The Developmentalist Project of the PT Governments: An Economic and Political Assessment

Abstract

An analysis of the endogenous and exogenous political and economic factors that conditioned the Partido dos Trabalhadores’s (PT) social-developmentalist project in 2003–2016 in the light of financialization and the “confidence game” conditioned by the volatility of external liquidity and commodities prices concludes that the first Lula administration faced the problem of a crisis of confidence and adopted orthodox policies but was able, with the improvement of international conditions, to launch policies of a more interventionist and distributive trend. Dilma Rousseff, facing a downright unfavorable international context, explicitly broke with the confidence game by applying the policy set of the new macroeconomic matrix. In her second term she radically reversed the policy orientation, moving toward a strong fiscal adjustment and monetary orthodoxy, and this eventually undermined her few sources of political support. The economic crisis from the second half of 2014 on undoubtedly contributed to the political crisis, which in turn made infeasible any attempt to implement policies to reverse the situation of economic crisis. Dilma’s impeachment finally interrupted the PT’s developmentalist project, allowing the emergence of new political actors.

Uma análise dos fatores endógenos e exógenos, políticos e econômicos que condicionaram o projeto social-desenvolvimentista do Partido dos Trabalhadores (PT) em 2003–2016 à luz da financeirização e do “confidence game” condicionado pela volatilidade dos ciclos externos de liquidez e preços de commodities conclui que o primeiro governo Lula enfrentou o problema de crise de confiança e adotou políticas ortodoxas, mas pôde, com a melhoria nas condições internacionais, adotar políticas de perfil mais intervencionista e redistributivista. Já Dilma Rousseff, embora enfrentando contexto internacional francamente desfavorável, rompe explicitamente com o “confidence game” ao assumir o conjunto de políticas da Nova Matriz Macroeconômica. Na transição do primeiro para o segundo mandato, Dilma inverteu radicalmente a orientação das políticas, partindo para um forte ajuste fiscal e a ortodoxia monetária, o que acabou minando os poucos focos de sustentação política com os quais contava na sociedade. A crise econômica a partir do segundo semestre de 2014 sem dúvida contribuiu para dar origem à crise política, e esta por sua vez inviabilizou qualquer tentativa de implementação de políticas para reverter o quadro de crise econômica. O impeachment de Dilma, por fim, interrompe o projeto desenvolvimentista do PT, permitindo a emergência de novos atores políticos.

Keywords

Developmentalism Partido dos Trabalhadores Lulism Lula da Silva Dilma Rousseff

After the disappointment regarding the outcomes of the liberal reforms carried out in Latin America in the 1990s, the region witnessed, at the beginning of the new century, a strong renewal of the discussion on the viability of developmentalist projects. In this context, two new theoretical perspectives emerged: social developmentalism and new developmentalism, both updating the classical developmentalism formerly advanced by the Economic Commission for Latin America and the Caribbean (ECLAC) but adding new dimensions to the concept. Specifically, both supported a domestic strategy whereby state activism was required for structural change focused on (re)industrialization and a broad process of social transformation.

Although the renewal of the debate originated in academia, its transformation into an economic policy agenda took place under the governments of the Partido dos Trabalhadores (Workers’ Party—PT) starting in 2003. The central objective of this paper is to defend the plausibility of this hypothesis, to which we add the following subsidiary questions: First, we refute the thesis of a lack of differentiation between left-wing and right-wing parties on the ideological spectrum of Brazil. Confirming the existence of a project specifically concerning the PT, we argue that elections and the political system offer the society public policy alternatives. Secondly, we briefly revisit the debate over Lulism (Singer, 2009) by examining the extent to which this phenomenon differs from the broader PT project. Supposing that a developmentalist inspiration was present from the beginning of the Lula administration and culminated in more radical efforts by the first Dilma administration, Lulism can be interpreted, in line with Samuels and Zucco (2018), as an epiphenomenon of the PT rather than as something ontologically distinct. This discussion brings us, finally, to whether the developmentalist project of the PT governments was classical, social, or new developmentalist. Reflecting on the reasons that policy makers adopted different strategies at different times, we seek to shed light on the external or internal, political or economic constraints on the broader project of political differentiation of the PT. From this perspective, we argue that two dimensions conditioned the dynamics of the project’s advance or retreat: whether the constraints were exogenous or endogenous and whether they were political or economic.

In this study we will present, in the light of the available literature, a critical assessment of the PT’s developmentalist project, identifying possible reasons for its peak and crisis. We divide the article into three major sections. The first deals with the nature of the party’s project. The second examines some conditioning factors of the performance of the PT governments: (1) financialization and the reduction of space for developmentalist policy, (2) how external and internal factors, both economic and political, combined to favor or limit the implementation of developmentalist policies, (3) the pattern of interlocution between political and societal actors, and finally (4) the economic effects of Operation Car Wash. The third and last section draws conclusions.

The Nature of the PT Governments’ Developmentalist Project

The Definition of Developmentalism

The concept of developmentalism encompasses different premises depending on the author and the period. Two essential references are Bielschowsky (1988) and Fonseca (2014). Bielschowsky (1988: 247) defines developmentalism as a “planned and state-supported industrialization project.” Its key concept is the transformation of Brazilian society according to an economic project including (1) comprehensive industrialization as a way of overcoming underdevelopment, (2) awareness that efficient and rational industrialization cannot be achieved through the spontaneous operation of market forces, (3) planning (identifying the desired expansion rates of economic sectors and instruments to promote that same expansion), and (4) state execution of the expansion, assembling the financial resources and promoting direct investment where private initiative is insufficient (Bielschowsky, 1988: 7).

Fonseca (2014), in turn, considers two analytical keys for developmentalism—ideas and political practice—and seeks to determine the intersection of these two dimensions. His method has two stages: researching the “attributes used by various authors in expressing their understanding of what developmentalism is” and determining the attributes of the “historical experiments normally listed in the literature as examples of developmentalism” (Fonseca, 2014: 7). From this examination, he extracts the common core of the concept: interventionism, industrialization, and a national project. Industrialization is understood as the only way to promote national development—“to accelerate economic growth, productivity, and technical progress” (13). Interventionism is the involvement of the state administrative apparatus as the coordinating entity of the project, an instrument of national development. Nationalism is the prioritization of the national interest instead of fragmentary interests.

The PT Project: Social Developmentalism or New Developmentalism?

Two questions arise in the discussion of the nature of the PT governments’ project. The first is whether it was actually developmentalist. As we have seen, the canonical concept of developmentalism assumes a national project, state interventionism, and industrialization. From this perspective, therefore, there was no developmentalist project, since, although the first Dilma government did adopt an unsuccessful industrialist agenda (Mello and Rossi, 2018), the party did not reverse or deepen the ongoing deindustrialization. The reasons for the failure of Dilma’s industrialist agenda are the subject of significant discussion, with some writers highlighting changes in the international scenario (Sarti and Hiratuka, 2018) while others stress inconsistencies in the policies adopted by the PT governments, especially the first Dilma government (Carvalho, 2018; Paula, Modenesi, and Pires, 2015). As for social advances, there is some consensus on important gains in terms of reducing unemployment, moderate improvement in income distribution, and greater access for the lower-income segments to utilities and basic appliances. Thus, in terms of the canonical concept of developmentalism, it is doubtful that the PT’s project was in fact developmentalist, but it can perhaps be identified as “social developmentalist.”

This brings us to the second question, which is what kind of developmentalism the project represented. Because of the failure of neoliberal policies in the 1990s, neodevelopmentalist strands (new developmentalist and social developmentalist) have emerged to revive the discussion of developmentalism as part of the ECLAC structuralist tradition.¹ Social developmentalism (Bielschowsky, 2012; Carneiro, 2012) argues that the growth of the domestic mass market must be stimulated by both employment expansion and improved income distribution. Further, since a mass-consumption-based growth strategy loses momentum over time, expansion must be complemented by public investment in infrastructure. The policies advocated include wage policies (minimum wages being powerful instruments for increasing demand and redistributing income), social transfers directed to the poorest segments of the population, promotion of consumer credit, public investment in infrastructure to create incentives for private investment, industrial policies, and financing subsidized by public banks. Therefore, this approach is a kind of wage-led growth strategy—pursuing growth through an increase in the proportion of wages in the national income, which induces investment and profit (Ferrari-Filho and Paula, 2016).

New developmentalism (Bresser-Pereira, 2011; Bresser-Pereira, Oreiro, and Marconi, 2016), in contrast, emphasizes the two fundamental problems of middle-income economies: the tendency of wages to increase less than productivity because of the abundance of labor supply and the tendency toward overvaluation of the exchange rate as a result of the combination of the “Dutch disease”² with abundant flows of foreign capital toward emerging economies, attracted by their generally higher profit and interest rates and also the policy of growth with foreign savings. This type of developmentalism advocates policies for catching up: an income policy that promotes wage increases in line with productivity and an exchange-rate policy that neutralizes exchange-rate overvaluation, aiming at a rate that allows domestic producers to compete in foreign markets with fair profit margins. The (re) industrialization process is, thus, directly linked to the goal of obtaining surpluses for the manufactured exports in order to stimulate new investments in this sector. Under this strategy, a developing economy must resort to an export-led growth strategy³ for catching up in a short time.

Prates, Fritz, and Paula (2019) have shown that the PT governments’ developmentalist project combined elements of both social developmentalism and new developmentalism together with orthodox economic policies. In general, however, these governments sought to stimulate, by an actual increase in minimum wages, and social transfers and credit expansion, growth based on mass consumption—a predominantly wage-led growth strategy⁴ that was closer to the social developmentalist approach. The first Lula administration, because of the 2002 crisis of confidence, started with a very orthodox economic policy, with high interest rates and high primary surpluses (averaging 3.5 percent of the gross domestic product [GDP] in 2003–2006), to cope with external imbalances and rising inflation whose contractionary effects were felt less acutely because of the beginning of the commodities boom that favored Brazilian exports.

After 2006 the government adopted a less conventional set of policies, among them the accumulation of foreign exchange reserves, an increase in the minimum wage and in disbursements from the Brazilian Development Bank, and a plan to accelerate economic growth through increased public and private investments in infrastructure. In response to the international financial crisis in September 2008, a set of countercyclical policies that included a credit policy via public banks, tax exemptions for consumption of durable goods, increases in public investment, expansion of credit through the Brazilian Development Bank, and the creation of the popular housing program My House, My Life was successful in addressing its impacts on the Brazilian economy.

In the first Dilma administration, there was a change in the direction of economic policy in the midst of the deterioration of the external scenario (the international crisis and deceleration of the Chinese economy). From mid-2011 on, the government adopted a set of measures called the “new macroeconomic matrix,” which included a reduction in the SELIC interest rate by the Central Bank (from 12.5 percent in July 2011 to 7.5 percent in August 2012), devaluation of the currency (by 25 percent from August 2011 to May 2012), intensive use of the Brazilian Development Bank in subsidizing credit lines to finance investments, the promotion of reindustrialization with the Greater Brazil Plan, tax exemptions (particularly for payrolls), an infrastructure plan (the Logistics Investment Program), a reform of the power sector, capital controls, a tightening of monitored prices (energy and oil), and protectionism (Singer, 2018).

Finally, in 2015, at the very beginning of her second term, Dilma made a sharp turn, adopting a policy switch and the questionable discourse of “expansionist fiscal contraction,”⁵ with a set of measures that included fiscal adjustments in public spending, even higher interest rates, accelerated adjustment of managed prices (energy and oil), and the elimination of subsidized credit.

Regarding social policies, these were active throughout the entire PT government, being key instruments for achieving the goals of income redistribution and the promotion of domestic consumption. The two main factors that contributed to improving the distribution profile were the significant increase in the minimum wage (66.9 percent in real terms from December 2003 to December 2014) and the income transfer policy anchored in the increase of social security benefits and the Bolsa Familia (family subsidy) program.

In summary, the PT governments adopted a social-developmentalist approach in both social policy (in particular the minimum wage increase) and economic policy (public investment, credit to families, and the actions of public banks). The key instrument of new developmentalism—maintaining the exchange rate at a competitive level—was adopted only for a short period of time and in the context of a poorly coordinated economic policy.

Implications for the “PT-Versus-Lulism” Debate

In an influential work, Singer (2009) traces the origin of Lulism to the 2006 elections, which, according to him, triggered a deep and significant electoral realignment, a change in the boundaries of electoral loyalties in a long political cycle. In the context of a favorable international environment that allowed President Lula to promote redistributive and antipoverty policies, a distinction emerged between the bases of support for the president and those for his party. Lulism was characterized by the manifestation of support for Lula by the subproletariat, a class fraction that benefited from the public policies that were being implemented. This subproletariat, heavily concentrated in the North and Northeast, was highly vulnerable because of its low income, employment, and general subsistence conditions but according to Singer nevertheless contributed to a change in the sociopolitical base of PT support, rearranging the class foundations of the existing power bloc.⁶ In a sense it was stronger than the PT in that it configured a change in the social composition of voters for leftist presidential candidates, and it became the pivot of a pragmatic turn toward “weak reformism”—a gradual shift in the state’s emphasis on social policy and the abandonment of any radical program or direct confrontation with capital (Singer, 2009; see also Loureiro and Saad-Filho, 2019).

Samuels and Zucco (2018) adopt a different perspective. Analyzing the distinct socioeconomic underpinnings of Lulism and PT supporting bases, they detect, as does Singer, a poorer, nonwhite, and less educated profile for supporters of the presidential candidate while the profile of the PT’s declared militants and supporters remained that of a generally organized and activist middle class, mostly white, especially employees of the formal public sector. They declare that, despite the former president’s capacity for political articulation and rhetorical eloquence, Lulism as a psychological and sociological phenomenon is considerably weaker than the PT and quite independent of it. They see Singer’s argument as ambiguous in that it does not examine the ideological basis of Lulism, which combines a progressive view of state interventionism with conservative attitudes toward social interests. Moreover, they argue, notions such as realignment are problematic in this context because they fail to take into account the executive’s vote on legislative choices. Thus, the debate should include broader considerations of parliamentary results and party identification profiles beyond the presidential election.

Through surveys and experimental techniques, Samuels and Zucco show that the PT’s predilection for Dilma Rousseff’s election and reelection survived even troubled social episodes such as the street protests of June 2013.⁷ Identification with the party derived mainly, on the one hand, from its associative roots in civil society and, on the other hand, from the economic vote of government approval characteristic of multiparty presidential systems anywhere in the world. They conclude that partisanship remained very important in helping to solidify long-term support for the PT. They attribute the success of the PT in building its identity to organizational efforts to put down roots in civil society with a more participatory approach to politics. Lulism, expressed in support of the former president and his style of government, will eventually be moderated, but PT partisanship will survive, drawing its strength from elements beyond Lula. Over time, they suggest, the legacies of Lula and the PT will converge in the ideals of greater democracy and socioeconomic equality.

What our working hypothesis allows us to consider regarding this debate is that programmatic motivation has never been absent from economic policy decisions. In other words, while developmentalism has always served as a horizon for actions of the PT governments, it does not seem to us credible that the actions of Lula and later Dilma were detached from an attempt to create a political identity different from the one that marked liberal center parties in Brazil such as the Brazilian Democratic Movement Party and the Brazilian Social Democratic Party. Thus Lulism, although very important for understanding the outcomes of particular elections, becomes less important than the attempt to construct PT’s own brand in the context of macroeconomic choices.

The Heyday and Crisis of the PT Developmentalist Project

Financialization, Rentierism, and the Limits of the Policy Space

An important impact of contemporary capitalism on the degree of freedom for developmentalist and interventionist policies in developing countries has been the financialization of the economy—a worldwide trend in countries with some sophistication in their respective financial systems. As measured by various indicators, Brazil is widely regarded as a highly financialized economy but one with particular characteristics (Bruno et al., 2011). “Financialization” is understood as “the increasing role of financial motives, financial markets, financial actors and financial institutions in the operation of domestic and international economies” (Epstein, 2005: 3) or “a pattern of accumulation in which profit-making occurs increasingly through financial channels rather than through trade and commodity production” (Krippner, 2005: 174). Finance-led capitalism has spread around the world because of neoliberalism, promoting policies such as financial liberalization, labor market flexibility, and the draining of the social and developmental state.

An important issue for emerging economies is the asymmetric international financial integration with regard to the dynamics and magnitude of capital flows directed to peripheral countries experiencing globalization. These flows depend mainly on exogenous factors, making these countries permanently vulnerable to their reversal either because of changes in monetary conditions at the center or because of the increased preference for liquidity of global investors (Paula, Fritz, and Prates, 2017). International studies show that financialization significantly reduces the autonomy of national states in formulating domestic economic policies independent of international or long-term conditions, consistent with the conditions of production and interests of domestic nonfinancial sectors (Becker et al., 2010).

Bresser-Pereira (2018: 27) argues that since the late 1980s we have seen the emergence of a financial-rentierist capitalism “in which capitalists are predominantly rentiers, while highly placed technobureaucrats are the top executives of companies or financiers.” Rentiers, most of them inheritors, replace entrepreneurs in the ownership of big firms and leave the management of companies to financiers. Financialization in Brazil since 1994 has been brought about by the replacement of a monetary regime characterized by inflationary gains to one based on gains from interest. This process was spurred by the increasing liberalization of capital accounts, given the speculative nature of capital flows from both residents and nonresidents. Under the new regime the government has sought to reconcile the interests of rentierist-based accumulation with redistributive social policies favoring segments whose income derives from interest and other financial gains. Thus financialization has been stimulated by two interrelated factors—high actual interest rates and an “overnight” circuit in the Brazilian economy inherited from the period of high inflation but maintained in the post-real period—into which high-yield investments of economic agents are channeled⁸ (Paula and Bruno, 2017). Some of the detrimental effects of this process in Brazil are worth mentioning. On the one hand, financialization converts into a paroxysm the capitalists’ (including industrial entrepreneurs-turned-rentiers’) preference for liquidity, reducing gross fixed capital formation in favor of short-term financial investments that compete with investments in capital assets by increasing the liquidity premium (Bruno et al., 2011). On the other hand, it has had clear concentrating effects on high-income segments, considering that financial income, dividend income, and inheritance and gifts represent almost 40 percent of the country’s income (Menezes Filho, 2017).

Internal and External Economic Constraints

The Brazilian economy is highly integrated into the international system in terms of capital flows and partially dependent on commodity exports. Although the significance of commodity exports is less pronounced than in other Latin American economies, their direct (the income effect) and indirect (through financial conditions, country risk, exchange rate, inflation, and monetary policy) impacts are considerable. In addition, the form of its international financial integration is crucial in terms of policy space for the implementation of developmentalist and interventionist policies. The Brazilian economy has undergone an intense liberalization of its capital account and balance of payments since the early 1990s, becoming quite integrated from this perspective (Paula, 2011). The Lula administration was complaisant regarding these flows, which increased considerably after 2004 and, with a brief gap at the height of the international financial crisis, returned strongly starting in 2009 in the absence of any effective measures to moderate them. The Dilma administration, in contrast, faced an avalanche of external resources resulting from the quantitative easing of the Federal Reserve and began regulating capital flows, even on the derivatives market, which contributed to the Central Bank’s devaluing of the exchange rate in August 2011. However, when the flow of capital began to reverse, it eliminated those controls.

In this context of strong international financial integration and dependence on commodities’ exports, Campello (2015) describes a “confidence game” in which emerging countries, especially in Latin America, vulnerable to global liquidity and price cycles, are required to adjust their domestic policies to changing external conditions, especially in downturn periods. In good times of abundant financial flows and high prices, governments have some leeway to implement an agenda of income distribution. However, during times of reversal in financial flows and/or commodity prices, governments, regardless of their political orientation, are exposed to pressures to reverse market conditions and curb resource leakage. Thus, the imposition of supply-side policies and the general convergence toward orthodoxy are based not on autonomous decisions but on exposure to global market volatility.

Lula’s government was pressured to adopt orthodox policies at the beginning of its first administration in 2003, and since international conditions (the commodities boom) were favorable it implemented a more interventionist and distributive agenda while maintaining an orthodox macroeconomic policy. However, in the first Dilma administration, given a less favorable international scenario with a sharp deterioration in the country’s terms of trade, policies moved away from more conventional measures, thus breaking away from the confidence game. In her second term Dilma tried rather desperately to regain market confidence through an orthodox adjustment, but it turned out to be counterproductive in a recession that was already under way and in a quite adverse international and domestic context.

Undoubtedly, the Lula administration was favored by the international economic context, but discretionary domestic policies played an important role in increasing the autonomy of its economic policy, in particular the combination of a policy of accumulating international reserves starting in 2005 (from US$28 billion in 2004 to US$352 billion in 2011) and the reduction of the external public debt (from US$136 billion in 2003 to US$86 billion in 2006), which made the Brazilian government a creditor in dollars starting in 2007. In contrast, domestic policy during the Dilma administration was faltering. Initially orthodox, it later combined orthodoxy with heterodox policies and finally returned to strongly contractionary policies in 2015. The countercyclical policies adopted by the first Dilma administration were poorly conceived on the fiscal issue—with ambitious targets being maintained through nonrecurring budget operations and exemption-oriented fiscal expansionism and excessive intervention in regulated markets (energy and oil), among other things.⁹

Although it is common to attribute the deceleration of the Brazilian economy starting in 2011 and the recession from mid-2014 on to policy errors and particularly to the new macroeconomic matrix, Borges (2017) has asked whether the crisis was a consequence of bad luck or bad policy. He concludes that, despite mistakes made (especially in the fiscal area), exogenous factors played an important role in slowing growth. Using varied methodologies, he estimates that between 40 and 60 percent of the 2012–2016 GDP growth slowdown was brought about by exogenous factors (both international and domestic). He highlights four factors: (1) the water crisis between 2013 and 2015, (2) the impact of Operation Car Wash on the economy, particularly the civil construction and oil and gas sectors, (3) a sharp drop in commodity prices, and (4) the virtual insolvency of Petrobras in 2015–2016. Usually attributed to corruption, accumulated losses between 2011 and 2014 with the sale of gasoline and diesel in the domestic market at prices below those paid on imports, and the ambitious program of investments implemented after the discovery of the Pre-Salt oil layer in 2007, the Petrobras situation was, according to Borges, mainly the result of “the strong and highly unexpected collapse of international oil prices starting at the end of 2014.” The second Dilma government faced a “perfect storm,” caused by a set of exogenous shocks. While her management was marked in part by bad policy, there was also a lot of bad luck.

From Societal Corporatism to the Crisis: Conspiracy and the Return to the Neoliberal Route

The 2002 elections significantly changed the correlation of political forces between representatives of capital—particularly financial capital, until then hegemonic—and labor, with the rise of previously marginalized sectors (Boito Jr. and Berringer, 2013). The difficulties of the Washington Consensus–inspired policies in leveraging industrial growth favored a partial convergence between some domestic businesses and wage earners (Ianoni, 2018). This “productivist” coalition was responsible for the gradual flexibilization of the macroeconomic tripod via the mediation platform created by segments of Brazil’s political, business, and civil society actors (Bresser-Pereira, 2014; Ianoni, 2018). The result of easing measures was a mix of policies now including actions focused on the productive sector and the fight against poverty.

From a path-dependence perspective, the Lula administration was marked by both advances and concessions reflecting the difficulties of maintaining a capital-labor consensus in the framework of financial globalization and the power of the financial-rentier sectors. Still, growth was the guarantor of the support for the PT from the “center,” an eclectic coalition enlisting new actors and reflecting the influence of the financial sector, productive-sector entrepreneurs, and workers on government policy making (Ianoni, 2018). Other significant changes were the creation of broad forums for civil society sectors to participate in public policy formulation, the assignment of key positions in the executive branch and in large state-owned companies to leaders drawn from the union movement or heterodox economic thinkers, and, finally, greater participation of labor representatives in the dynamics of development through pension funds (Boito Jr. and Berringer, 2013; Boschi, 2014).

In the first Dilma administration, Lula’s political consensus was maintained, and, from 2011 on, there was an inflection with the new macroeconomic matrix (Carvalho, 2018; Ianoni, 2018).¹⁰ The new matrix proved controversial and sparked debate about future tensions that would lead to the collapse of the government’s supporting coalition. By bringing about changes in interest and exchange-rate policies, with a drastic reduction in the SELIC rate and bank spreads, Dilma eventually strained the relationship with representatives of this financial-rentier segment (Singer, 2018).

In addition to the strife with the Brazilian Federation of Banks, in June 2013 widespread protests erupted in the country’s capitals that produced a fundamental cleavage in the Dilma government (Singer, 2018: 99). The political climate of the country was rattled by the episode. The drop in the president’s popularity put the federal government in a defensive stance, with setbacks such as the restoration of higher interest rates by the Central Bank in mid-2013 and “the beginning of the mobilization of the middle class, which would eventually play a decisive role in the fall of Dilma” (Singer, 2018: 103). This new window of opportunity caused many business representatives and actors to detach themselves definitively from the governing coalition and protest against the prevailing economic policy (Singer, 2018). Still, Dilma won the 2014 elections, albeit by a narrow margin (51.8 percent against 48.2 percent for Aécio Neves). The opposition began to enjoy a more favorable margin of support and number of representatives, displaying and intensifying the same regional and income polarization hinted at in the 2006 and 2010 elections. Partisan fractionalization in Congress increased (with a considerable increase in small parties and a reduction of the deputies of the PT), bringing about a correlation of forces that was more complex and heterogeneous and showing the future difficulties regarding the conciliation of the legislative and executive branches (Miguel and Assis, 2016; Santos and Canello, 2015).

Reelected after harsh campaigning with strong criticism of Neves and his orthodox-liberal agenda, Dilma then announced a financial-sector representative (Joaquim Levy) as minister of economics and adopted the defeated program. This change led to the accusation of “electoral swindle,” with mass protests intensifying in the streets concomitant with the deepening recession and dissatisfaction with the president, whose rejection rate jumped from 30 percent to 70 percent from September 2014 to September 2015 (Carvalho, 2018; Datafolha, 2018). The erosion of middle-class support for the PT was marked, as was middle-class backing of the protests together with right-wing political actors, movements, and business entities (Singer, 2018). The liberal rhetoric critical of interventionism ascribed to the PT—holding it responsible for the mistakes that had led the country to the crisis—gained momentum.

At the political level, the concatenation of a more conservative Congress and a new president of the Chamber of Deputies (Deputy Eduardo Cunha) colliding directly with the executive branch hindered any consensus on a common agenda between the two powers. It is in this context that Vice President Michel Temer gained strength. Given the president’s lack of popularity and communication problems, he became a relevant political coordinator, mediating with the Congress and specifically with his party Brazilian Democratic Movement—increasingly detached from Dilma—for the approval of liberalizing and austerity measures.¹¹

Operation Car Wash was closely intertwined with all this turbulence in the country, and the investigation of it, which began in 2014, had a great impact on both the political and the economic crisis and in particular on two powerful and interconnected national production chains: oil and gas, and construction (Belluzzo, 2018; Campos, 2019). Officially launched with the arrests of the Petrobras director Paulo Roberto Costa and the black-market banker Alberto Youssef in April 2014, it quickly resulted in an investigation of several Petrobras contracts with other companies and the blocking of hundreds of accounts totaling more than US$400 million. The investigation culminated in October 2014 with the leaking of Youssef’s unveiling of the scheme to the Federal Prosecutor’s Office and the Doomsday Operation (within Car Wash) that led to the arrest of top executives of major construction firms.

It is no easy task to estimate the aggregate impact of Operation Car Wash on the economy. Consulting companies such as GO Associados e Tendências, for example, attribute around a 2 to 2.5 percent contribution, respectively, to the 2015 and 2016 GDP contractions to impacts on the metalworking, naval, construction, and heavy engineering sectors, whose losses may amount to R$142 billion (BBC, 2015; Valor Econômico, 2016). The main effects of the crisis were concentrated in the construction industry, which was paralyzed by the sharp downturn of state investment. Between 2014 and 2017 it registered a negative balance between hiring and dismissals of 991,734 formal vacancies (predominantly in the Southeast of Brazil). Between 2014 and 2016, this balance was 1,115,223 out of 5,110,284 (or 21.8 percent) of the total job loss of the period, and from the second quarter of 2014 until the last quarter of 2018 there was a strong downturn.¹²

The dismantling and decapitalization of major construction firms are also noticeable. Data collected by the magazine O Empreiteiro made available by the Brazilian Chamber of the Construction Industry (CBIC, 2019) show that between 2015 and 2016, for example, the gross revenues of Queiroz Galvão, Andrade Gutierrez, and Camargo Corrêa dropped by 37, 31 and 39 percent, respectively. Odebrecht, the largest national construction company, had, in 2014, gross revenues of R$107 billion, with 168,000 employees and operations in 27 countries. By 2017—less than four years after the outbreak of the scandal and after its president had been arrested—they were R$82 billion, with 58,000 employees and activities in only 14 countries. Other industry giants—Queiroz Galvão, OAS, Andrade Gutierrez, and Camargo Corrêa—also saw their consolidated financial assets melt from around R$ 25.77 billion in 2014 to approximately R$8.041 billion in 2017, a loss of 68.6 percent. Many firms, forced to carry out divestment plans to fit the new scenario of fewer projects and comply with heavy leniency agreements, also sold many assets to foreign groups. Odebrecht initiated the sale of its subsidiary Braskem—until then the largest petrochemical company in Latin America, producer of biopolymers with significant participation from Petrobras—to the Dutch group LyondellBasell; Andrade Gutierrez sold its control over OI to Dutch and Portuguese shareholders, and Camargo Corrêa sold CPFL to State Grid, a Chinese company (Campos, 2019).

With regard to the oil sector, the scandal involving the cartel made up of state and other companies occurred in the context of a sharp drop in the price of the commodity, costing Petrobras net losses of R$26.6 billion in the last quarter of 2014 and R$36.9 billion in the last quarter of 2015 (Petrobras, 2018a). The crisis required the company to reduce its investment volume from approximately US$48.8 billion in 2013 to US$15.1 billion in 2017, a decrease of almost 70 percent (Petrobras, 2018b; Petrobras, 2018c). Its investments fell from 1.97 percent of GDP in 2013 to 0.73 percent of GDP in 2017 and from 9.44 percent of total investment to 4.69 percent in the same period, and its share of public investments declined from 49.3 percent in 2013 to 36.5 percent in 2017. This sharp downturn affected staff and numerous projects with other firms, contributing to a reduction in the number of formal employees in the Petrobras system from 86,108 to 68,829 between 2013 and 2016 and from 360,180 to 117,555 among outsourced workers for the same period. Within four years the company’s direct production chain lost nearly 260,000 formal and informal jobs (DIEESE, 2017). The crisis due to the Petrobras scandal, coupled with the Temer government’s new liberalizing program, led to a radical policy reversal for the oil sector and the massive sale of refineries and state-owned assets. Petrobras transferred 90 percent of its assets related to a Southeast pipeline network, the Nova Transportadora Sudeste, to the Canadian group Brookfield and the gas pipeline and transport network in the North and Northeast to the French group ENGIE (Carta Capital, 2019; Valor Econômico, 2017).

In summary, the oil and gas segment was the spearhead of the dismantling of Brazil’s engineering and infrastructure, accentuating a trend toward denationalization of the country’s productive activities ongoing since the Real Plan, with serious impacts on employment and income.¹³ The disruption of these two sectors—civil construction and oil and gas—greatly contributed, on one hand, to the deepening of the economic crisis from 2015 on and, on the other hand, to the disruption of some of the few sectors that had substantial domestic capital and were internationally competitive.

Conclusions

In this article we have highlighted the existence of a political-economic project of the PT that we characterize as rather close to social developmentalism. This project was different from those of previous governments and their liberal centrist tendencies. In addition, we have described how the financialization of the economy in Brazil has, at various times, imposed policies in favor of the interests of financiers/rentiers, dramatically limiting the state’s scope for adopting measures to stimulate growth. Finally, we have discussed the internal and external constraints on deepening the developmentalist project in Brazil: financially integrated and commodity-dependent emerging countries are subject to a confidence game in which leftist governments with distributive aims are exposed to the ups and downs of global cycles of liquidity and commodity markets and therefore required to adjust their domestic policies to these global conditions.

The first Lula administration faced the problem of a crisis of confidence and adopted orthodox policies but was able, with the improvement of international conditions, to launch policies that were more interventionist and distributive. Dilma Rousseff, facing a downright unfavorable international context, explicitly broke with the confidence game by applying the policies of the new macroeconomic matrix. In the transition to her second term, she radically reversed the policy orientation, moving toward strong fiscal adjustment and monetary orthodoxy, and this eventually undermined her few sources of political support. The political conspiracy that produced her impeachment ended up occurring with surprising ease, fueled by the effects of Operation Car Wash and the erratic trajectory of economic policies once the government became aware of continuing difficulties regarding the new macroeconomic matrix. The economic crisis from the second half of 2014 on undoubtedly contributed to the political crisis, which in turn made infeasible any attempt to implement policies in order to reverse the situation. Dilma’s impeachment finally interrupted the PT’s developmentalist project, with its mistakes and successes, allowing the emergence of new political actors.

Footnotes

Notes

Luiz Fernando de Paula is a professor in the Institute of Economics of the Federal University of Rio de Janeiro and a volunteer professor in the Institute of Social and Political Studies of the State University of Rio de Janeiro (IESP-UERJ). Fabiano Santos is also a professor in the latter institute, and both he and Paula are researchers in Brazil’s National Council for Scientific and Technological Development (CNPq) and State of Rio de Janeiro Research Foundation (FAPERJ). Rafael Moura is a Ph.D. student in the Post-Graduate Program in Political Science of IESP and also a Class 10 Fellow of the FAPERJ.

References

BBC (British Broadcasting Corporation) 2015 “Escândalo da Petrobras ‘engoliu 2,5% da economia em 2015.’” https://www.bbc.com/portuguese/noticias/2015/12/151201_lavajato_ru (accessed December 2, 2015).

Becker

Joachim

Jäger

Johannes

Leubolt

Bernhard

Wissenbacher

Rudy

2010 “Peripheral financialization and vulnerability to crisis: a regulationist perspective.” Competition and Change 14 (3–4): 225–247.

Belluzzo

Luiz Gonzaga

2018 “As consequências econômicas da Lava Jato,” pp. 21–35 in Kerche

Fábio

Feres Júnior

João

(eds.), Operação Lava Jato e a democracia brasileira. São Paulo: Contracorrente.

Bielschowsky

Ricardo

1988 Pensamento econômico brasileiro: O ciclo ideológico do desenvolvimentismo. Rio de Janeiro: Contraponto.

Bielschowsky

Ricardo

2012 “Estratégia de desenvolvimento e as três frentes de expansão no Brasil.” Economia e Sociedade 21: 729–748.

Boito Júnior

Armando

Berringer

Tatiana

2013 “Brasil: classes sociais, neodesenvolvimentismo e política externa nos governos Lula e Dilma.” Revista de Sociologia e Política 21 (47): 31–38.

Borges

Bráulio

2017 “Impacto dos erros (reais) da Nova Matriz tem sido muito exagerado.” Blog do Ibre, September 8. https://blogdoibre.fgv.br/posts/impacto-dos-erros-reais-da-nova-matriz-tem-sido-muito-exagerado (accessed June 22, 2019).

Boschi

Renato Raul

2014 “Politics and trajectory in Brazilian capitalist development,” pp. 123–143 in Becker

Uwe

(ed.), The BRICS and Emerging Economies in Comparative Perspective: Political Economy, Liberalisation and Institutional Change. London: Routledge.

Bresser-Pereira

Luiz Carlos

2011 “From old to new developmentalism in Latin America,” pp. 108–130 in Ocampo

José

Ros

Jayme

(eds.), The Oxford Handbook of Latin American Economics. Oxford: Oxford University Press.

10.

Bresser-Pereira

Luiz Carlos

2014 A construção política do Brasil: Sociedade, economia e estado desde a Independência. São Paulo: Editora 34.

11.

Bresser-Pereira

Luiz Carlos

2018 “Capitalismo financeiro-rentista.” Estudos Avançados 32 (92): 17–28.

12.

Bresser-Pereira

Luiz Carlos

Oreiro

José Luis

Marconi

Nelson

2016 Macroeconomia desenvolvimentista. Rio de Janeiro: Elsevier.

13.

Bruno

Miguel

Diawara

Hawa

Araújo

Eliane

Reis

Anna Carolina

Rubens

Mário

2011 “Finance-led growth regime no Brasil: estatuto teórico, evidências empíricas e consequências macroeconômicas.” Revista de Economia Política 31: 730–750.

14.

CBIC (Câmara Brasileira da Indústria da Construção) 2019 “Banco de Dados.” http://www.cbicdados.com.br/menu/empresas-de-construcao/maiores-empresas-de-construcao (accessed September 21, 2019).

15.

Campello

Daniela

2015 The Politics of Market Discipline in Latin America: Globalization and Democracy. Cambridge: Cambridge University Press.

16.

Campos

Pedro Henrique

2019 “Os efeitos da crise econômica e da operação Lava Jato sobre a indústria da construção pesada no Brasil: falências, desnacionalização e desestruturação produtiva.” Mediações 24 (1): 127–153.

17.

Carneiro

Ricardo

2012 “Velhos e novos desenvolvimentismos.” Economia e Sociedade 21: 749–778.

18.

Carta Capital 2019 “Venda de refinarias aprofunda desmonte da Petrobras.” June 29. https://www.cartacapital.com.br/economia/venda-de-refinarias-aprofunda-o-desmonte-da-petrobras/amp/?__twitter_impression=true&fbclid=IwAR3F8x5hMzubSwuzkQw1G62MudJYUEjW8sj8pS_KXjQM0fbMgNWLqj178eg (accessed July 13, 2019).

19.

Carvalho

Laura

2018 Valsa brasileira: Do boom ao caos econômico. São Paulo: Editora Todavia.

20.

Datafolha 2018 “Avaliação Datafolha da Presidente Dilma.” https://www1.folha.uol.com.br/paywall/signup.shtml?https://www1.folha.uol.com.br/infograficos/2015/02/118652-avaliacao-datafolha-da-presidente-dilma.shtml (accessed December 10, 2018).

21.

DIEESE (Departamento Intersindical de Estatística e Estudos Socioeconômicos) 2017 “Conjuntura econômica, indicadores do setor de petróleo, da Petrobras e outros números de interesse para os trabalhadores.” https://sindipetro.org.br/wp-content/uploads/2017/09/DIEESE_Subs%C3%ADdios_SINDIPETRO_RJ.pdf (accessed June 19, 2019).

22.

Epstein

Gerald

(ed.). 2005 Financialisation and the World Economy. Cheltenham, UK: Edward Elgar.

23.

Ferrari-Filho

Fernando

Paula

Luiz Fernando de

2016 “Padrões de crescimento e desenvolvimentismo: uma análise keynesiano-estruturalista.” Nova Economia 26: 775–807.

24.

Fonseca

Pedro Cézar Dutra

2014 “Desenvolvimentismo: a construção do conceito,” pp. 29–78 in Calixtre

A. B.

Biancarelli

André Martins

Cintra

Marcos Antonio Macedo

(eds.), Presente e futuro do desenvolvimento brasileiro. Brasília: IPEA.

25.

Fritz

Bárbara

Paula

Luiz Fernando de

Prates

Daniela

2017 Developmentalism at the Periphery: Can Productive Change and Income Redistribution Be Compatible with Global Financial Asymmetries? desiguALdades Working Paper 101.

26.

Ianoni

Marcus

2018 Estado e coalizões no Brasil (2003-2016): Social-desenvolvimentismo e neoliberalismo. Rio de Janeiro: Contraponto.

27.

Krippner

Greta

2005 “The financialisation of the American economy.” Socio-Economic Review 3 (2): 173–208.

28.

Loureiro

Pedro

Saad-Filho

Alfredo

2019 “The limits of pragmatism: the rise and fall of the Brazilian Workers’ Party (2002–2016).” Latin American Perspectives 46 (1): 66–84.

29.

Mantega

Guido

2012 “O primeiro ano da nova matriz econômica.” Valor Econômico, December 19. http://www.valor.com.br/brasil/2945092/o-primeiro-ano-da-nova-matriz-economica (accessed June 19, 2019).

30.

Mello

Guilherme

Rossi

Pedro

2018 “Do industrialismo à austeridade: a política macro dos governos Dilma,” pp. 245–282 in Carneiro

Ricardo

Baltar

Paulo

Sarti

Fernando

(eds.), Para além da política econômica. São Paulo: Editora UNESP.

31.

Menezes Filho

Naercio

2017 “A desigualdade caiu no Brasil?” Valor Econômico, September 22. https://www.valor.com.br/opiniao/5129416/desigualdade-caiu-no-brasil (accessed June 19, 2019).

32.

Miguel

Luis Felipe

Assis

Pedro Paulo

2016 “Coligações eleitorais e fragmentação das bancadas parlamentares no Brasil: simulações a partir das eleições de 2014.” Revista de Sociologia e Política 24 (60): 29–46.

33.

Paula

Luiz Fernando de

2011 Financial Liberalization and Economic Performance: Brazil at the Crossroads. London: Routledge.

34.

Paula

Luiz Fernando

Bruno

Miguel

2017 “Financeirização, coalizão de interesses e taxa de juros no Brasil.” Revista Princípios 151: 66–71.

35.

Paula

Luiz Fernando

Fritz

Bárbara

Prates

Daniela

2017 “Keynes at the periphery: currency hierarchy and challenges for economic policy in emerging economies.” Journal of Post-Keynesian Economics 40 (2): 183–202.

36.

Paula

Luiz Fernando

Modenesi

André

Pires

Manoel

2015 “The tale of the contagion of two crises and policy responses in Brazil: a case of (Keynesian) policy coordination.” Journal of Post-Keynesian Economics 37: 408–435.

37.

Petrobras 2018a “Resultados consolidados do quarto trimestre de 2015.” https://www.investidorpetrobras.com.br/ptb/3198/RMF4T2015_R_Portugus.pdf (accessed October 19, 2018).

38.

Petrobras 2018b “Resultado do Quarto Trimestre de 2013.” https://www.investidorpetrobras.com.br/ptb/4603/RMFUSPort_2013_1.pdf (accessed October 19, 2018).

39.

Petrobras 2018c “Relatório ao Mercado Financeiro - Resultados Consolidados de 2017.” https://www.investidorpetrobras.com.br/ptb/1163/RMF4T17USPortugues.pdf (accessed October 19, 2018).

40.

Pinto

Eduardo Costa

Pinto

João Paulo Guedes

Saludjian

Alexis

Nogueira

Isabela

Balanco

Paulo

Schonerwald

Carlos

Baruco

Grasiela

2017 A guerra de todos contra todos: A crise brasileira. Texto para Discussão IE/UFRJ 6.

41.

Prates

Daniela

Fritz

Bárbara

Paula

Luiz Fernando de

2019 O desenvolvimentismo pode ser culpado pela crise? Uma classificação das políticas econômica e social dos governos do PT ao governo Temer . Texto para Discussão IE/UFRJ 9.

42.

Salto

Felipe

Ribeiro

Leonardo

2015 “Operações compromissadas, gosto de subdesenvolvimento.” Valor Econômico, June 8. https://www.valor.com.br/opiniao/4083296/operacoes-compromissadas-gosto-desubdesenvolvimento (accessed June 22, 2019).

43.

Samuels

David

Zucco

Cesar

2018 Partisans, Antipartisans, and Nonpartisans: Voting Behavior in Brazil. Berkeley and Los Angeles: University of California Press.

44.

Santos

Fabiano

Canello

Júlio

2015 “Brazilian Congress, 2014 elections and governability challenges.” Brazilian Political Science Review 9 (1): 115–134.

45.

Sarti

Fernando

Hiratuka

Célio

2018 “Desempenho recente da indústria brasileira no contexto de mudanças estruturais domésticas e globais,” pp. 127–170 in Carneiro

Ricardo

Baltar

Paulo

Sarti

Fernando

(eds.), Para além da política econômica. São Paulo: Editora UNESP.

46.

Singer

André

2009 Os sentidos do Lulismo: Reforma gradual e pacto conservador. São Paulo: Cia das Letras.

47.

Singer

André

2018 O Lulismo em crise: Um quebra-cabeça do período Dilma (2011–2016). São Paulo: Cia das Letras.

48.

Valor Econômico 2016 “‘O efeito da Lava-Jato no PIB se confirmou,’ diz Gesner Oliveira.” September 15. https://www.valor.com.br/valor-investe/casa-das-caldeiras/4672327/o-efeito-da-lava-jato-no-pib-se-confirmou-diz-gesner-olivei (accessed July 13, 2019).

49.

Valor Econômico 2017 “Estatal fecha venda de 90% da NTS para Brookfield.” April 5. https://www.valor.com.br/empresas/4927334/estatal-fecha-venda-de-90-da-nts-para-brookfield (accessed July 13, 2019).

50.

Valor Econômico 2019 “Construtoras encolhem 85% em 3 anos.” July 1. https://www.valor.com.br/empresas/6326143/construtoras-encolhem-85-em-3-anos (accessed July 13, 2019).