The Global History of Inequality

INTRODUCTION

Inequality has increased in most Western countries since the early 1980s. In a recent report, the international non-governmental organization Oxfam noted that the twenty-six richest people in the world own as much wealth as the poorest fifty per cent of the world's population.¹ Discontent with the growing disparities in wealth and income has soared in recent years, especially in the wake of the 2007/2008 financial crisis and the “Great Recession” that followed. The Occupy movement protested against the greed of the “one per cent”, referring to the highly skewed income distribution in the US. Former US president Barack Obama proclaimed the growth of within-country economic inequality as “the defining challenge of our time”.² Yet, he enacted few policies that reduced inequality during his two terms in office; the Gini coefficient in the US actually increased slightly between 2007 and 2016.³ His successor, whose election has often been explained as a consequence of these high levels of inequality, has slashed taxes for the wealthy, probably causing further rises in inequality in the future. In this essay, I will review two recent economic history books that examine the historical roots of within-country inequality on a global scale: Branko Milanovic's Global Inequality (2016) and Walter Scheidel's The Great Leveler (2017). Formerly a lead economist at the World Bank, Milanovic is a well-known scholar working in the field of economic inequality, while Scheidel has a background as a specialist in the economic, social, and demographic history of antiquity.

The classical economists Adam Smith and David Ricardo had been much concerned with inequality.⁴ In the opening paragraphs to On the Principles of Political Economy and Taxation (1817), David Ricardo writes “[t]o determine the laws which regulate this distribution [between rent, profit and wages], is the principal problem in Political Economy”.⁵ For much of the twentieth century, however, distribution had been of relatively minor importance in economics. As the Nobel-Prize-winning economist Robert Lucas wrote in 2003: “Of the tendencies that are harmful to sound economics, the most seductive, and in my opinion most poisonous, is to focus on questions of distribution […] The potential for improving the lives of poor people by finding different ways of distributing current production is nothing compared to the apparently limitless potential of increasing production.”⁶

Similarly, until the Great Recession (2007–2009), economic historians had devoted much attention to the question of what caused economic growth in the West, as well as the reverse question of what caused the lack of economic growth in other parts of the world. This has been the crucial issue in the debate on the “Great Divergence”, which has dominated economic historical research since the early 2000s. The research on long-term trends in within-country economic inequality has been limited.⁷ Jan Luiten van Zanden noted in 1995 that the comparative study of early modern inequality in Western Europe was still “virgin territory”.⁸ The study of economic inequality also remained largely absent from major social history outlets, such as the International Review of Social History and the Journal of Social History. In the Journal of Social History, of the 1,598 research articles published since 1967, only one (!) was specifically focused on economic inequality.⁹ There were a few additional studies on earnings inequality between men and women,¹⁰ and two studies (by the same author) focusing on racial inequality.¹¹ In the International Review of Social History, of the 3,135 articles published since 1935, only fourteen contain the word “inequality” (and two of those are book reviews).¹² Only three have inequality as a main theme of the article.¹³ To an extent, this is the result of a different vocabulary in social history; all IRSH articles contain the word “class”, suggesting an interest in inequality between different groups.

This trend has radically reversed in the past decade. The most famous publication on inequality is, of course, Thomas Piketty's Capital in the Twenty-First Century, whose sales by 2016 had already exceeded 2.5 million,¹⁴ making it one of the few academic books to feature on global bestseller lists. But Piketty's work does not stand alone. A great number of articles on inequality have appeared in social and economic history journals since the Great Recession. As a result of this work, we now have a lot of information about long-run trends in inequality in many countries of Western Europe from the early modern period onwards, as well as for countries in the Americas since the nineteenth century. Most recently, inequality estimates have been published on some African and Asian countries, though the evidence is less abundantly available. Milanovic and Scheidel have synthesized the findings of these studies in their books.

LONG-RUN TRENDS IN INEQUALITY

Economic inequality is generally measured by the so-called Gini coefficient, which is scaled from 0 to 1, with 0 reflecting perfect equality (with all people in a society having exactly the same amount of income/wealth) and 1 reflecting perfect inequality (one person has everything, the rest have nothing). Computing an accurate Gini coefficient for an entire country requires a lot of data as you need information on the total amount of income earned as well as information on the distribution of this income across all members of the society in question. Therefore, some scholars have limited their scope to assessing the share of total income accruing to the top 1–10 per cent, as this requires less data about the distribution of income among the lower, and often less well-documented classes of society. Another way of dealing with this issue is to estimate a Gini based on information on the incomes of several groups in society, for example using “social tables”.¹⁵ Others have used proxies for the level of inequality, for example by looking at changes in the ratio of wages (reflecting the income of the lower classes) to rents (income of landholding classes) in a society.

There are various forms of economic inequality; there is inequality within countries, between countries, as well as between world citizens, which combines both. Milanovic's book looks at all three, while Scheidel focuses solely on inequality within countries/societies. In addition, there is the differentiation between income and wealth inequality. Whereas income is a flow, capital is a stock. Whereas Milanovic's book is concerned mainly with income inequality, Scheidel discusses both wealth and income inequality, even if greater emphasis is placed on the latter. Both books could have given more consideration to trends in wealth inequality as this indicator is generally more skewed than income inequality. In the present-day Netherlands, for example, post-tax income inequality is rather low (with a Gini coefficient of 0.29 in 2016),¹⁶ whereas wealth inequality is very high (a Gini coefficient of 0.89 in 2016).¹⁷ Furthermore, high wealth inequality tends to result in high income inequality as the income from capital tends to grow faster than income from labour (as Piketty has famously argued). Finally, income inequality can be measured before (market income) and after taxation and public transfers (disposable income). Substantial differences between these figures emerged, especially after the rise of the welfare states in the West.

While the timeframe of the two books is different, both studies find consecutive long-run cycles of increasing and declining inequality. Scheidel studies a much longer period – starting his discussion with pre-agrarian societies thousands of years bce. As one might expect for pre-historic times, the evidence on which this discussion is based is thin, yet the breadth of the material used and interpreted by Scheidel is huge. For example, he cites evidence from a Pleistocene burial site near Moscow with remains from about 30,000 to 34,000 years ago to show that even primordial hunter-gatherer groups were not entirely egalitarian, as some of the graves contained a much larger number of ivory beads and more prestigious items than others (Scheidel, p. 31). Inequality really took off with the rise of sedentary agricultural societies, when more complex social hierarchies were created. Early state formation allowed for the rise of a small ruling class, able to cream off much of the surplus created by the mass of cultivators. Evidence of rising inequality has, for example, been found from records on inheritance and dowries in ancient Mesopotamia for the period between roughly 1500 and 500 bce (Scheidel, p. 48), while archaeological evidence on house sizes showed that economic inequality peaked in the Roman Empire at the height of its power, in the first centuries of the Christian calendar, and declined after its fall.

Much of the evidence on medieval and early modern cycles of growing and declining inequality cited by both Milanovic and Scheidel stems from Europe. Inequality peaked just before the Black Death in 1347, when the dramatic loss of population pushed up incomes of workers across the globe. In the fifteenth century, inequality was once again on the rise as a result of population growth, which weakened the position of workers vis-à-vis landlords (Milanovic, pp. 60–63). Little is known about inequality trends in most other parts of the world in this period. Only for the Ottoman Empire is there data from probate inventories suggesting growing inequality between the early 1500s and the early 1800s (Scheidel, p. 102). The evidence on inequality becomes much more abundant for the nineteenth and the twentieth centuries. Data from the United States show continuously rising inequality from the late eighteenth century to the 1860s and a stabilization thereafter until the Great Depression. Subsequently, the Gini declined until its historical low point of 0.35 in 1979, after which it has steadily increased up until the present.¹⁸ Similar long-run swings in rising nineteenth-century inequality, a decline during the period of the two world wars, and a rise from the late 1970s/early 1980s can be found throughout Western Europe. For some parts of South America estimates suggest that the overall income Gini increased from 1870 to 1920 (Scheidel, p. 109). For Chile, however, the evidence suggests declining inequality from the 1870s to around the turn of the twentieth century. It then increased until the 1930s, after which it again declined (Milanovic, pp. 82–85).

Whereas Scheidel has focused on amassing an assortment of evidence from very deep into the past,¹⁹ Milanovic digs deeper into the development in (after-tax) income distribution of world citizens between 1988 and 2008. This development is shown in his now famous graph, which has become known as the “Elephant” due to its peculiar shape (resembling an elephant with a raised trunk) (Milanovic, p. 11). The figure shows that, in this period, real incomes increased most for those people around the median of the global income distribution. Most of these people with high income gains are part of the middle classes in emerging Asian economies such as China, India, and Indonesia, whose incomes increased by between fifty and eighty per cent. Other big gainers are those at the very top of the global income distribution (the infamous global top “one per cent” – mostly from the United States and Western Europe), whose (already high) incomes increased by some sixty-five per cent. The people who saw almost no increase in their incomes are the lower middle classes of the developed countries in Western Europe and North America. The poorest five per cent of the global population also saw only minor gains in their incomes over these twenty years.

Overall, it becomes clear that, across the world, countries experienced long cycles of rising and declining inequality. In the early modern and modern periods, these cycles took about fifty to a hundred years, but in earlier times they could have stretched longer (as shown, for instance, by Scheidel's evidence on the Roman Empire). These cycles do not occur simultaneously across the globe (see the evidence on Chile, which shows trends different to those for the US and Western Europe in the nineteenth and twentieth centuries). In general, for two “global” overviews, the works are disproportionally focused on the West (Europe and the Americas), while almost no reference is made to trends in within-country inequality in Africa and Asia. While this is, of course, the result of a lack of data, Milanovic and Scheidel do not reflect on how this could influence their results. Can the trends, as well as the explanations for those trends, really be extrapolated to African and Asian societies without any adjustments? This seems highly unlikely.

THE CAUSES OF RISING AND DECLINING INEQUALITY

What drove these long-run trends in inequality? The Kuznets curve, which predicts that with rises in average levels of income the level of inequality will initially rise and then decline, was long accepted as a general pattern in the evolution of inequality. According to Nobel laureate Simon Kuznets, who gave his name to this particular inverted U-curve, inequality in societies initially rose with economic development as the early stages of industrialization increased the incomes of factory owners faster than those of their workers, while incomes in agriculture stagnated or declined. After a certain level of average income has been reached, inequality is then expected to decline as the growth of the service sector and the welfare state will allow for a broader distribution of the benefits of economic growth. Yet, the rises in inequality in the West since the 1980s were clearly at odds with this theory. In his famous recent book, Capital in the Twenty-First Century, Thomas Piketty presented an alternative theory, suggesting that peaceful capitalist economies have a general tendency to become more unequal over time, as the returns to capital are greater than the general rate of economic growth (and thus the returns to labour), captured in the now famous formula r > g. Decreases in inequality occurred only as a consequence of special events, most importantly the two world wars of the twentieth century.

Milanovic is unconvinced by Piketty's theory, as he suggests that there have also been periods of “declining inequality driven by economic forces under capitalism” (Milanovic, p. 49). Here, Scheidel and Milanovic are in disagreement, because whereas Scheidel believes inequality can be levelled only by large-scale violence and disaster, Milanovic points to both benign and malign forces of levelling. In terms of benign forces, Milanovic draws on the work of Kuznets – he also christened the observed long-run swings in inequality “Kuznets cycles” – by emphasizing the role of urbanization (diminishing rural/urban inequality), the rise of schooling (reducing educational inequality), and population ageing (which increased demand for social services and therefore required higher levels of taxation) (Milanovic, pp. 93–95). Rising taxation and government spending, not only on education, but also on healthcare and other public goods, reduced inequality in the twentieth century. In contrast, Scheidel completely discards the possibility of benign, or Kuznetsian, forces of levelling. According to him, peaceful economic reform, education, democratization, or redistributive fiscal and welfare policies can be of a large enough scale and scope to lead to durable decreases in inequality, without the pressure of wide-scale violence.

Regarding the malign forces causing equalization, Milanovic and Scheidel are in general agreement. Scheidel's work is devoted entirely to what he calls the “four horsemen” of levelling: mass-mobilization warfare, transformative revolution, state collapse, and lethal pandemics. Milanovic similarly emphasizes these forces of levelling. There is, however, an important difference in how these two authors treat these malign forces. Whereas Scheidel seems to view these levellers as largely exogenous shocks, Milanovic views them as endogenous. Scheidel does not discard the possibility that high and rising levels of inequality played a role in causing state collapse or transformative revolution, but he simply does not include such considerations in his analysis. He writes: “for the purposes of this study, I treat violent shocks as discrete phenomena that act on material inequality” (Scheidel, p. 11). This means he does not consider the context in which the levelling takes place, thereby omitting important information about the mechanisms involved. Milanovic suggests that the outbreak of World War I and the decline of inequality caused by that war are related to the high levels of inequality predating the war. High levels of inequality led to high savings among the elites and limited domestic demand from the lower classes. This caused the wealthy to look for profitable uses for their money outside of their own country. As in the late nineteenth and early twentieth century, this meant “being in physical control of a place, and making such investment profitable required that other possible competitors be excluded even at the cost of a war” (Milanovic, p. 95). By endogenizing wars and other violent shocks, Milanovic makes them part of his Kuznets cycles, in which periods of rising inequality are inevitably followed by a period of decline.

Scheidel and Milanovic pay relatively little attention to another development that took place at the same time as the great twentieth-century decline in inequality, namely the growing strength of the labour movement in the interwar and postwar period. A recent study has demonstrated a strong and significant negative relationship between the proportion of workers who joined a labour union and various measures of inequality for the US in the twentieth century.²⁰ Scheidel (pp. 164–169) and Milanovic (p. 98) link the increase in unionization in this period to the wars, but given that the same rise took place in non-belligerent countries such as Sweden, and given the slow decline after 1945, war is clearly not the only story here.

In general, Scheidel seems overly keen to link all levelling to violence, death, and destruction, and this sometimes leads him to attribute levelling to violence even when there was none. This is mostly the case when discussing the relationship between the two world wars and the decline in inequality. For example, Sweden and Switzerland, which were neutral during World War II, both saw declines in inequality during the first half of the twentieth century.²¹ While Scheidel explains that both countries mobilized for war, he does not acknowledge the fact that violence, and huge losses of population, were not necessary for substantial levelling in these cases. Furthermore, he pays little attention to the fact that inequality in all Western countries continued to decline until well into the 1970s, long after the end of World War II. Conversely, Scheidel has suggested the Bolivian Revolution of 1952 as an example of violent revolution leading to greater levelling. Yet, he missed the research by Jonathan Kelley and Herbert Klein, which showed that the decreases in inequality in Bolivia were only temporary.²² In fact, Kelley and Klein formulate a general theory about why inequality after revolutions is always likely to return to previous levels in the long run – a theory that seems highly relevant to his book, as one of the chapters deals entirely with the relationship between revolutions and inequality.

Scheidel's pessimistic conclusion stems, at least in part, from the lack of a formal research design, leaving ample room for cherry-picking the cases that lead to a certain conclusion. While Scheidel's main research question (“why did inequality fall?”) is open enough to allow for the study of peaceful levelling, his sub-questions seem to relate primarily to different types of disaster (his “four horsemen”). He does not provide a systemic analysis of a clearly specified and consistent set of cases, nor does he define what he considers a substantial enough decline in inequality to consider it a successful case of levelling. Thus, he writes some 230 pages dealing with a wide range of examples of how death and destruction have led to greater equality, while often not giving an indication of the degree of levelling that took place. Scheidel then devotes only about forty pages to discussing the possibilities of peaceful levelling. The substantially reduced quantity of materials considered for peaceful levelling means that several such cases are neglected. Scheidel does not discuss why the Gini went down considerably in Spain after 1953 (during a period of great economic growth). Nor does he discuss the decline in inequality in Chile from a Gini of 0.60 to 0.45 between 1870 and 1900, and that of a similar magnitude between 1930 and 1970 (data in Milanovic, p. 83). Furthermore, as he has not defined what can be considered significant levelling, Scheidel easily brushes aside as “unsubstantial” declines from 0.49 to 0.44 in the Gini for disposable income in fourteen Latin American countries between 2000 to 2010. The fact that he has entirely missed the possibilities of peaceful levelling is particularly odd because he is familiar with the work of Milanovic, and cites him affirmatively on several occasions, yet he does not engage at all with the arguments and evidence on peaceful levelling that Milanovic puts forward.

What caused the rise of inequality? Scheidel seems to be in general agreement with Piketty, as well as with Bas van Bavel,²³ and assumes that the natural tendency for inequality is to rise (in market economies). He agrees with Milanovic that technological progress, commercialization, and economic development are factors leading to rises in inequality. He adds to these the role of state formation and the exercise of power by predatory elites in causing rising inequality (especially in the early historical episodes) (Scheidel, p. 86). Milanovic adduces a number of different driving forces of inequality in his book. In the pre-industrial era, growing inequality was associated with urbanization and the creation of economic surplus, which increasingly ended up in the pockets of rent-seekers. With population growth, the tendency was for returns to labour to decline relative to returns to land and capital, further pushing up inequality. Following Kuznets, he emphasizes the role of structural change in the rise of inequality in the modern era. Capital and high-skilled-biased technological change pushed up inequality by causing the rewards to capital and skills to rise.

Both authors suggest that globalization has pushed up inequality. Globalization increases inequality by putting downward pressure on wages (due to competition from low-wage countries) and also by making it more difficult to tax capital, which further increases inequality. Now, while this may have been the case for some countries, it was probably not the case for all. Martin Ravallion, for example, has noted that for the developing world within-country inequality has remained largely flat since 2000, while inequality in Latin America has been falling since the 1990s.²⁴ In France, as a consequence of its high statutory minimum wage, globalization has not led to increases in inequality.²⁵ Writing about the wave of globalization during the belle époque, Peter Lindert and Jeffrey Williamson find that globalization has very different effects on within-country inequality in different countries, depending on local resource endowments.²⁶ From my own research (in progress) on globalization and inequality in Southeast Asia, it becomes clear that local land market institutions play an important role in distributing the gains from trade.²⁷

Milanovic views globalization as an unstoppable force of nature (a misconception he shares with many economists). Thus, he writes “‘Deglobalization’ with return to the ‘local’ is impossible because it would do away with the division of labor, a key factor of economic growth” (Milanovic, p. 192). While the division of labour is certainly important in the rise of economic growth, with the high current levels of globalization, there are serious diminishing marginal returns to this mechanism, while it comes at great distributional costs. Dani Rodrik calculated that with average tariffs as low as they are today (below five per cent) “a move to complete free trade would reshuffle more than $50 of income among different groups for each dollar of ‘net’ gain created”.²⁸ Rodrik makes the compelling argument that globalization has gone too far and suggests a number of measures to reduce globalization and its disruptive distributional impact. Rodrik's work on globalization, which should be well-known to everyone dealing with the topic, remains conspicuously absent from Milanovic's book.

THE FUTURE AND CONSEQUENCES OF INEQUALITY

On the basis of their assessment of trends in the past, both authors attempt some predictions about the future of inequality. Milanovic believes that the economic catch-up of Asia with the West will continue in the coming decades – thereby reducing global inequality among countries and among world citizens. Inequality within Western countries will continue to rise and previous remedies to reduce inequality have run their course. Education levels are already nearing their upper limits in developed countries and globalization makes taxation of the most important contributor to inequality – capital – extremely difficult. Inequality in China may soon start falling as a result of rising levels of education and the ageing of its population (and thus the demand for more social spending), unless this development is forcefully counteracted by a Chinese rent-seeking political elite. In the West, the most promising options to reduce inequality are policies that distribute endowments in capital and human capital more equally among the population (Milanovic, p. 220). However, Milanovic doubts not only whether even very drastic policies will be enough to change the tide, but also whether such policies are likely to be implemented. Here, he finds himself in agreement with Bas van Bavel, who, in The Invisible Hand, also argues that economic inequality leads to political inequality and notes that the beneficiaries from the economic system that gave rise to this inequality are unlikely to implement correction mechanisms. After a tipping point has been reached (as it has been in advanced capitalist societies), ever-growing political and economic inequality becomes inevitable.²⁹ Regrettably, no analyses of possible solutions to this problem are offered by either Milanovic or van Bavel.

After having discussed Scheidel's most important levelling forces – death and destruction – in the previous section, it will come as no surprise to discover that he turns out to be a prophet of doom. Globalization will continue to be a potent force increasing within-country inequality in the future. Technological change, now including the ability to alter genes and modify human bodies, “will open up new frontiers in the evolution of inequality” (Scheidel, p. 431), and there is nothing that can be done about it: “even a combination of several quite radical and historically unprecedented government interventions would reverse the effects of resurgent inequality only partially” (p. 435).

The data on within-country income inequality for the recent period that both authors present can also be interpreted rather differently. In my view, throughout the twentieth century, taxation has been an extremely potent force for decreasing inequality. In Germany, for example, while the Gini for market incomes increased from below 0.4 to over 0.5 between 1970 and 2010, the disposable income Gini consistently hovered around 0.3 during that same period. Milanovic shows disposable Ginis that seem to fluctuate around that level (or only slightly above it) also in Spain, Italy, and the Netherlands, while Scheidel (p. 425) shows that the Gini for disposable income was about 0.27 for Denmark, Finland, France, and Sweden in 2011. The ability of these states to keep the Ginis floating around that level in the face of powerful market forces pushing them up over the past thirty to forty years, and without large-scale death and destruction, should give cause for optimism. Neither of the authors presents evidence on wealth inequality, which is generally much higher than income inequality, and has been increasing in recent decades; this might have supported their pessimism better.

Relatively little attention is devoted in either book to the consequences of inequality. This is unfortunate, since it is not clear to everyone why we should care about inequality. In fact, many people think that inequality may actually be good (especially those leaning to the right of the political spectrum). It spurs economic growth by incentivizing hard work, creativity, and human capital formation. What we should care about instead is poverty (these people would argue). Some go even further and argue that we should not even care so much about poverty, as we are already so much richer today than we have been at any time in the past. As a result of huge leaps in technological progress and economic growth, a poor person in the West today has a higher level of material well-being than a medieval English king.³⁰

Milanovic does not discuss the relationship between economic inequality and other indicators of well-being at all, while Scheidel briefly cites some research on the relationship between inequality, happiness, economic mobility, and civil war and conflict in the present. It is regrettable that neither of them cites the recent book by Richard Wilkinson and Kate Pickett that shows a wealth of evidence about the correlation of high levels of inequality with higher infant mortality, obesity, mental illness, crime rates, and drug abuse, and lower life expectancy and decreased levels of trust.³¹ Furthermore, neither Milanovic, nor Scheidel engages with the influential literature suggesting that high levels of political and economic inequality, resulting from colonial institutional legacies, have hindered long-run economic growth in developing countries.³² In fact, whether inequality is always bad for economic development remains a contentious issue and seems to depend a lot on the kind of inequality (structural or market-based), as well as on the wider context.³³ Engaging with these discussions could have made more compelling the argument that it is important to study inequality. While studies of the causes of rises and declines in inequality have mushroomed in recent years, more future research should be devoted to studying the consequences of inequality in a variety of contexts.³⁴ Recent research dealing with the effects of economic inequality on the abilities of societies to cope with disasters shows the promise of such lines of research.³⁵

Milanovic does, however, discuss the crucial question of whether within-country income and wealth inequality threatens the sustainability of Western democracy at present.³⁶ As a result of increasing inequality, the relative size of the middle class is declining across the Western world, which is problematic as the middle classes are generally seen as the most important supporters of democracy. In the US, the greatest threat is posed by the development of a plutocracy. We can already observe the much greater political power held by the wealthy vis-à-vis the middle classes and the poor in the United States. Research has shown that US senators are “5 to 6 times more likely to respond to the interest of the rich than to the interest of the middle class” (Milanovic, p. 194). Furthermore, the high cost of running for political office in the US essentially means that everyone except for the very rich is excluded from political power. American plutocracy is unlikely to effectively address the growing problems of the middle and lower classes associated with globalization. While in Europe entry into politics is less influenced by money and multiparty systems are effective at blocking the path to plutocracy, it faces other problems, as the pressure of globalization (that is both trade and migration) pushes the resurgence of nationalism and populist politics.³⁷

CONCLUSION

Inequality is back on the public and academic agendas, with a multitude of interesting studies on inequality appearing over the past decade. These include Global Inequality and The Great Leveler, both important and well-written works. With grand temporal and geographic scope, both these books have uncovered long-run cycles of growing and declining inequality in different parts of the globe. In more recent times, these cycles typically lasted between about fifty to a hundred years, while in antiquity and medieval times they may have lasted longer. In terms of the forces driving these cycles, there is substantial agreement between the two authors. Rising inequality is caused by technological change, globalization, and economic development. Further back into the past, rising inequality was also associated with the process of state formation and the increasing ability of elites to extract rent from individuals. They also agree that violence, death, and destruction are powerful forces reducing inequality. However, whereas Milanovic also believes in peaceful forces that level, such as unskilled-biased technological change, education, urbanization, and rising social transfers, Scheidel discards that possibility entirely. In the absence of violent shocks, both consider it unlikely that within-country inequality will decline substantially in the near future. It remains to be seen to what extent the developments observed for the West actually apply to all non-Western societies.

As a result of the huge breadth of these books, they have sacrificed some depth. In their efforts to generalize global trends, they missed important variation between cases; globalization, for example, has had very different effects on inequality across the globe, depending on the local context. More research on the interaction between local conditions and the factor that is thought to push up inequality could have led to more useful insights into how to combat rising inequality. Furthermore, these generalizations also lead them to overly pessimistic conclusions. Scheidel's one-sided suggestions that only death and destruction lead to substantial levelling are untenable, considering the substantial evidence on peaceful decreases in inequality. Milanovic, on the other hand, needlessly speculates that the policies proposed to combat inequality are unlikely to be implemented. Even if this were to be correct, it is unclear in what ways such pessimistic speculations are helpful in combating the problem of high inequality. Furthermore, since the Western world currently still (largely) consists of (more or less) functioning democracies, these conjectures do not hint at a high appreciation of people's ability to assess their own interest and vote in favour of parties and policies that can steer the world back into the right direction. If global history is full of examples showing how death and destruction led to decreased inequality, it is also full of examples of workers organizing themselves and bargaining for better labouring conditions, and of people voting for politicians who implemented social policies leading to the rise of the modern welfare state.