Abstract
This article investigates whether Brazil’s regional press mirrors subnational electoral cleavages during the impeachment of President Dilma Rousseff. Drawing on debates about political and regional parallelism, we compare editorials from two leading state newspapers located in electorally opposed contexts: O POVO is located in the state of Ceará, a Workers Party (PT) stronghold and Gazeta do Povo in Paraná, a predominantly anti-PT state. Our corpus comprises 60 editorials published between March 2015 and August 2016. Using content analysis, we coded editorial framings in terms of problem definition, causal attributions, consequences, those deemed responsible, proposed solutions, and explicit stances on the impeachment of Rousseff. We test two competing hypotheses: (H1) regional parallelism, whereby newspapers align with their state’s prevailing partisan preferences; and (H2) a “default” hypothesis, whereby regional outlets reproduce the national press’s anti-PT bias regardless of local electorates. Mixed results support partially both hypotheses. Gazeta do Povo consistently frames the impeachment as a legitimate constitutional response to a crime of responsibility and endorses Rousseff’s removal as the only viable solution. O POVO, by contrast, foregrounds the economic crisis, avoids projecting consequences, and remains largely non-committal, oscillating between criticisms of the impeachment and acceptance of its formal legality. We argue that these patterns indicate neither clear regional parallelism nor uniform anti-left bias, but an asymmetric regional parallelism in which conservative regions exhibit strong anti-PT advocacy while PT-leaning regions display moderated, centrist editorial postures constrained by political and economic elites.
Keywords
Introduction
The debate on how the Brazilian media system relates to the political system has in the concept of political parallelism one of its most recurring analytical axes. Although the notion was originally proposed by Seymour-Ure (1974), it was through its incorporation and adaptation by Hallin and Mancini (2004, 2011) that the concept gained traction to assess similarities and differences in relationships between political and media systems across national contexts.
In Brazil, however, its application is not without controversy. Some authors such as Albuquerque (2012) emphasize the instability of relationships between media and politics in Brazil as a characteristic that makes the concept inevitably inadequate for unstable countries. Despite this alleged inadequacy, adaptations are suggested by Albuquerque et al. (2015), who identified, for 2006–2014, an asymmetric political parallelism consisting of the following scenario: a PT (Workers Party) that was electorally strong governing without a counterweight in Congress, but that finds in the media system a situation inverse to the political system,i.e., a predominantly oppositional and therefore anti-PT stance.
More recent phenomena and studies show that this diagnosis would likely no longer hold. After all, the country has undergone rapid and profound transformations in both political culture and its media system (e.g. Dos Santos, 2019; Lycarião et al., 2018; Porto et al., 2020; Tavares, 2020). Moreover, the Brazilian journalism and political communication literature have highlighted that media systems in Latin America and other Majority World countries assume different configurations across subnational contexts, thus opening a research agenda attentive to the regional/local dimension of the relationship between media and politics (Chakravartty and Roy, 2013; Durazo-Herrmann and Pereira, 2022; Echeverria et al., 2024; Pinto, 2017). Nonetheless, no study has yet investigated the hypothesis of regional parallelism (Yusha’u, 2018) in Brazil’s regional/local quality press.
This is a relevant gap, given that political behavior in Brazil has been marked precisely by divisions between regions and states more inclined to vote for the PT, such as Ceará (in the Northeast of Brazil), and those with an anti-PT trait, such as Paraná (in the South).
This article explores this gap and offers an original contribution to the literature by shifting the focus from the national reference press to the regional press, testing whether territorially rooted electoral preferences help to explain editorial frames during a political crisis. In the selected case, this is the impeachment of Dilma Rousseff (PT).
Rousseff was the first woman elected and reelected president of Brazil and was definitively removed from office on 31 August 2016 by a Congress composed mostly of conservative senators and deputies. In less than 2 years of her second term, she was accused of committing a crime of fiscal responsibility, the so-called “pedaladas fiscais” (fiscal pedaling), which consisted in deliberately delaying transfers of funds to public banks to artificially improve the public accounts.
This study analyzes editorials on Dilma Rousseff’s impeachment published by the leading newspapers in their respective states, Gazeta do Povo (Paraná) and O POVO (Ceará). These states show sharply different electoral preferences in major elections (Nicolau, 2020).
Inspired by “multiple modernities” and “beyond the West” approaches (Chakravartty and Roy, 2013; Echeverria et al., 2024; Waisbord and Mellado, 2014), we formulate two competing hypotheses. The first, drawing on formulations by Yusha’u (2018), predicts the existence of a regional political parallelism, with the media system in each subunit of the nation approximating the predominant preferences of its electorate (
The corpus comprises 60 editorials, 31 from O POVO and 29 from Gazeta do Povo, published between March 10, 2015, a key date for pro- and anti-government mobilizations, and August 31, 2016, when the Senate voted to remove Rousseff. We conducted a content analysis guided by Entman’s framing theory (1993). Two coders assessed five variables: causes, attribution of responsibility, consequences, proposed solutions, and the editorial’s stance. Reliability was established after three pretests and an intercoder reliability assessment. Krippendorff’s α ranged from 0.83 to 1.00, and nominal agreement from 86.7 to 100%. Our analytical strategy combined a time series of editorial volume with a comparison of frames and editorial stances across the two newspapers.
The results point to convergences and divergences that do not fully adhere to either hypothesis. In light of this, we propose that Brazil’s regional press tends toward asymmetric regional parallelism: in states with a more anti-PT profile, the bias in that direction is clear and consequential; but in strongholds more favorable to the PT, we observe a more moderate posture, seeking to appear equidistant from ideological poles. At the subnational level in Brazil, therefore, the pronounced elite orientation characteristic of legacy news outlets across Latin America appears to persist, to the detriment of mass politics (Hallin et al., 2025).
The article is organized as follows. The next section reviews the Brazilian debate on political parallelism. We then situate the political and market changes that have reconfigured the Brazilian media system, motivating the present study. The subsequent section presents the research question and two hypotheses. Next, we detail the methodological design, case selection (O POVO and Gazeta do Povo), the corpus, and the coding procedures. We then present the results. The final section discusses their implications for understanding Brazil’s media system and, in light of the study’s main limitations, outlines a research agenda for other contexts.
Political parallelism in Brazil
One of the most prominent problems in Brazilian political communication research concerns the role played by the press and media (more broadly) in the country’s political dynamics. Within this debate, the concept of political parallelism is among the most prominent and certainly the one in greatest dialogue with the international specialized literature. Originally proposed by Colin Seymour-Ure in 1974, the concept was updated and adapted by Hallin and Mancini (2004, 2011) for comparative political communication research, establishing a framework to identify similarities and differences among national and regional realities (e.g., Western Europe vs. the Americas). Hallin et al. (2025) have recently renewed this endeavour by arguing that, in Latin America, political parallelism is often structured not primarily by stable party alignments, but by a cleavage between populism and anti-populism, especially in contexts marked by left-wing populist rule. In their account, one of the main reasons for this cleavage lies in the strong elite orientation of legacy news media in the subcontinent.
This way of conceiving political parallelism in Latin America addresses Albuquerque’s (2012) criticism regarding the use of the concept in Brazil. Broadly, the author points to the absence of stable relationships between agents of the media and political systems in the country, which would be a prerequisite for operationalizing political parallelism. Underlying this instability is precisely the hypertrophy of the media system in Brazil, which projected itself over the last decades of the 20th century as a political force invested with its own interests yet cloaked in the rhetoric of the common good (Albuquerque, 2005).
In line with this perspective, Pimentel and Marques (2021) consider that news companies in Brazil do not necessarily reflect the internal divisions of the political system (political parallelism) nor are they manipulated by corresponding groups (instrumentalization). According to the authors, they act as independent political players, historically intervening and participating in debates while promoting their own interests.
Azevedo (2017), in turn, takes 25 years (1989 to 2014) of coverage by O Globo, O Estado de S. Paulo, and Folha de S. Paulo as evidence that there would be a medium degree of political parallelism in Brazil. This degree implies a “sharing of ideological beliefs and values of a particular cause” (p. 153). In the specific case of those newspapers, it implies that they have historically maintained an alignment with center-right coalitions opposing liberalism vs. national developmentalism. Given the PT’s platform is convergent with the latter, the author bases the inevitable ideological opposition with the traditional Brazilian press as the reason why those outlets can be characterized as notably “anti-PT.”
Despite diverging from Azevedo (2017) as to the use of political parallelism to explain the relationship between Brazil’s media and political systems, Albuquerque et al. (2015) describe this relationship quite similarly in its anti-PT character. The authors identify a moment in Brazilian history (2006–2014) in which PT governed without an effective political opposition. At the same time, the press played a strongly oppositional role, without any sympathetic (or at least non-hostile) counterpart in the media system. Thus, they perceived a strong asymmetry between the role and power occupied by the PT in the political system and its corresponding support in the media system. From this asymmetry, the authors formulated the concept of asymmetric political parallelism to describe that historical moment.
However, the concept of asymmetric political parallelism rests on a partial and transitory view of Brazil’s media system for two fundamental reasons. First, after the period analyzed, there emerged a quite expressive right-wing parliamentary representation (Zucco and Power, 2024). Second, relevant sectors of the traditional press have felt compelled to modify their editorial lines in light of transformations in the information market and emergent political forces (Dos Santos, 2019; Lycarião et al., 2018; Porto et al., 2020; Tavares, 2020). Moreover, Durazo-Herrmann and Pereira (2022) and Pinto (2017) make important contributions by disaggregating the Brazilian media system and challenging analyses that treat it as a nationally homogeneous system. Durazo-Herrmann and Pereira demonstrate that political stability and instability affect the organization of the public sphere, the political identities of media actors, and journalists–source relations. Pinto, in turn, shows that regional subsystems differ in market concentration, spatial organization, local programming, and the links between media ownership and political actors. Taken together, these studies foreground the internal heterogeneity of Brazil’s media system and establish the subnational and regional scale as indispensable to its analysis.
This, in our view, raises the prospect of possible impacts on the behavior of the regional/local press. As we will see in the next section, these transformations and historical characteristics of Brazilian media system motivate the present study, which investigates the impact on the editorial politics of Brazil’s regional/local press when, in part of society and in the political system, a strong movement of opposition to the PT emerges, culminating in the 2016 impeachment of Dilma Rousseff.
Political and information-market transformations in Brazil
Electoral polarization in Brazil has been intense since redemocratization. For more than 20 years, the poles gravitated around PSDB (The Brazilian Social Democracy Party) and PT. With small ideological differences, the two parties dominated presidential elections far more because of their capacity to efficiently coordinate national and subnational alliances than due to voter engagement (Borges and Vidigal, 2018). However, the specialized literature has perceived growing voter polarization fueled by the structure of presidential competition and a shift to the right of the ideological spectrum, turning to conservative agendas (Borges and Vidigal, 2018; Cervi et al., 2021; Ortellado et al., 2022; Singer, 2022), with clear consequences for the ideological composition of Brazil’s parliament (Zucco and Power, 2024).
Chaia and Brugnago (2014: 102) identify the radicalization of the right in the 2014 elections, when the PT’s Dilma Rousseff defeated the PSDB’s Aécio Neves, consisting of asymmetric polarization between left and right, where the left strengthened a moderate stance and the right radicalized, with expressions verging on hate speech. This belligerent behavior became even sharper 4 years later, in 2018, when the far-right candidate Jair Bolsonaro (PL) was elected president.
Scholars seek to identify the factors that drove the electorate’s shift to the right. In this regard, Borges and Vidigal (2018) are categorical in asserting that Lula’s leadership helped strengthen identification with the PT among new party voters, considering that personal evaluations of Lula are strongly linked to party identification; i.e., the more a voter likes the PT’s principal leader, the greater the probability that they also identify with the party and vice versa. On the other hand, the PT’s centrality in the Brazilian political system generated a feeling of rejection, dubbed antipetismo (anti-PT sentiment). Nicolau (2020) observes that, since 2020, surveys show a robust contingent of voters who evaluate parties negatively, but only the Workers’ Party generated its own term for such rejection.
Moreover, Dos Santos (2019), Lycarião et al. (2018), Porto et al. (2020) and Tavares (2020) point to transformations in Brazil’s media system toward a more advocacy-oriented behaviour among media companies. These studies indicate, among such transformations, the emergence of a set of outlets within Brazil’s media system aligned with the views and ideological stances of the (far) right.
This is observed alongside the rightward shift of a large portion of the electorate. This movement was detected from 2010 and culminated in 2018 with Jair Bolsonaro’s election. Richmond and McKenna (2023) consider it not merely a product of historical contingencies, the party system, or factors operating at the national or individual levels, but of attitudes and electoral preferences linked to social conditions, organizational ecologies, and the subjective experiences of everyday life in particular places.
Given the country’s regional socioeconomic differences, the political polarization identified by the authors does not occur uniformly across regions. The Northeast and the South illustrate this logic. Nicolau (2020) identifies the Northeast as a PT stronghold from the 2006 elections onward, a pattern that has persisted. The numbers reinforce the region’s support for PT candidates. The researcher attributes these victories to major investments by PT governments in social programs and, above all, the Bolsa Família Program (one of the country’s most successful social assistance programs).
Results (2nd round) of the presidential elections in Brazil (2002 - 2022).
Note. Authors’ elaboration.
At the regional level, Richmond and McKenna (2023) perceive a clear split between presidential candidates, with the Southeast, South, and Center-West (the country’s wealthier regions) preferring right leaning candidates by significant margins, while the poorer Northeast and some Northern states favor the PT. In their view, this picture represents a continuation of earlier electoral trends, whose origins date back to when the country’s political polarization still revolved around PT and PSDB. These characteristics are in line with those identified by Albuquerque (2012) and Pinto (2017) for the Brazilian media system.
This scenario raises questions inspired by studies of media/politics relationships in Majority-World countries that offer analytic angles more adherent to the Brazilian case. In this regard, Echeverria et al. (2024) contend that Latin American media systems are better understood through the lens of multiple modernities and institutional legacies, in which centralized access to power, state-driven differentiation, and deep intra-national inequalities shape the structures and dynamics of media/politics relations. From this perspective, uneven economic development and territorial variation foster the emergence of distinct subnational media systems, making differences within countries as analytically significant as differences between nations.
Chakravartty and Roy (2013) show that this seems to be valid for India’s media system. This is because the country’s national subunits operate under distinct regimes of ownership and partisanship. The authors also emphasize differences between India’s political system and those of the Global North, with India displaying more contingent or flexible patterns of division and political order.
Regarding the Brazilian media system, Durazo-Herrmann and Pereira (2022) and Pinto (2017) challenge nationally homogenizing approaches by foregrounding the analytical importance of subnational variation. Pinto (2017) shows that regional media should not be treated as a residual or uniform sphere, but rather as a set of media subsystems whose configurations vary according to territorial, demographic, and economic conditions. Durazo-Herrmann and Pereira (2022) reinforces this perspective by showing that subnational political dynamics shape the organization of the public sphere, the political identities of media actors, and journalists–source relations. Taken together, these studies make the regional and subnational scales central to understanding the relationship between media and politics in Brazil.
The idea of regionalization guided by distinctions among national subunits is so central to Yusha’u (2018) that the author coined the term “regional parallelism” to explain the dynamics and complexity of relations between the press and political power in African countries. In his empirical research on Nigerian press coverage of scandals, Yusha’u (2018) observes how ethno-religious cultural differences between journalists in the North and South shape how clientelism and corruption allegations are addressed.
The contributions of Chakravartty and Roy (2013), Durazo-Herrmann and Pereira (2022), Echeverria et al. (2024), Pinto (2017), and Yusha’u (2018) motivate the present research insofar as they suggest that the relationship between media and political systems in Majority-World societies cannot be adequately understood without attention to subnational variation. In this regard, the article offers an original contribution to the literature by investigating whether the concept of regional parallelism (Yusha’u, 2018) helps to explain the editorial behaviour of Brazil’s regional quality press in states marked by contrasting electoral preferences. To the best of our knowledge, this is the first attempt to adapt and test the explanatory power of that concept in the Brazilian context.
Research question and hypotheses
Given the regional differences that Brazil presents in political behavior, is it possible to observe, from regional newspaper editorials, some level of regional parallelism (Yusha’u, 2018), in which each region’s reference press is oriented by the predominant electoral preferences of its respective national subunit? That is, would the regions and states most supportive of the PT have a press more aligned with the party, while the press in more anti-PT states would be more adversarial to it? Or would the anti-PT character observed in the national reference press in previous studies (e.g., Albuquerque, 2019; Azevedo, 2017) be equally observable in the regional press, regardless of its predominant electoral demography?
These alternative formulations of the same research question correspond to two distinct hypotheses. The first concerns the existence of regional parallelism (
Study design and methodological procedures
To investigate the explanatory power of each hypothesis above, the present work examines two regional newspapers. One (Gazeta do Povo) comes from the state of Paraná and the other (O POVO) from the state of Ceará. Table 1 shows that these states have exhibited an electoral polarisation since 2006.
For the state of Ceará we selected O POVO since it is the longest-running and highest-circulation print newspaper in the state of Ceará during Rousseff’s impeachment (Poder360, 2018). In terms of its editorial line, the paper presents itself as a forum for the pluralist debate of ideas, placing democracy, the rule of law, and free enterprise as highly valued principles (O Povo, 2023).
The newspaper belongs to a family-owned company devoted exclusively to the communications sector, which also includes radio stations, a television channel, a news portal, a magazine, and the Fundação Demócrito Rocha (Marques and Rebouças, 2019; O Povo, 2023). Its relationship with the local power structure cannot be reduced either to straightforward dependence or to full autonomy. If, on the one hand, its trajectory suggests historically close and at times strategic articulations with Ceará’s political and economic elites (Silva, 2016; Vidal, 1994), on the other hand the newspaper has made a concentrated effort to counter the image of political attachment to specific groups by cultivating a discourse of editorial independence grounded in transparency toward readers. The clearest expression of this strategy was the creation, in 1994, of the position of ombudsman, which has been maintained ever since (O Povo, 2023). Over the last decade, moreover, O POVO has expanded its investment in digital platforms and social media, thereby broadening its reach beyond Ceará and diversifying its audience profile (Marques and Rebouças, 2019).
For Paraná, we selected Gazeta do Povo because, like O POVO in Ceará, it has the largest reach and circulation in the state (Poder360, 2018). Beginning in 2020, outside the period analyzed here, the newspaper shifted to digital only and ended its print edition (Poder360, 2020). Editorially, Gazeta do Povo presents itself as a voice of Brazilian conservatism and pursues advocacy journalism aligned with the right (Tavares, 2020). The newspaper currently belongs to the Grupo Paranaense de Comunicação (GRPCOM), but its original capital reportedly came from dozens of wealthy Paraná families, especially owners linked to the yerba-mate and timber sectors, whose interests were addressed in its inaugural editorial (Santos, 2011).
From the outset, therefore, the media outlet was economically anchored in dominant regional sectors, and this was later complemented by a strong dependence on advertising markets. With regard to the local power structure, Gazeta do Povo’s self-image of independence was contradicted by a persistent alignment with political and economic groups. This proximity later assumed a new configuration during the Lava Jato years, when Curitiba (capital of Paraná) became a central node of national political conflict and the newspaper converted that privileged territorial position into editorial projection, attracting conservative readers and reshaping itself as a politically oriented, anti-PT, and right-leaning outlet (Tavares, 2020).
Both O POVO and Gazeta do Povo sought to expand their audiences through their websites and social media platforms. However, the nationwide picture available for the period suggests that such efforts had not yet translated into significant national penetration since fewer than 3% of newspaper readers in Brazil identified either title as their primary newspaper reference (Secom, 2016).
Thus, surveying these newspapers offer an opportunity to test the two hypotheses laid out in the previous section. Moreover, given the paucity of works dedicated to understanding local press behavior on moments of political stress, surveying newspapers with a more local reach helps to overcome this gap in the literature since much of it focuses on national media with strong agenda-setting power.
The analytical corpus (Nocrato and Lycarião, 2026a) consists of 60 editorials (OP = 31; GP = 29), published from 10 March 2015 (when pro- and anti-federal government mobilizations began) through 31 August 2016 (when the president was formally deposed by the Senate).
Material retrieval followed rules previously established in the research codebook (Nocrato and Lycarião, 2026b), using the terms “impeachment”, “impedimento,” “cassação,” or “deposição.” The coding of opinion pieces underwent three pretests using editorials of other newspapers (e.g. O Globo, Estadão), and cases were eliminated when any of the four words appeared in contexts unrelated to the president’s deposition process.
Variables and categories created for content analysis.
Note. Authors’ elaboration.
Results of the intercoder reliability test.
Note. Authors’ elaboration.
With the methodological structure defined, we began the analysis by observing editorial publication frequency. We initially divided data analysis into two parts. First, we observed temporal oscillation by identifying the volume of editorials published by O POVO and Gazeta do Povo over the selected time period. Second, we adopted frame analysis to assess the two newspapers’ positions on the legitimacy or not of Dilma Rousseff’s removal.
Results
The investigation observed that both outlets began to address Dilma’s deposition in March 2015, 3 months after her second-term inauguration (Figure 1). Thus, the editorials placed the impeachment on the agenda even before the process began institutionally in Congress. In 2015, both O POVO and Gazeta do Povo showed an oscillating movement and a reduced volume of texts (OP = 7, GP = 9). In 2016, both newspapers presented publication peaks between March and May, as shown in Figure 1. Monthly volume of editorials mentioning “impeachment,” “cassação,” and “impedimento” – OP and GP.
Presence of variables in the editorials.
aThe variable “stance” was the only one with the editorial as the unit of analysis; it could not occur more than once per editorial. The other variables were coded by utterance (sentence within the editorial) and could therefore occur multiple times within the same editorial.
Note. Authors’ elaboration.
Categories present in the variable “Causes of Impeachment”.
Note. Authors’ elaboration.
Gazeta do Povo, in turn, presents the legal argument of the crime of responsibility (the main claim used by the opposition) as the central argument for supporting Dilma’s removal. The argument began to be addressed in a July 2015 editorial, which asserted that the Federal Court of Accounts (TCU) had rejected the government’s accounts due to illegalities in fiscal management; it extended over a year and 9 months. This was reinforced on 15 April 2016, 2 days before the lower house’s vote on the admissibility of Dilma’s impeachment.
Categories present in the variable “Those responsible for the crisis”.
Note. Authors’ elaboration.
Dilma’s responsibility appears in 22 opinion utterances in Gazeta do Povo. In one of them, it denounces what it labels a maneuver by the president and the PT to stop impeachment, putting the government up for sale by offering ministries and senior posts to any deputy who committed to voting in Dilma’s favor.
In the left–right ideological war, the newspapers diverge on the responsibility of other actors. While Gazeta do Povo blames Lula and the PT for the moral, political, and economic deterioration (high inflation, low purchasing power, and corruption allegations), O POVO blames the government for being unable to implement what it considers necessary measures to get Brazil’s economy back on track. Lula is cited only once.
In contrast, Gazeta do Povo adopted a strongly oppositional stance toward Lula, also holding him responsible for the crisis, associating his strong influence over Dilma’s government and the use of subterfuges to hinder the progress of the impeachment process in Congress and prevent the deposition of his political ally.
Categories present in the variable “Consequences of Impeachment” a .
aIn this variable, totals equal the number of editorials because more than one consequence was not presented per unit. As shown in the table, most editorials did not include this dimension of framing.
Note. Authors’ elaboration.
Gazeta do Povo also shows a strong tendency not to present the consequences of removal (76% of cases). Nor does it discuss important themes that guided the debate, such as whether there would be improvement or deterioration in the economic and political scenarios. Most opinion pieces do not make projections about the country’s future.
Categories present in the variable “Solution for the political situation”.
Note. Authors’ elaboration.
Moving in another direction, O POVO presents no solution in 56.3% of its editorial utterances.
Categories present in the variable “Editorial stance regarding the impeachment of Dilma Rousseff”.
Note. Authors’ elaboration.
In all texts reinforcing the process’s legitimacy, Gazeta do Povo underscores that interrupting the mandate is an action backed by the Constitution. This thesis is defended in early 2015 editorials and follows most opinion pieces throughout nearly 2 years of debate on the potential removal of Dilma Rousseff. Accordingly, the newspaper argues that Congress respected the Constitution; the president’s removal was therefore legal and legitimate.
O POVO, in turn, adopts a largely neutral stance. In 16 opinion pieces, it takes no position on the impeachment. Early in 2015, however, the paper ran three editorials describing the interruption of the presidential mandate as harmful to the political order, while refusing to call the process a “coup,” as Dilma Rousseff and her supporters did. As the congressional proceedings advanced, its framing shifted: in 12 opinion pieces, O POVO maintained that the process complied with Brazil’s institutional norms.
Discussion and conclusion
This article set out to examine how O POVO and Gazeta do Povo framed and positioned themselves in their editorials regarding Dilma’s impeachment, in view of two hypotheses formulated to answer the following research question: is it possible to observe, from regional/local newspaper editorials, some level of regional parallelism (Yusha’u, 2018) in which each region’s reference press is oriented by the predominant electoral preferences of its national subunit? Or does the regional/local press behave uniformly across subunits, invariably revealing an anti-left, anti-PT or even anti-populist bias (Albuquerque, 2019; Albuquerque et al., 2015; Azevedo, 2017; Hallin et al., 2025)?
The results presented above do not fully adhere to either hypothesis. This is because we neither observed regional parallelism with clearly antagonistic positions nor can we ignore the divergences. Both media adopted frames and stances that were at times similar and at times different regarding the impeachment of Dilma Rousseff.
Gazeta do Povo emphasized the legal argument of “crime of responsibility,” bringing it to the center of the discussion. This was the main basis presented to justify removal of Rousseff, an attempt to deconstruct the argument of a parliamentary coup defended by the president and her supporters. O POVO, in turn, shifted the center of discussion from the political or legal aspects to the economic field, associating especially rising unemployment and a sharp drop in industrial production as determining factors for part of Brazilians’ support for the impeachment.
As for responsibility for the crisis, both blamed the president for the crisis that befell the country. However, Gazeta do Povo went further and also held Lula responsible, something much less emphasized by O POVO.
On solutions for the crisis, the newspapers adopted divergent answers. While O POVO does not point clearly for a solution, Gazeta do Povo embraced a pro-impeachment defense. In every editorial addressing a solution for Brazil, it defends impeachment as the only solution. Emphasizing the crime of responsibility, Gazeta do Povo constructed the idea that the president’s removal was legitimate and supported by the Constitution. In 75.9% of its editorials it reinforces that the impeachment was desirable.
O POVO predominantly adopted neutrality. In more than half of its editorials (51.5%), it did not take a stance on the impeachment. But it was also the outlet that expressed some degree of opposition to the impeachment. It was framed in a negative tone in three editorials, but in other 12 pieces stated that the process respected the country’s institutional order.
In view of these mixed results, we identify in Brazil a form of regional parallelism, though of an asymmetric nature. This is because, as suggested by O POVO’s oscillation in stance over the analysed period, the behaviour of both newspapers appears to be shaped less by the electoral profile of their states than by the social profile of those who own and predominantly read them. In both cases, these are media outlets historically tied to economic elites and directed primarily at higher-income readers (Albuquerque, 2012; Azevedo, 2006; Secom, 2016). At the subnational level in Brazil, therefore, one finds the same strong elite orientation of the main legacy news outlets in Latin America that tends to relegate mass politics (Hallin et al., 2025). Thus, rather than a press polarized according to the political realities of its respective states, what emerges is a regional press that assumes strongly right-leaning tones in states most faithful to anti-PT sentiment and, in partial and asymmetric contrast, a press that, even in states with electoral preferences toward the PT, tends to accommodate the ideological inclinations of the country’s economic elites. In this sense, we argue that Brazil’s regional press tends to display an asymmetric regional parallelism.
Having said that, the present study presents two important limitations. First, the analysis is restricted to editorials and does not encompass news content. Although this choice is consistent with an important part of the Brazilian literature on media and politics, it nonetheless leaves an analytical gap, since political parallelism tends to be more robust when it extends beyond the opinion pages and also structures news coverage. Second, the empirical corpus is limited to two newspapers, which evidently constrains the scope for broader generalization about the Brazilian regional press as a whole.
Notwithstanding these limitations, the article offers an innovative contribution insofar as, to the best of our knowledge, it constitutes the first attempt to adapt and test the explanatory power of the concept of regional parallelism (Yusha’u, 2018) in the Brazilian context. In doing so, it helps advance a more nuanced understanding of the internal heterogeneity of the Brazilian media system, already noted by earlier studies (Durazo-Herrmann and Pereira (2022); Pinto, 2017).
Future research could build on both the findings and limitations of the present study. In this regard, the regional parallelism hypothesis could be examined in other national contexts characterized by pronounced subnational political polarization. The United States, for instance, constitutes an especially suggestive case, since states such as Wyoming and the District of Columbia display sharply contrasting electoral preferences. Examining how local media in such politically divergent settings frame contentious national issues, both in news coverage and in editorials, could offer valuable comparative insights into whether regional media in federal systems tend to reflect local partisan alignments or, alternatively, gravitate toward a more centralized elite consensus, as observed in Brazil. Such comparative efforts would help clarify the conditions under which regional parallelism (symmetric or asymmetric) emerges in different media systems.
Footnotes
Acknowledgments
We thank the funding agencies Funcap and CNPq for the financial support received for this research, and we extend special thanks to Giselly Correa Barata for her work and support in the reliability testing of the content analysis. We also thank the editorial team of Journalism, especially the anonymous reviewers, who helped us significantly improve the quality of the present paper.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by CNPq (406504/2022-9; 403286/2023-9; 309636/2025-6), and Funcap (PRH-0212-00069.01.00/23).
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
We declare that the data and codebook are available through the following links: Nocrato and Lycarião (2026a, 
.