DOI: 10.60625/risj-vt3v-wx80

Key findings

In this Reuters Institute factsheet continuing the work we began in 2020, we analyse the percentage of people of colour in top editorial positions in a strategic sample of 100 major online and offline news outlets in five different markets across four continents: Brazil, Germany, South Africa, the United Kingdom (UK), and the United States (US).

Looking at a sample of ten top online news outlets and ten top offline news outlets in each of these markets, we find:

• Overall, 17% of the 70 top editors across the 100 brands covered are people of colour, despite the fact that, on average, 44% of the general population across all five countries are people of colour. This is down six percentage points from last year and is one percentage point lower than when we first started collecting these data in 2020.
• As in previous years, in every single country covered, the percentage of top editors of colour remains below – often far below – the percentage of people of colour in the general population. If we set aside South Africa – which pushes up the topline figure in the previous bullet point – and look at the four other countries covered, only 4% of the top editors are people of colour, compared with, on average, 32% of the general population.
• In Brazil, Germany, and the UK, none of the outlets in our sample have a person of colour as top editor. In South Africa, the percentage of top editors of colour dropped from 71% in 2024 to 63% in 2025. In the US, the percentage of top editors of colour also decreased – to 15%, compared with 29% last year.
• In the four countries where data are available on the number of journalists of colour, there continues to be no simple relationship between the percentage of journalists of colour and the percentage of people of colour in top editorial positions. In Brazil, and to a lesser degree the UK, there are fewer top editors of colour than there are journalists of colour. Despite the decreases in South Africa and the US, there are still more top editors of colour than journalists of colour in both countries.
• The overall share of internet news users who say that they read news from at least one major outlet with a top editor of colour has slightly decreased from 27% to 23% and ranges from 0% in Brazil, Germany, and the UK, where no editors in the sample are people of colour, to 85% in South Africa.

General overview

Who occupies top editorial positions in major news organisations is of both symbolic and practical importance. Symbolically, people in these roles often represent the news organisations they lead, and the media industry at large, to the public. As key figures in the leadership of news outlets, top editors model what journalism can and should be (Perreault and Tham 2023). Practically, they make decisions that influence what news and newsrooms look like, and their backgrounds and experiences are among the factors that influence such decisions (Duffy 2021). The racial and ethnic composition of newsroom management can shape diversity in hiring, retention, and promotion; the professional culture in newsrooms; and news content itself, influencing editorial decision-making and attention to stories and experiences that reflect – well or poorly – the communities that news organisations seek to serve (Jenkins and Powers 2023; Richardson 2022; Cruz and Holman 2022).

These premises are central to our ongoing work studying diversity in leadership in the news media, both from the perspective of gender (Ross Arguedas et al. 2025) and race (Ross Arguedas et al. 2024). We continue our work on race and ethnicity in news leadership here, documenting the diversity of top editors of major outlets to assess one aspect of how the news media industry has responded to conversations about racial justice, equity, and representation in many countries.

It’s worth underscoring how, in the six years since the murder of George Floyd and the subsequent resurgence of the Black Lives Matter movement, the public discourse surrounding diversity, equity, and inclusion (DEI) efforts has in many places shifted from a sense of urgency to fading interest and, increasingly, explicit hostility, encouraged in part by right-wing public figures such as the US President Donald Trump and Elon Musk.¹

Even in organisations still sympathetic to DEI, diversity is unlikely to improve without sustained and concerted efforts. In practice, news organisations are juggling many competing priorities and face talent pipeline issues as long as they recruit top leaders from a profession that has long underrepresented people of colour on the upper rungs. One recent report about diversity inside media organisations found that while many organisations now have diversity among their goals, it typically ranks low compared to other commercial priorities, especially in legacy publications (FT Strategies 2024).

This short factsheet contributes to important research by others working on race in the news media, including work gauging journalists’ perceptions of newsroom diversity (Boaventura Teixeira 2022; Jastrzebski and Willnat 2024), as well as studies documenting the challenges of underrepresentation, discrimination, and hostility faced by journalists of colour in predominantly white newsrooms (for example, Al-Kaisy 2023; Bhowmik et al. 2024; Hull et al. 2024).

Methods and data

Continuing our work from previous years (Eddy et al. 2023; Eddy et al. 2022; Nielsen et al. 2020; Robertson et al. 2021; Ross Arguedas et al. 2024), we focus on top editorial leadership positions in a strategic sample of five markets: South Africa in Africa; Germany and the UK in Europe; the US in North America; and Brazil in South America. These countries have different demographics and histories of white imperialism, colonialism, and slavery. To get an overview of similarities and differences across markets, as well as any changes in leadership over time, we examine the same countries in 2025 as we did in previous years. This choice of markets also allows us to leverage available data on the journalistic profession and on news and media use. We include five markets from those covered in Worlds of Journalism (Hanitzsch et al. 2019) and in the Reuters Institute Digital News Report 2024 (Newman et al. 2024). (While the Digital News Report provides fresh data each year, by now the available Worlds of Journalism data are, in some cases, over ten years old.)²

Where possible, we compare our findings to data on racial and ethnic diversity in the journalistic profession and in the general population, using the most recent census and other official data. However, data are not always available and comparisons not always possible, not only because of limited research, but also because of legal restrictions on collecting and retaining certain statistics on race (for instance, in Germany).

Following work on the representation of race in editorial positions (Chakravartty et al. 2018) and citation patterns (Freelon et al. 2023) in communications research, we operationalised race by adopting a conceptualisation that contrasts institutionally dominant white populations and dominated racialised communities as a means of calling attention to the reification of racial categories as part of structural inequality and the exercise of power. This language and history differs across the countries we study but what we are specifically referring to here are present-day hierarchies tied to colonial legacies of empire, militarism, and Apartheid. This year, and consistent with the shift we made in 2023, we have opted to use the terminology ‘people of colour’ instead of ‘non-white’ throughout our report, in order to move away from references to a white default and to avoid terminology that in the South African context is associated with Apartheid.

As in the past, we want to be clear: race and racial discrimination work in complex ways not always tied to skin colour; for example, where it has a religious dimension. They are the product of what has been defined as ‘racecraft’: categories in part produced by racist practices that in turn shape society and how we see it (Fields and Fields 2012). There are also dimensions of ethnicity that are not always related to conceptions of race. However, the conceptualisation we use here captures some important aspects of this in the countries we cover.

We therefore deploy a simple and reductionist, but hopefully still illuminating and relevant, binary and code each top editor based on whether they are a person of colour or not. This coding was based on editors’ self-identification when publicly available as well as their biographies and images. Relying on physical appearance and biographies to analyse race is imperfect, as related research has noted (Freelon et al. 2023), so we also drew upon local partners’ expertise on differing contexts of race and ethnicity in each market we study here. Using the terminology ‘people of colour’ is in no way meant to suggest an identity, let alone a homogeneous group, given the great diversity and complexity of people’s identities. Rather, it provides a way to categorise otherwise very different people who come from institutionally dominated ethnic and racial groups. It helps us point to a dimension of inequality in representation at a macro level.

Our overall approach is identical to that of previous years. In each market, we focused on the top ten offline (TV, print, and radio) and top ten online news brands in terms of weekly usage, as measured in the Reuters Institute Digital News Report 2024 (Newman et al. 2024). When compiling the top ten lists, we only included specific brands, leaving out catch-all categories, such as ‘local news’ or ‘regional news’. We also excluded foreign news outlets that do not have newsrooms within the country of interest. Adhering to these lists also means that news aggregator brands such as Yahoo! News are included in the analysis if they: (1) are widely used in that market; (2) have local teams based there; and (3) produce at least some original content.

Our focus on the most widely used offline and online brands means that some important outlets with editors of colour are not included in the sample (in the UK, for example, the Financial Times, edited by Roula Khalaf, is not in the sample). Because of changes in what the most widely used brands are, and our focus on the top ten offline and online brands, there has been some turnover in the specific brands included in the analysis: 93 of the 100 brands covered in 2024 are included in the analysis again this year, with seven new brands included.

For each brand, we identified the top editor by checking their official webpages, press releases, and related news coverage, alongside other public information, including professional social media accounts, such as LinkedIn and X. The data were collected in late January and early February 2025. We looked for editor-in-chief or the nearest equivalent; for example, executive editor, or head of news for TV. The exact terminology varies from country to country and organisation to organisation, but in most cases it is possible to identify a single person. Exceptions to this can be found in Germany, where organisations sometimes have shared top leadership. Three such cases with joint top editors were present in this year’s data, but this did not impact the coding because all were white.

In some instances, brands on our lists belong to (or have been absorbed by) larger media groups or conglomerates. In such cases, we privileged the name of the person in the top leadership role at the level of the brand listed – rather than the entire group – or in the case of some media groups in South Africa with leadership organised around medium (for example radio or broadcast) rather than brand, at the medium level. Likewise, when there appeared to be clear convergence between the offline and online version of a brand, with a single person steering the ship, we used the same top editor for both. However, when we identified a different person in charge of the online version of a brand, and especially when we understood this to be tied to distinct content and distinct editorial teams or decision-making, we used different names for the online and offline versions of the brand.

We refer to the individuals identified collectively as the top editors. It is important to note, of course, that this does not mean the top editor is the only person who matters or, in fact, is the most important person in terms of day-to-day editorial decision-making. We coded one observation as missing in cases where both online and offline versions of the same brand share a top editor as well as where two brands with the same ownership share a top editor, to avoid double-counting. Together, the analysis covers a total of 70 individuals across the 100 brands included in 2025, compared with 75 individuals in 2024. Some top editors had publicly or privately announced they were stepping down at or around the time of data collection. In these cases, where no replacement had been publicly announced by mid-February 2025, we chose to keep the outgoing top editor listed as is, and we include here the top editor as of late February.

The individuals identified were double-checked in consultation with local partners within every market, including current and former Reuters Institute Journalist Fellows as well as academic experts. In some cases, we also contacted the brands or their press offices to confirm who their top editor was. Where organisations responded, we always deferred to their judgement.

In a constantly evolving media environment where many news organisations offer little or no transparency about who is actually in charge, it sometimes requires a qualitative judgement to determine who the top editor is. Indeed, in some cases where there is no single, clearly designated editor-in-chief, or roles and responsibilities across online and offline parts of the same outlet are unclear, we have made a judgement call as to who to code as the top editor of the outlet in question. We have tried to be as clear and consistent as possible about the criteria used to code an opaque and inconsistent world. The primary point of the factsheet is to capture the overall pattern, even if in some cases individuals could have been coded differently.

All individuals in the dataset were coded independently by the authors. Race and ethnicity are complicated phenomena, and so are statistics on race and ethnicity. What we present here is based on our coding of the individuals covered. Top editors may not in every instance see themselves in the same way, or always be seen by everyone else in the same way, though we prioritised editors’ self-identification when this information was publicly available. The numbers presented here, both from our own data collection and from secondary sources we rely on, should be seen with this in mind.

Findings

Based on this dataset, we find that only 17% of the 70 top editors across the 100 brands covered are people of colour. Whereas last year we noted stagnation in the overall trend (following small increases between 2021–2022 and 2022–2023), this year we see a reversal, as our overall figure is six percentage points lower than the 23% we documented last year and is similar to our 2020 figure, when 18% of top editors were people of colour. This is the largest drop we have documented from one year to the next since we first started collecting our data. As in previous years, it is substantially below the, on average, 44% of people in the general population across all five markets who identify as people of colour (based on census data and other official estimates).³ It is also below the average percentage of journalists of colour (21%) in the four countries where we have available data (Hanitzsch et al. 2019).

The shift in this year’s top-line figure is driven both by personnel changes and turnover in the brands included for 2025. If we look exclusively at the 93 brands covered in both 2024 and 2025, 18% of the top editors in 2025 are people of colour, which is five percentage points lower than the same figure last year. Of the 11 new top editors identified and coded among these repeat brands only two, both in South Africa, are people of colour. Meanwhile, none of the top editors of colour among new brands included this year were people of colour, far below the 18% for this same data point last year.⁴

As is clear from Figure 1, the percentage of top editors of colour varies considerably across the five markets analysed. In Brazil and Germany, as has been the case since 2021, none of the outlets we cover has a person of colour as top editor. This year, they are joined by the UK, which has returned to 0% top editors of colour in the sample year, down from 7% last year, after the only brand with an editor of colour in the sample slipped off the list of brands with the highest reach.

In the other two countries, where the increases we documented between 2021 and 2023 seemed to be tapering off last year, this year we find decreases across the board. South Africa, while still having the largest proportion of people of colour in top editorial positions, has seen a decrease from 71% in 2024 (and 80% in 2023) to 63% this year, driven by one personnel change and one new brand entering the top lists. This brings South Africa below its 2020 number (68%). Likewise, in the US the percentage of top editors of colour decreased to 15% from 29% in 2024 (and 33% in 2023), because of a reduction in the number of editors of colour (from four to two) and the slightly smaller base this year (13 editors total compared to 14 last year). Across the board, the figures for this year in all five countries have nearly returned to their 2020 levels (slightly above in the US and slightly below in South Africa).

Figure 1.

If we look at our data intersectionally and examine the gender composition of the subsample of editors of colour, we see that in South Africa 60% of the editors of colour are men – closer towards gender parity than last year, when the figure was 67%, and the previous year, when it was 75%. This shift was the result of two personnel changes. Meanwhile, in the US we see a change in the opposite direction, from 75% of men editors of colour to 100%, also resulting from new incoming top editors. This further extends the shift we noted last year, when the percentage of men among the top editors of colour increased from 50% to 75%. However, jumps in the data are large in the US due to the overall low (and, as noted above, shrinking) number of editors of colour. Nonetheless, these market-level differences underscore the importance of tracking intersectional disparities to better understand structural inequalities in news leadership and the media industry more broadly (Banjac 2022; Lachover 2022).

Relying on data from the Worlds of Journalism project collected between 2012 and 2016 (see Hanitzsch et al. 2019), shown in Figure 2, we compare the relationship between the percentage of people of colour in top editorial positions across four of the five markets and the percentage of people of colour working in journalism. (Data on the racial identity of journalists in Germany are not available.) Here, as in previous years, we find a mixed picture. The percentage of editors of colour is considerably higher than the percentage of journalists of colour in South Africa (63% vs. 34%) and to a lesser extent the US (15% vs. 9%). Meanwhile, we see the opposite pattern in Brazil, and to a lesser degree the UK, where the proportion of editors of colour (0% in both places), is below the 34% and 6% of journalists of colour, respectively.

Figure 2.

In Figure 3, we plot data on the percentage of top editors of colour in each market relative to the demographics of the population as a whole. The marked disparities and overrepresentation of white people among top editors across all five countries, which we had identified in previous years, have been magnified this year, in line with the overall decrease in top editors of colour.

As with last year, the gap is largest in Brazil, where people of colour make up 57% of the population yet there is not a single top editor of colour. In South Africa, the country with the highest percentage of editors of colour in our sample, we still find a 30 percentage point (pp) gap relative to the percentage of people of colour in the general population (63% versus 93%), which is up from the 21pp difference last year. And in the US, at 15%, the percentage of top editors of colour remains 28pp below the 42% of people of colour (including Hispanic/Latinx) in the general population. Setting aside South Africa, which is something of an outlier relative to the other countries analysed, we can see that just 4% of top editors across Brazil, Germany, the UK, and the US are people of colour, compared with, on average, 32% of the general population across these markets.

Figure 3.

Finally, by combining the data collected for this Reuters Institute factsheet with the Reuters Institute Digital News Report 2024 (Newman et al. 2024) data, we can identify the proportion of people in each market who access news from at least one major news outlet with a person of colour as top editor. In Figure 4, we show the share of online news consumers who say that they got news from at least one major outlet (online or offline) with a top editor of colour. The average across all five countries, at 23%, is slightly lower than the 27% we found last year. However, the numbers vary considerably from country to country. Audiences in Brazil, Germany, and the UK accessed no news in the past week from a major outlet included in our sample and edited by a person of colour, for the simple reason that there are no people of colour in top editorial positions among the outlets we cover. In the US, in 2025, 28% of online audiences used at least one source with a top editor of colour, far below the 42% that did so in 2024. In South Africa, the figure dipped 5pp from 90% in 2024 to 85% in 2025.

Figure 4.

Conclusion

In this Reuters Institute factsheet we have analysed the racial breakdown of top editors in a strategic sample of 100 major online and offline news outlets in five different markets across four continents. While we found very little change in 2024, this year we see decreases in the proportion of top editors of colour across all three markets that weren’t already at 0%, resulting in the largest decline in the top-line figure since we started collecting the data. As in previous years, we continue to find that people of colour are significantly underrepresented, and white people are significantly overrepresented as top editors, relative to their share of the general population in all five countries.

Furthermore, in countries where we had seen some indication of growth in recent years, following reckonings around racial and ethnic injustice – particularly in the US, South Africa, and the UK – this year the gains seem to have slipped away, as the figures in all five markets have returned to levels similar to those of 2020. Beyond the stagnation we identified in 2024, this year we find some evidence of backsliding, although more data over time will make clear the extent to which this is an ongoing trend.

As we alluded to in the opening of this factsheet, these findings emerge against the backdrop of a rapid cooling of and, in some quarters, growing backlash against DEI initiatives, epitomised by US President Donald Trump’s recent executive order halting DEI programmes across the US government, followed by the rollback of such policies at large US-based tech and media companies.⁵ Last year we noted that sceptics have criticised the DEI boom of the early 2020s as more performative than principled,⁶ and policies that in the immediate aftermath of the Black Lives Matters uprisings may have reaped companies reputational rewards may now be seen as having a limited, or even contrary, effect among some segments of the public. This may also be the case with dominant political elites, who sometimes seek to weaponise DEI discussions for their own purposes. To the extent that efforts to address disparities within organisations – in general and in the media more specifically – have been motivated by public pressure, and to the extent that public support of DEI weakens and/or becomes more polarised, commitments to DEI may continue to wane.

At the same time, failing to address disparities within newsrooms can also erode news organisations’ relationships with audiences from marginalised or underserved groups who, as previous research has shown, have long expressed grievances around how their communities are portrayed in the news media. The belief that news is not made by or for people like them – or worse yet, is complicit with systems of oppression – can contribute to low levels of trust in news (Ross Arguedas et al. 2023; Peterson-Salahuddin 2023). As a special report for the Columbia Journalism Review put it: ‘Ultimately, the value of diversity to journalism is not about skin colour, gender, sexual orientation, or social class. It’s about the stories people can tell.’ (Arana 2018).

We will know more when we repeat this analysis next year and publish new data tracking developments in race equality among top editors across the world.

References

• Al-Kaisy, A. 2023. Structural Racism in UK Newsrooms. London: Ethical Journalism Network.
• Arana, G. 2018. ‘Decades of Failure’, Columbia Journalism Review, 5 November,  (Accessed Mar. 2025).
• Banjac, S. 2021. ‘An Intersectional Approach to Exploring Audience Expectations of Journalism’, Digital Journalism, 10(1), 128–147. https://doi.org/10.1080/21670811.2021.1973527
• Bhowmik, S., Smith, M., Wenger, D. 2024. ‘Then and Now: Newsroom Diversity Matters for Mississippi’s Black Journalists’, Journalism Practice, 1–19. https://doi.org/10.1080/17512786.2024.2418912
• Boaventura Teixeira, L.F. 2022. The Human Impact of Racial Representation in Brazilian Newsrooms. Oxford: Reuters Institute for the Study of Journalism.
• Chakravartty, P., Kuo, R., Grubbs, V., McIlwain, C. 2018. ‘#CommunicationSoWhite’, Journal of Communication, 68(2), 254–66.
• https://doi.org/10.1093/joc/jqy003
• Cruz, C. Y., Holman, L. 2022. ‘The Media and Race in the Trump Era: An Analysis of Two Racially Different Newsrooms’ Coverage of BLM and DACA’, Howard Journal of Communications, 33, 197–215. https://doi.org/10.1080/10646175.2021.2012853
• Duffy, A. 2021. ‘Out of the Shadows: The Editor as a Defining Characteristic of Journalism’, Journalism, 22, 634–49.
• https://doi.org/10.1177/1464884919826818
• Eddy, K., Selva, M., Nielsen, R. K. 2022. Race and Leadership in the News Media 2022: Evidence from Five Markets. Oxford: Reuters Institute for the Study of Journalism
• Eddy, K., Ross Arguedas, A., Mukherjee, M., Nielsen, R. K. 2023. Race and Leadership in the News Media 2023: Evidence from Five Markets. Oxford: Reuters Institute for the Study of Journalism.
• Fields, K. E., Fields, B. J. 2012. Racecraft: The Soul of Inequality in American Life. London: Verso.
• Freelon, D., Pruden, M. L., Eddy, K. A., Kuo, R. 2023. ‘Inequities of Race, Place, and Gender Among the Communication Citation Elite, 2000–2019’, Journal of Communication, 73(4), 356–67. https://doi.org/10.1093/joc/jqad002
• FT Strategies. 2024. ‘Diversity Inside Media Organisations: 2024 Trends’, (Accessed Mar. 2025).
• Hull, K., Walker, D., Romney, M., Pellizzaro, K. 2024. ‘“Through Our Prism”: Black Television Sports Journalists’ Work Experiences and Interactions with Black Athletes’, Journalism Practice, 18(3), 511–28. https://doi.org/10.1080/17512786.2022.2050468
• Jastrzebski, S., Willnat, L. 2024. ‘Attitudes of US Public Broadcasters: A Liberal Helping of Interpretive Journalism’, Electronic News, 18(3), 139–68. https://doi.org/10.1177/19312431241246239
• Jenkins, J., Powers, E. 2023. ‘Listening, Reflecting, and Protecting: Empathy as a Priority for Local Journalists Covering Underserved Communities’, Journalism Practice, 1–18.
• https://doi.org/10.1080/17512786.2023.2202648
• Lachover, E. 2022. ‘Ethiopian-Israeli Women Working in Mainstream Journalism: Linking Ethiopian, Black and Israeli Identities through News-Making’, Journalism, 23(10), 2041–59. https://doi.org/10.1177/14648849211018599
• Nielsen, R. K., Selva, M., Andı, S. 2020. Race and Leadership in the News Media 2020: Evidencefrom Five Markets. Oxford: Reuters Institute for the Study of Journalism.
• Perreault, G. P., Tham, S. M. 2023. ‘Effective Leadership in Journalism: Field Theory in How Journalists Evaluate Newsroom Leadership’, Journalism, 25(11) 2283–2300, https://doi.org/10.1177/14648849231222694
• Peterson-Salahuddin, C. 2023. ‘Trust Signals: An Intersectional Approach to Understanding Women of Color’s News Trust’, Media and Communication, 11(4), 332–43 http://dx.doi.org/10.17645/mac.v11i4.7106
• Richardson, R. J. 2022. ‘Local TV Newsroom Diversity: Race and Gender of Newscasters and Their Managers’, Journal of Broadcasting & Electronic Media, 66, 823–42. https://doi.org/10.1080/08838151.2022.2121834
• Robertson, C. T., Selva, M., and Nielsen, R. K. 2021. Race and Leadership in the News Media 2021: Evidence from Five Markets. Oxford: Reuters Institute for the Study of Journalism.
• Ross Arguedas, A., Banerjee, S., Mont’Alverne, C., Toff, B., Fletcher, R., Nielsen, R. K. 2023. News for the Powerful and Privileged: How Misrepresentation and Underrepresentation of Disadvantaged Communities Undermines Their Trust in News. Oxford: Reuters Institute for the Study of Journalism.
• Ross Arguedas, A., Mukherjee, M., Nielsen, R. K. 2024. Race and Leadership in the News Media 2024: Evidence from Five Markets. Oxford: Reuters Institute for the Study of Journalism.
• Ross Arguedas, A., Mukherjee, M., Nielsen, R. K. 2025. Women and Leadership in the News Media 2025: Evidence from 12 Markets. Oxford: Reuters Institute for the Study of Journalism.

Acknowledgements

The authors would like to thank Luiz Fernando Boaventura Teixeira, Kirsten Eddy, Nic Newman, Paul Herman, and Anne Schulz for their valuable time, input, and feedback.

About the authors

Amy Ross Arguedas is a Research Fellow at the Reuters Institute for the Study of Journalism.

Mitali Mukherjee is the Acting Director and Director of the Journalist Programmes at the Reuters Institute for the Study of Journalism.

Rasmus Kleis Nielsen is a Professor at the Department of Communication of the University of Copenhagen and a Senior Research Associate at the Reuters Institute for the Study of Journalism.

This report can be reproduced under the Creative Commons licence CC BY.