Thomas Piketty is one of the most influential living social scientists. His first book, Capital in the Twenty-First Century, was translated into numerous languages, sold millions of copies, and has reshaped the way we understand and debate inequality [2014]Footnote ¹. Appreciative of and involved in debates beyond economics, he has since ventured into studying the relationship between ideology and economic inequality [Piketty 2020]Footnote ² and has taken on controversial topics such as reparations for colonialism and slavery in recent work [Piketty 2022a]Footnote ³.

Given his stature and theoretical sophistication, it is surprising that Anglophone social scientists have paid little attention to a book Piketty published two years ago in French. The book, which has yet to be translated into English though some of the themes are explored in chapter 8 of A Brief History of Equality, explores ways to measure and ameliorate inequality related to national origin, ethnoracial identity, and religion in France. Mesurer le racisme, vaincre les discriminations (Measuring racism, overcoming discrimination) was published by Éditions Seuil in 2022 as part of a collection of short books written by public intellectuals to briefly address pressing public debates.Footnote ⁴ At only 72 pages, this book lays out Piketty’s vision of the problem of ethnoracial inequality in France, suggests how to potentially solve it, and presents a critique of the Anglo-Saxon tradition of ethnoracial categorization and measurement. As one of the world’s most important economists attempting to provide a new perspective on a topic that has previously been studied primarily by sociologists, anthropologists, and political scientists, it is worth considering carefully what he has to say.

Piketty’s main explanandum in the book is the puzzle that, “[…] no country has invented a perfect system permitting it to combat racism and discrimination” [front of the book]. Writing to what he imagines as a universalist French audience, perhaps suspicious of the possibility of tracking and quantifying ethnoracial identity in a country where no questions about origins are asked on the census, Piketty argues for a universalist system that can at once fight against racism while not “[…] freezing identities, which are always plural and multiple” [12].

Piketty starts by arguing that France can begin to remedy ethnoracial inequality by addressing broader social equality, since minority groups are often overrepresented in the popular classes [12-13]. He sees ethnoracial categories as being “always socio-racial categories,” and thus suggests that place-based inequalities in teacher salaries or local budgets are an important place to start in addressing ethnoracial inequality in France [25]. However, despite his insistence on the importance of creating equal access to state funding for those living in poorer areas, he notes that “this is unfortunately not sufficient” in a context where “certain origins are the specific object of particular discrimination” [25]. With this goal in mind, Piketty proposes that France should create a new “Observatoire national des discriminations” (National Discrimination Observatory) which would introduce an annual regime of audit testing. This would include audits of job applications to test how observed ethnoracial identity impacts on the probability of an individual receiving an interview. It would also repeat other experiments, such as a famous study of police ID checks in France that audited whether the police stop those of different observed ethnoracial identities at different rates [Jobard et al. 2012]Footnote ⁵. This authority would “objectify, quantify, and compare different forms of discrimination as they are perceived in society, and would become the official and incontestable barometer of discriminatory practices” [34].

Piketty suggests that this new authority and its testing regime would allow for a quantitative method of tracking and eliminating racial discrimination. However, the types of events that could be audited under this system would be limited. While we might be able to observe who receives a first-round interview, discrimination occurs in the interviews themselves, the workplace, and promotion decisions. Discrimination at these stages cannot be audited the same way. Thus, the author admits that there is a need for some type of administrative classification that can be used to track these types of inequalities over time, and suggests that the French state could collect information on parents’ country of origin on surveys like the census, allowing the state to study those in the second generation but not beyond. This is Piketty’s solution because he is worried that the state might lead to the “freezing of identities” by asking about ethnoracial identification and instead believes that asking about parental origin will be helpful and not intrusive [34].Footnote ⁶

Piketty argues that this method is reasonable because it would be surprising for an area to have discrimination against third and fourth generation immigrants from sub-Saharan Africa, but not against second generation immigrants. He is also critical of the US context, where he notes that self-identified ethnoracial classification has not led to racial equality [42-43]. Further, he argues that in contexts like Germany or France, where most immigration comes from Turkey or North Africa, respectively, there is less variation in physical appearance and also extensive intermarriage [44-45]. Piketty suggests that those who are the product of intermarriage, where each of their grandparents might be from a different country, could have trouble deciding whether to self-identify as Maghrebian, Black, Asian, Mixed Race, or White. Additionally, he uses research from 2006, which studied how different individuals felt about the use of ethnoracial self-identification in France, and suggests that north Africans felt particularly uncomfortable with these forms of self-identification.

Ultimately for Piketty, the question of introducing a US or British style system of ethnoracial identification is reduced to the question of “whether or not it will effectively fight against discrimination” [52]. For him, the answer is not so clear, and he believes that a national testing regime and administrative data on immigrant origin might be sufficient. Piketty does not want his position on these categories to be the last word on the topic, and calls for further debate about how to best fight discrimination in the French context [52]. He then makes a brief digression to discuss religious neutrality and religious tax breaks, before concluding with an impassioned argument for addressing issues of discrimination in a society beset with debates about identitarianism.

This book provides a generative theoretical perspective, by considering the statistical tracking of ethnoracial identity primarily as a method, among others, to reduce inequality, and in terms of its potential secondary impacts. These points have often been missed in US debates around these issues, where scholars frequently assume that we have nothing to learn from France’s current approach to studying ethnoracial inequality except for the dangers of administrative colorblindness. For instance, what is the long-term effect of an American marking race and ethnicity on every form they fill out, from their medical intake forms to surveys about television programing? Moreover, it is not always clear that when an ethnoracial group recognizes themselves as a coherent group and experiences discrimination, the best way to remedy that fact is to create a statistical category for that group on the census rather than a regime of testing like the one Piketty suggests. Ethnoracial categories used in administrative contexts can have unexpected consequences. Further, the author’s attempt at a pragmatic approach to this issue in France, where it is difficult to imagine ethnoracial categories being added to the census in the near future, is interesting and worthy of further exploration.

However, I believe Piketty may overstate the risk of “freezing” identities and French reticence in using ethnoracial identification. First, we need additional research on the effects of using ethnoracial categories on individuals, and the question of whether these categories “freeze identities” is up for debate. Just because these identities are salient in the United States, where they are used on the census, does not mean they would necessarily have a “freezing” effect in France. Second, the existing empirical work on reticence to respond to ethnoracial questions in France is less pessimistic than Piketty suggests. For example, he discusses the 2006 Simon and Clément study, which shows the use of ethnoracial identification in the census or scientific research is less controversial than the collection of these data in personnel files, and is more consistently approved of across groups [Simon and Clément 2006a]Footnote ⁷. Additionally, views on this topic in France have not remained static. A 2016 survey by the Defenseur des Droits, “Access Aux Droits,” asked an ethnoracial identification question, with only 15% of the sample refusing to respond to the question, fewer than the percentage Piketty cites from the earlier Simon and Clément study [Clément 2016: 85]Footnote ⁸. This Defenseur des Droits study has been critical for demonstrating ethnoracial inequality in France, although its impact has been limited because of its small sample (approximately 5,000 respondents) and because it has only been fielded once. Whether these questions would be refused in future iterations of such a survey is an empirical question worthy of further analysis. Additionally, in Simon and Clément’s original study, one respondent was perturbed by the ethnoracial questions because it was unclear “à quoi sert la réponse” (“what purpose the response serves”) [Simon and Clément 2006b]Footnote ⁹. A public education campaign about how these categories would be used could placate many French citizens and alleviate their concerns about collecting ethnoracial data.

Finally, there are important implications of this work for Piketty’s own project. Much of his research is based on having comparable historical data, centered around the production of his impressive World Inequality Database (WID). For him, a “debate without data” risks conceding an important topic like wealth inequality to research based on “[…] an abundance of prejudice and a paucity of fact” [Piketty 2014: 2]. Yet, what are the implications of missing the third generation of immigrations in administrative records or missing a group like those who identify as “Black” in France, since Black French individuals expand beyond the bounds of immigration origins, coming from places such as Brazil, the United States, and Belgium. Though the collection of these data has its flaws in the US context, it has allowed for exceptional work by scholars who have tracked racial wealth gaps since the civil war [Derenoncourt et al. 2023]Footnote ¹⁰, and who have shown the historical evolution of racial inequality in incarceration [Muller 2021]Footnote ¹¹. Just because racial inequalities still exist in the United States does not mean that having data to study them has not been helpful or productive.

Perhaps for Piketty, the social cost of formalizing these categories and risking “fixing” these identities is not worth these benefits.Footnote ¹² But these are important considerations for the future of collecting self-identified ethnoracial data beyond the second generation. If understanding changes in broad wealth concentration over time and place can help suggest policies to reduce these inequalities, would not the same also be true for racial inequalities in wealth? These are unsettled questions in a punchy book well worth reading for scholars interested in deepening their ways of thinking about ethnoracial statistics and reducing inequality.