Keywords

Introduction

While Brazil and India have emerged as important global players, they display key differences in the way their democracies have deepened and what this means for their state capacities . Since the liberalization of the economy in the early 1990s, Brazil pursued incorporation of previously excluded lower- and middle-class groups, creating a kind of state capacity equipped for broad sacrifices to provide universal public goods, what might be called state capacity for broad collective action. In India, by contrast, incorporation of previously excluded regional, ethnic, and communal groups created state capacity to work with elites atop strategic sectors, what might be called embedded capacity. These differences appear in tax regimes as Brazil has increased its tax take significantly, and to some degree progressively, while India has structured its tax system to favour the most dynamic and globally integrated sectors.

To explain these differences in state capacity, as expressed in tax, this chapter focuses on the political economy of democratic deepening. The concept of democratic deepening refers to the ever-broader incorporation of excluded groups into public, political processes. While both countries have deepened their democracies, the groups excluded in Brazil have largely defined themselves and mobilized around class identities while in India excluded groups have largely defined themselves and mobilized around ascriptive identities, such as region, caste, and religion.

Distinct patterns of democratic deepening connect to tax through the mechanisms of incorporation by which previously excluded groups engage with politics and the state. By articulating demands for protection against neoliberal adjustment and institutionalizing links to a party system organized around left-right cleavages, class-based democratic deepening in Brazil produced a social base in support of raising revenues to provide universally distributed social policies. By contrast, with demands articulated to expand participation in the benefits of neoliberal expansion, ascriptively based democratic deepening in India produced coordination among domestic capital to use tax incentives and subsidies to strategically upgrade exports.

This chapter evaluates changes to state capacity in Brazil and India by looking at the political economy of democratic deepening. The first section outlines the main concepts that orient the argument: democratic deepening and state capacity as viewed through tax. The second section applies these concepts to Brazil and India, analysing state capacity and taxation. The third section links tax capacity with democratic deepening, focusing on the type of incorporation of new groups and the particular coalitions that have been built in the two countries and how this affects revenue mobilization.

Analytical Framework

Deepening Democracy

A crisis of representation and participation in developed countries (Putnam 1994) and new democracies (Roberts 1998) has provoked debate about the quality of democracies. Weaknesses in democracy are evidenced by declining voter participation, feeble political associations, and limited social mobilization in the aftermath of transitions to democracy (O’Donnell 1996), while democratic practice is constrained by economic inequality, traditional relations of clientelism, and state fragility. These shortcomings truncate citizenship, diminishing the full range of individual rights and responsibilities and excluding social groups (Goldfrank 2011).

Deepening democracy requires the recognition of excluded social actors as legitimate participants in politics and the establishment of new mechanisms of institutionalized interaction with public authority. This has been referred to by some in the context of citizenship regimes, which “define who has political membership, which rights they possess, and how interest intermediation with the state is structured” (Yashar 2004: 6). One implication is that deepening democracy is a never-ending process in which new identities are constantly constructed, leading to the formation of new social groups and new mechanisms of linkage between state and society (Held 1995). This allows for constant innovation as collective actors engage in “insurgent citizenship” and mobilize for incorporation into democratic practice as excluded actors (Holston 2009).

Collective actors can emerge from a variety of ideologies and patterns of exclusion. One pattern defines exclusion along class and sectoral lines, such as working class, middle class, and rural and urban popular sectors. Alternative bases for mobilization include ascriptive ethnic, religious, and regional identities. While the intensity of different types of cleavages within societies may vary, the most important determinant of mobilization is the political dynamic by which groups come to view themselves as excluded, the strategies they choose to gain inclusion, and the outcomes of their struggles for inclusion, including the response of existing members of the polity.

Given the turn of events in Brazil and India in recent years, the concept of deepening democracy must also consider the possibility for its opposite—shifts towards authoritarianism. Just as excluded groups can mobilize for access and struggle to deepen democracy, they can be blocked or re-excluded in revanchist efforts by existing elites. There is no single arc of history that always and forever bends towards justice; political struggles between those who seek inclusion and those who seek to exclude are uncertain and contingent. This is especially the case in developing countries where advanced productive and social actors adapted to contemporary international trends coexist in multiple and overlapping relations with actors entrenched in prior patterns of international and domestic activity (Rosenberg 1996).

State Capacity

Deepening democracy has implications for state capacity. Elitist theories argue that excessive mobilization threatens to overwhelm state capacity for institutionalization and lead to breakdown. Such approaches presume a trade-off between social mobilization and state capacity for order (Huntington 1968). When considering episodes of state capacity increase, elitist approaches at best seek explanations in accommodations among elites, especially business groups and state elites (Mahon 2004; Fairfield 2015), regional and national elites (Schneider 2019), and globally-oriented, national capitalist, and oligarchic elites (Schneider 2012).

Alternatively, perspectives that emphasize the role of social coalitions in support of the state note the potential for deepening democracy to strengthen capacity in the sense of Michael Mann’s notion of “infrastructural power.” Infrastructural power is the ability of the state to penetrate and implement policies across a territory. In contrast to the “despotic power” that authoritarian regimes use to limit social mobilization, democratic deepening offers infrastructural power for states to work through society, engaging citizens in national projects (Mann 1984).

The incorporation of new social groups into an institutionalized relationship with the state buttresses infrastructural power in two relevant ways. First, it allows the state to stimulate greater sacrifice on the part of the citizenry to generate public goods. Typically, public goods are under-provided, as they face problems of free-riding and time-inconsistency that prevent sufficient contributions from self-interested individuals and groups (Olson 1965). States that have incorporated a broader range of social actors can draw on a wider selection and greater quantity of public contributions to pursue national projects. For example, Eugene Weber describes how the French revolution allowed the state to construct a common identity and set of institutions to incorporate previously scattered rural peasants, turning “peasants into Frenchmen,” who were willing to contribute soldiers and taxes in support of national development (1976).

A useful indicator of state capacity to stimulate collective action appears in public finance. Tax payments are “unrequited and compulsory” contributions of private wealth to the state (Buchanan 1963), in which increases in taxation can be seen as a collective action problem of getting individual taxpayers to sacrifice their resources in pursuit of social ends (Lieberman 2005). While every contributor might like others to contribute more, no revenues will be gathered if everyone puts the burden on someone else. Contributions can be conceived even more explicitly as examples of collective action when they are closely linked to spending policies allocated on a universal basis of social provision and protection (Sanchez-Ancochea and Martinez Franzoni 2016). While every beneficiary might prefer benefits allocated privately to them, this would lead to far fewer social benefits overall. In sum, broad-based collective action requires individuals and groups to accept collective sacrifice in pursuit of a common good, and strong public finance indicators for such collective action can result from an increase in tax capacity linked to expanded social spending and broader social protection.

The second kind of infrastructural power relevant here refers to “embeddedness,” the ability of the state to engage strategic sectors through specialized relationships and policies. This close relationship allows the state to engage in “husbandry” to modernize leading actors and cultivate their international competitiveness (Evans 1995). This draws on the Polanyian observation of the state role in creating and “governing” markets (Wade 1990), and fits with studies of global integration, in which the state helps fit national economic actors into global value chains (Kaplinsky 2005), and to move them into higher value activities of international scale and scope (Mazzucato 2015). This also includes government fiscal, regulatory, and financial efforts to attract foreign capital and technology.

Capacity for embeddedness has its reflection in tax regimes, as elite actors negotiate fiscal bargains in exchange for shifting into riskier dynamic sectors. Especially in developing countries, where domestic capitalist classes tend to be weaker and lack access to capital and technology, government tax policy can be especially important. For such policies to be effective, however, they cannot devolve into rent-seeking and protection; rather they must be oriented towards pushing economic actors into ever-new markets and higher value activities (Kurtz and Schrank 2005). Capacity for embeddedness can be indicated by tax policies that offer inducements to competitiveness, such as export-processing zones, often with sunset clauses or other incentives that wean domestic producers off protection and stimulate dynamism. Such incentives are especially indicative of capacity for embeddedness when they are connected to outlays to build the infrastructure for high-value exports.

State Capacity and Tax in Brazil and India

Both Brazil and India have undertaken important changes to their tax regimes in the past two decades, but those changes have produced quite different results. In Brazil, changes to the tax regime have expanded revenues, with particular expansion in direct taxes and taxes drawn from newly incorporated social groups linked directly to social programmes. In India, reforms targeted the most dynamic and internationally integrated sectors with exemptions and incentives, promoting privately-held high-value exports, but limiting increases in revenues.

Brazil: Expanded Capacity for Broad-Based Collective Action

Following a concerted effort to increase tax capacity from the mid-1990s to the present, Brazilian taxes are today among the highest in the developing world.Footnote 1 Tax revenue increased steadily, with notable reforms to the tax system including an increase in income taxes and the implementation of numerous contributions linked to social spending . These increases more than compensated for decreases in tariffs as Brazil liberalized international trade.

Figure 6.1 shows the increase in tax as a percentage of gross domestic product (GDP) between 1991 and 2017. Taxes increased from 23.7 per cent to more than 32 per cent. These revenues were necessary as the country faced fiscal insolvency during the 1980s and had to muster revenues to balance the budget. In addition, the government needed additional income to support expanded social spending mandated in the 1988 Constitution and to respond to demand pent up during the 20-year military regime that ended in 1985.

Fig. 6.1
figure 1

Tax as percentage of GDP in Brazil (1991–2017). (Source: Author calculations from CEPALSTAT 2015)

Full size image

A glimpse at the distribution of taxes and their attribution to different levels of government, illustrated in Fig. 6.2, highlights several details. The tax system is complex, made more so with a large number of contributions tied to specific social spending outlays. These are mostly collected at the federal level (social security, labour, health, welfare), and some use payroll as a base, while others calculate contributions on the basis of gross receipts. The single largest tax is the Tax on Circulation of Goods and Services (ICMS) levied by the states, accounting for 7.3 per cent of GDP.

Fig. 6.2
figure 2

Tax as percentage of GDP, by base and level of government, Brazil (2009). (Source: Author calculations from Secretaria de Receita Federal n.d.)

Full size image

With the most important tax on consumption—the ICMS—controlled by state governments, the federal government was forced to exchange trade taxes lowered by liberalization for income taxes. This replacement was reasonably effective, as federal taxes stood at 7.97 per cent of GDP in 1994 and 7.67 per cent in 2010. The effort to raise income taxes resulted in an increase in direct taxes in the 2000s. By 2009, direct taxes were almost 60 per cent of the total, up from barely 44 per cent in 2000. As a percentage of GDP, this represented an increase from 13.59 per cent to 17.83 per cent (Afonso et al. 2013).

To increase revenues further, the federal government made use of earmarked taxes, known in Brazil as contributions. The use of earmarks is a good indicator of the type of fiscal contract offered to citizens, with revenues collected tied explicitly to social spending . These increased from 8.95 per cent of GDP in 1994 to 12.94 per cent of GDP in 2010. Federal contributions included social security, a social security contribution on business receipts (COFINS), a contribution to unemployment benefits for dismissed workers (FGTS), a health contribution on receipts (CSS), a welfare contribution on receipts (PIS), a public sector social security contribution (CPSS), a contribution on profits towards social security (CSLL), a contribution on fuel towards education and health (CIDE), and a temporary contribution towards education and health on financial transactions (CPMF).

In addition to considering changes in the levels and structure of revenues, it is also possible to evaluate the incidence of tax in terms of impacts on distribution. For example, while the increase in direct taxes is likely to have positive impacts on vertical equity, indirect taxes (IT), which continue to account for about half of all taxes, are likely to have negative impacts on vertical equity (DIEESE 2009). A more precise way to explore the equity impact of changes to the tax system is in terms of the relative burden on income groups. The chart (Table 6.1) is taken from a study that uses household surveys to calculate the incidence of tax and displays the change from 1996 to 2004 in direct, indirect, and overall taxes. First, the change from 1996 to 2004 shows an increasing burden for all deciles. People pay more of their incomes in tax. Second, the increase is steadily greater for poorer deciles than it is for richer deciles (except a portion of the upper middle class), and the increase weighs heavier on the poor. Still, though they appear to have become less progressive over time, the burden of direct taxes continues to increase with wealth.

Table 6.1 Changes in direct tax and indirect tax incidence within total tax, Brazil (1996–2004)
Full size table

India: Capacity for Embeddedness

In India, tax capacity has largely held steady since the onset of neoliberal reforms in the early 1990s, as is shown in Fig. 6.3. The figure displays total revenue as a percentage of GDP in the upper line, with the amount collected by central government and states below.

Fig. 6.3
figure 3

Tax as percentage of GDP, India (1991–2017). (Source: Author calculations based on data from Indian Ministry of Finance 2014–2015. Notes: FY: fiscal year)

Full size image

In fiscal year 1991, taxes stood at 15 per cent of GDP while the measure for 2017 was 18.16 per cent of GDP. Within this relatively steady overall trend, there were some fluctuations, as taxes trended downwards most years between 1991 and 2001 and again during the international slowdown in 2008–2009. A combination of reforms and rapid growth offered reasonable upward movement that made up for drops in revenues between 2001–2002 and 2006–2007, and revenues have fluctuated around the 16–18 per cent level since then.

It may not be surprising to observers that levels of tax capacity in India are lower than in Brazil, bearing in mind the country’s economic structure, poverty, and other characteristics that typically predict tax effort. Recent studies of tax effort however show that, even taking into account typical determinants of tax burden, India’s tax capacity is quite low. According to the International Monetary Fund (IMF), India’s actual tax effort is only 52.2 per cent of the predicted level when controlling for income per capita, economic openness, agriculture as a share of GDP, spending on education, income inequality, corruption, and inflation (Fenochietto and Pessino 2010).

In terms of distributional impact of tax, there have not been the same studies of incidence in India as in Brazil, but some conclusions can be drawn from the relative burden of direct and indirect taxes. Direct taxes on income, wealth, property, and capital gains tend to fall more heavily on those who are wealthier, and in India there would appear to have been a shift in the tax structure towards direct taxes, from 16 per cent of total revenues to a peak of 43 per cent in 2009–2010. Since 2009–2010, direct taxes have fallen a bit, steady at around 37 per cent over the past few years. The changing relative proportion of direct and indirect taxes is displayed in Fig. 6.4. Still, the low overall burden of taxes and the ongoing dependence on indirect taxes suggest that there is significant room to expand direct taxes and increase the progressivity of the tax regime.

Fig. 6.4
figure 4

Direct and indirect tax as percentage of total tax, India (1991/1992–2018/2019). (Source: Author calculations based on data from the Reserve Bank of India n.d.)

Full size image

More worrying still is that most of the direct taxes come from corporate income tax, which reflects the rapid growth in the country and the degree to which profits, as opposed to wages, have absorbed that growth (Sood et al. 2014). Further, reforms to the Income Tax Act of 1961 and the Wealth Tax Act of 1957, which were initially advanced in the 2009–2010 budget, caused revenue losses, drops in marginal rates on corporate income tax, and shifts in personal income tax brackets upwards such that most people fell into the lowest bracket. These changes had regressive impacts overall, as seen in an upward movement in the share of indirect taxes in total revenues to 61.6 per cent in 2012–2013.Footnote 2

A number of specific differences emerge when exploring the role of social contributions in the fiscal revenues of the two countries. Brazil, as discussed earlier, gathers a significant portion of its revenues in contributions, 34.8 per cent in 2017 according to World Bank Development Indicators. By contrast, only 0.19 per cent of central government revenues in India came from social contributions in 2013,Footnote 3 whereas labour taxes as a percentage of commercial profits are nearly twice as high in Brazil (39.4 per cent in 2018) compared to India (20.3 per cent in 2018) (World Bank n.d.). This is explained on the one hand by the lack of mobilization of social contributions in India and the large size of the informal sector, hovering around 88.2 per cent in India according to the International Labour Organization (ILO 2018).Footnote 4

While in India the capacity for broad-based collective action is limited, the state has instead cultivated its ability to engage with leading market actors. The estimated revenues foregone by India from incentives reached 5.9 per cent of GDP in 2011–2012 and 5.7 per cent in 2012–2013, more than a third of total revenues (CBGA 2013). Incentives are oriented to boost the fortunes of the wealthiest and most dynamic sectors; corporate tax incentives provide proportionally greater benefit to companies with larger profits, who pay lower tax rates than firms with smaller profits. Also, most exemptions apply to indirect taxes, customs, and excise duties, with 80 per cent of total taxes foregone occurring in these categories.

In addition, Indian exemptions are particularly oriented towards dynamic sectors deeply integrated in the international economy. In total, India has over 170 special economic zones in which tax incentives, among other benefits, are offered, often in the form of a deduction from corporate income of export profits (India Ministry of Commerce and Industry 2015). Software technology parks also enjoy special incentives, and Information and Technology enabled services and Business Process Outsourcing providers have seen tax rates as low as 7.4 per cent in 2006–2007, averaging 15.98 per cent, far lower than the 33.21 per cent statutory rate (CBGA 2013).

The difference between capacity for broad-based collective action and embeddedness appears starkly in Brazilian and Indian public finance. While Brazil has greatly expanded its revenues, especially through direct taxes and the use of earmarked contributions tied to social outlays, India has allowed revenues to stagnate while using the fiscal regime to favour exporters and dynamic, internationally integrated sectors. An explanation for these differences in capacity can be found in patterns of democratic deepening.

Deepening Democracy and Tax Capacity in Brazil and India

In India and Brazil, struggles for political access during the 1970s and 1980s legitimated excluded groups as collective actors, and in the 1990s they consolidated their organizational and partisan strategies and articulated demands in the context of neoliberal adjustments. With renewed growth during the 2000s (slightly earlier in India), previously excluded groups sought policies and institutions that would provide them with stable linkages to the state. This incorporation of excluded groups deepened democracy in both countries, but the results have been quite different with regard to state capacity as expressed in tax. In Brazil, popular sectors and middle classes formed a social coalition in support of efforts to expand tax revenues, make them more progressive, and connect them to universal social policies. In India, efforts to sustain internationally competitive leading sectors focused on targeting publicly financed benefits to urban middle classes and caste, regional, and other identity-based groups in shifting social coalitions. The countries differ in terms of which excluded groups mobilized, what demands they articulated, and the mechanisms of incorporation that linked them to the state.

Brazil: Cross-Class Coalitions

The following paragraphs trace the extension of citizenship regimes in Brazil that occurred with the mobilization of certain excluded groups, their articulation of demands, and linkage to the state during the period from the 1970s to the 2000s. The military leaders who governed Brazil from 1964 to 1985 promoted industrial deepening at the same time as they shut down democratic politics and brutalized opponents. Middle classes, including public sector workers and private urban professionals, had originally been a privileged segment within Brazilian state-led development, but certain segments, such as students and intellectuals, were among the most repressed by the military. Middle-class struggles for human rights brought them together with a wide range of social movements addressing citizenship and quality of life issues, including Afro-Brazilian rights, women’s rights, environmental protection, neighbourhood services, and public health movements, among others (Escobar and Alvarez 1992). Their efforts came to be understood as the struggle for the “right to have rights” (Dagnino 2007).

Working classes also mobilized in opposition to the military. They bore the brunt of both repression and regressive growth strategies, even as deepening import-substitution industrialization expanded their role in the economy. Gradually, urban formal sector workers created a space for autonomous organizing and struggle, convened on the shop-floor to avoid the reactionary control of sector-wide unions, and expanded their workplace gatherings to include community demands for public services.Footnote 5 Strike waves in the late 1970s were especially vigorous in the manufacturing belt around São Paulo, where an alternative national federation of unions, Unified Workers Central (CUT), formed in 1983.

The urban formal sector was joined also by other groups, such as informal sector workers and especially the rural landless population, which swelled in number and particularly suffered from political repression and biased growth strategies. Available land dwindled as national policy prioritized large-scale surplus production for export and industrial requirements of power and natural resources. Politically, the rural poor experienced renewed subjugation, as the military responded to faltering electoral support with dependence on rural elite allies and their clientelist networks in the poorest regions of the country (Hagopian 1996). Rural workers resisted, and with the help of the progressive Catholic Church (Houtzager 2000), developed a rural landless workers’ movement with a series of occupations in 1979, becoming one of the largest social movements in Latin America, the Landless Workers’ Movement (MST) (Woolford 2010).

These popular-sector and middle-class movements increasingly established their legitimacy as collective actors supporting democratization, as the military regime offered a unified target. Strikes and protests accelerated over the course of the 1970s, and the gradual opening of electoral competition after 1973 oriented at least part of the democratization struggle into the party system, with the main aggregator of opposition in the Democratic Movement of Brazil (MDB later PMDB), later joined by the Workers’ Party (PT), formed in 1980. The PT was a new kind of partisan organization in Brazil, considered an “anomaly” compared to previous political formations, as it emerged from a “solid base in labour and social movements,” with much of “its leadership drawn from the labour movement” (Keck 1992). As a party that emerged “from the bottom up” (Nylen 1997) with “extra-parliamentary” origins (Meneguello 1998), the PT was committed to the autonomy of its movement and union allies.

Once the military finally exited power in 1985, the 1988 Constituent Assembly included many of the demands of social movements and popular sectors. These demands called for expanded welfare state policies, including mandates for universal provision in health and education, expanded funding for housing and sanitation, as well as greater decentralization and participation (Draibe 2003). These policies would require greater resources, but the first priority of the candidate who entered office in 1990, Fernando Collor, was to stabilize an economy facing runaway inflation. Though he was eventually removed for corruption in 1992, Collor began the liberalization of trade and deregulation of the Brazilian economy that would accelerate under Fernando Henrique Cardoso’s Party of Brazilian Social Democracy (PSDB) government from 1994 to 2002.

For those who had mobilized for democratization, the battleground shifted to the struggle over neoliberal adjustment. Cardoso earned convincing electoral victories on a base of middle- and upper-class supporters attracted by the technocratic and social democratic credentials of PSDB party founders, and defeating inflation extended his support to popular sectors who could finally begin consuming again (Baker 2010). Backed by this coalition, he took advantage of expanded contributions tied to social policy outlays, fulfilling some of the demands of the 1988 Constitution and securing resources for macroeconomic stabilization.Footnote 6

Yet, many previously excluded actors were increasingly repelled by neoliberal elements of stabilization. Social movements that had fought for democratization opposed Cardoso’s efforts at centralization, as well as the extensive use of decrees and other executive privileges that removed aspects of policy from public debate. Formal sector and manufacturing workers opposed privatization and the deindustrialization that followed trade liberalization (Hunter 2010), and those who had supported expanded social policies objected to the prioritization of macroeconomic stabilization.

Increasingly, the coalition of working classes, social movements, and middle-class actors who had mobilized for democratization articulated their demands in opposition to neoliberal adjustment. In the process, they drew closer to the PT, which remained in opposition to the adjustment strategy of the Cardoso government (Roman 2012). As an indicator of the articulation of demands around neoliberal adjustment, shifts in the party system showed an increasing left-right orientation and polarization (Hagopian et al. 2009). From 1990 to 1998, the PT steadily increased its support from 35 to 49 to 58 Congressional representatives, while Cardoso’s PSDB increased from 37 to 67 to 99. Right-wing parties associated with the military regime disappeared and were absorbed into right-wing supporters of the PSDB’s neoliberal stabilization strategy.

When Luiz Inácio Lula da Silva, a union leader and founding member of the PT, finally won the presidency in 2002, it was not clear that he would be able to successfully institutionalize linkages to working classes, social movements, and middle classes. To win the presidency, he had signalled that he would moderate, and the government maintained Cardoso’s inflation targeting regime. Yet, a rebounding economy allowed the PT to pay off IMF obligations early and provided much-needed policy space. The government adopted an expansionary macroeconomic stance beginning in 2006, as Lula won re-election to a second term. During the 2008 international financial crisis, the government embarked on counter-cyclical fiscal expansion, led by the Growth Acceleration Programme inaugurated that year that spent a total of Brazilian reals (BRL) 1.7 trillion (USD 817 billion) (Ministerio de Fazenda 2014).

The expansionary fiscal strategy targeted precisely those previously excluded groups who had mobilized for democratization, articulated demands against neoliberalism, and now achieved policy and institutional linkage to the state through the PT. Middle-sector professionals benefited from employment and increased salaries and pensions, and government banking institutions further expanded consumer and housing credit to reach lower middle classes. The middle class increased to 55 per cent of the population, 105.5 million people (Barbosa 2010).

Even more significant were the efforts to expand the incomes and consumption of working classes and popular sectors. Successive increases in the monthly minimum wage, from BRL 350 in 2002 to BRL 560 in 2012Footnote 7 have improved conditions across the board, as many low-income jobs and pensions are indexed to minimum wages. Further extending the impact is the extension of pensions to informal sector workers, reaching 28 million people, as well as small farmer credits and agricultural extension that reached almost two million small producers (Melo 2008).

The flagship social policy, an income transfer programme called Bolsa Família, has absorbed and expanded various social programmes to extend income support, education, and health to low-income families. At its peak, Bolsa Família reached more than 40 million Brazilians and was credited with cutting poverty, improving health outcomes, and improving education attainment (Castiñeira et al. 2009; Hall 2006). According to World Bank data, the percentage of the population in poverty fell from 22.4 per cent in 2002 to 6.9 per cent in 2014, and the Ministry of Finance reports the incomes of the poorest three deciles had annual average rates of growth of 7.2, 6.3, and 5.9 per cent (as compared to 1.4, 2.5, and 3.3 per cent for the highest deciles) (2012). This has produced a fall in the Gini coefficient from 0.596 in 2001 to 0.519 by 2012 (Ministry of Finance 2012).Footnote 8

The PT scaled up participatory innovations it had experimented with while governing at the local level during the 1990s (Lavalle et al. 2005). Participatory institutions had been mandated in the 1988 Constitution, and the PT had developed local-level participatory budgeting into a global reference.Footnote 9 With national power after 2002, the PT significantly expanded participatory councils for public policy. In 2010 they operated in 31 different policy areas, included 1350 members—with slightly more civil society (55 per cent) than government (45 per cent) representation—and undertook both deliberative and advisory tasks (Lopez and Pires 2010). As of 2005, there were over 300,000 registered civil society organizations, and by 2009, they were receiving over BRL 14 billion (USD 8 billion) in government transfers.

In targeting working class, social movement, and middle-class actors, policies and institutions stabilized linkages to the state for previously excluded groups. This once again shows up in the party system. While class-based voting patterns had been somewhat muddied by the anti-inflation consensus of the 1990s, a class cleavage consolidated in which the PT polled better among poorer voters and in poorer regions (Roman 2012). Voting intention for the PT was higher at lower income levels, with average support over the four elections from 2002 to 2014 at 57 per cent for those in the lowest income bracket and 42 per cent for those in the highest income bracket. Further, support from those with only primary education averaged 61 per cent over the same four elections, while support from those with tertiary education averaged 43 per cent (Datafolha 2014; IBOPE 2013).

PT support for social spending sought to institutionalize linkages to the cross-class coalition that had formed in democratization and which had articulated its opposition to neoliberalism in the 1990s. The counterpart to these social spending expansions was tax increases, including efforts to increase progressive taxation.

Two puzzles require further explanation. The first is the failure of the PT to construct more embedded forms of capacity in terms of forging alliances with elements of the national bourgeoisie and cultivating them into internationally competitive positions such that Brazil was not locked into a dependence on commodity exports. The second puzzle was revealed in the 2016 removal of the Workers’ Party from power and rapid disintegration of their incorporation project by right-wing leadership.

At the heart of both puzzles is the historical dependence of the government on oligarchic factions of Brazilian capital, whose economic base rested on commodity exports and whose social base rested on clientelist links to voters. Dependence on oligarchic elites was in part a function of the party system, which was characterized by high levels of fragmentation and volatility, resulting in minorities in Congress for the party of the president (Kinzo 2004). To survive defections during legislative battles, executives built super-majority coalitions, a practice labelled “presidential coalitionism” (Abranches 1988). The most easily attracted partners for such coalitions were oligarchic and clientelist politicians whose need for patronage made their price relatively cheap.

Critical in such coalitions has been the PMDB, always among the top-three holders of seats in the legislature and an anchor of coalitional presidential strategies since the 1980s (Abranches 1988). From its origins as a partisan umbrella for a wide range of opponents against the military regime, the PMDB evolved into a catch-all vehicle dependent on oligarchic elites based in commodity-exporting states (Nobre 2010). The PMDB, and other conservative parties such as the Partido Republicano Brasileiro (PRB), Partido Progressista (PP), and Partido da Republica (PR), offered support for the government, but their leading actors were relatively uninterested in diversifying away from a commodity-export economy (Bresser-Pereira 2015).

This dependence on oligarchic factions of local elites also helps explain the 2016 impeachment of PT President Dilma Rousseff and the rapid roll back of universal social provision. The impeachment paved the way for explicit efforts to de-incorporate exactly the sectors incorporated since democratization, including a 20-year spending cap that will greatly reduce social spending and protection, a labour reform that stripped rights existing since the 1930s, and a proposed pension reform that would weaken coverage, lower benefits, and raise the age of retirement.

What sealed the fate of the inclusionary project was the sudden defection of middle classes to an alliance with oligarchic interests (Souza 2015). While the Workers’ Party was unable or unwilling to manipulate judicial and legislative tools to escape an onslaught of corruption allegations, oligarchic elites showed no such restraint. Middle-class acceptance, and even encouragement, of a witch-hunt against leftist politicians while leaving oligarchic interests unscathed, exposed their essential suspicion of popular sectors and leftist projects as well as their essential weakness as political actors, unable to drive an agenda and able only to support those projects offered to them by other class actors.

In sum, the policies and institutions introduced since democratization stabilized linkages between the state and previously excluded groups such as working classes, social movements, and middle classes in support of expanded revenue mobilization and greater progressivity, especially when linked to social outlays. Yet, there was limited ability to coordinate dynamic sectors and escape commodity dependence, and the project was ultimately undermined when oligarchic elites turned on the Workers’ Party and middle classes abandoned the cross-class coalition, restoring a neoliberal agenda to national politics.

India: Cross-Group Coalitions

In India, the Congress Party led the struggle for independence gained in 1947 and dominated politics until the 1970s, during which time it built a support base buoyed by a nationalist and developmentalist consensus. In the heavily populated and therefore electorally crucial north of India, Congress cohered a social coalition that combined a largely upper-caste leadership with a mass base among lower caste and Dalit (those who were previously labelled “untouchable” and officially labelled Scheduled Caste/SC)—a “coalition of extremes” as described by Brass (1984). To delay and deter potential rivals, Congress leaders were adept at absorbing “aspiring social groups” (Weiner 1967), preserving an elite coalition of political-bureaucratic state actors, large farmers, and industrial elites benefiting from state-led development (Bardhan 1984).

This was labelled the “Congress system” by Kothari (1964), and it left room at its margins for alternative groups to mobilize around identities not incorporated in the coalition of extremes. The “Congress system … (is) a party of consensus and parties of pressure. … These groups outside the margin do not constitute alternatives to the ruling party. Their role is to constantly pressurize, criticize, censure and influence it” (Kothari 1964: 1162). Such groups grew in strength as the Congress organization decayed and emerged especially in regions where the social and political coalition undergirding Congress dominance was already weaker (Jaffrelot 2003). The alternative to Congress’s upper- and lower-caste/class coalition has been labelled a “sandwich coalition” (Swamy 2010), uniting Other Backward Classes (OBC), regional interests, middle classes, and Hindu nationalists left out of the Congress system.

The breakthrough for alternatives to Congress dominance happened in the North a decade later. Indira Gandhi was able to win a landslide election for Congress in 1971 with aggressive efforts to target the poor, symbolized by the slogan “garibi hatao” (abolish poverty). Yet, her strategy weakened the large-scale farmers that brokered Congress control in the countryside, opening space to increase mobilization among middle-peasant groups who had seen increases in resources and leverage as a result of the Green Revolution (Varshney 1998).

Eventually, the anti-Congress opposition united and won elections in 1977 led by the Janata Party (Kohli 1987). The Janata coalition included Hindu nationalist Bharatiya Jan Sangh, the market-oriented Swatantra Party, the middle-peasant-dominated Bharatiya Lok Dal, and the OBC-oriented Samyukta Socialist Party. This marked the arrival of a previously excluded middle class, OBC, medium-sized farmers, and regional interests as collective actors, and they formed a similar anti-Congress coalition in the 1989 election of the National Front.

For several reasons, previously excluded groups articulated demands in opposition to the Congress system by demanding recognition and benefits targeted on the basis of caste, regional, and other ethnic identities (Frankel 2004). One reason can be traced to the decision of the National Front to implement the recommendations of the 1980 Mandal Commission, reserving 27 per cent of government jobs for OBCs (Jayal 2005). This fulfilled one demand of the OBC movements but “divided the ‘backward’ from the ‘forward’ segments of the middle-peasant/backward-caste coalition, pitting peasant castes from different regions against each other” (Swamy 2010). Allocations of reservations depended on regional balances of caste demographics and conditions, producing regionally specific polarities of competition among different OBC, SC, and other identity-based movements, many of which formed caste-specific political parties to campaign for recognition and benefits (Jaffrelot 2003).

This identity-based orientation of demands received further impetus as Hindu nationalists in the Bharatiya Janata Party (BJP) stimulated controversy around the destruction of a Muslim shrine at Ayodhya, wooing upper classes that opposed protections for minorities and upper castes that opposed reservations. With the fragmentation of the party system preventing any outright majorities, partisan competition stimulated efforts by both the BJP and Congress to prime identity-based cleavages to patch together governing majorities of shifting, regionally specific, caste-based partisan allies (Yadav and Palshikar 2006).

The final factor focusing demands on identity-based issues was the essential agreement of Congress and BJP on the strategy of economic adjustment. Both parties pressed liberalizing reforms, starting with the Congress government of 1991 and deepening under the BJP governments of 1996 and 1998. Their implicit agreement kept debates over stabilization out of the public eye and reforms passed by “stealth,” meaning through bureaucratic regulations and easily passed parliamentary measures (Jenkins 2000). At the same time, more difficult reforms were pushed to state governments, whose chief tools included market liberalization and export promotion (Sinha 2005), frequently emphasizing tax privileges and incentives for investment.

Over time, the BJP appeared to be more capable of forming alliances on a state-by-state basis, and its 1998–2004 government at times included as many as 24 parties. The BJP developed an “elite revolt” of Hindu nationalism and neoliberal market promotion with a base in upper castes and middle classes (Corbridge and Harriss 2000). Middle classes rejected what they perceived as corrupt and populist policies of Congress attempts to appeal to poor voters, and upper caste Hindus rejected preferential policies targeted at lower castes as a result of the Mandal Commission.

The BJP party cadre structure and its mass organization, the Rashtriya Swayamsevak Sangh (RSS), operate within a broader network of Hindu social organizations, the Sangh Parivar (Jaffrelot 1996). They have displayed the capacity to cultivate an internationally oriented middle class who blend Hindu nationalist and pro-business agendas, as evidenced by the deep ties between Sangh Parivar and non-resident Indians (NRI). NRI international success is taken as an indication of “Hindus’ world historic mission,” and the donations they send home sustain the RSS and similar organizations (Gopalakrishnan 2006).

Interestingly, the RSS has also proved sufficiently effective at penetrating poorer districts to balance efforts by other actors, especially Congress, in advancing a pro-poor agenda. While Congress reverts to its Congress-system practice of depending on local elite brokers, the RSS sends campaigners and charitable enterprises into poor districts, generating information about local political dynamics, stimulating Hindu supremacist cleavages, and neutralizing class-based opposition among urban and rural working classes and poor (Thachil 2011).

The 2014 election landslide expresses the electoral potential of the BJP coalition of middle classes, upper castes, and subaltern groups. Such a combination had been successfully exhibited by new Prime Minister Narendra Modi during his term as chief minister of the state of Gujarat, as he overcame the traditional Congress coalition to oversee a period marked by both rapid growth and anti-Muslim programmes.

Despite the dexterity of BJP alliance strategies and its potential coalition, it was the Congress government that attempted to institutionalize mechanisms of incorporation for previously excluded groups from 2004 to 2014. Congress’s 2004 victory came somewhat as a surprise, as the Indian economy had already begun to achieve high rates of growth under the BJP, but the Congress-led United Progressive Alliance (UPA) won with allies such as the OBC-led Samajwadi Party (SP) and Dalit-led Bahujan Samaj Party (BSP), along with the Left Front of Communist Parties. With a mandate to respond to those ignored during neoliberal stabilization, the coalition partners agreed a Common Minimum Programme that stabilized linkages to previously excluded groups through institutions, policy, and legislation (Saez and Singh 2012).

The chief institutional innovation was the National Advisory Council (NAC), formally outside government, led by Sonia Gandhi, and composed of notable civil society activists with a mandate to provide policy advice to the government (Hasan 2012). As an autonomous and appointed body, the NAC appealed to the cohort of middle-class activists eager to provide input to public policy. On the other hand, when NAC initiatives stalled with the first UPA government, the council had to be reconstituted for the second UPA government starting in 2009, and its lack of governmental authority or civil society mobilization meant that it could be easily disbanded when the BJP returned to power in 2014 (Arora and Kailash 2015).

Still, a number of NAC policy innovations stabilized linkages to previously excluded groups. The most important was MNREGA, the Mahatma Gandhi National Rural Employment Guarantee Act, targeting small farmers and landless workers, many of whom had provided the support base to Dalit or OBC movements. Passed in 2005, MNREGA provided up to 100 days of work to each rural household at a minimum wage rate, and the average wage across all states has increased 180 per cent between 2005 and 2019. By 2010–2011, it had been rolled out in all districts that were not completely urban, reaching over 110 million people in 2019. On average, households received 50.85 days of work per year, and 19.48 per cent of beneficiaries were scheduled caste and 16.31 per cent scheduled tribe, well above their population percentage. As a result of the programme, which displayed 54 per cent women beneficiaries (Money Control 2018), rural wages increased an estimated 4.3 per cent, and households began to receive up to 8 per cent of their annual income from the programme.Footnote 10

What distinguished the social policy regime pursued under the UPA governments was the legislation of social policy as rights, including legislation, covering rights to information, food, education, compensation for land acquisition, and forest rights. Rights-based legislation offered a justiciable mechanism by which individuals could hold the state to account, meeting one of the chief demands of middle-class activists hoping to weaken the control of local elites and corrupt or inefficient state bureaucrats (Jenkins and Goetz 2005).

Yet, Congress efforts at incorporation failed to unite a cross-class alliance in favour of mobilizing revenues to support its ambitious agenda. Segments of the middle class were unconvinced by the rights-based approach, especially as Congress’s reputation was undermined by corruption scandals related to telecommunications licensing, coal, and construction for the 2010 Commonwealth Games, among others. It became increasingly difficult to expect middle classes to support mobilizing greater revenues from leading sectors, especially as they were the core of the urban, educated sector that was leading Indian emergence in international exports of technology and services (Fernandes and Heller 2006). Middle-class Indians oppose transfers for poor citizens as examples of “vote-buying” and efforts to target lower castes as “populism” (Kapur 2010).

Mechanisms of incorporation initiated by the UPA governments remained targeted to identity-based ethnic groups, with variable impacts across regions depending on political and caste configurations. For example, the UPA adopted what some labelled Mandal II by extending reservation of 27 per cent of seats to the premier federal government university system, including Indian Institutes of Technology, Indian Institutes of Management, and the All India Institute of Medical Sciences. As before, membership in an OBC category remained defined by states (Thorat 2005). Such targeting is cheaper than universal policies but reinforces divisions among subaltern groups and failed to generate a stable majority cross-class coalition in favour of progressive expansions of state capacity. Those policies that passed as universal rights, such as rights to information, education, and food, putatively altered the relationship between state and society, but were not supported with fiscal resources that made such rights a reality.

The result for India is a citizenship regime in which regional, caste, and ethnic actors have secured access to politics, but the targeted mechanisms of incorporation allow cycling among different combinations of support groups. The resulting instability limits the potential for collective projects such as increasing tax capacity, and progressivity in revenues is blocked by the elite consensus around neoliberal strategies of growth and middle-class opposition to redistribution. Decentralizing authority to states has intensified rivalry among different identity-based actors and across regions, producing incentive regimes that compromise both progressivity and universality in the tax system.

Conclusion

This chapter explored the nature of democratic deepening in Brazil and India and their relationship to state capacity. More specifically, India and Brazil differ in developing state capacity to raise revenue and channel these into social programmes or provide incentives to economic actors to support efforts at economic diversification.

In Brazil, middle-class and popular-sector actors legitimated themselves as collective actors in the struggle for democratization and constructed demands around opposition to neoliberalism. Their linkage to the state was institutionalized through social policies, state institutions, and left-right partisan competition. Cross-class democratic deepening built state capacity for broad-based collective action for social policies.

In India, ascriptive groups legitimated themselves in opposition to the Congress system and constructed demands for inclusion into the benefits of neoliberalism. Their linkage to the state was institutionalized through symbolic and targeted appeals channelled through bi-polar competition between national partisan coalitions led by the Congress and BJP. Cross-group democratic deepening built state capacity for embeddedness in local elites and cultivated international upgrading.

The patterns observed in these two countries suggest a number of implications. Development promotion is strengthened when both state capacity to stimulate broad collective action, as found in Brazil, as well as embeddedness in strategic sectors, as found in India, are present. Yet, the political determinants of these capacities are quite different. The kinds of social coalitions that produce one kind of capacity differ from the social coalitions that produce other capacities.

In addition, both kinds of state capacity are extremely vulnerable to reversal, especially in developing countries. In Brazil, dependence on the cooperation of oligarchic elites left state capacity vulnerable, as evidenced by the rapid defection of oligarchic allies once commodity prices dropped. In India, dependence on the cooperation of elites atop strategic sectors left state capacity vulnerable in other ways, as they could turn even further away from excluded minorities at times of slowdown, as evidenced by the rise of the Hindu nationalist BJP. For late developers, state capacity including capacity to mobilize domestic resources through taxation is essential to catching up. Unfortunately, the political determinants of state capacity are subject to variation across context and vulnerable to shifts in domestic and international relations.