                20
Box 2: Cadastro Único (CAD), Single Registry for Social Programs

In 2002, a self-identification registry callled Ca-         called RANI in the case of indigenous peo-
dastro Único (CAD) was created to facilitate the            ple) and national insurance number (Carteira
distribution of social programs and to act as the           de Trabalho).
‘gateway for poor people to access a set of over 20     •   The condition for enrollment in the CAD is
policies’ (MDS 2015). This includes Bolsa Família           that the household’s income should not ex-
and Tarifa Social (as per Decree 9357/2018) as well         ceed three minimum wages. In the case of
as other sectoral programs such as the Water for            a single person household, the beneficiary
All Program. The CAD is ‘the main tool for the              should not earn more than one minimum
state to select and include low-income families             wage monthly. In 2019, the national minimum
in federal, state and municipal programs aimed              wage was equivalent to BRL 998 per month
at inclusive social participation in public policies’       (USD 246), but state governments can vary
(IBGE/PNAD 2014:22).                                        the level of the minimum wage depending
                                                            on local living standards.ii
The enrollment guidelines developed by the MDS          •   The beneficiary must notify the CAD of any
outline the application procedure as follows:               changes in the household’s situation (house-
                                                            hold income, change of address).
•   A member of the family (generally the house-
    hold head) is responsible for providing infor-      In a household survey released by IBGE in 2014,
    mation about the other family members. The          40.6 percent of households surveyed in Brazil
    MDS clearly states: ‘the person – named as          were not aware of the existence of CAD and
    Person Responsible for the Family Unit (RF) –       were not fully aware of all social assistance pro-
    must be at least 16 years old, and preferably       grams to which they may be entitled (IBGE 2016
    be a woman’ (MDS 2015).                             p.26). This finding was corroborated during
•   The person must visit a Social Assistance Of-       the first stakeholder workshop organized by
    fice (CRAS) to enroll. In many cases the com-       COPPE/UFRJ for this research project in June
    pilation of relevant information is carried out     2019. Several participants flagged how citizens,
    by social workers.                                  mainly in remote areas, are often unaware of the
•   The only documents requested for the enroll-        main information regarding social policies in the
    ment in CAD are the RF’s identify card (also        country.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                       21
LUZ PARA TODOS                                               i)   Initiatives led by foreign investments and pri-
                                                                     vate companies (1879-1930)
   Electricity access prior to Luz para                         ii) Creation of the state-owned company
   Todos                                                             Eletrobras, which held a monopoly over the
                                                                     sector and expansion of the hydroelectric
   Prior to the introduction of Luz para Todos in                    park (1930-1990)
   2003, household electricity coverage was 93 per-             iii) Privatization of the electricity sector and cre-
   centiii (urban and rural) with 97 percent access in               ation of the Electricity Regulator – ANEEL
   urban areas and 70 percent in rural areas (IBGE                   (1990-2003)
   2002). This result was realized by alternating               iv) Reform of the electricity sector, rural electri-
   stages of public and private investments that can                 fication and energy development that is now
   be divided into four main phases:                                 subsidized by the government (2003-today).

   Figure 4
   Population with access to electricity, 1960-2000

                                                                                 TOTAL         URBAN          RURAL

                        180
                                                                                                168.2

                        160

                                                                                  145.6
                                                                                                137.0
                        140

                                                                       117.3
                        120
POPULATION (MILLIONS)

                                                                                   110.1

                        100
                                                        89.9

                                                                       79.3
                        80
                                            69.2

                        60
                                     51.5               50.3

                                             38.6                       38.0
                        40                                                          35.5
                                     33.2
                                                         39.5                                    31.2

                                             30.5
                                     18.3
                        20

                         0
                              1940   1950   1960         1970          1980        1990        2000         2010

   Source: Authors’ elaboration based on ANEEL 2002 and IBGE 2000

  ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                22
The second and third phases were when the bulk         Electricity sector reform and the
of residential electricity access was achieved, es-    launch of Luz para Todos
pecially in urban areas where every urban house-
hold that requested a connection from the local        In 2002, a significant shift in the electricity sec-
energy company distributor was provided with           tor occurred when the rules for achieving uni-
one. However, in rural areas, consumer demand for      versal electricity access became stricter. Law
new electricity connections was largely ignored,       Nº10,438/2002 required electricity distribution
resulting in the development of private initiatives.   companies to respond to new requests for res-
Rural households organized themselves into en-         idential electricity connection within three days
ergy cooperatives that would carry the costs of        in urban areas and five days in rural areas and
rural electrification. In return for covering these    prevented them from charging the final consum-
costs, rural consumers received large discounts        er for the costs (ANEEL 2016). The 2002 law also
on their electricity bills from the local energy en-   required energy distribution companies to main-
terprise responsible for that area. According to       tain all new electric connections free of charge to
one of the energy stakeholders interviewed for         the final user. Funding comes from various sourc-
this report: "rural electrification from 1930s to      es but flows through the Conta de Desenvolvi-
1990s was largely driven by private initiatives and    mento Energético or CDE (Account for Energy
the creation of energy cooperatives" (AMX 2019).       Development) (see Box 3).
From 1990 to 2002, rural electrification became
a concern for the government and several initia-       Building on reforms, Brazil launched Luz para To-
tives, including Luz no Campo (Light in the Coun-      dos in 2003, with the aim of providing access to
tryside) and PRODEEM (the Brazilian Program for        electricity in still-unserved rural areas. According
Rural Electrification Using Photovoltaics – 1996-      to the IBGE, more than 2 million rural households,
2003),iv set out to provide rural households with      or over 12 million Brazilians, lacked access to elec-
electricity access in the form of renewable elec-      tricity in 2002. Further aggravating their exclusion,
tricity. These programs were largely funded by         approximately 90 percent of these households
foreign enterprises and development finance            were estimated to be living in areas with a low HDI
institutions.                                          and on less than three times the minimum wagev

Box 3: Conta de Desenvolvimento Energético (CDE), Account for
Energy Development

Conta de Desenvolvimento Energético (CDE),             provided through Tarifa Social (affordability)
Account for Energy Development was established         and supports Conta de Consumo Combustivel
in 2002 under Law Nº10,438/2002 and is regulated       (CCC), the cross-subsidization account. The
by the ANEEL. The CDE covers the costs of              CDE is financed through levies added to
universal energy access, funds the subsidies           consumer electricity bills.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                         23
2

                                                           i.e. the level at which 2households were eligible                                                                                                                                                                         ued expansion of the hydropower park and ex-
                                                          2
                                                           for support from Bolsa
                                                                                2
                                                                                     Família2 and
                                                                                              2
                                                                                                  other social                                                                                                                                                                       tension of the national grid, starting with urban,
                                                           safety nets.
                                                                      2
                                                                            2
                                                                                  2      3
                                                                                                                                                                                                                                                                                     densely populated areas and then moving to ru-
                                                                                                                                                                                                                                     2
                                                                  2       2                                                           5
                                                                                                                                                   2        2                                                                                                                        ral and
                                                                                                                                                                                                                                                                                           2      isolated areas. The reliance on hydropow-
                                                                                                                                                                                                                                     2                                2            2            3
                                                                                          4
                                          2    Initially, the Luz para Todos program             2        was to 2end                                                                                                                                                                er for3 electricity generation entails a complex
                                                                                                2          2
                                               in 2008, but it was subsequently    3    3           extended sev-     2                                                                                                                                                              organization
                                                                                                                                                                                                                                                                                             3
                                                                                                                                                                                                                                                                                                           of
                                                                                                                                                                                                                                                                                                           2    resources and interconnections
                                                                                                                                                                                                                                                                                                          2 2
                                               eral times, most recently to 20222 (under Decree
                                                  2                         2                                                                                                                                                                                                        between the reservoirs,   3        in addition to distribu-
                                                               2                 2   2      2                                                                                                                                                                                                               2 3
                                                             2                          2              2          2                                                                                                                                                                                     3
                                              29357 of 2018). The       2 initial target of connecting 12                                                                                                                                                                            tion to the    4 final2 consumer.
                                                                                                                                                                                                                                                                                              3
                                          2
                     2                                                                                                                                                                                                                                                            2
            2
                     2 2
                                      2
                                               million
                                                2        rural
                                                         2
                                                           3
                                                                  2
                                                                people    2 by 2008
                                                                               2    3     has been
                                                                                                  3
                                                                                                        2 revised
                                                                                                             2          to                                                                                                                                                               2     E 2 33
                                  2                                                                                                                                                                                                                                                               2
2                                           52 22               2     5                                                                                                                                                                                                                                                                       2
                                        2         2 2                       2 2                      2                                                                                                                                                                                                     2
                                  4
                                               that of achieving universal access     2       2
                                                                                                 3  by 202232
                                                                                                                 (MME
                                                                                                                    2     2
                                                                                                                                                                                                                                                                                              In2 20173 2
                                                                                                                                                                                                                                                                                                             the national interconnected grid covered
                                                           4                  2     4
                                            2                  2 2                         2
                                               2015; 2017;22019).
                                                    2
                                                                2   2   2
                                                                                 2
                                                                                         2           2
                                                                                                         3     2                                                                                                                                                                              141,388
                                                                                                                                                                                                                                                                                                   3       2 km across the south and east of the coun-
                                                                              3                                                                    3                                                                                                                                                    3
                                                                                          3                                                                 23                          2                                                    2
                                                                  2               2
                                                                                          2               2                                                                                                  3                                        2
                                                                                                                                                                                                                                                                                              try 3(see Figure
                                                                                                                                                                                                                                                                                                  3                2
                                                                                                                                                                                                                                                                                                                          5) but remained largely absent
                                                                                                                                                                                                                                                     2 2                                                         2 2
                                                              The2 2limitations
                                                                     2   2
                                                                               2 of2 grid
                                                                                       2    2
                                                                                              extension
                                                                                                 4
                                                                                                          2
                                                                                                           for2 33                                                              2
                                                                                                                                                                                                                                                                                              in the state2 of
                                                                                                                                                                                                                                                                                                                     3
                                                                                                                                                                                                                                                                                                                      3
                                                                                                                                                                                                                                                                                                                         Amazonas and in parts of Acre,
                                                                             2                     2
                                                              providing33electricity       access
                                                                                     2               3   2  3                                                                                                                                                                                      32          3
                                                                       2
        2                                                                    2                 2
                                                                                                       4        2
                                                                                                                  2                                                                                                                                                                           Mato Grosso,4        2    Roraima, Pará and Amapá (the Le-
                                                                                                                                                                                                                 E                                                                                  E
    2                                                                                    2 3   3        2     3                                                                                                                                                                                            2     3
                 2                                                                                                                             2                                 2                2                                          3                                                 2              3
                                                                                                                          2                                                                      2
                      2
                          2                           2                                       2
                                                                                                                                                                                         22
                                                                                                                                                                                                                             22
                                                                                                                                                                                                                                                                                  2           gal2 Amazon).            The grid is unlikely to be expanded
                                  2                                                                   3
                                                                                                                                                                        2 2
                                                                                                                                                                          3                                          2                                                                                2 2
                                                                                                                                                             3
                      4
                                          2
                                                        The preferred strategy 2for increasing
                                                                                          D 2 2 electricity                                                                                                                                                                                   into2 the Amazon forest for a number of reasons
                              2                                      2
                                                         2                                        2     2                                                                                                                                                                                             2
                          2           2                 access in Brazil
                                                                   2     2 has been characterized
                                                                                           6
                                                                                                3 by2 contin-
                                                                                                   2      2                                                                                                                                                                                   (ANEEL
                                                                                                                                                                                                                                                                                                  3           2010). These include the geomorpholo-
                                                                                          C
                                                                       2                2       2
            2             2                                  2    2                                2
    2                                                 2    2
                                                                                A         2
                                                                                              3
                                                                                                                                                                                                                                                                                          2
                                                                                                                                                                                                                                                                                                  3

    2                                         2                       2                               4           2                                         8               4               2
                                                                                                                                                                                                     2                                   2                                                    2
                                                                                                                                           B2
        2                                                                 2                                                                                                                                                                                            3
                          3           2                       Figure
                                                                 2 3 5
                                                                     3
                                                                     2
                                                                                                          2
                                                                                                                                                        3
                                                                                                                                                                    2
                                                                                                                                                                             7
                                                                                                                                                                                                             4
                                                                                                                                                                                                                                 3
                                                                                                                                                                                                                                             2
                                                                  2                                                                    2
                                                                              4                           2                                                                     23               2                                                                                        2
                                                              National
                                                                  3
                                                                    2 2 Grid Coverage
                                                                              3 2   3
                                                                                         3
                                                                                         2 2
                                                                                                                                                                                  2                                  2

                                                                2       D 2     2
                                                                                  2
                                                                                      2
                                                                                       2     D                                                                                           2                                           2
                                                                                                      6                           2        2                2           26                               2                                   2
                                                                                                                                           2                                                                         22
                                                                                                      C                   2
                                                                                                                                  2                                     C                            2
                                                                                                                                                                                                         2
                                                              A                                   2                               A                                     2           2                                                                                       2
                                                                                                              2                             28                                                                                                                                                                    2
                                                                              8                           2                                                                     2
                                                                                                                                       2
                                                                                                                                         2 2                                                                                 2                                                                                                                    2
                                                                                                                                                                                                                                                                                                                                                          2

                                                          B                                                                   B
                                                                                                                                                                    3                                                        2                        3                                               2
                                                                                              7
                                                                                              2                       2               5 3                           7                                                3
                                                                                                                                                                                                                                                                                  2
                                                                                                                       4                    2                                                    4                           2                                    2                                       2                               3                                       2
                                                                                                                                                                                                                                 2           2                                                    5
                                                                                                                                                                                                                                                                                                                          2
                                                                                                  3                                                 2               3
                                                                                                                                                                                                                                                                                                                  2                                                                                                           2
                                                  4                           3                                       4                            3                                                                                                                                                                                                                          2                       2   2                           3
                                                                                                                                                                                                                                                                                                                                          4
                                                  3                                               2                   3                                                 2                                                                                                                                                                             2
                                                                                                                                                                                                                                                                                                                                                                                              2                                   3
                                                                                                                                                                                                                                                                                                                                                  2                       2
                                                          2                                                                   2                                                                                                                                                                                               3       3                                                           2
                                                                          2                                                                    2                                                                                                                                                                                                                                                                                  3                 2
                                                                                                                                                                                                                                                                                                                                                                                                                                                   2 2
                                                          2                               2                                   2                                 2                                                                                                                                                                                         2                                                                                               3
                                                                                                                                                                                                                                                                                                                      2           2           2                                                                                                        2 3
                                                                                                                                                                                                                                                                                                                                                                  2                       2                                       3            3
                                                                                                                                                                                                                                                                                                                                                                                                                                          4            2
                                                                                                                                                                                             2                                                                                                                                                                                                            2
                                                                                                                                                                    2                2
                                                                                                                                                                                                     2
                                                                                                                                                                                                                                                                                                                                                          3
                                                                                                                                                                                                                                                                                                                                                                  2
                                                                                                                                                                                                                                                                                                                                                                                                                          2           E    2
                                                                                                                                                                                                                                                                                                                                                                                                                                               3
                                                                                                                                                                                                                                                                                                                                                                                                                                                   3
                                                                                                                                                                                                         2                                                                                                                                                                                                        2
                              2                                   2                               2                                    2                                                                         2
                                                                                                                                                                                                                                 2                                     2
                                                                                                                                                                                                                                                                                                                                                      3                                                                               2   2
                                                      2                                                                   2                                                                              4                                                                                                                                                                                                                        2
                                                              2                                                                   2                                                                                      2                                                    2
                                                                                                                                                                                                                                         2                                                                                                                                                                                            2
                 2                    2           5                                   2                       2       5                                                                                                                                                   2       2
                                                                                                                                                                                                                                                                                                                                                                                                                                  3
                                                                                                                                                                                                                                                                                                                                                  2
                                                                                                                                                                                                                                                         2             2
                                                                      2
                                                                                          Existing                        Planned
                                                                                                                               2
                                                                                                                                                                        Complex                                                  2               2                                                                                                                                                                        3

                                                                      138 kV                                                                                                                                                                     2                                        2                                                           4
                                                                                                                                                                        PARANÁ                                                                       2                                                                                                                                                                    2
                                                                      230 kV                                                                                                                                                                                            3         2
                                                                                                                                                                                                                                                                                                                          2       3
                                                                                                                                                                                                                                                                                                              2                               2
                                                                      345 kV                                                                                            PARANAPANEMA                                                                                                                                                                      2                                                           2
                                                                                                                                                                                                                                                                                                                                      2 2                                     2
                                                                      440 kV                                                                                                                                                                                                                                                  3
                                                                                                                                                                        GRANDE                                                                                                                                        2                           D           2                               2
                                                                      500 kV                                                                                                                                                                                                                                                                      6                       2                       2
                                                                                                                                                                                                                                                                                                                                                                              2
                                                                      750 kV                                                                                            PARANAÍBA                                                                                                                                                                 C                   2
                                                                                                                                                                                                                                                                                                                                                                          2
                                                                                                                                                                                                                                                                                                          A                                       2                                                           2
                                                                                                                                                                                                                                                                                                                          8                               2
                                                                                                                                                                        PAULO AFONSO                                                                                                                                                                                              2
                                                                                                                                                                                                                                                                                                      B                                                                       3                       3
                                                                                                                                                                                                                                                                                                                                          7
                                                                                                                                                                                                                                                                                                                                                                  4                   2

                Existing                  Planned          2 Number
                                                       Existing
                                                           Complex
                                                                 Planned            Complex
                                                                     of existing circuits                                                                                                                                                                                                     4
                                                                                                                                                                                                                                                                                              3
                                                                                                                                                                                                                                                                                                                          3
                                                                                                                                                                                                                                                                                                                                              3
                                                                                                                                                                                                                                                                                                                                                  2
8 kV                                            138 kV                                                                                                                                                                                                                                                2
                                                           PARANÁ                   PARANÁ                                                                                                                                                                                                                            2
0 kV                                            230 kV                                                                                                                                                                                                                                                2                               2

5 kV                                                          345 kV                      PARANAPANEMA                                                          PARANAPANEMA
0 kV                                                          440 kV                                                                                                                                                                                                        2                                 2
                                                                                          GRANDE                                                                GRANDE                                                                                                                            2
                                                                                                                                                                                                                                                                                                          2
0 kV                                                          500 kV                                                                                                                                                                                          2                       2       5

0 kV                                                          750 kV                      PARANAÍBA                                                             PARANAÍBA                                                                                                                                         2

                                                                                          PAULO AFONSO                                                          PAULO AFONSO

            2                                2 Number of existing circuits
                      Number of existing circuits

                                                              Source: ONS 2017                                                                                                                                                       138 kV
                                                                                                                                                                                                                                                             Existing                 Planned                                         Complex
                                                                                                                                                                                                                                                                                                                                      PARANÁ
                                                                                                                                                                                                                                     230 kV
                                                                                                                                                                                                                                     345 kV                                                                                           PARANAPANEMA
                                                                                                                                                                                                                                     440 kV
                                                                                                                                                                                                                                                                                                                                      GRANDE
                                                                                                                                                                                                                                     500 kV
                                                                                                                                                                                                                                     750 kV                                                                                           PARANAÍBA

                                                              ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                                                                                                                                                                  PAULO AFONSO                                                                                                            24
                                                                                                                                                                                                                                                         2            Number of existing circuits
gy of the Amazon forest with its thick vegetation;              eas with the lowest electrification rates and low-
low-density population and the low return over                  est HDI values in Brazil.
investment on energy development projects in
this area. Moreover, the Amazon forest hosts one                The Ministry of Mines and Energy (MME) iden-
of the most ethno-culturally diverse populations                tified specific contexts in which access to elec-
living in natural parks and sustainable reserves                tricity was particularly challenging, such as vul-
where development plans are not encouraged                      nerable indigenous or traditional communities
by conservation plans.                                          living in remote and sensitive ecosystems (nat-
                                                                ural parks, extractive reserves and indigenous
The area outside national grid coverage is called               lands). In these areas, the MME ascertains that,
the isolated system. It is composed of medi-                    as the national grid can not be expanded, local
um-size thermopower plants supplying energy to                  populations should be served by renewable and
the consumers of the northern states (Amazonas,                 decentralized systems (ANEEL 2017). Among
Pará, Rondonia, Roraima, Amapá and Mato Gros-                   the available technological options, the MME
so as well as the island of Fernando de Noronha).               encourages mini- and micro-hydropower plants,
As Figure 6 shows, the territory covered by the                 thermopower plants fueled by biofuels or nat-
isolated system corresponds in large part to ar-                ural gas, wind power, and solar photovoltaics.

Figure 6
Access to Electricity by State, 2010

                                            Roraima                   Amapá

                                     Amazonas
                                                                    Pará                     Maranhão
                                                                                                                      Ceará
                                                                                                                                     Rio Grande
                                                                                                                                     do Norte

                                                                                                                                         Paraíba
                                                                                                        Piauí
                                                                                                                                     Pernambuco
                      Acre

                                                                                 Tocantins                                         Alagoas
                                      Rondônia
                                                                                                                               Sergipe
                                                                                                            Bahia
                                                      Mato Grosso

                                                                                       Distrito Federal
                                                                              Goiás

                                                                                             Minas Gerais

                             98% – 98.99%               Mato Grosso
                                                        do Sul                                                      Espírito
                                                                                                                    Santo

                             97% – 97.99%                                      São Paulo
                                                                                                            Rio de
                                                                                                            Janeiro
                             95% – 96.99%                            Paraná

                             94% – 94.99%
                                                                           Santa
                                                                           Catarina
                             91% – 93.99%
                                                            Rio Grande
                             77% – 90.99%                   do Sul

Source: Authors' elaboration based on UNDP Atlas 2012 and IBGE 2010.

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                                             25
In addition, there are hybrid systems involving a      example, if the electricity distribution company in
combination of options including diesel/renew-         the state of Amazonas received two connection
able hybrids (ANEEL 2017).                             requests in the same period, one in a peri-urban
                                                       village and the other in a quilombola village, they
Shifting focus to prioritize those left                should connect the quilombola one first, even if
furthest behind                                        the peri-urban request (because of geographical
                                                       vicinity) were easier to connect in terms of dis-
The priorities of Luz para Todos have evolved over     tance and cost.
time. Initially the program prioritized low-HDI
communities located near the grid, whose con-          TARIFA SOCIAL
nection was at the lowest cost. As these near-grid
communities were connected, the focus shifted          The history of subsidized electricity
to more remote communities and then, in the fi-        consumption
nal phase, to families located in regions farthest
from the grid, especially in the Amazon. In recent     The Luz para Todos program aims to achieve
years, Luz para Todos has targeted low-income          universal electricity access, but once con-
families, ethnic minorities and vulnerable popu-       nected, how do low-income families pay for
lations living in protected areas such as national     their electricity services? Gomes and Silveira
parks, extractive reserves and conservation areas      (2012:160) suggest that the creation of social
(Decree Nº 9357/2018). These areas correspond          tariffs is ‘instrumental in the process of ensur-
to the regions with the lowest HDI in Brazil – see     ing the sustainability of the [universal electrici-
Figure 6.                                              ty access] program’. This led to the creation of
                                                       a social tariff to complement Luz para Todos
To accelerate the completion of Luz para Todos in      and ensure affordability of electricity by con-
rural areas, the MME issued a regulation in 2012       sumers. The Tarifa Social program was created
requiring energy distributors (currently still rely-   under Law Nº12,212/2010 and is the most im-
ing on the CDE to finance universal electricity ac-    portant subsidy for electricity consumption for
cess)vi to give priority to remote communities and     low-income people. While the tariff was only in-
‘traditional’ groups. The Ministry of Environment      troduced in 2010 (considerably after the launch
(MMA) categorizes these as specific groups of          of Luz para Todos), discounts on electricity
people who share distinctive low socioeconomic         consumption had been used since 1985 in the
status, geographies and sociocultural identities,      context of economic recession, inflation and
including: a) indigenous Indians; b) Seringueiros      high national debt. Prior to 2010, discounted
(rubber workers); c) Quilombolas (people of Af-        tariffs were not introduced explicitly as a tool
rican-Brazilian heritage); Castanheiros (nut col-      to reduce energy poverty. Due to economic re-
lectors); d) Quebradeiras de coco-de-babaçu            cession, the demand for electricity in the late
(babaçu coconut workers); e) Pescadores Arte-          1980s and 1990s decreased. This coincided
sanais (artisanal fishermen); f) Ribeirinhos (river-   with several hydropower plants (commissioned
ine people); g) Ciganos (Romani people); and h)        during the 1970s) coming onstream. As a result,
Caatingueiros (people living in the Caatinga bi-       the supply of electricity exceeded demand and
ome) among others (MMA 2017). Luz para Todos           consumers were encouraged to consume more
relies on these categories to prioritize future con-   electricity with the introduction of discounted
nections. The 2012 regulation stipulated that the      tariffs (Tavares 2004). Hence, discounted tar-
most vulnerable and marginalized groups should         iffs were originally a way to increase electricity
be given priority for electricity connection. For      consumption and avoid financial losses on in-

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                       26
vestment in the hydropower plants operating                permanent disabilities (of any age).vii Households
during the economic crisis.                                with a BPC beneficiary also receive discounts on
                                                           their total electricity bill.
A volume-differentiated tariff that
varies by region and beneficiary                           Electricity tariffs in Brazil also vary by region.
group                                                      The average tariff in 2018 was 0.568 BRL/kWh
                                                           (ANEEL 2019). This means that without the dis-
Tarifa Social has two components – one targeted            count, a family consuming 30kWh would pay BRL
at low-income households and the other at the              17.00 (USD 4.10) per month. With a discount of
most vulnerable marginalized social groups.                65 percent the family would pay BRL 5.95 (USD
                                                           1.44). In addition to the volume-differentiated
To be eligible for the low-income Tarifa Social,           discount, electricity is also cross-subsidized in
households must be enrolled in the CAD. The Tar-           some regions, especially in the northern regions
ifa Social is a volume-differentiated tariff. A lifeline   where the costs of generation and distribution
block mandates that if a household consumes 30             of energy are higher than in other parts of the
kWh per month or less, they will receive a 65 per-         country. Here, consumers of isolated systems
cent discount on their electricity bill. As a house-       receive electricity generated by decentralized
hold’s monthly electricity consumption increases,          thermal power plants that rely on fossil fuels. The
the percentage discount decreases (see Table 3).           lack of energy infrastructure together with low
                                                           population density and large distances between
The same tariff structure is also available to vul-        urban centers contribute to high prices of goods
nerable groups of people likely to be affected by          and services. Higher tariffs than in other regions
energy poverty, such as the elderly and disabled.          would be needed to cover the extra costs of gen-
The Benefício Assistencial ao Idoso e à Pessoa             eration compared to electricity generated by the
com Deficiência (BPC) or Benefit for Elderly and           national grid. Given that these northern areas
Disabled People, is a cash transfer program tar-           have some of the lowest HDI values in Brazil, fos-
geting older people (over 65) and people with              sil fuels for transportation, cooking and electricity

Table 3
Tarifa Social for low-income households

                  MONTHLY CONSUMPTION
                  QUOTA (MCQ)                                                  DISCOUNT (%)

                  MCQ
Table 4
Tarifa Social for indigenous groups

                  MONTHLY CONSUMPTION
                                                                        DISCOUNT (%)
                  QUOTA (MCQ)

                  MCQ
BOLSA FAMÍLIA,
LUZ PARA TODOS
AND TARIFA SOCIAL:
IMPACTS AND INSIGHTS

ENERGY SAFETY NETS | BRAZIL CASE STUDY   29
BOLSA FAMÍLIA AND THE IMPACTS                                                                  fers lifted 3.4 million people out of extreme pov-
  OF THE LPG ALLOWANCE                                                                           erty and 3.2 million out of poverty (IPEA 2019).
                                                                                                 Currently in Brazil there are 13.9 million people
  Poverty reduction                                                                              receiving the benefit (MDS 2019).

  The numbers of poor and extremely poor people                                                  Bolsa Família represents an important resource
  in Brazil totaled 59 million and 25 million, respec-                                           for low-income families. However, two issues—
  tively, in 2001 (IBGE 2019). While these numbers                                               diversion to other goods and services, and failure
  decreased slightly in the 2000s, it was only after                                             to keep pace with real cost increases in LPG—
  the launch of Bolsa Família in 2006 that the num-                                              raise serious questions over the continued effec-
  bers decreased significantly, to 14 million and 5                                              tiveness of the program in supporting energy ac-
  million respectively in 2014 (IBGE 2019). Between                                              cess for the most disadvantaged groups.
  2006 and 2015, Bolsa Família benefited 36 million
  people and increased living standards in Brazil                                                Challenges of subsidizing energy
  (IDB 2015:38). Its coverage among the poorest 20                                               access within Bolsa Família
  percent of the Brazilian population has increased
  over time, reaching 60 percent in recent years.                                                During the second stakeholder workshop, par-
  Its incidence coefficients—measuring the redis-                                                ticipants (energy stakeholders, academics and
  tributive impacts of the first real disbursed by                                               NGOs) were asked about the effectiveness of
  the program—have also become more negative,                                                    consolidating diverse safety nets under Bolsa
  which indicates the measure has become increas-                                                Família. The participants disagreed with the anal-
  ingly progressive. About 70 percent of the ben-                                                ysis in the literature discussed above that this in-
  efits of Bolsa Família have reached the country’s                                              tegration helped the government to have better
  poorest 20 percent. In 2017, Bolsa Família trans-                                              control over the transfer, thus minimizing the risk

  Figure 7
  Evolution of poverty in Brazil

                                                                                                             PEOPLE IN EXTREME POVERTY                 PEOPLE IN POVERTY

                 70

                                                                                                      62
                                          61                                                                60
                              59    59                                                     59   59
                 60    58                                                            58
                                                 56                          56                                   56
                                                                   53
                                                       52   52          52
                                                                                                                       49
                 50                                                                                                           47

                                                                                                                                    42
                                                                                                                                          40
MILLION PEOPLE

                 40                                                                                                                             37
                                                                                                                                                      34

                                                                                                                                                           30
                                          29                                                                                                                     29
                 30     28    28    28
                                                 26                                       26          26                                                               26
                                                                   24        24      25         24          24
                                                             23
                                                       22               22
                                                                                                                 21
                 20                                                                                                     17
                                                                                                                              17
                                                                                                                                    14    14    13    12
                                                                                                                                                           10    10
                 10                                                                                                                                                     8

                  0
                        90

                             91

                                   92

                                           93

                                                94

                                                      95

                                                              96

                                                                   97

                                                                        98

                                                                               99

                                                                                      0

                                                                                            1

                                                                                                  2

                                                                                                        3

                                                                                                             4

                                                                                                                  5

                                                                                                                         6

                                                                                                                               7

                                                                                                                                    8

                                                                                                                                           9

                                                                                                                                                 10

                                                                                                                                                          11

                                                                                                                                                         12

                                                                                                                                                                 13

                                                                                                                                                                        14
                                                                                           0

                                                                                                                              0
                                                                                                 0

                                                                                                       0

                                                                                                                  0

                                                                                                                                    0
                                                                                                                        0

                                                                                                                                          0
                                                                                                             0
                                                                                     0

                                                                                                                                                      20

                                                                                                                                                       20

                                                                                                                                                                20
                             19

                                                                                                                                                                      20
                                                                                          20

                                                                                                                                               20
                                                                   19
                                   19

                                         19

                                                      19

                                                                        19
                                                            19

                                                                             19
                                                19

                                                                                                                             20
                                                                                                20

                                                                                                      20

                                                                                                                 20

                                                                                                                                   20
                      19

                                                                                                                       20

                                                                                                                                         20
                                                                                                            20
                                                                                    20

  Source: Authors' elaboration based on IBGE 2019

 ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                                                                      30
of fraud, arguing that the consolidation neither              the market price of LPG and the fall (in real terms)
simplified the process nor reduced bureaucracy.               of the value of the Bolsa Família with inflation
                                                              averaging 6.5 percent in the 2000-2017 period
In addition, participants suggested the delivery              (IPEADATA 2019). This inability to keep pace with
of Bolsa Família limited its ability to promote               inflation may also have led to recent increases
energy access. Families receiving Bolsa Família               in the numbers of poor and extremely poor be-
simply receive a lump sum cash transfer each                  tween 2015 and 2017.
month with no limitations attached to how it may
be spent. No assessment was found as to wheth-                Table 5 shows the nominal annual price of LPG in
er the LPG portion of Bolsa Família contributed               comparison to the value of the Bolsa Família in Bra-
to household energy expenditure, however, it                  zilian reals, both of which have increased in recent
appears that ‘invisibility’ of the energy access              years. The figures are not directly comparable be-
subsidy within the non-earmarked Bolsa Família                cause Bolsa Família is paid monthly, while a cylin-
means families have the freedom to allocate their             der may last longer than one month. However, the
LPG allowance to other goods (mostly food and                 bottom row clearly illustrates that for a household
educational materials). As a result, the govern-              to consume the same amount of LPG, they had to
ment is considering a return to the old Vale Gás              use more of their Bolsa Família transfer in 2019 than
voucher program exclusively aimed at subsidiz-                in 2015, i.e., the cost of a 13kg cylinder represent-
ing the monthly purchase of an LPG cylinder.                  ed 58 percent of the monthly Bolsa Família in 2015,
                                                              while in 2019 it represented 79 percent of its value.
The value of Bolsa Família has not
kept pace with the cost of LPG                                Moreover, the liberalization of the LPG market ini-
                                                              tiated in the mid-90s leading to a complete with-
The workshop participants also highlighted that               drawal of subsidies by the end of 2000 (Coelho et
the amount of money transferred under the Bol-                al. 2018) also created regional disparities. For ex-
sa Família has become increasingly insufficient               ample, monthly reports released by the MME show
to cover the purchase of LPG. The impact varies               that the northern region has the highest prices for
between regions, due to disparities in delivery               LPG due to the distribution margin and transporta-
costs, and has been compounded by the rise in                 tion costs (MME 2015 N109, 2015: 9).

Table 5
Nominal price of LPG and value of Bolsa Família

                                          2015          2016           2017         2018          2019

         Value of monthly
                                           77            82             85           89            89
         Bolsa Família transfer (BRL)

         Cost of 13 kg LPG (BRL)           45            50             55           65            70

         Cost of LPG cylinder as
         share of value of monthly        58%           61%            65%           76%          79%
         Bolsa Família transfer

Source: Authors' elaboration based on ANP 2019 and Ministry of Social Development 2019

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                31
41.6                                                                                         41.6

                                                38.6                                                                                        38.6

   2010                                        38.2                                                      2010                               38.2

                                          36.2                                                                                            36.2

       Figure 8                                 38.5                                                                                        38.5

       Real price variation of LPG (to final user) by region in Brazil, 2007-2016
                                                      41.4                                                                                       41.4

                                                39.0                                                                                        39.0

   2011                                        38.5                                                      2011                               38.5
          CENTRE-WEST           SOUTH                   SOUTHEAST                          NORTHEAST      NORTHERN REGION
               CENTRE-WEST             SOUTH
                                         36.4                  SOUTHEAST                     NORTHEAST      NORTHERN
                                                                                                                CENTRE-WEST
                                                                                                                     REGION         SOUTH
                                                                                                                                      36.4                 SOUTHEAST                    NORTHEAST        NORTHERN
        PRICE VARIATION OF LPG (BRL/CYLINDER)
                                                 39.6                                                                                         39.6

       0% PRICE
             10%VARIATION
                     20% OF30%
                            LPG (BRL/CYLINDER)
                                   40%    50%                                   60%           70%               PRICE VARIATION OF LPG (BRL/CYLINDER)

          0%     10%      20%          30%              40%            50%            60%       70%             0%   10%     20%    30%             40%            50%            60%      70%
                                   35.3

                                          35.3         41.9                                                                            35.3        41.9
                                 33.6

                                        33.6      39.9                                                                               33.6        39.9
2007                            32.2

   2012
   2007                            32.2
                                32.7
                                                39.4                                                     2012
                                                                                                         2007                      32.2      39.4

                                   32.7 37.3                                                                                        32.7 37.3
                                  34.3

                                         34.3 40.4                                                                                    34.3 40.4

                                   35.0

                                         35.0           43.7                                                                          35.0          43.7
                                 33.8

                                        33.8          41.3                                                                           33.8         41.3
2008                            32.3

   2013
   2008                             32.3              41.1                                               2013
                                                                                                         2008                       32.3          41.1
                                 33.3

                                     33.3 39.4                                                                                       33.3 39.4
                                   34.8

                                         34.8           43.6                                                                          34.8          43.6

                                          38.9

                                                 38.9        45.6                                                                            38.9         45.6
                                       36.4

                                              36.4 42.4                                                                                    36.4 42.4
2009                               35.2

   2014
   2009                                35.2             43.1                                             2014
                                                                                                         2009                         35.2          43.1
                                   35.0

                                          35.0        42.0                                                                            35.0        42.0
                                       35.8

                                              35.8            46.5                                                                        35.8            46.5

                                                41.6

                                                      41.6            51.7                                                                         41.6           51.7
                                         38.6

                                                38.6           48.5                                                                         38.6            48.5
2010                                     38.2

   2015
   2010                                        38.2            47.5                                      2015
                                                                                                         2010                               38.2           47.5
                                       36.2

                                          36.2                47.0                                                                        36.2            47.0
                                         38.5

                                                38.5                   53.3                                                                 38.5                   53.3

                                               41.4

                                                      41.4                          60.5                                                         41.4                           60.5
                                         39.0

                                                39.0                         55.7                                                           39.0                         55.7
2011                                     38.5
   2016
   2011                                        38.5                   52.6                               2016
                                                                                                         2011                               38.5                  52.6
                                   36.4

                                          36.4                        52.2                                                             36.4                       52.2
                                          39.6

                                                 39.6                           59.5                                                          39.6                          59.5

          0%     10%      20%          30%              40%            50%            60%       70%             0%   10%     20%    30%             40%            50%            60%      70%
                                                41.9

                                                       41.9                                                                                        41.9
                                              39.9

                                                  39.9                                                                                           39.9
2012                                      39.4

   2012                                 37.3
                                                39.4                                                     2012                                39.4

                                    37.3                                                                                                   37.3
       Source: Authors'            40.4
                          elaboration based                     on ANP 2017
                                                     40.4                                                                                        40.4

                                                                                                                                                                                                    32
                                                 43.7
       ENERGY SAFETY NETS | BRAZIL CASE STUDY
                                                        43.7                                                                                        43.7
                                               41.3

                                                      41.3                                                                                        41.3
2013                                           41.1

   2013                                   39.4        41.1                                               2013                                     41.1
Impact of the increase in LPG price on                                           However, since 2013, levels of residential fire-
energy access                                                                    wood consumption have shown a slight increase
                                                                                 (to 20.4 kilotons in 2015) while consumption of
According to Gioda (2019:373) ”the use of fire-                                  LPG has remained constant (Figure 9). Although
wood by the low-income population seems to                                       their analyses use different methodologies,viii this
be directly associated with the price of LPG. In                                 increase was corroborated by the IBGE that cal-
general, the firewood used by this population                                    culated that there was an increase to 17.6 percent
is of ’picking’ quality (gathered from the land                                  in households using firewood as fuel for cooking
without cost) and the stoves are rustic”. Figure                                 in 2017 compared to 16.1 percent in 2016 (IBGE
9 presents data from the state-owned Energy                                      2019; Gioda 2019). More research is needed to
Research Company (EPE 2017) showing a de-                                        understand how the recent economic crisis com-
crease in the use of residential firewood con-                                   bined with the increase of LPG prices has impact-
sumption from 2008 to 2013 (from 24.9 to 18.5                                    ed fuel choices in low-income households.
kilotons) along with a corresponding increase
in LPG consumption. This supports the asser-                                     To minimize the impacts of the increase in LPG price
tion by Coelho et al. (2018) that the LPG sub-                                   on energy access, workshop participants also rec-
sidization policies contributed to a 22 percent                                  ommended decreasing fees and taxes on LPG and
reduction in households consuming firewood                                       providing subsidies for the purchase of solar stoves
between 2002 and 2012.                                                           in rural areas with poor LPG distribution networks.

Figure 9
Residential consumption of firewood and LPG in Brazil, 2008-2017

                                                                                                       FIREWOOD            LPG

                                   11                                                                                      26

                                        24.85
                                                                                                                   10.81   25
                                 10.8
                                                 24.28
                                                                                           10.69
                                                                                                                                 Annual firewood consumption (103t)
                                                                                 10.67                     10.75
Annual LPG consumption (103m3)

                                                                                                                           24
                                                         23.47                                     10.70
                                 10.6
                                                                                                                           23
                                                                 10.41
                                                                         10.46
                                 10.4                                                                                      22
                                                    10.30
                                                                         20.87
                                 10.2                                                                                      21
                                                                 20.98                             20.43
                                                                                                                   19.72   20
                                                                                         19.70
                                  10
                                                 10.00                                                     19.56           19
                                        9.89
                                  9.8
                                                                                 18.52                                     18

                                 9.6                                                                                       17
                                        2008    2009     2010     2011   2012     2013      2014    2015   2016    2017

Source: Authors' elaboration based on EPE 2017

ENERGY SAFETY NETS | BRAZIL CASE STUDY                                                                                                                                33
You can also read