{ SIXTEEN }

Social Nativism: The Postcolonial Identitarian Trap

In the previous chapters, we looked at the transformation of political and electoral cleavages in the United Kingdom, United States, and France since World War II. In particular, we saw how, in all three countries, the “classist” party systems of the period 1950–1980 gradually gave way in the period 1990–2020 to systems of multiple elites, in which a party of the highly educated (the “Brahmin left”) and a party of the wealthy and highly paid (the “merchant right”) alternated in power. The very end of the period was marked by increasing conflict over the organization of globalization and the European project, pitting the relatively well-off classes, on the whole favorable to continuation of the status quo, against the disadvantaged classes, which are increasingly opposed to the status quo and whose legitimate feelings of abandonment have been cleverly exploited by parties espousing a variety of nationalist and anti-immigrant ideologies.

In this chapter, we will begin by verifying that the evolutions observed in the three countries studied thus far can also be found in Germany, Sweden, and virtually all European and Western democracies. We will also analyze the singular structure of political cleavages in Eastern Europe (especially Poland). This illustrates the importance of postcommunist disillusionment in the transformation of party systems and the emergence of social nativism, which can be seen as a consequence of a world that is at once postcommunist and postcolonial. We will consider the extent to which it is possible to avoid the social-nativist trap and outline a form of social federalism adapted to the European situation. We will then study the transformation of political cleavages in non-Western democracies, notably India and Brazil. In both cases we will find examples of incomplete development of cleavages of the classist type, which will help us to gain a better understanding of both Western trajectories and the dynamics of global inequality. Finally, with all these lessons in mind, we will turn in the final chapter to the elements of a program for creating, in a transnational perspective, new forms of participatory socialism for the twenty-first century.

From the Workers’ Party to the Party of the Highly Educated: Similarities and Variants

To be clear from the outset: we will not be able to treat each of the subsequent cases in as much detail as we studied France, the United States, and the United Kingdom, partly because to do so would take us beyond the scope of this book and partly because the necessary sources are not available for all countries. In this chapter, I will begin with a relatively succinct presentation of the main results currently available for the other European and Western democracies. I will then analyze in greater detail the results for India (and in somewhat less detail for Brazil). Not only does India’s democracy include more voters than all the Western democracies combined, but study of the structure of the Indian electorate and of the transformation of India’s sociopolitical cleavages from the 1950s to the present illustrates the urgent need to go beyond the Western framework if we are to gain a better understanding of the political-ideological determinants of inequality as well as the conditions under which redistributive coalitions can be assembled.

In regard to Western democracies, the principal conclusion is that the results obtained for the United Kingdom, United States, and France are representative of a much more general evolution. First, we find that the reversal of the educational cleavage took place not only in the three countries already studied but also in the Germanic and Nordic countries that constitute the historic heart of social democracy: Germany, Sweden, and Norway (Fig. 16.1). In all three countries, the political coalition associated with the workers’ party in the postwar decades (which did particularly well among more modest voters) became the party of the educated in the late twentieth and early twenty-first centuries, achieving its highest scores among those who obtained higher degrees.

In Germany, for example, we find that, in the period 1990–2020, the vote for the Social Democratic Party (SPD) and other parties of the left (especially Die Grünen and Die Linke) was five to ten points higher among highly educated voters than among the less educated, whereas it was around fifteen points lower in the period 1950–1980. To make the results as comparable as possible across time and space, I will focus here on the difference between the vote of the 10 percent most diplomaed voters and the 90 percent least diplomaed (after controlling for other variables). Note, however, that as in the French, American, and British cases, the trends are similar if one compares voters with and without college degrees or the 50 percent most educated with the 50 percent least educated, both before and after controlling for other variables.¹ In the case of Germany, note that the amplitude of the reversal of the educational cleavage is almost identical to that observed in the United Kingdom (Fig. 16.1). Note, too, the role played by the emergence of the Greens (Die Grünen) in shaping the German trajectory. From the 1980s on, the ecological party has attracted a substantial share of highly educated voters. Yet one still sees a reversal of the educational cleavage (though less pronounced at the end of the period) if one focuses exclusively on the SPD vote.² Broadly speaking, although the institutional structure of parties and factions varies widely from country to country as we saw in comparing France with the United States and United Kingdom, it is striking to see how limited the effect of these differences is on the basic trends that interest us here.

FIG. 16.1.  The reversal of the educational cleavage, 1950–2020: United States, France, United Kingdom, Germany, Sweden, and Norway

Interpretation: In the period 1950–1970, the vote for Democrats in the United States and for various left parties in Europe (labor, social-democratic, socialist, communist, radical, and ecologist) was higher among less educated voters; in the period 2000–2020, it was higher among more educated voters. The change came later in Nordic Europe but went in the same direction. Note: “1950–59” includes elections from 1950 to 1959, and “1960–69” includes those from 1960 to 1969, and so on. Sources and series: piketty.pse.ens.fr/ideology.

In Sweden and Norway, the politicization of the class cleavage in the postwar period stands out starkly. Specifically, the social-democratic vote among the 10 percent most highly educated voters was on the order of thirty to forty points lower than among the remaining 90 percent in 1950 to 1970. By contrast, this same gap was on the order of fifteen to twenty points in Germany and the United Kingdom and ten to fifteen points in France and the United States (Fig. 16.1). This shows the degree to which the Nordic social-democratic parties were built around an exceptionally strong mobilization of the working class and manual laborers.³ This mobilization was necessary, moreover, to overcome the particularly extreme proprietarian inequality that persisted into the twentieth century (particularly in Sweden, where the electoral system weighted votes in proportion to wealth); it led to the establishment of unusually egalitarian societies in the postwar period.⁴ Nevertheless, both in Sweden and Norway we find an erosion of this electoral base, which begins in the 1970s and continues throughout the period 1990–2020. Less educated voters little by little withdrew their confidence from the social democrats, which by the end of the period were racking up their highest scores among the highly educated. To be sure, compared with what we see in the United States and France and to a lesser degree in the United Kingdom and Germany, the social-democratic vote did hold up better among the disadvantaged classes in Sweden and Norway. But the basic tendency, which has been under way now in all countries for more than half a century, is clearly the same.

The postelection surveys available in each country do not always allow us to go all the way back to the 1950s. The types of survey that were conducted and the state of the surviving records are such that in many cases we cannot begin systematic comparison until the 1960s or even the 1970s or 1980s. The sources we have gathered nevertheless allow us to conclude that the reversal of the electoral cleavage is an extremely general phenomenon in Western democracies. In nearly all the countries studied, we find that the vote profile of left-wing parties (labor, social-democratic, socialist, communist, radical, and so on, depending on the country) has reversed over the past half century. In the period 1950–1980, the profile was decreasing with educational level: the more educated the voter, the less likely he or she was to vote for the parties of the left. This gradually turned around in the period 1990–2020: the more highly educated the voter, the more likely to vote for the parties of the left (whose identity had clearly changed in the meantime)—increasingly so as time went by. We find this same evolution in countries as different as Italy, the Netherlands, and Switzerland as well as Australia, Canada, and New Zealand (Fig. 16.2).⁵ When the questionnaires and surveys allow, we also find results comparable with those obtained for France, the United States, and the United Kingdom regarding the evolution of the vote profile as a function of income and wealth.⁶

Within this general scheme, several countries exhibit distinct variations owing to their socioeconomic and political-ideological configurations. These specific trajectories deserve more detailed analysis, which would take us far beyond the scope of this book.⁷ I will say more later, however, about Italy, a textbook case of advanced decomposition of a postwar party system leading to the emergence of a social-nativist ideology.

FIG. 16.2.  Political cleavage and education, 1960–2020: Italy, the Netherlands, Switzerland, Canada, Australia, and New Zealand

Interpretation: In the period 1950–1970, the vote for Democrats in the United States and for various left parties in Europe (labor, social-democratic, socialist, communist, radical, and ecologist) was higher among less educated voters; in the period 2000–2020, it was higher among more educated voters. We find the same result in the United States and Europe, Canada, Australia, and New Zealand. Note: “1960–1969” includes elections between 1960 and 1969, “1970–1979” includes elections between 1970 and 1979, and so on. Sources and series: piketty.pse.ens.fr/ideology.

The only real exception to this general evolution of the cleavage structure of electoral democracy in the developed countries is Japan, which never developed a classist party system of the sort observed in the Western countries after World War II. In Japan, the Liberal Democratic Party (LDP) has held power almost continuously since 1945. Historically, this quasi-hegemonic conservative party has done best among rural farm voters and the urban bourgeoisie. By placing postwar reconstruction at the center of its program, it bridged the gap between the economic and industrial elite and traditional Japanese society. It was a complicated moment, marked by the US occupation and extreme anticommunism induced by the proximity of Russia and China. By contrast, the Democratic Party usually did best among modest-to-middling urban wage-earners together with more highly educated voters, who were eager to contest the US presence and the new moral and social order represented by the LDP. It was never able to constitute an alternative majority, however.8 More generally, the specific structure of Japanese political conflict should be seen in relation to long-standing cleavages around nationalism and traditional values.⁹

Rethinking the Collapse of the Left-Right Party System of the Postwar Era

To recapitulate: Compared with the very high concentration of income and wealth observed in the nineteenth century and until 1914, income and wealth inequality fell to historically low levels in the period 1950–1980. This was due in part to the shocks and devastation of the period 1914–1945, but a more important cause of change was the far-reaching critique to which the dominant proprietarian ideology of the nineteenth and early twentieth centuries was subjected. Between 1950 and 1980, new institutions and social and fiscal policies were created for the express purpose of reducing inequality: these included mixed public-private ownership, social insurance, progressive taxes, and so on.10 During this period, the political systems of all the Western democracies were structured around a “classist” conflict between left and right, and political debate centered on redistribution. The social-democratic parties (broadly understood to include the Democratic Party in the United States and various coalitions of social-democratic, labor, socialist, and communist parties in Europe) drew their support mainly from the socially disadvantaged classes, while the parties of the right and center-right (including the Republicans in the United States and various Christian-democratic, conservative, and conservative-liberal parties in Europe) drew their support mainly from the socially advantaged. This was true regardless of which dimension of social stratification—income, education, or wealth—one looked at. This classist structure of political conflict became so widespread in the postwar decades that many people came to believe that no other form of political organization was possible and that any deviation from this norm could only be a temporary anomaly. In reality, this classist left-right structure reflected a particular historical moment and was the product of specific socioeconomic and political-ideological conditions.

In all the countries studied, this left-right system has gradually broken down over the past half century. In some cases the names of the parties have remained the same, as in the United States, where the labels “Democrat” and “Republican” endure despite their countless metamorphoses. Elsewhere, parties have sometime updated their nomenclature, as in France and Italy in recent decades. But whether the names have changed or remained the same, the structure of political conflict in the Western democracies in 1990–2020 no longer has much to do with that of the period 1950–1980. In the postwar period, the electoral left was everywhere the workers’ party, but in recent decades it has become the party of the educated (the “Brahmin left”): the more educated the voter, the more likely to vote for the left. In all the countries studied, less educated voters have little by little ceased to vote for the parties of the left, leading to a complete reversal of the educational cleavage; the same voters have sharply reduced their participation in the political process.¹¹ When a divorce of such magnitude occurs in so many countries as the result of a long-term process extending over six decades, it is clear that something real is happening and that this cannot be attributed to some unfortunate misunderstanding.

The decomposition of the postwar left-right system and in particular the fact that the disadvantaged classes gradually withdrew their confidence from the parties to which they had given their support in the period 1950–1980 can be explained by the fact that those parties failed to adapt their ideologies and platforms to the new socioeconomic challenges of the past half century. Two of those challenges stand out: the expansion of education and the rise of a global economy. With the unprecedented growth of higher education, little by little the electoral left became the party of the highly educated (the “Brahmin left”) while the electoral right remained the party of the highly paid and wealthy (the “merchant right”), though less so over time. As a result, the social and fiscal policies of the two coalitions have converged. Furthermore, as commercial, financial, and cultural exchanges began to develop on a global scale, many countries experienced the pressure of heightened social and fiscal competition, which benefited those with the most educational capital on the one hand and the most financial capital on the other. Yet the social-democratic parties (in the broadest sense of the term) never really revised their redistributive thinking so as to transcend the limits of the nation-state and meet the challenges of the global economy. In a sense, they never responded to the critique Hannah Arendt proposed in 1951 when she observed that to regulate the unbridled forces of global capitalism, new forms of transnational politics would need to be developed.¹² Instead, the social-democratic parties contributed in the 1980s–1990s to liberalize the flow of capital everywhere without regulation, compulsory information sharing, or common fiscal policy (even on the European level).¹³

Other important factors explaining the decomposition of the postwar party system include the end of the old colonial empires; the growth of trade and increasing competition between the old industrial powers and the poor but developing countries where labor was cheap; and finally, the influx of immigrants from the former colonies. Taken together, these factors contributed in recent decades to the emergence of unprecedented identitarian and ethno-religious political cleavages, especially in Europe. New anti-immigrant parties arose on the right while existing parties (such as the Republicans in the United States, the Conservatives in the United Kingdom, and various traditional right-wing parties on the continent) took a tougher line on immigration-related issues. Two points should be noted, however. First, the decomposition of the classist left-right structure of the postwar period unfolded very gradually; it began in the 1960s, well before the immigration cleavage became salient in most Western countries (which generally did not happen until the 1980s and 1990s and in some cases even later). Second, when we look at the various Western countries, what stands out is that, over the past half century, the reversal of the educational cleavage proceeded nearly everywhere at the same pace and without apparent relation to the magnitude of cleavages over race or immigration (Figs. 16.1–16.2).

In other words, while it is clear that anti-immigrant parties (and anti-immigrant factions within older parties) have exploited identity cleavages in recent decades, it is just as clear that the reversal began for other reasons. A more satisfactory explanation is that the disadvantaged classes felt abandoned by the social-democratic parties (in the broadest sense) and that this sense of abandonment provided fertile ground for anti-immigrant rhetoric and nativist ideologies to take root. As long as the absence of redistributive ambition responsible for this sense of abandonment remains uncorrected, it is hard to see what might prevent that fertile ground from being exploited further.

Finally, an additional reason for the collapse of the left-right system of the postwar era is undoubtedly the fall of Soviet Communism and the ensuing shift in the political-ideological balance of power. For many years the mere existence of a communist countermodel put pressure on capitalist elites and political parties that had long been hostile to redistribution. But it also limited the redistributive ambitions of the social-democratic parties, which were de facto integrated into the anticommunist camp and therefore felt little incentive to seek an internationalist socialist alternative to capitalism and private ownership of the means of production. Indeed, the collapse of the communist countermodel in 1990–1991 convinced many political actors, especially among social democrats, that redistributive ambition had become superfluous. Markets, they now believed, were self-regulating, and the new goal of political action was to extend them as far as possible, both within Europe and around the world. The 1980s and 1990s were the crucial years when many key measures were decided, beginning with the complete liberalization of capital flows (without regulation). This effort was to a large extent led by social-democratic governments, and social-democratic parties remain unable even today to perceive alternatives to the situation they themselves created.

The Emergence of Social Nativism in Postcommunist Eastern Europe

The case of Eastern Europe clearly illustrates the role played by postcommunist disillusionment and the ideology of competitive markets in the collapse of the postwar party system (based on a clear left-right cleavage). In the transition to democracy following the collapse of communism in Eastern Europe, former ruling parties transformed themselves into social-democratic parties, in some cases fusing with newly emerged political movements and in others splintering and then recombining with them. Although much of the public remained hostile to the old parties (for understandable reasons related to their past errors), former state bureaucrats and managers of state enterprises affiliated with those parties often exercised important responsibilities in the early stages of the transition.

Consider, for example, the Democratic Left Alliance (SLD), which held power in Poland from 1993 to 1997 and again from 2001 to 2005. Eager to forget the communist past and join the European Union, the SLD adopted a platform that was social-democratic in name only. Its first priority was to privatize firms and open Polish markets to competition and investment from Western Europe, thus forcing the country to meet the criteria for EU admission as rapidly as possible. To attract capital, and in the absence of the slightest fiscal harmonization at the European level, a number of East European countries, including Poland, also set very low tax rates on corporate profits and high incomes in the 1990s and early 2000s.

Yet when the postcommunist transition was complete and Poland finally joined the European Union, the results did not always live up to expectations. Income inequality increased sharply, and large segments of the population felt left behind. German and French investments often generated large profits for shareholders while promised raises for wage-earners failed to materialize. This fostered strong resentment of the dominant powers within the EU, which were always quick to remind the Poles of the generous transfers of public funds they had received while forgetting that the reverse flow of private profits out of Poland (and other East European countries) significantly exceeded the influx of public transfers.14 In addition, since the 1990s, political life in Eastern Europe has been scarred by a large number of financial scandals, often linked to privatizations and involving individuals close to the party in power. Several corruption cases (such as the Rywin Affair in Poland in 2002–2004) revealed alleged links between the media and political and economic elites.¹⁵

Such was the deleterious climate in which the SLD went down to defeat in the 2005 Polish elections, in which the party received barely 10 percent of the vote and the “left” was virtually wiped off the political map. Since then, political conflict in Poland has revolved around the conservative liberals of the Civic Platform (PO) on one side and the conservative nationalists of the Law and Justice Party (PiS) on the other. It is striking to see how the electorates of these two parties have evolved along classist lines since the early 2000s. In the elections of 2007, 2011, and 2015, the conservative liberals of Civic Platform did best among highly paid and highly educated voters while the conservative nationalists of PiS appealed mainly to the less well paid and less educated. SLD voters occupy an intermediate position, although the party scarcely registers in the current political lineup.¹⁶ Their income is slightly below average, and their educational level slightly above, but the cleavage is smaller than with either the PO or PiS (Figs. 16.3–16.4).¹⁷

FIG. 16.3.  Political conflict and income in Poland, 2001–2015

Interpretation: Between 2001 and 2015, the PO (Civic Platform) (conservative liberal) vote shifted toward higher income groups, whereas the PiS (Law and Justice) (conservative nationalist) vote shifted toward lower-income voters. Sources and series: piketty.pse.ens.fr/ideology.

In Brussels, Paris, and Berlin, officials often worry that PiS seems hostile to the EU which PiS party leaders frequently accuse of treating Poland like a second-class partner. By contrast, Eurocrats love PO because it is always quick to endorse EU decisions and rules and religiously adheres to the principle of “free and undistorted competition.” PiS is rightly accused of championing authoritarian government and traditional values: for instance, it vehemently opposes abortion and same-sex marriage.18 Note, however, that since PiS came to power in 2015, it has also enacted fiscal and social measures to help people of low income, including a sharp increase of family benefits and pension hikes for the poorest retirees. By contrast, PO, in power from 2005 to 2015, generally preferred policies that favored the well-to-do. In short, PiS has been readier to flout budget rules than PO and more willing to spend on social programs.

FIG. 16.4.  Political conflict and education in Poland, 2001–2015

Interpretation: Between 2001 and 2015, the PO (Civic Platform) (conservative liberal) vote shifted toward more highly educated groups, while the PiS (Law and Justice) (conservative nationalist) vote shifted toward the less educated. Sources and series: piketty.pse.ens.fr/ideology.

Thus, PiS developed an ideology that one might characterize as “social-nationalist” or “social-nativist,” offering redistributive social and fiscal measures together with an intransigent defense of Polish national identity (deemed to be under threat from unpatriotic elites). The issue of immigration from outside Europe took on new importance in the wake of the refugee crisis of 2015, which provided PiS with an opportunity to voice its vigorous opposition to an EU plan to relocate immigrants throughout Europe (a plan that was quickly abandoned).19 Note, however, that the classist structure of the PO and PiS electorates was already in place before the issue of immigration really figured on the political agenda.

Unfortunately, it is not possible at this stage to systematically compare the evolution of the electoral cleavage structure in the various countries of Eastern Europe since the postcommunist transition of the 1990s owing to the inadequacy of the available postelection surveys.²⁰ Circumstances varied widely from country to country, and there was rapid turnover of political movements and ideologies. Social nativism has been spreading, with a mixture of strong hostility to non-European immigration (which has become the symbol of what the much-reviled Brussels elites would like to surreptitiously impose, despite the fact that the actual numbers of refugees are extremely small compared with the European population) and with a variety of social policies intended to show that the social-nativist parties care more about the lower and middle classes than do the pro-European parties.

The Hungarian case is in some ways similar to the Polish. Since 2010 the country has been governed by the conservative nationalist party Fidesz and its leader Victor Orban, who has without a doubt become one of the most prominent proponents of social-nativist ideology in Europe. Although officially a member of the European Peoples’ Party, the same parliamentary group to which the German Christian Democratic Union (CDU) and other “center-right” parties belong, Orban did not hesitate to plaster his country with provocative anti-refugee posters in 2015 along with giant billboards denouncing the alleged nefarious influence of George Soros, a Hungarian-born billionaire said to symbolize a conspiracy of globalized Jewish elites against the peoples of Europe. For the “social” component of its program, Fidesz (like PiS) increased family benefits, which for obvious reasons served as a symbol of social nativism.²¹ The Hungarian government also created subsidized jobs aimed at putting the unemployed back to work under the control of local governments and mayors loyal to the ruling party. Fidesz also sought to encourage Hungarian entrepreneurs and companies by offering government contracts in exchange for political fealty. These measures demonstrated Fidesz’s readiness to resist EU budget and competition rules, in contrast to its political rivals, especially the social democrats, who were regularly accused of marching to orders from Brussels.²²

It is worth recalling the circumstances in which Fidesz came to power in 2010. In 2006 (as in 2002), the social-democratic Hungarian Socialist Party, or MSZP (a direct descendant of the Hungarian Socialist Workers’ Party, or MSZMP, which had ruled from 1956 to 1989 and then been reorganized as the MSZP in 1990) won a narrow victory over a coalition led by Fidesz, whose popularity was rapidly increasing. The social-democratic leader—Ferenc Gyurcsany, who served as prime minister of Hungary from 2004 to 2009—was also an entrepreneur and one of the wealthiest men in the country, having amassed a fortune out of the privatizations of the 1990s. Shortly after his reelection in 2006, he gave a speech to party leaders that was supposed to remain confidential but somehow leaked to the media. In it, Gyurcsany candidly explained how he had lied for months to secure his victory, in particular concealing from voters planned budget cuts, which he regarded as inevitable. Social spending was to be slashed and the health care system overhauled. The news came as a bombshell; an unprecedented wave of demonstrations ensued. Orban saw the scandal as the long-awaited proof of the social democrats’ shameless hypocrisy and shrewdly exploited the chaos. For Fidesz, which began as a conservative nationalist movement, the situation provided a perfect opportunity to demonstrate that it was more sincerely devoted to the disadvantaged than the so-called social democrats, who had in reality become pro-market elitist liberals. Gyurcsany was ultimately forced to resign in 2009 in circumstances further complicated by the financial crisis of 2008 and the imposition of budgetary austerity throughout Europe. This sequence culminated in 2010 with the ultimate collapse of the social democrats and the triumph of Fidesz, which then easily won the legislative elections of 2014 and 2018.²³

The Emergence of Social Nativism: The Italian Case

It would be quite wrong to think of the development of social nativism as a specifically East European phenomenon without implications for Western Europe or the rest of the world. Eastern Europe should rather be seen as a laboratory where conditions were perfect for the combination of two ingredients that we also find in only slightly less extreme forms elsewhere. Together, these two factors give rise to social nativism: first, a strong sense of postcommunist and anti-universalist disillusionment leading to extreme identitarian withdrawal, and second, a global (or European) economy that prevents the establishment of coordinated, effective, and nonviolent policies of social redistribution and inequality reduction. In this light, it is particularly instructive to look at how a social-nativist coalition was formed in Italy after the legislative elections of 2018.

Compared with other Western democracies, Italy is distinctive in several ways. For one thing, the Italian postwar party system imploded in the wake of financial scandals unearthed by the anti-Mafia judges who conducted the Mani Pulite (Clean Hands) investigation in 1992. This led to the downfall of the two parties that had dominated Italian politics since 1945: the Christian Democrats and the Socialists. On the right of the political spectrum, the Christian Democrats were replaced in the 1990s by a complex and changing set of parties, including Silvio Berlusconi’s conservative-liberal Forza Italia and the Lega Nord (Northern League). The Lega was initially a regionalist antitax party, which advocated fiscal autonomy for “Padania” (northern Italy) and opposed transfer payments to the south, a region deemed to be lazy and corrupt. Since the refugee crisis of 2015, the Lega has become a nationalist anti-immigrant party devoted to ridding the country of foreigners. Lega leaders regularly denounce the alleged invasion of Italy by blacks and Muslims, who they claim threaten to take over the peninsula. The party appeals to anti-immigrant voters among the least favored classes, especially in the north, where it has also retained a base of antitax voters from the ranks of management and the self-employed.

On the left the situation is no less complicated. The collapse of the Socialist Party in 1992 and its ultimate dissolution in 1994 inaugurated a cycle of political realignment and renewal. The Italian Communist Party (PCI), long the most powerful in Europe along with its French counterpart, was hard-hit by the fall of the Soviet Union and chose in 1991 to transform itself into the Democratic Left Party (PDS). The PDS then joined other movements to create the Democratic Party (PD) in 2007, with the ambition of unifying “the left,” like the Democratic Party in the United States. In 2013, the party organized an internal election to choose its new leader. The winner, Matteo Renzi, became prime minister in 2014 and served in that post until 2016 as the head of a coalition led by the PD.

What happened on the left went beyond name changes. The left electorate (Socialist Party, or PS; PCI; PDS; PD) has been totally transformed in recent decades. In the 1960s and 1970s these parties racked up their highest scores among the disadvantaged classes, but that is no longer the case. In the 1980s and 1990s, the PS and PCI (and later the PDS) obtained their best results among the highly educated. This trend continued in the 2000s and 2010s. In the 2013 and 2018 elections, the vote for the PD was twenty points higher among the highly educated than among the rest of the population.24 The PD’s policies, especially the loosening of restrictions on layoffs (“Jobs Act”) enacted by Renzi soon after he took power, provoked strong opposition from the unions and huge demonstrations (a million people took to the streets of Rome in October 2014), which made the party even less popular among workers. Renzi’s reforms received strong public support from Christian Democratic German Chancellor Angela Merkel, and they would not have been approved by the Italian parliament had the PD not entered into a de facto coalition with Forza Italia. These developments convinced many Italian that the PD no longer had much to do with the postwar Socialist and Communist Parties that figured in its pedigree.

The latest arrival on the Italian political scene was the Movimento Cinque Stelle (or Five-Star Movement—M5S). Founded in 2009 by the humorist Beppe Grillo, M5S portrays itself an antisystem, antielitist party that cannot be placed on the usual left-right spectrum. One of its signature issues is a universal basic income. Compared with the PD, the M5S runs up its highest scores among less educated voters, among the disadvantaged classes in the south, and among the disillusioned of all the other parties, who are drawn to the movement’s promises to spend more on social measures and develop neglected regions. Within a few years, M5S had capitalized on discontent with the former governing parties, starting with Forza Italia and the PD, and had begun winning a quarter to a third of the vote in each new election.

In the legislative elections of 2018, the electorate divided into three large blocs: M5S got 33 percent of the vote, the PD 23 percent, and a coalition of right-wing parties 37 percent.²⁵ The right-wing coalition was quite heterogeneous because it included the anti-immigrant Lega (17 percent), the conservative-liberal Forza Italia (14 percent), and a number of smaller conservative nationalist parties (6 percent). Since no single bloc obtained a majority of the seats, a coalition was necessary to form a government. An M5S-PD alliance was briefly considered, but mutual suspicion made this impossible. M5S and the Lega, which had already joined forces to oppose Renzi’s Jobs Act both in parliament and in the streets in 2014, ultimately reached an agreement to govern the country on the basis of a synthesis of their two platforms, including both the guaranteed minimum income advocated by M5S and the uncompromising anti-refugee policy championed by the Lega.²⁶ The heterogeneous nature of this coalition agreement is reflected in the structure of the government: Luigi di Maio, the young leader of M5S, became the minister of economic development, labor, and social policy, which is responsible for overseeing the minimum income, territorial development, and public investment in the south while Matteo Salvini, the leader of the Lega, occupies the strategic post of minister of the interior, from which in the summer of 2018 he launched several spectacular anti-immigration operations, including closing all Italian ports to humanitarian ships engaged in rescuing refugees adrift on the Mediterranean.

The M5S-Lega coalition that has ruled Italy since 2018 is clearly a social-nativist alliance; it naturally calls to mind the PiS government and Poland and the Fidesz government in Hungary. Of course, there is no guarantee that this coalition will survive. Either of the two pillars on which it stands could collapse, bringing down the whole edifice. Relations between the two governing parties are very tense, and all indications are that the nativists are on the verge of emerging as the dominant partner in the coalition. Salvini’s sallies against refugees have made him increasingly popular and may allow the Lega to outstrip M5S in the next elections or even to win an absolute majority. In any case, the mere fact that such a social-nativist coalition could come to power in an old West European democracy like Italy (the third-largest economy in the Eurozone) shows that the phenomenon is not limited to postcommunist Eastern Europe. Social-nativist leaders in several countries, including Orban and Salvini, have not been shy about publicizing their shared antielitist attitudes and common view of Europe’s future in both immigration and social policy.²⁷

On the Social-Nativist Trap and European Disillusionment

It is natural to ask whether a similar political-ideological coalition could emerge in other countries, especially France. This would have significant consequences for the political equilibrium of the European Union. When we look at the distribution of votes in the 2018 Italian elections, we find three voting blocs (or four, if we distinguish between the two components of the right bloc, the Lega and Forza Italia, which split over the question of alliance with M5S). This configuration of Italy’s ideological space has some important points in common with the four-part division we saw in the 2017 French presidential election as well as some significant differences.28 In the French context, the closest equivalent to the M5S-Lega alliance would be a hypothetical alliance between the radical left, La France Insoumise (LFI), and the Front National (which in 2018 renamed itself the Rassemblement National, or RN). At this stage, an LFI-RN alliance seems out of the question, however. The RN electorate includes voters most fiercely opposed to immigration while the LFI electorate (on the evidence of 2017) includes the voters most favorable to immigration.²⁹ The social and redistributive policies favored by LFI voters and party leaders, such as progressive taxation, are directly descended from the historical policies of the socialist and communist left. The RN draws on a very different ideological corpus, which makes it hard to imagine any way that the two parties could agree on a common plan of action, at least in the near future. Despite many attempts to win respectability and bury its historic origins (in Vichy, colonialism, and Poujadism), even to the point of changing its name, the RN remains the heir of a movement that the vast majority of LFI voters regard as untouchable.³⁰

Nevertheless, the rapidity of change in Italy suggests a need for caution in anticipating what trajectories might be possible in the medium term. Several things made the social-nativist alliance of 2018 possible in Italy. One was the damage done by the collapse of the postwar party system in 1992. As the integrity of all the postwar parties was called into question, people lost faith in the old faces and promises to the point where they could no longer find their political bearings. Ideologies that had once seemed solid shattered into a thousand pieces, and previously unthinkable alliances became acceptable.³¹

One reason why the social-nativist cocktail became thinkable in Italy has to do with specific features of the Italian immigration controversy. Because of its geographical situation, Italy became the destination of choice for large numbers of refugees fleeing Syria and Africa via Libya.³² The other countries of Europe, always prompt to lecture the rest of the world—including Italy—on the need for generosity, mostly refused to consider any plan to apportion responsibility for the refugees in a rational and humane way. France showed itself to be particularly hypocritical on this score: French border police were ordered to turn back any immigrants attempting to cross from Italy. Since 2015, France has admitted only one-tenth as many refugees as Germany.³³ In the fall of 2018, the French government decided to close its ports to the humanitarian vessels turned away by Italy, and it went so far as to refuse to allow the Aquarius to sail under a French flag, condemning the ship chartered by the humanitarian organization SOS Méditerranée to remain tied up in port while refugees drowned at sea. Salvini had a field day attacking the attitude of France and particularly its young president Emmanuel Macron, elected in 2017, who in Salvini’s eyes was the very embodiment of Europe’s hypocritical elite. French hypocrisy thus became his justification for cracking down on immigrants in Italy.

The charge of hypocrisy is of course one of the classic rhetorical devices of the anti-immigrant right. The Front National and other parties of its ilk have always denounced the self-righteousness of elites quick to defend open borders as long as they do not have to bear the consequences.³⁴ But rhetoric of this type (pioneered in France by Jean-Marie Le Pen in the 1980s) is usually convincing only to those who already believe because it is clear that those who use it are interested mostly in stirring up hatred as a stepping stone to power for themselves. In Salvini’s case, in the context of a Europe-wide conflict over immigration with a particularly bitter clash between France and Italy, the charge of hypocrisy has acquired a certain plausibility. The specific nature of this conflict is part of the reason for the Lega’s growing popularity in Italy. It also helps to explain why the M5S, although relatively moderate on the refugee issue, could agree to a coalition with the Lega: the tough anti-immigrant line could be presented as part of a broader attack on elite hypocrisy.

Last but perhaps not least, the social-nativist coalition in Italy is fueled by a widespread distaste for European rules and in particular European budgetary rules, which allegedly prevented Italy from investing and from recovering from the 2008 crisis and the purge that followed. Indeed, it is hard to deny that the European decision, pushed by Germany and France in 2011–2012, to impose deficit reduction throughout the Eurozone led to a disastrous “double-dip” recession and a sharp spike in unemployment, especially in the south.³⁵ It is also clear that Franco-German conservatism on the issue of pooling public debt and establishing a common interest rate at the European level—a policy change that would be consistent with having a common concurrency and would protect the countries of the south from speculation in the financial markets—is due largely to the fact that France and Germany would prefer to continue enjoying the benefits of near-zero interest rates by themselves, even if it means leaving the European project at the mercy of the markets in any future financial crisis.

Of course, the alternatives proposed by the Lega and M5S are far from perfect or well thought out. Some in the Lega seem to be contemplating an exit from the euro and return to the lira, which would allow for debt reduction through moderate inflation. The majority of Italians worry about the unpredictable consequences of such a move, however. Most leaders and voters of the Lega and M5S would prefer a change of Eurozone rules and of the policy stance of the European Central Bank (ECB). If the ECB could print trillions of euros to save the banks, people ask, why can’t it help Italy by deferring its debt until better times return? I will say more later about these complex and unprecedented debates, which remain underdeveloped. What is certain is that answers to these questions cannot be postponed indefinitely. Social discontent with the EU and deep incomprehension of the authorities’ inability to muster the same energy and deploy the same resources to help large numbers of people as they did to save the financial sector will not magically disappear.

The Italian case also shows that the sense of disillusionment with Europe, which the Lega shares with M5S, can serve as a powerful bond for a social-nativist coalition. What makes the Lega and its leader Matteo Salvini so dangerous is precisely Salvini’s ability to combine nativist rhetoric with social rhetoric—attacks on immigrants with attacks on speculators and financiers—and to wrap all of it up in a critique of hypocritical elites. A similar formula could be used to build social-nativist coalitions in other countries, including France, where disillusionment with Europe runs high among supporters of both the far left and the far right. The fact that Europe is so often instrumentalized for the pursuit of antisocial policies, as was clear in the sequence of events leading up to the Yellow Vest crisis of 2017–2019 (which followed abolition of the wealth tax in the name of European competition, financed by a carbon tax that fell heavily on the poorer half of the population), unfortunately makes such an evolution plausible. Indeed, if the nativist party opportunistically tones down its anti-immigrant rhetoric and concentrates instead on social issues and resistance to the European Union, it is not out of the question that we could someday see a social-nativist coalition similar to the Lega-M5S coalition in Italy coming to power in France.

The Democratic Party: A Case of Successful Social Nativism?

Some readers, even among those generally hostile to anti-immigrant politics, might nevertheless be tempted to welcome social-nativist movements in Europe. After all, wasn’t the Democratic Party that backed the New Deal in the United States in the 1930s and ultimately supported the civil rights movement in the 1960s and elected a black president in 2008 originally an authentic social-nativist party? Having supported slavery and contemplated sending slaves back to Africa, the Democratic Party reconstructed itself after the Civil War around a social-differentialist ideology, combining very strict segregation in the South with a relatively egalitarian social policy for whites (especially white Italian and Irish immigrants and indeed the white working class generally). In any case, whatever its shortcomings, Democratic social policy was certainly more egalitarian than Republican social policy. Yet it was not until the 1940s that the Democratic Party tried to do something about the segregationist element within it, which it finally purged in the 1960s under pressure from the civil rights movement.

With this example in mind, one might imagine a trajectory in which PiS, Fidesz, the Lega, and the Rassemblement National follow a similar course in the coming decades, offering relatively egalitarian social measures to “native Europeans” combined with a very harsh crackdown on non-European immigrants and their children. Later, perhaps half a century or more in the future, the nativist component would fade away or perhaps even transform itself to the point of embracing diversity once conditions were right. There are several problems with this idea, however. First, before becoming the party of the New Deal and civil rights, the Democratic Party did a great deal of damage. From the 1870s to the 1960s, Democrats in the South imposed segregation on blacks, kept black children from attending the same schools as whites, and supported or covered up lynchings organized by the Ku Klux Klan and similar vigilante groups. It makes no sense to suggest that there was no other path to the New Deal and the Civil Rights Act. There are always alternatives. Everything depends on the ability of political actors to mobilize and search for them.36

In the current European context, the potential damage if social nativists were to take power would likely be of the same order. Indeed, the damage has already begun where social nativists currently hold power: not only have they cracked down on immigrants in their own countries, but they have also pressured timorous governments elsewhere in Europe to enact more restrictive immigration policies. Meanwhile, thousands of migrants die in the Mediterranean and hundreds of thousands molder in camps in Libya and Turkey. If the social-nativist parties were free to do as they wished, they might very well turn to more violent attacks on non-European immigrants and their descendants living in Europe, retroactively stripping them of nationality and deporting them, as purportedly democratic regimes have done in the past in both Europe and the United States.³⁷

What is more, there are serious reasons to doubt that today’s social-nativist movements are capable of enacting genuinely redistributive policies. In the late nineteenth and early twentieth centuries, the Democratic Party in the United States helped develop tools for social redistribution, including the federal income and inheritance taxes enacted in 1913–1916 and social insurance (pensions and unemployment) and minimum wage programs in the 1930s—and remember that under Democratic leadership the United States led the way in progressive taxation, raising top marginal rates to the highest levels ever seen anywhere in the period 1930–1980.³⁸ Contrast this record with the rhetoric and accomplishments of PiS in Poland, Fidesz in Hungary, and the M5S-Lega alliance in Italy. It is striking to see that none of these parties has proposed any explicit tax increase on the wealthy, even though they sorely need the revenue to finance their social policies. True, PiS did reduce certain tax deductions beneficial to people with high incomes so that they ended up paying somewhat more than before, but the Polish government still has not dared to raise tax rates on the wealthy.³⁹

Interstate Competition and the Rise of Market-Nativist Ideology

In Italy, it is noteworthy and revealing that M5S agreed to include the Lega’s campaign proposal for a “flat tax” (a legacy of the Lega’s origins as an antitax party) in its coalition agreement with Salvini’s party. If this measure were fully implemented, it would mean that every taxpayer, no matter how high his or her income, would pay the same flat rate—thus completely dismantling the progressive tax system. This would result in an enormous loss of tax revenue, the benefits of which would go to people of middle-to-high income but which would be paid for by increased public borrowing, on the model of Ronald Reagan’s tax reforms of the 1980s. Because this would pose a serious problem for a heavily indebted country like Italy, this part of the coalition’s reform program has been postponed and will no doubt be enacted only in some very limited form with a reduction of top marginal rates rather than complete elimination of progressive taxation. Nevertheless, the fact that M5S could have agreed to such a proposal says a great deal about the movement’s lack of ideological backbone. It is hard to see how one can possibly finance an ambitious basic income proposal and a vast program of public investment while eliminating progressive taxes on top incomes.

Why do today’s social nativists lack appetite for progressive taxation? There are several possible explanations. It may be that they do not want to be associated with the legacy of the social-democratic, socialist, Labour (United Kingdom), or New Deal left. M5S has embraced the idea of a universal basic income, which it sees as innovative and modern, but rejects the progressive tax system that could finance it, which it finds complicated and tired. Another point that bears emphasizing is the degree to which the ECB’s policy of massive monetary creation since 2008 has changed people’s perceptions. Because the ECB created trillions of euros to save the banks, it is difficult for social nativists to admit that complex and potentially unjust and evadable new taxes are needed to pay for a universal basic income or new investment in the real economy. One finds repeated references to the need for just monetary creation in the rhetoric of M5S, the Lega, and other social-nativist movements. Until European governments propose a more convincing means of mobilizing resources, such as a European tax on the wealthy, the idea of paying for social expenditures by contracting new debt and creating new money will continue to attract strong support among social-nativist voters.

The lack of appetite for progressive taxes is also a consequence of several decades of antitax propaganda and sanctification of the principle of competition of all against all. Today’s hypercapitalist economy is one of heightened interstate competition. Competition to attract high earners and wealthy capitalists already existed in the late nineteenth and early twentieth centuries. But it was less intense than competition today. This was partly because the means of transportation and information technology were different back then. More importantly, the international treaties that have defined the global economy since the 1980s have ensured that the new technology would be used to protect the legal and fiscal privileges of the rich rather than the majority. That technology could be used instead to create a public financial register, which would allow countries that so desire to impose redistributive taxes on transnational wealth and the income it generates. Such a system is not only possible but also desirable: it would replace existing treaties, which allow the capital to circulate freely, with new treaties that would create a regulated system built on the public financial register.40 But this would require substantial international cooperation and ambitious efforts to transcend the nation-state, especially on the part of smaller countries (such as the nations of Europe). Nativist and nationalist parties are by their very nature not well equipped to achieve this kind of cross-border cooperation.

Therefore, it seems quite unlikely that today’s social-nativist movements will develop ambitious plans for progressive taxation and social redistribution. The most probable outcome is that once they arrive in power, they will find themselves (whether they like it or not) caught up in the mechanism of fiscal and social competition and thus be forced to do whatever it takes to promote their national economies. Only for opportunistic reasons did the Rassemblement National in France oppose abolition of the wealth tax during the Yellow Vest crisis. If the RN were to come to power, it would likely cut taxes on the rich to attract new investment, not only because such a course would be in keeping with its old antitax instincts and its ideology of national competition but also because its hostility to international cooperation and a federal Europe would force it to engage in fiscal dumping. More generally, the disintegration of the EU (or just the reinforcement of state power and anti-migrant ideology within the EU) to which the accession of nationalist parties to power could lead would intensify social and fiscal competition, increase inequality, and encourage identitarian retreat.⁴¹

On Market-Nativist Ideology and Its Diffusion

In other words, social nativism is highly likely to lead in practice to a market-nativist type of ideology. In the United States, Donald Trump has clearly gone this route. In the 2016 presidential campaign, Trump tried to give his politics a social dimension by portraying himself as the champion of the American worker, whom he described as the victim of unfair competition from Mexico and China and as citizens abandoned by Democratic elites. But the actual policies of the Trump administration have combined more or less standard nativist measures (such as reducing the influx of immigrants, building a border wall, and supporting Brexit and nativist governments in Europe) with tax cuts for the rich and multinational corporations. Reagan’s 1986 Tax Reform Act featured a reduction of the top marginal income tax rate (to 28 percent, later raised to 35–40 percent under George H. W. Bush and Bill Clinton but never restored to previous levels). The tax reform that Trump negotiated with Congress in 2017 pushed this logic even further by focusing cuts on corporations and “entrepreneurs.” The federal corporate tax rate, which had been 35 percent since 1993, was abruptly cut to 21 percent in 2018, with an amnesty on profits repatriated from abroad. This reduced corporate tax receipts by half and very likely triggered a global race to the bottom on corporate taxes, an essential component of public finance.42 On top of that, Trump obtained an additional tax reduction focused specifically on self-employed entrepreneurs (like himself), whose business income will henceforth be taxed at a maximum rate of 29.6 percent, compared with 37 percent on top salaries. The combined impact of these two measures is that the rate at which the wealthiest 0.01 percent of taxpayers (including the 400 richest people in the country) are taxed has for the first time fallen below the rate assessed on people lower down in the top centile or even the top thousandth; top rates are sinking closer and closer to the effective rate paid by the poorest 50 percent.⁴³ Trump also sought complete elimination of the progressive tax on inheritances, but Congress refused to go along with him on that point.

It is particularly striking to note the similarity between the tax reforms enacted by presidents Trump and Macron in 2017. In France, in addition to elimination of the wealth tax (ISF) discussed previously, the new government passed a gradual reduction of the corporate tax from 33 to 25 percent and also cut the tax on dividend and interest income to 30 percent (compared with the 55 percent rate on the highest salaries). The fact that a purportedly nativist government like Trump’s adopted a tax policy similar to that of a supposedly more internationalist government like Macron’s shows that political ideologies and practices have converged to a considerable degree. The rhetoric varies: Trump praises “job creators” while Macron prefers to speak of the “climbers at the head of the rope” (“premiers de cordée”). Ultimately, however, both adhere to an ideology according to which the competition of all against all requires offering ever greater tax cuts to the most mobile taxpayers while the masses are exhorted to honor their new benefactors, who bring innovations and prosperity (while omitting to mention that none of this would exist without public support for education and basic research and private appropriation of public knowledge).

Meanwhile, both the French and US governments risk increasing inequality and contributing to the feeling of the lower and middle classes that they have been left to face the consequences of globalization on their own. Trump tries to win them over by claiming that he is doing a better job than the Democrats of stopping immigration and is far more vigilant when it comes to opposing unfair competition from abroad.⁴⁴ He cleverly portrays “job creators” as more useful than the Democrats’ intellectual elites when it comes to winning the global economic war the United States is waging against the rest of the planet.⁴⁵ Trump regularly denounces intellectuals as condescending and hectoring, always ready to follow the latest cultural fad no matter how threatening to American values and society. In particular, he loves to denounce the newfound passion for the climate: the idea of climate change is “a hoax,” he says, invented by scientists, Democrats, and foreigners jealous of American prosperity and greatness.⁴⁶ Anti-intellectual sentiments have also been mobilized by nativist governments in Europe and India, illustrating the crucial need for more education and for citizen appropriation of scientific knowledge.⁴⁷

The French president has made the opposite wager. He hopes to hold on to power by branding his opponents as nativists and antiglobalists, betting that a majority of the French believe in tolerance and openness and will therefore vote against the social nativists when the moment of truth arrives (in any case, by then the social nativists will have turned into market nativists a la Trump). At bottom, both ideologies insist that there is no alternative to tax cuts benefiting the rich and that the progressive-nativist cleavage is the only remaining axis along which political conflict can occur.⁴⁸ Both are based on misleading simplifications and a healthy dose of hypocrisy. Indeed, it is still possible for individual countries to pursue ambitious programs of redistribution, even small countries like those found in Europe.⁴⁹ If even small states can redistribute, the federal government in the United States has all the power it needs to enforce its fiscal policies—provided it can muster the necessary political will.⁵⁰ Furthermore, there is nothing to prevent efforts to develop greater international cooperation, especially on tax issues, for the purpose of achieving more equitable and durable economic growth.

On the Possibility of Social Federalism in Europe

The most natural way to escape the social-nativist trap would be to develop social federalism in one form or another. International cooperation and political integration can achieve social justice and redistribute wealth by democratic means. Unfortunately, such a harmonious and nonviolent reform of European institutions is not the most likely outcome. It is probably more realistic to prepare for somewhat chaotic changes ahead: political, social, and financial crises could tear the European Union apart or destroy the Eurozone. Whatever lies ahead, reform is essential. No one envisions a return to autarky, and new treaties will therefore be necessary and, if possible, more satisfactory than the existing ones. Here I will focus on the possibility of social federalism in the European context. The lessons are of more general import, however, partly because European social and fiscal policies can have an important impact on other parts of the world and partly because similar forms of transnational cooperation may be applicable to other regions (such as Africa, Latin America, or the Middle East) as well as to relations between regional organizations.

The European Union is a novel and sophisticated attempt to organize an “ever closer union among the peoples of Europe.” In practice, however, European institutions, established in stages from the Treaty of Rome (1957), which constituted the European Economic Community (EEC), to the Maastricht Treaty (1992) establishing the EU and the Lisbon Treaty (2007), which set current EU rules, have mainly sought to organize a vast market and guarantee free circulation of goods, capital, and workers but not to arrive at a common social or fiscal policy. Recall the basic principles on which these institutions operate.51 Broadly speaking, in order for decisions taken by the European Union, whether regulations, directives, or other legislative acts, to take effect, they must be approved by the two institutions that share legislative power: first, the European Council, which consists of heads of state and government (and which also meets, as the Council of the European Union, at the ministerial level depending on the issue under discussion, so there can be a council of ministers of finance, ministers of agriculture, and so on); and second, the European Parliament, which since 1979 has been elected by universal suffrage and represents the member states on the basis of population (with overrepresentation of smaller states).⁵² Decisions are prepared and promulgated by the European Commission, which acts as a kind of executive body and European government with a president of the Commission as its head and Commissioners in charge of various domains appointed by the Council, which consists of heads of state and government, and then approved by the Parliament.

Formally, the setup resembles a classic federal parliamentary structure with an executive and two legislative chambers. Two particular features make the EU arrangement very different, however. One is the key role played by the unanimity rule and the other is the fact that the council of ministers is totally unsuited for parliamentary deliberation of a pluralistic, democratic kind.

Recall first that most important decisions require a unanimous vote of the council of ministers. In particular, unanimity is required in all matters relating to taxes, the EU budget, and systems of social protection.⁵³ As for regulation of the internal market, free circulation of goods, capital, and people, and trade agreements with the rest of the world, which are ultimately the matters at the heart of the European project, the rule of “qualified majority voting” applies.⁵⁴ But once any matter touching on common fiscal, budgetary, or social policy—and especially anything touching on taxation or the public finances of the member states—is at issue, the unanimity rule applies. Concretely, this means that every country has veto power. For instance, if Luxembourg, with a population of half a million or barely a tenth of a percent of the total EU population of 510 million, wants to tax corporate profits at zero percent at the expense of its neighbors, nobody can stop it from doing so. Any country, no matter how small—be it Luxembourg, Ireland, Malta, or Cyprus—can block any tax measure that comes up. Moreover, since the treaties guarantee free circulation of capital with no obligation of fiscal cooperation, conditions are ripe for a race to the bottom. Fiscal dumping is the result, and the beneficiaries are those whose capital is most mobile.

Furthermore, the absence of any common tax or budget is what makes the European Union more of a commercial union or international organization than a true federal government. In the United States or India, the central government also has a bicameral legislature, but it has the power to levy taxes for collective projects. In both cases, federal income, inheritance, and corporate taxes bring in revenues of about 20 percent of gross domestic product (GDP), compared with barely 1 percent for the European Union, which in the absence of any common tax system depends on contributions by member states set by unanimous agreement.

On the Construction of a Transnational Democratic Space

Can this be changed? One possibility would be to allow fiscal and budget issues to be decided by qualified majority vote. Leave aside the fact it would not be easy to persuade the small countries to give up their fiscal veto. It would probably take a coalition of countries exerting very strong pressure on the others and threatening them with significant sanctions. In any case, even if one succeeds in imposing the qualified majority rule on all twenty-eight member states (soon to be twenty-seven if the United Kingdom goes through with Brexit, which as of this writing remains uncertain) or if a smaller group of countries agrees to forge ahead on some other basis, the problem remains that the council of finance ministers (or heads of government) is totally unsuited to the task of developing a true European parliamentary democracy.

The reason is simple: the council is a body consisting of one representative per country. As such, it is designed to pit the (perceived) national interests of member states against one another. In no way does it allow for pluralist deliberation or the construction of a majority based on ideas rather than interests. In the Eurogroup,55 the German finance minister alone represents 83 million citizens, the French finance minister 67 million, the Greek finance minister 11 million, and so on. Under these conditions, it is simply impossible to deliberate tranquilly. The representatives of the large countries cannot allow themselves to be publicly placed in the minority on a fiscal or budgetary matter of importance to the home country. As a result, decisions of the Eurogroup (or of any of the European bodies consisting of ministers or heads of state and government) are almost always unanimous, taken under the cover of consensus following deliberation behind closed doors. In these bodies, none of the usual rules of parliamentary debate apply. For example, there are no procedural rules governing amendments, speaking time, or the manner of voting. It makes no sense that such bodies can decide tax policy for hundreds of millions of people. Since at least the eighteenth century and the age of Atlantic Revolutions we have known that the power to levy taxes is the quintessential parliamentary power. Setting tax rules, deciding who and what can be taxed and how much, requires free and open public debate under the watchful eye of citizens and journalists. All shades of opinion in every country need to be fully represented. By its very nature, a council of finance ministers cannot satisfy these requirements.⁵⁶ To recapitulate: European institutions, in which ministerial councils currently play the central and dominant role, relegating the European Parliament to a supporting role, were designed to regulate the broad market and conclude intergovernmental agreements; they were not designed to make fiscal and social policy.

A second possibility, which is widely supported by European leaders who favor a federal system, would be to transfer all power to approve new taxes to the European Parliament. Elected by direct universal suffrage, subject to the usual rules of parliamentary debate, and taking decisions by majority vote, the European Parliament is clearly better suited than the ministerial councils to deliberate on new taxes and budgets. Although this is clearly a better option than the first, several difficulties remain. We need to consider all the implications and understand why this option is unlikely to succeed. First, note that one essential step to creating a viable European democracy is to completely rewrite the rules governing the lobbies that currently loom so large in Brussels politicking, whose lack of transparency raises serious problems.⁵⁷ Second, transferring fiscal sovereignty to the European Parliament would mean that the political institutions of the member states would not be directly represented in the vote on European taxes.⁵⁸ This would not necessarily be problematic, and such bypassing of national institutions already happens in other contexts, but the point nevertheless calls for careful consideration.

In the United States, the federal budget and taxes, like other federal laws, must be approved by the Congress, whose members are elected for the purpose and do not directly represent the political institutions of the individual states. Bills must be approved in identical terms by both the House of Representatives and the Senate. Seats in the House are apportioned according to the population of each state while two senators from each state (regardless of size) sit in the Senate. This system, in which neither chamber takes priority over the other, is not a model of its kind and frequently leads to deadlock. But it does function, more or less, perhaps because there exists a certain equilibrium among states of different sizes.⁵⁹ In India there are also two chambers: the Lok Sabha, or House of the People, directly elected by citizens with districts carefully drawn to ensure proportional representation of the population throughout the country and the Rajya Sabha, or Council of States, whose members are elected indirectly by the legislatures of the states and territories of the Indian Union.⁶⁰ Laws must in principle be approved in identical terms by both chambers, but in case of disagreement it is possible to convene a joint session to agree on a final text, which in practice gives a clear advantage to the Lok Sabha owing to its numerical superiority.⁶¹ In addition, when it comes to fiscal and budgetary measures (“money bills”), the Lok Sabha automatically has the last word.

Nothing stands in the way of imagining a similar solution for Europe: the European Parliament could have the last word on European taxes and a European budget financed by those taxes. There are, however, two key differences that render such a solution unsatisfactory. First, it is unlikely that the EU’s twenty-eight member states would agree to delegate fiscal sovereignty, at least initially. Therefore, those states that wish to forge ahead would need to be allowed to constitute a subchamber of the European Parliament. This could happen, but it would mean a fairly sharp break with the remaining member states. Second, and more important, assuming that all twenty-eight countries are in agreement or that some subgroup is prepared to forge ahead, one key difference remains between the European Union and the United States or India: the nation-states of Europe existed as such before the EU. In particular, each member state is free via its own national parliament to ratify or reject international treaties. In addition, these national parliaments—be it the Bundestag in Germany, the Assemblée Nationale in France, or any of the others—have been voting for decades (in some cases since the nineteenth century) on taxes and budgets; over the years, these have grown to considerable proportions, on the order of 30–40 percent of GDP.

With the taxes approved by these national parliaments, Europe’s nation-states were able to implement novel social and educational polices, pioneering an immensely successful new model of development. They achieved the highest standard of living ever attained while limiting inequality (at least compared with the United States and other parts of the world) and providing relatively equal access to health and education. These national parliaments will continue to exist and continue to levy taxes and approve budgets. No one believes that all decisions should be made in Brussels or that EU spending should jump overnight from 1 to 40 percent of GDP, supplanting all national, regional, and local budgets and social insurance programs. Just as the property regime should be decentralized and participatory, the political regime should also be as decentralized as possible and involve actors on all levels.

Building European Parliamentary Sovereignty on National Parliamentary Sovereignty

For these reasons, if one wants to construct a truly transnational democratic space appropriate to Europe as currently constituted, one had better allow some role for national parliaments. One possibility might be to create a European Assembly (EA) composed partly of representatives of participating national parliaments and partly of members of the European Parliament (MEPs). Each participating country would be represented in proportion to its population, and each political party would be represented in its national delegation in proportion to its representation in the home-country parliament or the European Parliament, as the case may be. These questions of apportionment are too complex to be settled here. One proposal that has emerged as a working hypothesis from recent discussions is that the EA should consist of 80 percent members of national parliaments and 20 percent MEPs.62

The advantage of this proposal, which is based on a draft Treaty on the Democratization of Europe (T-Dem),⁶³ is that it can be adopted by countries that wish to do so without modifying existing European treaties. Although it would be best if it were adopted by as many countries as possible—especially Germany, France, Italy, and Spain (which by themselves account for 70 percent of the population and GDP of the Eurozone)—there is nothing to prevent a smaller subset of countries from moving forward to form, say, a Franco-German Assembly or a Franco-Italo-Belgian Assembly.⁶⁴ In any case, this EA would be granted the power to approve four important common taxes: a tax on corporate profits, a tax on high incomes, a tax on large fortunes, and a carbon tax. To complement these taxes, the Assembly would also vote on a common budget. Assuming tax revenues of, say, 4 percent of GDP, the money could then be apportioned as follows: half would revert to the member states for their own use (for instance, to reduce taxes on the lower and middle classes, which have thus far borne the brunt of European fiscal competition) while the other half would finance research, education, and the transition to renewable sources of energy, with an additional portion set aside to defray the cost of welcoming new immigrants.⁶⁵ These suggestions are merely meant to be illustrative; obviously it would be up to the EA to set its own priorities and levy taxes accordingly.⁶⁶

The key point is to create a European space for democratic deliberation and decision making in which it would be possible to adopt strong measures of fiscal, social, and environmental justice. As we saw in analyzing the structure of the French and British referendum votes in 1992, 2005, and 2016, the breach between Europe and the disadvantaged classes has grown to significant proportions.⁶⁷ Without concrete, visible measures to demonstrate that the European project can be made to serve the goal of greater fiscal and social justice, it is hard to see how this can change.

The T-Dem proposal does not depend on any particular composition of the EA: for example, 50 percent of its members could be drawn from national parliaments rather than 80 percent. Deciding on the precise makeup of the EA will require broad debate and deep reflection. Technically, the system proposed here could even work with zero percent national deputies, in which case the EA would simply be a subset of the current European Parliament (including only MEPs from countries willing to proceed with this plan). If enough countries agreed to move in this direction and entrust fiscal sovereignty to a subset of the European Parliament, then it would represent a significant improvement over the status quo. In my view, however, reducing the proportion of national deputies too much (below 50 percent, say) would entail significant risks. The most obvious of these is that if a national parliament strongly disagreed with the fiscal and social policies adopted by the EA, it could always decide to withdraw from the project and repudiate the authorizing treaty. Since no one denies that national parliaments retain sovereignty in ratifying (and therefore repudiating) international treaties—this is one of their most important powers—it seems strange to deny them the right to participate in voting on European taxes.⁶⁸

More importantly, if national parliaments were deeply involved in the composition of the EA, this would have the effect of transforming national legislative elections into European elections. If national deputies were strongly represented in the EA, it would be impossible for parties and candidates standing in national elections to keep on blaming Brussels for everything that goes wrong while claiming that they have nothing to do with EU institutions (the favorite sport, unfortunately, of many European political leaders). If a subset of national deputies were to represent their party in the EA, they would have to explain in their national campaigns what European-level policies they intended to support (including taxes, budgets, and amounts to be returned to the national treasury).⁶⁹ National political life would thus be profoundly Europeanized. For this reason, I think that the project of building European parliamentary sovereignty on top of national sovereignty is ultimately a more ambitious form of federalism than the alternative project, which is to construct a European Parliament entirely independent of national parliaments.⁷⁰ Above all, this novel way of constructing transnational parliamentary sovereignty seems better adapted to European political and historical realities: Europe is quite different in this respect from other countries with federal systems (such as the United States, India, Brazil, Canada, and Germany, among others). Hence a new approach is required.⁷¹

Rebuilding Trust and Developing Common Norms of Justice

The T-Dem proposal would also place a strict ceiling on transfer payments between states signing the agreement. The purpose of this provision is not only to facilitate acceptance of the proposal but also to signal that the main objective is to reduce inequality within countries. This might seem like a technical point, or even a blemish, but in view of the climate of distrust that currently prevails in Europe, it is doubtless the only way of making progress.

Under the current EU budget framework, the European Commission annually publishes each country’s “budget balance”—that is, the difference between what it contributes to the total EU budget (currently around 1 percent of GDP) and the amount it receives in return. In the period 1998–2018, the largest net contributors were Germany, France, and the United Kingdom, with net contributions of 0.2–0.3 percent of GDP, depending on the year.72 The issue of transfers to the European Union played a significant role in the Brexit campaign.⁷³ The new budget envisioned under the current T-Dem proposal (4 percent of GDP or more) would supplement the current EU budget for the states signing the agreement. To reduce the risk of rejection, the proposal envisions that the difference between monies received from the signatory states and monies returned to those same states under the supplementary budget should not exceed 0.1 percent of GDP.⁷⁴ Of course, this figure could be raised or lowered if the signatories agree without altering the substance of the proposal.

This is a crucial point because the fantasy of a “transfer union” has become a major impediment to fresh thinking about the EU. Since the crisis of 2008, German political leaders in particular have been quick to decry any hint of a Transferunion. Members of Chancellor Merkel’s Christian Democratic Union have led the way, but the Social Democrats have not been far behind, and they have been joined by others in northern Europe (especially in the Netherlands). The argument is that every proposal to levy a common European tax or increase the common EU budget is an attempt by the countries of southern Europe, including France (which is said to be badly managed), to lay hands on the wealth painstakingly produced by the virtuous and hard-working Europeans of the north. This is not the place to explain how such distrust reached this level, which at times resembles an identity conflict. No doubt the French government’s recurrent tendency to complain about European budget rules that it helped to define (without proposing new rules to replace them) has long been irritating to Germans and others. Remember, too, that the Greek debt crisis began when Greek officials revealed that they had been seriously understating their country’s deficit—a cause of significant mistrust.⁷⁵ On the other hand, it is clear that the German view—that all of Europe’s problems could be solved if only every country would adopt the German model—makes no sense at all: if every country in Europe ran a trade surplus the size of Germany’s, the rest of the world could not absorb it. In any case, focusing exclusively on public transfers is not the right way to look at things. There are large flows of private money between states as well, with much of it going to countries like Germany that have invested heavily and profitably in their neighbors. Recall that the outflow of private profits from Eastern Europe vastly exceeds the inflow of public transfers.⁷⁶ In the future, it will be important to consider the flows of capital and profits made possible by the integration of the European economy (and the way in which these are affected by current laws and fiscal policies) to avoid focusing exclusively on public balances.⁷⁷

In any event, given the state of mistrust that currently exists in Europe after ten years of financial crisis in which every country has felt misused by the others, it is highly unlikely that any German (or French or other) government could persuade its citizens to transfer fiscal and budgetary authority to a European Assembly without setting a ceiling on any transfers that might result. If raising the proposed ceiling of 0.1 percent turns out to be possible, so much the better.⁷⁸ But the transfer ceiling should not be used as an excuse to reject the T-Dem proposal, which would remain useful even if explicit transfers were banned outright. The reason for this is that average incomes are not very different within the principal countries of the Eurozone, so the real goal is to reduce inequality within (rather than between) countries.⁷⁹ In other words, the lower and middle classes in all countries (including Germany) have much to gain from a more just tax system: for example, a system that would tax large companies at higher rates than small ones, high incomes and large fortunes at higher rates than low incomes and small fortunes, and heavy carbon emitters at higher rates than low carbon emitters. To sum up, the mere fact of establishing more just taxes within each country and protecting against the risk of fiscal competition (because the new taxes would be applied simultaneously in several countries) would in itself constitute decisive progress, even without any transfers.

Furthermore, the calculation of public transfers should of course exclude expenditures and investments by one country for the benefit of all, such as money spent on preventing climate change or housing refugees or educating students from other signatory states. Since the purpose of the common budget is to pay for public goods that will benefit all signatories, citizens of each country should see themselves as members of one and the same political community and view the common budget as something of benefit to all; one might then hope that over time the very concept of balancing each country’s contributions against its rebates would cease to be meaningful. Until then, however, one has to accept that trust must be built gradually, lest nationalist reflexes derail the plan.

Ending the Permanent European Public Debt Crisis

The social-federalist project presented here is driven by an ambition to achieve fiscal, social, and environmental justice. The goal is to enable a community of states (in this instance in Europe, but the idea could easily be extended to other contexts) to show that internationalism can lead to more just public policy and not simply to the merciless competition usually associated with European integration (and globalization more generally). In the specific context of the Eurozone, where nineteen countries chose to create a common currency while maintaining nineteen separate public debts and nineteen different interest rates, our proposal also includes the possibility (if the European Assembly so decides) of borrowing at a common rate of interest.80

Once again, in view of the climate of mistrust alluded to earlier, it is important to be clear to avoid misunderstanding and to allow for progress. The point is not to mutualize debt. In other words, it is not to take Germany’s debt (64 percent of GDP in 2018) and throw it into the same basket as Italy’s debt (132 percent of GDP) and then to ask German and Italian taxpayers to pay off the total amount with no regard to who threw what into the basket. Not that this idea is totally preposterous on its face: young Italians are no more responsible than young Germans for the debt they inherited from their forebears. The point is simply that no German party could possibly win if it assented to debt mutualization. If we are to achieve transnational justice and redraw European borders, we must consider history and politics when dealing with the debt or any other major issue. Specifically, our proposal for dealing with European public debt is inspired by the 2012 German debate on a “public debt redemption fund,” with one important difference: we rely on a democratic body, the European Assembly, rather than an automatic rule to decide the pace at which the debt will be repaid.⁸¹ In other words, the EA could decide to pool all or part of the debt of signatory countries in a joint refinancing fund and decide each year, as bonds come due, what portion of the debt to refinance by issuing new bonds. Each country’s debt would be placed in a separate account, however, to be serviced by that country’s taxpayers but at a rate of interest identical for all. That is the key point.

This point may seem technical, but in reality it is fundamental. Indeed, it was the chaotic course of interest rate spreads between the various countries of the Eurozone that led to the European debt crisis (even though European public debt on the eve of the crisis was no higher than that of the United States, Japan, or the United Kingdom). Why did the Eurozone perform so poorly after the economic crisis of 2008? Because of a lack of organization and an inability to create a common Eurobond. The crisis began in the private financial sector in the United States, but the Eurozone alone must bear the blame for transforming it into a persistent crisis of public debt. The consequences have been dramatic, particularly in terms of rising unemployment, identitarian retreat, and growing anti-immigrant sentiment. Prior to the crisis, however, European integration seemed to be succeeding: unemployment was down, the extreme right was in retreat, and migrant flows were higher than in the United States.⁸²

The emergency measures to which the Eurozone countries agreed to cope with the debt crisis did nothing to resolve the long-term issues, however, and they will need to be revisited in one way or another (unless their terms are simply ignored, which will only make everyone unhappy and exacerbate tensions). The new rules set by the 2012 budget treaty (the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union, or TSCG) stipulate that, in theory, deficits must not exceed 0.5 percent of GDP.⁸³ Barring “exceptional circumstances,” failure to obey the debt and deficit rules is supposed to be met with an automatic penalty. In practice, however, the rules are so absurd that they are unenforceable. The deficit in question is the secondary deficit; that is, the deficit after payment of interest on public debt. If a country’s public debt equals 100 percent of GDP and the interest rate is 4 percent, it needs to realize a primary surplus of 3.5 percent of GDP to stay within the rules. In other words, taxpayers must pay more in taxes than they receive back in government spending, with a gap between the two equal to 3.5 percent of GDP, possibly for decades.

In the abstract, the approach envisioned by the TSCG is not illogical: once one rules out exceptional measures such as debt restructuring and cancellation, if inflation is near zero and growth limited, then running large primary surpluses is the only way to pay down debt on the order of 100 percent of GDP over a period of decades. The social and political consequences of such a choice have to be borne in mind, however. Running large primary surpluses means decades of devoting huge resources to repaying principal and interest on bonds held in the portfolios of wealthy investors while the country forgoes investment in the transition to clean energy, medical research, and education.

In practice, the TSCG rules have never been and never will be enforced. For example, in the fall of 2018, a new crisis erupted between the European Commission and the social-nativist government in Italy. The Italians wanted to increase their deficit to 2.5 percent of GDP, whereas the previous government had promised 1.5 percent. The Commission objected, and a compromise was struck, allowing Italy to run a deficit set officially at 2 percent of GDP but in reality probably somewhere between 2 and 2.5 percent (in any case, significantly above the official limit of 0.5 percent, which no one seemed to take seriously). Given that interest on the debt currently represents about 3 percent of Italian GDP, this means that the country is running a primary surplus of 0.5 to 1 percent of GDP, which is not insignificant: with such a sum Italy could double (or even triple) its total spending on higher education (of just over 0.5 percent of GDP).

Some might find it reassuring to say that the required primary surplus would have been much larger had the Commission and the Eurogroup decided to apply the rules more strictly and to rejoice in such flexibility. But the truth is that it makes no sense to lay down such hyper-rigid rules only to ignore them because they are so absurd, thus ending up with a murky compromise negotiated behind closed doors without open deliberation.⁸⁴ Still, one might make do with a requirement that future primary surpluses should be positive but small (less than 1 percent of GDP). In other words, debtor countries could be asked from now on to levy taxes sufficient to cover their spending plus a little bit more, but they would not be expected to pay off old debt at a rapid rate. Such a solution would be tantamount to postponing repayment of the old debt to the distant future (which could be seen as a reasonable compromise). In practice, however, none of this is ever spelled out clearly, and what is expected of one country is not the same as what is expected of another.

In 2015, a clear political decision was made to humiliate Greece, which in the eyes of European (and especially German and French) authorities had elected a “radical left” party, Syriza (a coalition of communist, socialist, and green parties to the left of Pasok, the Greek socialist party, which had been discredited by having held power between 2009 and 2012 at the height of the financial crisis). Having won the election, Syriza sought to soften the terms of the austerity policy imposed on Greece by Europe’s leaders. But to avoid handing Syriza a symbolic victory, which European leaders feared might lead to a contagious spread of left-wing resistance (especially in Spain, where Podemos was on the rise), they decided to force the new Greek government to accept an even harsher austerity policy, requiring a primary surplus target of 3 percent of GDP even though Greek output had fallen 25 percent below its 2007 peak.⁸⁵ Meanwhile, Europe’s leaders ignored the fact that Syriza, for all its faults, was an internationalist party, open to Europe and supportive of immigrants arriving on Greek shores. It would have been wiser to work with the new Greek government to develop a more just fiscal policy for the EU, which might have included higher taxes on wealthy Greeks—and wealthy Germans and French as well.⁸⁶

The European approach to the Greek crisis may have disheartened the radical left, but it emboldened the radical right: three years later, in 2018, a social-nativist government came to power in Italy. This was a coalition held together mainly by hostility to foreigners, but because of Italy’s size, European officials were obliged to take a more conciliatory line in dealing with it.⁸⁷

Although today’s interest rates on sovereign debt are unusually low—a situation that may not last forever—interest payments on the debt currently amount to 2 percent of Eurozone GDP (the average deficit is 1 percent, and the primary surplus is 1 percent). In other words, more than 200 billion euros a year is being spent on interest payments, compared with a paltry 2 billion euros invested in the Erasmus student exchange program. Is this really the best way to achieve a better future? If such sums were devoted to education and research, Europe could lead the world in innovation, surpassing the United States. In any case, there should be a democratic forum for debating such choices. If there is another financial crisis, or even just an increase in interest rates, the flaws in the budget rules laid down in 2012 will quickly give rise to an explosive situation: the rules will be impossible to enforce, and latent tensions and hostilities among countries will rise to the surface because there is no legitimate democratic institution for seeking a better compromise.⁸⁸

Relying on the History of the Debt; Finding New Solutions

The solution I am proposing here is to place our trust in parliamentary democracy. Open, pluralistic, public deliberation is the only way to achieve the legitimacy necessary for such decisions and to respond in real time to the changing economic, social, and political situation. It is time to reconsider the erroneous belief first enshrined in the Maastricht Treaty of 1992 (and compounded by the TSCG of 2012) that Europe, merely by applying automatic budgetary rules, can have a common currency without parliamentary democracy, a common debt, or common taxes. Under the proposed plan, the European Assembly would be competent to decide on a common refinancing rate for all or part of the debt of member states as well as for rescheduling debt and deciding the pace at which it will be repaid. Countries wishing to benefit from the common bond and common rate of interest must agree to accept the will of the majority of the EA (in which the influence of each country will be limited by design). If a country wishes to retain full sovereignty over its debt and deficit, then it will not be allowed to benefit from the common interest rate. As for the pooled portion of the debt, the EA will be free to choose how payments of principal and interest will be scheduled and serviced.89 One solution might be to require member states to maintain a primary budget balance: tax revenues are exactly equal to expenditures, neither more nor less. This would amount to a rescheduling of existing debt over a long time horizon. If interest rates on the common debt are low (and kept there, whatever the financial markets do, by the action of the ECB, which will naturally purchase a significant share of the common debt)⁹⁰ and nominal growth is significantly higher in the future (which is not guaranteed), then the stock of past debt would gradually shrink in relation to GDP over the decades to come.⁹¹

Some might be tempted to engrave the rule of primary budget balance in stone.⁹² After all, once the possibility of democratically levying just taxes exists and the EA is authorized to tax high incomes and large fortunes in all signatory states, the idea of collecting just enough in taxes to cover all expenditures is an excellent principle to follow—as a general rule. The problem is that in certain circumstances—an economic crisis, say, leading to a large temporary decrease in tax revenues—such a rule is clearly too rigid. The same is true when long-term interest rates are unusually low (as they are now, partly because private investors are short of investment opportunities),⁹³ and governments, conversely, are in a position to promote strategic investments. Top priorities among those strategic investments include the transition to renewable sources of energy, the fight against global warming, research, and education.⁹⁴ To what extent are governments capable of identifying suitable investment opportunities and channeling funds to where they will do the most good? This is of course a very complex question. Still, we need to create public bodies with the legitimacy to make such decisions. Absent evidence to the contrary, there is no reason to think that we can do better than pluralistic, public deliberation in a parliamentary setting followed by a vote of representatives elected under the most egalitarian possible conditions. The idea that it might be preferable to replace democratic decision making with rigid and automatic rules expresses a nihilistic disillusionment with democracy (which no historical experience justifies).⁹⁵

In practice, the European Assembly could also decide to hasten the liquidation of debt by adopting specific measures such as exceptional progressive taxes on private wealth. Such measures played an important positive role at the end of World War II: they allowed public debt to be reduced quickly, creating room for public investment in reconstruction and growth, especially in Germany and Japan.⁹⁶ In hindsight, the most problematic aspect of the panoply of methods used in the postwar period was undoubtedly the recourse to inflation, which did contribute to rapid reduction of the debt but at the price of eating away the savings of the lower and middle classes. In light of such experience, it seems reasonable to maintain the ECB’s mandate to keep inflation low and to focus on other proven methods of debt reduction, this time taking advantage of explicit coordination at the European level both to reschedule debt (relying on the ECB to keep interest rates very low) and to levy exceptional taxes (by way of the EA). The EA could decide to reschedule debt, for instance by delaying repayment until the countries of the Eurozone return to levels of employment and growth comparable to those of the pre-crisis period (especially in southern Europe but also throughout the Eurozone). The EA could also decide to delay repayment of the debt until sufficient progress has been made toward other goals, such as combating climate change. This could fairly easily be justified.⁹⁷

To conclude, I want to emphasize that my purpose is not to decide in advance what course should be followed. It is simply to illustrate the need for a democratic body of incontestable legitimacy such as the European Assembly, which would draw on both national parliaments and the European Parliament and be empowered to take the complex decisions that the situation requires. The idea that the significant problems raised by European public debt can be resolved by the kind of automatic budget rules included in the TSCG of 2012, which assumes that lower- and middle-class taxpayers will quietly agree to pay the taxes required to achieve large primary budget surpluses for decades to come, is totally unrealistic. Since 2008, the debt crisis has exacerbated existing tensions among the countries of Europe. In the end, it sowed reciprocal misunderstanding and mistrust among the countries primarily responsible for the construction of the European Union, most notably Germany, France, and Italy. The potential remains for serious political disturbances or even disintegration of the Eurozone. If we go on pretending to resolve these problems behind closed doors, in meetings of heads of state and ministers of finance in which naked power takes precedence over reason, then new crises are likely to erupt. Only the constitution of a true transnational parliamentary democracy offers the possibility of an open and thorough examination of the various options available in the light of historical experience. Without such a thorough examination, no lasting solution is possible.

On the Political Conditions for a Social-Federalist Transformation of Europe

The advantage of the social-federalist approach just outlined is that it would allow a core group of European countries that wished to move toward a stronger political and fiscal union to do so without undermining the current European Union of twenty-seven or twenty-eight member states. Call this new union the European Parliamentary Union, or EPU, to distinguish it from the current European Union (EU). Ideally, the core group of EPU members would include the four largest countries of the Eurozone (Germany, France, Italy, and Spain); at a minimum, two or three of these countries would be needed to make the EPU viable. Of course, it would be best if all the countries of the Eurozone joined up right away, but some non-Eurozone countries might be more eager to join.98 Regardless of whether the initial core group consists of five, ten, or twenty countries, there is no reason why it could not peacefully and durably coexist with the EU for as long as it takes to convince all EU member states to join the EPU, at which point the two entities could merge. During the transitional phase, the member states of the EPU would participate in both its institutions (including the European Assembly, which would approve the EPU budget and taxes) and the institutions of the EU. If EPU members successfully demonstrate that their more empowered union can achieve greater fiscal, social, and environmental justice that the existing EU, then hopefully most EU member states would eventually, if not immediately, want to sign up.

Such a peaceful transition, though desirable, is unfortunately not the only imaginable scenario. In practice, it is likely that states that have invested heavily in fiscal dumping, such as Luxembourg and Ireland, would fiercely resist. Not only would they refuse to participate in the project; more than likely they would try to sabotage it by arguing that the EPU somehow violated existing treaties. They might even bring suit before the Court of Justice of the European Union (CJEU) on the grounds that only a general revision of the European treaties (requiring a unanimous vote of member states) can end the rule of fiscal unanimity and create a European Assembly allowing decisions by majority vote. The argument that unanimity is required to end the rule of unanimity may seem particularly specious and high-handed, but the national interests at stake (or perceived to be at stake) are so enormous that it would be a mistake to think that such arguments would not be made. The CJEU did validate the intergovernmental treaties signed in 2012 to cope with the financial emergency after determining that there existed no other legal way to respond to the crisis. So it is possible that it would respond in the same way to the Treaty on the Democratization of Europe (or a similar text) on the grounds that there is no other way to deal with the democratic and social emergency.⁹⁹ That said, law is not an exact science, so there is no guarantee that the CJEU would approve, in which case the states backing the EPU would have no choice but to renounce the existing EU treaties so as to force other countries to negotiate new ones.

Furthermore, regardless of how the T-Dem or any similar text comes into effect, any attempt by the core countries to establish a common tax system would almost inevitably give rise to tensions with the countries that chose to remain outside. In particular, during the transitional phase, if the EPU decided to tax corporate profits, high incomes, large fortunes, and carbon emissions, it would need to make certain requests of nonmembers for information about cross-border profit flows, income, financial asset holdings, and carbon content of traded goods. Past experience suggests that it would not be easy to obtain their cooperation on such matters. Trade sanctions would likely have to be imposed to obtain the desired information. For example, in regard to taxing corporate profits, one way to deal with the lack of adequate international cooperation might be to apportion the profits of multinational corporations on the basis of the amount of goods and services sold in different countries (independent of the location where the profits are officially—and often fictitiously—reported).¹⁰⁰ All signs are that if the larger countries of the Eurozone imposed sanctions on Luxembourg and Ireland, those countries would quickly yield.¹⁰¹ But the will to play hardball is indispensable, especially since the sanctioned countries would surely denounce the sanctions as violations of existing treaties.¹⁰²

Consider, for example, the US threat in 2010 to withdraw the banking licenses of Swiss banks doing business in the United States. This threat broke a negotiating deadlock, compelling the Swiss government to amend its laws to allow banks to transmit to US tax authorities information about accounts held in Switzerland by US citizens. In Europe, if Germany, France, and Italy were to make similar threats against Luxembourg or Switzerland, the threatened countries would surely protest that sanctions are inconsistent with current European treaties. Unfortunately, such sanctions may be necessary to change the status quo, and they would probably need to be enforced for some period of time before having any real impact.

To sum up, the real obstacle is neither legal or institutional; it is primarily political and ideological. The central question is whether the countries that are suffering most from tax competition—chiefly large countries such as France, Germany, Italy, and Spain—consider the issue important enough to justify a proactive strategy that might include punitive sanctions against states that refuse to cooperate (which might require a unilateral exit from existing treaties). To date, the approach taken by most governments and parties, including socialist and social-democratic parties of one stripe or another, has been to regard tax competition as a problem, to be sure, but a problem that unfortunately cannot be solved as long as Luxembourg, Ireland, and all other countries refuse to give up their veto power. But it has been clear for some time that such an approach leads nowhere. Unfortunately, the governments of the large countries have thus far not felt that the issue was important enough to risk dividing the EU by creating separate political institutions (such as the European Assembly I am proposing) for a subset of countries prepared to move forward. Their hesitation is understandable. But ultimately the risks inherent in the status quo—namely, a definitive and potentially fatal breach between the disadvantaged classes and the proponents of the European project—seem to be greater. Furthermore, the construction of a transnational parliament exercising fiscal sovereignty by way of democratic deliberation is likely to be a fragile process, and it is therefore almost inevitable that it should begin with a small number of countries; only after it has demonstrated its viability need it be extended to the others. In other words, if the process (which would have been easier had it begun sooner) is delayed until all twenty-seven or twenty-eight member states are ready to move forward, it will probably never get started at all.¹⁰³

Why has the process not already begun? Ultimately, the reason is no doubt that many political leaders and parties, especially in Germany and France and not only of the center-right but also of the center-left, continue to believe that the benefits of fiscal competition (in pressuring states to hold down spending at a time when taxes are already at historically high levels) outweigh the costs of the endless race to the bottom (which benefits those whose capital is most mobile) or at least do not justify the significant political complications that would arise from trying to end it.¹⁰⁴ Another equally important ideological factor is that the European project has long relied on the sacrosanct right of states to enrich themselves through trade and free circulation of goods, capital, and people and then further enrich themselves by siphoning off their neighbors’ tax base. In reality, that sacrosanct right does not exist: it is a consequence of a very specific ideological interpretation of the history and politics of the European Union, whose benefits to the upper class in all the member states (including France and Germany) far outweigh any benefits accruing to the lower and middle classes of Ireland and Luxembourg. But leaders have insisted on that right for so long that it has come to be perceived as legitimate.¹⁰⁵

Finally, although readiness to renounce existing treaties is no doubt a necessary condition for reaching agreement on new ones, it is by no means sufficient. Since the crisis of 2008, various political parties such as Podemos in Spain and LFI in France have toyed with the idea of threatening to exit as a way of forcing the EU to agree to new policies, especially in the area of fiscal and social harmonization.¹⁰⁶ The problem is that these parties have not thus far indicated precisely what new political system they would like to see established in Europe. In short, we know what treaties they would like to renounce but not what treaties they would like to endorse in their place. The problem with this strategy is that is easily caricatured as anti-European, as it has been since 2008 by German and French governments, which in effect instrumentalize the European project to impose their inegalitarian ideology and refusal to consider common taxes at the European level. This is a powerful argument for discrediting these upstart parties in the eyes of a public worried about the prospect of dismantling the EU—an effective strategy for keeping them out of office.

Furthermore, if one of these parties were somehow to come to power in France, for example, the accumulated mistrust between member states (and between France and Germany in particular) could trigger a chaotic and uncontrollable breakdown of the European treaties. Resentment and misunderstanding among countries could ultimately outweigh their attachment to the European ideal. Another risk, in my view at least as likely as the first, is that devotion to Europe would keep the European Union together but, in the absence of any specific commitment to new institutions or precise plans for fiscal and social harmonization, would end in an insipid, disappointing compromise, especially if there is no prior public debate and citizens fail to grapple with these complex yet eminently political questions.¹⁰⁷

The Separatist Trap and the Catalan Syndrome

What is at stake in the social-federalist transformation of Europe extends far beyond the boundaries of Europe itself. The question is whether a different organization of the global economy is possible. Can the treaties that currently govern free trade and customs unions be replaced by a broader set of international accords based on a model of durable and equitable development, with concrete and attainable goals of fiscal, social, and environmental justice? In the absence of such accords, the risk is that the race to the bottom will continue: fiscal dumping will increase; inequality will continue to rise; and xenophobic, identitarian, anti-immigrant political parties will continue to exploit the situation in their pursuit of power.

Another risk involves what one might call the separatist trap. An example of this can be seen in the attempt to organize a referendum on self-determination in Catalonia in 2017. It is striking to see the degree to which regionalist sentiment in Catalonia varied with income and education. When Catalonian voters were asked whether they supported the demand for greater regional autonomy (potentially leading to independence), it turned out that support increased with increasing income and education: support for the regionalist idea ran as high as 80 percent of those polled in the top decile of income or education compared with 40–50 percent among the bottom five deciles (Figs. 16.5–16.6). If we look only at voters supporting a referendum of self-determination (and thus eliminate those favoring greater autonomy within Spain), we find that the cleavage is even more pronounced: support for independence is dramatically higher among the upper classes, particularly those with the highest incomes.¹⁰⁸ Note, too, that support for self-determination increased sharply after the economic crisis, which hit Spain hard after 2009, with a second dip in 2011–2013 after austerity policies were imposed at the European level. Only 20 percent of Catalan voters favored self-determination in 2008, compared with 32 percent in 2011 and 35 percent in 2016.¹⁰⁹ It was because of this rapid increase of support for self-determination that the Catalan government organized an independence referendum in 2017 against the will of the government in Madrid; the election was boycotted by parties in favor of keeping Catalonia in Spain, and this precipitated a serious constitutional crisis, which is still ongoing.¹¹⁰

FIG. 16.5.  Catalan regionalism and income, 2008–2016

Interpretation: In 2008, 47 percent of Catalan voters belonging to the bottom 50 percent of the income distribution supported greater regional autonomy or a referendum of self-determination (answers to both questions were added), compared with 64 percent of the next 40 percent and 74 percent of the top 10 percent. Sources and series: piketty.pse.ens.fr/ideology.

FIG. 16.6.  Catalan regionalism and education, 2008–2016

Interpretation: In 2016, 44 percent of Catalan voters with only a primary education supported greater regional autonomy or a referendum on self-determination (independence). The two scores were added. Compare this with 60 percent among those with secondary diplomas and 74 percent of those with tertiary degrees. Sources and series: piketty.pse.ens.fr/ideology.

It is quite striking to discover that Catalan regionalism is much more pronounced among those with more advantages. It is instructive to compare the social profile of the Catalan vote with the profiles observed in the referenda on Europe conducted in France in 1992 and 2005 and in the United Kingdom in 2016. In all cases we find that the advantaged classes voted heavily for Europe while the disadvantaged rejected it.111 These vote profiles are perfectly consistent, moreover, since the advantaged classes that supported Catalan independence (or increased autonomy) had no desire to quit the EU—quite the opposite. They wanted Catalonia to remain in the EU but as an independent state so as to continue to benefit from commercial and financial integration with Europe while keeping Catalan tax revenues in Catalonia.

Of course, it would be wrong to reduce Catalan regionalism to a fiscal motive. Cultural and linguistic factors are also important, as is the memory of Francoism and the brutality of the central government in Madrid. Still, the issue of fiscal autonomy played a key role in the Catalan regionalist movement, especially since Catalonia is wealthier on average than the rest of Spain. It is natural to think that the wealthiest taxpayers were particularly exasperated by the thought that some of what they paid in taxes was being transferred to other regions. By contrast, the lower and middle classes may be somewhat more sensitive to the virtues of social and fiscal solidarity. Note, however, that Spain is already a country with one of the most decentralized tax systems in the world, even when compared with much larger federal states. Specifically, since 2011, the income tax has been shared equally between the federal government and the regions.¹¹² There are many problems with such a system: it undercuts the very idea of solidarity among citizens and pits region against region, which is particularly problematic when it comes to a tool like the income tax, which is supposed to make it possible to reduce inequality between rich and poor regardless of regional or professional identity.¹¹³

By way of comparison, in the United States the income tax has always been mainly a federal tax, even though the population is seven times as large as Spain’s and despite the American penchant for decentralization and states’ rights. Since the creation of the federal income tax in 1913, it has been the main tool for achieving fiscal progressivity, applying the highest rates to the highest incomes.¹¹⁴ No doubt the wealthy taxpayers of California (a state almost as populous as Spain, with six times the population of Catalonia) would have liked to keep half of the income tax paid by the state’s highest earners for themselves and their children, but they never succeeded in doing so (and never really tried, since the idea would have been interpreted as a secessionist declaration of war). Or consider an example closer to Spain: in the German Federal Republic the income tax is exclusively federal. The Länder are not allowed to levy additional taxes or keep any of the revenue for themselves, no matter what the taxpayers of Bavaria may think. To be clear, there is nothing necessarily wrong with levying additional taxes at the regional or local level, provided they remain moderate. But Spain, by choosing to divide income tax revenues fifty-fifty with the regions, probably went too far and now finds itself in a situation where some Catalans would like to keep 100 percent for themselves by becoming independent.

Europe also bears heavy responsibility for the Catalan crisis. In addition to its calamitous handling of the Eurozone crisis, to the detriment of Spain in particular, the European Union has for decades been promoting a development model based on the idea that it is possible to have everything at once: an integrated European and global market without any genuine obligation of solidarity or financing of public goods. Under such conditions, why shouldn’t Catalonia try its luck and become a tax haven like Luxembourg? For many pro-independence Catalans, that is indeed the goal: as an independent state, Catalonia could keep all of its tax revenues for its own development while at the same time cutting taxes on foreign investors to draw new capital into the region. Not having to share revenues with the rest of Spain would make it that much easier to cut taxes on foreigners. There is no doubt that the politics of Catalan independence would have been totally different if the EU had had a federal budget comparable to that of the United States, financed by progressive federal income and inheritance taxes. If the taxes paid by high earners in Catalonia went to the EU federal budget, just as the US income tax goes to the US federal budget, Catalonia would have only a limited financial interest in separating from Spain. To escape the bonds of fiscal solidarity, it would need to exit Europe with the risk of being barred from the vast European market, the cost of which would be prohibitive in the eyes of many pro-independence Catalans. I am not claiming that the Catalan regionalist and independence movement would then immediately disappear or that it should disappear. But it would be seriously weakened, and its focus would turn to cultural, linguistic, and educational issues, which are important and complex, rather than being obsessed with tax issues and obscure bargains between regions. The Catalan crisis in its present form is a symptom of a Europe that pits region against region in a race to the bottom with no fiscal solidarity whatsoever. Every country seeks advantage for itself by undercutting its partners. The Catalan case shows how the organization of the political system is intimately intertwined with the issues of inequality, borders, and property rights.

Ideological Dissonance, Fiscal Dumping, and the Small-Country Syndrome

The temptations of fiscal competition can be strong, even in communities not initially inclined that way ideologically. Before Luxembourg became a tax haven, it had no particular ideological disposition to assume that role.115 But once globalization (and in particular the treaties governing the free circulation of capital) developed in such a way as to make this strategy appealing, the temptation became too strong to resist. Small countries are particularly susceptible because the amount of (real or fictitious) investment they can hope to attract is quite large relative to the size of their economies. Neighboring countries may have large tax bases, which can more than make up for whatever domestic revenues may be lost by cutting taxes on the wealthy.¹¹⁶

The Swedish case offers a particularly extreme example of ideological dissonance.¹¹⁷ During the Swedish banking crisis of 1991–1992, Swedes realized that a small country in a world of major financial flows and capital movements is quite exposed and vulnerable. The crisis might have been seen as an occasion to reconsider the dangers of the financial deregulation of the 1980s. In practice, however, it was instrumentalized by people who had believed for decades that the Swedish social model had been pushed too far, that the social democrats had been in power too long, and that it was time for the country to move toward the new Anglo-American liberal model that had emerged from the conservative revolution of the 1980s. The conservative liberals briefly came to power in 1991–1994, long enough to sharply reduce the progressivity of Swedish income and wealth taxes and to institute a flat tax of 30 percent on interest and dividends, which for the first time were exempted from the progressive tax regime. Conservative ideology continued to make inroads in the 1990s and 2000s, and in 2005 and 2007 the progressive tax on inheritances and wealth were abolished.¹¹⁸

The Swedish decision to abolish the inheritance tax in 2005, at practically the same time as Hong Kong (2006), illustrates the strength of the “small-country syndrome.” Larger countries such as Germany, the United Kingdom, France, Japan, and the United States have all maintained the progressive inheritance tax, assessing rates of 30–55 percent on the largest estates in the late 2010s.¹¹⁹ But Sweden’s social democrats decided it would be a good idea to eliminate any tax on intergenerational wealth transfers, even though Germany’s Christian Democrats, Britain’s Conservatives, France’s Gaullist liberals, and even US Republicans judged it preferable to keep them with reduced but still substantial rates on the largest fortunes.¹²⁰ During Swedish debates on these issues, the fear of capital leaks toward other countries of the region played an essential role. Whether justified or exaggerated, these fears did not induce the Swedish government to push for reform of the directives on the circulation of capital or for greater fiscal cooperation in Europe. As in the case of Catalonia, the solution was nevertheless simple: it would have sufficed to levy a progressive tax at the EU level. The fact that the Swedish social democrats never considered making such a proposal shows the degree to which the ideological and political agenda of social democracy remains confined for the moment to the nation-state. To be sure, Sweden remains more egalitarian than other countries thanks to an advanced system of social insurance financed by substantial taxes and social contributions assessed on the entire population as well as to a free and high-quality educational system (including higher education). Still, the abolitions of 2005–2007 increased inequality at the top of both the wealth and income distributions in Sweden since 2000 and may ultimately weaken the Swedish model.¹²¹ This resistance to international cooperation made it more difficult to maintain progressive taxes elsewhere, including both rich countries and poor and emerging ones.¹²²

What is more, the “small-country syndrome” may spread to larger countries. As emerging economies claim an ever greater share of the global economy, which has grown to unprecedented size, nearly all countries are small in relation to the global economy, including France, Germany, the United Kingdom, and even to a certain extent the United States. For many Conservative leaders, the purpose of Brexit is precisely to turn the United Kingdom into a tax haven and lightly regulated financial center (a postindustrial conversion process that in some respects began in the 1980s). Absent a social-federalist turn, globalization is likely to have the same effect in many other countries.

The Social-Localist Trap and the Construction of the Transnational State

It will not be easy to follow the social-federalist route to building a transnational governmental authority. For that reason, some political movements may be tempted by a social-localist strategy—promoting equality and economic alternatives at the local level. For instance, the Catalan independence movement includes a minority left-wing faction that sees Catalonia as more friendly to social experimentation than the government in Madrid (and that also wants to break with the Spanish monarchy and turn Catalonia into a republic). Unfortunately, it is quite possible that this left-wing group would be outflanked and dominated in any future Catalan state by conservative liberals wedded to a very different model of development (of the tax haven variety).

It is of course perfectly legitimate to promote a social-localist agenda, particularly since action at the local and municipal level can indeed offer opportunities to reshape social and property relations complementary to what can be achieved at the central level. Still, it is important that local action be conducted within a more general social-federalist framework. To clear up ambiguities about the various forms of Catalan regionalism and distinguish itself from those who simply want to keep regional tax revenues for themselves and their children, the pro-independence republican left should make clear that it favors common progressive wealth and income taxes at the European level. Just because the path to social federalism is complex is not a reason to be unclear about the broader strategy—quite the contrary.

The broader strategy is especially important because when it comes to the kinds of political action that social localism inspires, there are often fairly obvious limits to what can be achieved unless those actions are complemented by higher level regulations and policies. Take, for example, a recent effort to keep Google out of Berlin. As a result of anti-Google demonstrations, the company decided not to build a new campus in the Kreuzberg district of Berlin. This “campus,” like others that Google already operates in London, Madrid, Seoul, São Paulo, Tel Aviv, and Warsaw, was to occupy an old red-brick factory and serve as a place for meetings, events, and training for information technology professionals. The local associations that organized the “Fuck Off Google” movement could legitimately proclaim victory. They had persuasively led the charge against real estate speculation, higher rents, and evictions of low-income families—which, for this already gentrifying neighborhood, would be the inevitable consequences of Google’s decision to move in, even though it paid virtually no taxes in Germany and other countries where it earned most of its profits. This successful effort to block Google, a large-scale tax evader, drew a great deal of attention in Berlin, where the Christian Democrats blamed the governing coalition of SPD, Greens, and Die Linke for creating a climate “hostile to entrepreneurs” (which the coalition denies).123

Mobilizations of this type raise complex issues. Of course, it is close to unbearable to hear the CDU use the word “entrepreneur” to describe a corporation that pays virtually no taxes, especially since the party has led the federal government in Germany (Europe’s leading economic power) from 2005–2019 without doing anything to make Google accountable. But it is also clear that local mobilizations like the one in Berlin are not enough, partly because other cities will undoubtedly welcome a “Google campus” and partly because the real goal is to be able to tax and regulate a company of Google’s size at the European level. And the fact is that the SPD, Greens, and Die Linke have thus far proposed no common plan of action that would make it possible to, say, levy a European tax on the profits of the largest corporations or, at a minimum, a Franco-German tax or a tax levied by the largest possible subset of EU member states. Sticking to social localism and refusing to join an ambitious social-federalist movement also offers adversaries particularly effective lines of attack.

In other contexts, especially in the United States, it is sometimes easier to go from a social-localist commitment to a social-federalist one. Consider the example of Alexandria Ocasio-Cortez (known as AOC), a Democrat from New York who was elected to the House of Representatives in November 2018. A member of the Democratic Socialists of America, AOC took a leading role in the fight to prevent Amazon from building a new headquarters in Brooklyn. As in Berlin, the movement zeroed in on the fact that the company not only pays virtually no tax on its profits but was asking for generous public subsidies, which various cities interested in hosting the new headquarters were competing to provide. Amazon’s refusal to allow any union representation added fuel to the fire. The conflict ended when Amazon decided not to go ahead with the Brooklyn project in January 2019. To no one’s surprise, Republican and Trumpist pressure groups unleased their wrath against AOC.¹²⁴ Unlike the anti-Google activists in Berlin, an elected representative like AOC can champion policies to regulate of large corporations and can also vote for progressive federal taxes (AOC is among those supporting a marginal income tax rate above 70 percent on the highest incomes).¹²⁵ In the European context, by contrast, no such social-federalist platform is even possible unless people mobilize both to transform Europe’s institutions and to build transnational coalitions to that end.

The Construction of Indian Political Parties and Cleavages

We have just taken a relatively detailed look at the conditions under which social federalism might develop in Europe and how it might provide a way out of the social-nativist trap. While the European case offers some lessons of general applicability, it remains a fairly special case. If we wish to gain a better understanding of the transformation of political cleavages and the structure of political-ideological conflict in large federal communities as well as of the risk of identitarian withdrawal in electoral democracies, it is absolutely essential that we not confine our attention to Europe and the United States alone. For that reason we will now turn our attention to political cleavages in India and Brazil.

The evolution of the party and cleavage structure of the Indian Union is particularly interesting, in part because it is the largest parliamentary federal republic in the world (with 1.3 billion citizens, compared with 510 million in the European Union and 320 million in the United States) and in part because, as we will see, the Indian party system has evolved since the 1960s toward a classist system while Western electoral democracies have evolved in the opposite direction. The Indian case is highly instructive because it shows that the construction of egalitarian coalitions and classist cleavages can follow a number of different paths and does not depend on exceptional events (such as the two world wars and the Great Depression in the West). This decentering of our gaze outside the West is also essential for rethinking the issue of federalism and deepening our understanding of the identity and ethno-religious cleavages that have emerged in Europe in recent decades. Comparable cleavages exist in India, which has a much longer experience of multiconfessionalism. It is instructive to compare the ways in which these issues are politicized in different countries.

In India’s first elections after independence and the partition of Pakistan in 1947, the Congress Party (the Indian National Congress, or INC) played a clearly dominant role. Founded in 1885, INC had led India to independence by peaceful parliamentary means and therefore enjoyed great legitimacy. The Congress Party had always held a “secularist” multiconfessional view of India and insisted on respect for all religions (whether Hindu, Muslim, Christian, Sikh, Buddhist, Jewish, or atheist). It was also under Congress leadership that the Constitution of 1950 established a system of quotas and “reservations” aimed at giving former untouchables and aboriginal tribes (“scheduled castes/scheduled tribes,” or SC/ST) access to higher education, public employment, and elective office. The intent of these policies was to rid the country of the inegalitarian heritage of the old caste system, which British colonialism had helped to rigidify. In practice, Congress relied on traditional local elites, often drawn from the highest castes, especially the Brahmin literati (like the Nehru-Gandhi family). INC combined a certain progressivism with various forms of social and political conservativism with respect to issues of property and education as indicated by the absence of real agrarian reform in India and insufficient investment in public services, health, and education for the socially disadvantaged.126

In the legislative elections of 1951, 1957, and 1962, INC took 45 to 50 percent of the vote, enough to win a comfortable majority in the Lok Sabha given the fragmentation of the opposition and the nature of the voting system.¹²⁷ The rest of the votes were scattered among a host of ideologically very different parties: regionalists, communists, nationalists, socialists, and so on, none of which seriously threatened the dominance of the Congress Party. In the 1957 and 1962 elections, the country’s second leading party was the Communist Party of India (CPI), which took roughly 10 percent of the vote at the federal level.¹²⁸ The Hindu nationalists of Bharatiya Jana Sangh (BJS, or party of Hindu people) finished third, with less than 7 percent of the vote. INC’s unchallenged dominance began to crumble in the 1960s and 1970s. The Congress vote fell below 40 percent, and it lost power for the first time in 1977 with the victory of the Janata Party (party of the people). But this was an ad hoc anti-INC coalition of left- and right-wing opponents of Indira Gandhi’s Congress with no real common program.¹²⁹ It did not last. INC bounced back, more united and coherent than before, and regained power in the 1980 elections. All in all, India was governed virtually without interruption by INC and by prime ministers of the Nehru-Gandhi family for four decades, from the late 1940s to the late 1980s.¹³⁰

After the first phase of Indian democracy, which the Congress Party dominated from 1950 to 1990, came the second phase (1990–2020), characterized by the gradual development of a true multiparty system with parties alternating in power at the federal level. When we look at the results obtained by the various parties in elections to the Lok Sabha, we find that INC’s position began deteriorating around 1990: from 40 percent of the vote in 1989, its score fell to 20 percent in 2014. If we count the various centrist parties allied to Congress, however, the 2014 score comes to around 35 percent—much reduced from the postwar decades but still substantial (Fig. 16.7).¹³¹

Since 1990, India has witnessed the rise of the Bharatiya Janata Party (BJP).¹³² Over time the BJP has become a vast and well-oiled political machine. It describes itself as “the largest political party in the world.”¹³³ The BJP (like its predecessor, the BJS) is also the political and electoral arm of a huge Hindu missionary organization, the Rashtriya Swayamsevak Sangh (RSS), which is a federation of various youth movements ranging from a Hindu version of the Boy Scouts to actual paramilitary organizations.¹³⁴ Founded in 1925, the RSS is an organization whose ideology is in many ways the exact opposite of the ideology of INC (founded in 1885). Whereas INC proposed to unite India on the basis of secularism and religious diversity, the RSS has always preached a strictly Hindu and violently anti-Muslim version of nationalism. For instance, one of the founders of RSS, M. S. Golwalkar, alluded to an “800-year war” between Hindus and Muslims in a text that he wrote in 1939 that is one of the movement’s foundational documents. In it, he explained how Islam had profoundly handicapped the development of Hinduism and of Indian civilization more generally; a civilization, Golwalkar bluntly explained, which over the millennia has achieved a degree of refinement and sophistication never rivaled by Christianity or Islam.¹³⁵ Feelings of humiliation and the need for revenge after nearly two centuries of British colonial rule also played a crucial role.

FIG. 16.7.  Legislative elections in India (Lok Sabha), 1962–2014

Interpretation: In the 2014 legislative elections, the Congress Party (INC) and allied centrist parties won 34 percent of the vote (19 percent for the INC alone), the BJP (Hindu nationalists) and allied right-wing parties won 37 percent, and the parties of the left and center-left (SP, BSP, CPI, etc.) 16 percent, and other parties 13 percent. Note: In the 1977 legislative elections (after the state of emergency), Janata Dal grouped both left and right opponents of the INC and is here classed with “other parties.” Sources and series: piketty.pse.ens.fr/ideology.

To encourage the rebirth of Hindu civilization, the RSS and BJP have proposed an elaborate vision of the ideal society, which clearly cannot be reduced to religious hostility. Specifically, the principles of social harmony and moderation, embodied in vegetarianism and respect for traditional families, the Hindu religion, and Sanskrit culture, play an essential role in the doctrines they espouse. Nevertheless, hostility to Islam is never very far below the surface. The increasingly violent riots that the RSS and other Hindu religious organizations began to foment in 1984 with an eye to rebuilding a Hindu temple at Ayodhya (Uttar Pradesh), the mythical city of the god Rama as described in the Ramayana, played a central role in the BJP’s rise. The destruction of the Babri Masjid (a sixteenth-century mosque) by Hindu activists in Ayodhya in 1992, following years of violence backed by the RSS and BJP then in power in the region, marked a decisive step.136 Numerous similar riots ensued and continue to plague the country.¹³⁷ In the BJP’s 2019 campaign manifesto, the promise to rebuild a temple of Rama on the site of the Ayodhya mosque is still listed among the party’s top priorities.¹³⁸

In addition to the two main electoral blocs led by INC and BJP, there is a persistent third bloc consisting of parties of the left and center-left (Fig. 16.7). This group includes not only the various communist parties (CPI, CPI[M] [Marxist], and so on) but also a large number of parties that describe themselves as socialist or social-democratic, such as the Samajwadi Party (SP, a socialist party descended from the secularist branch of the Janata Party coalition of 1977–1980 and its brief rebirth as Janata Dal in 1989–1991) as well as lower-caste parties such as the Bahujan Samaj Party (BSP, “party of the majoritarian society”), about which I will say more later.¹³⁹ These parties play a central role in certain states and receive about 20 percent of the vote at the federal level. Ideologically, they are generally closer to INC than to the BJP but do not officially endorse either camp. The SP and BSP formed an explicit alliance in the 2019 elections. Whether they will join Congress or not is one of the key political questions facing India today.¹⁴⁰

Indian Political Cleavages: Between Class, Caste, and Religion

We turn next to the question of how the structure of India’s various electorates has evolved in relation to their respective ideologies. Let us begin with the vote for the BJP and its allies as a function of caste and religion (Fig. 16.8).¹⁴¹ Broadly speaking, we find that there has always been a very strong cleavage in the structure of the BJP vote. Unsurprisingly, voters identifying as Muslim have never been tempted to vote for the BJP (barely 10 percent of them do so). In other words, 90 percent of Muslim voters have always voted for parties other than the BJP. In view of the BJP’s violent anti-Muslim rhetoric, this is hardly surprising. Within the Hindu electorate, we find that the BJP vote has always been an increasing function of caste, in the sense that the likelihood that a voter will vote for the BJP or its allies is systematically lower among the lowest castes, especially the former untouchables and members of aboriginal tribes (SC/ST). It is slightly higher among the “other backward classes” (OBC), and it peaks among the highest castes, especially Brahmins. For instance, in the 1998 and 2014 elections, we find that 60 percent of Brahmins voted for the BJP.

FIG. 16.8.  The BJP vote by caste and religion in India, 1962–2014

Interpretation: In 2014, 10 percent of Muslim voters voted for the BJP (Hindu nationalists) and allied parties, compared with 31 percent of the SC/ST (scheduled castes/scheduled tribes, lower castes), 42 percent of OBC (other backward classes, intermediate castes), 49 percent of other FC (forward castes, high castes excluding Brahmins), and 61 percent of Brahmins. Sources and series: piketty.pse.ens.fr/ideology.

To interpret these results properly, remember that Muslim voters represented 10–15 percent of the Indian population from 1960 to 2010, compared with 25 percent for the SC/ST, 40–45 percent for the OBC, and 15 percent for the high castes (6–7 percent Brahmins).142 Note, moreover, that it is fairly logical that the BJP electorate is so tilted toward the higher castes. This cleavage reflects the widespread perception among the lower castes that Hindu nationalists attach great value to the traditional social order and to the symbolic and economic domination of the high castes. In particular, the BJP and its allies have often opposed quota systems that favor the lower castes, which they see as an unhelpful cause of division within a supposedly harmonious Hindu society as well as a reduction of the number of places for their children in universities, public employment, and elective office. In view of these positions, it is not surprising that the castes benefiting from the “reservations” system (SC/ST and OBC) are not generally drawn to the BJP.

Looking next at the vote for Congress and its allies as well as for the parties of the left and center-left, we find profiles that are the inverse of those we saw in the BJP vote (Figs. 16.9–16.10). The propensity to vote for INC and parties of the left is highest among Muslim voters, slightly lower among low-caste voters (SC/ST and OBC), and sharply lower among high-caste voters, especially Brahmins. At first sight, this reflects the fact that Congress and the parties of the left have always championed a secularist idea of India, notably by coming to the defense of Muslims against the BJP. They have also fought to reduce inequality between the lower castes and the upper castes by backing various quota systems.

Several points deserve further comment, however. First, the magnitude of the observed cleavages is striking. Among Muslim voters we regularly find votes of 50–60 percent in favor of INC and its allies and 20–30 percent for the parties of the left and center-left (for a total of 80–90 percent). The levels observed among lower-caste voters (especially SC/ST) are nearly as high. By contrast, high-caste support for these parties is very low, particularly toward the end of the period.

FIG. 16.9.  Congress party vote by caste and religion in India, 1962–2014

Interpretation: In 2014, 45 percent of Muslims voters voted for Indian National Congress (INC) and allied parties, compared with 38 percent of SC/ST (scheduled castes/scheduled tribes, lower castes), 34 percent of OBC (other backward classes, intermediate castes), 27 percent of other FC (forward castes, high castes excluding Brahmins), and 18 percent of Brahmins. Sources and series: piketty.pse.ens.fr/ideology.

FIG. 16.10.  The left vote by caste and religion in India, 1962–2014

Interpretation: In 2014, 23 percent of Muslim voters voted for parties of the left/center-left (SP, BSP, CPI, etc.), compared with 17 percent of SC/ST (scheduled castes/scheduled tribes, lower castes), 15 percent of OBC (other backward classes, intermediate castes), 11 percent of other FC (forward castes, high castes excluding Brahmins), and 12 percent of Brahmins. Sources and series: piketty.pse.ens.fr/ideology.

It is particularly interesting to note that in the 1960s (and probably also in the 1950s, although the absence of postelection surveys prior to 1962 makes it impossible to be precise), the Congress Party enjoyed substantially higher support among high-caste voters, especially Brahmins, who were more likely to vote for INC than were the other high castes (Kshatriyas, Rajputs, Banyas, and so on) in the 1962 and 1967 elections (Fig. 16.9). This reflects the fact that in the early decades of the Indian Republic, INC was a quasi-hegemonic party, which captured a very high proportion of the vote—around 40–50 percent on average, among all social groups, including local elites and in particular Brahmins, the caste from which the Nehru-Gandhi family sprang and which played a key role in organizing the party at the local level both before and after independence.¹⁴³ In the 1960s, Congress was still doing nearly as well among Brahmins as among Muslims and lower-caste Hindus. Since then, the profile of INC vote has been totally transformed. Support from the high castes fell away in the 1970s and 1980s and even more between 1990 and 2010 as the upper-caste vote was captured by the BJP. In the 2014 elections, the structure of the Congress vote no longer bore any relation to that of the 1960s: Muslim and lower-caste voters continued to place their confidence in INC, but support fell off rapidly in the higher levels of the caste hierarchy.

To sum up: over the past half century, India gradually moved from a system in which one party, Congress, was quasi-hegemonic, owing to its role in achieving independence (which won it the support of all social classes), to a “classist” party system, in which the Hindu nationalists of the BJP receive a disproportionate share of upper-caste votes while Congress and the parties of the left capture the bulk of the lower-caste vote. In other words, while the classist system has been disappearing in the Western democracies, which are increasingly characterized by systems of multiple elites (a “Brahmin left” that captures the votes of the highly educated and a “merchant right” that appeals to the well paid and wealthy), a classist system has emerged in India as the higher castes (Brahmins, warriors, and merchants) have quit Congress and joined the BJP.

The Difficult Emergence of Classist Cleavages in India

One key point remains to be clarified, however. Can the electoral cleavages observed in India really be described as “classist,” or should they rather be called “casteist”—that is, more closely related to caste and religious identity than to socioeconomic characteristics? There is no simple answer to this question, in part because there is a high degree of correlation among the variables involved and in part because the available data are not adequate to yield a clearer answer.

Consider the correlation among the key variables. Remember that the high castes were on average more highly educated than the rest of the population and also had higher incomes and greater wealth. In particular, individuals who identified themselves as Brahmins, who were already more educated and owned more property in the colonial era, remained at the top of all three hierarchies in the late twentieth and early twenty-first centuries. Other high-caste individuals were distinctly less educated but almost always high in the income and wealth distributions. By contrast, Muslims on average remain fairly low in all three dimensions, scarcely higher than the SC/ST, while the OBC fall in between those two groups and the high castes.144 In other words, the caste hierarchy used to represent the electoral cleavages in Figs. 16.6–16.8 roughly corresponds to the socioeconomic hierarchy based on educational level, income, and wealth.¹⁴⁵

Nevertheless, while the two hierarchies overlap on average, they do not perfectly coincide at the individual level. In other words, there are many high-caste voters, including Brahmins, who are less educated and have lower incomes and less wealth than many OBC, Muslim, or SC/ST voters. Note, too, that the correlations among the three dimensions of social inequality vary from state to state (for example, the proportion of high-caste individuals is higher in northern India than in southern India), and the politicization of caste and class also varies widely from region to region. To clarify things, the most natural way to proceed is to introduce control variables so that we can reason in terms of “other things being equal.” Unfortunately, Indian postelection surveys do not contain variables that would allow us to correctly measure income and wealth (on a comparable basis over time). If we introduce controls for state, age, sex, education, and urban size, we obtain the following results.

FIG. 16.11.  The BJP vote among the high castes, 1962–2014

Interpretation: In the period 1962–2014, high-caste voters (FC, or forward castes) always voted more heavily than others for the BJP (and allies) before and after application of control variables. The caste effect (after controls) seems to have increased over time. Sources and series: piketty.pse.ens.fr/ideology.

Looking first at the BJP vote among the high castes (relative to other voters), we find that introducing controls somewhat reduces the magnitude of the “high-caste effect.” It remains quite strong, however, and even increases over time (Fig. 16.11). We obtain a comparable result when we look at the BJP vote among the lower castes (SC/ST) relative to other voters (Fig. 16.12). Finally, the religious cleavage between Hindus (of all castes) and Muslims is barely diminished when we introduce controls; what is more, it strongly increases over time (Fig. 16.13).

It is hard to say how these results might change if we had better socioeconomic control variables (especially for income and wealth). Clearly, the religious cleavage would remain, which is not very surprising, given the BJP’s very antagonistic attitude toward Muslims. Given the small effect of introducing controls (other than for region), it seems likely that the caste effect would also remain quite pronounced. The fact that caste can have an effect on voting independent of socioeconomic characteristics is in any case not very surprising given the importance of caste-based quotas in the Indian debate. If redistribution in India were based primarily on income and wealth—for example, by using taxes and money transfers dependent on those variables—or if preferential admission to universities and public employment depended on parental income or family wealth (rather than on caste as such), then it would be more surprising that caste remains the principal determinant of political cleavages. But since India makes limited use of social redistribution policies based on income and wealth, and quotas play a key role in structuring political conflict, the fact that politicization depends more on caste than on class should not be surprising.

FIG. 16.12.  The BJP vote among the lower castes, 1962–2014

Interpretation: In the period 1962–2014, low-caste voters (SC/ST, scheduled castes/scheduled tribes) were always less likely than others to vote for the BJP (and allies) both before and after application of controls. Sources and series: piketty.pse.ens.fr/ideology.

FIG. 16.13.  The BJP and the religious cleavage in India, 1962–2014

Interpretation: In the period 1962–2014, Hindu voters (of all castes: SC/ST, OBC, and FC) were always more likely than Muslim voters to vote for the BJP (and allies) before and after application of controls. The size of the religious cleavage has significantly increased over time. Sources and series: piketty.pse.ens.fr/ideology.

On the Perception of a Common Fate Among the Disadvantaged Classes

These results are particularly instructive for a Western observer because they show that electoral cleavages are historically and political constructed. They depend on the mobilization strategies of the parties and on the tools available for social redistribution. They are not fixed for all time and may evolve as political and ideological constructs change in complex ways. Note, too, that in contrast to what we find in recent decades in Europe and the United States where the white working class tends not to vote for the same parties as the Muslim or black minorities, in India lower-caste Hindus vote for the same parties as the Muslim minority (namely, Congress and the parties of the left). This again is valuable information because it shows that racism and Islamophobia are no more natural among the disadvantaged classes than among the elite. These attitudes are historically and socially constructed and depend on the tools available for building social solidarity and on the mobilization strategies of all parties.

In the Indian case, if low-caste Hindus and Muslims vote for the same parties, it is not just because both groups see themselves as targets of the high castes and Hindu nationalists of the BJP. It is also because the quota system has created de facto solidarity between the OBC and Muslims. The new quotas implemented in 1990 in favor of the OBC had important consequences. Recall that the original postwar “reservations” system covered only the former untouchables and aboriginal tribes (SC/ST). The constitution of 1950 did anticipate its extension to “OBC,” but the issue was so explosive that it was not until 1990 that the new quotas were actually implemented following the proposals of the Mandal Commission (1978–1980).146 The key point is that, unlike the quotas benefiting the SC/ST from which Muslims were excluded, the 1990 “reservations” favoring the OBC applied to both disadvantaged Hindus and disadvantaged Muslims. A system was established to gauge the living conditions and material needs of various social groups, based mainly on employment, housing, assets, and land (and not religion). An identical income ceiling was set for all groups: anyone who made more than that was ineligible for quota benefits.¹⁴⁷

The new quotas were fiercely opposed by the higher castes, which feared, not without reason, that precious places would be taken from their children. The BJP was particularly hostile to the new system, which not only disadvantaged the children of its electorate but also awarded some of their places to the reviled Muslim minority. By contrast, among the disadvantaged, the new system played a key role in fostering a community of interests and fates between Hindus and Muslims, which came together in defense of the system. Several political parties emerged to defend the rights of the lower castes (SC/ST and Hindu and Muslim OBC) against the historical high-caste monopoly of India’s top posts. I am thinking in particular of the lower-caste party BSP, whose name is usually translated as “party of the majoritarian society.” Created in 1984 to defend the interests of the disadvantaged and denounce the privileges of the upper classes, the BSP—led by the charismatic Kumari Mayawati, the first women descended from the former untouchables (SC) to head a regional government in India—formed an alliance with the socialist SP (Samajwadi Party) in the 1993 regional elections to oust the BJP from power in Uttar Pradesh. These electoral clashes have continued from the 1990s to the present and have had a significant impact on the country.¹⁴⁸

Whatever the limitations of a political program based on “reservations,” and despite the sometimes chaotic nature of the coalitions associated with the spate of new parties, the fact remains that the emergence of parties representing the lower castes in the period 1990–2020 has played a decisive role in the politicization of inequality and mobilization of the disadvantaged classes. In a sense, just as the New Deal program of public works and social insurance helped to create a community of interests between disadvantaged blacks and whites in the United States (at least for a time), the reservations in favor of the OBC have created solidarity between disadvantaged Hindus and Muslims in India.

Classist Cleavages, Identitarian Cleavages: The Social-Nativist Trap in India

To what extent will classist cleavages and redistributive issues shape Indian democracy in the decades ahead? Although it is obviously impossible to foresee how things will evolve, various hypotheses are possible. There are contradictory forces at work. Several factors tend to reinforce identitarian cleavages. The fact that quota systems play so central a role in Indian politics is problematic. Quotas of course have their place in a broader set of social and fiscal policies, but by themselves they are not enough. Furthermore, “reservations” can lead to endless conflicts over the boundaries of subcastes and jatis, which may perpetuate and exacerbate identity conflicts.

In recent decades, moreover, the BJP has deliberately tried to deepen the religious cleavage and stir up hostility against Muslims. Having tried in vain to block the quotas in favor of the OBC in the 1990s, the BJP gradually changed its strategy in the 2000s and 2010s. Aware that it could not win a majority solely by appealing to high-caste voters, the party embarked on a campaign to win over disadvantaged Hindus. This strategy took concrete form when Narendra Modi became party leader. Modi was the first BJP leader to come from the OBC rather than from one of the high castes. Under his leadership the party won the 2014 elections. If one looks at the evolution of the vote structure, it is striking to see how much the BJP and its allies increased their share of the SC/ST and OBC vote in that election (Fig. 16.8). In effect, the BJP successfully split the lower-caste Hindu vote off from the Muslim vote. The split was not as severe in some states, such as Uttar Pradesh, where the lower-caste parties successfully unified the disadvantaged electorate, but the strategy clearly worked in many northern Indian states, including Modi’s home state of Gujarat (Fig. 16.14).¹⁴⁹

FIG. 16.14.  The BJP vote by caste, religion, and state in India, 1996–2016

Interpretation: In all Indian states, the BJP (and allies) did better among FC voters (forward castes, high castes) than among OBC (other backward classes, intermediate castes), SC/ST (scheduled castes/scheduled tribes, lower castes), and Muslims. Note: The results indicated here show the average of regional election results from 1996 to 2016. Sources and series: piketty.pse.ens.fr/ideology.

The BJP’s strategy of wooing the lower-caste Hindu vote rests on several pillars.150 Modi of course has emphasized his modest background as a humble tea seller in Gujarat (which borders on Pakistan). He joined the RSS at the age of eight. The Congress Party, he insisted, was not only run by a dynasty of the privileged but also incapable of defending India from both its domestic enemies (Muslims) and its foreign enemy (Pakistan). In this connection, it is important to remember that the partition of Pakistan and the subsequent exchange of Hindu and Muslim populations has left deep scars.¹⁵¹ The conflict is in a sense still ongoing in the state of Jammu and Kashmir, whose ties to India are challenged by Muslim separatists who, according to India, use Pakistan as a rear base for preparing acts of terrorism. During the anti-Muslim riots that Modi and the BJP fomented in Gujarat in 2002—the most violent disturbances India had seen since 1947—tracts were circulated accusing the Muslim population, quite implausibly, of preparing an insurrection in case of a Pakistani invasion.¹⁵² India has been permanently traumatized, moreover, by the Muslim terror attacks in Delhi in 2000–2001 and in Mumbai in 2008–2009 (carried out by Pakistani and Indian commandos).¹⁵³

The idea that India’s Muslim population of some 180 million—some of whom had ancestors who became Muslims as long ago as the eleventh century—somehow bears responsibility for these terror attacks (or that they were actively preparing for a Pakistani invasion) of course makes no sense, any more than the repeated allegations that Congress and the parties of the left are colluding with Islamist jihadis. But in a context where everyone is seeking explanations of truly traumatic events, it is unfortunately quite common to look for accomplices and scapegoats.¹⁵⁴ In such a climate, it is hardly surprising that the attack by Muslim separatists on Indian police in Pulwama (Jammu and Kashmir) a few months before the 2019 elections, followed by air raids on camps in Pakistan (an ideal opportunity for Modi to show his strength), had a decisive impact on the campaign to the advantage of the BJP.¹⁵⁵

Note that the political issues raised by the exacerbation of the religious cleavage do not end with violence and rioting. In two states—Gujarat and Maharastra—where the BJP has held power, laws were passed to put pressure on Muslims (and to a lesser extent on Christians and Buddhists). The government tightened regulations on the slaughter of animals (extending the religious prohibition from cows to all cattle, violations of which have become a pretext for recurrent lynchings) and on religious conversion (Hindu nationalists argued that the old rules were too lax, leading to abuse by Muslim and Christian missionaries; young Muslims were accused of practicing “love jihad” by seducing gullible Hindu girls). What is ultimately at stake in these controversies is quite clearly the definition of who does and does not belong to the national community. Since 2014, BJP and RSS officials have made many statements showing that their goal is to challenge the very existence of secularism and multiconfessionalism in India, despite guarantees written into the 1950 constitution (which thus far the BJP has been unable to amend for lack of the required two-thirds majority).¹⁵⁶ School and university curricula are being revised to present the entire history of India in an exclusively Hindu and anti-Islamic light. This is a debate about the very boundaries of the community, and in this case the frontier is internal: Hindu nationalists are arguing that only some members of the community are legitimate, and the rest must either submit or leave.

Boundary conflicts such as these have taken other forms in other contexts. In the United States in the nineteenth century there was talk of sending blacks back to Africa.¹⁵⁷ Then, from 1865 to 1965, the solution was segregation, designed to keep blacks out of white space. Latinos with US citizenship were expelled in veritable pogroms in the 1930s, and children born to undocumented immigrants are threatened today. In Europe, debate has focused on the rules for acquiring citizenship, the legitimacy of past naturalizations, and even the possibility of stripping the nationality of undesirable immigrants and their children and then deporting them. The issues and circumstances vary from case to case, but all illustrate the way in which conflicts over the boundaries of the community come to take precedence over issues of ownership and redistribution, which presuppose agreement on the contours of the community.

The Future of the Classist Cleavage and of Redistribution in India: Intersecting Influences

Despite the forces conspiring to deepen identitarian and religious cleavages, it is important to note that other, no less powerful forces are pushing in the opposite direction. First, the BJP’s pro-market, pro-business economic strategy, which was supposed to strengthen India’s international position, has led in practice to an extremely inegalitarian distribution of the fruits of growth. The BJP thus finds itself facing the same type of dilemma as Trump and the Republicans in the United States. Since much of the electorate is deriving little benefit from globalization and pro-business policies, one option open to these parties is to turn up the identitarian rhetoric, be it anti-Muslim or anti-Latino, and indeed they have done so. But availing themselves of this option affords other parties the opportunity to propose more appealing alternatives. In the Indian context, it is interesting to note that in the 2019 campaign, the Congress Party proposed a basic income policy: the Nyuntam Aay Yojana (NYAY).158 The amount proposed was 6,000 rupees per month per household, the equivalent of about 250 euros in purchasing power parity (or one-third that amount at the current exchange rate), which in India is a considerable sum since the median monthly income is less than 400 euros per household. This basic income would go to the poorest 20 percent of the population. The cost would be significant (a little over 1 percent of GDP) but not prohibitive.

In any case, this proposal deserves credit for highlighting the need for redistribution and taking a step beyond the old system of quotas and “reservations,” which despite having allowed some lower-caste individuals to attend university, find public jobs, and seek elective office, is by itself not enough. By design, measures like the basic income also encourage disadvantaged groups—whatever their origin or religion—to think of themselves as sharing a common fate. Like other basic income proposals, however, this one should not be regarded as a miracle solution or panacea. Public spending on health in India has stagnated in recent years, and spending on education has actually decreased (as a percentage of GDP).¹⁵⁹ These are precisely the areas in which India has fallen behind China, which has somehow found the resources necessary to improve education and health, on which future development depends.¹⁶⁰ It is important to strike the right balance between reducing poverty and investing in growth-enhancing social programs.

Another important shortcoming of the Congress Party proposal is that it says very little about how it should be financed. What is particularly unfortunate is that this could have been an opportunity for INC to rehabilitate progressive taxation and finally turn the page on its neoliberal phase. In the 1980s and 1990s, the Congress government, under the influence of Reagan and Margaret Thatcher so common in those years, chose to sharply reduce the progressivity of the income tax, which contributed to skyrocketing inequality.¹⁶¹ No doubt afraid of vicious attacks from the BJP and its business backers if it proposed higher taxes on top earners and large firms, Congress pretended that the measure could be financed by growth alone, with no new taxes. As comprehensible as this choice was, it undermined the credibility of the proposal and would have limited INC’s ability to invest more in health and education had it won, as during its previous stints in government.

Furthermore, without potent measures of fiscal and social justice for which demand in India is high, a more explicit alliance between INC and the parties of the left (such as the CPI, SP, and BSP) seems unlikely. Yet in view of the evolution of voting patterns and electorates in recent decades, such an alliance seems natural as a way of countering the BJP and its allies. In particular, the Congress electorate has become quite similar to the electorate of the parties of the left, in contrast to that of the BJP (Figs. 16.8–16.10). It is also interesting to note that the new alliance (Gathbandan) forged during the campaign between the socialist and lower-caste parties (SP and BSP) resulted in a joint proposal calling for the creation of India’s first federal wealth tax, which would bring in roughly enough revenue to pay for the NYAY (basic income) proposed by Congress.¹⁶²

Given the climate of heightened worries about national security in which the campaign unfolded, which benefited the BJP, and owing to the weakness of the other parties (Congress and left), which did not really form a coalition, the Hindu nationalists were reelected in 2019.¹⁶³ Debate on these issues will continue in the years ahead. The outcome will become increasingly important to the rest of the world, partly because of India’s growing share of the global economy and partly because the debate revolves around issues of identity and inequality similar to those that are roiling the Western democracies. Two key differences stand out, however, and these are particularly instructive for other countries. First, although the importance of the classist cleavage has declined in the West since 1980, it has increased in India since 1990.¹⁶⁴ Second, while the white working class has parted company with black and Muslim minorities in Western democracies, in India the Hindu and minority Muslim classes vote for the same parties.¹⁶⁵ Several future trajectories are conceivable, ranging from intensification of conflicts of identity, religion, and inequality to the advent of a secularist redistributive coalition. Whatever choice Indians make and whatever new balance of power emerges among India’s parties will have an impact well beyond India’s borders.

Note, too, that India’s affirmative action “reservations” system, enshrined in the 1950 constitution, is itself being reshaped and redefined. It was initially intended to afford greater upward social mobility to the lower castes (untouchables and aboriginal tribes). The goal was to use quotas to attenuate the persistent effects of a very oppressive heritage of inequality from the old caste society and its rigidification under the British. In 1990, the system was extended to “OBC,” and in 1993 a parental income threshold was introduced for deciding who was eligible under the quota system (regardless of class or caste origin). This threshold was extended to the former untouchables and aborigines (SC/ST) in 2018. The BJP, finding itself unable to reduce the quotas open to the lower classes as much as it wished, passed a law in 2019 establishing new quotas for high-caste individuals (including Brahmins), whose parental income fell below the new threshold, at the expense of high-caste individuals with incomes above that level.¹⁶⁶ Interestingly, the BJP decided to do this because a large part of its electorate consisted of impoverished upper-caste individuals whose socioeconomic status and educational level were too low to derive full benefit from the country’s economic growth. It was adopted by a nearly unanimous vote of the Lok Sabha. These developments suggest that in the future the quota system will likely be transformed from a system based on caste and jati to one based more on parental income, wealth, and other socioeconomic criteria.

At a time when Western societies are questioning the low representation of the disadvantaged in the most selective educational institutions, legislative bodies, and top political and administrative posts, India’s evolving ways of dealing with such problems deserve close scrutiny, not to idealize or condemn them but to learn from them.¹⁶⁷ To be sure, there is no substitute for adequately financed educational and health services open to all coupled with ambitious policies to reduce income inequality and redistribute wealth. Still, it is possible to justify ways of compensating for social origins in student admissions and selection procedures in conjunction with these other policies.¹⁶⁸

Conversely, the political-ideological evolutions that take place in Europe and the United States will have a decisive impact on India’s future trajectory as well. I have already alluded to the effect of the conservative revolution of the 1980s in the United States and United Kingdom on tax policy in the rest of the world, including India. There will be more of the same in the future. Today, when the SP and BSP propose a progressive wealth tax as a way of paying for the Congress Party’s basic income proposal, the BJP can easily argue that these are socialist fantasies that no other country is currently applying and that India’s prosperity depends above all on stability of the social order and the property regime. If Europe were to turn in earnest toward social federalism, or if the United States were to return to the steeply progressive tax system it so successfully applied in the past (as more and more Democrats are advocating), then the debate in India and elsewhere would likely take a different turn. By the same token, if the rich countries continue to undercut one another in a race to the bottom on taxes, it will be that much more difficult for a coalition like that of the SP-BSP to persuade the Indian public, given the business community’s strong opposition to the wealth tax and its role in financing the parties and the media. In that case, the BJP will likely take an even harder line toward Muslims. The world’s various inequality regimes are more intimately related than ever before.

The Incomplete Politicization of Inequality in Brazil

In India we have just seen an example of a democracy in which a post-independence party system evolved in recent decades in a classist direction—the opposite of what we found in our study of Western democracies. Obviously, we cannot study the transformation of cleavage structures in all non-Western postcolonial societies. This would take us far beyond the scope of this book. Brazil is nevertheless an interesting case, in which we once again see a classist party system emerging in the period 1989–2018 with important consequences for redistribution and significant interactions with other parts of the world.

Recall that Brazil was the last country in the Euro-Atlantic world to abolish slavery in 1888. Thereafter the country remained one of the most inegalitarian in the world. Not until the end of the military dictatorship (1964–1985) and the constitution of 1988 was the right to vote extended to all with no educational restriction.169 The first presidential election under universal suffrage took place in 1989, and the former union worker Luiz Inácio Lula da Silva, backed by the Workers’ Party (PT), qualified for the second round, in which he took 47 percent of the vote. In 2002 he was triumphantly elected with 61 percent of the second-round vote, and in 2006 he was reelected by the same margin despite having been mercilessly mocked by traditional Brazilian elites for his lack of education. Although some said that he was not worthy of representing the country abroad, Lula’s election marked the beginning of the era of universal suffrage in Brazil. After Dilma Rousseff replaced Lula as the PT candidate, the party continued to win elections, although with shrinking margins (56 percent in 2010, 52 percent in 2014). Finally, in 2018, the nationalist conservative Jair Bolsonaro won with 55 percent of the second-round vote versus 45 percent for PT candidate Fernando Haddad, marking a new turn in Brazil’s political history.¹⁷⁰

It is interesting to observe that the structure of the PT electorate and of the Brazilian party system more generally developed gradually over the three decades following the end of the dictatorship. At its inception in the 1980s, the PT did best among industrial workers, lower- to middle-class urban employees, and intellectuals who had mobilized against the dictatorship.¹⁷¹ Because the lowest levels of education and income were found primarily in rural areas and poor regions, the PT electorate in the 1990s had a slightly higher level of education than the national average (but a slightly lower average income). In short, when the military dictatorship ended, as when independence came in India, the structure of the vote was not inherently classist. Only after Lula came to power did the social composition of the PT clearly change. In the 2006, 2010, 2014, and 2018 elections, PT always did best among less educated and lower-income voters (Fig. 16.15).¹⁷² The evolution was equally dramatic at the regional level. The poorest regions of the country, particularly in the northeast, voted more and more heavily for the PT while the wealthier regions gradually turned away from the party. In the 2014 and 2018 elections the Nordeste continued to deliver large majorities for Rousseff and Haddad while the south (including São Paulo) decisively rejected the party. This social and geographic cleavage coincided with a very pronounced racial cleavage. After 2006 we find that voters identifying as black or mixed race (a little more than half of the population) were much more likely to support the PT than those who identified as white, even after controlling for other socioeconomic characteristics.¹⁷³

The fact that the PT vote evolved in this direction is consistent with the policies the party pursued while in power. From 2002 on, PT governments focused their efforts on reducing poverty, specifically through the social transfer program known as Bolsa Familia. The income of Brazilians in lower-income groups, particularly in the poorest regions of the country, rose sharply, which made Bolsa Familia and PT very popular among farm workers, poor peasants, domestic servants, low-paid service and construction workers, and so on. By contrast, among the employers of such workers, these social programs were often seen as very costly and as incitements to disruptive wage demands. PT governments also significantly increased the minimum wage, whose real value had fallen sharply under the dictatorship but which was restored in the 2010s to the level of the 1950s and early 1960s.¹⁷⁴ The PT also developed programs to offer preferential access to universities for disadvantaged blacks and persons of mixed race, previously underrepresented on university campuses.

FIG. 16.15.  The politicization of inequality in Brazil, 1989–2018

Interpretation: In the period 1989–2018, voting for the PT (workers’ party) in Brazil became more and more clearly associated with lower levels of income and education, which was not the case in the first elections after the end of the military dictatorship. Sources and series: piketty.pse.ens.fr/ideology.

There is little dispute that these redistributive policies, in conjunction with the widening of the class cleavage, drove traditional Brazilian elites to seek ways to regain control of the situation. This desire led to the impeachment and removal of Rousseff in 2016 and then to the election of Bolsonaro in 2018. Bolsonaro portrays himself as the leader who will put an end to the country’s drift toward socialism. He does not hide his sympathy for the military dictatorship or his zeal for restoring the social order, respecting property rights, and maintaining security. Like Trump, he relies on the exploitation of racial cleavages and nostalgia for white supremacy in a country in which “whites” are officially no longer in the majority.175 Despite this, it is clear that the erosion of popularity that comes naturally with holding power in any democracy also played a role in the PT’s demise, as did the obvious limitations of the policies it pursued between 2002 and 2016. The party failed to make a serious assault on Brazil’s corruption problem, to which it contributed by accepting secret payments in a country where the financing of political parties and the media has never been adequately regulated. To be sure, these deficiencies were related to the fact that Brazil’s electoral system and institutions make it very difficult to form a parliamentary majority. Despite repeated massive presidential victories, with substantially more than 50 percent of the second-round vote from 2002 to 2010, the PT could never muster a majority of deputies in support of its policies. It had to bargain with other political groupings to pass laws and approve its budgets.¹⁷⁶ Nor did the PT ever clearly explain the need for greater transparency in public life and for campaign finance reform so that it gave the impression of accommodating to the existing system, with all its flaws.

Note, too, that the PT’s record in reducing inequality had its pluses and minuses. Although people at the lower end of the income spectrum clearly benefited from its policies with an increase in the share of national income going to the bottom 50 percent between 2002 and 2015, this improvement came entirely at the expense of the middle class, or more precisely, of those situated between the bottom 50 percent and the top 10 percent of the income distribution; those in the top 10 percent lost nothing and were able to maintain their position (which in Brazil’s case is an unusually powerful one). Indeed, the top 1 percent of the income distribution increased its share of national income between 2002–2015: it remained twice as large as the share of the bottom 50 percent.¹⁷⁷ These disappointing and paradoxical results are easy to explain: the PT never attempted real fiscal reform. The middle class, not the wealthy, paid for its social policies, for the simple reason that the PT never attacked the country’s regressive tax structure: Brazil has high indirect and consumption taxes (of up to 30 percent on electric bills, for example) while for historical reasons there is little in the way of progressive taxation of income or wealth (for instance, the tax on the largest inheritances is capped at 4 percent).

Once again, these policy shortcomings stem not only from doctrinal and ideological limitations but also from the absence of an adequate parliamentary majority. Whether in Brazil, Europe, or the United States, it is impossible to reduce inequality as much as one would like without also transforming the political, institutional, and electoral regimes. Note, too, as in India, the importance of outside influences. Quite clearly, it would have been much easier for Lula and the PT to forge ahead with progressive taxes on income and wealth had such tax policies enjoyed widespread international approval, as they might in the future.¹⁷⁸ On the other hand, the current race to the bottom among countries engaged in tax competition with one another might reinforce the inegalitarian and identitarian orientation of Bolsonaro and his nationalist conservative backers, as it has encouraged Modi and the BJP in India.

Identity and Class Cleavages: Borders and Property

The Brazilian case, like the Indian, shows why it is essential to look beyond the West to understand the political dynamics associated with inequality and redistribution. In the period 1990–2020, as the left-right classist cleavage that had prevailed in Europe and the United States from 1950 to 1980 was crumbling and on the verge of collapse, other types of classist cleavage were emerging in India and Brazil, each following its own specific sociopolitical path and exhibiting its own specific fragilities and potentialities. The variations among these different trajectories show that political and ideological conflict is fundamentally multidimensional.

In every case we have studied, we can clearly distinguish two types of cleavage: one identitarian, the other classist. The identitarian cleavage revolves around the question of borders—that is, the boundaries of the political community with which one identifies and the ethno-religious origins and identities of its members. The classist cleavage revolves around issues of socioeconomic inequality and redistribution and especially the issue of wealth. These cleavages take different forms in Europe and the United States, India and China, Brazil and South Africa, and Russia and the Middle East. But we find both dimensions in most societies, usually with multiple ramifications and subdimensions.

Broadly speaking, the classist cleavage can win out only if the identitarian cleavage can be overcome. In order for political conflict to revolve around inequalities of wealth, income, and education, there must first be agreement about the boundaries of the political community. But the identitarian cleavage is not simply invented by politicians who seek to instrumentalize it to gain power (even if it is easy to identify such politicians in all societies).

The boundary question is fundamental and complex. In a global economy in which different societies are linked by flows of many kinds—commercial, financial, migratory, and cultural—but continue to function as separate political communities, at least in part, it is crucial to describe how those societies dynamically interact. In the postcolonial world, groups of human beings who had never previously had much contact (other than through war or colonial domination) have begun to mix and interact within the confines of a single society. This marks an important step forward for human civilization, but it has also given rise to new identity cleavages.

At the same time, the collapse of communism has at least temporarily stifled hopes for achieving a just economy and transcending capitalism by striving for greater social and fiscal justice. In other words, as the identity cleavage deepened, the class cleavage receded. This is surely the main reason for the rise of inequality since the 1980s. Technological and economic explanations miss the crucial point, which is that economic and property relations can always be organized in more than one way, as exemplified by the extraordinary political and ideological diversity of the inequality regimes we have studied in this book.

We find the same pattern in virtually every region of the world: the identity cleavage deepened and conflicts over boundaries intensified while the wealth cleavage weakened and criticism of wealth became muted. Yet although the pattern may be the same, the variations from society to society remain significant. No deterministic explanation can account for such diversity; what matters are social and political mobilization strategies. Here, the long-term comparative perspective is essential. Inequality regimes were extensively transformed well before the two world wars of the twentieth century. To argue that similar shocks are necessary before inequality can again be so dramatically reduced is to read the past in a very conservative and misleading fashion. The cases of India and Brazil show that identity cleavages need not take precedence over class cleavages. In both countries, disadvantaged classes were able to overcome differences of origin and identity in order to join forces in political coalitions built around seeking more redistributive policies. Everything depends on equipping groups of different origins and identities with the institutional, social, and political tools they need to recognize that what unites them outweighs what divides them.

Studying other national electoral patterns would yield further confirmation of this general fact.179 The case of Israel is probably the most extreme example of an electoral democracy in which identity conflict has taken precedence over everything else. The relation of the Jewish population to the Palestinian and Israeli Arab populations has become virtually the only significant political issue. In the period 1950–1980, the Israeli Labor Party dominated the party system; one of its main goals were to reduce socioeconomic inequality and foster novel cooperative practices. But because it was unable to come up with a viable political solution acceptable to all constituent communities, which would have required either the creation of a Palestinian state or a novel form of binational federal government, Labor has all but disappeared from the Israeli political scene, leaving more security-minded factions to outbid one another in an endless round of escalation after escalation.¹⁸⁰ In the Muslim countries, the religious and social dimensions of electoral conflict have combined in different ways at different times. In Turkey in the period 1950–1970, the Kemalist Republican People’s Party (CHP) was both more secular and more popular among lower-class voters. Those voters differed with more religious voters over agrarian reform and redistribution of land to poor peasants, measures opposed not only by landlords but also by groups determined to protect land held by religious organizations (along with the social role those organizations played). In the period 1990–2010, the Party of Justice and Development (AKP) won over a substantial share of lower-class voters with a program of Muslim and nationalist revival while the CHP vote shifted more toward the cities.¹⁸¹ In Indonesia, agrarian reform played a similar but more lasting role.¹⁸² I alluded earlier to the absence of land reform in South Africa, where the existence of a hegemonic post-apartheid party has complicated the emergence of any type of class cleavage.¹⁸³ By bringing all these cases together and looking closely at many different historical experiences, we gain a better idea of the complex interactions between cleavages based on income and wealth and cleavages based on ethno-religious identity. Looking beyond the West, we find that historical trajectories varied widely with respect to these two dimensions.

Still, despite all these cultural, national, and regional differences, the global ideological context should not be neglected. We saw this in the cases of India and Brazil: the ability of local political forces to promote credible redistribution strategies and give voice to class differences depended in important ways on ideological evolutions in the Western countries.¹⁸⁴ Given the economic, commercial, and financial weight of the United States and the European Union and their decisive influence on the legal framework within which the global economy operates, the political-ideological transformations under way in both regions will be crucial. What happens in China and India and in the medium run in Brazil, Indonesia, or Nigeria will also play a growing role as the world’s ideologies become increasingly interconnected. What is certain is that the influence of ideology is not about to decrease—quite the contrary. Questioning of the property regime and the system of borders has never been more intense. Uncertainty about how to respond to recent changes has never been greater. We live in an era that wants to see itself as postideological but is in reality saturated in ideology. Nevertheless, I am convinced that the history recounted in this book can serve as the basis for new thinking about participatory internationalist socialism. The past can teach us how to restructure property regimes and borders to move us closer to a just society and quell the identitarian menace. I will explore these ideas further in the final chapter of the book.

The Dead Ends and Pitfalls of the Populism Debate

Before I do that, however, I need to clarify one bit of terminology. In this book I have deliberated avoided invoking the notion of “populism.” The reason is simple: the concept is not useful for correctly analyzing the evolutions currently under way. The political-ideological conflicts we see in various parts of the world are profoundly multidimensional. In particular, they involve both cleavages over borders and cleavages over the property regime. “Populism,” a word used ad nauseam in public debate, mixes everything up in one indigestible stew.

All too often, the notion is instrumentalized by political actors to designate whatever they do not like and from which they wish to dissociate themselves. It is taken for granted that any party that is anti-immigrant or hostile to foreigners should be regarded as “populist.” But parties that call for higher taxes on the wealthy are also characterized as “populist.” And if a party dares to suggest that public debt need not be repaid in full, then clearly it deserves to be called “populist” as well. In practice, the term “populism” has become the ultimate weapon in the hands of the objectively privileged social classes, a means to dismiss out of hand any criticism of their preferred political choices and policies. Gone is the need for any debate about novel social and fiscal arrangements or alternative ways of organizing globalization. It is enough to brand dissenters as “populists” to end all discussion with a clear conscience and foreclose debate. In France, for example, it has become commonplace since the 2017 presidential election to classify voters who opted for either Jean-Luc Mélenchon or Marine Le Pen in the first round as “populists,” neglecting the fact that Mélenchon voters were on average the most open to immigration of all voters while Le Pen voters were the most fiercely hostile.185 In the United States, it was not uncommon in 2016 to classify both the internationalist socialist Bernie Sanders and the nativist businessman Donald Trump as “populists.” In India, one might characterize as “populist” both Modi’s anti-Muslim BJP and the socialist, communist, and lower-caste parties that take a diametrically opposite tack. In Brazil, the label “populist” has been applied to both the authoritarian conservative Bolsonaro movement and the workers’ party of former president Lula.

To my mind, the term “populism” should be strictly avoided because it does not help us to understand the world. In particular, it is silent about the multidimensionality of political conflict and about the fact that attitudes toward wealth and borders can diverge sharply. It is important to distinguish carefully among the various dimensions of political conflict and to analyze carefully and rigorously the political and institutional responses on offer. The debate about populism is vacuous: it licenses a lack of precision. A low point was reached in the debate about public debt during the Eurozone crisis. Any politician, demonstrator, or citizen who dared to suggest that sovereign debt might not be fully and immediately repaid became straightaway the target of pundits: no more “populist” idea was imaginable.

Yet all these enlightened pundits appeared to be almost totally ignorant of the history of public debt, not least the fact that debt had been canceled many times over the centuries and particularly in the twentieth century, often with success. Debt in excess of 200 percent of GDP weighed on any number of countries in 1945–1950, including Germany, Japan, France, and most other countries of Europe, yet it was eliminated within a few years by a combination of one-time taxes on private capital, outright repudiation, rescheduling, and inflation. Europe was built in the 1950s by wiping away past debt, thereby allowing countries to turn their attention to the younger generation and invest in the future. Of course, every situation is different, and we must seek new solutions to resolve our current debt problems in constructive ways, drawing on the successes of the past and circumventing their limitations, as I have tried to show. But for critics who know virtually nothing about history to dismiss as “populist” those who seek to launch a necessary and unavoidable debate is simply unacceptable. To be sure, the leaders of the Lega and M5S in Italy and of the “Yellow Vests” in France who have called for referenda on debt cancellation may not fully appreciate the complexity of the issue, which cannot be settled by a simple “yes” or “no.” There is an urgent need for debate on the fiscal, financial, and institutional arrangements necessary to reschedule the debt because it is “details” like these that determine whether debt reduction comes at the expense of the wealthy (by way of a progressive wealth tax, for example) or of the poor (by way of inflation). The social demand to do something about the debt may be confused, but it is also legitimate, and the response should be not to shut down debate but rather to open it up in all its complexity.

To conclude, note that the worst consequence of the populism debate may be that it fosters new identity conflicts and impedes constructive deliberation. Although the term is usually used pejoratively, it is sometimes taken up by those accused of being populists as a badge of identity. This clouds the issue even further because the positive use of the term is just as nebulous as the negative one. For instance, some anti-immigrant movements brandish the term “populist” to show that they take the side of “the people” (who are assumed to be unanimous in their hostility to immigration) against the “elite” (who are said to favor open borders everywhere). Some would-be “radical” left movements (such as Podemos in Spain and LFI in France) have also taken up the term “populist,” not always prudently, to set themselves apart from other “left” (socialist or social-democratic) parties, which they accuse of having betrayed the working class. There are better ways to make such a critique than by using a loaded, totemic, and dangerously polysemic word like “populist.” In practice, the term pits “the people” against “the elite” (financial, political, or media as the case may be) while avoiding any discussion of what institutions (e.g., at the European level) are needed to alleviate the condition of the disadvantaged. At times “populism” is used to deny the importance of ideology: the implicit assumption is that pure force is all that matters and that institutional details can be settled once “the people” have asserted their strength and carried the day.¹⁸⁶

The history of all the inequality regimes studied in this book proves the opposite. Historical changes of great magnitude occur when the logic of events comes together with short-term mobilizations and longer-term institutional and intellectual changes. In the late nineteenth and early twentieth centuries, the People’s Party in the United States played a useful role not because it laid claim to the name “populist” (which in itself was neither necessary nor sufficient) but because it was part of a fundamental political and ideological shift that culminated in the Sixteenth Amendment to the US Constitution and the creation of the federal income tax in 1913—a tax that would become one of the most progressive in history and make it possible to finance the New Deal and reduce inequality.

For all these reasons, I think it is important to be wary of the dead ends and pitfalls of the “populism” debate and focus instead on content and, specifically, on new thinking about the property regime; fiscal, social, and educational systems; and the organization of borders. In other words, we should be thinking about the social, fiscal, and political institutions that can help to create a just society and allow class cleavages once again to take priority over identity cleavages.

---

1.     1.  For a detailed analysis of the results for Germany and Sweden, see the online appendix (piketty.pse.ens.fr/ideology) and F. Kosse and T. Piketty, “Changing Socioeconomic and Electoral Cleavages in Germany and Sweden 1949–2017,” WID.world, 2019.

2.     2.  See the online appendix, Fig. S16.1.

3.     3.  As in Germany and France, I have included in Fig. 16.1 not only the main social-democratic or labor party (the SAP in Sweden and the Arbeiderpartiet in Norway) but also other parties of the left (socialists, communists, etc.) and ecologists. If one focuses exclusively on the SAP, one still finds an extremely clear reversal of the educational cleavage. See the online appendix, Fig. S16.1.

4.     4.  On the Swedish case, see Chap. 5.

5.     5.  For a comparative analysis of the results obtained in twenty-one countries, see the online appendix and A. Gethin, C. Martinez-Toledano, and T. Piketty, “Political Cleavages and Inequality. Evidence from Electoral Democracies, 1950–2018,” WID.world, 2019.

6.     6.  In particular, unlike the vote profile as a function of education, which reversed in nearly all countries, the profile as a function of income remained decreasing (though less and less so), giving rise to a system of multiple elites of the “Brahmin left” versus “merchant right” type. See the online appendix, Fig. S16.2.

7.     7.  The very similar evolutions of the structure of electoral cleavages in so many different countries shows that certain common factors are ultimately more important than the national specificities that are often invoked. But as important as this general scheme is, it should not be allowed to mask the many national and regional political-ideological variants that contribute to the basic trend. For example, variants of electoral behavior linked to family structure can explain many local variations without undermining the basic similarity. On the links between family structure and political ideology, see these two classic studies: H. Le Bras and E. Todd, L’invention de la France (Hachette, 1981); E. Todd, L’invention de l’Europe (Seuil, 1990); L’origine des systèmes familiaux (Gallimard, 2011).

8.     8.  See A. Gethin, “Cleavages Structures and Distributive Politics” (master’s thesis, Paris School of Economics, June 2018), pp. 89–100. See also K. Mori McElwain, “Party System Institutionalization in Japan,” in Party System Institutionalization in Asia, ed. A. Hicken and E. Martinez Kuhonta (Cambridge University Press, 2015), pp. 74–107.

9.     9.  In The Silent Cry (Kodansha International, 1967 [Japan]; English trans. J. Bester, Kodansha International, 1981), Kenzaburo Oe magnificently evokes the complexity and violence of relations between the intellectual elites and the poorer classes in Japan, particularly in relation to the urban-rural cleavage, traditional values, and the question of modernization of the country since the beginning of the Meiji era (1868), to say nothing of the role played by the island nation’s geopolitical position, its relation to the United States, and the antagonisms aroused by the presence of Korean laborers.

10.   10.  See Chaps. 10–11.

11.   11.  Figs. 16.1–16.2 and Figs. 14.8–14.9.

12.   12.  See Chap. 10.

13.   13.  See Chap. 11.

14.   14.  See Fig. 12.10.

15.   15.  Lew Rywin, a well-known Polish film producer, was convicted of attempting to extort millions of euros from Agora, the leading Polish media company. Invoking the name of the incumbent prime minister, Rywin tried to get Agora to pay him to back an amendment to a law governing radio and television broadcasting.

16.   16.  In the 2007, 2011, and 2015 elections, the SLD captured less than 10 percent of the vote, compared with 30–40 percent each for the PO and PiS.

17.   17.  For a detailed presentation of these results, see A. Lindner, F. Novokmet, T. Piketty, and T. Zawisza, “Political Conflict and Electoral Cleavages in Central-Eastern Europe, 1992–2018,” WID.world, 2019.

18.   18.  Remember that since 2015 the president of the European Council has been Donald Tusk, who served as prime minister of Poland and leader of PO from 2007 to 2014. See Chap. 12.

19.   19.  The 2015 refugee crisis arose after civil war broke out in Syria in 2012–2014, sending nearly 1 million Syrians fleeing to Europe (an influx that amounted to 0.2 percent of the EU population of 510 million). Most of these ended up in Germany. EU member states then decided to stanch the plow by signing an agreement with Turkey in 2016; in return for financial aid, Turkey agreed to keep most refugees fleeing Syria in camps on its soil. We will see later that the flow of migrants entering the European Union has actually decreased since the economic crisis of 2008. The wide media coverage of “columns of immigrants” crossing the Balkans to reach Germany and Northern Europe—coverage that was intensely exploited by political parties—had a profound impact on the way the crisis was seen.

20.   20.  For an analysis of the available materials, see Lindner et al., “Political Conflict and Electoral Cleavages in Central-Eastern Europe.”

21.   21.  Family benefits and other pro-natalist social and fiscal policies reduce inequality (a system providing equal allotments for each child is more of a boost for low-income families than for higher-income families) while also expressing the need to increase the country’s population without bringing in immigrants. PiS’s and Fidesz’s family policy since 2015 can be compared with the bonus of $10,000 per child (starting with the second child) introduced in Russia in 2007, which apparently had a significant effect on the birth rate. See E. Yakovlev and I. Sorvachev, “Short-Run and Long-Run Effects of Sizable Conditional Child Subsidy” (presentation, Paris School of Economics, Applied Economics Lunch Seminar, September 18, 2018).

22.   22.  In practice, this did not prevent Fidesz from exercising lingering liberal-authoritarian instincts in response to economic issues. For instance, in late 2018, the government enacted a law reinforcing the power of employers to force workers to work overtime.

23.   23.  The coalition led by Fidesz (which included the Christian Democratic People’s Party, or KDNP) won 53, 45, and 49 percent of the vote in the elections of 2010, 2014, and 2018 respectively versus 24, 19, and 12 percent for the MSZP (whereas the two blocs had been close to equal at 41–43 percent each in 2002 and 2006). These two groups are flanked on the right by Jobbik (with about 20 percent of the vote in the most recent elections). Jobbik is even more violently opposed than Fidesz to black and Muslim immigrants (even though they are totally absent in Hungary).

24.   24.  See Fig. 14.2. For detailed results on educational cleavages in Italian elections, see the online appendix and Gethin et al., “Political Cleavages and Inequality.”

25.   25.  The remainder of the vote (7 percent) went to small parties independent of the three major polls. In the 2016 elections, M5S received 26 percent of the vote, the PD 30 percent, and the coalition of the right 29 percent.

26.   26.  This compromise was never very coherent because the Lega proposed cutting the taxes needed to pay for the universal basic income. I will come back to this point.

27.   27.  For instance, during a joint press conference in Milan in August 2018, the Hungarian prime minister declared: “We proved that immigration could be stopped on land, he is proving that it can be stopped on the sea.” The Italian minister of the interior responded: “Today begins a common journey that will be followed by numerous other stages in the coming months to focus on the right to work, health, and security. Everything that the European elites refuse to give us.” C. Ducourtieux, J.-B. Chastand, and M. Nasi, “Migrants: Viktor Orban et Matteo Salvini prennent Emmanuel Macron pour cible,” Le Monde, August 29, 2018. The European elections of 2019 also confirmed that the nativist wing of the Italian coalition was on the verge of gaining the upper hand over M5S.

28.   28.  See Table 14.1 and Fig. 14.19.

29.   29.  I am referring here to the “egalitarian internationalist” quarter of the 2017 presidential electorate, which includes not only LFI but the other parties of the left (although most of the votes went to the LFI candidate for tactical reasons related to the possibility of access to the second round). See Table 14.1.

30.   30.  The same is true of the party leadership, which has always stood for welcoming immigrants.

31.   31.  The series 1992, an Italian political drama broadcast in 2015, offers an illuminating and instructive view of that key year in Italian political history and helps us to understand the lengthy process of political decomposition. We meet Leonardo Notte, a dishonest but likable press agent who assists in Berlusconi’s rise. We also meet Pietro Bosco, a battered veteran of the Gulf War, who somewhat by accident finds himself elected to parliament on the Lega Nord ticket, in a milieu of old Roman politicians and their wheeling and dealing. Had the drama unfolded twenty-five years later, Bosco might have wound up in the M5S instead of the Lega.

32.   32.  Libya had been plunged into chaos in 2008 by the joint Franco-British invasion, which can be compared for both lack of preparation and devastating consequences to the Anglo-American invasion of Iraq in 2003 (although the human cost of the latter invasion was significantly larger).

33.   33.  See the online appendix. The official position of the French government, first under François Hollande and then Macron, was that, although the so-called Dublin Accord (under which any demand for political asylum must be examined in the country where the applicant first set foot on European soil) needed to be reformed, reform was unfortunately impossible because Poland and Hungary refused to agree to it, and therefore France would continue to send refugees back to Italy. If Germany had taken the same position in 2015, it would not have admitted more than a million refugees.

34.   34.  As noted earlier, the Democrat John Calhoun made similar accusations of hypocrisy against the industrial and financial elites of the North, whom he charged with calling for the abolition of slavery only to secure a cheap, readily exploitable, and easily discardable work force for their factories. See Chap. 6.

35.   35.  See Chap. 12.

36.   36.  For instance, the federal troops that occupied the South at the end of the Civil War could easily have imposed desegregation. In fact, they tried (too shortly), and it is not difficult to imagine a sequence of events and individual actions that might have ended in success. Similarly, Republicans might have won the White House since the 1960s without resorting to the racial attacks that Richard Nixon, Reagan, and Trump used so effectively, if only they had searched for more ambitious political and ideological alternatives.

37.   37.  See Chap. 14.

38.   38.  See Chaps. 10–11.

39.   39.  See the online appendix.

40.   40.  See Chaps. 11, 13, and 17.

41.   41.  A more optimistic view is the following: the disintegration of the European Union and the demise of its rules regarding budgets, finance, and competition could revive the old left-right competition. The left bloc would once again enjoy some room to maneuver and could propose new social and ecological policies, while the right bloc would be free to pursue a pro-business and anti-immigration agenda. This is the implicit hypothesis underlying B. Amable and S. Palombarini, L’illusion du bloc bourgeois. Alliances sociales et avenir du modèle français (Raisons d’agir, 2017). But there is reason to believe that the return to the nation-state would lead instead to intensified interstate competition, of which the primary beneficiaries would be the nativist and nationalist parties.

42.   42.  The corporate tax rate in the United States was 45–50 percent from the 1940s to the 1980s before dropping to 34 percent in 1988–1992 and then rising to 35 percent in 1993–2017 (to which one must add state taxes of 5–10 percent). Until 2018, the United States had resisted the race to the bottom on corporate taxes that began in Europe. See Chap. 11. The sudden cut to 21 percent risks restarting the race to the bottom.

43.   43.  See E. Saez and G. Zucman, The Triumph of Injustice: How the Rich Dodge Taxes and How to Make Them Pay (Norton, 2019). These conclusions follow from strict application of the tax code without even considering the various methods of tax avoidance employed by the very rich.

44.   44.  In practice, the new trade agreement negotiated with Mexico and Canada in 2018 is a carbon copy of the previous North American Free Trade Agreement (NAFTA), with just enough minor symbolic differences to lend plausibility to the idea that something has changed, such as the addition of a clause marginally increasing the percentage of automobile parts manufactured by workers paid more than $16 an hour, with a sanction of just 2–4 percent in customs duties in case of violation. In financial terms, the measure is totally insignificant compared with the corporate tax cuts approved in 2017.

45.   45.  Just as Reagan’s election in 1980 can be linked in mind and spirit to the book that Milton Friedman published in 1963 with Anna Schwartz on the monetary history of the United States (the bible of monetarist economics; see Chap. 12), Trump’s election in 2016 can be linked to a deeply Trumpian book that Samuel Huntington published (The Clash of Civilizations and the Remaking of the World Order [Simon and Schuster, 1996]). In sum, Huntington proposed that the ideological conflict between capitalism and communism would give way to a war of cultures and identities, which he took to be fixed, ahistorical essences (the “West” versus Islam, Hinduism, etc.).

46.   46.  Trump’s anti-Brahminism is also clear in fiscal measures targeting university presidents (who are paid too much compared with “job creators,” according to Trump) and taxing tuition rebates for graduate students as wages (a measure that ultimately did not pass). By contrast, Trump’s demands that nations benefiting from US military protection pay for services rendered is a clear throwback to the old warrior class and the military tributes it received in the premodern trifunctional order.

47.   47.  Note than in Italy, M5S and the Lega have also agreed on anti-vaccination legislation as punishment for allegedly know-it-all elites (and rapacious pharmaceutical companies). PiS in Poland and BJP in India regularly accuse scholars of abusing the Polish or Indian nation by questioning established truths. Bolsonaro in Brazil has engaged in similar criticism of scholars.

48.   48.  This cleavage is in reality rather artificial in light of the French government’s actual immigration and climate policies.

49.   49.  As shown, for example, by the fact that tax receipts from the French wealth tax and the declared tax base both increased significantly from 1990 to 2018, despite fiscal competition and lax enforcement. See Fig. S14.20.

50.   50.  As shown by measures like the Foreign Account Tax Compliance Act (2010), which put pressure on Swiss banks to reveal hidden assets, or Elizabeth Warren’s proposed wealth tax (with an exit tax of 40 percent on individuals who renounce US citizenship and try to take their wealth abroad). See Chap. 11.

51.   51.  See also Chaps. 11 and 12. The reference to “ever closer union” is taken from the first sentence of the preamble to the Treaty on the Function of the European Union, adopted in Lisbon in 2007 along with the Treaty on European Union. These two texts, which came into effect only gradually between 2009 and 2014, constitute the current legal foundation of the European Union. They were supplemented in 2012–2013 by the budget treaty (TSCG), which set new deficit rules, and by the European Stability Mechanism (ESM). The treaties adopted in Lisbon in 2007 are essentially the same as those rejected in France in the 2005 referendum on the European Constitutional Treaty (Chap. 14). They were simply cleaned up: the term “constitution” was eliminated, the principle of “free and undistorted competition” in the old preamble was replaced by “fairness in competition,” and most importantly, the whole thing was ratified by parliament rather than by referendum. For links to these various texts, which are worth consulting, see the online appendix. See also D. Chalmers, G. Davies, and G. Monti, European Union Law: Text and Materials (Cambridge University Press, 2014).

52.   52.  Technically, the legislative function is exercised by the council of ministers in the relevant area (along with the European Parliament), while the European Council (consisting of heads of state and government under the authority of a President of the European Council who is named by them) focuses on the broad orientation of policy and treaty reform. But because ministers generally act under the authority of their heads of government, the difference is more legal and practical than truly political.

53.   53.  The unanimity rule also applies in the areas of foreign policy and common security, police cooperation, admission of new member states, European citizenship, and so on.

54.   54.  The qualified majority rule is as follows: a decision is adopted if it is supported by 55 percent of the countries representing at least 65 percent of the population of the European Union. This rule, adopted after lengthy debate, is the principal innovation of the Treaty on European Union adopted in Lisbon in 2007 (and before that in the now-defunct European Constitutional Treaty). It has been in effect since 2014. Previously, a system based on assigning a certain number of votes to each country was used. This was reviewed periodically and was the subject of constant controversy.

55.   55.  Although the term “Eurogroup” is not mentioned in any of the treaties, it has come to designate the council of finance ministers of the Eurozone, a body that took on an important role in the aftermath of the financial crisis of 2008–2009.

56.   56.  Remember that the decision inopportunely taken behind closed doors by the Eurogroup after the financial crisis of 2008–2009 explains why Europe absurdly fell back into recession in 2011–2012. See Chap. 12 and the online appendix, Figs. S12.12a–S12.12c. The poor performance of the Eurogroup shows how ill suited it was to the job.

57.   57.  On this subject see S. Laurens, Les courtiers du capitalisme. Milieux d’affaires et bureaucrates à Bruxelles (Agone, 2015).

58.   58.  To date, not even the most federalist proposals for European reform have gone that far. In particular, the treaty instituting the European Union that the European Parliament adopted in 1984 (the so-called Spinelli plan) accorded an important role to the Parliament, which was to have the power to appoint and remove Commissioners and to examine and amend laws and directives proposed by the Commission while leaving intact the requirement that such laws and directives must also be approved in identical terms by the council of ministers of member states (possibly by qualified majority vote). The optimistic federalist hypothesis (still entertained today) is that the council would ultimately bow to majority decisions by the European Parliament even though it would retain its formal right of veto over parliamentary decisions.

59.   59.  In particular, it makes no sense to compare the US Senate to the European Council. The equivalent of the European Council would be a senate composed of the governors of the individual states, where two states, say California and New York, accounted for half of the country’s GDP, which is roughly the situation of Germany and France in the Eurozone. Such a system would probably function very poorly, and the two governors would probably meet often without coming to any agreement. The comparison sometimes made between the Bundesrat (which represents the German Länder) is also not very convincing. Note that US senators have been directly elected by universal suffrage only since passage of the Seventeenth Amendment in 1913. Previously they were chosen by state legislatures.

60.   60.  The members of the Rajya Sabha are chosen from lists presented by the parties and are not themselves members of the state legislatures.

61.   61.  The Lok Sabha has 545 members, and the Rajya Sabha, 245. In practice, the joint session procedure has been used only three times since the adoption of the Constitution of 1950, including once in 1963 to pass the law prohibiting dowries.

62.   62.  See the “Manifesto for the Democratization of Europe,” published in December 2018 and available at M. Bouju, L. Chancel, A.-L. Delatte, S. Hennette, T. Piketty, G. Sacriste, and A. Vauchez, “Manifeste pour la Démocratisation de l’Europe,” website, www.tdem.eu. See also M. Bouju, L. Chancel, A. L. Delatte, S. Hennette, T. Piketty, G. Sacriste, and A. Vauchez, Changer l’Europe, c’est possible! (Seuil, 2019).

63.   63.  The draft of this treaty, the Treaty on the Democratization of Europe (T-Dem), is also available at www.tdem.eu.

64.   64.  Broadly speaking, there is nothing to prevent countries that wish to sign bilateral or multilateral treaties from doing so while respecting their other commitments. Because the fiscal competences discussed here are not EU competences, the Democratization Treaty could be adopted without violating existing rules. For a legal analysis of these issues, see S. Hennette, T. Piketty, G. Sacriste, and A. Vauchez, How to Democratize Europe (Harvard University Press, 2019). Note that the so-called Elysée treaty, providing for bilateral cooperation between France and Germany, was renewed in 2019, at which time a Franco-German parliamentary assembly consisting of one hundred members drawn from the French Assemblée Nationale and the German Bundestag was created. For now, this assembly is purely consultative, but there is nothing to prevent it from being granted the kinds of decision-making competences discussed here.

65.   65.  The proposed taxes include a 15 percent tax on corporate profits (together with a 22 percent minimum on corporate taxes at the national level, for a total of 37 percent); a common tax on high incomes of 10 percent on incomes above 200,000 euros and 20 percent above 400,000 euros (on top of the 40–50 percent currently applied at the national level, for a total of 60–70 percent for the highest incomes); a common tax on large fortunes of 1 percent above 1 million euros and 2 percent above 5 euros million (on top of existing property, land, and other national wealth taxes, which could be complemented by a common tax on inheritances of 10 percent above 1 million euros and 20 percent above 2 million euros); and a common carbon tax (with an initial price of 30 euros per ton, to be reevaluated annually). All details are available at www.tem.eu. These proposals are meant simply to orient thinking about what kind of budget a European Assembly might adopt; they should in no sense be taken as representing an ideal of fiscal progressivity on high incomes and large fortunes (about which I will say more in the next chapter; see Tab. 17.1).

66.   66.  For example, the European Assembly could also decide simply to return all revenues to its member states, in which case the arrangement would simply allow member states to tax their most powerful economic actors more effectively at the federal level to reduce the fiscal pressure on the lower and middle classes, which would already be a fine achievement.

67.   67.  See Fig. 14.20 and Fig. 15.18.

68.   68.  The hypothetical possibility of amending the constitutions of European countries to prevent them from leaving the European Union or repudiating various European and international treaties seems unrealistic for the foreseeable future. It would arouse fierce and probably irresistible opposition in Germany, France, and elsewhere. In the United States, the US Constitution did not allow the South to secede, but that did not stop them from trying. In Europe, the only conceivable treaties at this point are those based on voluntary and reversible decisions of the member states. Obviously, this does not mean that things will always be this way.

69.   69.  Presumably, each political party would send deputies well versed in these matters to the European Assembly. The EA would meet less often than national parliaments (perhaps one week a month) and could schedule its sessions so as to conflict as little as possible with national parliamentary sessions.

70.   70.  The fact that the European Parliament, prior to its election by direct universal suffrage in 1979 when its role was purely consultative, consisted of representatives of national parliaments probably explains the reluctance of many of the most convinced federalists (particularly among MEPs) to approve of involving the national parliaments in a federalist project. The proposal I present here clearly goes in this direction, however, because the proposed European Assembly consists in part of national deputies and has the last word on European budgets and taxes (which are the most important elements of federal sovereignty). Hence the situation would be totally different from that which existed before 1979. Concertation with the European Council is envisioned, but in case of disagreement it is the Assembly that decides. See the Treaty on the democratization of Europe (www.tdem.eu), article 8.

71.   71.  The approach I take here builds European sovereignty on national political institutions but not on national governments (as has been done until now). Rather, I look to national parliaments (which represent the full range of opinions and allow for deliberation and decision by majority vote). Joschka Fischer gave a speech at Humboldt University in 2000 based on similar premises but received little in the way of reply from the French government at the time.

72.   72.  See the online appendix.

73.   73.  In particular, Nigel Farage, the leader of UK Independence Party, spent the entire campaign totaling up the amounts transferred to Europe that could have been used to pay for the National Health Service. Of course, he seriously inflated the numbers to make his point.

74.   74.  Treaty on the democratization of Europe (www.tdem.eu), article 9.

75.   75.  In late 2009, the Greek government announced that its deficit was 12.5 percent of GDP and not 3.7 percent as previously stated. See Chalmers, Davies, and Monti, European Union Law, pp. 704–753, for a narrative of events and European responses.

76.   76.  See Fig. 12.10 and the online appendix, Fig. S12.10.

77.   77.  Note that private flows are partly considered already in the sense that contributions to the European budget (and annual balances) are calculated on the basis of gross national income (GNI), which is equal to GDP corrected by net income flows to and from other countries.

78.   78.  For example, a raise to 0.5 or 1 percent or even higher if agreement could be reached.

79.   79.  This is obviously less true outside the Eurozone: if the countries of Eastern Europe are included, then significant transfers and investment flows must be factored in.

80.   80.  Bouju et al., “Manifesto for the Democratization of Europe,” article 10.

81.   81.  In the German proposal for a “redemption fund” (a rather moralistic name), the goal was to reduce total outstanding debt to 60 percent of GDP over a period of twenty to thirty years at a rate (and therefore a primary budget surplus) to be decided in advance. But it is neither realistic nor wise to lock such a decision in without regard to changing economic circumstances.

82.   82.  According to UN demographic and migration data, the net flow of migrants into Europe (minus returns) reached 1.4 million a year between 2000 and 2010 and then fell to 0.7 million between 2010 and 2018 despite the refugee spike in 2015. In the United States, which recovered more quickly than Europe from the 2008 recession, the flow remained stable (1 million a year from 2000 to 2010, 0.9 million from 2010 to 2018). See the online appendix, Fig. S16.4. On average, migratory flows into the rich countries were barely 0.2–0.3 percent a year in the period 2000–2020. What is new is that these flows are occurring in a context of demographic stagnation: the annual birth rate is below 1 percent of the population in many rich countries, which means that an inflow of 0.2–0.3 percent a year can over time noticeably alter the composition of the population. Recent experience shows that this can lead to political exploitation of identity differences, especially if policies to promote job creation, new housing, and necessary infrastructure are absent.

83.   83.  Except in countries where the debt is “significantly lower than 60 percent of GDP,” where the deficit can go as high as 1 percent. See TSCG, Article 3.

84.   84.  The best proof that the new budgetary rules are not taken seriously is that many European leaders continue to refer to “the 3 percent rule,” apparently unaware that the new deficit target is 0.5 percent. This also illustrates the urgent need to bring these issues back within the realm of democratic debate.

85.   85.  See Chap. 12 and the online appendix, Fig. S12.12c. For good measure, in July 2015 Greece was threatened with expulsion from the Eurozone if it refused these conditions (despite the absence of any legal grounds for such a threat).

86.   86.  The hostility aroused by Syriza was due in part to the clumsiness of its leaders, who at times during the 2015 crisis gave the impression of seeking special exemptions from the rules for the benefit of Greece alone, whereas resolving the Eurozone’s public debt crisis actually called for a global solution applicable to Italy, Portugal, and other countries as well as the establishment of parliamentary bodies in which the influence of each country (including Greece) would have been limited. Nevertheless, all signs are that if European (and especially French and German) leaders had proposed a global response aimed at achieving social justice, Syriza’s leaders would have been the first to embrace it. Ultimately, the hostility aroused by Syriza attests to a more general ideological climate of postcommunism and very deep conservatism on economic and financial matters. In particular, East European leaders were often the most hostile to Syriza, which they suspected of retailing socialist-communist promises similar to those for which they had paid the price in the past (in a very different context). They also resented the arrogance that the older member of the European club displayed toward the newcomers (identity conflicts are always close to the surface in Europe). In this respect, the fact that Angela Merkel was born in East Germany was not incidental to her conservatism on these issues (coupled with significant historical amnesia, in that Germans conveniently forgot the debt cancellations from which they benefited in the 1950s) as well as to her willingness to open the gates to Syrian political refugees.

87.   87.  Small countries like Greece and Portugal can find themselves forced to accept high primary surplus targets (currently 3–4 percent) lest they find themselves in an even worse situation if forced out of the Eurozone (in view of the small size of their domestic markets and their vital need of European developmental assistance). In a country like Italy, any attempt to impose too high a primary surplus would probably give rise to irresistible internal pressure to exit the European Union. By definition, a primary budget balance means that a country can cover its expenses with its tax revenues and has no need of financial markets (reinforcing the temptation of autarky).

88.   88.  The ultimate circumvention of democracy was probably the 2012 treaty establishing the European Stability Mechanism (ESM). Loans from the ESM are conditional upon signing memoranda with the ECB, the Commission, and the International Monetary Fund regarding reforms to be undertaken by the country receiving the loan. These reforms can involve all kinds of issues (including health, education, pensions, and taxes), all without parliamentary oversight or public deliberation. See Chalmers, Davies, and Monti, European Union Law, pp. 741–753.

89.   89.  There is no need to formally amend the TSCG because the rules it established would not apply to the common bond that would be issued by the states signing the T-Dem democratization treaty.

90.   90.  The T-Dem proposal also provides for increased oversight of the ECB and the European Monetary System, including public hearings and confirmation of nominations, compared with what currently exists. See Bouju et al., “Manifesto for the Democratization of Europe,” articles 12–17.

91.   91.  With primary budget balance, the stock of debt increases at the rate of interest (since interest is paid by issuing new debt) while GDP increases at the nominal growth rate (the sum of real growth plus inflation). Hence the ratio between the two decreases if the nominal growth rate is higher than the rate of interest. However, if two are equivalent (say 2 percent a year), then the stock of debt does not decrease as a percentage of GDP.

92.   92.  Note that such a rule would in any case be less constraining than the 0.5 percent secondary deficit required by the TSCG.

93.   93.  Today’s low long-term interest rates are also a result of new prudential rules (which for excellent reasons require private financial institutions to hold large amounts of the safest public debt—though without reducing their unprecedentedly large private balance sheets) especially because there are few financial instruments as safe as US and European public debt (which can give a lasting advantage to these countries in a situation in which the share of the global economy deemed less safe is growing and, with it, the amount of savings in search of safe investment opportunities).

94.   94.  For an interesting project that would allow willing EU member states to set up a Bank for Climate and Biodiversity (in conjunction with the European Investment Bank and the ECB), see the proposed “Treaty Establishing a Union for Climate and Biodiversity,” made public in 2019 under the auspices of the so-called Finance-Climate Pact: Pacte Finance-Climat, “Treaty Establishing a Union for Climate and Biodiversity” (2019), https://www.pacte-climat.net/wp-content/uploads/2019/03/Treaty_Finance_Climate_Pact.pdf. More broadly, it bears emphasizing that the T-Dem was built primarily around the issue of fiscal justice and in particular around the idea of a European Assembly able to levy common taxes to finance a balanced budget. Many issues (particularly financial and banking issues) are barely touched on. The goal is not to end discussion but to open it up around a specific project, which can then be amended and extended with other ideas such as those contained in the Finance-Climate Pact (which, by contrast, may put too much emphasis on banking and borrowing and not enough on fiscal justice and democratization).

95.   95.  That said, the challenges of transnational democracy are themselves real and unprecedented. If primary budget balance is a necessary condition for moving in the direction of mutualized interest on European public under the democratic supervision of a European Assembly, this would mark a clear improvement over the current situation.

96.   96.  See Chap. 10.

97.   97.  In practice, rescheduling debt with low (or near-zero) interest and nonindexing to real growth or inflation leads to a significant decrease in the value of the debt in relation to GDP within a few decades, which drains the drama from the issues raised by partial or total cancellation of the debt. For example, this is what happened with the German foreign debt, which was frozen in 1953 by the London conference and ultimately canceled in 1991. See Chap. 10. The creditors in that case were a consortium of Western countries and banks. In the proposal outlined here, it would be mainly the ECB and certain special-purpose entities (such as the ESM).

98.   98.  For example, it is quite possible that some East European countries (where broad segments of the public reject the conservative-nativist evolution of their homelands as indicated by the high rate of abstention, especially in Poland) and Nordic countries would want to join the EPU before countries like Luxembourg and Ireland, whose governments have invested all their political capital since the 1990s in explaining that their future lies in fiscal dumping.

99.   99.  For an analysis of this point, see Hennette et al., How to Democratize Europe, pp. 46–52. The TSCG and ESM treaties were designed as simple intergovernmental treaties signed outside the general framework conceived for the revision of the European treaties, and some states refused to participate (notably the United Kingdom and the Czech Republic), which did not stop the CJEU from approving the method.

100. 100.  On this system, which has always been used to allocate taxable profits among the states of the United States and could be applied by the United States or European countries to other countries, see Saez and Zucman, The Triumph of Injustice. The larger the core group of the EPU, the more likely it will be to secure significant international cooperation. But the important point is that any country could on its own tax profits fictitiously recorded in noncooperating countries.

101. 101.  For example, in 2010, the United States did not need to withdraw the licenses of Swiss banks operating in the United States. The passage of a law providing for such sanctions in case of noncooperation was enough to induce the Swiss to change their banking laws to allow information about accounts held by US citizens to be transmitted automatically to US tax authorities.

102. 102.  In particular, apportionment of taxable profits on the basis of sales is equivalent to imposing tariffs on exports of goods and service, except that the tariffs in question are proportional to profits earned at the global level (a firm that makes no profit would not be taxed).

103. 103.  The process could have been initiated when the EEC consisted of just six countries (France, Germany, Italy, Belgium, the Netherlands, and Luxembourg) from 1957 to 1972. No doubt these countries were preoccupied with other challenges, such as decolonization and the reconstruction of their political systems on the basis of a viable left-right competition at the national level.

104. 104.  As I indicated earlier, it is possible to meet the concern about high tax levels by returning to member states all or part of the revenues from common taxes, mainly to finance tax cuts for the lower and middle classes. If need be, this could be included in the treaty; although this is not the most desirable option, it would still represent a major improvement over the status quo.

105. 105.  Another factor accounting for the slowness of the movement toward a European Union with a parliament allowed to levy taxes is related to the fact that among the very highly educated elite (and especially among economists) a certain distrust of elected assemblies has emerged, coupled with a growing attraction to governance by unelected committees and rules decided by such committees.

106. 106.  The strategy is often formulated as Plan A versus Plan B: Plan A calls for agreement on certain treaty changes while Plan B threatens exit or disobedience in case of failure of Plan A, with the ultimate goal of forging a new agreement with a smaller number of countries. Various versions of the Plan A/Plan B approach were developed in the wake of the 2015 Greek crisis and the threats of the German government (and particularly its finance minister) to exclude Greece from the Eurozone to force the Greek government to accede to Germany’s wishes.

107. 107.  To be clear, I am not saying that it would be enough for a French government to be elected on a platform of remaking the European Union to impose its wishes on other countries. By contrast, it is clear that a French politician who simply declares his desire to renegotiate certain European treaties, as the Socialist candidate did in the 2012 presidential campaign without offering the slightest indication of what he wished to obtain, will not be in a very strong position to get what he wants once elected.

108. 108.  See the online appendix, Figs. S16.5–S16.6. See also A. Gethin, C. Martinez-Toledano, and M. Morgan, “Rising Inequalities and Political Cleavages in Spain,” WID.world, 2019.

109. 109.  All told, 57 percent of Catalan voters in 2008 favored the idea of greater regional autonomy (potentially including self-determination), compared with 61 percent in 2011 and 59 percent in 2016. Between 2008 and 2016, the proportion of those favoring increased regional autonomy within Spain (but without self-determination) decreased while the percentage favoring self-determination increased, even though Catalan obtained greater autonomy in 2010 (as we will see). Note that the questions posed in the survey referred to a regime applicable to all Spanish regions. If one looks at answers to questions specifically dealing with Catalonia, support for self-determination by Catalan voters is ten points higher (up to 45–50 percent in 2017–2018).

110. 110.  In the September 2017 independence referendum, “yes” won by 90 percent versus 8 percent for “no” and 2 percent blank, but the participation rate was only 42 percent.

111. 111.  See Fig. 14.20 and Fig. 15.18.

112. 112.  In 2018, the federal government taxed income at rates of 9.5 percent (for incomes below 12,450 euros) to 22.5 percent (for incomes above 60,000 euros). If a region decides to apply these same rates, then taxpayers in that region would pay total income taxes of 19 to 45 percent, with revenues divided fifty-fifty between Madrid and the region. Each region can also set its own tax brackets and its own additional rates, higher or lower than federal rates. In any case, it keeps the additional revenue and does not have to share it with other regions. These new fiscal decentralization rules were put in place in 2010 for Catalonia and the other regions of Spain. On the other hand, in 2010 the Spanish constitutional court invalidated other aspects of the new autonomy statute (such as the regionalization of justice, a controversial provision) that Catalonia approved by referendum in 2006 after negotiation with the Spanish parliament, which then had a socialist majority. This episode contributed to the hardening of pro-independence sentiment.

113. 113.  This system of internal competition has also led since 2011 to tax fraud by wealthy individuals and companies, which for tax purposes misrepresent their residence or place of business, potentially undermining the progressivity of the tax system. See D. Agrawal and D. Foremny, “Relocation of the Rich: Migration in Response to Top Tax Rate Changes from Spanish Reforms,” Review of Economics and Statistics, 2019.

114. 114.  Admittedly, top marginal federal income tax rates have varied widely over time (from an average of more than 80 percent between 1930 and 1980 to around 40 percent since the 1980s), but the fact is that the federal income tax has always played a more important redistributive role than the additional income taxes enacted by the states (generally between 5 and 10 percent).

115. 115.  See, for example, the statements by Jean-Claude Juncker cited in Chap. 11.

116. 116.  Note, too, that under current European law, it is quite possible to cut taxes only on new taxpayers enticed into the country, for instance, by using the special tax regime for the “stateless,” as Denmark has recently done. See H. Kleven, C. Landais, E. Saez, and E. Schultz, “Migration and Wage Effects of Taxing Top Earners: Evidence from the Foreigners’ Tax Scheme in Denmark,” Quarterly Journal of Economics, 2014.

117. 117.  A somewhat different example of dissonance is provided by Norway, a country that likes to portray itself as social-democratic and environmentally conscious but that did not hesitate to extract hydrocarbons that should have been left in the ground (if one worries about global warming) to accumulate vast financial reserves. In the series Occupied (Netflix, 2015), guilt ultimately leads Norway to stop pumping oil, which triggers a Russian invasion to restart the flow, with the support of a European Union more concerned about its energy supply than about the climate. The EU is depicted as especially cowardly, notably in the person of a not very inspiring French European Commissioner.

118. 118.  See Chap. 11 and G. Du Rietz, M. Henrekson, and D. Waldenstrom, “Swedish Inheritance and Gift Taxation (1885–2004),” in Swedish Taxation: Developments since 1862, ed. M. Henrekson and M. Stenkula (Palgrave, 2015). The conservative-liberal government of 1991–1994 also had an impact on the educational system, where competition was encouraged. On the influence of some Swedish economists on the liberal turn of the early 1990s, see A. Lindbeck, P. Molander, T. Persson, O. Peterson, A. Sandmo, B. Swedenborg, and N. Thygesen, Turning Sweden Around (MIT Press, 1994).

119. 119.  See Fig. 10.12, and the online appendix, Figs. S10.12a–S10.12b. There was an attempt to abolish the inheritance tax in the United States under George W. Bush in 2001–2002 and then again under Trump in 2017–2018, but these attempts did not succeed because some Republicans believed they went too far. But the tax threshold was raised significantly, reducing the effectiveness of the tax.

120. 120.  In Sweden, it was the social democrats who abolished the inheritance tax in 2005 while it was the liberal conservatives who abolished the wealth tax in 2007 (they were in power from 2006 to 2014).

121. 121.  The abolitions of 2005–2007 also reflected the perception of some Swedish social-democratic leaders that the country had become so egalitarian that it would not have been useful to tax the very wealthy. They may have forgotten that the country was extremely inegalitarian until the beginning of the twentieth century and that maintaining substantial social equality over the long run requires appropriate institutions. On Sweden’s hyper-inegalitarian past, see Chap. 5.

122. 122.  The Swedish case was widely instrumentalized in the French debate on abolition of the wealth tax in 2017–2018. It may be exploited again to abolish the inheritance tax. The effect on less developed countries has been even greater because on their own they lack the means to influence the global system of registering and taxing wealth.

123. 123.  See T. Wieder, “A Berlin, le mouvement ‘Fuck Off Google’ plus fort que Google,” Le Monde, October 26, 2018.

124. 124.  One such group was the “Job Creators Network,” which supported Trump’s corporate tax cuts and “the fight against socialism” (its website features the headline “JCN’s Next Mission: Fighting Socialism”). In early 2019 the JCN financed a campaign to plaster the walls of New York with posters attacking AOC: “Amazon Pullout: 25,000 Lost NYC Jobs. $4 Billion in Lost Wages. $12 Billion in Lost Economic Activity for NY. Thanks for Nothing, AOC!” The theme of fighting socialism has also become a top priority of the White House Council of Economic Advisors (as is evident from the report “The Opportunity Costs of Socialism” that the CEA published in October 2018). This shows that the threat of socialism is taken seriously and that significant material resources have been mobilized in the ideological battle.

125. 125.  It remains to be seen whether this new rhetoric will lead to real change if the Democrats win the next election (which is not a sure thing, to judge by the record of past Democratic administrations).

126. 126.  See Chap. 8.

127. 127.  A uninominal one-round system as in the United Kingdom and the United States.

128. 128.  In 1964, the CPI split into several parties, including the CPI and the CPI(M), mainly over the issue of the Russia-China spit and political strategy (whether to ally with the INC or pursue an independent strategy). The CPI(M) has held power in a number of Indian states since the 1970s, including West Bengal and Kerala, generally as the leader of a “Left Front” or “Left Democratic Front” coalition involving several left-wing parties.

129. 129.  Faced with growing social protest of many varieties, Indira Gandhi declared a state of emergency from 1975 to 1977, which temporarily united all the malcontents against her in the 1977 elections.

130. 130.  The successive prime ministers were Jawaharlal Nehru from 1947 to 1964, followed by his daughter Indira Gandhi from 1966 to 1977 and then 1980 to 1984, and then Rajiv Gandhi, who held the post from 1984 to 1989 after his mother was assassinated by her Sikh bodyguards in 1984. Note that the Nehru-Gandhi family is not related to Mahatma Gandhi, who joined the INC between the wars and remained a member until he was assassinated by a Hindu nationalist in 1948.

131. 131.  As allies of Congress we include the parties that frequently joined coalitions with the INC, especially as part of the United Progressive Alliance (UPA): the National Congress Party, Dravida Munnetra Kazhagam, Telangana Rashtra Samithi, and Biju Janata Dal. See A. Banerjee, A. Gethin, and T. Piketty, “Growing Cleavages in India? Evidence from the Changing Structure of the Electorates, 1962–2014,” WID.world, 2019.

132. 132.  The name BJP can be translated as “party of the Hindu people” or “party of the Indian people.” Since Bharata is the traditional name of India in Sanskrit and the BJP promotes an ideology that clearly stresses India’s Hindu identity, the first translation seems preferable.

133. 133.  The BJP officially claims this title, pointing out that in 2015 its membership exceeded 110 million (see www.bjp.org), compared with about 90 million for the Chinese Communist Party (see Chap. 12).

134. 134.  RSS translates to “National Volunteer Organization.”

135. 135.  See M. S. Golwalkar, We, or Our Nationhood Defined (1939), pp. 49–50. Golwalkar’s vehement opposition to Islam and Hindu nationalism call to mind Chateaubriand’s remarks in his 1802 Génie du christianisme (see Chap. 7). The RSS and BJP have at times used anti-Christian rhetoric and participated in anti-Christian actions (notably directed against Christian missionaries and their efforts to convert certain aboriginal tribes). But in the Indian context it was naturally the rivalry with Islam, which over the centuries had attracted numerous converts from the lower castes, that played the central role (see Chap. 8). Another classic RSS theme is that Hinduism, owing to its sense of moderation, respect for the planet, and vegetarianism, is the only viable alternative to Western ideologies and particularly to the opposition between capitalism and communism.

136. 136.  The RSS, which was briefly banned in 1948 after Gandhi was assassinated by a former member of the organization, was banned again in 1992 after members of the group participated in the destruction of the mosque. It was rehabilitated the following year, however, when the courts found that direct involvement of RSS leaders in organizing the riots had not been proven. According to Hindu activists, the Babri Masjid was built in the sixteenth century on a site formerly occupied by a temple dedicated to Rama. Archeological excavation has shown that many structures were built in the neighborhood of the site without definitively settling the matter.

137. 137.  In addition to destroying mosques, the principal motives of anti-Muslim riots have to with the illegal slaughter of animals and failure to respect certain holidays. See Chap. 8.

138. 138.  See BJP, Sankalp Patra: Lok Sabha 2019 (Shahdol, India, 2019), https://www.bjp.org/en/manifesto2019, section on “Cultural Heritage.” The case is still in the courts, with new excavations under way to determine how the site might be shared between Hindus and Muslims.

139. 139.  This group includes the following parties: CPI, CPI(M), SP, BSP, Janata Dal (United), Janata Dal (Secular), Rashtriya Janata Dal, and All India Trinamool Congress. See Banerjee, Gethin, and Piketty, “Growing Cleavages in India?”

140. 140.  In practice, the two coalitions, SP-BSP and INC-allies avoid running against each other in some states and strategic districts where alliance against the BJP and its allies seems indispensable, but they have not yet come to the point of concluding an explicit national alliance.

141. 141.  For a detailed presentation of the results and the postelection surveys used, see the online appendix and Banerjee, Gethin, and Piketty, “Growing Cleavages in India?” Records of the postelection surveys have been kept from 1962 to 2014, although some files are unfortunately missing and some surveys from the 1980s and early 1990s are defective.

142. 142.  See Figs. 8.2–8.5 and Tables 8.1–8.2. Voters of other religions (Christians, Buddhists, Sikhs, etc.; roughly 5 percent of the population) voted on average in a way similar to Muslims and the low castes. The sample sizes are too small, however, to analyze their behavior separately, and they have been omitted from the analysis presented here. See Banerjee, Gethin, and Piketty, “Growing Cleavages in India?”

143. 143.  In the 1946 provincial elections organized by the British, with a censitary suffrage (under which about 20 percent of the adult population enjoyed the right to vote), Congress won 80 percent of the seats, of which nearly 50 percent were occupied by Brahmins, provoking the rage of B. R. Ambelkar. See A. Teltumbde, Republic of Caste (Navayana, 2018), p. 143. On the subject of Ambelkar, the political leader of the lower castes, who was at odds with Congress in the 1930s and 1940s, see Chap. 8.

144. 144.  See Chap. 8 and N. Bharti, “Wealth Inequality, Class and Caste in India, 1951–2012,” WID.world, 2018.

145. 145.  With one important difference: Muslim voters vote more to the left than the SC/ST, whereas they stand slightly higher in the socioeconomic hierarchy.

146. 146.  The decisive contribution of the governments led by the Janata Party in 1977–1980 and by Janata Dal in 1989–1991 was to create this commission and introduce the OBC quotas. See Chap. 8 and especially C. Jaffrelot’s analyses of democratization by caste in India. To a large extent, the disintegration of the Janata Party coalition in 1980 was a result of the clash between Mandal and Mandir: the secularist and socialist parties chose to support the process initiated by the Mandal Commission, which led to the OBC quotas, while the Hindu nationalists quit the coalition and created the BJP, with the symbolic goal of building a Hindu temple (Mandir) in Ayodhya. See S. Bayly, Caste, Society and Politics in India from the Eighteenth Century to the Modern Age (Cambridge University Press, 1999), pp. 297–300.

147. 147.  The threshold for belonging to the “creamy layer” is currently 800,000 rupees per year (which excludes about 10 percent of the population). See Chap. 8.

148. 148.  Recall that Uttar Pradesh (in northern India) is the most populous Indian state (with 210 million people in 2018) and that elections in the region are widely covered by the Indian media.

149. 149.  Broadly speaking, party dynamics vary widely at the state level, where we find variable combinations of classist and casteist cleavages. In Delhi, the INC won in 1998, 2003, and 2008 with support from the BSP and by relying on the disadvantaged and immigrant populations whereas the BJP (who won in 1993) did best among wealthy and anti-immigrant voters. See the illuminating book by S. Kumar, Changing Electoral Politics in Delhi. From Caste to Class (Sage, 2013). The citizens’ anticorruption party—the Aam Aadmi Party—largely claimed the heritage and electorate of the INC-BSP to defeat the BJP in 2013 and 2015—a victory that earned it the implacable antipathy of the federal government.

150. 150.  For an illuminating analysis of BJP strategy in the 2014 elections and afterward, see C. Jaffrelot, L’Inde de Modi. National-populisme et démocratie ethnique (Fayard, 2019).

151. 151.  On this subject, see M. J. Akbar, India: The Siege Within. Challenges to a Nation’s Unity (UBS Publishers Distributors, 1996).

152. 152.  The riots began after Hindus returning from Ayodhya (where they had demonstrated in favor of building a temple dedicated to Rama) died when their train was attacked by projectiles launched from a Muslim neighborhood. Modi, who was then chief minister of Gujarat, publicly accused the entire Muslim community of collective responsibility and implicitly urged his followers to riot. Modi’s stirring of religious conflict contributed to his regular reelection as chief minister of Gujarat from 2001 to 2014 and served as a launching platform for the 2014 federal elections. See Jaffrelot, L’Inde de Modi, pp. 69–75. See also C. Thomas, Progroms et ghetto. Les musulmans dans l’Inde contemporaine (Karthala, 2018).

153. 153.  In 2018, ten years after the Mumbai attacks, 80 percent of people polled continue to rank Islamist terror as the principal threat facing the country. Modi—unlike his predecessor, the BJP’s Atal Vajpayjee—has always refused to participate in public ceremonies celebrating the end of Ramadan, which he sees as an “appeasement strategy” comparable with Chamberlain’s way of dealing with the Nazis. In the 2014 elections, Modi compared the ballot weapon to “bows and arrows under the Moghuls” and regularly referred to Rahul Gandhi as shehzada (heir apparent of the Muslim dynasties under the Moghuls). He never misses an opportunity to fan the flames of religious antagonism or to evoke past Muslim domination of India. See Jaffrelot, L’Inde de Modi, pp. 124–143.

154. 154.  Although the situation in France is admittedly less violent, though more and more agitated, the charge that journalists from Mediapart were somehow complicit with Islamist extremism or even with the authors of the 2015 attack on Charlie Hebdo, or more generally, the routine accusations of “Islamo-leftism” leveled at any individual or political movement that defends Muslims and non-European immigrants against the xenophobic, anti-immigrant, anti-refugee right, clearly draw on the same inspiration as the strategy of the BJP.

155. 155.  The BJP rose sharply in the polls after these events.

156. 156.  In 2017, RSS leader Mohan Bhagwat defended the Hindu reconversion movement known as Ghar Wapsi (“coming back home”) as follows: “India’s Muslims are also Hindus.… We are bringing back our brothers who have gone astray. They did not leave us of their own volition. They were stolen, spirited away.… Now the thief has been caught, and everyone knows that my property is in his possession. I am going to take it back, and there is no reason to make a big thing out of this.… We need not fear. Why should we be afraid? We are not foreigners. This is our homeland. This is our country. This is the Hindu Rashtra.” See Jaffrelot, L’Inde de Modi, pp. 172–173.

157. 157.  Such talk led to the creation of Liberia, although the forced return of blacks never reached the level envisaged by its promoters. See Chap. 6.

158. 158.  Nyuntam Aay Yojana means system of guaranteed minimum income.

159. 159.  Health expenditures have stagnated at 1.3 percent of GDP on average between 2009–2013 and 2014–2018 while investment in education fell from 3.1 to 2.6 percent. See N. Bharti and L. Chancel, “Tackling Inequality in India: Is the 2019 Election Campaign Up to the Challenge?” WID.world, 2019.

160. 160.  See Chap. 8.

161. 161.  See A. Banerjee and T. Piketty, “Top Indian Incomes, 1922–2000,” World Bank Economic Review, 2005; L. Chancel and T. Piketty, “Indian Income Inequality, 1922–2015: From British Raj to Billionaire Raj?” WID.world, 2017. Income taxes on top earners were cut under both Rajiv Gandhi (1984–1989) and P. V. Narasimha Rao (1991–1996). See also J. Crabtree, The Billionaire Raj: A Journey Through India’s New Gilded Age (Tim Duggan Books, 2018).

162. 162.  The 2019 SP-BSP proposal called for a federal wealth tax of 2 percent on fortunes above 25 million rupees (roughly 1 million euros in purchasing power parity). This would affect about 0.1 percent of the Indian population (or 10 million people) and bring in revenues of roughly 1 percent of GDP. See the online appendix. Note that the introduction of a national minimum income in France in 1988 (revenu minimum d’insertion, or RMI) also coincided with the introduction of a new wealth tax (ISF), receipts from which largely covered the cost of the RMI.

163. 163.  The BJP and its allies in the National Democratic Alliance (NDA) had 336 of 545 seats in the previous Lok Sabha (281 of which belonged to the BJP); they won 352 (of which 303 belonged to the BJP) in 2019.

164. 164.  Note, too, that the lower classes in India still turn out heavily in Indian elections (sometimes more heavily than the well-to-do), in contrast to what we have found to be the case in Western democracies in recent decades. Some scholars see this as a consequence of a state so weak that the wealthy do not need to mobilize to protect themselves from it. See K. Kasara and P. Suryanarayan, “When Do the Rich Vote Less than the Poor and Why? Explaining Turnout Inequality Around the World,” American Journal of Political Science, 2015. It may also be that the new lower-caste parties such as the BSP are able to turn out the vote effectively.

165. 165.  The same is true of the Christian, Buddhist, Sikh, and other minorities, although the numbers are smaller.

166. 166.  See Chap. 8. A constitutional amendment adopted in January 2019 created an additional quota of 10 percent (above the 50 percent of places reserved for SC/ST and OBC) for groups not previously covered by quotas (meaning, in practice, the high castes), whose annual income fell below the threshold of 800,000 rupees. In practice, this income threshold (which excluded about 10 percent of the population) was linked to other criteria having to do with the size of housing and amount of land owned. These rules were supposed to be enforced with the help of a new and much more reliable system for recording income and wealth. See Bharti and Chancel, “Tackling Inequality in India.”

167. 167.  See Chap. 8 for a more detailed discussion of India’s experiments with quotas and reservations.

168. 168.  I will come back to this in the next chapter.

169. 169.  The postslavery constitution of 1891 stipulated that illiterates would not be allowed to vote, and this provision was perpetuated in the constitutions of 1934 and 1946. In 1950, more than 50 percent of the adult population still could not vote. See Chap. 6.

170. 170.  The 2018 election took place in unusual circumstances. Lula, designated as his party’s candidate, was in prison and prevented by the courts from running.

171. 171.  Lula had been a unionized worker in the industrial region of São Paulo, while Rousseff spent three years in prison under the military dictatorship before attending university.

172. 172.  As in India, the postelection surveys available in Brazil do not allow us to break down the vote by wealth. For a detailed presentation of the results for Brazil, see A. Gethin, Cleavage Structures and Distributive Politics (master’s thesis, Paris School of Economics, June 2018), pp. 29–41; A. Gethin and M. Morgan, “Brazil Divided: Hindsights on the Growing Politicization of Inequality,” WID.world, 2018.

173. 173.  See A. Gethin, “Cleavage Structures and Distributive Politics,” p. 38, fig. 3.5. Note that the education and income effects shown in Fig. 16.15 are estimated after controlling for other variables (including region and race). Without controls, the effect of belonging to the top educational or income decile would be approximately twice as high (on the order of fifteen to twenty points rather than six to ten). See the online appendix, Fig. S16.15.

174. 174.  See M. Morgan, Essays on Income Distribution Methodological, Historical and Institutional Perspectives with Applications to the Case of Brazil (1926–2016), (PhD diss., Paris School of Economics and EHESS, 2018), p. 106, fig. 3.5, and pp. 135–316, figs. 3.24–3.25. In retrospect, it is clear that it was the 1964 coup, backed by the United States against labor president João Goulart (who had been minister of labor in 1953 at the end of the Vargas era, during a period of important wage hikes), that put an end to the era of inequality reduction in Brazil (1945–1964). The military seized power largely to end what was seen as a socialistic trend subversive of Brazil’s proprietarian social order.

175. 175.  In the 2010 census, individuals identifying as “white” represented only 48 percent of the population, compared with 54 percent in 2000. In the states of southern Brazil, whites remain in the majority, however. They account for 70–80 percent of the population of the state of São Paulo and of the states close to Uruguay and Argentina, compared with barely 20–30 percent in the state of Bahia and in the Nordeste.

176. 176.  The bargaining occurred because the system under which the Brazilian federal congress is elected makes it very difficult for one party to win a majority (even when it wins more than 60 percent of the vote in the second round of the presidential election, as PT did in 2002 and 2006). In particular, the Brazilian National Congress is chosen by proportional representation at the state level, which leads to a proliferation of regional parties.

177. 177.  According to available estimates, the share of national income going to the bottom 50 percent rose from 12 to 14 percent in Brazil between 2002 and 2015 while that of the next 40 percent fell from 34 to 32 percent and that of the top 10 percent remained stable at 56 percent. At the same time, the share going to the top 1 percent rose from 26 to 28 percent. See M. Morgan, “Falling Inequality beneath Extreme and Persistent Concentration: New Evidence on Brazil Combining National Accounts, Surveys and Fiscal Data, 2001–2015,” WID.world, 2017, figs. 3–4.

178. 178.  Note that the reduction of inequality in Brazil in the period 1945–1964 took place at a time when the international ideological context was much more favorable to progressive taxation and redistribution.

179. 179.  Note, too, that the multidimensionality of political-ideological conflict is also a factor in countries that are not pluralist electoral democracies, such as China, Russia, Egypt, and Saudi Arabia. But it is harder to see there except in fleeting ways, which makes the collective learning process even more difficult.

180. 180.  The Labor vote, which was quite high among the lower classes in the 1960s, began to center on the more highly educated in the 1980s, apparently reflecting both the transformation of the global ideological context (the turn away from socialism) and the evolving cleavage between Israelis of European and North American origin (Ashkenazim) and those of Middle Eastern and North African origin (Mizrahim and Sephardim). See Y. Berman, “The Long-Run Evolution of Political Cleavages in Israël, 1969–2015,” WID.world, 2018. Note, moreover, the near-total absence of fiscal data on income and wealth in Israel despite the country’s parliamentary Labor tradition.

181. 181.  F. M. Wuthrich, National Elections in Turkey. People, Politics and the Party System (Syracuse University Press, 2015).

182. 182.  In the early 1960s, some traditional elites transferred land to religious foundations (waqf) to avoid agrarian reform, and similar strategies continued to influence the geographic voting patterns for Islamist parties in 2000–2010. See S. Bazzi, G. Koehler-Derrick, and B. Marx, The Institutional Foundations of Religious Politics: Evidence from Indonesia (Sciences Po, working paper, 2018). See also P. J. Tan, “Explaining Party System Institutionalization in Indonesia,” in Party System Institutionalization in Asia, ed. A. Hicken and E. Martinez Kuhonta (Cambridge University Press, 2015), pp. 236–259.

183. 183.  See Chap. 7.

184. 184.  Other cases touched on in this chapter also illustrate the importance of foreign influences on party dynamics. The Cold War heritage and crushing of communist and socialist movements slowed the formation of class cleavages and encouraged the emergence of religious parties not only in Indonesia but also in Malaysia, Thailand, and Turkey. In South Africa, the defense of white landlords by Western governments (and most recently by Donald Trump) impeded ambitious agrarian reform. In Israel, it is obvious that the structure of political conflict could change completely if the United States and European Union decided to force a political settlement of the Israel-Palestine conflict.

185. 185.  See Fig. 14.1.

186. 186.  In this connection, it is striking to see the degree to which works associated with “left populism” avoid dealing with the question of what kinds of institutions are needed to transcend capitalism. Although some of these books are quite interesting, they avoid tackling the social-federalist issues raised in this chapter and also eschew discussion of the property regime, voting rights within firms, and progressive wealth taxes. See, for example, E. Laclau, La raison populiste (FCE, 2005); J. L. Mélenchon, L’ère du peuple (Fayard, 2014); C. Mouffe, Pour un populisme de gauche (Albin Michel, 2018). These works start with the assumption that the first priority is to heighten the antagonism between people and the elite to mobilize voters tired of the false alternative between left and right (and sometimes attracted by xenophobic rhetoric). The implicit hypothesis is that questions of programmatic and institutional content (concerning, say, Europe or the property regime) will be dealt with once a new balance of power among parties has been achieved.