Failed Honeymoon: Dilma Rousseff’s Third Election Round

The 2014 Presidential Election in the Media

A year after the protests of June 2013, when demonstrators occupied the streets of Brazil’s main cities, the most acrimonious electoral process in the history of the New Republic began. The campaign’s fierce competition impacted the strategies adopted by candidates already in the first round of elections. As Borba, Veiga, and Martins (2015) have pointed out, Dilma Rousseff used a higher percentage of negative ads—attacks against her adversaries in the free-ad time of the electoral campaign—than did previous incumbents who led in the polls at the time of their own reelection efforts (Borba, Veiga, and Martins, 2015).

The electoral campaign and social media embody important aspects of political communication, but we cannot overlook the critical role of journalism in the configuration of public debates in electoral periods. In fact, notwithstanding the growth in the number of Internet users in Brazil, almost all of the journalistic content in circulation is produced by the major communications companies. Studies on the role of the media during elections in Brazil have identified a solid disposition on the part of these companies to give negative coverage to candidates on the left during presidential elections (Aldé, Mendes, and Figueiredo, 2007b; Arruda, 1995; Azevedo, 2000; Colling, 2000; Figueiredo, 1996; Gonçalves, 1996; Kucinski, 1998; Miguel, 1999; 2002; Souza, 2009). Historically this bias has been so acute that sometimes the media giants have taken the initiative to defend themselves, arguing that their position is simply the result of a vigilant stance with regard to political power with the interest of society at heart, all of this very much in line with the role the media should have in modern democratic societies. This is also known as the watchdog function of the media. We have shown elsewhere that in Brazil the media giants do not act as watchdogs when center-right candidates run for the presidency, but only in regard to left and center-left candidates, a bias that reveals the watchdog argument to be rather unsound (Feres Júnior and Sassara, 2016).

In fact, the study of media and politics has an important role to play in present-day democracies. By critically analyzing the media in action, it watches the so-called watchdogs. Continuing a tradition established in media studies in Brazil (Aldé, Mendes, and Figueiredo, 2007a) and abroad (Altman and King, 2007; Andreev, Diggory, and King, 2001), we applied the methodology of sentiment analysis to a corpus composed of articles published on the front page of Brazil’s top newspapers: Folha da S. Paulo, Estadão, and O Globo. ² The results (Table 1) show that Rousseff’s bias index was by far the most negative among the main candidates, while Aécio’s was the least.

OPEN IN VIEWER

Table 1

Sentiments of Front-Page Newspaper Articles for 2014 Presidential Candidates

	Neutral	Negative	Positive	Bias Indexa
Dilma	273	212	10	.74
Marina	231	59	24	.15
Eduardo	76	19	14	.06
Aécio	221	44	41	.01

Source: Manchetômetro (http://www.manchetometro.com.br).

a.

The bias index is calculated by subtracting the number of positive articles from the negative and then dividing the result by the number of neutral articles. Simply subtracting negatives and positives (a method employed by other researchers in calculating comparative bias) overlooks the importance of the neutral coverage. Descriptive texts such as the ones we find in neutral pieces dilute the impact of the negative coverage and give the candidate more exposure, which in electoral terms is almost always positive. Therefore, neutral articles are included in the calculation as a bias dilution factor.

While the numbers of neutral articles the two candidates received were similar, Rousseff had five times more negative articles published on the front page of the newspapers than Aécio did. At the same time, Aécio had four times more positive articles written about him than the candidate of the Partido dos Trabalhadores (Workers’ Party—PT). If the data analysis from the 2014 election were extended to the coverage of parties, scandals, and other political figures such as Lula and Fernando Henrique Cardoso, the results would show that in all categories the bias against PT was immense, ³ regardless of which big media organization in our sample is singled out. In spite of the unbalanced coverage, Dilma Rousseff was reelected in the second round of the election with 51.64 percent of the popular vote, the narrowest victory since the country’s return to democracy in the 1980s.

The Postelection Political Context

Immediately after the 2014 election, the losing party, the Partido da Social Democracia Brasileira (Brazilian Social Democracy Party—PSDB), started to give signs that it was not going to accept the results. Only four days after the second round, Carlos Sampaio, who was responsible for the legal section of the PSDB campaign, filed a petition with the Electoral Superior Court for an inquiry into the voting system that included a recount of the votes conducted by a committee of specialists selected by the parties. In the petition, the party claimed “lack of belief that the vote counting was trustworthy and that the electronic system was infallible based on various accusations made after the voting process.” The petition, however, contained no concrete charges. ⁴

On December 18, hours before the electoral court swore in Dilma Rousseff and her vice president, Michel Temer, PSDB and its party coalition requested the annulment of Rousseff’s and Temer’s candidate registrations and the cancellation of their titles. Among the charges used to justify this move were economic abuse, manipulation of socioeconomic indicators to achieve electoral goals, and use of tainted money from Petrobras in the campaign. The request was filed in February but reopened in October 2015 and had not received a final ruling at the time this article was finalized.

In spite of these efforts, Dilma Rousseff’s second term began in time, just as determined by the statute regulating the electoral process. On March 8 she gave the inaugural speech, taking advantage of International Women’s Day to defend a fiscal adjustment. The 16-minute video with her speech sparked a wave of protest that peaked on March 15, when thousands of protesters went to the streets of Brazil’s major cities asking for her impeachment. ⁵ The opposition actively incited the movement through social media, inviting the population to occupy the streets of the country’s main cities.

In June 2015, Brazil’s Tribunal de Contas da União (Federal Court of Accounts—TCU [equivalent to the U.S. General Accounting Office]) decided to postpone the approval of the 2014 budget. In his written opinion, the president of the TCU, Augusto Nardes, asked the president for further explanation on 13 items, which included accounting measures that had the alleged intent of disguising the size of the budget deficit and decrees that, according to him, created credit without congressional authorization. In her defense, the president denied any administrative wrongdoing and argued that the TCU had never questioned previous administrations when they adopted the same practices with regard to the government’s accounts.

On October 7 the TCU, in unprecedented fashion, voted unanimously to reject the government’s accounts. On October 21 the opposition sent the House a new impeachment request, the second one crafted by the jurists Hélio Bicudo, Miguel Reale Júnior, and Janaina Paschoal at the behest of PSDB. In early December Eduardo Cunha—who was the speaker of the House at the time—accepted the request in retaliation for PT’s failure to support him in an investigation he faced in the House Ethics Committee. The following April, in the longest Congress session ever, representatives voted to impeach Rousseff in the name of their families and God. ⁶ On May 12, the senate started the impeachment process, thereby suspending Dilma Rousseff. On August 31, 2016, the impeachment was approved and the president elected in 2014 was permanently removed from her post.

An analysis of the 2014 election shows the media’s strong anti-PT and anti-Rousseff bias (Feres Júnior et al., 2014; Feres Júnior and Sassara, 2016). Journalistic coverage of electoral campaigns is usually characterized by negative bias, but the 2014 period was atypical in that the negative bias continued after the election and increased exponentially until Rousseff’s impeachment.

The Media After the Election

The vast majority of the studies on media and politics in Brazil focuses on elections. In this article, we extend the analytical scope into the postelection phase as we try to unveil the role the media played in the political crisis that followed Dilma Rousseff’s election. We call this phase “Rousseff’s third election round” by analogy with the two rounds typical of the election process in Brazil. The third round culminated in Rousseff’s impeachment. We hypothesize that in the months that followed the election the main Brazilian newspapers, in contrast to what the literature on the subject describes, promoted a smear campaign against Dilma Rousseff very much in tune with the opposition political forces that she had just vanquished at the ballot box.

During the period between January 2014 (the beginning of the electoral year) and August 2018, her final deposition, the number of positive articles pales in comparison with those of negative and neutral ones (Figure 1).

OPEN IN VIEWER

Figure 1.

Coverage of Dilma, 2014–present (Manchetômetro [http://www.manchetometro.com.br]).

We observe that the number of negative pieces began to exceed the number of neutral ones around October 2014, that is, right after the election. Before this turning point, however, one can observe several trends. First, although the general tone of the coverage of the election was strongly biased against Rousseff, in the less heated time period between January and July of 2014 the number of negative articles was similar to that of neutral ones. As the campaign officially began, in July, the coverage of Rousseff increased in volume, but until the second round of elections the number of neutral articles was still larger than the number of negatives. The month after the second round, in November 2014, the curves started to diverge: negative articles increased in number while neutral ones started to decline.

The number of neutral articles during the electoral campaign might give the erroneous impression that Rousseff received a balanced coverage. Coverage of her during the campaign was extremely negative compared with that of other candidates, and after the election it became even more so. Before the campaign the ratio of negative and neutral articles was one to one. During the campaign the number of neutral articles exceeded the number of negative ones, but after the election negative articles were three times as frequent as neutral ones—a real media massacre. Acting in tandem with the political opposition, the big media denied Dilma Rousseff a honeymoon period. In fact, the last time that the numbers of neutral and negative articles were equal was in December 2014. From that moment on negative articles were more than twice as frequent as neutral ones.

The peaks of Rousseff’s negative coverage coincided with important events in the political process that culminated with her impeachment. In March 2015, when the first pro-impeachment demonstrations took place, Rousseff’s coverage in the media reached a record high of negatives. There was another peak in December, when Eduardo Cunha accepted the impeachment request in Congress, and yet another in March and April of 2016, coinciding with more unfavorable political events: the beginning of the impeachment process, Lula’s indictment, more pro-impeachment demonstrations, leakage of the content of phone conversations between Rousseff and Lula, and the blocking of her attempt to make Lula her chief of staff by the Supreme Court.

The skeptical reader may wonder whether it is not only natural for media coverage to become negative in moments of political tension such as the one Brazil has been experiencing. In fact, this is not the case. Each of the events mentioned above has elicited profound moral and practical disagreement. For instance, according to some analysts, Judge Sérgio Moro broke the law in leaking the conversation between Rousseff and Lula on at least three counts: the conversation was recorded after the authorization for the wiretap given by Moro himself had already expired, the country’s president was being recorded without prior Supreme Court authorization, and the wiretap was leaked to the press. Other jurists argue that there are no constitutional grounds for denying Rousseff the right to appoint a minister in her cabinet or for denying the citizen Luis Inácio Lula da Silva the basic political right of being nominated minister. However, the media framed this event only as crude evidence of Rousseff and Lula’s criminal behavior, giving no voice to those who had a different interpretation of the facts.

Standards for Comparison

Our skeptical reader may also ask whether the media’s negative bias with regard to Rousseff was simply the result of the watchdog behavior we should expect from them vis-à-vis the political establishment. One can dismiss this claim from the start by leaning on the academic literature on media and politics in Brazil, which shows consistent bias against the PT and the left in general throughout the short history of the New Republic and also that the media do not behave as a watchdog when political power is in the hands of right-leaning incumbents. Insisting further, our skeptical reader may retort that past media behavior should not be used to explain the present—that the observed historical bias does not prove that there was bias in the most recent postelection period. At least hypothetically, the argument would go: it is possible that the media have decided to change their stance and in fact started playing the role of watchdog of the public interest. However, the strong anti-Rousseff bias revealed here is evidence that this hypothesis does not stand up to scrutiny. We have tried to show, through an internal comparative logic, that the intensity of the negative coverage is not entirely justifiable, since there is a great deal of controversy surrounding the role played by the chief of the actors in this narrative. Rousseff, Moro, Lula, Eduardo Cunha, Michel Temer, the PSDB, and the PT are all objects of intense debate. In other words, there are various interpretations, and the media favored the interpretations that opposed Rousseff, silencing competing versions. In order to further verify this claim, we chose to compare Rousseff’s coverage with that of her main opponent in the presidential campaign of 2014, Aécio Neves.

Senator Neves has had a long political career. The grandson of former president Trancredo Neves, he was elected congressman for the state of Minas Gerais in 1987. In 2001 he became speaker of the House and subsequently was elected governor of Minas Gerais for two consecutive terms. In 2011 he became senator for the same state. After his narrow defeat in the presidential election of 2014, Neves continued to be the president of his party, PSDB. In addition to his public image, primarily as a champion of the political opposition to PT governments, Neves is appropriate for our comparison because he has been accused of corruption and other illegal practices many times in the recent past.

In March Delcídio do Amaral, a former senator caught by corruption investigations known as “Operation Car Wash,” stated in his plea bargain deposition that Neves had received bribes from Furnas, a state-owned company in the energy sector, ⁷ and that he had falsified Banco Rural data obtained during the national post office hearings in Congress. ⁸ In the same month the attorney general, Rodrigo Janot, asked the Supreme Court to investigate Neves on the basis of Amaral’s accusations. Amaral also claimed that, when he presided over the national post office hearings, he had to work hard to stop information about Marcos Valério, an advertisement executive with previous involvement in other corruption scandals, from becoming public because this information would have harmed Neves directly. ⁹

In June Neves was accused of having favored his own father illegally while still governor of Minas Gerais. ¹⁰ He was also accused of having received bribes from the public oil company Transpetro and of having received illegal funds to finance his campaign for House speaker in 2001. ¹¹ Neves was also held responsible for appointing Irany Varella, who was accused of taking bribes during the Cardoso administration, to a Petrobras directorship. ¹² Also in June, the developer Léo Pinheiro, partner and former president of the construction company OAS, claimed that he had bribed then-governor Neves’s assistants during the construction of the Administrative City, a project that Neves had commissioned as governor of Minas Gerais and whose price tag was 2.1 billion reais. ¹³

In July the former president of the construction company Andrade Gutierrez reported that he had known of a negotiation involving Codemig, a state-owned company in Minas Gerais, which included a donation to Neves’s presidential campaign. The donation was to have been mediated through Oswaldo Borges da Costa Filho, a former president of Codemig and known as the informal treasurer of PSDB in Minas Gerais. ¹⁴

All the news pertaining to allegations of corruption against Neves were reported in the same media outlets analyzed in this article. However, they were not given much attention by editors and journalists, did not figure prominently on the front page, and did not generate follow-up coverage and journalistic investigations. Even during a period of intense accusations, Neves’s neutral curve was equal to the negative one or even exceeded it. The comparison between the negative reporting on Rousseff and Neves in 2016 shows clearly how disproportionate the coverage was (Figure 2).

OPEN IN VIEWER

Figure 2.

Negative articles about Dilma (top) and Aécio (bottom), January–August 2016 (Manchetômetro [http://www.manchetometro.com.br]).

The ratio between the number of negative articles for Rousseff and for Neves varies from 7 (in June) to 82 (in August), the month Rousseff was suspended. Only one negative article on Neves was published in the period. At any rate, the curves’ visual difference is enough to prove the point.

Conclusion

This brief analysis of the postelectoral political context is sufficient to show that Rousseff did not enjoy a honeymoon period after her reelection. In fact, while her government faced serious difficulties in passing legislation in Congress, her second term was threatened even before inauguration by attacks on the legitimacy of the electoral process. Eduardo Cunha’s election as speaker of the House ensured that no honeymoon would occur at least in the realm of party politics. From the beginning, Cunha sabotaged the alliance between Partido do Movimento Democrático Brasileiro (Brazilian Democratic Movement Party—PMDB) and PT, the two largest parties in Congress and the pillars of her presidency’s support coalition, thus eliminating any hope for the stabilization of the political process.

The opposition went beyond denying Rousseff a honeymoon period in the political arena. It violated Przeworski’s principle of democratic stability by using both official and unofficial channels to remove the elected president. Lawsuits filed in the electoral court, rejection of government accounts, and petitions for impeachment obviously fell into the category of official channels. But they were doing much more than using official channels to oppose her. Unofficially, the opposition, with the help of the big media, managed to organize semiautonomous right-wing movements that coalesced around the pro-impeachment campaign demonstrations—an innovation in recent Brazilian history, since the political right had been absent from the streets since the start of the democratization period in the late 1970s.

Things were no easier for Rousseff in the media. As we have shown above, in the media realm the honeymoon hypothesis should be fully rejected. The abrupt rise in the number of negative articles in November, the month after the second round of elections, shows that the major Brazilian media conglomerates had no intention of waiting for the organization of the newly elected government. Part of the negative coverage Rousseff received had to do with Operation Car Wash, a corruption investigation that so far has produced no evidence of her involvement with illegal practices.

Former President Lula was also the target of an intense negative campaign in the media. He and Rousseff were often linked in articles that discussed his alleged participation in corrupt practices involving Petrobras. The most scandalous event was Rousseff’s frustrated attempt to appoint Lula her chief of staff. The Supreme Court blocked the nomination, thereby violating his right to occupy the public post. Its decision took place amidst an intense negative media campaign fueled by illegal leaks of private phone conversations between Lula and Rousseff recorded under the authority of Judge Sérgio Moro and then handed over by Moro himself to be publicized by Rede Globo.

Rousseff’s impeachment was the result of a confluence of factors. It is true that among these factors were a highly politicized and biased judiciary (from Moro to the Supreme Court), the Attorney General’s Office, and Federal Police, state organizations that often seemed out of control either because there were no mechanisms of checks and balances (in the case of the Attorney General’s Office) or because these mechanisms did not work (in the case of the judiciary). However, it is hard to imagine that these elements would have reached such synergy without the active collaboration of the media. They promoted Moro and the federal prosecutors in Curitiba to the status of national heroes, pressuring the Supreme Court justices into action or paralysis depending on the circumstances, encouraging the middle class to demonstrate in the streets of major cities side by side with right-leaning groups, fascists, and military-coup supporters, and selectively targeting Rousseff, Lula, the PT, and several people linked to them—all this with a barrage of innuendo and accusations that were often baseless.

Brazil since Rousseff’s impeachment is in the hands of a government with very little legitimacy. This government is now implementing policies that systematically turn back social rights and achievements gained in recent decades. The country has a failed justice system, which is in the hands of people who claim to pursue justice but can barely conceal their own political interests. Moreover, Brazilian society is highly unequal and increasingly disillusioned with democratic political representation. Brazilians have faced many crises in their history, and even the present one will surely pass. However, the country’s return to the virtuous enhancement of democratic institutions and practices will not happen as long as political communication is controlled by a small group of families who own the major media companies. Quality information and access to multiple opinions are the sustenance of democratic life. Without it, there is no actual democracy, be it in Brazil or elsewhere in the world.

Supplemental Material