TRUMP,BOLSONARO,AND THE FRAMING OF THE COVID-19 CRISIS

Abstract

In the aftermath of the global COVID-19 crisis, whereas many world leaders enacted swift lockdown orders and robust testing regimes to preserve public health and to speed up economic recovery, Donald Trump in the United States and Jair Bolsonaro in Brazil responded to outbreaks by publicly downplaying the significance of the crisis and argued that overly restrictive health measures would create too sizable an economic risk. These two presidents have done much to weaken democracy and trust in government. In this article, we examine the extent to which two institutions in each country––federalism and the party system––impacted the ways in which they framed the COVID-19 crisis and policy responses to it in 2020, especially during the first months of the pandemic. Our evidence suggests that each of these institutions provided opportunities for both leaders to reconstruct public understandings of the crisis while deflecting blame for negative public-health outcomes.

As the COVID-19 pandemic reveals, public-health emergencies constitute a threat to service-oriented economies (Brinca, Duarte and Faria e Castro 2020). Even so, political elites have embraced divergent ideas about the appropriate relationship between the pandemic and economic management. In countries as different as Australia, New Zealand, South Korea, and Singapore, leaders have responded to this reality by prioritizing public health, enacting swift quarantines, and strong testing regimes. Indeed, the evidence suggests that these measures have worsened economic outcomes relative to countries with less stringent measures (Balmford et al. 2020). Yet in others, such as Brazil and the United States, political leaders have attempted to maintain economic productivity by downplaying the severity of the crisis, eschewing responsibility for negative public-health outcomes. By the fall of 2020, Presidents Donald J. Trump and Jair Bolsonaro had paid a personal cost for their leadership during the pandemic: each had contracted COVID-19. ¹ In neither case did these personal experiences alter their general approach to managing the pandemic, which had made both countries global leaders in confirmed cases and deaths. In neither country did the virus elicit a shift in the major ideological frameworks each leader applied in their crisis response. Trump's conflict with public-health officials––while extreme––bore a striking resemblance to prior Republican attacks on science and drew on a deep reservoir of conservative distrust of scientific experts (Motta 2021). Bolsonaro supported revisionist views, challenging the effectiveness of social distance measures and vaccines and supporting medicines with no scientific evidence for COVID-19 treatment (Fonseca et al. 2021). At the same time, in each of these two countries, this shift took place in a broader political context where a president sought to weaken democratic institutions and expert systems to consolidate their power, regardless of the human and public health cost of this strategy amidst a major global pandemic.

What distinguishes both the U.S. and Brazilian cases, however, are the unique configurations of institutions in both countries that shaped each leader's capacity to rhetorically reconstruct the crisis to suit their ends, deferring blame for deaths while claiming credit for delivering economic relief (see Weaver 1986). Although Trump and Bolsonaro are exceptional presidents, in this article we argue that their framing strategies during the COVID-19 crisis reflected well-known political strategies and existing institutional legacies. More specifically, we examine the role played by two institutions––federalism and party systems––in providing a political support structure for the idea espoused by both Trump and Bolsonaro that strong public health measures would hurt the strength of the national economy. The analysis of framing processes surrounding a deadly pandemic such as COVID-19 also allows us to gauge the actual impact of these processes on public policies that can save or endanger people's lives and livelihoods. Thus, any analysis of framing processes must circle back to the broad human consequences of political discourses and strategies which, in the case of both presidents, have been disastrous.

Empirically, the present article analyzes the public statements of Presidents Trump and Bolsonaro (speeches, press conferences, media interviews, and Tweets) during the first year of the COVID-19 crisis (2020), with a particular focus on their initial responses to the pandemic. The objective of this analysis is to elucidate the relationship between their statements and specific blame avoidance and credit claiming strategies embedded in U.S. and Brazilian political institutions, especially federalism and the party system. More concretely, the analysis explores the framing of specific policy alternatives such as confinement and economic stimulus in the context of a rapidly evolving pandemic, which soon became a contested crisis in these two countries with multiple meanings and ramifications. As suggested below, in the institutional context of federalism and partisan politics, both Presidents Trump and Bolsonaro attempted to exploit this situation to deflect blame for the growing number of COVID-19 victims while claiming credit for their plans to foster a swift reopening of the economy.

Four main sections comprise the remainder of this article. First, we formulate an ideational-institutional theoretical framework about framing contests and political strategies during crises. Second, we explore the presidential framing of COVID-19 in the United States. Third, we turn to the presidential framing of the pandemic in Brazil. Fourth, using our ideational-institutional framework, we compare the two countries directly to stress the similarities and differences between them before summarizing the main findings of this study and sketching an agenda for future research.

Framing, Credit Claiming, and Blame Avoidance

As we argue here, under Trump and Bolsonaro, the COVID-19 crisis in these two federal countries has led first and foremost to traditional framing contests shaped by existing political institutions: namely, federalism and party systems. Thus, this section focuses primarily on the framing of collective threats and how they interact with these political institutions.

A central aspect of crises is how political actors frame collective threats while pursuing blame avoidance and credit claiming strategies. This is the case because these actors can affect the ways in which such collective threats are perceived to advance their political and policy goals (Béland 2007; Rojecki 2016). To understand how this is possible, we turn to the literature on framing processes, before discussing their relationship to blame avoidance and credit claiming strategies.

First, frames are simply “symbols and concepts” (Campbell 2004, 94) used by political actors to shape the perception of reality and promote their political and policy agenda (on framing processes see also Benford and Snow 2000; Schön and Rein 1994). As understood here, the frames political actors deploy are strategic in nature because they are used to win electoral and policy battles “that are dialogical in nature in the sense that actors respond to their opponents by putting forward alternative frames to attack them and weaken support for their policy solutions” (Béland 2019, 20). These remarks stress the interactive character of framing processes, which are a form of strategic political discourse (Schmidt 2008). Because of their strategic nature, framing processes as studied in this article are inseparable from the actors who produce them. In other words, the analysis of these processes, just like the studies of ideational processes more generally, is necessarily actor-centric and grounded in a recognition of the agency of political actors, who combine existing cultural and ideological repertoires to produce certain strategic effects (Campbell 2004).

Second, during crises and beyond, frames are regularly employed in the blame-avoidance and credit-claiming strategies of political actors (see Weaver 1986, 2018). According to Kent Weaver (2018, 260), blame avoidance refers to how political actors seek to minimize their own “responsibility for unpopular actions taken.” As for the concept of credit claiming, it refers to how political actors seek to claim responsibility for popular actions and positive outcomes (Weaver 1986).

There is strong evidence that, during crises, framing contests among political actors—to define collective threats and legitimize policy responses to them—mesh with the concrete blame avoidance and credit claiming strategies of these actors. As Boin, ‘t Hart and McConnell (2009, 181) claim, crises “generate framing contests to interpret events, their causes, and the responsibilities and lessons involved in ways that suit their political purposes and visions of future policy directions.” This quote points back to the role of perceptions during crises and how political actors fight one another over the ways in which both collective threats and policy responses to them are framed. In the case of public health crises, across different political regimes, information control and expert advice are likely to play a direct role in the framing contests over the nature of the collective threats at hand and the potential policy responses to them involving political actors pursuing concrete political strategies, including blame avoidance strategies (Baekkeskov and Rubin 2017).

The ways in which these framing contests and political strategies unfold during crises are shaped by broader institutional legacies that set up the specific “rules of the game” that political actors must navigate to gain power and score electoral points. In the case of the United States and Brazil, federalism and the party system are central institutional factors shaping framing contests and related political strategies.

First, federalism diffuses power vertically by constitutionally guaranteeing the institutional autonomy of subnational entities, which play a direct role in many policy areas. Federal actors such as the president could blame subnational actors for negative policy outcomes during a crisis but also that, second, these actors can accuse the president and the federal government more generally of incompetence or wrongdoing to avoid taking responsibilities for policy failures that require intergovernmental cooperation. Conversely, federalism can blur the responsibility for “good news,” which facilitates credit-claiming attempts on the part of political actors located at different orders of government (Weaver 1986; on federalism and blame avoidance see Brown 2010; Maestas et al. 2008). This remark should not hide key differences between the two federal systems under consideration, as the U.S. system is more decentralized on average than the Brazilian one, something especially apparent in the field of health care.

Second, the vertical integration of political parties in federal systems and the relationship between existing parties and the president in power is another crucial institutional factor that may affect the ways in which framing contests and political strategies unfold during a crisis (see Ware 1995). In the case of Presidents Trump and Bolsonaro, we have two right-wing populist leaders who seek to foster political support from their base while seriously weakening democracy and public trust in government. Yet, President Trump is affiliated with the Republican Party, which has existed since the mid-nineteenth century as one of the two parties that dominate electoral politics in the United States. However internally divided and weakly institutionalized American parties are, party identification in the electorate remains strong (Azari 2016). Hence, Trump's support from the Republican Party in 2020 helps explain why he remained a highly competitive presidential candidate against Democrat Joe Biden, even if the latter ended up winning. As for President Bolsonaro, in the context of a more fragmented and weakly embedded party system, he has proved much more isolated since the failure of his Alliance for Brazil back in 2019, a situation that puts him in a vulnerable position vis-à-vis both subnational leaders (governors and mayors) and the National Congress. This is key as it pushed Bolsonaro to negotiate support with members of the National Congress.

In the following analysis, the features of federalism and the party system in each country are taken into consideration to analyze the framing and political strategies of Presidents Trump and Bolsonaro during the first months of the COVID-19 crisis. At this stage, a few methodological remarks about our empirical analysis are necessary. First, the rationale for our case selection is rather straightforward: the United States and Brazil are two federal presidential systems that have each been hit particularly hard by the COVID-19 crisis, as suggested below. Second, in both countries, the president played a central and controversial political role during the crisis. Third, Presidents Trump and Bolsonaro use a harsh, confrontational rhetoric against their enemies, which increased political tensions in each country during the COVID-19 crisis. Finally, despite these similarities, the existence of key institutional differences between the two countries allows us to see whether and how distinct institutional legacies (regarding federalism and the party system) shape the framing and political strategies of these two presidents.

In this article, we use a qualitative comparative case-study design to identify and explain key similarities and differences between the two countries under consideration. To analyze framing processes, we study the political discourse of each president as formulated in social media content, official speeches, and press conferences, among other primary sources. To link these framing processes to concrete blame-avoidance and credit-claiming strategies, we use a qualitative process tracing approach grounded in a systematic reading of both media accounts and academic literature on the two presidents and their respective countries and how they have handled the COVID-19 crisis. As for the timeframe of the analysis, we concentrate primarily on the first months of the crisis (especially from March to June 2020), as this period was especially crucial in terms of the framing contests over the pandemic, in part because COVID-19 was at the center of political discourse in both countries at the time.

United States

International rankings released prior to the pandemic suggested that the U.S. public health system was highly equipped to deal with COVID-19. Indeed, the United States ranked first among 195 countries on pandemic preparedness (NTI 2019). Yet, in contrast to these strong rankings, the U.S. political economy had deep defects that made collective action during a pandemic quite difficult. Institutionally, the public health authority in the United States remained highly decentralized, allowing for buck-passing from the federal government to state and local officials. Politically, partisan elites were starkly polarized––including on questions of public health and the role of experts––and public trust in government hovered at an all-time low (Pew Research Center 2021). Nevertheless, faith in the president among co-partisans in the electorate appeared buoyant to any number of prior crises (Gallup 2021).

Hence a Republican president, especially a figure like Trump, had few short-term political incentives to adopt a highly precautionary response to the pandemic. At the start of 2020, President Trump's hopes for re-election centered on the growing economy. In the State of the Union address delivered at the start of February the president proclaimed, “The unemployment rate is the lowest in over half a century. And very incredibly, the average unemployment rate under my administration is lower than any administration in the history of our country” (Trump 2020a). While Trump misrepresented the economic data, it was true that the official unemployment rate at the end of 2019 was a long-term low of 3.5 percent (Bureau of Labor Statistics 2020). For a president seeking re-election, this was a positive news. By this stage, COVID-19 was a known quantity and Trump had already put in place measures restricting travel from China to the United States and cases had already been reported in Washington state. Further, although it was not known until late April, the first COVID-19 related death in the United States occurred two days after Trump's speech on February 6 (Chapell 2020). Through February, however, Trump dismissed concerns that COVID-19 might disrupt American life. In September 2020 it became evident that President Trump knew at the start of February that COVID-19 was highly transmissible and posed a serious risk (Woodward 2020), but in a series of public statements through February and into March the president ignored the emerging scientific consensus and repeated that the virus would soon “go away” (Bump 2020).

That prediction proved reckless and, by early July, the United States had suffered over 130,000 deaths (John Hopkins Coronavirus Resource Center 2020). Many state governors responded by issuing “shelter in place” or “stay at home” orders through April and May (see Lee et al. 2020, for state-by-state closings and re-openings). These measures, along with growing public anxiety, inevitably had severe economic consequences. In the opening quarter of the year GDP fell by 5 percent (BEA 2020), even though economic activity had remained unrestricted through much of that period. The unemployment rate for May stood at 13 percent, which represented a much sharper and deeper decline than had even been the case during the Great Recession of 2008 and 2009 (Kocahhar 2020).

In this context, Trump started holding daily press briefings in an effort to regain control over the COVID-19 response narrative. From March 13 onward, Trump held court with the press for 24 straight days. In the early stages of the pandemic, the president's intent was to present himself as a wartime leader against the “invisible enemy.” The briefings themselves were seldom brief and were often erratic and confounded logic at times. Yet, amid the mixed messages, what was clear was that the White House was determined to maximize credit claiming and deny responsibility for any problems associated with the deteriorating public health situation. This meant shifting blame for the deteriorating public health situation through April on to other actors and institutions. Hence, the president insisted that the pandemic preparedness cupboard had been left bare by the Obama administration and he also regularly laid the blame for ongoing problems, such as shortages of necessary medical equipment, with state authorities.

This blame avoidance strategy was seen at its boldest in a press conference on March 13. When asked about the delay in ramping up the country's testing capacity for COVID-19, President Trump responded, “I don’t take responsibility at all.” Furthermore, he defended his administration's efforts by condemning the response of the Obama administration to the H1N1 pandemic in 2009, maintaining, “If you go back — please, if you go back to the swine flu, it was nothing like this. They didn't do testing like this” (Oprysko 2020). At a press briefing five weeks later, Trump elaborated again on why the failings of the previous administration explained ongoing problems with the roll out of testing: “I started with an obsolete, broken system from a previous administration,” he then added, “[u]nfortunately, some partisan voices are attempting to politicize the issue of testing, which they shouldn’t be doing, because I inherited broken junk” (Rupar 2020). Through this messaging, Trump was engaging in what Weaver (2018, 260) describes as “blame generating.” As a narrative, these statements were targeted at reinforcing the president's existing base of support by shifting responsibility for evident failures. This shift represented a backward-looking form of blame avoidance (i.e., blaming one's predecessors), simultaneously bringing Obama (one of Trump's favorite political foils) and his legacy into play. In blaming his predecessor, Trump's aim was to “weaken negative images” of his own (Weaver 2018, 268).

President Trump also mobilized blame-avoidance strategies by “passing the buck for decision-making to other policymakers” (Weaver 2018, 260): namely, state governors. These criticisms had two distinct aspects. First, Trump deflected worries about shortfalls in testing and the supply of medical equipment such as ventilators and personal protection equipment away from the federal government and on to the states. This was founded on the idea that the primary responsibility for tackling the pandemic did not lie with the federal government in the first place. Second, as the administration's focus shifted from combatting the virus to re-opening the economy the president attacked those governors he saw as too slow to ease lockdown restrictions. Emboldening Trump's move here was the structure of American fiscal federalism itself, specifically state and local governments’ inability to deficit spend. By refusing to sign legislation with general aid for state and local revenues, Trump forced states to choose between reopening their economies, accepting higher rates of infection and mortality, or enduring massive budget deficits (Rocco et al., 2020).

Notably, Trump's rhetorical strategy on economic reopening was not simply about “passing the buck” to the states but was underpinned by partisan motivation as the attacks concentrated on the actions of Democratic governors. For example, in mid-April the president said that states were not using all the testing facilities that they had to hand: “they don’t want to use all of the capacity that we’ve created. We have tremendous capacity… They know that. The governors know that. The Democrat governors know that; they’re the ones that are complaining” (White House 2020). Bizarrely, this blame avoidance was interspersed with credit claiming based on praise from some of those very same governors. In mid-April a video was played at the press briefing that included clips praising the president's response to the crisis from prominent Democratic governors: Andrew Cuomo of New York, Gavin Newsom of California, and Phil Murphy of New Jersey (Phillips 2020).

These contradictory communications about the president's relationship with the nation's governors extended to his public understanding of federalism. Trump, in fact, briefly claimed that he had the authority to force states to re-open their economies when attacking governors who he said were not acting quickly enough. In a tweet, the president wrote, “For the purpose of creating conflict and confusion, some in the Fake News Media are saying that it is the Governors decision to open up the states, not that of the President of the United States and the Federal Government. Let it be fully understood that this is incorrect” (Forgey and Gerstein 2020). Trump quickly backtracked from this claim as it was manifestly false, but the urgency of the administration's demand that states open as soon as possible contrasted with the White House's diffidence when governors asked that the federal government take a greater role in coordinating a response to the pandemic.

The messaging through these crucial months was therefore highly inconsistent if examined in a logical fashion. Trump veered from claiming absolutist-style authority to asserting that governors should lead the pandemic effort. He took credit for any progress and placed blame on others for bad news with little regard for the evidence. Simultaneously, he claimed to be leading the fight against the “invisible enemy,” yet played down the threat posed by the virus. But the overarching narrative had an internal logic, as it constantly found a rationale for blame avoidance and looked for ways of blame generation with regard to other political actors. Further underpinning this messaging was Trump's insistence that the economy be unlocked as quickly as possible, directly in defiance of his own administration's guidelines on when it would be safe for states to re-open. Again, Democratic governors were a primary target. For example, on a visit to Maine in early June, the president compared the state's governor, Janet Mills, to a “dictator” and told residents “She's going to destroy your state” if she did not lift restrictions in time for the summer tourist season (Whittle and Colvin 2020). Statements such as this reflected how Trump's focus had switched away from the dangers of COVID-19 as he concentrated on promoting the notion that the economy was on a path to recovery and that it was safe to re-open.

Further, even as thousands of Americans were still dying every day in early to mid-June, Trump insisted that the spread of the virus had been checked and that the economy was rebounding. In one briefing he asserted, “We’ve made every decision correctly” and, while acknowledging that “We may have some embers or some ashes or we may have some flames coming,” he insisted “we’ll put them out. We’ll stomp them out” (Trump 2020a). The primary focus was the improvement in the job numbers announced in June. Talking about the shape of the recovery Trump declared, “And now we’re opening, and we’re opening with a bang. And we’ve been talking about the ‘V.’ This is better than a ‘V’; this is a rocket ship.” (Trump 2020a). The same priority was emphasized at another briefing in mid-June, “last month, we added a record 2.5 million jobs. It's the highest in the history of our country in one month. That's the highest number of jobs: 2.5 million. The Dow rose above 26,000 points. And we saw the largest surge in retail sales ever recorded” (Trump 2020b).

In his remarks, President Trump did refer to the actions taken by Congress, such as a payment of $1,200 per adult and $500 per dependent child for people that earned up to $75,000 in 2019 and noted how the Payment Protection Plan had helped many businesses damaged by the pandemic. In fact, the major stimulus packages that were enacted by Congress, notably the CARES Act, which amounted to a nearly $2 trillion commitment were passed with near unanimous votes in a manner belying the constant partisan friction in Washington D.C. The White House was represented in the discussions over the CARES Act primarily by Treasury Secretary Steve Mnuchin, with media reports suggesting that he was willing to make more concessions to Democrats than some Republicans were comfortable with (Hulse 2020). Pre-existing policy legacies and institutional fragmentation hindered the distribution of monies promised in these initial economic support packages (Rocco, Béland, and Waddan 2020), but the spending commitments involved were in line with ‘big’ rather than ‘small’ government visions.

In fact, across the world the response to COVID-19 illustrated the underlying authority of the state as well as its unmatched fiscal capacity in a way that pushed past conservative principles. Orders to lockdown and huge, unfunded, spending commitments trespassed against libertarian instincts and established conservative economic orthodoxy. Yet, despite these tangles, it might be argued that Trump's campaign in 2016—with its defense of Social Security and Medicare—had shown that he was not moored to those conservative orthodoxies and that he might therefore be receptive to continued expansive government intervention if it helped sustain the economy. But, as the crisis evolved, there was little sign of a coherent set of ideas guiding the White House's response, beyond the insistence on an end to ‘stay at home’ instructions. Furthermore, by the end of June it was increasingly evident that the initial relatively swift and bipartisan response by Congress was unlikely to be sustained in the longer term. For example, the House did pass a further stimulus package known as the HEROES Act, which offered a further $3 trillion in aid packages, but did so along partisan lines as Republicans in both Congress and the White House denounced the bill as a liberal fantasy (Cochrane 2020).

That the president engaged in blame avoidance in an election year over life-and-death issues is hardly surprising. Further, the problem of specifying quite where responsibility lies given the complexities of intergovernmental relationships in the United States makes it plausible to shift responsibility for poor outcomes. As Hood (2010, 5) notes, “Crooked and ambiguous accountability trails may not serve democracy or good governance. But they can protect the political and administrative class from blame after failure.” Also, while institutional fragmentation in Washington D.C. can be frustrating and presidents will often blame Congress for inaction, especially if one or both chambers are controlled by the opposition party, the relatively quick legislative action as the crisis developed allowed Trump to claim credit. When signing the CARES Act, Trump (2020c) said “This is a very important day. I’ll sign the single-biggest economic relief package in American history and, I must say, or any other package, by the way. It's twice as large as any relief ever signed.” At that point, the president acknowledged the co-operation of the Democrats, but as the death toll rose through April and May the partisan finger-pointing became sharper and the normal institutional barriers to legislative action came back into view.

In the final stages of the presidential campaign, as the infection rate across the United States hit new highs, Trump shifted blame for public anxiety about COVID-19 on to the media. At a rally in North Carolina, he insisted: “That's all I hear about now. Turn on television, ‘Covid, Covid, Covid, Covid, Covid, Covid,’…By the way, on November 4, you won’t hear about it anymore” (Berenson and Bennett 2020). Yet, overall, when discussing the blame avoidance and blame-generating and credit-claiming strategies employed by the White House through the opening months of the COVID-19 crisis, it is important to note that these were largely unsuccessful in political terms, at least if measured by Trump's popular support. While Trump gained a slightly larger share of the popular vote than he did in 2016, his Democratic rival—former Vice President Joe Biden––ultimately won a decisive majority in both the popular vote and the electoral college. True to form, however, Trump refused to concede, spreading myths about massive voter fraud, filing baseless lawsuits, and pressuring state election officials to recalculate vote tallies. This culminated in the extraordinary scenes of January 6, 2021, as rioters, claiming to act on Trump's behalf, stormed the Capitol Building to prevent the certification of the Electoral College vote. By that point, Trump paid minimal attention to COVID-19, even as 4,000 Americans died from the infection on the day of the insurrection. Hence, even as the country was gripped by crisis, the patterns of partisan division and institutional fragmentation that confused lines of responsibility not only persisted but were exacerbated by the White House's emerging political strategies.

Brazil

Brazil's health care system was relatively well prepared to deal with the COVID-19 pandemic for two reasons. First, the Unique Health Care System (Sistema Único de Saúde, SUS) was institutionalized through different changes promoted since the approval of the 1988 Constitution to ensure health care universalization and promote coordination in service provision (Viana et al. 2008). Second, Brazil had a legacy of successful public health responses to crises such as HIV (Nunn 2009). Yet, the Ministry of Health's role in national coordination, an important feature of SUS, as health care decision-making is highly centralized, has been weakened since the beginning of Bolsonaro's government. Even though states and municipalities play a key role in service provision, the absence of national coordination created challenges for the coordination of governmental responses during the COVID-19 pandemic (Abrucio et al. 2020).

In the first months of 2020, Bolsonaro was starting his second year as president of Brazil. Since the beginning of his presidency in early January 2019, Bolsonaro had faced challenges in having the National Congress approve his legislative proposals. Back in January 2019, he had significant support of the newly elected members of the National Congress, but he had not built a coalition with other parties and lost support there over time, especially when left the Social Liberal Party (Partido Social Liberal, PSL). Although the presidency has powers over setting the national agenda (Figueiredo and Limongi 1995), it must build a coalition that involves negotiations and distribution of “goods” such as cabinet positions due to the highly fragmented nature of the legislature (Raile, Pereira and Power 2011). As a result, by the time the social security reform (one of the major changes made in 2019) was finally adopted, the executive's proposal had been heavily revised by the National Congress. Moreover, President Bolsonaro had not advanced key issues related to both his conservative and liberal agendas.

The COVID-19 pandemic emerged within a context of stagnant economic growth and rising unemployment, which were both aggravated by the economic and fiscal crises the pandemic caused. Considering this dire pre-existing economic situation in the country, President Bolsonaro feared the pandemic's negative economic and fiscal impacts. As he said at the beginning of March 2020, as the threat of the global pandemic became increasingly direct: “if the economy sinks, my government will end, any government would end” (Bolsonaro, quoted in Vasconcelos 2020). The impact COVID-19 could have on the economy led him to minimize its relevance, saying that “it is just a flu,” “other flus killed more people,” and “a lot of what people are saying about it is fantasy” (Bolsonaro, quoted in Vasconcelos 2020). He delegitimized media coverage, insisting that COVID-19 was not as bad as the media had made it look. The president stopped speaking to the press and, instead, started to address his supporters directly in front of his cabinet office. He also publicly opposed physical distancing and the closure of schools and businesses, denying international guidelines and scientific evidence. Instead, he only supported physical isolation for elderly people and other at-risk individuals, as the others should “go back to normal” so that “jobs should be maintained, people's income should be preserved” (Bolsonaro 2020). Reinforcing this message, he himself attended a public demonstration organized by his supporters, who backed the closure of the National Congress and Supreme Court.

From the end of January until the beginning of March (when Brazil had 252 suspected cases) states and municipal governments in jurisdictions with high infection rates such as São Paulo took the lead, elaborating public health plans and creating committees with the support of the federal Ministry of Health. After being notified of the first cases, state governors and mayors promoted physical isolation by telling residents to stay home while restricting economic activities. This was followed by more restrictive measures, including closing schools and non-essential businesses. As the virus spread, other states and municipalities across the country followed suit. States and municipalities also spearheaded efforts to increase hospital capacity and acquire testing material and health care equipment, especially respirators (see Pereira, Oliveira and Sampaio 2020).

While states took the lead in responding to the pandemic, political conflict between them and President Bolsonaro emerged. At this point, he was more concerned with deploying blame avoidance, shifting blame to governors for both the economic downturn and the increase in the number of COVID-19 deaths and infections. He also stated that governors and mayors were overacting to the public health crisis: “certain governors are taking extreme measures that are not within their power such as closing airports, roads, malls, and street markets” (Bolsonaro, quoted in Vasconcelos 2020). National decisions, particularly presidential decrees, sought to limit subnational powers or excluded them from decision making. Examples of this include the Law No. 13,979 (February 6) approved by the National Executive and Legislature, naming the Ministry of Health as the main actor in pandemic response and requiring subnational governments to ask for federal authorization when getting involved in such a response; the President's Decree No. 10,282 (March 20) that allowed churches and other organizations to reopen during the pandemic, contradicting governors and mayors’ pandemic guidelines, which was annulled by the judiciary a few days after it was issued; the President's Decree No. 10,289 (March 24) that created a committee to coordinate and monitor different Ministries’ pandemic initiatives that only included representatives of federal agencies, thereby excluding states and municipalities; and the President's Decree No 10,292 (March 25) that granted the federal government exclusive power to define regulations for interstate and inter-municipal activities.

Moreover, there was an intensification of an ongoing framing and blame contest between the president and the governors centered on the issue of who was responsible for the bad economic and public health news stemming from COVID-19. Even though the federal government transferred state resources for tackling the pandemic to subnational governments, states complained about delays in receiving respirators, testing material, and federal funds and demanded increased resources to alleviate their difficult fiscal situation. They also called on the federal government to suspend the collection of debts and other payments states make. In this context, the Governor of Rio de Janeiro, Wilson Witzel, stated that “the Union concentrates revenues, funding” but that the federal government had to understand that “we need to decentralize [fiscal] resources to states; in this moment, we need liquidity. It is necessary to ensure the federative pact” (Witzel quoted in Caramuru et al. 2020). As for the Governor of Maranhão, Flávio Dino, he claimed that “yesterday Bolsonaro talked about the ‘pact’ but started the day today by blaming governors for [Brazil's] economic problems, solutions for which states depend on the federal government.… Bolsonaro should allow the Ministry of Health to coordinate a unified national action plan based on technical criteria” (Dino, quoted in Moraes 2020).

At this point, against the president, the federal Ministry of Health took the lead in supporting state and municipal decisions. Minister Luz Henrique Mandetta became the most important actor during this episode, giving updates to the media at daily press briefings. He supported the creation of a “favorable environment” for cooperation between the federal government, states, municipalities, and the private sector (Mandetta, quoted in Cantanhêde 2020). He also contradicted President Bolsonaro, framing his decisions as evidence-based and, Bolsonaro's as anti-expertise. Simultaneously, when President Bolsonaro advocated for the use of the controversial drug hydroxychloroquine, Mandetta stated: “I only work…with science” (Mandetta, quoted in Nomura 2020). He also openly criticized President Bolsonaro for walking in a crowd without wearing a mask while shaking people's hands. President Bolsonaro, like President Trump, defied his own Ministry's public health guidelines, and, on April 16, 2020, he fired Mandetta altogether.

After Mandetta's dismissal, the Ministry of Health's response to the pandemic demonstrated a lack of national leadership and coordination in the public health policy response to the pandemic. Even though there is a national health system with shared responsibilities between the three levels of government, the federal dynamics that characterize the response to the pandemic are similar to those witnessed in the United States. States and municipalities issued public health guidelines on their own, without a stronger form of institutionalized coordination. This fragile coordination enhanced political competition among governors, who fought over issues such as the acquisition of respirators. ²

Despite this lack of leadership, in the following months, the president continued to advocate for the use of hydroxychloroquine and ivermectin, and the Ministry of Health allowed physicians to recommend these drugs to treat COVID-19 patients. In mid-late January 2021, in the city of Manaus (Amazonas), hospitals ran out of oxygen, which immediately led to hundreds of additional COVID-19 deaths. A few days before this “complete massacre,” the Minister of Health, Eduardo Pazuello, “an army general with no medical experience,” had visited the city to promote the use of these two drugs, instead of addressing the looming oxygen crisis in the city's hospitals (Phillips 2021).

However, the determinations of the 1988 Constitution regarding a decentralized model in which subnational governments are responsible for health care provision played a role in Brazil. In April 2020, the federal government's decision that it was responsible for issuing public health guidelines about “essential” activities and that states and municipalities had a secondary role was revoked by the federal Supreme Court. The Supreme Court decided that the Union, states, and municipalities have concurrent responsibilities in managing public health crises. The federal government, the Court ruled, is responsible at the legislative level but it should respect subnational autonomy (Supremo Tribunal Federal 2020).

Here President Bolsonaro lost on the legal front, but he used the ruling to avoid blame, stating that the Supreme Court decision had indicated that primary responsibility for tackling the pandemic lay with subnational units (i.e., states and municipalities), not the federal government. He said that governors and mayors “can re-open Brazil, not me… I do not have any power. …The Supreme Court said so. What do you want me to do? The Supreme Court decided: governors and mayors close and open [Brazil]” (Bolsonaro, quoted in Frazão 2020). The president continued: “[if you] believe that quarantine, the measures taken by your state are harming you – the appropriate forum for your complaints is your respective governor” (Bolsonaro, quoted in Frazão 2020). In this context, by the end of May and the beginning of June, governors and mayors re-opened or eased lockdown restrictions. It is important to mention that the blame contest did not end there, as the president publicly criticized state governors about vaccine agreements, especially the case of São Paulo's governor, who had signed an agreement with a Chinese company.

Another difference with the United States is that, unlike the situation prevailing there, the blame the Brazilian president put on governors was not partisan in nature. Regardless of partisan identities, Bolsonaro blamed all governors for the economic downturn and “passed the buck” to them and to the mayors, regardless of their partisan affiliation. This reality is related to the temporally changing nature of President Bolsonaro's own partisan identity. At the time of his election to the presidency in October 2018, Bolsonaro was affiliated with the PSL. ³ He left the party at the end of 2019 and tried to create a new party, the Alliance for Brazil, which failed. This means that he was not affiliated to a political party during the first months of the pandemic. Though only two governors were affiliated with the PSL, during the 2018 elections half of the 26 governors had declared support for Bolsonaro's candidacy. However, he lost some of this support during the pandemic, and almost all governors criticized or opposed his statements and decisions at one point or another, regardless of the differences in their own public health decisions (i.e., more or less restrictive) (Pereira, Oliveira and Sampaio 2020).

President Bolsonaro also claimed credit for the Auxílio Emergencial (Emergency Benefit) for individuals who had lost their income during the pandemic approved in beginning of April, a result of the National Congress’ efforts. The federal government opposed the initial proposal of transferring 200 reais ($35 USD) per month to citizens. Despite this opposition, at the end of March, the National Congress approved the benefit of 500 reais ($86 USD). The federal government then increased it to 600 reais ($104 USD) a month and sought to claim credit for the policy, and the federal government extended payments until December 2020, with a reduced amount from September to December. In April, more than 40 million people received the benefit, and by August, this number increased to 67 million.

By the end of April 2020, President Bolsonaro was politically isolated. The National Congress approved bills that he had initially opposed, such as the Emergency Benefit and Law No. 173 (passed May 27, 2020), which offered financial relief to subnational governments. At the same time, the National Congress rejected various executive orders (Medidas Provisórias, MP). President Bolsonaro also feared the National Congress would initiate impeachment proceedings against him because the body had received multiple requests to do so. Brazil's multiparty system played a role in changing Bolsonaro's precarious situation, allowing him to negotiate with Congress members linked to centrist parties.

Amidst all of this political posturing and strategic framing, the number of COVID-19 infections and deaths in Brazil kept increasing. By August 2020, the number of reported COVID-related deaths stabilized to approximately 1,000 per day, and there was still considerable under-reporting. According to Barberia and Gómez (2020), “playing to the public's fears during a pandemic does not seem to have strengthened his political support and authority.” Yet it seemed that Bolsonaro's strategies served to maintain his support levels. If Bolsonaro lost supporters in the first months of the pandemic due to the federal government's problematic policy responses, he gained other supporters “attributed to the emergency aid money given out during the pandemic” (Gomes 2020). Moreover, when the country reached a total of 100,000 COVID-19 deaths in August 2020, only 11 percent of Brazilians stated that they blamed President Bolsonaro for this negative outcome while 47 percent did not believe Bolsonaro was responsible for it (DataFolha 2020). This suggests that, during the first months of the pandemic at least, presidential framing strategies proved quite successful in blurring accountability lines and avoiding blame for almost half of Brazilians, even though, at the beginning of 2021, he lost support and more Brazilians blamed him for the increasing number of COVID-19 deaths, the crisis in Manaus, and the delays in vaccination.

Discussion

Our comparative analysis reveals that, in their framing strategies, both President Trump and Bolsonaro emphasized the economic menace posed by the public-health response to COVID-19 while deemphasizing the public-health threat itself. These approaches had tremendous human and policy consequences while leading to dramatic, yet country-specific, political conflict shaped by distinct institutional legacies tied to federalism and the party system. In other words, despite the exceptional nature of their presidencies, both Trump and Bolsonaro relied on traditional political strategies embedded in existing institutional legacies.

First, regarding federalism, in both countries framing contests and blame games took place between the president and subnational leaders, especially governors. In the United States as well as Brazil, the president blamed subnational leaders for the COVID-related economic downturn because of their allegedly excessive public health measures. Simultaneously, as the death toll increased in each country, these presidents blamed governors and sometimes mayors for the situation, a reality made possible by the accountability challenges stemming from federalism as an institutional framework. In Brazil, the shared public health responsibilities among the three orders of government (federal, state, and municipal) as well as the Supreme Court's decision that reiterated them also blurred accountability lines. Overall, in both the United States and Brazil, federalism shaped framing contests and the blame games they reflected and amplified.

Second, and relatedly, partisanship played a major structuring role in how both leaders engaged in framing contests. Yet, while federalism accounts for key similarities between the two countries, the divergence in party systems helps to explain differences between them as far as the presidential framing of the COVID-19 crisis is concerned. On one hand, in the United States, partisanship in the form of the Democratic-Republican dichotomy shaped framing contests and blame games involving the president and subnational actors. This is the case because the Republican president mainly attacked Democratic governors and vice versa. In the United States, such a close relationship between federalism and partisanship in the politics of blame is a well-documented trend that long preceded the advent of the Trump administration, in normal times and during crises (Brown 2010; Maestas et al. 2008). On the other hand, in Brazil, partisanship played a much more limited role in framing contests. This is tied to the fact that the party system in Brazil is much more unstable and fragmented than the system of ‘weak parties and strong partisanship’ found in the United States (Azari 2016). Such a chaotic situation made it possible for President Bolsonaro to change party affiliation or even govern as an independent, something that would be unlikely to happen in the United States. This partisan reality changes the rules of the blame games and framing contests by making President Bolsonaro particularly vulnerable to attacks coming from different sides of the political spectrum, at both the national and the subnational level. Conversely, the fact that he did not belong to any political parties allowed him to both criticize all the governors and to strike deals with various members of the National Congress, regardless of partisan affiliation. In terms of framing contests and blame games, President Bolsonaro's lack of partisan affiliation during the first months of the pandemic was thus both a curse and a blessing

A focus on the framing strategies of both presidents reveals how gaps in leadership can accentuate the weaknesses, and undermine the strengths, of political institutions’ crisis-response capacity. In both the United States and Brazil, however, that leadership and communication was not forthcoming through 2020. In an analysis published in the British Medical Journal in 2019, a team investigated national preparedness for an infectious disease epidemic or pandemic (Oppenheim et al. 2019). The results showed five levels of preparedness, with the United States in the top rank and Brazil in the second. Critical to these rankings were assessments of public health and physical infrastructure. Also central were what the authors describe as ‘institutional capacity’ and ‘public health communication.’ The former reflected on public administration systems and their capacity for implementing policy quickly and effectively. The latter required “effective systems to identify salient information gaps (or potentially hazardous rumors and misinformation), craft and adapt messaging and rapidly disseminate it to the population” (Oppenheim et al. 2019, 4). In both regards the leadership of Trump and Bolsonaro, by adopting political strategies that focused on what they perceived to be their personal advantage rather than promoting good governance, undermined institutional capacity and turned the potential strengths of federalism into weaknesses. Trump's and Bolsanaro's contestation of the evidence and the resultant contradictory public advice led to disjointed implementation of public health measures. In the U.S. case, while some Republicans did break from Trump in unambiguously acknowledging the public health threat of COVID-19, partisan polarization mostly further reinforced the different messages coming from state leaders. Brazil exhibited a less starkly two-dimensional partisan messaging, but this did not lead to a uniform narrative about the balance of risk caused by COVID-19. Bolsonaro's was the most loudly expressed opinion but the multiple voices criticizing and sometimes supporting him undermined the chance of consistent communication in a system with dispersed political authority embedded throughout federalism.

The gravity of the COVID-19 crisis did not displace the framing contests and blame games that have become a dominant feature of presidential politics in the United States and Brazil over the last few decades. In fact, if anything, it provided ample opportunity for opportunistic political leaders to leverage party systems and federal structures to reassign blame and turn the crisis into a political cleavage. While Trump and Bolsonaro did much to weaken democracy and reduce trust in government, their behavior and framing strategies during a pandemic they both mismanaged in a deadly fashion suggest that they still relied on rather traditional tactics deeply embedded in existing political institutions. This is a reality that future research on these two presidents and, more broadly, on framing processes in times of crisis should keep front and center.

Theoretically, the above analysis points to the fact that existing political institutions, including federalism and party systems, shape the framing strategies of political leaders, during crises and beyond them. Although political leaders have the autonomy and pursue their own strategies, political institutions create constrains and opportunities for them, a reality consistent with institutionalist approaches to politics and framing processes (Campbell 2004; Schmidt 2008). Our analysis suggests that even political leaders such as Trump and Bolsonaro, who attempt to weaken and sometimes even subvert democratic institutions, adopt framing strategies that reflect at least in part the enduring weight of these very institutions. Future comparative scholarship on framing processes during and beyond global crises would benefit from taking these findings into account.