This paper introduces the notion of geostructures of inequality, examining how inequality indicators are conceptualized and utilized in regional and development studies. This conceptual contribution is based on the use of partially ordered set theory (posets). The paper also offers an empirical contribution based on the use of Hasse diagrams to build regional benchmark profiles of the human development index (HDI). This paper identifies, among its key findings, regions characterized by more vertically oriented geostructures of inequality and, through sensitivity analysis, establishes how such structures are associated with more fundamental development indicators, including life expectancy.

A situação dos indígenas brasileiros é de exclusão, onde a imensa maioria reside nas reservas a eles destinadas. Este artigo objetiva explorar o conceito de aporofobia em relação ao indígena brasileiro. Os resultados demonstram que a aporofobia indígena emerge como um campo discursivo mensurável, com similaridade de 0,61 entre phobia e poverty no FastText e acima de 0,85 no BERT, revelando uma sobreposição semântica entre índio, pobreza e discriminação. A pesquisa analisou 676 notícias online publicadas ao longo de 12 meses em quatro veículos de comunicação, e utilizou análise léxica e de contexto com métodos FastText e BERT. Os resultados apontam ainda que 76% das notícias tratam dos direitos indígenas, porém foi detectado que a aversão e a pobreza têm comportamentos de grande similaridade. Encontramos ainda que o indígena é mais relacionado aos termos de pobreza que os de hostilidade e aversão. Este artigo contribui para preencher a lacuna sobre estudos de aporophobia no Brasil, e avança no entendimento da percepção de conceitos que moldam a sociedade brasileira contemporânea. Os resultados indicam que a aporofobia indígena emerge como uma região semântica concreta — um subconjunto do discurso social onde pobreza e rejeição se cruzam no campo indígena— oferecendo evidências empíricas que podem orientar políticas públicas voltadas aos povos indígenas e ao enfrentamento do preconceito.

PurposeThis study aims to examine how national higher education systems and institutional strategies foster high-quality education and promote societal impact. By analyzing country and university outcomes in the QS World University Ranking (QSWUR) and QS Sustainability Ranking (QSS), it explores how governance, research capacity, institutional collaboration and diversity are associated with university performance in both academic excellence and sustainability.Design/methodology/approachCross-sectional analysis of 436 universities in 48 countries using mixed-effects models and bootstrapped ordinary least squares regressions.FindingsCountries with regulatory governance perform better in both academic and sustainability rankings, though most variation occurs at the university level (93% in QSWUR; 73% in QSS). Research output is the strongest predictor. Industry collaboration contributes to academic scores, while open-access publishing negatively affects both rankings. Gender equity improves academic performance only when women's authorship exceeds 29%, indicating that greater representation is needed for positive effects.Research limitations/implicationsFindings are limited to the QSWUR and QSS. Further research is needed to assess whether results apply to other ranking systems.Practical implicationsThis study offers insights into improving university performance in global rankings, by prioritizing high-quality research, fostering industry collaboration, advancing gender equity and strengthening internal governance within supportive national regulatory environments.Social implicationsThe negative effect of open-access publishing suggests that current rankings may undervalue practices that promote public access to knowledge. This study also highlights structural barriers to gender equity, as most universities fall below the threshold where women's research contributions begin to positively affect performance.Originality/valueThis study advances existing research by analyzing how both country- and institutional-level factors are associated with university rankings in traditional academic and sustainability domains. This study offers practical guidance for improving performance while advancing sustainable development goals.

A desigualdade é uma característica persistente da economia e da sociedade brasileira. Este artigo explora o conceito de aporofobia fiscal por meio de duas abordagens empíricas complementares. A primeira analisa a estrutura semântica do discurso midiático brasileiro em torno de três debates fiscais centrais ocorridos em 2025: a isenção do imposto de renda para rendas mais baixas, a criação de um imposto mínimo sobre altas rendas e o mecanismo de cashback tributário para famílias de baixa renda. Utilizando análise semântica baseada no modelo FastText, foi examinado um corpus de 1.748 notícias publicadas em cinco grandes portais nacionais, permitindo identificar padrões discursivos de apoio ou aversão às políticas fiscais pró-pobres. Os resultados mostram que o discurso explicitamente aporofóbico aparece em apenas 1,6% das notícias, enquanto 95% apresentam enquadramento ambivalente, combinando apoio às políticas redistributivas com narrativas que enfatizam custos fiscais, mecanismos de controle ou riscos de dependência. Nenhuma notícia foi classificada como exclusivamente pró-pobre, sugerindo que o debate público tende a legitimar políticas redistributivas apenas sob condições de disciplina fiscal e responsabilização. A segunda abordagem consiste na construção do Índice de Aporofobia Fiscal (IAPF) aplicado aos estados brasileiros, baseado na articulação entre um subíndice de despesa em assistência social e um subíndice de receita relacionado à composição da arrecadação tributária, com ênfase na tributação direta. Os resultados revelam significativa heterogeneidade interestadual: o índice variou entre 0,10 e 0,62. Estados como Bahia e Espírito Santo apresentam os menores valores do indicador, enquanto casos como Goiás exibem desempenho comparativamente mais favorável. A distribuição espacial do índice também sugere a presença de padrões regionais na configuração fiscal subnacional. Ao integrar análise computacional de discurso com indicadores fiscais, o estudo contribui para ampliar a compreensão das relações entre desigualdade, justiça fiscal e construção simbólica da pobreza.

This paper critically engages with the LNOB principle of the 2030 Agenda, highlighting its conceptual, methodological, and structural limitations. Building on Amartya Sen's social choice theory and Rawlsian justice, it reconceptualizes "sustainability as justice," emphasizing real-world comparative assessments grounded in intersectionality. It develops a novel methodological framework combining the CART algorithm and its descriptive and statistical outputs with the D-index to systematically identify, measure and assess exclusion across plural informational spaces-resources, capabilities, rights and liberties, and subjective well-being. Applying this framework to MICS data across nine countries, the paper reveals how conventional SDG disaggregation masks structural inequalities and fails to capture the realities of the worst-off groups. Instead, it uncovers context-specific patterns of deprivation and prioritization, offering targeted, empirically grounded insights for policy reforms. Ultimately, this approach reorients LNOB from an aspirational slogan to a justice-centered, operational tool capable of diagnosing and addressing systemic disadvantage in sustainable development.

The paper provides three key contributions to examine how youth development can be used to approach the future: first, it introduces a new age-disaggregated human development indicator focused on the youth, named ‘Youth HDI’. It centres on educational, living standard and health issues. Secondly, it shows how partial ranking analysis (posets) can be used to unfold incomparabilities in ordering regions, overcoming traditional limitations associated with the use of composite indicators. Finally, it employs Phillips and Sul (2009) convergence test and clustering algorithms to assess the evolution of the Youth HDI for different regions. These contributions are quintessentially methodological about evidence for 'perspective of futures'. This illustration shows how youth’s vulnerability and marginalisation are not an exclusivity of developing countries. It unlocks the debate about the consequences of current youth’s human development to sustainability and how this kind of disaggregated indicator is essential for fulfilling 2030 Agenda’s transformative promise of “leaving no one behind”

There is considerable debate in the literature as to the precise definition of ‘left-behind’ places, and the appropriate metrics for identifying them, and for successfully targeting regional policy interventions. We propose an evaluation structure for ‘left-behindness’ based on Amartya Sen’s capability approach, and argue that the diversity of criteria should not be seen as a shortcoming, but rather as a richness to be explored. We show, through the use of ‘posets’ and Hasse diagrams, that there are several distinct ‘structures of left-behindness’ across European NUTS 2 regions, revealing spatial imbalances that do not conform to a ‘one size fits all’ narrative.

The capability approach is a versatile framework rooted on issues of justice and multidimensional assessment of quality of life developed in the 1980s as an alternative approach to prevailing mainstream development ideas focused narrowly on economic development. Most closely associated with the work of Amartya Sen, it has become of great interest to development scholars from a variety of different disciplines. Much has already been done exploring the conceptual foundations of the capability approach and discussing Sen's contribution to the field, but few books have explored the links between social choice (another field with rich contributions by Sen) and human development issues. Featuring many of the world's leading experts on social choice theory and capability indicators, Social Choice, Agency, Inclusiveness and Capabilities combines these interrelated themes into one volume and fully explores the relevance of social choice to human development.

During the military regime in Brazil (1964-1985), enrollment ratios in primary education grew substantially in the first decade under dictatorship, but stagnated in the mid-1970s. This paper shows that education spending might depend on the levels of centralisation in tax matters. Using panel data regressions and qualitative evidence, we argue that a massive big push industrialisation programme increased the pressure on external accounts, leading the government to intensify an export incentive policy based on tax subsidies that decreased the income of subnational governments. As a result, the capacity of funding mass education was compromised in the second half of the 1970s. Durante el r & eacute;gimen militar en Brasil (1964-1985), las tasas de matr & iacute;cula en la educaci & oacute;n primaria crecieron sustancialmente en la primera d & eacute;cada bajo la dictadura, pero se estancaron a mediados de los a & ntilde;os setenta. Este art & iacute;culo muestra que el gasto en educaci & oacute;n podr & iacute;a depender de los niveles de centralizaci & oacute;n en materia tributaria. Utilizando regresiones de datos de panel y evidencia cualitativa, sostenemos que un programa masivo de industrializaci & oacute;n aument & oacute; la presi & oacute;n sobre las cuentas externas, lo que llev & oacute; al gobierno a intensificar una pol & iacute;tica de incentivos a las exportaciones basada en subsidios fiscales que disminuyeron los ingresos de los gobiernos subnacionales. Como resultado, la capacidad de financiar la educaci & oacute;n masiva se vio comprometida en la segunda mitad de los a & ntilde;os setenta.

Among the myriad of technical approaches and abstract guidelines proposed to the topic of AI bias, there has been an urgent call to translate the principle of fairness into the operational AI reality with the involvement of social sciences specialists to analyse the context of specific types of bias, since there is not a generalizable solution. This article offers an interdisciplinary contribution to the topic of AI and societal bias, in particular against the poor, providing a conceptual framework of the issue and a tailor-made model from which meaningful data are obtained using Natural Language Processing word vectors in pretrained Google Word2Vec, Twitter and Wikipedia GloVe word embeddings. The results of the study offer the first set of data that evidences the existence of bias against the poor and suggest that Google Word2vec shows a higher degree of bias when the terms are related to beliefs, whereas bias is higher in Twitter GloVe when the terms express behaviour. This article contributes to the body of work on bias, both from and AI and a social sciences perspective, by providing evidence of a transversal aggravating factor for historical types of discrimination. The evidence of bias against the poor also has important consequences in terms of human development, since it often leads to discrimination, which constitutes an obstacle for the effectiveness of poverty reduction policies.

This paper analyses whether the human capital levels embodied in immigrants can explain xenophobic trends for 126 regions in 14 EU-15 countries from 1998 to 2018. It tests if xenophobic regions may be rejecting immigrants because they are poor, a phenomenon recently defined as 'aporophobia'. The results indicate that larger inflows of low-educated immigrants working in low-skilled occupations are significantly correlated with a higher rejection of migrants, thus confirming the aporophobia hypothesis. The findings in this paper bring light to the discussion of a powerful concept which underpins the need for a more just society.

This paper provides a composite analysis of children's academic development grounded on the capability approach. The study utilises a panel dataset comprising 8,422 Chinese children and adolescents aged 6 to 16, observed between 2012 and 2018. It introduces a series of innovative indicators, including a parent advantage index to capture how parents influence their children and a ranking indicator for spending priorities to reify the value of children's education that families have reasoned. To address unobserved heterogeneity, we adopted fixed-effects models, multilevel modelling, and heteroskedasticity-based instrumental variables. Our primary results show that a 1% increase in the parent advantage index yields an increase of 13.85% to 21.31% in children's academic development, and the biggest leap in prioritising education-relevant spending increases the child outcomes by 2.88% to 6.57%. By highlighting the influence of parents' beings and doings, particularly the value they assign to education, this research contributes to the existing literature on child development, which often focuses predominantly on material dimensions. In sum, it expands the frontiers of the capability approach and related research on parental practices. It offers novel insights into how policies can be reinforced to equalise educational opportunities and to boost human capital.

We argue for a broader and more deliberative regional policy-making process that can be used to better identify the needs of diverse left-behind communities and develop appropriate policies. We argue that the capability approach’s quintessentially inclusive and broad scope, and focus on real opportunities, agency, and process might better address the challenges of regional development. We use these insights to lay out a practical guide for how the capability approach could be used in policymaking, breaking down the implementation approach into steps, and providing examples from a variety of contexts to show how each step might be achieved in practice.

This paper clarifies why bias cannot be completely mitigated in Machine Learning (ML) and proposes an end-to-end methodology to translate the ethical principle of justice and fairness into the practice of ML development as an ongoing agreement with stakeholders. The pro-ethical iterative process presented in the paper aims to challenge asymmetric power dynamics in the fairness decision making within ML design and support ML development teams to identify, mitigate and monitor bias at each step of ML systems development. The process also provides guidance on how to explain the always imperfect trade-offs in terms of bias to users.

The aim of this paper is to investigate the casual relationship and the mechanisms behind school gender composition and scholastic achievement. We use four cohorts of 5th grade student’s assessment of Math and Portuguese in Brazilian elementary public schools and apply an identification strategy relying on school level gender peer effects and school fixed effects to control for self-selection of students across schools. We identify a positive relationship between scholastic achievement and the proportion of female students. This effect is underlined by improvements in student behavior, as indicated by teachers’ assessment of student’s academic potential, self-esteem, interest, discipline, absenteeism, and perception about syllabus coverage and school violence. Hence, this research draws attention to gender as an important factor in the allocation of students and teachers within schools. The consideration of our findings in the formulation and execution of policies can result in effective low-cost measures aimed at increasing achievement.

This chapter examines some relevant ethical elements behind the theme of the commons. It aims at raising some questions about what has been, in my opinion, ignored in the debate on the commons so far: the role of moral sentiments. The chapter examines Jean Tirole's (Economics for the common good. Princeton University Press, Princeton, 2017) Economics for the Common Good book and explores how moral sentiments are pivotal to better understanding social choice and collective action challenges. To conclude, I focus on Fratelli Tutti, discussing how social love can be part of a new ethical structure of the commons.

This paper proposes and estimates three novel higher education indices for 31 Chinese provinces: i) the Chinese Higher Education Density Index (CHEDI) to analyze the evolution of the quantitative distribution of higher education institutions (HEIs) in each province from 2001 to 2017, which is further decomposed into subgroups based on the type of college, i.e., four-year undergraduate colleges, two-year vocational colleges, and private institutions; ii) the Chinese Higher Education Quality Index (CHEQI) to examine the supply of higher education in terms of quality using a university ranking system; and iii) the Chinese Higher Education Index (CHEI), a composite indicator that incorporates both the quantity and quality dimensions of higher education institutions for each province, providing a weighted measure of the supply of higher education in China. The empirical findings indicate a significant and persistent heterogeneity in the supply of higher education between provinces. The indices identify which regions have been substantially rewarded by the higher education expansion of recent decades, going from an undersupply to a proportionate supply of higher education institutions. On the other hand, a significant share of regions still has a low supply in terms of either the quantity or quality of HEIs, or both.

Sustainability requires balanced development. The economy, society and the environment all need to be pursued simultaneously. In this context, the issue of incomparabilities among different dimensions of indices is at the heart of the discussion. However, this crucial issue is not fully addressed in the existing literature. Whereas dashboards leave to the readers the difficult task of making sense of a complex array of symbols indicating distinct goal levels and trends, composite indices sidestep the issue altogether. To overcome this state of affairs and provide a structural picture of sustainability, we introduce a poset (i.e. partially ordered set) analysis as a middle ground between the two extreme techniques used in assessing progress towards SDGs. By doing so, it aims to improve the use of SDG indicators. Our study finds that partial ordering offers a more transparent, simpler and more intuitive approach than the alternatives. It not only corrects any imbalance in the joint performance of different dimensions but also accords the environment the highest impact upon the overall SDGs. In this it is unlike the existing composite indices, which accord the economy the highest impact. The poset analysis is thus an appropriate technique for the pursuit of sustainable development.