Chapter 2—Bolsa Familia: A Case Analysis of the Brazilian State’s Embedded Autonomy Under Lula da Silva (2024)

Brazil’s developmental state and its social policies have been the focus of much debate in the last few decades. The Latin American giant is one of the developing nations going through a momentous process of modernization and economic growth. In less than a decade, its GDP quadrupled, from US $.5 Trillion to US $2 Trillion (World Bank 2011, 606).¹ The country has become, together with Russia, India and China (BRIC group), one of the leading forces behind worldwide growth.² One of the main achievements of the Brazilian state has been the deployment of the largest welfare programs in the world (Hall 2006, 697), which is symbolized by the Bolsa Familia.

The purpose of this paper is to assess to what extent the state has been capable of pushing back the market to create a public space from which to enact such ambitious social policies. What are the political processes through which the state negotiated with the dominant classes in Brazil, and achieved acceptance of state intervention on matters of development strategy, taxation policies, social expenditures, etc.? How fragile is the accord, and what are its weaknesses? These questions shall bring into light theoretical debates about the nature and role of the state, international competition, and other highly contentious issues.

Through the examination of the Brazilian state, I hope to critically assess its accomplishments and limits. Ultimately, I hope to link this to the broader debate about the ability and capacity of capitalist developmental states to solve social issues arising out of the process of capital accumulation.

¹ “World Bank 2011” consists of an excel spreadsheet. Therefore, page numbers for this source refer to the row #.

² The four nations, which represent 40% of the world population, accounted for about half of global growth between 2000 and 2008. According to the International Monetary Fund, BRIC will account for 61% of global gross domestic product growth in 2014 (France-Presse 2010).

When the Partido dos Trabalhadores (PT)³ finally came to power in 2003, it faced a country that had been socially devastated through decades of economic development, modernization and rapid urbanization. From the industrialization efforts of Vargas, to the period of conservative modernization led by the repressive military dictatorship, Brazil developed at the cost of millions of dispossessed and disenfranchised people. Many of these people gave rise to earth shattering and heart wrenching favelas. Many others perished along their fight for survival. The contrasts between the haves and the have-nots created a country that was yearning for benign social and economic policies. This thirst for welfare was a palpable and tangible force in the political arena of the nation that was mostly channeled through massive social movements. The following historical recollection is a quick overview of the creation of this social condition, which became the inheritance of the PT once it came to power.

³ Launched in 1980, it is recognized as one of the largest and most important left-wing movements of Latin America. It governs at the federal level in a coalition government with several other parties since January 1, 2003. For more details, see Baiocchi (2003, p. 10).

Until the beginnings of the 20th century, Brazil had an export oriented economy based on primary goods and an oligarchic political structure. This configuration entered into a process of deep structural change around 1930. The main cause behind this shift was the collapse of the financial markets and the ensuing global depression. The Brazilian elite had hitherto depended upon the export of primary products, with coffee providing over 61 percent of Brazil’s export earnings during the period 1889-1933 (Peláez 1972, 64). However, this model ceased to be viable when global trade plummeted, dragging down with it the prices of raw materials and the influx of foreign capital into Brazil (Hilton 1975, 758). In dramatic fashion, the international price of coffee fell approximately 65 percent in the wake of the crisis (Peláez 1972, 33). The immediate response of the Brazilian government was to avoid the complete collapse of the coffee industry. Between 1931 and 1944, the Brazilian state purchased approximately 78 million sacks (equivalent to the worldwide consumption for three years) of excess coffee, and destroyed them, in an attempt to reduce the supply and maintain the prices (Fausto 1999, 200).

Nevertheless, these agrarian subsidies would prove to merely be one piece of the response to the crisis. The rise into power of Getúlio Vargas in October 1930 marked the beginning of a shift towards a broader plan of industrialization that transformed Brazil. Official documents from the time show that Vargas andkey members of his administration came to understand industrialization as paramount to national security. They argued that failure would condemn the Brazilian elite to play the role of a colony among the industrial powers (Vargas 1938; Hilton 1975, 758). The new strategy required an increased centralization of power into the hands of the State. Among other policies, Vargas dissolved the Congress and State Legislatures, replaced state governors with hand-picked Federal representatives, and monopolized military, monetary, and fiscal authority (Fausto 1999, 199). The crisis had created the perfect opportunity for this shift in the governing elite. Weakened in their economic power and dominance, the landed oligarchy lost the control over the state apparatus. The power to dictate Brazil’s policies and its future now resided with a new cadre composed of industrialists, politicians, and military leaders, among others (Fausto 1999, 196).

Despite the shift in power, patterns of land ownership did not change. The oligarchy still presided over the latifundios, and exercised absolute power over peasants and workers. Vargas only subordinated the agricultural sector to the state, gained control over the social surplus generated through export crops,⁴ and used it to subsidize the industrial sector. The establishment of autarquias, semi-autonomous administrative units used to supervise and direct the activities of selected areas of the economy, is a good illustration of how this was achieved. It allowed thecentral government to regulate, through production quotas and other measures, the supply of a commodity to meet the target market demand (Gordon-Ashworth 1980, 88-89).

As a result of Vargas’s policies, industry became the leading growth sector of Brazil’s economy for the first time in history (Baer and Villela 1973, 227). Between 1933 and 1939, the annual industrial growth rate was 11.2 percent (Hilton 1975, 757). Furthermore, by 1938 Brazil had nearly achieved self-sufficiency in light manufactures (import substitution). This is reflected in the fact that industrial raw materials, fuels, and machinery constituted over 90 percent of imports during 1935-1937 (Hilton 1975, 761). The import of these capital goods would also diminish rapidly, and by the late 1940’s Brazil would be producing well over half its equipment needs (Leff 1969, 474).

Moreover, structural changes in this period encompassed the use of labor and technology in agriculture, both of which contributed to the massive rural exodus and the formation of urban slums. Firstly, labor relations changed drastically. Landowners decided to bring the colono system⁵ to an end. They progressively rolled back usufruct rights, and opted for cash wage systems instead. This was partly because the colono system itself became increasingly uneconomic with the introduction of new varieties of coffee plants, and the realization that planting subsistence crops in between rows of coffee bushes hindered productivity (Martins 2002, 313). The mechanization of agriculture and the introduction of chemical inputs further complemented the switch towards wage salaries. The use of tractors, which grew from 1,706 units in 1920, to 156,529 by 1970 (Stitzlein 1974, 39), and herbicides, meant that agricultural production became increasingly fragmented between tasks that still required manual labor, and tasks that did not anymore (Martins 2002, 315). These changes turned agricultural labor into a seasonal, rather than year-round activity. The wage system, therefore, became the most efficient system for capital accumulation; it allowed landowners to use workers when needed, and dispose of them when not.

As part of the structural changes of the epoch, landowners systematically evicted peasants from the land. This catastrophic situation, in which millions of women and men lost their ability to subsist, had a deep impact on urbanization patterns and rural migration. A great number of the dispossessed flocked into Brazil’s major cities in hopes of finding jobs in the burgeoning industrial sector. Consequently, the percentage of urban population jumped from 15 percent in 1940 to more than 50 percent in 1970 (Wagner and Ward 1980, 249). Unfortunately, the numbers of jobs in the formal economy, partly due to the technical modernization of Brazilian industry, was limited in comparison with the influx of migrants. The great majority of the newly arrived was relegated to the informal economy,⁶ where working conditions were poor, jobs were insecure, and wages were low. For the most part, migrants were denied access to health, education, housing and jobs. They had no other choice but to congregate in the peripheries of the city and build shacks as a last recourse for survival (Pino 1996, 436). The growth of favelas⁷ in Brazil, with all of its dehumanizing implications,⁸ would only get worse with the decades. In Rio de Janeiro, the numbers of favelas would swell from 7 percent (160 thousand) of Rio’s total population in 1950, to 18 percent (over 1.1 million) in the year 2000 (Perlman 2006, 158). It is also important to point out that Vargas fully endorsed and facilitated the establishment of these settlements. The governmentpassively looked the other way as the landless built their shacks, and even actively extended railroad transportation to the poor areas (Pino 1996, 436).

Vargas’ administration marked a stark shift in the economic orientation of Brazil. More crucially, labor, whose scarcity was once a chronic problem for the landowners, became redundant. This changing condition for capital accumulation created massive levels of social exclusion. The ensuing social upheaval is the manifestation of one of the most fundamental contradictions in capitalism. The lack of resolution of this contradiction—i.e. increasing human misery and economic growth—would become the most central issue in Brazilian society thereafter. The cruel disposal of countless human lives, presided over by those who owned the means of production and by extension the state apparatus, would not go unchallenged. Dispossessed peasants organized themselves into Peasant Leagues (Ligas Camponˆesas). Together with the Maoist branch of the Communist Party of Brazil⁹ and the Catholic Church, they demanded a radical agrarian reform program (Juli˜ao 1972; Hewitt 1969; Stepan 1973; Welch 1995). As we shall see, the reactionary countermovement to the popular uprising was swift and brutal.

⁴ For a review of the role of agriculture in industrial development, see Johnston and Mellor (1961). For case studies of the contribution of agriculture to general economic growth, see Wilber (1969); Macrae (1971); Ohkawa and Rosovsky (1960).

⁵ The colonato system generally consisted on year-long contracts, and the colonos were required to perform three main kinds of agricultural tasks. The first was the cultivation of coffee plants, which involved the general task of taking care of the plants, including keeping the plantation weed-free. This entailed two or three weedings annually. The second task consisted of harvesting the coffee, which was usually done between May and July. The third task was to provide the landowner with several days of unpaid work per year. This included jobs such as clearing pasture, cleaning and maintaining paths and roads, fixing fences, and putting out fires (Martins 2002, 305; Brannstrom 2000, 330). In exchange for this work, colonos had three sources of income. They received a monthly wage according to thenumber of coffee trees they were responsible for maintaining. They were also paid a certain amount per sack of coffee harvested. Lastly, colonos had limited usufruct rights. This allowed them to plant subsistence crops - such as corn, beans and even rice - between the rows of coffee bushes, as well as in the small plot of land within the landowner’s estate where they lived (Brannstrom 2000, 330; Martins 2002, 306).

⁶ For a discussion of the role of informal sector economies in Latin America, see Soto (1989).

⁷ For a literature on Brazilian favelas, see Perlman (2006); Pino (1997); Pino (1996); Wagner and Ward (1980). For a debate about slums across historical and geographical expanses, see Davis (2007); “United Nations Human Settlements Programme” (2003); Clinard (1970).

⁸ “In 1970 only 36.9 percent of Brazil’s urban households had gas service, 42.8 percent had sewerage links and 53.2 percent had water. Malnutrition in urban areas is prevalent. A study of migrants living in Fortaleza in 1976 indicated that 68 percent of the urban migrants suffered from caloric deficiencies” (Wagner and Ward 1980, 256).

⁹ There were two branches of the Communist Party in Brazil at the time; one was Maoist, the other pro-Moscow. For the “Maoist Communist Party of Brazil (Partido Communista do Brasil, or PC do B) . . . the struggle in the countryside would be spearheaded by dispossessed peasants for land, and not by landless agricultural workers exploited through the wage relation. The pro-Moscow Brazilian Communist Party (Partido Communista Brasileiro, or PCB) took the opposite view, and argued that the struggle would be a peaceful one, involving the implementation of existing rural labour legislation recognizing the claims of both colonos and permanent labourers not as peasants with rights to land but rather as agricultural workers with rights to a decent wage, reduced working hours, and improved working conditions. In short, groups with a shared political outlook were fighting for different policy objectives (land reform; improvements in payand conditions) on the basis of socio-economic identities that were equally distinct (peasants; agricultural wage labourers)” (Martins 2002, 318).

The military coup, which took place in March 1964 with U.S. support,¹⁰ had two clear goals: To “. . . forestall the communist plan for seizing power . . . and to reestablish order so that legal reforms [could] be carried out” (Dines 1964, 144). Forestalling communists meant, in this context, the violent repression of those who fought for social inclusion through agrarian reform i.e. distribution of land and wealth. The military intelligence and the DOPS (the political police), carried out massive state terrorism schemes, such as “Operation Clean-Up”.¹¹ They targeted Catholic Church organizations,¹² political parties on the left,¹³ dissidents within the military, union officials, peasant militants, lawyers, human rights advocates, among others. Political persecution, assassinations, and torture were all part of a structural strategy that was especially strong in the Northeast of Brazil¹⁴ (Skidmore 1990, 23-27).

The second stated goal of reestablishing order to carry out legal reforms was, in effect, the continuation of the process of agricultural and industrial modernization, which started during Vargas’ first term. Agricultural policy continued to favor not land redistribution but increased production, most of which was destined for export in a time when 30 million Brazilians were going hungry (Skidmore 1990, 299). The military regime implemented what was effectively a Green Revolution. They created the conditions for the use of new inputs (from hybrid seeds and chemical fertilizers) and mechanized implements, all of which led to profound changes in agricultural production,and the creation of agro-industrial complexes. These policy goals were achieved through subsidies to big landowners, who received generous allocations of financial resources and other benefits (Delgado 1985; Müller 1985). For example, the National Monetary Council waived taxes on all key farm products and inputsfertilizer, tractors, processing equipment, etc. Subsidies also took the form of special interest rates for agricultural credit that had lowor even negativereal rates. The government also extended a program of guaranteed minimum prices for agricultural products, invested in agricultural research, and created upper-level governmental educational programs for the training of agricultural specialists (Skidmore 1990, 92; Sauer 2006, 2).

The Brazilian Green Revolution exacerbated the trend of rural exodus, social exclusion, income inequality, and land concentration. While in 1960 the rural population comprised 55 percent of the population, by the end of the military dictatorship in 1985 this number dropped to only 28.7 percent (World Bank 2011, 955). Simultaneously, the urban population skyrocketed. São Paulo, Brazil’s largest city, offers a good sample of the general situation. Within the same period, its population rose from 3.6 million to 9 million, favela dwellers grew from 30 thousand to 629 thousand (Lloyd-Sherlock 1997, 291), and real wages fell by two-thirds (Wolford 2004, 411). Furthermore, income inequality widened to extraordinary levels. From 1960 to 1990, the share of the total national income of the wealthiest five percent of Brazil’s population jumped from 27.7 percent to 35.8 percent (Skidmore 1999, 198). Last but not least, land concentration also increased, with ten percent of the landowners in control of 80 percent of the land (Wolford 2004, 411). Thisis why many historians and academics call this period either painful or conservative modernization. The military government, clearly enacting policies that favored the capitalist class, modernized the latifundios and the industrial sectors while reinforcing the socially pernicious effects of the model (Sauer 2006, 2; Silva 1982).

Within this social context, economic growth proceeded at a breakneck pace. Led by industrial growth,¹⁵ GDP increased at an annual rate of 11.2 percent between 1967 and 1973 (Fausto 1999, 291). The Brazilian “miracle,” however, came to a halt when the turmoil in the international market, which started with the oil price shock of 1973, unleashed the perfect economic storm on much of Latin America. To understand the debt crisis that subsequently hit Brazil, an overview of the weaknesses of its growth model is crucial.

Brazil’s economy was built around oil. This was the case for agriculture, which, thanks to the Green Revolution, became highly mechanized and dependent on oil-based inputs such as fertilizers and pesticides. But this oil dependency was true for industry as well. In the late 1950s, Brazil’s government decided to commit the expansion of Brazil’s transportation future to highways rather than railways, mainly due to the lower initial investment costs (Skidmore 1990, 178). Thereafter, the stateoil company, Petrobrás, neglected the exploration of national reserves, and focused on refining and distributing cheap imported oil from the world market. As a result of these policies, Brazil imported 80 percent of its oil consumption (Skidmore 1990, 179).

This structural weakness was exposed when the oil crisis of 1973 and 1979,¹⁶ coupled with a global recession, put Brazil’s growth strategy in jeopardy. The sudden increase in oil prices, coupled with the diminishing demand for Brazilian products, threw Brazil’s international balance of payments in disarray. While imports and exports were almost in balance in 1973, from 1974 onwards, imports followed the rising international prices for oil, oil-derived products, and costs of transportation (Skidmore 1990, 180). This trend continued for more than a decade, and Brazil would run an uninterrupted deficit in its trade balance from 1970 to 1982, which would amount to a total of US $34 billion (World Bank 2011, 535).

To make up this deficit, Brazil had three options: to reduce imports, to deplete foreign reserves, and to incur foreign debt. The first option would have slowed growth, so it was not a viable choice for the government. Insofar as Brazil did not have nearly enough reserves to balance its trade, its only remaining option was to incur foreign debt,¹⁷ which went from approximately US$5.7 billion in 1970 to US $103 Billion in 1985 (World Bank 2011, 103). Consequently, Brazil now not only had to cope with the deficit in the trade balance, but also with the servicing of interests on its enormous debt. In only three years, payments on interest had more than tripled, from US $2.7 billion in 1978 to US $9.2 billion in 1981 (Skidmore 1990, 231). Inevitably, this debt treadmill started falling apart. The first indication came when interest rates went from 8.7 percent in 1978 to 17 percent in 1981, and continued soaring in the following years (Skidmore 1990, 236). Then, due to Mexico’s default on its debt servicing in August 1982, commercial banks became suspicious of Brazil, whose debt had surpassed that of Mexico, and refused to lend new money (Skidmore 1990, 232). Trapped without any other possible recourse, and faced with the prospect of a total collapse, Brazil was forced to seek help from the IMF (International Monetary Fund) (Skidmore 1990, 274).

The savage recession of 1981-83, which had been Brazil’s worst since the Great Depression, was the immediate consequence of the debt crisis (Skidmore 1990, 236-254). As the military government started to lose its grip on the economic situation, its legitimacy began to suffer as well. Internal class warfare and strife, which had never quite subsided, intensified when clamors for democracy and agrarian reform resurfaced into the center stage of Brazilian politics. Eventually, different social movements and political parties exercised enough pressure to force a return to democratic rule in 1985.

¹⁰ The U.S. intervention in Brazil and Latin America in general has to be considered in the context of the Cold War, and the attempts to curtail Communist expansion around the globe. For detailed accounts of the U.S. role in facilitating the military coup in Brazil, see Parker (1979; Black (1977). For accounts of U.S. involvement with Operation Condor, a campaign of political repression in the Southern Cone of South America, see Menjívar and Rodríguez (2005); Dinges (2004); McSherry (1999); McSherry (2000); McSherry (2001); McSherry (2002); McSherry (2005)

¹¹ For an overview of the repression, see M. H. M. Alves (1988); Fiechter (1975, p. 44). For a complete inventory of denunciations of torture, see M. M. Alves (1966); M. M. Alves (1973)

¹² MEB (Movimento de Educafdo de Base, or Basic Education Movement), JUC (Juventude Universitdria Catholica, or Catholic University Youth), and others whose organizing or charitable activities aroused suspicion.

¹³ The Moscow-line Communist Party (Partido Comunista Brasileiro, or PCB), the Maoist Communist Party (Partido Comunista do Brasil, or PC do B) and the Trotskyists, such as the ORM-POLOP (Organizafdo Revoluciondria Marxista-Politico Operdria, or the Revolutionary Marxist Organization/Workers’ Politics)

¹⁴ This area was the strongest center for Communist Party activities, and harbored many leaders deemed dangerous, such as Pernambuco Governor Miguel Arraes, SUDENE Superintendent Celso Furtado, literacy specialist Paulo Freire, peasant league lawyer Francisco Juliao, and long-time Communist Party activist Gregorio Bezerra (Skidmore 1990, 24).

¹⁵ “The most dynamic industrial sector was motor vehicles, growing at an annual rate of 34.5 percent. Of that production, which reached an annual output in 1969 of 354,000 units, 67 percent were passenger cars, the rest trucks and buses. That ratio contrasted sharply with the 1957-69 period, when the passenger car share was only 49 percent. Production was tilting toward the least fuel-efficient form of transportation” (Skidmore 1990, 139).

¹⁶ For a history of OPEC and the oil crisis, see Yergin (1993); Sampson (1991); Raymond (1976); Terzian (1985).

¹⁷ Total external debt is debt owed to nonresidents repayable in foreign currency, goods, or services. Total external debt is the sum of public, publicly guaranteed, and private nonguaranteed long-term debt, use of IMF credit, and short-term debt. Short-term debt includes all debt having anoriginal maturity of one year or less and interest in arrears on long-term debt.

When Lula came to power, there were already social protection policies in place and multiple cash transfer programs. Three programs were inherited from Fernando Henrique Cardoso’s (1993- 2003) administration. First was Bolsa Escola, which provided R$15 to each child attending school, up to a maximum of R$45per family (three children). Second was Bolsa Alimentação, which provided between R$15 and R$45 per month to families at risk to assist them in covering their basic food needs. Lastly, Auxilio Gás provided a monthly stipend of R$7.50 to low-income families to assist them in purchasing cooking gas (Fenwick 2009, 113).

Table 2.1: Bolsa Familia: Beneficiaries, annual expenditure (in nominal reales) and average value per household (in reales of 2004), 2004-10

Source: Sánchez-Ancochea and Mattei 2011, 305

Lula enacted Federal Law 10.836 and officially combined these three programs (plus a fourth one started by Lula himself called Cartão Alimentação) under the Bolsa Familia on January 9th 2004. He also tremendously increased the reach and scope of the social assistance. From 6.5 million families in 2004, Bolsa Familia was subsidizing almost 13 million families by 2010.With an average family size of four, Bolsa Familia was affecting an estimated 52 million citizens, or an astounding 26.6% of the 196 million people in Brazil.¹⁸ Official annual expenditures on the program also increased from approximately US $3 billion in 2004 to US $7 billion in 2010. Accumulated total expenditures since Bolsa Familia’s inception are roughly US $34 billion (Fenwick 2009, 115; Sánchez-Ancochea and Mattei 2011, 305).

In terms of the concrete benefits that families receive, the rules have been revised multiple times. When Bolsa Familia began in 2004, there were two categories of benefits targeting households in poverty and extreme poverty. Households in extreme poverty, who qualified as those earning up to R$50 a month, received a monthly cash allowance of R$50 independently of household characteristics. They could also receive an additional R$15 per child below 15 years of age with a maximum of three children per family. Households in poverty, who qualified as those earning between R$51 and R$100, could only benefit from the conditional children-based allowance.

These benefits were revised in 2006, 2009 and in 2011. In Bolsa Familia’s latest iteration, the cut-off point to receive benefits is R$70 for extremely poor and R$140 for poor households. Furthermore, the fixed transfer for extremely poor families is R$70 while the cash transfer per child in school has increased to R$32. A third type of benefit was also introduced in 2009. Households with children between 15 and 17 were entitled to receive R$38 per child, with a maximum of two per family. This was implemented to encourage continuing secondary schooling. With these reforms, the total value of benefits transferred varies between R$32 (US $20) and R$242 (US $152) per family per month (Sánchez-Ancochea and Mattei 2011, 304).

In the face of such an enormous government program, the lingering question is how the PT managed to accomplish its goal. We will explore the process by which the state procured the financial resources and the political backing of international and national capital. We will also discuss the results of Bolsa Familia in terms of the level of efficiency with which the state managed the cash transfers, as well as the ultimate goal of poverty alleviation.

¹⁸ Geographically speaking, the Northeast region of Brazil receives the largest portion of the cash transfer (about 50 percent in 2005). Historically this has been where poverty rates are highest (Fenwick 2009, 123; Marques and Mendes 2007, 26).

The pivotal shift in the conceptualization and discourse of welfare in Brazil came with the return of democracy. The massive social movements, which had created enough pressure to delegitimize and ultimately dismantle the military government, played a crucial role in the institutionalization of welfare programs in the policy agendas of subsequent administrations. The National Constitution of 1988, which reflected a hotly contested process of establishing new ground rules for a young democracy, embraced social inclusion as a citizenship right (Title VIII, Articles 203-204). With this shift, the welfare of all Brazilians became apublic responsibility. Unrestricted from the domains of private philanthropy and the church, this public conception provided the legal and discursive framework for a universal expansion of benefits (Fenwick 2009, 110).

It is worth emphasizing the role of the underlying societal pressure vis-à-vis the state apparatus and the larger context of IMF imposed neoliberal structural adjustment programs. During the 1990s, while most of Latin America was going through a period of market reforms predicated on the non-intervention of the state, Brazil saw a series of contradictory state policies. On the one hand, the Brazilian state was dismantling protectionist policies, subsidies, and regulatory frameworks. In a dramatic turn after six decades of the developmental state, neoliberal reforms became the order of the day (Nylen 1993; Green 2000). Simultaneously, the state also established the National Secretary of Social Assistance (SNAS, Secretaria da Assistência Social) in 1995 and created Bolsa Escola in 1999, the first national noncontributory cash-transfer program in Brazil’s history (Fenwick 2009, 111). In other words, the demands from radical groups such as the MST—which politically advocated for the end of capitalism, and demanded land redistribution, played a key role in creating a political space for the state to legitimize the enactment of welfare policies in its tussles with domestic and international capital.

It is also interesting to note that these welfare policies and the general shift towards social inclusion in Brazil started decades before the PT came to power, and were always presided by an administration not known for its radical politics. While, for example, Cardoso never implemented a program as ambitious inscope as Bolsa Familia, he nevertheless took steps in that direction.¹⁹ Rather independently from the state’s political leanings, societal pressure was a key factor in shaping state policy regarding welfare.

The discourse of welfare as a citizenship right, however, could only go so far when confronted with the stony veins of domestic and international capital. When the PT set out to gain their support, it astutely framed Bolsa Familia under a purely neoclassical discourse. On an Inter-American Development Bank loan proposal for Bolsa Familia, it was stated that “cash transfers directed at poor families, conditional on school attendance, can play a critical role in poverty reduction, given . . . their contribution to the accumulation of human capital” (Inter-American Development Bank 2004a, 1). In a similarfashion, the World Bank stated that “the broad objectives of the Bolsa Familia Program are to reduce poverty and inequality and promote human capital investments among poor families through the provision of direct monetary transfers to poor families and incentives for investing in human capital [i.e. schooling for children]” (The World Bank 2004a, 4).

Beyond adjusting the discursive and theoretical framework, the PT also made sure the implementation of Bolsa Familia followed mechanisms deemed as efficient—i.e. market oriented. As a policy of targeted conditional cash transfers, Bolsa Familia was perceived as a social risk management program that took pains to weed out those who are not poor (as opposed to policies that reach all citizens, such as universal health systems, price controls on food and services, etc.). Furthermore, cash transfers signified an emphasis on individual choices in the market that favored privatization: Goods and services purchased with the money go to private businesses, constituting an indirect business subsidy (Ansell 2011, 24; Hall 2006, 695).

Theoretically speaking, the concept of conditional cash transfers is not dissimilar to a negative income tax, a policy supported by libertarians (in the American sense of the term) such as Milton Friedman (2002). In adopting a market oriented discourse and policy, the PT obtained all the political support it needed from powerful classes both within and without its borders. It avoided a conflict that could have potentially created capital flight, disinvestment issues and the consequent negative impact on the state’s fiscal revenues. Quite the contrary, the PT was generously rewarded in terms of resources to bankroll Bolsa Familia.

¹⁹ One could argue that Cardoso’s administration was forcefully induced to do so. On August 11, 1995, a conflict between police and land occupiers in the rural town of Corumbiara, Rondonia, left 13 dead, 11 of them occupiers. Then, on April 17, 1996, a larger massacre occurred in the desolate mining town of Eldorado do Carajás, in southeastern Para. Police opened fire on 1,200 MST protestors blocking a highway, leaving 19 dead. As with Corumbiara, there was evidence that some of the victims had been executed (Ondetti 2007, 13; Bergamo and Camarotti 1996). These popular uprisings and police massacres attracted international media attention, and forced Cardoso to respond immediately. First, he sought to decelerate the quickly escalating level of land occupations (Pereira 2003, 53) and issued a decree making occupied farms ineligible for expropriation for at least two years. As a result, by 2002 land occupations had declined to slightly above 1995 levels (Ondetti 2007, 14). Simultaneously, Cardoso jumpstarted a state led agrarian reform program. He quintupled the annual budget for INCRA (Institute for Colonization and Agrarian Reform) (Wolford 2005, 248), and settled 287,000 families on expropriated private land during his first term (Pereira 2003, 45-51).

The launching of Bolsa Familia in 2004 was received with widespread support from the World Bank and the Inter-American Development Bank. Their political endorsement, which signified the wider support of the business community, was materialized in two loans for a total of US $2.57 billion that covered approximately 25 percent of total costs of Bolsa Familia in 2004 (Hall 2006, 698). First, the World Bank provided a US $572 million loan in June 2004. The funds were not mainly intended to be used for the cash transfers. Rather, its goals were to support (a) the consolidation of conditional cash transfer programs and reductions in gaps and duplications in coverage; (b) a strengthening of the system for identifying the target population; (c) the development of a monitoring and evaluation system; and (d) a strengthening of the basic institutional functioning of the program (The World Bank 2004a, 4; The World Bank 2004b, 6). Second, in December of the same year, the Inter-American Development Bank approved a loan of US $1 billion for Bolsa Familia, with the promise of up to US $2 billion (Inter-American Development Bank 2004b; Inter-American Development Bank 2004a).

Despite these sizeable inflows of foreign exchange, the Brazilian state financed most of the US $34 billion that Bolsa Familia has cost to date. In the decade since Lula came to power, the Brazilian state was able to increase revenues to finance increased expenditures. This was possible not because taxation rates rose(total taxation as % of GDP rose only slightly), but because the total revenue increased accordingly with the four-fold increase in GDP from 2003 to 2010. As many economists argue, economic growth, high productivity and improved competitiveness in global markets were the backbone upon which Lula’s social policies were enacted. But whose taxes financed Bolsa Familia?

While it is hard to trace the exact sources of the state’s fiscal revenue that directly funded Bolsa Familia, an examination of Brazil’s taxation system gives us some insights. As Figure 1 shows, indirect general taxes on goods and services (at about 13 percent of GDP in 2010) comprises the main source of tax revenue. This category of taxes includes value added taxes, which is considered to be highly regressive. Value added taxes impact working class peoples the most because they consume most of their income while having less savings and investments.

Moreover, Figure 1 also shows that taxes collected on the working class and the taxes collected on the income, profits and capital gains of the richest segment of the population have not changed dramatically during Lula’s administration. From 2003 to 2010, taxes on income, profits and capital gains timidly increased from 6.35 to 6.89 percent of GDP, while general taxes on goods and services barely increased from 12.02 to 12.89 percent of GDP.

These insights suggest that the increased public expenditures on welfare programs were not financed by an increased taxation on the rich. Rather, I suggest we are in the presence of an intraclass transfer of resources, rather than an inter-class transfer. In other words, the taxes paid by the working class matched the social expenditures directed toward it, making Bolsa Familia aself-financed program. While the evidence at hand does not prove this in a conclusive manner, case studies about welfare programs in the United States, Australia, Canada, Germany, Sweden and the United Kingdom show that intra-class transfers are the norm (Shaikh 2003, 537).

Further evidence of the self-financing nature of Bolsa Familia can be found in the reforms to the public pension system under Lula. In 2003, the state negatively readjusted the formula for calculating pensions, effectively reducing the amount received by all workers in the formal economy. In 2007, Lula extended pension benefits to informal workers who make up more than half of Brazil’s workforce. Therefore,“. . . while many formal workers had their pensions reduced, 28 million informal laborers who previously had no social security will now be entitled to pensions, which will largely be subsidized by the contributions of their formally employed counterparts” (Ansell 2011, 26).

If Bolsa Familia is indeed a case of self-financed welfare program, then this would explain, to a large extend, the positive reaction of the landowning, industrial and financial elites of Brazil towards Lula’s social policies. Ultimately, they did not see Bolsa Familia as a redistribution of resources from the rich to the poor, and therefore had little incentive to oppose the measure.

Even though the PT secured the funding and had earned the support of domestic and international capital, success was far from secured. As we will explore, the task of coordinating a bureaucracy that can efficiently enlist eligible households, channel the resources to them without overtly large administrative costs, monitor results, compile statistics, etc., becomes a herculean challenge when confronted with partisan opposition and potential corruption at multiple levels of government, among other problems. I will argue here that there were two key factors that allowed the PT to navigate the politics of implementation in a successful way.

The first factor is the creation of the Ministry of Social Development and Fight against Hunger (MDS), a centralized coordinating state agency that brought the much needed streamlining and efficiency boost to the entire social assistance complex in Brazil. This agency was created in January 2004, and integrated the Ministry of Food Security and Fight against Hunger with the Ministry of Social Welfare (Hall 2006, 697). As previously noted, under MDS’s tutelage Bolsa Escola, Bolsa Alimentação, Auxilio Gás and Cartão Alimentação were combined under one single program: Bolsa Familia. This integration had the powerful impact of resolving intra-bureaucratic chaos and reducing administrative costs. Before the changes, families who were eligible to the four redistributive programs had to go through fourseparate bureaucratic processes in order to register and present themselves to four separate local offices. After MDS’s creation, each family had to complete only one application, show up to one local office, receive one single payment per month, and be registered in one federal registry.²⁰ As a result, the administrative cost to run Bolsa Familia is about 6 percent of the total value of the disbursed cash transfer, or an average of R$3.66 per household per month (Fenwick 2009, 116).

Even though MDS was the central coordinating agency, Bolsa Familia was actually a highly decentralized operation. This is a direct result of PT’s embeddedness and experience with grassroots social movements, worker unions, identity groups, etc. across the country during the years of the party.²¹ I argue that this embeddedness and the consequent decentralized governing style allowed the PT to avoid corruption and political resistance to the program based purely on party line divisions. In other words, PT’s uniquely horizontal state-society relation is the second reason for Bolsa Familia to have achieved its self-established goals.

Bolsa Familia is in essence a central-local collaboration, while local municipalities (which are constitutionally autonomous entities) work directly with MDS and the central government. Crucial tasks such as data collection, registration of potentialbeneficiaries and monitoring of adherence to conditionalities remain decentralized to municipal level (Hall 2006, 697). In channeling the funds to the municipalities directly, Lula was able to leapfrog powerful state governors and other state level power brokers in the legislative branch. This was important, as Lula would have had an uphill struggle to garner the support of opposition party leaders in powerful positions. In 2002, when Lula won the elections, PT’s representation at the state level was extremely low, and only 13 percent of the 27 states had a PT governor. In other words, the PT had to implement a nationwide policy in a politically hostile environment (Fenwick 2009, 122).

Working with municipalities, creating incentives for them to voluntarily join Bolsa Familia, and making sure funds were not being misused was not easy either. As a matter of fact, in 2004 only 7.9 percent of the 5564 municipalities had a PT governor. In a rather ironic situation, the ace of spades that allowed Lula to break the partisan line in the municipalities was none other than the budget cuts and fiscal decentralization process, both of which were part of Cardoso’s austerity measures.

On the one hand, the 1988 constitution turned municipalities into the federation’s primary social services providers. This mandate to offer social services, as per article 15 of the Organic Law of Social Assistance, also became legally enforceable post-2000 through fiscal regulation imposed by the federal government. On the other hand, municipalities were given little financial support to accomplish this legal mandate. In a rather paradoxical move, the fiscal Responsibility Law of 2000 turned a dire situation into a full-blown fiscal crisis from which municipalities could not escape (Fenwick 2009, 110-117). As a result, municipalities were to a large extent fiscally coerced into participating and supporting Bolsa Familia. The inflow of federal assistance for Bolsa Familia became a godsend for local governors seeking to balance their fiscal accounts while also complying with the legal requirement of offerings social services. By 2006, all municipalities in Brazil had voluntarily adhered to the program, and some of them derived as much as 40 percent of their overall budgets from Bolsa Familia (Hall 2006, 705; Fenwick 2009, 115).

Participation, however, did not guarantee transparency. To avoid municipal level corruption, Lula installed conditions for the disbursement of funds to municipalities. Under Brazil’s decentralization law, each municipality was required to set up a social council (Conselho de Controle Social) for this purpose, whose members are chosen by the mayor from public and civil society sectors (Hall 2006, 697). Furthermore, fraud hotlines were established. These measures allowed Bolsa Familia to provide cash transfers to the poor without subjecting them to the manipulation of local political patrons (Ansell 2011, 25; Sánchez-Ancochea and Mattei 2011, 312). In an equally important way, these measures promote a democratic system of checks and balances and foster the creation of public spaces for debate and for the practice of citizenship. Ultimately, the implementation of these policies was possible due to PT’s historical involvement with grassroots organizing, their deeply rooted commitment to democracy, and an emphasis on the autonomy of local authorities and social movements.

²⁰ The Cadastro Único is a census of the Brazilian poor based on a standard questionnaire elaborated by the National Secretary of Income and managed by the municipalities. It is reported that MDS was inputting up to one hundred thousand families a day into the database (Fenwick 2009, 116).

²¹ For an account of this process, see (Moreira Alves 1990).

In order to implement Bolsa Familia, the PT had to strike deals with multiple social actors, especially with those who hold economic and political power domestically and internationally. The party had to walk an extremely fine line between complying with specific group and class interests, and exercising its power to twist the arm of those who stood in its way - e.g. state governors. However, the PT’s willingness to negotiate with opposition conservative parties and business elites caused a tremendous backlash. If during its inception the PT could be characterized as being part and parcel of the broader social movements that swept through Brazil in the second half of the 20th century, its rise to power signified an abrupt break with these same grassroots political groups. Not only did the PT come under heavy criticism from its former allies, but also from an increasingly hostile intellectual left that saw in the PT the disappointing story of a hero fallen from grace. So the question remains: Did the PT betray the left?

The main indictment about PT’s betrayal is a comparison between pre-presidency promises and post-presidency action. On the side of promises, the left argued that Lula’s and the PT’s strengths laid in their focus on radical agrarian reform, suspension of payment on Brazil’s enormous foreign debt, and redistribution of income and wealth (Foster 2007, 16). To a large degree, many on the left expected the PT to completelybreak with international capital, dominate domestic elites in the agricultural, industrial and financial sectors, and enact revolutionary economic and social policies that would ultimately lead to a post-capitalist society. Unfortunately, when this did not come to pass, hope turned into a sour disillusionment. In March 2006, a group composed of the leading grassroots movements in Brazil²² delivered a document to the Food and Agricultural Organization of the United Nations denouncing measures implemented by the Lula government that impeded campesino agriculture and the larger fight against business interests.²³

Many academic scholars have also been highly critical of the PT. James Petras, a professor Emeritus in Sociology from Binghamton University, analyzed in an article on the Journal of Peasant Studies the phenomenon of the 'centre-left' regime that has emerged recently in Latin America. He concluded that the government of Lula in Brazil, when measured against a set of criteria designating espousal of “leftist politics,” fails the test (Petras 2006). He argued, for example, that the composition and agenda of the Social Economic Development Council, which was tasked with creating a social pact between labour, business and the government, revealed Lula’s pro-business, anti-working class bias.

Of the 82 members of the Council, 41 are businessmen and 13 are trade unionists, a better than three to one proportion favouring the bosses. The purpose is to discuss tax reform —reduce business taxes, in other words— and social security reform, decrease payments to workers, pensioners and other state beneficiaries. When Lula was confronted with the preponderance of the business elite among his inner circle, he roundly defended his proindustrial/agribusiness bias, embellishing his choices with an apolitical, meritocratic varnish and accusing his critics of nepotism. Lula conveniently forgets that his businessmen’s ’disinterested talent for thinking for the country’ has resulted in the greatest social inequalities in the world. Like his ’Third Way’ counterparts elsewhere in the world, Lula deliberately overlooks the class interests of the business elite precisely because they are his strategic allies in the pursuit of a neoliberal project (Petras and Veltmeyer 2003, 13).

John Bellamy Foster, a professor of sociology at the University of Oregon and editor of Monthly Review, echoed these criticisms. Foster argued the following:

In 2002 . . . the PT . . . indicated a greater willingness to accept the conditions imposed by neoliberalism, including full repayment of Brazil’s debt. Taking care of economic “fundamentals” was to be prioritized even at the expense of the PT’s broader social program . . . Lula’s first term consequently was characterized by . . . very stringent economic programs aimed at debt repayment and “fiscal responsibility.” This was coupled with a much less ambitious program than originally conceived on behalf of the poor . . . the PT has also promoted neoliberal structural reforms that directly undermine the overall position of workers. This has then constituted a kind of Latin American social-democratic “third way” strategy in which neoliberal ends are hegemonic (Foster 2007, 16).

While it is certainly accurate that the PT changed its policy agenda in route to winning elections in 2002, the argument that this constitutes a betrayal of the left is a simplistic—yet understandable—reaction that stems from an incomplete theory of the state. The belief that the state apparatus can unilaterally make economic and policy decisions; that its autonomy and agency, while certainly substantial, is such that if only the right people were in power anything could happen, is nothing more than a ludicrous assumption.

In his Embedded Autonomy: States and Industrial Transformation, Peter Evans argued that successful state policies require bureaucratic autonomy from the dominant societal forces, as well as embeddedness with those same forces. While the autonomy argument stresses the value of government agency, the embeddedness argument reflects the fact that government agencies cannot effectively involve themselves in policy matters without obtaining detailed information from nongovernmental actors and generating trust with those actors by more or less representing their interests (Evans 1995). What is lacking in this picture, however, is an analysis of limits and constraints. To what extent can governments assert their autonomy oversocietal constraints? Can the state lose its agency by being co-opted by special interests?

This theoretical void is filled by David Held. In recounting Napoleon Bonaparte’s France, he said:

The state is portrayed as an immense set of institutions, with the capacity to shape civil society and even to curtail the bourgeoisie’s capacity to control the state . . . There were ultimate constraints on the initiatives Bonaparte could take, however, without throwing society into a major crisis, as there are on any legislative or executive branch of the state. For the state in a capitalist society . . . cannot escape its dependence upon that society and, above all, upon those who own and control the productive process. Its dependence is revealed whenever the economy is beset by crises; for economic organizations of all kinds create the material resources on which the state apparatus survives. The state’s overall policies have to be compatible in the long run with the objectives of manufacturers and traders, otherwise civil society and the stability of the state itself are jeopardized (Held 1989, 35).

In other words, while class interests can definitely take over the state’s autonomous agency, this does not preclude the possibility that the state can retain a scaling degree of autonomy. However, the degree to which the state can be an autonomous decision maker has very concrete material limits. In the case of Brazil, the fact that Lula chose to pursue a conditional cashtransfer program, instead of the promised radical agrarian reform, was because of this exact limitation. In Brazil, agricultural products are used as inputs into other manufacturing and industrial sectors, such as the biofuel industry.²⁴ Agriculture also plays the crucial role of earning foreign exchange through exports. In the case of Brazil, 34 percent of total exports come from the agribusiness sector²⁵ (World Bank 2011, 1024), which amounts to approximately US $52 billion (World Bank 2011, 1031). Without this influx of foreign currency, the Brazilian state would not be able to import essential inputs to sustain industrial activity. As a matter of fact, 75 percent of Brazil’s imports are manufactures,²⁶ (World Bank 2011, 1001) for a total of approximately US $101 billion (World Bank 2011, 1004). Because agriculture is such an essential element within the economic model, it creates a partnership of mutual dependency between the state and corporate agribusinesses, whereas the survival of the former is a function of the profits and economicviability of the latter (Stedile 2007, 50). If Lula was to disrupt the agricultural elites, the state apparatus and society at large would have collapsed under the storm of a crisis.

Does this mutual dependency mean that the structural conditions nullify the state’s agency? I would argue that they only constrain it. Popular pressure is important too for legitimation purposes. While Lula would have never been able, under the specific historical and economic conditions, to enact the kind of agrarian reform that the left was clamoring for, the state had enough leverage to create the largest welfare program in the world (as discussed earlier, social movements were largely responsible for creating such a political space). A more conservative administration would have not implemented a welfare program as ambitious as Bolsa Familia, even if it had the capacity to do so. The fact that the PT was in power —and the kind of agency and political commitments it had—mattered a lot.

In short, contrary to what the left was arguing, there was never a betrayal from the PT, because the tool in question—i.e. state apparatus—never had the ability to carry out the demands for agrarian reform in the context of 21st century Brazil. The PT only created a misconception due to its own political inexperience and lack of theoretical insight. When it realized the mistake of the promises they made, it was too late to reverse the hopes and expectations of millions of people.

²² The Movimento dos Pequenos Agricultores (MPA), the Movimento dos Trabalhadores Rurais Sem Terra (MST), the Movimento dos Atingidos por Barragens (MAB), the Movimento das Mulheres Camponesas (MMC), the Comissão Pastoral da Terra (CPT), and the Associação Brasileira de Reforma Agrária (ABRA).

²³ Among other points, they argued: “. . . the government has wholeheartedly embraced neoliberal policies and supported international organizations such as the WTO and the World Bank. At the Montreal round of the WTO, the Brazilian government helped to block the initiative to make it mandatory worldwide for transgenic products to be labeled, thus defending the interests of multinational agribusiness companies. Specific policies biased toward the big farm sector include: the continuation of the tax-exempt status for supplies used for export-oriented agribusiness (a hidden subsidy to the foreign commodity trade); legalization, through a presidential decree, of the use and trade of transgenic soy; ignoring any and all environmental research and the actual infringement of law through the smuggling of banned cotton and corn transgenic seeds; ignoring campesino and environmentalists” demands in the drafting process of the biohazards law; lack of enforcement of the law ordering the food industry to carry warning labels on all products containing more than 1 percent of transgenics; continuation of financial support through public banks for large agribusiness concerns, for a total sum that went from 20 to 42 billion reals per year . . . ; granting of credits through a federal development bank, the Banco Nacional de Desenvolvimento Social (BNDES), for paper mills and eucalyptus foresting; and taking the initiative to pass a law opening national parks to logging interests. The government has also . . . failed to fulfill its promises to settle the landless families occupying large estates; implement an encompassing agrarian reform program . . . ; pass a law to expropriate estates that use slave labor; stop . . . the final conclusions that define land occupations as a major felony; push for judicial punishment of rural massacres such as those in Corumbiara (1995), Carajás (1996), and Felisburgo (2004); stop the rise in violence in rural areas; remove older laws and statutes that block agrarian reform; demarcate native land belonging to several ethnic groups, especially the Xavantes, Guaranis, and Pataxós; control the advance of cultivation of soy and cotton in the Amazon and bush areasa process which could have dire environmental consequences in the future; and create a wide network of cooperative agri-industries among campesinos (Stedile 2007, 54).

²⁴ Brazil is a biofuel industry leader. For favorable assessments, see World Bank (2010); Silveira et al. (2009). For a critical assessment of the social and environmental impacts of biofuel production in Brazil, see McMichael (2010); White and Dasgupta (2010).

²⁵ Agribusiness exports comprise the commodities in SITC (Standard International Trade Classification) sections 0 (food and live animals), 1 (beverages and tobacco), and 4 (animal and vegetable oils and fats) and SITC division 22 (oil seeds, oil nuts, and oil kernels). For a detailed structure and explanatory notes of SITC, see http://unstats.un.org/unsd/cr/registry/regcst.asp?Cl=28.

²⁶ Manufactures comprise the commodities in SITC sections 5 (chemicals), 6 (basic manufactures), 7 (machinery and transport equipment), and 8 (miscellaneous manufactured goods), excluding division 68 (nonferrous metals).

Now that we have explored the multiple negotiations surrounding the creation and implementation of Bolsa Familia, one task—maybe the most important one—remains: assessing the efficiency of the program. The matter seems rather trivial and uncomplicated when compared against technical measurements of poverty. Nevertheless, the full complexity of the issue comes to light when we address the question of whether this kind of program can solve poverty around the globe. In other words, whether Bolsa Familia is something other capitalist developmental states with high poverty rates should emulate. This exploration will lead us into different theoretical frameworks underlying the origins and causes of contemporary poverty.

Technically speaking, poverty in Brazil diminished in a dramatic fashion since the inception of Bolsa Familia. As Figure 2 shows, the percent of people living below US $2 a day dropped from 21.74% in 2003 to 9.87% in 2009, and those living below US $1 a day dropped from 10.45% to 3.8% (World Bank 2011, 813-814). It is estimated that approximately 35 percent of this drop was attributable to Bolsa Familia (Hall 2008, 811). Quite conclusively, Bolsa Familia brought a marked improvement in the material livelihoods of millions of Brazilians. This was ultimately reflected in the elections of 2006, when Lula was rewarded with the votes he needed to attain reelection. In fact, there is a general consensus in the literature that the success of Bolsa Familia played a pivotal role on this matter (Hall 2008, 812; Hunter and Power 2007). Studies have actually shown that in 2006, 22 out of 26 states displayed an inverse proportionalitybetween the votes obtained by Lula and the level of socioeconomic development at the municipal level. In other words, the poorer the municipality, the more votes Lula captured (Zucco 2008, 33).

The newly formed relationship between the poorest in Brazil and Lula prompted critics to argue that Bolsa Familia creates greater dependence of the poor on government hand-outs. While the political implications that Bolsa Familia had on Brazilian politics is palpably real, the argument that this creates a relationship of dependence that fosters laziness and idleness lacks empirical support. Contrary to these accusations, a study conducted by the International Poverty Centre concluded that Bolsa Familia was not detrimental to labour force participation among program beneficiaries (Soares, Ribas, and Osório 2007). Other studies have shown that 75 percent of Bolsa Familia recipients are actually employed, which represents an almost identical proportion to that of the wider population (Hall 2008, 815).

The fact that most beneficiaries use Bolsa Familia as a supplementary income to their jobs is a consequence of Brazil choosing to use the World Bank’s arbitrary benchmark to measure poverty.²⁷ Even with Bolsa Familia benefits, the income of recipient families remains well below the minimum wage (Marques and Mendes 2007, 27). While immediate starvation and death is staved off, Bolsa Familia does not provide much more. Thus, the argument that such a meager income creates idleness can only be the product of the imagination of subjects who are completely out of touch with the lives of the poor.

More astute critics of Bolsa Familia, on the other hand, point out the fact that Bolsa Familia, as a policy that intends to lift people out of poverty, is a failed enterprise because it does nothing to create employment opportunities (Hall 2008; Hall 2006). Furthermore, they argue that “Bolsa Familia cannot . . . deliver a sustained improvement in health and education outcomes and a reduction of inequality in the long run” (Sánchez-Ancochea and Mattei 2011, 313). Lastly, they point out that the expansion on Bolsa Familia has been accompanied by a drop of investment on other social services.

. . . expansion in noncontributory benefits in Brazil through . . . welfare programmes has been accompanied by a significant drop in longer-term social investment . . . between 2002 and 2004, federal spending on basic sanitation and housing fell in real terms by 46 per cent. Over the same period at state level, investment in education saw a reduction of 12 per cent, housing by 14 per cent and basic sanitation by 18 per cent. Very similar results were recorded for social investments by municipal governments in terms both of overall as well as per capita spending . . . [in conclusion] there has been no coordinated effort at federal or sub-national levels to provide essential public services indispensable for the reduction of inequalities in lifestyle and improvements inwelfare alongside improvements in individual and family income and consumption (Lavinas 2006, 7).

It is clear that these critics move beyond the narrow definition of poverty as a quantitative average that disregards quality of life, employment opportunities,²⁸ etc. If we accept these rather reasonable re-framings of what poverty is, then the accomplishments of Bolsa Familia are much less impressive in light of the massive resources and exceptional circumstances that made the program possible.

²⁷ As a reminder, Bolsa Familia’s cut off line was R$50 (Us$26.62) and R$100 (Us$53.23) in 2004, and R$70 (Us$37.26) and R$140 (Us$74.52) by 2011. If these monthly incomes are divided by 30, they reflect the two categories (Us$1.25 and Us$2 a day) the World Bank has to measure poverty almost exactly.

²⁸ UNDP’s Human Development Index is an example of this. The HDI indicates that Brazil advanced to 0.718 in 2011, compared to 0.600 in 1990. This reflects a life expectancy of 73.5 years; 7.2 years of schooling on average (for those 25 years old); 13.8 years of schooling expected for that younger; and annual per capita income of $10,162. However, the HDI does not indicate the quality of education, or whether the education is conducive to employment. It merely measures the years children and young adults remain in schools. Similarly, quality of life is measured in per capita income, and disregards important aspects such as access to water and sanitation, level of ecological and social vulnerability, etc. Most importantly, the measurements are taken as averages, blurring the huge issues of inequality

Despite Bolsa Familia’s accomplishments, it will never be enough by itself to solve poverty in Brazil. Surprisingly, this is a well-known fact that both Lula and the World Bank acknowledge. As the following quotes show, they both see Bolsa Familia as a temporary and complementary measure.

Higher quality and more equitable access to education (including secondary and tertiary education) are key to reducing poverty and inequality in the long run. However, the positive impacts of education reform take time to materialize. Reforms to social assistance programs and a more equitable pension system can achieve complementary results sooner and substantially reduce inequality and alleviate poverty (The World Bank 2004a, 1). Bolsa Famlia is not our salvation, merely an emergency measure . . . and the ideal is that in a few years’ time Bolsa Famlia will no longer be necessary (Hall 2006, 704).

The difference, however, is that the World Bank holds education as the ultimate long term eradicator of poverty due to its capacity to build up “human capital”.²⁹ The theoretical framework and the assumptions underlying this approach to poverty reduction remains, unfortunately, largely unchallenged and unexamined in the literature. The notion that education has the power to lift people out of poverty is an entrenched and mainstream narrative that is reproduced on academic and popular outlets alike. However, the entire notion of human capital rests on a flawed theory of marginal productivity that was challenged and disarmed by the Cambridge capital controversy in the 1960s.³⁰

The implications of the logical inconsistency in marginal productivity theory shattered the notion that value of the factors of production can be measured independently from income distribution—i.e. the illusion that we get what we deserve because income is somehow determined by what we contribute to the production process. The poor can get all the education in theworld, but their “human capital” will not help them get better wages.

A separate argument against the human capital approach would be the fact that the poor confront structural unemployment. College graduates without jobs will still remain poor and indebted, as many American citizens are finding out these days. To this criticism, neoclassical economists, such as those presiding over the World Bank, would reply that under perfectly competitive markets, there is full capacity utilization and therefore no unemployment (or a very low Natural Rate of Unemployment). To the extent that this ideal situation does not exist in any country in the world, the blame would be put on imperfectly competitive markets. However, as many economists in the classical vein have argued, full capacity utilization is not a feature of industrial production. In other words, unemployment is a structural feature endogenous to how capitalist firms operate (Moudud 2010).

The consequences of this theoretical argument are quite serious. It implies that poverty due to unemployment or underemployment will be a permanent feature of any capitalist economy. This insight, while surprising to some, is not new to those elites in power. The State, since the times of Vargas, has considered the exclusion of large segments of the population as inevitable and an acceptable price for growth and modernization. Cardoso, for example, openly admitted that a significant portion of the population was not part of the dynamic segment of the economy, and that he could not offer them a place in the new globalized order (Pereira 2003, 50).

²⁹ This quote reflects their human capital approach: “The objectives of the new Bolsa Familia Program itself include: (a) reducing poverty and inequality today, through the provision of direct monetary transfers to poor families; and (b) reducing poverty and inequality tomorrow, by providing incentives and conditions for investments in human capital on behalf of beneficiary families, and by linking beneficiary families to complementary services that could help them invest and grow out of poverty in the future” (The World Bank 2004b, 9).

³⁰ Marginal productivity theory stems from neoclassical economics, and is the formulation that determines income distribution- i.e. how much each factor of production is remunerated, wages for labour and interests/profits for capital. This marginal productivity (mp) is a function of the additional output over the additional unit of factor employed. In other words, mp is the economic value produced per each utilized input. If labour is the input, mp will be high if one hour of work can produce a lot of goods. If fewer goods are produced per hour of work, then mp will be low. The level of mp is then used to determine wages. In other words, if your work is very productive, then you will be paid a lot, and very little otherwise. It all depends on how much one’s labour can produce, that is to say, one’s human capital. The implication of this theory is that workers are paid what they contribute to the production process. If poverty exists, it is largely because people do not possess enough human capital to raise their mp/wages. Therefore, they need education to improve their contribution to the economy. To put it another way, the poor are responsible for being poor. Unfortunately, the Cambridge capital controversy brought this theory’s logical inconsistencies to light. To calculate the mp of capital, and settle the income distribution—i.e. profit/interest rates/wages—the formula needs to know the prices of heterogeneous capital goods such as machinery. However, the prices of machinery and other capital goods change according to the income distribution! The circular reasoning is impossible to overcome (Edwards 1985).

I believe there are two main lessons that can be drawn from the investigation of Bolsa Familia. The first one relates to the complexities of the relations and interlinkages between the state and diverse groups in society. Lula’s government had to adopt the human capital discourse to make Bolsa Familia a palatable project for domestic and international capital. Furthermore, the state had to procure the economic resources to fund the initiative without resorting to an economically and politically costly inter-class transfer of resources. This move, at the same time, depended largely on a booming economy that allowed for an explosion in fiscal revenue. The state’s agency, while conditioned and constrained by each of these factors, still played a crucial role in the creation of a program that offered invaluable aid to millions of people in poverty.

The second lesson refers to the inherent limits of state capacity in resolving the contradictions and pernicious consequences of the capitalist mode of production. The state may do many things, but it cannot challenge the underlying prerogatives of the business enterprise without undermining its own legitimacy. Unemployment and poverty are features of a system that is inherently unequal. While states like Brazil can certainly diminish the negative impact of these features on the population, it cannot eradicate the root of the problem. Capitalist states cannot be expected to hinder their own development. If society, whether due to environmental and/or social concerns, wishes to move beyond capitalism, it should not expect the change to come from the state apparatus.

[1] Alves, Marcio Moreira. 1966. Torturas e Torturados. Rio de Janeiro: [Idade Nova].

[2] Alves, Marcio Moreira. 1973. A Grain of Mustard Seed: The Awakening of the Brazilian Revolution. Doubleday Anchor Press.

[3] Alves, Maria Helena Moreira. 1988. State and Opposition in Military Brazil. Univ of Texas Pr.

[4] Ansell, Aaron. 2011. “‘Brazil’s Social Safety Net Under Lula.’” NACLA Report on the Americas 44 (2) (April): 23-26.

[5] Baer, Werner, and Annibal V. Villela. 1973. “Industrial Growth and Industrialization: Revisions in the Stages of Brazil’s Economic Development.” The Journal of Developing Areas 7 (2) (January 1): 217-234.

[6] Baiocchi, Gianpaolo. 2003. Radicals in Power: The Workers’ Party (PT) and Experiments in Urban Democracy in Brazil. Zed Books.

[7] Bergamo, Monica, and Gerson Camarotti. 1996. “’Brazil’s Landless Millions.’” World Press Review 43 (7) (July): 46.

[8] Black, Jan Knippers. 1977. United States Penetration of Brazil. Manchester University Press.

[9] Brannstrom, Christian. 2000. “Coffee Labor Regimes and Deforestation on a Brazilian Frontier, 1915-1965.” Economic Geography 76 (4) (October 1): 326-346. doi:10.2307/144390.

[10] Clinard, Marshall B. 1970. Slums and Community Development: Experiments in Self Help. The Free Press.

[11] Davis, Mike. 2007. Planet of Slums. Verso.

[12] Delgado, Guilherme Costa. 1985. Capital Financeiro E Agricultura No Brasil, 1965-1985. Coleção América Latina; Variation: Coleção América Latina (Icone Editora (São Paulo)). São Paulo, SP, Brasil: Icone Editora; Campinas, SP: Editora da UNICAMP.

[13] Dines, Alberto. 1964. Os Idos De Março e a Queda Em Abril. Rio de Janeiro: J. Alvaro.

[14] Dinges, John. 2004. The Condor Years: How Pinochet and His Allies Brought Terrorism to Three Continents. The New Press.

[15] Edwards, Chris. 1985. The Fragmented World: Competing Perspectives on Trade, Money, and Crisis. London; New York: Methuen.

[16] Evans, Peter B. 1995. Embedded Autonomy. Princeton University Press.

[17] Fausto, Boris. 1999. A Concise History of Brazil. Cambridge University Press.

[18] Fenwick, Tracy Beck. 2009. “Avoiding Governors: The Success of Bolsa Fam´ılia.” Latin American Research Review 44 (1): 102-131. doi:10.1353/lar.0.0073.

[19] Fiechter, Georges André. 1975. Brazil Since 1964: Modernisation Under a Military Régime: A Study of the Interactions of Politics and Economics in a Contemporary Military Régime. London: Macmillan.

[20] Foster, John Bellamy. 2007. “’No Radical Change in the Model’.” Monthly Review: An Independent Socialist Magazine 58 (9) (February): 15-16.

[21] France-Presse, Agence. 2010. “BRIC Nations to Drive Global Growth”, May 19. http://www.industryweek.com/ global-economy/bric-nations-drive-global-growth.

[22] Friedman, Milton. 2002. Capitalism and Freedom. 40th anniversary ed. Chicago: University of Chicago Press.

[23] Gordon-Ashworth, Fiona. 1980. “Agricultural Commodity Control Under Vargas in Brazil, 1930-1945.” Journal of Latin American Studies 12 (1) (May 1): 87-105.

[24] Green, James N. 2000. “Introduction.” Latin American Perspectives 27 (6) (November 1): 5 -8. doi: 10.1177/0094582X0002700601.

[25] Hall, Anthony. 2006. “From Fome Zero to Bolsa Família: Social Policies and Poverty Alleviation Under Lula.” Journal of Latin American Studies 38 (4) (November): 689-709.

[26] Hall, Anthony. 2008. “Brazil’s Bolsa Família: A Double-Edged Sword?” Development and Change 39 (5) (October 28): 799-822. doi:10.1111/j.1467-7660.2008.00506.x.

[27] Held, David. 1989. Political Theory and the Modern State. POLITY PRESS.

[28] Hewitt, Cynthia. 1969. “Brazil: The Peasant Movement of Pernambuco, 1961-1964.” In Latin American Peasant Movements, ed. Henry A Landsberger. First ed. Cornell University Press.

[29] Hilton, Stanley E. 1975. “Vargas and Brazilian Economic Development, 1930-1945: A Reappraisal of His Attitude Toward Industrialization and Planning.” The Journal of Economic History 35 (4) (December 1): 754-778.

[30] Hunter, Wendy, and Timothy J. Power. 2007. “Rewarding Lula: Executive Power, Social Policy, and the Brazilian Elections of 2006.” Latin American Politics & Society 49 (1): 1-30.

[31] Inter-American Development Bank. 2004a. Support for the Social Protection System: Loan Proposal. http://www.iadb.org/en/ projects/project,1303.html?id=BR-L1004.

[32] Inter-American Development Bank. 2004b. Project Concept Document. http://www.iadb.org/en/projects/ project,1303.html?id=BR-L1004.

[33] Johnston, Bruce F., and John W. Mellor. 1961. “The Role of Agriculture in Economic Development.” The American Economic Review 51 (4): 566-593.

[34] Juli˜ao, Francisco. 1972. Cambão - the Yoke: The Hidden Face of Brazil. Harmondsworth: Penguin.

[35] Kingstone, Peter, and Timothy J. Power, eds. 2008. Democratic Brazil Revisited. 1st ed. University of Pittsburgh Press.

[36] Lavinas, Lena. 2006. Transferˆencias De Renda: O “Quase Tudo” Do Sistema De Proteção Social Brasileiro. Anais do XXXIV Encontro Nacional de Economia [Proceedings of the 34th Brazilian Economics Meeting]. ANPEC - Associa¸c˜ao Nacional dos Centros de Pósgraduação em Economia [Brazilian Association of Graduate Programs in Economics]. http://ideas.repec.org/p/anp/en2006/157.html.

[37] Leff, Nathaniel H. 1969. “Long-Term Brazilian Economic Development.” The Journal of Economic History 29 (3): 473-493.

[38] Lloyd-Sherlock, Peter. 1997. “The Recent Appearance of Favelas in São Paulo City: An Old Problem in a New Setting.” Bulletin of Latin American Research 16 (3) (January 1): 289-305.

[39] Macrae, John. 1971. “The Relationship Between Agricultural and Industrial Growth, with Special Reference to the Development of the Punjab Economy from 1950 to 1965.” Journal of Development Studies 7 (July): 397-422. doi:10.1080/00220387108421378.

[40] Marques, Rosa Maria, and Áquilas Mendes. 2007. “Lula and Social Policy.” Monthly Review: An Independent Socialist Magazine 58 (9) (February): 22-31.

[41] Martins, Jose De Souza. 2002. “Representing the Peasantry? Struggles For/about Land in Brazil.” Journal of Peasant Studies 29 (April): 300-335. doi:10.1080/03066150412331311099.

[42] McMichael, Philip. 2010. “Agrofuels in the Food Regime.” Journal of Peasant Studies 37 (October): 609-629. doi:10.1080/03066150.2010.512450.

[43] McSherry, Patrice J. 1999. “Operation Condor: Clandestine Inter-American System.” Social Justice 26 (4) (December): 144-174.

[44] McSherry, Patrice J. 2000. “Breaking News: Operation Condor.” NACLA Report on the Americas 34 (1) (August): 1.

[45] McSherry, Patrice J. 2001. “Operation Condor: New Pieces of the Puzzle” 34 (6) (June): 1-2.

[46] McSherry, Patrice J. 2002. “Tracking the Origins of a State Terror Network: Operation Condor.” Latin American Perspectives 29 (1) (January 1): 38-60.

[47] McSherry, Patrice J. 2005. Predatory States: Operation Condor and Covert War in Latin America. Rowman & Littlefield Publishers.

[48] Menjívar, Cecilia, and Néstor Rodríguez. 2005. When States Kill: Latin America, the U.S., and Technologies of Terror. University of Texas Press.

[49] Moreira Alves, Maria Helena. 1990. “The Worker’s Party of Brazil: Building Struggle from the Grassroots.” In The Future of Socialism: Perspectives from the Left, ed. William K. Tabb. Monthly Review Press,U.S.

[50] Moudud, Jamee K. 2010. Strategic Competition, Dynamics, and the Role of the State: a New Perspective. Cheltenham, UK; Northampton, MA: Edward Elgar.

[51] Müller, Geraldo. 1985. A Dinâmica Da Agricultura Paulista. Série São Paulo 80;; V. 2; Variation: Série São Paulo 80;; V. 2. São Paulo: Fundação Sistema Estadual de Análise de Dados.

[52] Nylen, William R. 1993. “Selling Neoliberalism: Brazil’s Instituto Liberal.” Journal of Latin American Studies 25 (2) (May 1): 301-311.

[53] Ohkawa, Kazushi, and Henry Rosovsky. 1960. The Role of Agriculture in Modern Japanese Economic Development.” Economic Development and Cultural Change 9 (1) (October 1): 43-67.

[54] Ondetti, Gabriel. 2007. “An Ambivalent Legacy: Cardoso and Land Reform.” Latin American Perspectives 34 (5): 9-25.

[55] Parker, Phyllis R. 1979. Brazil and the Quiet Intervention: 1964. First ed. Univ of Texas Pr.

[56] Peláez, Carlos. 1972. História Da Industrialização Brasileira Crítica á Teoria Estruturalista No Brasil. [Rio de Janeiro]: APEC.

[57] Pereira, Anthony. 2003. “Brazil’s Agrarian Reform: Democratic Innovation or Oligarchic Exclusion Redux?” Latin American Politics and Society 45 (2) (July 1): 41-65. doi:10.2307/3176979.

[58] Perlman, Janice E. 2006. “The Metamorphosis of Marginality: Four Generations in the Favelas of Rio De Janeiro.” Annals of the American Academy of Political and Social Science 606 (July 1): 154-177.

[59] Petras, James. 2006. “’Centre-left’ Regimes in Latin America: History Repeating Itself as Farce?” Journal of Peasant Studies 33 (2): 278-303. doi:10.1080/03066150600819211.

[60] Petras, James, and Henry Veltmeyer. 2003. “Whither Lula’s Brazil? Neoliberalism and ’Third Way’ Ideology.”The Journal of Peasant Studies 31 (October 1): 1-44. doi:10.1080/0306615031000169116.

[61] Pino, Julio C. 1996. “Dark Mirror of Modernization.” Journal of Urban History 22 (4) (May): 419.

[62] Pino, Julio C. 1997. “Sources on the History of Favelas in Rio De Janeiro.” Latin American Research Review 32 (3) (July): 111.

[63] Raymond, Vernon. 1976. Oil Crisis. W. W. Norton and Company, Inc.

[64] Sampson, Anthony. 1991. The Seven Sisters: The Great Oil Companies and The World They Shaped. Bantam.

[65] Sánchez-Ancochea, Diego, and Lauro Mattei. 2011. “Bolsa Família, Poverty and Inequality: Political and Economic Effects in the Short and Long Run.” Global Social Policy 11 (2/3) (December): 299-318. doi:10.1177/1468018111421297.

[66] Sauer, Sérgio. 2006. “Market-Led ’Agrarian Reform’ in Brazil: The Costs of an Illusory Future.” In International Institute of Social Studies (ISS), The Hague. http://www.iss.nl/content/download/ 3611/35262/file/Sauer.pdf.

[67] Shaikh, Anwar. 2003. “Who Pays for the ’Welfare’ in the Welfare State? A Multicountry Study.” Social Research 70 (2): 531-550.

[68] Silva, José Francisco Graziano da. 1982. A Modernização Dolorosa: Estrutura Agrária, Fronteira Agrícola E Trabalhadores Rurais No Brasil. Coleção Agricultura e Sociedade; Variation: Coleção Agricultura e Sociedade. Rio de Janeiro: Zahar Editores.

[69] Silveira, Jos´e Luz, Lúcia Bollini Braga, Antonio Carlos Caetano de Souza, Julio Santana Antunes, and Rolando Zanzi. 2009. “The Benefits of Ethanol Use for Hydrogen Production in Urban Transportation.” Renewable & Sustainable Energy Reviews 13 (9) (December): 2525-2534. doi:10.1016/j.rser.2009.06.032.

[70] Skidmore, Thomas E. 1990. The Politics of Military Rule in Brazil, 1964-1985. Oxford University Press, USA.

[71] Skidmore, Thomas E. 1999. Brazil: Five Centuries of Change. Oxford University Press, USA.

[72] Soares, F.V., R.P. Ribas, and R.G. Osório. 2007. Evaluating the Impact of Brazil’s Bolsa Família: Cash Transfer Programmes in Comparative Perspective. International Poverty Centre.

[73] Soto, Hernando De. 1989. The Other Path: The Invisible Revolution in the Third World. 1st ed. Harpercollins.

[74] Stedile, João Pedro. 2007. “The Neoliberal Agrarian Model in Brazil.” Monthly Review: An Independent Socialist Magazine 58 (9) (February): 50-54.

[75] Stepan, Alfred. 1973. Authoritarian Brazil: Origins, Policies, and Future. First ed. Yale University Press.

[76] Stitzlein, John Noel. 1974. “The Economics of Agricultural Mechanization in Southern Brazil”. PhD. Dissertation, Ohio, US: The Ohio State University. http://pdf.usaid.gov/pdf docs/PNAAA621.pdf.

[77] Terzian, Pierre. 1985. OPEC, the Inside Story. Distributor, Biblio Distribution Center.

[78] The World Bank. 2004a. Project Information Document. http://documents.worldbank.org/curated/en/2004/04/ 3219327/brazil-first-bolsa-familia-apl-project.

[79] The World Bank. 2004b. Project Appraisal Document. http://documents.worldbank.org/curated/en/2004/05/ 4123829/brazil-bolsa-familia-project.

[80] United Nations Human Settlements Programme. 2003. The Challenge of Slums: Global Report on Human Settlements 2003. Routledge.

[81] Vargas, Getúlio. 1938. A Nova Política Do Brasil. Rio de Janeiro: J. Olympio.

[82] Wagner, F. E., and John O. Ward. 1980. “Urbanization and Migration in Brazil.” American Journal of Economics and Sociology 39 (3) (July 1): 249-259.

[83] Welch, Cliff. 1995. “Rivalry and Unification: Mobilising Rural Workers in São Paulo on the Eve of the BrazilianGolpe of 1964.” Journal of Latin American Studies 27 (1) (February 1): 161-187.

[84] White, Ben, and Anirban Dasgupta. 2010. “Agrofuels Capitalism: a View from Political Economy.” Journal of Peasant Studies 37 (October): 593-607. doi:10.1080/03066150.2010.512449.

[85] Wilber, Charles K. 1969. “The Role of Agriculture in Soviet Economic Development.” Land Economics 45 (1) (February 1): 87-96. doi:10.2307/3145264.

[86] Wolford, Wendy. 2004. “This Land Is Ours Now: Spatial Imaginaries and the Struggle for Land in Brazil.” Annals of the Association of American Geographers 94 (2) (June 1): 409-424.

[87] Wolford, Wendy. 2005. “Agrarian Moral Economies and Neoliberalism in Brazil: Competing Worldviews and the State in the Struggle for Land.” Environment & Planning A 37 (2) (February): 241-261. doi:10.1068/a3745.

[88] World Bank. 2010. Brazil Low-carbon Country Case Study. Washington DC: The World Bank. http://siteresources.worldbank.org/BRAZILEXTN/Resources/Brazil LowcarbonStudy.pdf.

[89] World Bank. 2011. “World Development Indicators Online: Brazil.” http://data.worldbank.org/country/brazil.

[90] Yergin, Daniel. 1993. The Prize: The Epic Quest for Oil, Money, & Power. Free Press.

91] Zucco, Cesar. 2008. “The President’s ’New’ Constituency: Lula and the Pragmatic Vote in Brazil’s 2006 Presidential Elections.” Journal of Latin American Studies 40 (01): 29- 49. doi:10.1017/S0022216X07003628.