After the so-called structural reforms of the 1970s and 1980s, most Latin American countries had shown that they could achieve fast growth and deal with structural change. However, income per capita failed to converge on the world leaders, and growth was followed by increasing inequality and, in some parts of Latin America, even increasing poverty. Noting this experience, observers began to wonder whether inequality had become a permanent feature of Latin American development and whether it had contributed to the region’s disappointing long-run development performance (Bértola et al. 2010a).

A few years later, we were discussing something quite different. By 2014, Latin America had recorded fast growth for more than a decade and, contrary to what was going on in other parts of the world, inequality was falling. Had Latin America permanently changed its long-term direction? To what extent was decreasing inequality dependent on those high growth rates, and thus was it only temporary? What roles did market forces, institutions, and political ideology play during the inequality downturn? To start a search for answers, we sent out a call for papers for a conference in Buenos Aires, which was organized with support from the Economic Commission of Latin America and the Caribbean (ECLAC), the Inter-American Development Bank (IADB), and the International Bank for Reconstruction and Development (IBRD). The conference gathered together economic historians that had long been working on the history of Latin American inequality and poverty, with economists engaged in the study of inequality in the recent period.

By the time of the conference in December 2014, and even more clearly at the time of publication 2 years later, the atmosphere in Latin America had changed yet again. Even if it is too early to guess what will happen to inequality in the near future, we know for sure that the commodity-driven boom during the first years of the twenty-first century has come to an end. In 2014, Latin America was already growing more slowly than the OECD, and prospects for 2015 were even worse. The expected GDP growth rate for the years ahead is sufficiently low to allow us to say that Latin American catch up on the world leaders has come to a halt.

Thus, the original question posed at our conference is even more dramatically posed by more recent events. Has Latin America changed direction and was the recent inequality downturn just the result of a globally induced economic boom, similar to so many other previous booms in Latin America since the nineteenth century? Were the recent inequality events transitory with no permanent change in political, institutional, and other fundamentals, ones that have been a feature of the region for the past two centuries or even longer?

1 The Origins of Latin American Inequality

Most studies of Latin American development written between the 1950s and 1970s stressed colonial heritage as the main explanation of its underachievement. Competing schools of thought at least agreed on this point: dependency theories, modernization theories, Marxist writings, and development economics all agreed that Latin America’s colonial heritage contributed two key features: dependency on foreign powers, and inequality in civil rights, property rights, and political power. Things were made even worse due to the fact that the imperialists, Portugal and Spain, were relatively backward themselves compared with Western Europe.

Each of these thought that independence was jeopardized by the lack of a real social revolution, by the weakness of the local elites, by the failure to create a Latin American federation, and the development of new forms of unequal international relations, led by British “informal” imperialism, later followed by US hegemony. National states were consolidated during the last quarter of the nineteenth century. They implemented liberal reforms, by which the lands of the church, the peasant communities, and the state were privatized thus passing it on to an increasingly powerful landowning elite. Wage labor became dominant, but a varied array of coercive labor market mechanisms persisted. The landowning elite, together with merchants, foreign powers, state bureaucracies, and the military, formed an alliance which left the majority—mainly those of Afro-American and Indian ethnic origin—without property, civil rights, and education.

This process took varied forms in different Latin American countries and regions. Three main groups were identified in the literature. The Indo-American group—the Andean, Central American, and Mexican regions—was the center in the colonial period, densely populated and rich in gold and silver. There, the interplay between the haciendas and the peasant communities, together with centralized forced labor for the mines, was at the heart of social relations as the region drifted towards capitalism (Salvucci 2014). The Afro-American regions (northern Brazil, coastal Columbia, coastal Venezuela, and the Caribbean) were those in which tropical crops were produced with slave labor on large plantations. The indigenous American mountain and inland regions (Peru, Ecuador, Mexico, and Central America) worked the mines and farmed the haciendas. The Euro-American regions (Chile, Argentina, Uruguay, and southern Brazil) were less densely populated, labor was scarce, and European settlement played a dominant role. Even in these southern regions the concentration of land holdings took place. These interpretations of Latin American underdevelopment help explain different attempts to introduce economic, social, and political reforms in the middle decades of the twentieth century, including an active state involvement in the economy (Cardoso and Brignoli 1979).

After the 1980s, however, the academic and political climate in Latin America changed significantly. The region was confronted with the crisis of state-led industrialization, by the debt crisis, and by a “lost decade” of very poor economic performance. Pro-market reforms were introduced, often combined with authoritarian regimes. The story told by a new political voice said that the problems faced by Latin America were not market failures that the state had tried to overcome, but rather state involvement itself. Efforts were made to liberalize trade and capital flows, as well as to implement macro stabilization policies, and these were consistent with pro-global policies flourishing all over the world.

The pattern of development adopted by most Latin American countries since the 1980s was fueled by export growth, but the domestic sector did not keep pace. Inequality increased and the region continued to experience significant macro volatility. After every crisis, divergence with world leaders was deepened.

While increasing dissatisfaction spread and developed a more powerful voice in the context of the continent’s democratization, the dominant liberal discourse was challenged by neo-institutional approaches. While the neo-institutionalists agreed with conventional theory that free trade and unfettered capital movements were good for growth, they emphasized that domestic institutions were the ultimate factor explaining economic performance. Furthermore, institutions could not be expected to quickly change their character, since they tended simply to reproduce themselves in a path-dependent way. Thus, it was naïve to expect that the imposition of rules from outside sources would produce the expected results. Quite the contrary, foreign attempts to foster liberal reforms instead reinforced the power of rent-seeking elites and their extractive institutions.

When modern neo-institutional thinking entered Latin American debate, it rediscovered only one part of what we already knew: that domestic institutions were extractive, that most of the population worked under coercive mechanisms, that large parts of the population did not have civil rights or wealth, and that ethnic features were the key carriers of social and political relations. Neo-institutional theories argued that colonialism was a problem only in colonial times, vanishing afterwards while the creation of those domestic institutions in different clothing became the center of the story. The origin of bad institutions might reside with the colonizing power (North et al. 2000), or with the endowments the conquerors found (Engerman and Sokoloff 1997, 2012), or the domestic interplay between economic and political institutions (Robinson 2006). In any case, the problems Latin America faced at the end of the twentieth century had to be interpreted in the light of the institutions established shortly after the region was colonized. Changes in international economic and political relations, due to events such as the Industrial Revolution, the war of independence, the second industrial revolution, the transport revolution, mass migration, the process of profound globalization, and more, were not taken into consideration by neo-institutional thinking. This literature is consistent with the traditional histories that believed that Latin America was a relatively unequal place in the mid-late nineteenth century, that this inequality was consistent with institutions that were extractive and rent-seeking, and that these facts helped explain Latin America’s disappointing growth performance.

Coastworth (2008) challenged neo-institutional approaches. He argued that the problem faced by the independent Latin American economies was their lack of inequality. The elites were not powerful enough to expropriate land from inefficient peasant communities and a strong state was missing that would otherwise have invested in growth-enhancing infrastructure. It was during the first global century’s trade boom, from the mid-nineteenth century to World War I, that nation states consolidated, inequality soared, accumulation rates rose, and economic growth quickened. Coatsworth’s view was reinforced by the Inequality Possibility Frontier approach, developed by Milanovic et al. (2007). These authors argued that per capita income had to be high in order to produce a large surplus that could be appropriated by the elite, thus generating great inequality. Williamson (2010) has shown that in spite of this colonial legacy, inequality was no higher in Latin America just prior to its belle époque and early industrialization (c1870s) than in the western leaders prior to their industrial revolution (c1800s). Lindert and Williamson (2016) recently reported evidence showing that inequality in Latin America was also no higher than in the USA in 1860 (under slavery) or 1870 (after emancipation). We also know that inequality in Latin America grew significantly from 1870 to 1929 (Prados de la Escosura 2007; Bértola et al. 2010b), due in part to the impact of a secular commodity price boom that raised land rents relative to wages, but also because of the class power structure. Of course, the kind of inequality matters (Marrero and Rodriquez 2013; Bértola 2011). Inequality can be good for growth if it simply reflects rewards to those accumulating skills, taking risks, and innovating. Inequality can be bad for growth if it simply reflects successful rent-seeking or rising rents to land and mineral resources. While the assertion has not been confirmed with historical evidence, it seems plausible to think that late nineteenth century Latin American inequality was more of the “bad” kind while it was more of the “good” kind in the industrial leaders. After all, inequality is not only a matter of income and wealth but also a matter of opportunities, property rights, access to education, and civil rights. A continent with 25 % slaves and 60 % Indian peasants with very limited access to property, civil rights, education, and political power, as the Latin American case was since colonial times, can hardly be considered egalitarian whatever the income inequality measure. Moreover, Bértola et al. (2015) have shown how dangerous it can be to work with aggregate Gini-coefficients alone: during the middle decades of the nineteenth century, rural areas of Buenos Aires underwent a fall in inequality due to the huge rise in the labor share, while land concentration and the rental-wage ratio were increasing. In short, a low Gini-coefficient may hide a highly polarized society.

Much of the debate about Latin America’s inequality history has now focused on the twentieth century. The new question, posed recently by Williamson (2015), is whether Latin America failed to join the steep inequality downturn experienced by the developed world from World War I to the 1970s. Our knowledge about Latin American inequality trends over the period is still limited. We know much more about the increasing inequality between the 1960s and 1990s. For the period up to the 1960s, the problem is that we cannot generalize from what we know about only a few countries.

Bértola and Ocampo (2012: Chap. 1) identified three Latin American societies and all with different inequality trends. During the twentieth century, the Indo-American and the Afro-American societies seemed to converge towards labor markets which behaved like the Lewis unlimited-labor-supply model. In these regions, state-led growth and industrialization may have created new forms of inequalities: a highly concentrated industrial sector was able to create many jobs for relatively skilled and well-organized workers, while a large share of the labor force remained marginalized in low productivity jobs. The Euro-American societies in the southern cone experienced a different growth pattern. In Uruguay, the increasing inequality of the belle époque was followed by a slight reduction up to the 1930s. Afterwards, until the 1960s, inequality declined significantly in spite of a terms-of-trade boom. Contrary to what happened in previous booms (and contrary to the logic of the preceding bust of the 1930s), the institutional and political environment produced a significant inequality reduction: wage councils, industrial protectionism, welfare policies, and multiple exchange rates were tools that favored the egalitarian trend in a way that resembled the “Keynesian” atmosphere in the advanced countries (Bértola 2005). Many of these features could also be seen in Argentina and Chile. What happened after the 1960s in the Southern Cone is better known. Thus, the century-long trends appear to be ones of increasing inequality up to WWI, decreasing inequality up to the 1960s, and increasing inequality up to the turn of this century. This pattern is also consistent with the Piketty (2014) description of post-World War I developments as a unique period of decreasing inequality in the middle of the twentieth century (see also Lindert and Williamson 2016: Chap. 8). Was this southern cone pattern repeated elsewhere in Latin America, or did most of the region miss the Great Levelling which occurred among the industrial leaders from WW1 to the 1970s?

This volume contains fresh new contributions to this Latin American inequality debate. Part I gives the economic historians a chance to speak to those issues by looking back before the recent downturn, and Part II gives the economists a chance to look again at the recent downturn with Part I in their rearview mirror.

2 Part I: Looking Backwards for Explanations

In Chap. 2, Pablo Astorga presents new data measuring the functional distribution of income, based on four income groups (three groups of wage earners and a group of capitalists) for six Latin American countries (Argentina, Brazil, Chile, Colombia, México, and Venezuela), covering the century 1900–2011. The first conclusion to be drawn from this new evidence is that no common pattern is shared by all countries. True, Astorga’s new evidence documents that they all underwent increased inequality during the last decades of the twentieth century, a result consistent with other evidence. However, prior to 1960 secular trends in inequality differ. Argentina, like Uruguay (Bértola 2005), underwent rising inequality in the early decades, falling inequality in the middle decades, and then rising inequality in the closing decades. This is similar to what Rodríguez Weber finds for Chile (Chap. 3). In sharp contrast, Brazil shows stable inequality in the early decades followed by a rising trend in the middle decades which levels off in the closing decades. This characterization is consistent with what Buccifferro reports regarding Brazilian racial inequality in Chap. 8. A common feature found by Astorga is the existence of a sustained, even if moderate, narrowing of wage inequality in the middle decades of the twentieth century. However, that narrowing was offset by a rising share received by the top group, so that polarization widened during the period in all six economies, with the exception of Mexico. After 1970, the labor share fell while the mean income of the top earners rose, the clear winners over the last century. Another stylized fact revealed by Astorga’s new estimates is that countermovements took place: there was increasing polarization between the income levels at the two extremes of the social scale, while the share of the middle increased. These contradictory trends appear to be consistent with a society composed of two parts, a traditional one in which unskilled rural labor coexisted with rich landowners, and a modern one in which migration from the countryside ended up in more diversified urban activities, with higher productivity and higher wage inequality. Later on, a large part of this urban labor force was transformed into informal low-skilled workers, thus increasing inequality. In any case, Astorga’s contribution begins to fill a gap in our knowledge which will influence our thinking about Latin American inequality across the twentieth century.

In Chap. 3, Javier Rodríguez Weber explores income inequality in Chile since the 1850s. He provides a continuous series that reveals historically high and persistent inequality, with a wave-like pattern. Inequality rose until the 1870s, then decreased after the Pacific War, due to a huge expansion of its northern and southern frontiers. Inequality rose again during the first decades of the twentieth century with the climax of export-led growth. The Great Depression mainly damaged the income of the elites, while, like Uruguay, inequality fell after World War II. As in Argentina and Uruguay, inequality rose significantly during the military dictatorship. After more than two decades of democracy and center-left governments which followed the dictatorship, inequality was little changed.

Chile shows very high inequality levels today, among the highest in the world. This fact calls for explanations. Rodríguez Weber argues that the elite have been able to shape the economic and political institutions in its favor. This power was rooted in the high concentration of wealth and the oligopolistic market structure which not only promotes high inequality but also undermines democratic institutions. The power of the elite has deep historical roots. Since the nineteenth century, the landowning interests in the Central Valley were able to control a weak peasant population that lived in the haciendas.

The Mexican case is analyzed in Chaps. 4 and 5 using quite different evidence. Moromay López Alonso and RobertoVélez Grajales exploit data on heights to measure improvements and inequality in well-being. Using two different sources (passports and the army) they find increasing social inequality up to the 1930s. Differences between regions were less important than the inequality within regions. Since the 1930s, the poor population benefited by national public health campaigns as well as a (modest) rise of the welfare state thus reducing inequality; before then, the majority of the population was excluded from health services and lacked opportunities for advancement. Using data for the period from the 1950s onwards, the authors conclude that inequality is higher among women, and that there exists a clear downward bias against the agrarian population. In short, it seems that inequality in living standards was reduced mainly during the central decades of the twentieth century. It continued to decline thereafter, but at a slower pace, and remaining high. In Chap. 5, Campos, Domínguez, and Márquez construct a Human Development Index for Mexico from 1895 to 2010 with data at the subnational level for illiteracy rates, urbanizations rates, and the number of physicians per capita, as proxies for education, income, and health, respectively. The authors find a significant increase in human development as well as some evidence of regional convergence, especially between 1940 and 1980. However, regional disparities increased again after the 1980s. In short, these studies of the Mexican experience seem to confirm two main features: inequality was high and persistent; and inequality was reduced only during the central decades of the twentieth century, but rose again during the last two decades of the century.

The Brazilian case is analyzed in Chaps. 6, 7, and 8 with quite different approaches.

In Chap. 6, Pedro Funari studies the varied development patterns in four Brazilian states (Minas Gerais, Pernambuco, São Paulo, and Rio Grande do Sul), which, given their different initial resource endowments and colonial experience, would be expected to have different de facto institutions. He explores the connection using standard land concentration measures, and a proxy for political concentration (the percentage of eligible voters), all at the municipal level. His results support the idea that it is difficult to find general rules for the relation between growth and inequality. In the fast growing states of Minas Gerais and São Paulo, he found a positive relationship between inequality in 1920 and development outcomes in 2000. The contrary is found in the southern state of Rio Grande do Sul. Pernambuco, a northeastern state, presents no evidence of a significant relationship between prior inequality and subsequent long-term development outcomes. With respect to political inequality and development, no significant relationship could be found. Funari explains this latter result as depending on the limited formal power of voters in relation to other more important de facto sources of power. It is also probable that inequality among landowners is not the best way to capture the impact of inequality on growth, especially when the economy becomes less dependent on agriculture and when the dynamic sectors are mainly urban. In any case, what seems to be confirmed is that Brazilian economic growth during most of the twentieth century was compatible with a very high level of inequality.

In Chap. 7, Eustaquio Reis studies spatial inequality in Brazil by means of an ambitious reconstruction of data at the municipality level for the period 1872–2000. His approach is quite different from that of Funari (Chap. 11), since Reis emphasizes the role played by geography (land quality, climate, and transport costs), rather than informal institutions. He concludes that spatial inequalities did not experience significant changes over those 130 years: the rate of convergence is much slower than those found in other countries. The period of import substituting industrialization appears to have lowered regional inequality, while export-led growth increased it. Infrastructure development (particularly the railways) was important: the lack of adequate infrastructure investment was a major determinant of relatively slow growth. The only “institutional” factor that had an important impact on growth was the immigration rate, while human capital, slavery, land concentration, and political participation were not significant determinants of per capita income and productivity.

In Chap. 8, Justin Bucciferro studies racial inequalities in Brazil since the early nineteenth century, by analyzing data on income, primary school completion rates, literacy rates, life expectancy at birth, and occupation, augmented with qualitative sources. White, black, mixed race, indigenous, and Asian individuals were considered. He finds that progress has been made in the reduction of racial inequality in terms of health, education, and access to certain occupations, like industrial employment. However, racial disparities are still very large, and income gaps have not been significantly reduced. Bucciferro emphasizes how small the impact of emancipation on racial income gaps was since European immigrants undercut the bargaining power of the libertos. The Vargas populist period was the one in which most progress was made, while 1945–1980 recorded years of regression or stagnation in racial inequality. However, important progress was made during the last decades of the twentieth century and racial inequality seems to be at its lowest level today. If it is assumed that once ethnic-based inequality is reduced through cultural change and political empowerment, it should prove difficult to revert to previous levels. Thus, reductions in ethnic inequality should be long-lasting.

In Chap. 9, José Alejandro Pérez-Cajías asks whether the Bolivian Revolution of 1952 led to a significant institutional change in the delivery of education. His findings are based on the construction of impressive new evidence. After the revolution and until the 1980s, tax collections remained among the lowest in Latin America. Nevertheless, education spending as a share of GDP converged on Chile and Uruguay, two of the most developed Latin American countries. However, the quality and distribution of Bolivian education was not significantly modified: public spending on education was low and educational outputs were often among the worst in the region. Obviously, the main explanation for this was that per capita GDP remained extremely low in Bolivia. The question still unanswered is whether an increased effort in the educational system could have enhanced growth performance in the decades following the revolution.

In Chap. 10, María Camou and Silvana Maubrigades explore gender inequality in five Latin American countries (Argentina, Brazil, Chile, Mexico, and Uruguay). They find a tight correlation between total income inequality and the gender income gap. Those countries showing a lower gender wage gap are those that were successful in incorporating a larger share of women in the labor market in the first decades of the twentieth century, together with higher educational achievement. The countries with bigger gender gaps are those where women were incorporated into the labor market more slowly and which had a higher share of Indian and black workers with whom women first compete. The gender wage gap also increased throughout the period, a result consistent with their relative supply growth. In spite of their rising educational attainment, women are still underrepresented in the high-wage occupations.

In Chap. 11, Leticia Arroyo-Abad and Peter Lindert present a multi-country history of how Latin American government spending and taxes have reshaped the distribution of income. They combine historical time series for six countries with evidence on fiscal redistributions in the twenty-first century. While Latin America has followed the international trend with social spending increasing with per capita GDP growth, this social spending has not had a big distributional impact. As it turns out, the elderly have been favored by social spending in the form of pensions, and not by education and infrastructure that might have targeted the non-rich. Even though the redistribution did not favor the poor, its impact was still moderately positive, since social spending is distributed less unequally than pretax income and since the tax system is slightly progressive. The Southern Cone countries distributed more in favor of the elderly, but they were also most progressive in redistribution.

3 Part II: The Recent Inequality Downturn

The recent inequality decline in Latin America poses several questions. The first is whether inequality actually declined or whether it was mainly the result of incomplete data or incorrect measurement. The recent literature on top incomes has shown how much our perception of inequality can change if the results of household surveys are corrected by accurate measures of the income of the elite. Unfortunately, information on top incomes is still scarce: series for the twentieth century are only available for Argentina from the mid-1930s, for the first decade of this century they are only available for Argentina and Colombia, and estimates for a few recent years are available for Uruguay. By combining household surveys and top income figures, a more complete view of income distribution can be achieved, even though no clear conclusions can yet be drawn from so few cases. However, the World Bank (2014, pp. 35–39) concludes that while the drop in inequality is certainly moderated when top incomes are considered (for the limited sample), the downturn shown by household surveys still holds.

Important research has recently been reported on the recent downturn. López-Calva and Lustig (2010), as well as the same authors together with Campos et al. (2016)), argue that a fundamental factor behind the decline in income inequality is the rise in supply of workers with post-secondary education. The case of Mexico is particularly important, as it is the case of an economy that experienced neither a commodity boom nor fast growth.

Cornia (2014) concludes his edited volume stating that the inequality decline across the whole of Latin America can be characterized as a Polanyian process, in the sense that several social and political forces reacted against the outcomes of globalization and liberalization in the late twentieth century, with the result of increasing inequality. He concludes that inequality cannot be further reduced if some structural biases, as the lack of industrial policy, low savings, dependence on foreign capital, dependence on a few commodities, and reprimarization, remain unaddressed.

ECLAC has also been active in this field, producing a trilogy of reports with inequality at the center (2010, 2012, 2014). These reports, while taking into consideration the role played by market forces, emphasize the role played by other social factors, as, for example, labor market institutions and the impact of the activity rate, including gender issues. Moreover, ECLAC emphasizes the political character of distributional issues and links inequality reduction to the structural transformation of the Latin American economies.

What follows is then a partial list of possible explanations for the inequality decline. First, improvements in education which augmented the supply of skilled labor, thus putting downward pressure on the skill premium. To that extent, schooling might have won the race with technology (Goldin and Katz 2008). Second, if growth based on maquila-industries and natural resource processing favors the demand for unskilled labor, inequality might diminish on this count as well. Third, fast growth associated with the commodity boom should have raised the employment rate, thus raising the income of the lowest deciles. Fourth, to the extent that political shifts favored pro-poor labor market policies—minimum wages, wage councils, formalization of employment, social transfers, and the like—an inequality downturn should be reinforced. Finally, the consolidation of democratic rule in most of Latin America should have weakened the tacit power of the elite and thus the importance of their opposition to pro-poor policies.

The following chapters confirm some of the previous studies, add new evidence, explore new topics (as structural change and fiscal policies), and offer some in-depth political economy studies of national cases.

In Chap. 12, Verónica Amarante and Antonio Prado present a summary of ECLAC’s findings published in what is called the equality trilogy (2010, 2012, 2014). While accepting the role played by demand and supply forces in the recent inequality decline, they stress the role played by labor market institutions. Minimum wage increases, for instance, accounted for a third of the recent drop in Argentina’s Gini-coefficient, for 84 % in Brazil and for 7 % in Uruguay. Their research also finds that changes in formal employment shares and wage differentials between formal and informal jobs are as important as returns to education and changes in schooling levels—although there is considerable variance across the countries studied. The chapter also documents participation rates and income gaps by gender: the region exhibits great variation, some countries exhibiting huge gender gaps, and others almost parity. An exercise is performed exploring the impact on income inequality if these gender gaps closed. In the case of the labor participation rate, a reduction of about 5 % points would be achieved in the poorer Latin American countries. In the case of income gaps, a reduction of about 4 % would be achieved.

In Chap. 13, Augusto de la Torre, Julian Messina, and Joana Silva confirm the recent decline in labor income inequality and conclude, in line with López-Calva and Lustig, that such changes were driven by labor market behavior. Furthermore, the reduction was largely driven by a decline in the skill premium (returns to tertiary vs. primary education). They note, however, that a similar increase in the supply of skilled workers took place in the 1990s, while inequality worsened. To account for the difference, they argue that in the recent decade technical change displaced old skills thus producing some knowledge obsolescence.

In Chap. 14, Diego Sánchez-Ancochea adopts a multidisciplinary approach to study the political economy of policymaking to learn about the explanations for Chile’s high inequality. He confirms the persistent concentration of income at the top also found by Rodriquez Weber (Chap. 3). A few business groups control large segments of the economy, featured by high market and capital concentration, conditions that limit action by the State at the sector level. On the other hand, the State has been playing an increasingly active role in redistribution through social policy, thus creating opportunities for new progressive cross-class coalitions. In this context, the progress to be made on the margin depends mainly on the capacity to increase the tax burden of the elites. Thus, no radical changes are to be expected, unless more radical political changes take place.

In Chap. 15, Martin Andersson and Andrés Palacio ask whether the recent reduction in income inequality was driven, at least in part, by an agricultural resurgence, that is, whether the reduction of the rural–urban income gap played an important role. This seems to be the case, as agricultural labor productivity grew faster than in other sectors, reducing the intersectoral productivity and income gap. Nevertheless, the authors point to the fact that a new situation has appeared, featured by the increase in the share of the labor force employed in low productivity urban (especially informal) sectors. Thus, a new pattern of urban inequality may be replacing the old one of rural inequality. Furthermore, agrarian inequality can remain an important influence since productivity growth has been taking place mainly in large and capital intensive estates.

In Chap. 16, Judith Clifton, Daniel Díaz-Fuentes, and Julio Revuelta also explore the impact of fiscal policy on Latin American inequality, this time from 1960 to 2012. They find that fiscal policy had a much bigger influence during the period 1999–2012, than before. For the earlier years, fiscal policy had either a regressive or insignificant impact on inequality. During the more recent period, they find that the impact was higher in revenues than in expenditures. Furthermore, expenditures had a much bigger impact in urban than rural areas. With respect to revenues, direct taxation had a much more important impact on inequality reduction than indirect taxes. Given the fact that the tax burden in Latin America is much lower than in OECD countries, it is clear that a policy which increases the direct tax burden as per capita income grows would erase some inequality—although that might be hard to achieve politically.

Finally, in Chap. 17, Suzanne Duryea and coauthors recognize the recent achievements made by Latin American countries in terms of poverty and inequality reduction. But they then ask whether those achievements are sustainable and explore the possibility that the inequality downturn might cease or even reverse. They focus on three main policy areas that should help insure continued progress: an improved efficiency of social policy and the quality of services provided; the promotion of income stability and protection without distorting workers’ incentives; and the exercise of caution with respect to the creation of potentially costly and irreversible commitments.

So, what do we conclude about recent Latin American inequality trends? While there has been great variety across Latin America, it appears that the recent inequality reduction was due, at least in part, to very specific conditions associated with the commodity boom. The boom raised labor participation and lowered unemployment, favoring those at the bottom of the distribution and thus reducing inequality. What the boom giveth, the bust can taketh away. But the boom combined the impact of long-run improvements in human capital formation (lowering skill premiums), with an increased demand for unskilled work linked to the export-led boom. More fundamentally, a cluster of socio-political shifts favored labor. The new wave of democratization strengthened the voice of the poor, and their demands for education, health, and social transfers. A number of leftist governments and center-left alliances promoted the development of different labor market institutions, which contributed to the income leveling. Tax reforms were introduced in many countries, which, combined with better designed spending, has helped improve conditions in the lower deciles. Minimum wages were raised in several countries, having an important impact on distribution.

An important conclusion is then that it is not possible to reduce the explanation of inequality reduction to a single cause. Inequality was reduced even in countries, as Mexico and El Salvador, that did not experience a commodity boom. The commodity boom cannot per se explain inequality reduction: Latin American economic history shows that commodity booms often led to increasing inequality and the actual results of booms and busts in terms of inequality cannot be understood without political economy considerations. When considering political economy issues, the role of the past becomes important, in the sense of understanding the origins and development of Latin American economic, social, and political forces.

The big question is whether this process of declining inequality will be sustained and deepened. Many of the forces that were driving the leveling of incomes now are gone. The commodity boom has disappeared, and some of its positive impacts may be reversed. The activity rate will probably decline and unemployment will probably rise. Demand for unskilled work will probably decline. On the other hand, investment in social spending may favor education in its race with technology (Goldin and Katz 2008), pressing inequality downwards.

Democratic rule is now a permanent feature in Latin America. We can, of course, doubt its quality. State powers are still weak, domestic and foreign elites still have an influence on governments, and state bureaucracies are still lacking important capabilities. The risk of mismanagement of social revenues and expenditures is always present, and the possibility that antidemocratic regression might prevail in critical situations cannot be excluded.

The future of inequality is not only a matter of pure market forces but also on how labor market institutions and the political climate evolve.