¹ Morning, Ann, ‘Ethnic Classification in Global Perspective: A Cross-National Survey of the 2000 Census Round’, Population Research and Policy Review, 27: 2 (2008), pp. 239–72CrossRefGoogle Scholar.

² Loveman, Mara, National Colors: Racial Classification and the State in Latin America (Oxford and New York: Oxford University Press, 2014), p. 252 CrossRefGoogle Scholar.

³ Nobles, Melissa, Shades of Citizenship: Race and the Census in Modern Politics (Stanford, CA: Stanford University Press, 2000)Google Scholar.

⁴ Loveman, National Colors.

⁵ Hochschild, Jennifer and Powell, Brenna M., ‘Racial Reorganization and the United States Census 1850–1930: Mulattoes, Half-Breeds, Mixed Parentage, and the Mexican Race’, Studies in American Political Development, 22: 1 (2008), pp. 59–96 Google Scholar; and Mora, G. Cristina, ‘Cross-Field Effects and Ethnic Classification: The Institutionalization of Hispanic Panethnicity, 1965 to 1990’, American Sociological Review, 79: 2 (2014), pp. 183–210 CrossRefGoogle Scholar.

⁶ There are two distinct words in Brazilian Portuguese that translate into English as ‘black’ and two that translate as ‘brown’. For each ‘black’ and ‘brown’ there is one word that is generally used to describe the colour of objects, and one word that is used in reference to people or socio-cultural concepts. In each case, the ‘object’ term is also the term the census uses. The census term for brown is pardo, and black is preto. The socio-cultural terms are moreno and negro, respectively. Neither preto nor negro translate precisely into the racial concept of ‘black’ as it is used in the United States. For greater clarity, whenever possible we use the Portuguese terminology.

⁷ The categories of branco and preto were used each year the census collected race or colour data (1872, 1890, 1940, 1950, 1960, 1980, 1991, 2000 and 2010). Pardo was used in 1872, in 1890 it was substituted by mestiço, in 1940 it was used to code the answers left in blank, and in 1950 it became a category for self-identification together with branco, preto and amarelo (yellow). The amarelo category was added in 1940 and the indígena (indigenous) category was added in 1991: IBGE, Características Étnico-Raciais da População (Rio de Janeiro: IBGE, 2011)Google Scholar. For more on the colour question in the early twentieth century see Loveman, Mara, ‘The Race to Progress: Census Taking and Nation Making in Brazil (1870–1920)’, Hispanic American Historical Review, 89: 3 (2009), pp. 435–70Google Scholar.

⁸ Statistical analyses of Brazilian census data have shown that the same individuals may select different racial categories over time. One study of census data from 1950 to 1991 found that around of a third of Brazilians who classified themselves as preto in the early period later reclassified themselves as pardo: de Carvalho, José Alberto Magno, Wood, Charlie and Andrade, Flávia Cristina, ‘Estimating the Stability of Census-based Racial/Ethnic Classifications: The Case of Brazil’, Population Studies, 58: 3 (2004), pp. 331–43Google Scholar. Other studies utilise survey data on racial identity and identification. Some compare how individuals classify themselves, versus how they are classified by others, e.g. Bailey, Stanley, ‘Unmixing for Race Making in Brazil’, American Journal of Sociology, 114: 3 (2008), pp. 577–614 Google Scholar. Others compare how individuals classify themselves when given the census categories versus open-ended options, e.g. Petruccelli, José Luis, A cor denominada: Estudos sobre a classificação étnico-racial (Rio de Janeiro: DP&A, LPP, UERJ, 2007)Google Scholar. Qualitative studies also show that racial identification varies according to context, e.g. Sheriff, Robin, Dreaming Equality: Color, Race and Racism in Urban Brazil (New Brunswick, NJ: Rutgers University Press, 2001)Google Scholar.

⁹ Nobles, Shades of Citizenship.

¹⁰ Loveman, National Colors.

¹¹ Mora, ‘Cross-Field Effects and Ethnic Classification’.

¹² Hochschild and Powell, ‘Racial Reorganization’.

¹³ Interviews were conducted by the authors between 2007 and 2010 with six former and currently serving high-level officials at IBGE. Interviewees were identified through snowballing, and we targeted those who currently worked on racial statistics or who had done so in the past. Several respondents were interviewed on multiple occasions. In addition, we interviewed two black activists and one indigenous activist who had been leaders on issues of racial classification policy.

¹⁴ Anderson, Benedict, Imagined Communities: Reflections on the Origin and Spread of Nationalism (New York: Verso, 1983)Google Scholar.

¹⁵ Tishkov, Valery, ‘The Population Census and the Construction of Identity’, Anthropology and Archeology of Eurasia, 44: 2 (2005), pp. 10–40 Google Scholar and Zuberi, Tukufu, Thicker than Blood: How Racial Statistics Lie (Minneapolis, MN: University of Minnesota Press, 2001)Google Scholar.

¹⁶ Bates, Robert, ‘Modernization, Ethnic Competition, and the Rationality of Politics in Contemporary Africa’, in Rothchild, Donald and Olorunsola, Victor A. (eds.), State versus Ethnic Claims: African Policy Dilemmas (Boulder, CO: Westview Press, 1983), pp. 152–71Google Scholar.

¹⁷ Posner, Daniel, Institutions and Ethnic Politics in Africa (New York: Cambridge University Press, 2005)Google Scholar.

¹⁸ Kertzer, David and Arel, Dominique, ‘Censuses, Identity Formation, and the Struggle for Political Power’, in Kertzer, David and Arel, Dominique (eds.), Census and Identity: The Politics of Race, Ethnicity, and Language in National Censuses (Cambridge: Cambridge University Press, 2002), pp. 1–42 Google Scholar.

¹⁹ Paul Schor has shown that some US groups risked repression and mobilised to be classified on the census solely for the prestige that may come from formal recognition as a group. Schor, Paul, ‘Mobilising for Pure Prestige? Challenging Federal Census Ethnic Categories in the USA (1850–1940)’, International Social Science Journal, 57: 183 (2005), pp. 89–101 Google Scholar.

²⁰ Loveman, National Colors; Morning, ‘Ethnic Classification in Global Perspective’; and Paschel, Tianna, Becoming Black Political Subjects: Movements, Alignments and Ethno-Racial Rights in Brazil and Colombia (Princeton, NJ: Princeton University Press, 2016)Google Scholar.

²¹ Mora, ‘Cross-Field Effects and Ethnic Classification’, p. 157.

²² In the words of Roszak, Theodore (The Making of a Counter Culture (Garden City, NY: Anchor, 1969), p. 8)Google Scholar, technocrats ‘justify themselves by appeal to scientific forms of knowledge’, quoted in Putnam, Robert, ‘Elite Transformation in Advanced Industrial Societies: An Empirical Assessment of the Theory of Technocracy’, Comparative Political Studies, 10: 3 (1977), pp. 383–412 Google Scholar.

²³ Putnam, ‘Elite Transformation’.

²⁴ Hochschild and Powell, ‘Racial Reorganization’.

²⁵ Hall, Peter, ‘Historical Institutionalism in Rationalist and Sociological Perspective’, in Mahoney, James and Thelen, Kathleen (eds.), Explaining Institutional Change: Ambiguity, Agency and Power (Cambridge: Cambridge University Press, 2010), pp. 204–23Google Scholar.

²⁶ Schneider, Ben Ross, Burocracia pública e política industrial no Brasil (Paulo, São: Ed. Sumaré, 1991)Google Scholar and ‘The Career Connection: A Comparative Analysis of Bureaucratic Preferences and Insulation’, Comparative Politics, 3 (1993), pp. 331–50Google Scholar.

²⁷ Adler, Emanuel, ‘Ideological “Guerrillas” and the Quest for Technological Autonomy: Brazil's Computer Industry’, International Organization, 40: 3 (1986), pp. 673–705 Google Scholar.

²⁸ Tulia Falleti, ‘Infiltrating the State: The Evolution of Health Care Reforms in Brazil, 1964–1988’, in Mahoney and Thelen (eds.), Explaining Institutional Change, pp. 38–62.

²⁹ Additionally, in contrast to that of the United States, the Brazilian academic community was relatively weak until the late twentieth century, giving more autonomy to IBGE officials to define census categories.

³⁰ Lieberson, Stanley, ‘The Enumeration of Ethnic and Racial Groups in the Census: Some Devilish Principles’, in Joint Canada–United States Conference on the Measurement of Ethnicity (eds.), Challenges of Measuring an Ethnic World: Science, Politics, and Reality (Ottawa, Ontario: Statistics Canada and U.S. Department of Commerce, 1993), pp. 23–36 Google Scholar.

³¹ During this period, other national surveys have also been conducted by non-IBGE actors to test and explore racial categories and their meanings, such as the Datafolha Racismo Cordial (1998), Pesquisa Social Brasileira (2003), Fundação Perseu Abramo (2003), Datafolha Racismo (2008), LAPOP (2010) and PERLA (2010).

³² Stepan, Nancy Leys, ‘The Hour of Eugenics’: Race, Gender and Nation in Latin America (Ithaca, NY: Cornell University Press, 1991)Google Scholar; Schwarcz, Lilia Moritz, The Spectacle of the Races: Scientists, Institutions and the Race Question in Brazil, 1870–1930 (New York: Hill and Wang Press, 1999)Google Scholar; Dávila, Jerry, Diploma of Whiteness: Race and Social Policy in Brazil, 1917–1945 (Durham, NC: Duke University Press, 2003)Google Scholar.

³³ Nobles, Shades of Citizenship, p. 115.

³⁴ Paschel (Becoming Black Political Subjects) characterises the period of military dictatorship as a ‘stable domestic political field’ where meaningful change should not have been possible. Importantly, our account runs counter to the idea that for domestic changes to occur, state discourse and policies on race need to radically shift.

³⁵ Authors’ interview with José Luis Petruccelli, 29 July 2010, Rio de Janeiro, Brazil (digital recording in possession of authors).

³⁶ Since 1950, census racial questions have been based on self-identification of the respondent (who would also classify the others in the household).

³⁷ IBGE, VIII Recenseamento geral: parecer sobre o âmbito em extensão e profundidade do censo demográfico (Rio de Janeiro: IBGE, 1969), pp. 1–15 Google Scholar.

³⁸ Ibid, p. 2.

³⁹ Nobles, Shades of Citizenship. Freyre was closely connected to the military regime during its early years (1964–8). His ideas had been politically influential since the Vargas New State (1930–7), although the political relationship between Vargas and Freyre was turbulent. For the trajectory of the idea of racial democracy in Brazilian politics see Antonio Guimarães, ‘Democracia racial: el ideal, el pacto y el mito’, Estudios Sociológicos (2002), pp. 305–33. Freyre openly accepted his relationship with the military dictatorship in an interview with journalist Ricardo Noblat in Playboy magazine, March 1980.

⁴⁰ de Oliveira, Jane Souto, ‘Brasil mostra a tua cara’: imagens da população brasileira nos censos demográficos de 1872 a 2000 (Rio de Janeiro: Escola Nacional de Ciências Estatísticas, 2003), p. 34 Google Scholar.

⁴¹ Loveman, National Colors, p. 217.

⁴² Costa, Tereza Cristina, ‘Considerações teóricas sobre o conceito de indicador social: uma proposta de trabalho’, Revista Brasileira de Estatística, 36: 142 (1975), pp. 167–75Google Scholar.

⁴³ Authors’ interviews with José Luis Petruccelli, 15 Aug. 2007, Rio de Janeiro, Brazil and with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil (digital recording in possession of authors).

⁴⁴ Loveman, National Colors, p. 245.

⁴⁵ Paschel, Becoming Black Political Subjects.

⁴⁶ Schneider, ‘The Career Connection’.

⁴⁷ Schwartzman, Simon, ‘Legitimidade, controvérsias e traduções em estatísticas públicas’, Teoria e Sociedade, 2 (1997), pp. 9–38 Google Scholar.

⁴⁸ De Oliveira, Brasil mostra a tua cara, p. 36.

⁴⁹ Ibid., p. 34.

⁵⁰ For example, Moema Teixeira recalls that Isaac Kerstenetzky was ‘a very important president for IBGE … [who] maintained IBGE's independence in moments of conflict with the government’: authors’ interview with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil.

⁵¹ A new idea in US policy circles, social indicators explicitly linked data collection to social policy. See Cobb, Clifford and Rixford, Craig, Lessons Learned from the History of Social Indicators, Vol. 1 (San Francisco, CA: Redefining Progress, 1998)Google Scholar, who quote Denis Johnston on p. 9: there was a ‘growing perception by policy makers and their advisors that the nation's rich array of economic statistics and related measures were simply inadequate indicators of emerging developments and issues under prevailing conditions of rapid social change and severe social strains’.

⁵² Authors’ interview with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil.

⁵³ Costa, ‘Considerações teóricas’, p. 173.

⁵⁴ Costa argued: ‘The lack of information is what explains the nonexistence of particular policies, or the irrationality and unsuccessfulness of others’: ibid., p. 173.

⁵⁵ Ibid., p. 171.

⁵⁶ Loveman, National Colors.

⁵⁷ Costa, Tereza Cristina, ‘O princípio classificatório “cor”, sua complexidade e implicações para um estudo censitário’, Revista Brasileira de Geografia, 36: 3 (1974), p. 100 Google Scholar.

⁵⁸ Authors’ interview with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil.

⁵⁹ Nobles, Shades of Citizenship, p. 117.

⁶⁰ The 1982 and 1986 PNAD surveys included pre-codified colour inquiries using the census categories, and after 1987 the basic PNAD questionnaire included the census colour inquiry. See de Oliveira, Brasil mostra a tua cara, p. 35.

⁶¹ Harris, Marvin, Consorte, Josildeth Gomes, Lang, Joseph and Byrne, Brian, ‘Who are the Whites? Imposed Census Categories and the Racial Demography of Brazil’, Social Forces, 72: 2 (1993), pp. 451–62Google Scholar; Telles, Edward, ‘Who are the Morenos?’, Social Forces, 73: 4 (1995), pp. 1609–11Google Scholar; and Harris, Marvin, Consorte, Josildeth Gomes, Lang, Joseph and Byrne, Brian, ‘A Reply to Telles’, Social Forces, 73: 4 (1995), pp. 1613–14Google Scholar.

⁶² Carlos Hasenbalg, ‘Race Relations in Post-Abolition Brazil: The Smooth Preservation of Racial Inequalities’, PhD diss., University of California, Berkeley, 1979; and do Valle e Silva, Nelson, ‘Updating the Cost of not being White in Brazil’, in Fontaine, Pierre-Michel (ed.), Race, Class and Power in Brazil (Los Angeles, CA: UCLA Center for Afro-American Studies, 1985), pp. 42–55 Google Scholar.

⁶³ Wade, Peter, Race and Ethnicity in Latin America (London, UK: Pluto Press, 1997)Google Scholar.

⁶⁴ The title of their report means ‘the place of the negro in the labour force’: de Oliveira, Lúcia Elena Garcia, Porcaro, Rosa Maria and Araújo, Tereza Cristina N., O lugar do negro na força de trabalho (Rio de Janeiro: IBGE, 1985)Google Scholar.

⁶⁵ de Oliveira, Nelson Fernando Inocêncio, Consciência negra em cartaz (Brasília: Editora Universidade de Brasília, 2001), p. 85 Google Scholar.

⁶⁶ Authors’ interview with Nilza Pereira, 15 Aug. 2007, Rio de Janeiro, Brazil (digital recording in possession of authors).

⁶⁷ Nobles, Shades of Citizenship, p. 122.

⁶⁸ de Oliveira, João Pacheco, ‘Entrando e saindo da “mistura”: os indígenas nos censos nacionais’, in de Oliveira, João Pacheco (ed.), Ensaios em antropologia histórica (Rio de Janeiro: UFRJ, 1999), pp. 124–54Google Scholar.

⁶⁹ As noted by other authors this change was directly related to the Indigenous and Tribal Peoples Convention no. 169 (ILO 169), promulgated in June 1989 by the International Labor Organization. Even though Brazil did not ratify it until later, they may have recognised 169's call for collecting information on indigenous people, considered the minimal recommendation. Loveman, National Colors; and Paschel, Becoming Black Political Subjects.

⁷⁰ Perz, Stephen, Warren, Jonathan and Kennedy, David P., ‘Contributions of Racial-ethnic Reclassification and Demographic Processes to Indigenous Population Resurgence: The Case of Brazil’, Latin American Research Review 43: 2 (2008) pp. 7–33 Google Scholar; IBGE, Tendências demográficas: uma análise dos indígenas com base nos resultados da amostra dos censos demográficos 1991 e 2000 (Rio de Janeiro: IBGE, 2005)Google Scholar.

⁷¹ The Census Advisory Committee was a committee of scholars, statisticians and IBGE officials convened to advise IBGE on the design and implementation of the 2000 census.

⁷² Authors’ interview with Elisa Reis, 29 Aug. 2007, Rio de Janeiro, Brazil (digital recording in possession of authors).

⁷³ Htun, Mala, ‘From “Racial Democracy” to Affirmative Action’, Latin American Research Review, 39: 1 (2004), p. 75 Google Scholar.

⁷⁴ For example, Lima, Márcia, ‘Desigualdades raciais e políticas públicas: ações afirmativas no governo Lula’, Novos Estudos –CEBRAP, 87 (2010), pp. 77–95 Google Scholar and Paschel, Becoming Black Political Subjects.

⁷⁵ Although Cardoso only became president in 1995, he served as Minister of Finance for Itamar Franco's government from 1993 on. This position allowed him to guide many political decisions and appointments.

⁷⁶ Authors’ interview with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil.

⁷⁷ Reis does not believe that changing the classification system was ‘among [Cardoso's] high priorities’: authors’ interview with Elisa Reis, 29 Aug. 2007, Rio de Janeiro, Brazil. Schwartzman resisted the idea that explicit political pressure prompted the re-examination of the classification regime. He recalls, ‘I never had a formal request to change [the system], but this was in the air, and some people were suggesting that we should do that’: authors’ interview with Simon Schwartzman, 21 Aug. 2007, Rio de Janeiro, Brazil (digital recording in possession of authors).

⁷⁸ Authors’ interview with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil.

⁷⁹ Authors’ interview with Simon Schwartzman, 21 Aug. 2007, Rio de Janeiro, Brazil.

⁸⁰ The PME surveyed six metropolitan regions (Belo Horizonte, Porto Alegre, Recife, Rio de Janeiro, Salvador and São Paolo) and included close to 90,000 people aged ten years and above.

⁸¹ Schwartzman, Simon, ‘Fora de foco: diversidade e identidades étnicas no Brasil’, Novos Estudos–CEBRAP, 55 (1999), pp. 8–96 Google Scholar.

⁸² Ibid.

⁸³ The conclusion was that Brazilians understand their race as phenotype not origin, following Oracy Nogueira's analysis from the 1950s. See Nogueira, Oracy, ‘Preconceito de marca e preconceito de origem: sugestão de um quadro de referência para a interpretação do material sobre relações raciais no Brasil’, in Bastide, Roger and Fernandes, Florestan (eds.), Relações raciais entre negros e brancos em São Paulo (São Paulo: Anhembi, 1955)Google Scholar. This question of origins also relates to the current debate on genetic studies in Latin America. See Wade, Peter, Beltrán, Carlos López, Restrepo, Eduardo and Santos, Ricardo Ventura (eds.), Mestizo Genomics: Race Mixture, Nation, and Science in Latin America (Durham, NC: Duke University Press, 2014)Google Scholar.

⁸⁴ Authors’ interview with Benedita da Silva, 3 Sept. 2007, Rio de Janeiro, Brazil (digital recording in possession of authors).

⁸⁵ Ibid.

⁸⁶ In response to da Silva's legislation, the Ministry of Justice convened a meeting that included IBGE President Schwartzman. Schwartzman regarded the proposal to assign every Brazilian a racial ID card as ‘a terrible idea’. He conveyed his feelings to the Minister of Justice and wrote a newspaper op-ed arguing against the proposal and defending the legitimacy of IBGE's classification system: authors’ interview with Simon Schwartzman, 21 Aug. 2007, Rio de Janeiro, Brazil.

⁸⁷ Paschel, Becoming Black Political Subjects.

⁸⁸ For a more detailed discussion of these programmes, and of how beneficiaries interpreted the racial categories as they were employed, see Schwartzman, Luisa, ‘Seeing Like Citizens: Unofficial Understandings of Official Racial Categories in a Brazilian University’, Journal of Latin American Studies, 41: 2 (2009), pp. 221–50Google Scholar.

⁸⁹ Paschel, Becoming Black Political Subjects.

⁹⁰ Kamel, Ali, Não somos racistas: uma reação aos que querem nos transformar numa nação bicolor (Rio de Janeiro: Nova Fronteira, 2006)Google Scholar.

⁹¹ Authors’ interview with José Luis Petruccelli, 15 Aug. 2007, Rio de Janeiro, Brazil.

⁹² Independent survey data from the 2000s suggested that increasing numbers of Brazilians chose the negro category in open-ended questions about their race or colour, but these numbers were still far below the 40–50 per cent of the population who chose preto or pardo when given the census options. For a study on black identification in Brazil and recent changes, see Sansone, Livio, Blackness without Ethnicity: Constructing Race in Brazil (New York: Palgrave Macmillan, 2003)Google Scholar.

⁹³ Authors’ interviews with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil, and José Luis Petruccelli, 29 July 2010, Rio de Janeiro, Brazil.

⁹⁴ The PCERP covered 25,000 households in six metropolitan areas. IBGE, Características étnico-raciais da população – um estudo das categorias de classificação de cor ou raça 2008 (Rio de Janeiro: IBGE, 2011)Google Scholar. Available at http://biblioteca.ibge.gov.br/visualizacao/livros/liv49891.pdf (accessed 6 June 2017)

⁹⁵ Introduced in the early 2000s by PT Senator Paulo Paim, the statute languished for years. The version that was finally passed lacked racial equality benchmarks and affirmative action quotas for an array of issue areas from education and health to sports and the arts, all features of the original version. However, a 2012 law made socio-economic quotas, with racial sub-quotas, mandatory in all federal universities (generally the most prestigious universities in the country).

⁹⁶ For a good discussion about the immigration laws of the early twentieth century and other ways in which law has traditionally reinforced race inequality in Latin America see Hernández, Tanya Katerí, Racial Subordination in Latin America: The Role of the State, Customary Law, and the New Civil Rights Response (New York: Cambridge University Press, 2012)Google Scholar.

⁹⁷ Full text of the law accessed online: http://www.seppir.gov.br/portal-antigo/Lei%2012.288%20-%20Estatuto%20da%20Igualdade%20Racial.pdf (accessed 6 June 2017).

⁹⁸ Telles, Edward, Pigmentocracies: Ethnicity, Race and Color in Latin America (Chapel Hill, NC: North Carolina University Press, 2014)Google Scholar.

⁹⁹ Paschel, Becoming Black Political Subjects.

¹⁰⁰ Schneider, ‘The Career Connection’, p. 331.

¹⁰¹ Ibid, p. 345.

¹⁰² Loveman, National Colors.

¹⁰³ Bailey, ‘Unmixing for Race Making in Brazil’.

¹⁰⁴ Sansone, Blackness without Ethnicity.

¹⁰⁵ Indeed, scholars outside IBGE have debated whether it represents a viable census category. See Harris et al., ‘Who are the Whites?’, Telles, ‘Who are the Morenos?’, and Harris et al., ‘A Reply to Telles’.

¹⁰⁶ Authors’ interview with Moema Teixeira, 27 Aug. 2007, Rio de Janeiro, Brazil.

¹⁰⁷ Loveman, Mara, Muniz, Jeronimo O. and Bailey, Stanley R., ‘Brazil in Black and White? Race Categories, the Census, and the Study of Inequality’, Ethnic and Racial Studies, 35: 8 (2012), pp. 1466–83Google Scholar.

¹⁰⁸ Muniz, Jerônimo Oliveira, ‘Sobre o uso da variável raça–cor em estudos quantitativos’, Revista de Sociologia e Política 18: 36 (2010)Google Scholar; and ‘Inconsistências e consequências da variável raça para a mensuração de desigualdades’, Civitas–Revista de Ciências Sociais 16: 2 (2016), pp. 62–86 CrossRefGoogle Scholar.