SciELO - Scientific Electronic Library Online

vol.1 número3Os direitos sociais como compromissosMedidas de austeridade em julgamento: da compatibilidade das medidas de austeridade com a Convenção Europeia dos Direitos Humanos índice de autoresíndice de assuntosPesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados

  • Não possue artigos similaresSimilares em SciELO


e-Pública: Revista Eletrónica de Direito Público

versão On-line ISSN 2183-184X

e-Pública vol.1 no.3 Lisboa dez. 2014



The brazilian “second bill of rights”: a cause for celebration?∗ 

A segunda declaração de direitos brasileira: motivo para celebração?


Octavio Luiz Motta FerrazI

IKing's College London Capital House, first floor 42 Weston Street London SE1 3QD. e-mail:




The Brazilian Constitution of 1988 established an extensive catalogue of fundamental rights and set, as the fundamental principles and objectives of the new Brazilian republic, citizenship, the dignity of the human being, and the construction of a free, just and solidary society, in which poverty should be eradicated and inequalities reduced, and everyone's well-being should be promoted without any discrimination. But how wide is the gap still between the social and economic citizenship promises of the Citizen's Constitution and the real life of the Brazilian population? Despite significant problems of quality, efficiency and underfunding, the imperfect quantitative data indicates that the public services and benefits system in Brazil, a great part of which created by the Constitution, has improved, often significantly, the well-being of the Brazilian population in the past couple of decades.

Keywords : Brazilian Constitution of 1988; social and economic rights; constitutional effectiveness; sham constitutions; Economic and Social Rights Fulfillment Index (ESRF).



A Constituição brasileira de 1988 estabeleceu um extenso catálogo de direitos fundamentais e consagrou, como princípios fundamentais e objetivos da nova República Brasileira, a cidadania, a dignidade da pessoa humana e a construção de uma sociedade livre, justa e solidária, em que a pobreza deve ser eliminada, as desigualdades reduzidas e o bem-estar de todos deve ser promovido sem discriminação. Mas quão largo é o fosso entre as promessas de cidadania social e económica da Constituição Cidadã e a vida real da população brasileira? Apesar de problemas significativos de qualidade, eficiência e falta de investimento, os dados quantitativos, ainda que imperfeitos, indicam que os serviços públicos e o sistema de benefícios no Brasil, em larga medida criados pela Constituição, melhoraram, muitas vezes de modo significativo, o bem-estar da população brasileira nas últimas duas décadas.

Palavras-Chave : Constituição brasileira de 1988; direitos sociais e económicos; efetividade constitucional; sham constitutions; Índice de Satisfação de Direitos Sociais e Económicos.


Sumário: 1. Introdução; 2. O Papel do Direito; 3. Medindo a Efetividade Constitucional; 4. O Ranking das Sham Constitutions de Law e Versteeg; 5. O Index de Efetividade dos Direitos Económicos e Sociais (IEDES); 6. Dissecando o IEDES do Brasil; 7. Considerações Finais

Summary: 1. Introduction; 2. The Role of Law; 3. Measuring Constitutional Effectiveness; 4. Law and Versteeg's Sham Constitution's Ranking; 5. The Economic and Social Rights Fulfillment Index (ESRF); 6. Dissecting Brazil's ESRF; 7. Concluding Remarks



1. Introduction

The adoption of a new constitution is always a hopeful moment, a moment of belief that the new institutional apparatus established by the new document, with its procedural and substantive norms, will lead to an improvement in society. This is especially so as most constitutions are adopted after difficult periods in the life of a nation, periods that the nation, or at least most of its population, want to leave behind.1 The promulgation of the Brazilian constitution of 1988 is a case in point. Dubbed the “Citizen's Constitution” (A “Constituição Cidadã”), its aim was to leave in the past 21 years of military dictatorship and inaugurate a new era where democracy and the rule of law would prevail and all Brazilians could achieve the status of true citizens. It established, for that purpose, an extensive catalogue of fundamental rights at the very beginning of the constitutional document (articles 5 to 17) and set, as the fundamental principles and objectives of the new Brazilian republic, citizenship, the dignity of the human being, and the construction of a free, just and solidary society, in which poverty should be eradicated and inequalities reduced, and everyone's well-being should be promoted without any discrimination (articles 1 and 3).

There is no doubt that Brazilians have a lot to celebrate in these now more than 25 years of the Citizens' Constitution. We just had at the end of 2014 the seventh consecutive direct presidential election with the participation of the whole population (now more than 140 million electors in a population that has just surpassed 200 million).2 This is the longest continuous democratic period in Brazilian history (the previous one lasted from 1945 to 1964). Our young democracy survived important crises and difficult moments, such as the impeachment for corruption of the first directly elected president in 1989 barely two years after his election. No one speaks seriously any longer of a potential return of the military whenever the Workers' Party grows in voters' intentions. It has actually won the last four elections!

But how has the constitution impacted on the lives of individuals in terms of their civil and social citizenship? Are these significant democratic advancements of a more procedural nature matched by improvements in the effective enjoyment of substantive civil and social rights? This is the question I aim to explore in this article, with particular focus on so-called social and economic rights, which Roosevelt famously named, as a group, as forming a “second bill of rights” in his State of the Union Address of 1944, since the American constitution did not, and to this date does not include these rights.

Brazil, as many, if not most countries that adopted new constitutions in the past few decades, has included an extensive list of these rights in its constitutional document.3 Yet, unlike the more optimistic picture just mentioned in the field of democratic rights of political participation, in the area of social and economic rights (and also to some extent in civil rights), the general view tends to be much less optimistic. Indeed, as put by the sociologist Paulo Sergio Pinheiro in an analysis encompassing the whole of Latin America yet with a significant focus on Brazil, “there exists in these countries a profound gap between social citizenship and political citizenship in the context of a democratic institutional framework.”4 As he emphasized:

“Constitutions in Latin American countries usually incorporate broad provisions for the protection of individual rights, which are then systematically disregarded.”5

To be sure, legislation, and in particular constitutional legislation, suffer everywhere from a degree of ineffectiveness. There is always a gap between the so-called law in the books and law in action. As David Law and Mila Versteeg remind us in their insightful paper “Sham Constitutions”, “the dilemma that constitutions may amount to nothing more than "parchment barriers" is as old as the practice of constitution-writing itself” and affect developed and developing countries alike.6 As they well put it:

“Even under ideal conditions, as in the case of a wealthy, well-established democracy, there inevitably exists a gap between the state as envisioned by a country's formal or "large-c" constitution, and the state that actually exists pursuant to the body of rules, understandings, and practices that make up the informal or "small-c" constitution.”7

Yet, as they also point out, in some cases, and Latin American in general and Brazil in particular have sadly represented glaring historical examples, “that gap can widen to a chasm.”8

How bad, however, the situation really is at the moment in its largest and second greatest economic power, lauded by many international institutions and the international press as a current role model for social and economic progress?9 How wide is the gap still between the social and economic citizenship promises of the Citizen's Constitution and the real life of the Brazilian population?

In what follows I will try to shed some light on this question and, with several caveats and disclaimers that will be explained as the article progresses, will tentatively conclude that there is indeed some cause for celebration, although not for any complacency, far from that.

2. The Role of Law

It is important to acknowledge, from the outset, a difficulty in our quest. As we shall see in detail below, Brazil has undoubtedly made significant social and economic progress in the past two decades and a half. To cite just a couple of indicators, from 1991 to 2010 Brazil's Index of Human Development, which measures advancements in education, health and income, has improved 47,5%, moving the country from the category of Very Low Human Development (0.493 in 1991) to High Human Development (0.727 in 2010).10 Life expectancy grew from 67 in 1990 to 74 in 2012, while infant mortality decreased from 52 to 13 per 1.000 live births in the same period. Illiteracy decreased significantly, achieving less than 10% for the first time in Brazil's history, and most children are now enrolled in primary education.11


Imagem 1


But what has been the role of the constitution, if any, in these undeniable developmental achievements? It is tempting to conclude, especially for a constitutional and human rights lawyer, that all these improvements are a direct consequence of the constitutionalisation of social and economic rights, and all shortcomings, and there are still many of course, a failure of the state (that is, its politicians) to comply with their constitutional obligations. But this rather simplistic view, unsurprisingly not shared by other social and political scientists, is not only inadequate from a purely analytical perspective (as it offers in effect an inaccurate explanation for a complex and nuanced social phenomenon), but also counterproductive. It motivates messianic and unrealistic perceptions of law and its operators, in particular judges, as the primary tools of progressive social change. The now widespread phenomenon of the so-called judicialization of social and economic rights in Brazil and other places is perhaps the most prominent symptom of this problem.12

In what follows, therefore, I will attempt to carry out a more nuanced and realistic analysis of the effect of constitutionalisation of social and economic rights, that is, one that tries to take into consideration as well as it can the complexity of the causal relationship between the formal inclusion of rights in a constitution and the actual enjoyment of these rights by the population. We will also need to grapple with the current lack of a well-established and effective methodology to tackle this complexity.

We must start by admitting a much more complex web of possibilities in the connections between constitutionalisation and rights enjoyment. We must accept that the enjoyment or lack of enjoyment of social and economic rights by individuals of a certain country might occur both as a result of state action or inaction or in spite of state behaviour. There are thus at least four possibilities that a nuanced analysis of the effect of constitutionalisation has to be able to capture. The following table expresses this in a graphical manner:

Rights Satisfaction

State behaviour

Enjoyment (+)

(+) Action

Enjoyment (+)

(-) Inaction

Non-enjoyment (-)

(+) Action

Non-enjoyment (-)

(-) Inaction

Any society is likely to present a nuanced combination of all four possibilities. Let us take an example to illustrate this related to the right to education, included in article 6 of the Brazilian constitution. Take the issue of literacy. According to the latest data, there are 13,2 million illiterate individuals in Brazil above the age of 15, representing 8,7% of the population of that age. What is the causal connection between the right to education expressed in the constitution and this situation? According to the simplistic model, we should conclude that the Brazilian state is respecting, in the field of literacy, the right to education of 91,3% of the population, and disrespecting that same right of the 8,7% of illiterate individuals. 13 Moreover, and since the illiteracy rate was around 20% when the constitution was adopted, we could also conclude (still within the simplistic model) that the constitution played a significant role in the reduction of illiteracy, but still has some way to go.14 The reality, however, is of course much more complex than that. There are millions of people who are literate despite the omission of the state to provide educational services in sufficient quality and quantity who have resorted either to private or charitable schools. There are millions of others who are literate due precisely to the effort of the state to provide free education, and still others who remain illiterate (around 13 million as we saw), some despite having had the opportunity to access state services but not having taking them up and some due to lack of access to these services.

The same applies to all other social rights. Take the right to health. The Brazilian constitution recognises this right in article 6 and has mandated the state to create a universal public health system in articles 196 to 200, that is, one that covers the health needs of the whole of the 200 million individuals in the population. It is known, however, that the system, which spends currently around 3,5% of Brazil's GDP, is insufficient both in coverage and quality of services to do that. As a consequence, around 40 million Brazilians (20% of the population) use almost exclusively private health services, through some sort of insurance or out of pocket payments. The 160 million who rely mostly on the public system receive health services of variable quality and coverage, having often also to resort to private providers (especially for drugs) or go without. Finally, we must not forget that some services are indeed widely available and accessible yet are not taken up by some individuals for reasons that vary from cultural resistance to lack of information and education. Tuberculosis is a classic example where unexpensive and widely available treatment exists yet more uneducated patients, often from lower socio-economic groups, fail to access or comply with treatment directives for lack of knowledge.15

The education and health cases briefly described above are sufficient to show, I hope, the complexity of the causal connection between state conduct and rights enjoyment by individuals. The challenge, if we want a more nuanced and accurate picture of the role and effect of rights constitutionalisation in the enjoyment of rights by a certain population, is to develop a methodology that can capture all this complexity. To determine whether a constitutional document has produced (or failed to produce) effects, and of what nature (positive or negative) and in what measure one has to be able to access and analyse complex data that reveals these four potential situations across the whole of the population.

Moreover, two further complicating layers must be added to this already complex scenario. The first one is the lack of a clear consensus on the exact content of these rights. Beyond a minimum and obvious core, such as for example literacy in the case of education, or basic health care in the case of health, it is not clear what level of these goods the state should be obliged to provide to their citizens. Moreover, given that these goods are dependent on limited resources, state omission to provide them can often be justified by lack of availability of resources. Although the combination of state inaction and lack of rights enjoyment might at first sight indicate rights violation, this is not necessarily the case when social and economic rights are in question. One needs to be able to distinguish inaction that amounts to rights violation from inaction that doesn't.

I have no ambition to develop such a methodology in this article, and am not even sure if it is possible to achieve a reasonably complete one, although I think the project might be a worthwhile one. Here I aim simply to use the Brazilian case, with empirical data and qualitative observations and analysis, to try and show what a more nuanced analysis of the question of the effectiveness of constitutionalisation of social rights can look like. In this endeavour, I will engage with a couple of recent interesting attempts by scholars to develop similar methodologies to the one I believe is required.

3. Measuring Constitutional Effectiveness

How can we gauge whether the constitution is producing in reality what it promised to do? To put it another way, how can we measure state compliance with the constitution? If what I have argued in the previous session is correct, establishing the gap between what the constitution promises and the level of individuals enjoyment is an important yet insufficient step in our enterprise. The fact that 8.7% of the Brazilian population above 15 years of age is illiterate shows simply that a significant portion of the population, almost 14 million people, do not enjoy this minimum level of education. There is a clear gap, therefore, between the promise of the constitution of education as a right of everyone, and the reality of the life of these individuals. But only a prima facie conclusion of constitutional ineffectiveness, or state violation, can be established, as already explained above. 16 We need to further determine whether this gap is indeed a result of state negligence in its duty to provide education to everyone. What makes this task particularly complex especially in the field of social and economic rights, as I mentioned briefly above and must now elaborate, are two main complicating factors. Firstly, there is no clear and consensual definition of the content of these rights, not even an approximate one as exists in relation to other types of rights. Secondly, there is no clear consensus about the level of resources that states are obliged to dedicate to these rights, which makes the task of determining violations even harder. Let me try and explain this in more detail through a couple of comparisons with civil and political rights. Violation of some rights such as the right not to be tortured, freedom of speech, the right to vote etc. can be more easily gauged though simple empirical observation. When a organisation, say Amnesty International or Human Rights Watch reports that several thousand individuals where tortured by the police of a certain state, or are jailed for their political opinions, or do not have the right to choose their government, one might dispute the accuracy of the report, the source of the data etc., but one can be sure that the conduct in question amounts clearly to violations of human rights. When another organisation, say the WHO or UNESCO, reports that a certain number of women have died during child birth, or that a certain percentage of the population has not completed secondary education, the same cannot be automatically concluded.

The twin and interrelated difficulties of indeterminate content/resource dependence affect much more the latter type of rights than the former, making it much harder to determine their content and their violation. Let us go back to education in Brazil. As an upper middle-income country according to the classification of the World Bank, one could perhaps argue that the Brazilian state is under a duty to provide at least universal access to primary education to the population. Access to secondary and higher education, however, and the quality of education as a whole, are more difficult to determine whether and how to include. Brazil has performed consistently poorly in all international assessments of the educational achievement of its pupils (see PISA test).17 That indicates that the quality of education in the country leaves a lot to be desired, but does it also establish a failure of compliance with the right to education in the constitution? To reach that conclusion one would need to demonstrate that the content of the right to education includes secondary education of a certain quality and also that the state has deliberately, or negligently failed to provide such education. The difficulties are readily apparent.

When one expands the analysis to all social rights included in the Brazilian constitution, i.e. health, housing, employment etc, the task becomes even more daunting. Rather than accept it as an impossible venture, however, we should make an effort to gradually develop, collectively and incrementally, better tools to allow us to acquire a deeper understanding of the important issue of whether and under what circumstances constitutions make a practical difference. As well put by David Law and Mila Versteeg, “Even the most difficult and profound causal questions can be investigated empirically … if scholars contribute incrementally and collectively to a research agenda wherein each wave of scholarship simultaneously improves upon what came before and lays the foundation for what must come next.”18

And this is exactly what some scholars have been doing in the past few years. Let us look at two different recent proposals, including Law and Versteeg's one, to see if and how they help us analyse the Brazilian case.

4. Law and Versteeg's Sham Constitution's Ranking

Law and Versteeg's effort in their article Sham Constitutions is an interesting development in the effort to measure the actual effects of rights constitutionalisation. They collected and analysed data from no less than 178 countries and compared the actual enjoyment of rights on the ground with the promises of the constitution. They elected, for reasons of data availability, the following 15 human rights to measure, divided into 3 broad categories: I- Personal Integrity Rights: 1. prohibition of arbitrary arrest and detention; 2. Prohibition of torture; 3. Right to Habeas Corpus; 4. Fair Trial Rights; 5. Prohibition of Death Penalty; II – Civil and Political Freedoms: 6. Freedom of Assembly or Association; 7. Freedom of Movement; 8. Freedom of Religion; 9. Right to Vote; 10. Frredom of the Press and Opinion; III – Socioeconomic and Group Rights: 11. The right to health; 12. Right to Education; 13. Gender Equality in Marriage; 14. Gender Equality in Labour Relations; 15. General Protection of Minority Rights.

They then analysed the actual enjoyment of these rights on the ground by using different sources of empirical data available. Countries that promise more than 10 of these rights in their constitutions but respect less than half of the rights promised were qualified by them as possessing “sham constitutions”. Of the 167 constitutions effective in 2010 for which they possessed data, 39 (22,4%) fell in the category of “sham constitutions”. Brazil was not one of them. According to their method, which scored countries from 0 (no compliance at all) to 1 (full compliance), Brazil scored an overall 0,750, made up by a top score of 1 in personal integrity rights but a half score of 0,5 in social and economic rights.19There are obviously several problems that could be raised in Law and Versteeg's study, as it is inevitable in an effort of such scope, trying to cover the whole world. But the effort is impressive nonetheless and the problems I will highlight about Brazil are therefore not meant in any way as a criticism but rather as a way of exploring where the methodology could be improved.

Their list is perhaps too heavily weighted towards so-called civil and political rights (2/3 of the rights in the list), which immediately makes it less appropriate for our purposes here, focused on social and economic rights. Their restriction to health and education, although doubtlessly two very important social rights, makes their analysis much less capable of representing adequately the vast array of social and economic rights currently recognised in many constitutions, e.g. housing, food, water, labour rights, minimum wage etc. Moreover, to gauge the enjoyment of these rights they elect a single proxy for each, life expectancy for health and literacy rate for education, which, again, restricts significantly the power of their conclusion. Let us use Brazil to illustrate this point. According to Law and Versteeg's methodology, the right to education is highly respected in practice if a country's adult literacy rate is higher than 93%, and lowly respected at all if it is lower than 36%. The right to health is highly respected when life expectancy in a country is above 72 years, and lowly respected when it is below 52. Anything in the middle of these two thresholds makes a country a moderate compliant with these rights. According to this methodology, Brazil would be on the verge of qualifying for high level of respect for the right to education with a literacy rate of 91,3% and is already there regarding the right to health with a life expectancy of 74. But is this a reliable and reasonably accurate evaluation of the effectiveness of Brazil's constitution even on the limited area of health and education? It seems that the limited scope of the data used by Law and Versteeg forces us to answer this question in the negative. Indeed, as we have already discussed above, literacy rates alone tells us very little about a state's compliance with the right to education. Firstly, because literacy is the very bare minimum achievement expected from any system of education. It seems inadequate to use such a low standard to gauge compliance with the right to education, especially “upper middle income” countries such as Brazil.20 Moreover, as also already noticed, this “outcome indicator” does not tell us much, if anything, about the actual reasons for the persistence of illiteracy in a country. The state might be doing its best to eradicate it but still failing for reasons beyond its control. It also fails to capture the possibility we have also highlighted above of rights enjoyment despite state's omission. The same applies to health. Although life expectancy is a better proxy than literacy, especially set at the high threshold of 72, it fails, again, to tell us much about the actual performance of the state in the field of health.

A more informative indicator of enjoyment would necessarily need to include more information on other components of these complex social and economic rights. It would also need to disaggregate them in order to capture disparities in the enjoyment of these goods by different socio-economic-ethnic-gender groups. Especially in very unequal countries like Brazil, average non-disaggregated indicators can be very misleading. 21 Take again life expectancy used by Law and Versteeg. Although it is reasonably high on average in Brazil at 74, the reality is that in some regions of the North-east, it is still below 60s for the poorest quintile of the population, whereas in the south and south east, it reaches almost 80 for the richest quintile of the population. The same is true of most other health indicators, as the table below on another important proxy often used for health, infant mortality, clearly shows. Whereas in the poorest quintile of the population in the poorest region, the Northeast, it is still very high at 56,4 per 1.000 live births, in the richest quintile of the richest region, the South, it is very close to European levels at 10.1. Thus the nickname Belíndia (Belgium + India) popularized in the 1970s by Brazilian economist Edmar Bacha to bring to life Brazil's stark inequalities.22


Table 1


It is true that Law and Versteeg define out of their aim any attempt to gauge why states fail to comply with the promises in their constitution. As they put it:

“We determine whether this definition of a “sham constitution” has been met by looking solely to the magnitude of the gap between what a country promises in its constitution and what it delivers in practice: the larger the shortfall, the more strongly that the constitution is identified as a sham. … our approach renders it unnecessary to speculate upon the reasons for which different countries breach or uphold their constitutional obligations.” 23

But even for their more modest aim of determining the gap between promise and reality it seems that their data is too restricted. Literacy alone cannot capture, for instance, Brazil's promise of education, which establishes clearly in its constitution that primary education should be free and accessible to all. (art. 6 and 205) Life expectancy, as an average, cannot determine compliance with the promise of health, when a significant number in the population are still dying very young in the poorest quintile. Let us look thus at different, and more comprehensive measures developed by other scholars.

5. The Economic and Social Rights Fulfillment Index (ESRF)

The Economic and Social Rights Fulfillment Index (ESRF), developed by Fukuda-Parr, Lawson-Remer and Randolph 24, is in my view a more developed attempt to measure state compliance with social and economic rights. They deliberately tried to overcome the limitations I noticed earlier about “outcome indicators” of the type used by Law and Verteeg. As they properly state:

“Standard socioeconomic indicators of human outcomes (such as infant mortality or school enrolment) are used as a proxy for fulfillment of social and economic rights. Such indicators go some way towards reflecting the extent to which a population is enjoying a particular social or economic right but do not reflect the obligations of the duty bearer that are complex …”25

They also use, as a starting point, outcome indicators, to gauge the extent of rights enjoyment in the relevant country. But they adopt a broader scope, including alongside health and education also food, housing and decent work (social security and non-discrimination are left out for data availability constraints). Another important refinement of their model is that they distinguish between low and middle income countries on one side and high income countries on the other, creating in fact two different indices, one for each (ESRF-1 and ESRF-2 respectively). Finally, they are careful to choose indicators that can capture deficiency of rights enjoyment more thoroughly (often using multiple indicators for each right) in a number of interrelated and interdependent rights dimensions (what they call “bellwether” indicators). The table below shows the indicators they use in ESRF-1, i.e. for low and middle-income countries, where Brazil qualifies.

Right to Food

Malnutrition (height for age)

Right to education

Primary school completion rates; gross secondary school enrolment rates; educational achievement

Right to Health

Under 5 mortality rate;

Life expectancy;

Assisted birth rates

Right to adequate housing

Access to improved water source; access to improved sanitation

Right to decent work

$1.25 a day PPP poverty rate; vulnerable employment

This broader and more detailed set of outcome indicators allows one to arrive at a much more accurate picture of rights enjoyment. But this as already noticed, is just the first step in the analysis. One needs then to gauge whether enjoyment deficiencies found in this first stage can be attributed to the state violation of its duties, which is one of the shortcomings of the index developed by Law and Versteeg as I argued above. For that, they incorporate in their index a component that reflects state's resource capacity, since the main justifiable obstacle for state failure to comply with social rights' obligations is resource constraints. The indicator used here is GDP per capita.

With all these indicators, it is possible to gather extensive information about the level of enjoyment of these five social and economic rights across a population and the level of wealth in the country in question. The challenge then is to produce a methodology through which these data can be used to determine whether the studied country has a level of enjoyment that matches its level of wealth or not. They offer here two different alternative approaches, the ratio approach, and another they call the achievement possibilities frontier (APF). They both involve complicated formulas which is beyond the scope of this article to try to explain in any detail, but the general idea is to build a composite index putting together indices for each of the five rights and establishing the average level of achievement that a country should display at a certain level of economic development expressed by GDP per capita. In a later article the authors applied the APF version to some 150 countries, finding that no country investigated reaches the maximum 100. The best ranked high income country, Finland, achieves 94.66 and Guyana, the best ranked country in the low/middle income group achieves 96,20. The worst in the low/middle income group for which there was available data was Equatorial Guinea, at 15,94, and the United States of America in the high income countries' group, with 76,97. Brazil achieved a respectable 14th position, with a ESRF-1 index of 90.14. The table below shows the breakdown of Brazil's score:


GDP per capita (2005 PPP)

Right to Food Index

Right to Health Index

Right to Education Index

Right to Housing Index

Right to Decent Work Index

ESRF-1 Index











It is of course possible to raise several challenges to this type of enterprise in general and to Fukuda Par and her colleagues' model in particular. The choice of indicators can be disputed, the way in which each of the indices is constructed can be challenged, as can the way they are aggregated, the criteria used to arrive at the APF etc. But there is no doubt, in my view, that the effort is not only impressive but necessary. As the authors properly affirm in the conclusion of their article, “quantitative indicators are an essential tool for assessing the performance of states in fulfilling their obligations to enable their citizens and residents to adequately realize economic and social rights”.26 It is naïve to expect, however, that such indicators can be refined to a level of accuracy, and achieve the level of consensus, which would allow their use as a definite tool to legally judge countries' compliance with their social and economic rights' obligations. They can certainly provide invaluable information in that difficult task. But more specific local information and qualitative analysis will always be needed for a more thorough assessment. I turn now precisely to this assessment in the case of Brazil to gauge how persuasive and accurate is its index achieved through the ESRF1 methodology.

6. Dissecting Brazil's ESRF

I think it would surprise most Brazilians to learn that its level of social and economic rights fulfilment is so high, achieving full marks in the case of education, almost full marks in the field of health, food and decent work, and falling a bit behind, although not much, only in the field of housing. Most Brazilians would instinctively think, rather differently, that Brazil could do much better with the level of economic development it has achieved. Has the ESRF got it wrong or are the negative perceptions just mentioned exaggerated? Or both?

Let us look more closely at the findings of ESRF in the light of other empirical data available to try to answer these questions. It is important to notice, first, that the ESRF, similarly to the “Sham Constitution's index” discussed above, looks only at averages. The authors themselves recognise the problems of this in the presentation of the data on Brazil when they affirm that Brazil's 14th position out of 101 countries “is an average that obscures performance”.27 But even so, for the average to be as high as it is, a significant proportion of the country should be enjoying high levels of social and economic rights adjusted for the level of economic development of the country. Is this the case?

It seems very doubtful in my view and, again, a significant part of the blame for these surprisingly positive results seems to lie in the inadequacy of the proxies used as indicators of rights enjoyment. Despite using a larger variety of such indicators than Law and Versteeg, the ESRF is also restricted by data availability and the authors themselves recognise that “these indicators are proxies and clearly cannot capture the entire breadth of the rights in question.”28 It is not surprising, thus, that the right in which Brazil did worst in the ESRF, the right to housing, is precisely that in which indicators are more complete and adequate to better capture the actual reality of the living conditions of the population. Indeed, enjoyment of this right was gauged through indicators of access to sanitation, improved water source and housing made of durable materials. With a large number of people living in shanty-towns and without access to sanitation and water it would be surprising if Brazil scored well in this right.

The results of the ESRF in the field of housing, despite using mostly different indicators, seem compatible with the latest study of IPEA (Instituto de Pesquisa Econômica Aplicada), a government funded think-thank. Indeed, in its latest evaluation of the so-called housing deficit, a concept adopted by the Ministry of Cities (Ministério das Cidades), it is estimated that between 2007 and 2012, this deficit was reduced from 5,59 to 5,24 million households, a reduction from 10% of the households in the country to 8,53%.29 By housing deficit is meant households that do not comply with a minimum criteria of quality (durable building materials), co-habitation (more than one family per household), affordability (more than 30% of household income) and density (more than 3 people per room). Another important piece of data for our purposes is that the deficit is unsurprisingly higher in the poorest sections of the population and has actually increased in that group, rising from 70% to 73%. 30

As I mentioned above, there can be no automatic extrapolation from a deficit in the enjoyment of the object of a right and a right's violation, and a much more refined methodology would be required to detect the latter. Yet the data just presented gives us a strong indication that the right to housing in Brazil, an upper middle-income country, is being violated when such a large proportion of the population is living below this very minimum criteria, adopted by government itself.

Is the situation of the rights to health and education that much better in order to justify the large difference in their respective scores? (to recall: education – 100; health – 91.93; housing – 76,20). In my view, it isn't, and the disparity here seems in great part due to the indicators chosen. Education is gauged by a single indicator: net enrolment of 7 to 14 year-olds. Now, as we saw above, even in Law and Versteeg's less nuanced model, Brazil did not get 100 in education since its illiteracy rate is still above he required threshold. It is true that the ESRF takes into account resource availability. Yet it seems odd to claim that Brazil is doing everything it can in terms of education with its available resources when the quality of our education, as measured by international exams such as the PISA, is so poor.31 The more plausible conclusion is that the mere enrolment is not an adequate indicator to capture rights enjoyment.

Some light can be shed here by the study carried out by UNESCO in its Education for All (EFA) project. According to its latest report, Brazil is among the 55 countries that has not yet achieved, and is actually not even near achieving, the 6 EFA goals set for 2015, which are: Goal 1. Early childhood care and education; Goal 2. Universal primary education; Goal 3. Youth and adult skills; Goal 4. Adult literacy; Goal 5. Gender parity and equality; Goal 6. Quality of Education.32 Brazil has certainly done well, and much better than in the housing field, in several areas of education. Primary education enrolment is now almost universal, there has been significant progress in gender equality and some progress in adult literacy. Importantly, as the UNESCO report stresses, there has been significant increase in investment backed by taxation revenue. According to the report:

“A well-functioning taxation system enables governments to support the education system with domestic revenue instead of borrowing or relying on external finance. … In some middle income countries, however, such as Egypt, India and the Philippines, there is far greater potential to mobilize domestic resources for education.

India became the world's 10th largest economy in 2011, but tax revenue was equivalent to only 16% of GDP, and government expenditure per person was just US$409. By contrast, in Brazil – the world's sixth largest economy – tax revenue was equivalent to 24% of GDP and expenditure per person was US$4,952 (IMF, 2013).

This huge difference is a key reason Brazil has managed to go further in improving education quality and narrowing learning inequality. The levels of current spending on education as a share of total government expenditure in the two countries also reflect the greater priority that Brazil affords to the education sector. In 2011, government spending on education in Brazil was 18% of total government expenditure, with US$2,218 being spent on each primary school child. India devoted 10% of the government budget to education, with US$212 spent per primary school child.” 33

Yet, as the report also concludes, “even these higher allocations per child often do not adequately equalize spending … as it is estimated that US$971 per pupil is required to attain a minimum level of quality for grades 1 to 4, but in 2009 the government allocated US$ 611 per pupil in the North-East region, about half as much as in the wealthier South-East region (PREAL and Lemann Foundation, 2009).”34 With this richer set of indicators, therefore, it is plausible to argue that the ESRF is overestimating Brazil's compliance with the right to education, although a much deeper analysis would be required, of course, to quantify in what magnitude.

The same goes for health, although perhaps at a lesser intensity. The combination of three indicators (life expectancy, maternal mortality and child mortality) seems better able to capture with more accuracy the health conditions of the population. Yet the score seems nonetheless too high, when, again, it is known that the Brazilian health system has important failures and consumes a meagre 3,5% of its GDP. As argued by Sonia Fleury:

“The disjunctive faced by the Unified Health System (“SUS”), when it guaranteed universal healthcare with such scarce public funds (3,5% of GDP), badly compared even with Latin American countries, ended up producing perverse effects such as the precarisation of labour relations and, often, the deterioration of the quality of services in public health units. … The challenge is to make the right to health, already incorporated in the national political culture, into a claimable one, where a safe, humane and quality service can be guaranteed.”35

This widely held perception of the state system is borne by the empirical data available which, as with education, shows important improvements but also significant deficiencies that have not been yet overcome and seem far from an adequate solution. The table presented above, showing huge disparities in the health indicator of infant mortality, is by and large consistent across all important health indicators such as life expectancy, maternal mortality, pre-natal care, access to health services in general etc.36 I need to reemphasise that none of these challenges are meant as a fatal criticism to the project of developing quantitative models to measure the effectiveness of social and economic rights. On the contrary, I have already expressed several times during this article my admiration and enthusiasm for such projects. They are a great improvement on the usual simplistic analysis carried out by legal academics and rights' activists which I mentioned in the beginning.

My challenges are rather meant as an attempt to contribute to the improvement of these efforts. As the authors of the ESRF methodology themselves recognise in the final sentence of their insightful paper, “other qualitative and quantitative measures are necessary to paint a more complete picture of economic and social human rights fulfilment in Brazil.”37

7. Concluding remarks

As discussed above, despite interesting progress in the development of more solid quantitative measures to gauge the effectiveness of constitutional provisions recognising social and economic rights, they remain limited and in inevitable need of supplement by qualitative analysis. In this article, I have discussed a couple of interesting recent developments in order to shed some light on the question, in particular on the effectiveness of the Brazilian constitution, which is now more than 25 years old.

Despite significant problems of quality, efficiency and underfunding, it is fair to say, as the imperfect quantitative data indicates, that the public services and benefits system in Brazil, a great part of which created by the Constitution, has improved, often significantly, the well being of the Brazilian population in the past couple of decades.38 Our 1988 constitution was at least in part responsible, although as I stressed in this article it is hard to gauge its actual impact, for an unprecedented increase in social expenditure by the Brazilian state. Tax collection by the state raised from 22% when the constitution was adopted to almost 37% currently. The budgets for education, housing and social assistance have grown steadily since the constitution and have achieved now a relatively safe state that not even the austerity measures imposed by the IMF after the 1998 crisis were able to affect significantly.39 Brazil's Index of Human Development moved from very low, at 0.493 in 1991 to high, at 0.727 in 2010, an improvement of 47,5%.40

There are also clear examples of improvements caused directly by the Constitution, such as the establishment of a minimum wage, and the so-called BCP (Benefício de Prestação Continuada), an income supplement provided to anyone who due to age or illness is not able to earn a living above a certain threshold. As well put by Eduardo Fagnani: “the BPC is directed at a group thus far by and large excluded from any public mechanism of income transfer. Its creation changed the standard of social protection … and its implementation has brought inclusion to 2,7 million old and disabled people in situation of extreme poverty.”

Brazil does certainly not have a “sham constitution”! But there is no reason for complacency and self-congratulation. The magnitude of poverty in Brazil remains striking at some 28 million people, and its huge historical inequality, despite also some important improvement, remains one of the largest in the world. The challenge now is to go beyond what has already been achieved, and fight potential retrogression.


t Octavio Ferraz is Reader in Transnational Law at King's College London, where he teaches Public Law, Human Rights and Law and Development. He holds an LLB and an LLM from the University of São Paulo, Brazil, and an MA and PhD from the University of London, UK (from King's College and University College respectively). He was a senior research officer to the UN Special Rapporteur on the Right to Health in 2006, and has published extensively on social and economic rights in Brazil and in transnational perspective, in particular in the field of health.
* This is an extended version of a presentation given at the Conference 70 Years of the Second Bill of Rights held at the University of Lisbon School of Law on the 19th and 20th May, 2014-. I am grateful to the organisers and participants for valuable questions and comments.
1 I am of course leaving out of this description what we might call “non-democratic” constitutions, that is, constitutions that are not the product of democratic deliberation but are rather imposed either externally or within the nation. Yet even those are sometimes genuine efforts to improve society, although most often they are what Giovanni Sartory called “façade constitutions” in his classic typology. Giovanni Sartori, “Constitutionalism: A Preliminary Discussion”, American Political Science Review, 56, 1962, 861.
2 Precisely 141.824.607 people will be apt to vote in the presidential elections of October 5th 2014 according to the Brazilian Superior Electoral Tribunal, an increase of 4,4% from the elections of 2010 (135.804.433).
3 David S. Law & Mila Versteeg, “The Evolution and Ideology of Global Constitutionalism”, California Law Review, 99, 1163, 2011, pp. 1200-01 tbl.2.
4 Juan Mendez, Guillermo O'Donnel and Paulo Sergio Pinheiro (eds), The (Un)Rule of Law & the Underprivileged in Latin America, Notre Dame, 1999, p. 14.         [ Links ]
5 Juan Mendez, Guillermo O'Donnel and Paulo Sergio Pinheiro (eds), The (Un)Rule , p. 9.         [ Links ]
6 David Law and Mila Versteeg, “Sham Constitutions”, 101 California Law Review, 101, 2013, p. 867, citing THE FEDERALIST No. 48, at 308 (James Madison)
Links ] Arial, Helvetica, sans-serif">7 David Law and Mila Versteeg, “Sham”, p. 868.         [ Links ]
8 David Law and Mila Versteeg, “Sham”, p. 868.         [ Links ]
9 At the end of 2009 the British magazine The Economist ran in its cover a picture of the Christ in Rio as a rocket being propelled from the Corcovado mountain with the words “Brazil takes off” and calling the country “Latin America's big success story”. As well as the growing economy, the piece compared Brazil favourably to the other so-called BRIC countries for being a democracy and praised president Luiz Inacio Lula da Silva's government for his social policy that “moved to reduce the searing inequalities that have long disfigured it”. “Brazil takes off”, The Economist, Nov 12 2009, It is true that the celebratory mood has turned a little recently with the stagnation of Brazil's economic growth and the street protests of July 2013. See “Has Brazil blown it?”, The Economist,
10 accessed on 21.8.2014.
11 Comprehensive historical data can be found at
12 Most of my research in the past few years has been focused on an attempt to measure the actual impact of the judicialization of health in Brazil. Although the vast majority of legal academics and human rights activists are supportive of it, my findings, despite not being conclusive, gives in my view justified cause for caution in celebrating judicialization. See, among others, Octavio Ferraz, “The right to health in the courts of brazil: worsening health inequities? ”, Health and Human Rights, An International Journal, Vol 11, No 2, 2009; “Harming the Poor through Social Rights Litigation”, Texas Law Review, Issue 7 of Volume 89, 2011; “Health Inequalities, Rights and Courts: The Social Impact of the “Judicialization of Health” in Brazil”, in Yamin and Gloppen (eds.), Litigating the Right to Health: Can courts bring more justice to health systems?, Cambridge, 2011, and, with Daniel Wang: “Reaching out to the needy? Access to justice and public attorneys' role in right to health litigation in the city of Sao Paulo”, Sur - International Journal on Human Rights, v. 10, n. 18, 2013.
14 This is of course a gross and immediately recognisable simplification by anyone familiar with the history of education in Brazil. Even during the military period illiteracy fell by half, from around 40 to 20%.
15 Danúzia Rocha and Rubens Adorno, “Abandono ou Descontinuidade do Tratamento da Tuberculose em Rio Branco, Acre” (Dropouts or Discontinuity of the Treatment of Tuberculosis in Rio Branco, Acre), Saúde Soc. São Paulo , v.21, n.1, pp. 232-245, 2012, available at
16 Note, here, that Law and Versteeg's invaluable work on “Sham Constitutions”, discusses below in the text, provides information of this first step. I would not use the language of “compliance” and “violation”, however, when referring to the gap between the formal constitution and the reality on the ground, since that gap may have a legitimate justification, or its inexistence might have nothing to do with the behaviour of the state, as I argued above.
17 In the latest edition of 2012 Brazil scored at its best performances 57 (mathematics) 54 (reading) and 57 (sciences) among 65 countries. Available at
18 Idem, at 935.
19 They found no reliable data for civil and political freedoms to use in their study.
20 According to the World Bank known classification, available at
21 I have explored the egalitarian dimension of social and economic rights elsewhere and won't discuss it in any further depth in this paper. See, e.g., Octavio Ferraz, “The Right”.
22 See his fable O Rei Da Belíndia (uma fábula para tecnocratas), 1974, available at
23 op. cit. at 880.
24 Sakiko Fukuda-Parr , Terra Lawson-Remer & Susan Randolph, “An Index of Economic and Social Rights Fulfillment: Concept and Methodology”, Journal of Human Rights, 8:3, 2009, pp. 195-221.         [ Links ]
25 Sakiko Fukuda-Parr , Terra Lawson-Remer & Susan Randolph, “An Index”, p. 197.         [ Links ]
26 Sakiko Fukuda-Parr , Terra Lawson-Remer & Susan Randolph, “An Index”, p. 256.         [ Links ]
27 Guyer, Parr, Randolph and Moreira-Daniels, Measuring the Progressive Realization of Economic and Social Human Rights in Brazil: A Disaggregated Economic and Social Rights Fulfillment Index , 2009.
28 Guyer, Parr, Randolph and Moreira-Daniels, Measuring, p. 10.
29 V. Lima Neto, B. Furtado and C. Krause, Nota Técnica IPEA. Estimativas do Déficit Habitacional brasileiro (PNAD 2007-2012), 2013.
30 Guyer, Parr, Randolph and Moreira-Daniels, Measuring, p. 12.
31 See note 17 above.
32 EFA Global Monitoring Report 2013/2014, Teaching and Learning: Achieving quality for all, UNESCO, 2014.
33 EFA Global Monitoring Report, p. 118.
34 EFA Global Monitoring Report, p. 126.
35 Sonia Fleury, “Políticas públicas em busca da qualidade dos serviços de saúde”, in Flávio Vaz, Juliano Musse, and Rodolfo Santos, (Coord.), 20 anos da Constituição Cidadã: Avaliação e desafios da Seguridade Social, 1ª ed., Brasília 2008, pp. 241-246.
36 For a comprehensive set of data and analysis, see the final report of the National Commission for the Social Determinants of Health (2008), available at
37 At p. 24.
38 Flávio Vaz, Juliano Musse, and Rodolfo Santos, (Coord.), 20 anos da Constituição Cidadã: Avaliação e desafios da Seguridade Social, 1ª ed., Brasília, p. 4. My translation from the Portuguese: “O BCP se destina a um público até então majoritariamente excluído de qualquer mecanismo público de transferência de renda. sua criação trouxe uma mudança no padrão de proteção social brasileiro no campo da garantia de renda, tradicionalmente identificado com os seguros sociais, e sua implementação tem significado a manutenção de níveis mínimos de bem estar para mais de 2,7 milhões de pessoas idosas e com deficiência em situação de extrema pobreza.”
39 As well remarked by a commentator: “Além da restauração do estado Democrático de Direito, o legado dos movimentos sociais foi a construção de um razoável sistema de proteção social conquistado na contramão do pensamento neoliberal hegemônico em escala mundial e do movimento em direção ao estado mínimo a que foram submetidos os países subdesenvolvidos, incluindo os da América latina. em grande medida, isso decorreu do fato de que a agenda da redemocratização do País, impulsionada pelo movimento social, não abriu brechas para os experimentos neoliberais – pelo menos até o final dos anos 80. O cenário hostil, a partir de 1990 e até os dias atuais, não impediu que parcela significativa das conquistas de 1988 fosse consagrada.” (Eduardo Fagnani, “Direitos sociais no fio da navalha”, in Flávio Vaz, Juliano Musse, and Rodolfo Santos, (Coord.), 20 anos da Constituição Cidadã: Avaliação e desafios da Seguridade Social, 1ª ed., Brasília, 2008, p. 23).
40 accessed on 21.8.2014.