Socio-Economic Review

Socio-Economic Review Advance Access originally published online on June 24, 2006
Socio-Economic Review 2007 5(1):81-115; doi:10.1093/ser/mwl017

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 5/1/81    most recent mwl017v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Google Scholar Articles by Mann, M. Articles by Riley, D. Search for Related Content Related Collections N30 - General, International, or Comparative D31 - Personal Income, Wealth, and Their Distributions Social Bookmarking          What's this?

© The Author 2006. Published by Oxford University Press and the Society for the Advancement of Socio-Economics. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Explaining macro-regional trends in global income inequalities, 1950–2000

Michael Mann¹ and Dylan Riley²

¹ Department of Sociology, UCLA, 264 Haines Hall—Box 951551, Los Angeles, CA 90095-1551, USA ² Department of Sociology, UC Berkeley, 410 Barrows Hall (1980), Berkeley, CA 94720-1980

Correspondence: mman{at}soc.ucla.edu; riley{at}berkeley.edu

This paper analyses levels and trends in within-country income inequality during the second-half of the 20th century. Two broad general trends were visible: beginning in the 1950s and 1960s inequality declined substantially in most countries, and in the 1980s and 1990s it increased more variably and more slightly. However, we focus here on presenting and explaining the distinct ‘macro-regional’ regimes of inequality that became evident in this period. We identify six: Anglo, Nordic, Continental West European, Latin American, East Asian and South Asian. We show that intra-regional variations in inequality have been much less than inter-regional variations, and that the six regional regimes did not remain constant but had distinct trajectories through time. The Anglos were at first the most egalitarian, only later developing the greater inequality supposedly characteristic of their ‘liberalism’. The Nordics only became the most egalitarian in the middle of our period. The Euro-Continentals began as the most unequal among the Western countries, but then became almost as equal as the Nordics. Latin America began and remained the most unequal region of all. East Asia was rather egalitarian, while South Asia converged towards a middling inequality. Though pressures towards more inequality strengthened from 1980, regional responses remained varied and path-dependent. Then, in more speculative vein, we seek to explain these regional patterns in macro-sociological terms. Beginning from a model of dynamic interactions between economic, ideological, military and political power relations, we trace the political institutionalization of distinct macro-regional ideologies of ‘best practices’ of socio-economic development.

Key Words: inequality • income • welfare state • globalization • world regions • JEL classification: N30 income and wealth: general • international • or comparative

	1. Introduction

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
Virtually all research on income inequality takes either the nation-state or the entire world as its basic unit of analysis. Work on within-country income inequality (Tsai, 1995; Alderson and Nielsen, 1999, 2002) uses the nation-state. Work on between-country inequality (Firebaugh, 1999, 2000) and synthetic work attempting to address both problems in a single perspective (Goesling, 2001) uses the concept of world income inequality and often emphasizes differences between the developed ‘North’ and the underdeveloped ‘South’. This paper develops a different approach. We suggest that what we call ‘macro-regions’ may be a useful way of thinking about income inequality. We conceptualize ‘macro-regions’ as clusters of countries occupying specific geo-political niches, and possessing similar political institutions, similar cultures, and similar economies. Our argument is that levels of within-country income inequality are similar across countries within macro-regions, but differ among countries in different macro-regions. We first paint a broad descriptive picture of cross-national income inequality, which shows the importance of regions. We then focus on some of the historical processes, such as defeat or success in war, colonial invasion and the spread of specific cultures that are potentially causally significant for understanding how these patterns of income distribution emerged. We conclude by suggesting that scholars look closely at macro-regions as well as nation-states in examining income inequality.

	2. International income inequality: the debate

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
We present the debate on income inequality in two stages. First, we review the consensus view on the dynamics of income inequality over the past couple of decades. Second, we consider two types of explanations for these dynamics (within-country explanations, and North–South explanations). Finally, we argue that neither of these explanations presents a compelling account of how countries cluster in terms of their levels of within-country income inequality. Indeed, we argue that although macro-regional differences are widely recognized, they have engendered remarkably little theoretical discussion. In the final section, we sketch a conceptual approach for addressing this problem focusing on macro-regions rather than nation-states, or centre–periphery models.

2.1 The state of the debate
Three conclusions seem to be relatively well established in the literature on comparative income inequality: first, within-country income inequality increased in most countries in the 1980s and 1990s after having fallen steadily during the post-war period; second, income inequality among countries, particularly among the developed, ‘North’ and the underdeveloped ‘South’ remained significant and may have increased, particularly in the 1980s; third, despite a widespread trend towards increasing inequality, the overall level of inequality among countries continued to differ. We briefly review the literature on each of these points.

In a 1978 Brookings institution publication, economist Henry Aaron argued that, ‘tracking changes in the distribution of income in the USA "is like watching grass grow" (quoted in Gottschalk and Smeeding, 1997, p. 633)’. That is now no longer the case. In most countries in the world both labour market inequality and household inequality increased in the past two decades of the 20th century (Gottschalk and Smeeding, 1997, p. 636). This view also seems to be supported by a recent study of inequality in 73 countries (Cornia and Addison, 2003), which found that during the 1980s and early 1990s inequality widened in 48 of them, with 16 remaining static and only 9 seeing more equality. Thus the 1980s and 1990s were decades of widening within-country inequality in most countries for which data are available.

What explains the trend towards widening income inequality in the 1980s and 1990s? Although this debate is far from settled two basic positions dominate the field: market-based accounts, and class and institutional accounts. Neo-classical economists argue that inequality primarily varies according to the demand for and supply of land, capital and labour of different skills. This model predicts the level of inequality should vary in an inverse ‘U' shape with trade liberalization. The classic Kuznets curve showed inequality first widening as trade expanded as a result of industrialization, then narrowing (Higgins and Williamson, 1999; Dollar and Kraay, 2004). Today trade pressures are assumed to vary between the North and the South of the world. The North is more endowed with skilled labour, and the South with unskilled. According to the Stolper–Samuelson Theorem of comparative advantage, as trade increases, resources will shift to export-oriented markets. As global trade barriers are lowered, the ‘North’ exports skilled-labour intensive goods and imports unskilled-labour intensive ones. The ‘South’ does the reverse exporting unskilled-labour intensive goods and importing skilled-labour intensive ones. Thus in the ‘North’, demand for skilled labour increases, but falls for unskilled labour while the opposite happens in the ‘South’. So trade opening increases inequality (because it increases returns to education) in the ‘North’, but lessens it in the ‘South (Wood, 1994, pp. 28–30; Gottschalk and Smeeding, 1997, p. 647)’. This provides a potential explanation for expanding income inequality in the ‘North’. There are market forces driving wage differences between highly skilled and unskilled workers.

A slightly different, although not necessarily incompatible, style of explanation focuses on institutions. These institutions are specified in different ways. They can refer to property relations, class organizations, or organizations of collective bargaining. Classical political economy and Marxists argue that the distribution of property generates class struggle determining income distribution (Dobb, 1963, p. 14; Marx, 1984, pp. 877–884; Smith, 1976, pp. 53–61). This has been confirmed by the positive correlations that exist between union strength and leftist governments and greater equality (Garrett, 1998; Hicks, 1999, pp. 145–147; Korpi, 2000; Rueda and Pontusson, 2000). Scholars in this tradition suggest that globalization affects income inequality by changing the balance of class power. Nationally organized labour weakens as unionized manufacturing declines and skill differentials grow, while globally organized capital strengthens. Thus unions and states become less sticky and less able to resist capital's drive to acquire more of the surplus.

Some recent work also links income inequality to the dynamics of international class formation. Finance capital, these scholars suggest, is the most globalized part of contemporary capitalism, able to outflank the state and weaken labour. This leads to a fiscal ‘race to the bottom’ as states are forced to lower taxes and welfare benefits in order to attract mobile capital. This widens inequality. This is seen as entwining with a further pressure: the geo-economic power of the USA. The USA is both the dominant economic power and its economy has also shifted most from manufacturing to finance (Arrighi, 1994, pp. 214–238; Brenner, 1998, p. 20; Brenner, 2002; Arrighi and Silver, 2003). In the 1970s, US global economic dominance shifted from the Bretton Woods system of global currency and credit management, diplomatically negotiated and maintained by the Powers, to a more US-and market-driven system based on the dollar as a reserve currency (Gowan, 1999). The USA favours relatively open markets, especially in the financial sector. In the late 1970s the neo-liberal Chicago School swept first through American economics and business schools, and then through those of many Southern countries. American control of international banking also helped it impose neo-liberal reforms on the world. This is also said to widen inequality, for even some neo-liberals accept that in the short run greater market-conformity will widen inequality.

Many dispute the connection between market-conformity and growth. They say that more statist ‘production regimes’ generate their own efficiencies, alternative to those of markets. Active labour market policy, human-capital and physical-capital investment strategies, and ‘social pacts’ organized by states between employers and workers may all confer comparative advantages over more market-conforming regimes. These strategies achieve equality and growth, while growth will also reduce the fiscal pressures on states, allowing them to break out of the debt cycle and reduce inequality further. Some note that the most active production regimes of the past were in small countries highly dependent on exports and imports. In the past globalization pressures led to more, not less, intervention in markets. Today they note there is almost no correlation between the internationalization of the economy and inequality. Growth, not a particular production regime or level of inequality is what will attract foreign capital (Boix, 1998; Garrett, 1998; Goodin et al., 1999; Weiss, 1999; Hall and Soskice, 2001; Rhodes, 2001; Swank, 2002). In the South of the world high inequality seems to generate low growth. Even a World Bank ‘viewpoint’ publication (De Ferranti et al., 2004, pp. 11–14) now advocates more redistribution in Latin America, to be financed by higher taxes. Among the more developed countries (which tend to have less inequality) there is no relationship between inequality and growth (Kenworthy, 2004, ch. 4). There is more than one way to get growth. Countries may choose either a low or a moderate level of inequality (though not a high one) without sacrificing growth. Market power and market efficiency are not the same thing. Markets privilege certain power actors who may then enforce higher levels of inequality, but this may be unrelated to or actually harm growth. These various positions have generated a sophisticated empirical literature examining the effects of foreign investment, institutions, demography and market forces on intra-country income inequality (Burtless, 1995; Tsai, 1995; Gottschalk and Joyce, 1998; Alderson and Nielsen, 1999, 2002; Gustafsson and Johansson, 1999; Kentor, 2001).

Closely related to the argument over the causes of income inequality, is a debate over the causes of differences in the level of inequality among countries (irrespective of trends). American levels of inequality are almost twice the level of Swedish ones; Brazilian inequality is almost three times the level of South Korean; the differentials between skilled and unskilled wages in Mexico widened, while Canada's did not, and there has been no overall race to the bottom in tax policies within the Organisation for Economic Co-operation and Development (OECD) countries. As Gottschalk and Smeeding (1997, p. 636) put the point, ‘There is substantial diversity in the inequality of household disposable income across major OECD nations, with the greatest inequality in the USA and the least inequality in Nordic and Northern European countries.’ Thus, whatever one's view of income inequality trends, national level differences remain important.

Work on comparative welfare states has most explicitly engaged in explaining these differences in levels. Scholars in this tradition recognize that most welfare states have proved resilient in the face of recent pressures. Esping-Andersen's tripartite model of welfare-state regimes is still widely used. His ‘liberal’ regime is the most market-conforming, though a safety-net welfare-state is targeted specifically at the poor. Rights are universalistic, but do not greatly restrain market-based inequalities, seen as necessary for economic growth. The ‘social–democratic’ regime extends welfare benefits universally to the population, which greatly reduces market (and gender) inequalities. In the ‘conservative corporatist’ regime benefits are based on social insurance principles, directly related to contributions (and therefore to incomes) though the state also rewards certain social statuses, especially male breadwinners and stay-at-home mothers. Though it does not particularly aim at redistribution, some of its targeting and the sheer scale of transfers do redistribute. Many scholars find that the social–democratic and conservative corporatist regimes have been ‘sticky’, modifying programmes slightly in the face of pressure, but essentially preserving their welfare strategies. Only liberal countries have seen extensive welfare cuts (Esping-Andersen, 1990, pp. 25–27; Kersbergen, 1995; Pierson, 1998, 2001; Esping-Andersen, 1999, pp. 73–95; Huber and Stephens, 2001, pp. 223–241; Wildeboer Schut et al., 2001; Swank, 2002, 2003).¹

Yet many doubt whether such ‘stickiness’ can hold out much longer against a barrage of current pressures. Some emphasize the globalization pressures noted above. Others emphasize economic pressures only half related to globalization, like rising unemployment or the transfer of employment towards the less productive, more unequal, and less unionized service sector. Others stress demographic pressures on state budgets such as population ageing, the growth of single-person households less able to shoulder welfare burdens, the entry of more women into the labour force (increasing the supply of low-wage labour and also widening inequality between households with two and no earners), and the gradual maturing of welfare entitlements in the population over time (e.g. Pierson, 2001; Kenworthy, 2004).

The third basic fact that has generated much debate, is that most income inequality is between, not within, countries. Here two positions have developed. One group of scholars argues that ‘...the gap between the populations of wealthy and poor nations has grown steadily since 1965 (Korzeniwicz and Moran, 1997, p. 1029)’. Many scholars date the shift to the recession of the 1980s, and link it specifically to the USA's high interest-rate regime (Hobsbawm, 1994, p. 407; Baumann, 1998, ch. 4; Arrighi and Silver, 1999, pp. 211–212; Held and McGrew, 2000, p. 29; dissenters are Castells, 1998, ch. 2 and Goldthorpe, 2002). Firebaugh (1999) challenges these conclusions, although he (2000, p. 330) agrees that ‘...the huge disparity in average incomes across nations is the major component of the total income inequality for the world’. Thus, even scholars who deny that the gap is widening, recognize that differences between the ‘North’ and ‘South’ are of fundamental importance. Overall trends seem to differ according to whether people or countries are the unit of analysis. In most Southern countries average incomes are worsening relative to the averages of most Northern countries. But because China and India, with the two biggest populations in the world, also have the world's highest growth rates, the average income of most people in the South is now drawing a little closer to the Northern average (Korzeniewicz and Moran, 1997; Firebaugh, 1999; Rodrik, 2002; Milanovic, 2003; Dollar and Kraay, 2004). Others have argued that a shift is now occurring in which within-country income inequality is becoming again relatively more important as a source of world inequality (Goesling, 2001). All the positions in this debate agree, however, that between-country income inequality, and specifically inequality between the developed ‘North’ and the underdeveloped ‘South’ is a fundamental dimension of the world distribution of income.

2.2 Regions: a theoretical lacuna
Discussion of within-country inequality and discussions of between-country inequality fit easily into established theoretical traditions. Theoretically the debates are between various kinds of Marxian analysis (class struggle analysis, dependency theory, world systems) and classical and neo-classical economic models. None of the positions in these debates focuses on regions, especially as culturally and geographically constituted clusters of nation-states. Rather literature remains focused either at the level of the nation-state, or at the level of the world as a whole. The only theoretically relevant regions are ‘North’ and ‘South’, which do not designate geographical, historical or cultural areas but rather refer to the developed and underdeveloped world, respectively. There are some examples of attempts to model regional effects more finely. Both Tsai (1995, p. 480) and Alderson and Nielsen (1999, p. 624) modelled regional differences in the impact of FDI on income inequality. While Tsai found that region mattered, Alderson and Nielsen (1999, p. 624) argued that ‘... net of the baseline model and of time invariant, regional specific factors, the stock of foreign direct investment continues to have a strong positive effect on income inequality’. But while both studies agree on the importance of regions, they do not integrate regional effects theoretically into their analyses.

The comparative welfare-state literature has dealt most directly with this question. While work on the welfare-state, and especially Esping-Andersen's scheme, is extremely valuable (and indeed constitutes the starting point for our analysis), its theoretical arguments remain within the basic institutional framework of regime variations at the level of nation-states sketched above. Welfare states, on this argument, are more egalitarian because they possess institutions that redistribute income in a particular way. It is fundamentally the characteristics of national level state and class structures that matters most. What is left unacknowledged in this scheme is the regional clustering of these states.

Our ambition is to put regions (especially as they are culturally constituted) at the centre of the analysis of inequality. Our key argument is that the characteristics of countries' regimes of inequality derive in part from the political institutionalization of macro-regional cultures in which they are embedded. Of course we do not argue that culture is the only cause of income inequality. We would agree with the De Ferranti et al.’s statement (2004, p. 22) about Latin America, that inequalities are produced by ‘...economic processes...mediated by social and political institutions that are very broadly understood to encompass the rules and norms of behavior in society'. We would prefer to express this more systematically, in terms of a model of the four ‘sources of social power’—ideological, economic, military and political (Mann, 1986, ch. 1).

	3. Method

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
We begin our analysis by presenting descriptive statistics of our six macro-regions. We then seek to explain the differences by tracing the historical processes that produced them. Our macro-regions are clusters of countries that share what are often called distinctive and historically constructed ‘cultures’. Figure 1 lists the countries that are included in the six regions—Latin America, East Asia, South Asia, Anglo, Nordic and European Continental. Three of these regions (the Anglo region, Latin America and the Nordic region) share either common languages, or a mutually intelligible group of languages. Two of the remaining regions (the European Continental region and East Asia) share common political cultures. The European Continental region is characterized in the post-war period by a Christian Democratic/Socialist political culture, while East Asia shared the model of the developmental state. The South Asia region is marked by enormous internal cultural diversity dominated by massive multi-ethnic states that have very specific patterns of inequality.

View larger version (13K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 1 Definition of the macro-regions.

 

	4. Data

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
We now turn to the evidence. Economic inequality may be of wealth or income. We examine only income because more comparative data exist and because most recent changes are in income dispersion (Atkinson, 2003, p. 488). Our data come mainly from the United Nations World Income Inequality Database (WIID), compiled from national sample surveys and arriving at Gini coefficients of equality. These are constructed using a Lorenz curve, which plots the share of the national income that each decile (or tenth) of the population possesses. In a case of perfect equality the Lorenz ‘curve’ is a diagonal straight line, with each decile receiving an additional 10% of the income. In the case of perfect equality each decile would consume exactly its proportional share of income, so the bottom 10% would consume 10% of the income and so on. In reality the Lorenz curve tends to bend inward and form more of a reverse ‘L’ shape. This means that bottom deciles of the income distribution do not consume their share of the income. For example, the bottom 60% might consume only 40% of the income. The Gini coefficient is a way of expressing this inequality with a number. The coefficient is the area between the Lorenz curve and the diagonal line of perfect equality divided by the total area under the diagonal line of perfect equality (for technical issues, see Atkinson, 1975, pp. 45–46). This measure has the advantage of being widely used and easy to interpret, especially in the form used here of the Gini index: the coefficient multiplied by 100. As the index increases, so does inequality. The lowest Gini index in our countries is about 25; the highest is about 60.

Comparing the results of different surveys conducted at different times and some with different methodologies is problematic. Debates rage over alternative measures and samples, and attempts at international standardization are still ongoing. The Gini index is also not a perfect expression of the actual income distribution. Similar figures may express income distributions that are quite different. Further measures of poverty often reveal rather different distributions and trends. Nonetheless, Ginis provide the simplest and broadest macro-evidence on inequality. Though the UN WIID data are widely used they are uneven in coverage, quality and categorizations. The data are generally thought to be more reliable when analysing trends through time within countries than when making comparisons among countries. Many errors will probably enter into such comparisons. We are comparing groups of countries, and so if there are such errors they are likely to reduce the chances of finding fairly cohesive macro-regions. Thus if we do find them, this might be doubly impressive. We have mostly used country data points defined by the WIID team as being ‘OK’ (i.e. of high quality), but we added some ‘Not OK’ data points where we thought this was justified. Thus, for example, some of the Latin American data was collected using urban-only surveys, but this can be justified by observing that these are highly urbanized countries. We did compare the patterns found in all country data points with those found in only the ‘OK’ data points and found no significant differences between them.

We selected data on household or family as the income-receiving unit. There seems to be very little difference between the two. Most scholars working on income inequality prefer to have evidence on net income, but the difference between gross and net income inequality is also interesting to us especially as it relates to the differing role of the state in different regions. Thus we present descriptive statistics of both net and gross income. The weight of our analysis however rests on the WIID gross income inequality data. For the gross income evidence for South Asia we included surveys that asked about household expenditure. Scholars suggest that expenditure is less unequally distributed than income. Following the procedure of Cornia (2001, p. 19), who faced the same problem with same data, we adjusted the Gini Index upward 7 points allowing us to include data on India, Malaysia and Indonesia. We supplemented the WIID with data based on dispersions in manufacturing wages, as collected by The University of Texas Inequality Project (UTIP, 2000) provided by Galbraith and Kum (2005). This is not based on surveys of individuals or households but on overall industry data. Thus it uses only an ‘imputed’ Gini coefficient calculated from the Theil statistic. It is based on labour market inequalities in wages, before much state intervention has occurred. In this respect it is closer to the WIID gross than net income data, and thus broadens our ability to distinguish between market and state determinants of degrees of inequality. The picture of income inequality that emerges from the WIID gross and the UTIP data for our purposes does not differ considerably. We will comment on the UTIP evidence only when it seems to be sharply at odds with substantive conclusions that we draw from the WIID evidence.

Figure 2 presents two sets of box-plots.² The box-plots show the median Gini coefficient, the full range of Ginis and any outliers for each decade on our two WIID measures. We now comment briefly on each macro-region.

View larger version (21K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 2 Gini index from WIID.

 
Latin American countries have seen the greatest inequality over the past 50 years, with Ginis ranging from the mid-forties to the mid-fifties for every decade. The net household data from the WIID are not much different from the gross household data, suggesting that Latin American states do not redistribute income in a way that affects inequality. These data suggest little trend. We note, however, that the UTIP data do show an upward trend in the 1980s, beginning to flatten in the 1990s. This indicates perhaps that recent pressures towards even greater inequality came more from market than state processes. Of course, other scholars and institutions have already noted that Latin America is probably the most unequal region in the world (Deininger and Squire, 1996; Inter-American Development Bank, 1999; Morley, 2001; De Ferranti et al., 2004).

In our data South Asian countries were the next most unequal group, with gross household Ginis ranging from the low-forties to the upper-forties for every decade. We have no net household Ginis, though the UTIP data broadly confirm the gross pattern. There appears to have been an upward trend in the 1970s, followed by a majority downward trend (though with outliers). This region contained the most variety.

After the 1950s, the East Asian region became more equal—and this began before most countries in the region industrialized. Almost all the regional Ginis for net income inequality were in the thirties after 1950. The WIID box-plot charts show no trend over time, though the UTIP data did show a somewhat ‘U’ shaped trend from the seventies to the nineties, especially in Hong Kong and Taiwan—perhaps again indicating recent market pressures counteracted by state actions.

Note that these three less-developed regions of the world show very different patterns. There is no single ‘Southern’ pattern. Indeed, the first ‘Northern’ region, the Anglo region has been quite similar to the East Asian region. Anglo gross household Ginis were also in the thirties, though with a slight upward trajectory from the sixties onwards (the UTIP data show most widening occurred in the 1980s, with a flattening in the 1990s, except for the USA and New Zealand where manufacturing wage dispersion continued to widen). Anglo net household Ginis had a wider range and with a more pronounced widening of inequality during the 1990s. This may mean that the Anglo states became less geared to redistributing incomes. This region began the 1950s with the lowest level of inequality of all regions, but this was followed by steadily rising inequality.

The countries in the Continental European region exhibited a steady decline in inequality in gross and a milder decline in net household inequality through the period. The UTIP data are (surprisingly) closer to the latter pattern, though with greater variations among countries: Austria, Italy and the Netherlands experienced a ‘U’ shape pattern with high inequality in the early 1970s and again in the nineties. In Spain, inequality fell in the 1980s, and the trend remained flat; while in Germany inequality increased a little in the 1990s. The balance of all the data for the 1990s show a more limited trend upwards in inequality, as those countries formerly most equal move closer to the median.

The Nordic region has been the most equal of all regions since the 1970s. There was no reversal of this in the 1990s, though the UTIP data show a slight upward trend in inequality from the 1980s and a short burst of increased inequality one in Norway in the early 1990s, and another one in Sweden in the late 1990s. This probably indicates that there were labour market pressures operating against their high equality regime, but that the region's states were still managing to counteract this with redistributive tax and welfare policies.

To more securely establish the ‘regioness’ of macro-regions we present two further data displays. The first is a line chart (Figure 3) of the mean Gini Index for Gross Household Income over the five decades of the WIID data. The chart shows not only no convergence among average national level Gini indices in these regions, but seems to show some divergence. It seems that Latin America, the Anglo, and the East Asia regions were becoming more unequal, while the Nordic, South Asia, and European Continental regions were becoming more equal.

View larger version (11K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 3 Mean Gini index by macro-region (gross household income WIID). Source: United Nations University WIDER (2000).

 
We also performed analysis of variance (ANOVA) (Table 1). This is a technique for assigning variance to within-group and between-group differences. Unfortunately there was not enough data to perform ANOVA on the net income evidence so our analysis is restricted to the gross household or family income from the WIID. The purpose of this exercise is simply to test, and in our view cast some doubt on, the hypothesis that regions fundamentally do not matter. ANOVA analyses variance as measured by the sum of squares. The total variance is constituted by the sum of squares of the between-group variance and the sum of squares of the within-group variance. The mean sum of squares is the total sum of squares divided by the degrees of freedom. This figure indicates the variance explained per degree of freedom. An ‘F’ statistic is derived by dividing the mean sum of squares of between-group differences by the mean sum of squares of within-group differences. The ‘F’ statistic is used to determine if the differences between groups are probably due to chance.³ Our analysis shows significant differences among the means of the macro-regions for all the decades. All the ‘F’ statistics are significant at the ‘0.01’ level. Further these differences did not disappear over the decades in our analysis. This analysis thus suggests that macro-regions have durably different levels of within-country income inequality.

View this table: [in this window] [in a new window]   Table 1 ANOVA of Gini indices of national gross household income inequality among and within macro-regions

 
Our description points to three conclusions. First, we found that our ‘macro-regions’ are indeed different in terms of their mean Gini coefficients. Further these differences have persisted over time. Second, we did not find a simple North–South divide. Patterns of income inequality differ within both groupings of countries. Third, there is little evidence of a convergence towards a single global pattern of income inequality. We found durable, though not static, macro-regions in both North and South. Thus, looking at the evidence descriptively suggests that the levels of within-country income inequality differ according to ‘macro-region’.

	5. Explaining macro-regional income inequalities

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
How should we explain these differences? Generally the literature on cross-national income inequality attempts to come up with explanations that are identical across countries, regions and historical periods. In this section we develop a different approach. We suggest, somewhat speculatively, that different causal mechanisms may have been at work in the different regions. We would argue that there are four sources of social inequality: ideological, economic, military and political. These correspond to the four sources of social power identified by Mann (1986, ch. 1). We briefly state their overall contributions before proceeding to more detailed explanations of the historical trajectories of the regions.

1. Ideological power is relevant to inequality in two ways. First ideologies such as liberalism, socialism and Christianity have diffused across the world, interacting with more local ideologies to generate taken-for-granted macro-regional notions of ‘best practices’. These shape the policies of states, the organizations of firms, and the types of inequalities that populations will tolerate. Remember that ideologies are less constrained by states than other forms of power and easily transcend their borders. Second, ideologies may also determine patterns of group closure. For example divisions among castes, nations and races can all form bases of inequality.
   
2. Economic power relations are also important, as previous scholars have argued. Some have taken a global market-centred view of these, yet different forms of class compromises entrenched within countries may be as crucial to understanding our results.
   
3. Military power relations expressed through colonial conquest, world wars and anti-colonial rebellions were important in our data in shaping the initial parameters of inequality in the 1950s and 1960s. But even in the era of relative international peace that followed, geo-political power relations continued to play a role, especially visible in the role of the USA in influencing Latin American patterns of inequality.
   
4. Political power relations have remained important because of the globalization of the nation-state: over 190 polities claim sovereignty over their territories in the name of the people or nation. Sovereignty includes policies to shape patterns of inequality, which therefore vary country by country. Different political institutions also matter. Degrees of state centralization and ‘inclusive’ electoral systems involving both proportional representation and multi-party regimes promote welfare generosity (Swank, 2002), and so probably more equality. How have these four sources of inequality interacted to produce the levels and trends in inequality that we have shown above?
   

5.1 Latin America
Latin America is by most accounts the most unequal region in the world. The top 5% of the income distribution in Latin America remains double that of the North and 60% higher than that of East Asia. How should we understand the specificity of this region? Ideologically the region has developed a thin national identity encouraging rapacity among elite groups and little social solidarity. At the same time Latin America has been characterized by surprisingly little group closure, apart from conflicts between indigenous populations and European immigrants. Economically hugely concentrated landholding has exacerbated inequality, and the emergence of a large informal economy has done so as well. Militarily the region remains subordinate to the USA. Politically it has had weak states capable neither of much development, nor much redistribution.

High levels of inequality originated in the land expropriations of the Conquest. A small elite of Spanish or Portuguese conquistadores, gentry and merchants acquired by conquest large estates, mines and businesses, exploiting but not exterminating non-free indigenous or imported slave labour. As an oligarchy they controlled the post-colonial states, and European immigration then filled out the middling classes. Differences among the descendants of indigenous peoples, African slaves, ordinary European immigrants and ruling oligarchies remain large, underpinned by ethnic differences which explain much of the inequality differences between countries (De Ferranti et al., 2004, ch. 4). These levels were then institutionalized through oligarchical states widening their client networks into ‘patrimonialism’. In fiscal terms, states were historically quite small and their military violence was largely confined to internal repression, with few international wars. The whole continent shares Catholicism, and all but Brazil share the Spanish language. This shared continental culture has been crosscut less by inter-national than by similar intra-national ethnic divides, weakening the cohesion of each nation-state (Centeno, 2002).

To the extent that states have been effective, their actions tend to have exacerbated income inequality. Dominant classes and interest groups have been successful in milking these states for rent, while universal policies of social welfare have rarely developed (Costa Rica is the main exception). Patrimonial states now reinforce Latin American inequality. For example, the education system is top-heavy in post-secondary education (compared with East Asia which has relatively more elementary and middle school places). Despite the relatively abundant supply of education, returns to education increased through the 90s. As Morley (2001, p. 52) writes ‘...the white collar differential has fallen everywhere but in Latin America, where it has actually risen sharply since 1988 despite increase in the share of college and high school graduates’. Market forces do not here determine returns to education. Rather, the highly educated can extract rents from their education, and the state plays an active role in overproducing them.

Regional, ethnic and class differences remained large, and most people did not have stable urban-industrial employment. Neutrality in the World Wars was important, for it ensured minimal disruption to dominant classes and ethnic groups and evaded the kind of land reform or ‘social citizenship’ that often followed 20th century wars in Europe and Asia. Across the mid-20th century, attempts at further economic development centred on creating a large internal market by protecting local industry and putting a floor under industrial and public sector wages. States combined import-substitution, tariff protection and nationalization of enterprises. Social citizenship, unlike the path in Great Britain, became half-universal, half-sectoral. States granted privileges to employees in modern industry and government over agricultural and marginal sector workers. They acted as patrons, redistributing resources to clients inside the more modern sectors, including labour unions, while the poor remained outside. De Ferranti et al. (2004) emphasize the ‘weight of history’ in Latin America. To be more exact, recent history has delivered no great crisis to disrupt conquest-generated inequalities.

By the late 1970s it looked as if some states (particularly Brazil) might break out of the historic mould to achieve growth and cut inequality. However, geopolitics intervened. Latin America is geo-politically subordinated to the USA. In its struggle against socialism the USA had supported (where necessary with military means) Latin American oligarchs and their regimes of inequality. Yet the debt crisis led to the USA seeking to roll back the patrimonial state by promoting market-friendly neo-liberal reforms. Oil price rises in the early 1970s created a huge reserve of petro-dollars deposited in US banks. Northern demand for money dropped as investment slowed with a long economic downturn, so banks lent freely to developing countries at very low rates of interest. The transfer was greatest in Latin America. But in 1979 the USA suddenly raised interest-rates to cut inflation, and by the early 1980s the foreign debt burden for Latin American states surged to between 33 and 100% of GNP (Stallings, 1992, pp. 58–63; Fourcade-Gourinchas and Babb, 2002, p. 557). American-dominated international financial institutions insisted on new credit terms as a solution to the crisis: neo-liberal structural adjustment programmes that on balance would widen inequality.

Yet sovereign states make and implement policy, and no state (not even Pinochet's Chile) embraced the entire neo-liberal package. Each tried to protect a mixture of key constituencies and universal social citizen rights. In Brazil there were even signs of movement towards more rights (Ferreira and Litchfield, 1996; Baumann, 2001). Overall, however, the structural adjustment packages widened inequality (whether slightly or greatly is hotly debated), especially through the banking reforms (Bulmer-Thomas, 1996; Panuco-Laguette and Szekely, 1996; Berry, 1998; Ocampo et al., 2000; Morley, 2001, pp. 46–50, 84–86; De Ferranti et al., 2004).

We attribute the slight widening of inequality in the continent in the 1980s to the initial neo-liberal reforms induced by US pressure, and the subsequent slight trends towards convergence in the 1990s to a more shared and mixed later response. The weak Latin American countries tended to divide under the pressure of US neo-liberalism. Propertied classes liked the policy of reducing inflation by cutting budget deficits and earning foreign exchange through freer trade and export-oriented growth (Bulmer-Thomas, 1996, pp. 10–12). Yet politicians were reluctant to withdraw protections from their clients, who included public sector workers and labour unions. The more internationally visible fiscal and commercial policies usually moved towards the required neo-liberalism, while state welfare and employment shifted more slowly or not at all, especially at provincial government level. Political struggles over such issues have dominated politics over the past two decades and their outcome continues to influence trends in inequality.

Latin America thus has proved vulnerable to two distinctive sources of inequality: a historical legacy of inequality rooted in the post-conquest social order, and a neo-liberal reform movement beginning in the 80s. The region was particularly prone to these reforms because of its geo-politically subordinate position in relation to the USA, and the relative weakness of both nationalism and effective states.

5.2 South Asia
The inequality figures on the South Asia region suggest a zone with a very high level of inequality, but which has been somewhat successful at reducing those inequalities over the past 50 years. The region is thus similar to Latin America in terms of its level, but differs in terms of its trend. In terms of all four of the sources of inequality we have discussed above, the countries in the region differed dramatically from those in Latin America. Ideologically, South Asia possesses both strong forms of post-colonial nationalism, and elaborate and ancient patterns of social closure associated with religious and caste differences. Economically the distribution of land in South Asia was probably more egalitarian than in Latin America. At least it seems relatively clear that India still possesses a large population of small owner cultivators. Militarily this region is very autonomous from the influence of the USA. India, in particular, is a nuclear power that cannot be bullied by the Americans. Politically the region is dominated by developmental states committed to reducing some forms of inequality, particularly those associated with caste.

Our data (supported by Dollar and Kraay, 2004) suggest that inequality in the region peaked in the early 1980s, followed by some lowering of Gini coefficients through the late 1980s and the 1990s. The extent to which this results from secular Kuznets curve effects, statist policies or liberalization, remains unclear. The South Asia region is not as cohesive as the others. The countries within it differ from East Asia, being later developers, much larger (except for Malaysia), and more regionally and ethnically diverse, though Malaysia is bi-ethnic rather than multi-ethnic. These factors favour inequality and weaken the cohesion of nations and states. Yet, influenced by the nearby East Asian development states, they consciously adapted parts of the developmental state package. They were initially protectionist, generated more of their own investment than in Latin America, and usually subsidized exports, though Indonesia has not (see various essays in Rowen, 1998 and Chan et al., 1998). These contrary tendencies make it difficult to predict responses to inequality pressures.

Within the South Asia region India is a particularly interesting and important case. Here two distinctive sources of inequality overlap: caste inequalities and class inequalities. Caste distinctions have historically been of two kinds on the subcontinent: the Varna and Jati systems (Deshpande, 2000, p. 322). Both were forms of social closure in which groups were assigned specific jobs in the occupational and social hierarchy. The Jati system tended to be more regionally variable and less rigid than the Varna system. Into the 1960s in rural areas the lowest castes lived outside the village in ‘...mud-and-thatch hovels, packed closely together in the middle of the paddy fields (Béteille, 1977, p. 39)’. Elaborate rituals of deference and social avoidance underlined the differences between the Brahmans and the other castes (Béteille, 1977, p. 41). The post-colonial Indian state made great efforts to overcome caste distinctions by establishing ‘protective discrimination’ for certain castes (the so-called scheduled castes). This effort was enshrined in a highly egalitarian constitution (Béteille, 1983, p. 79).

Many studies now suggest that caste inequalities have been steadily weakening over the past 50 years. At least some of this weakening is connected with the emergence of new occupations that undermine the economic foundations of the castes. For example rural labour is now largely performed by casual labour forces rather than through a system of services assigned to specific caste groups. The emergence of this casual labour force may decrease economic security, but it is also probably an egalitarian force from the point of view of caste since ‘...agricultural labour is the most caste-heterogeneous in the village (Jayaraman and Lanjouw, 1998, p. 15)’. In the village of Palanpur the lower caste Muraos enriched themselves through investing in agriculture and thus have grown wealthier than the previously upper caste Thakurs (Jayaraman and Lanjouw, 1998, pp. 37–38). Thus, in India greater market integration may tend to reduce income inequality, at least within the agricultural community.

The South Asia region then has been characterized both by more effective states, and more culturally entrenched forms of inequality than in Latin America. There seems to be some evidence for cautious optimism about inequality in the region. India in particular is an effective and independent state, and the development of markets is undermining at least some forms of inequality.

5.3 East Asia
The East Asia region is marked by a rather intermediate level of inequality, falling between the extreme inequalities of Latin America and South Asia, but generally higher than in the rest of the developed world outside the USA. Ideologically these states share a hierarchical/communitarian model of society. Economically most of the states in the region underwent extensive land reforms (unlike either Latin America or South Asia). Militarily the region is subordinate to the USA, but the USA has not been able to impose its models of development or political institutions on these countries. Politically the region is characterized by effective and highly interventionist states.

The Gini indices for East Asia have been kept relatively low by deliberate state intervention within cultures that do not easily legitimate economic inequalities. Japan's Ministry of International Trade and Industry and similar institutions in Korea and Taiwan funnelled private savings into industries that were competitive on the world market, pursuing export-led growth and market share more than short-term profitability. This presupposes a high rate of savings and a low level of domestic consumption, the opposite of the import-substitution model. Austerity has been made culturally and politically palatable by substantial economic equality (Johnson, 1982, 1995, p. 65). These states have inherited strong bureaucratic traditions that are particularly culturally embedded in universities (Evans, 1995, pp. 47–60). State managers are relatively insulated from debt and so from neo-liberalism, unlike their Latin American counterparts. They limited the operation of international capital markets until Korea began to liberalize in the mid-1990s, which contributed greatly to the 1998 Asian crisis (Gowan, 1999; Brenner, 2002). The current growth of democracy is unlikely to weaken state commitment to low inequality, because this may be seen as benefiting the electoral majority.

Most of the states and cultures in East Asia shared collectivist yet hierarchical institutions often inaccurately labelled as ‘Confucian’. But 20th century wars brought great disruption by eliminating landowning elites who became discredited for their collaboration with the overthrown European and Japanese Empires. Semi-authoritarian governments received substantial aid from the USA, seeking to build up their economies as counterweights to the power of communist China and the USSR except for Hong Kong (an entrepôt economy) these states promoted and subsidized export-oriented growth. Encouraging domestic investment, they avoided debtor relations with the West until briefly in the late 1990s. It is common to label these as ‘development states’ (except for Hong Kong). They have relatively homogenous populations in terms of ethnicity, language or religion, and this enabled them to develop inclusive national school systems and high national literacy (much more than in Latin America). This promoted relative economic equality, as we see from their first Gini coefficients. These states borrowed ‘best practices’ from each other, developing a common capacity to coordinate in state committee rooms the main national power actors. This system is sometimes called ‘embedded autonomy’ (Evans, 1995), sometimes ‘governed inter-dependence’ (Weiss and Hobson, 1995; Weiss, 1999). It involves a top-down corporatism generating dense interactions among elites. These states might possess the capacity to prevent market pressures from disrupting their developmental strategy and social cohesion.

5.4 The Anglo region
The Anglo region experienced the greatest recent surge in inequality. Ideologically this region is characterized by a high tolerance for income inequality, and some very rigid forms of social closure such as race. Economically countries in this region have highly pure capitalist economies. Militarily the Anglo region includes the dominant player on the world stage (the USA). This gives its social model great importance because it is likely to be adopted elsewhere. Politically ‘liberal’ welfare states characterize the Anglo region.

After World War II these countries had initially been the most equal. The Ginis for household gross income were lower than the Nordic and European Continental region in the 1950s and 1960s, and they remained lower than the latter up to 1980. This was probably the consequence of victory in World War II through common sacrifice in both the British Empire and the USA. These victories, in sharp contrast to the defeats that swept Europe and East Asia, produced a brief reform surge but little delegitimation of elites. The reform surge soon ran out of steam and never led to significant state intervention in labour markets (Klausen, 1998). The USA ceased reform first; from 1980 their inequality on all our measures increased. The USA still led, followed by the UK, and then in the 1990s by New Zealand, Australia and finally Canada. Luxembourg Income Study data on wage dispersion show big increases in the 1980s and 1990s in the US, UK and New Zealand, with little change in other countries (Smeeding, 2002; Kenworthy, 2004). The most substantial cuts in state spending on welfare occurred in liberal countries. They were greatest in Britain and New Zealand, but had the greatest effect in the US, where already niggardly welfare provisions were cut to the bone (except for the elderly). In Australia Liberal (i.e. conservative) governments introduced radical cuts, Labour governments countered with new welfare programmes, but there were more Liberal than Labour governments, and then Labour also began to cut. Cuts have been most recent in Canada (Hicks, 1999; Huber and Stephens, 2001; Saunders, 2002; Swank, 2002, ch. 6).

Globalizing or demographic pressures were no stronger for the Anglos than for other OECD countries. Indeed, their market income differentials are generally no greater than in other developed countries. Though US pay scales are indeed more differentiated, the USA now compensates with lower unemployment. The difference in inequality probably comes primarily from their states that redistribute less than others do. It is true that Australia and New Zealand also depended heavily on raw material and semi-finished exports and were badly affected by a secular downturn in prices, precipitating early slashing of budgets. More generally, Anglo policies of cyclical macro-economic management got into difficulties in the 1970s, generating stagflation and export of capital. British Labour governments briefly attempted corporatist policies of wage restraint, but these fell victim to voluntarism as neither unions nor employers could be held to agreements. Macro-economic failure, and inability to generate corporatist state-union and state-employer cooperation common elsewhere, more than globalizing or demographic pressures, led the way towards more market-conforming alternatives in the Anglo countries (King and Wood, 1999).

Anglo parties then embraced ideological neo-liberalism in the 1980s. This was especially so under the conservatives Thatcher and Reagan, though Mulroney in Canada also moved in the same direction. They privatized, attacked unions and tinkered with tax codes ‘to reward enterprise’. Some of their rhetoric was then endorsed by their more leftist political opponents, seeking more votes than their shrinking labour base could provide and affected by what they detected as a rightward shift in the country's ideological climate. Blair made market-friendly policies central to ‘New Labour’, as did Clinton's vision of ‘New Democrats’. Chrétien in Canada embraced similar policies, though with less rhetoric. Clinton's welfare reform represented the biggest slashing of American provisions. But the Australian and New Zealand Labour Parties were the first leftist parties to try out market-conforming policies in arenas like international trade, privatization and welfare reform (Swank, 2002, ch. 6). In Britain, Blair acknowledged the influence on him of Australian Premier Paul Keating, though Blair's own electoral successes and the global stature of Clinton then reversed the influence flow. By the end of the 20th century the Antipodean Labour Parties were emulating the liberalizing ‘Third Way’, a trend that continues today. Though the USA remained somewhat insular, new notions of best practices diffused freely among the other Anglo countries. Political and state elites in Australia, New Zealand and (to a lesser extent) Canada often receive their higher education in the UK. This was a collective shift across a single macro-regional culture.

The UK, the US, Canada, Australia and New Zealand form an English-speaking cultural bloc. They had all been part of the British Empire, with British emigrants having provided the core of the colonial settlement. All but the USA remain members of the (British) Commonwealth. They had all entrenched liberal institutions of civil and political citizenship before World War I. Having exterminated most of the natives, these colonists enjoyed rather low levels of material inequality—except for the one major racial difference in the US. Victors in the World Wars, they were unthreatened by revolution, and they peacefully extended what Marshall (1963) called social citizenship to workers in two main war and post-war bursts. These produced ‘lib-lab’ compromises, with the ‘Lab’ element most marked in Britain, Australia and New Zealand, followed by Canada, with the USA trailing. Except perhaps for the US, they shared a perception that liberalism had failed and needed some collectivist corrections for the common good (Klausen, 1998). But though they all institutionalized class compromises, especially in the aftermath of World War II, these were only lightly entrenched in their states. Compromises were more ‘voluntaristic’ than corporatist, so that (as in liberal ideals) parties could freely withdraw from them. This was later to be the main element of ‘lab’ weakness in the lib-lab regime type.⁴

In contrast, the Nordics and Continentals have more corporatist states. Western's (1993) study of work-related welfare benefits shows that in both of them labour unions penetrated into the hearts of states to participate in the actual assessment and disbursement of benefits. Under the ‘Ghent’ and similar systems of disbursement union membership held up much better than among the Anglos where benefits are organized by civil servants, with unions confined to labour market bargaining. The Anglos lack class compromise entrenched in law and the state and buttressed by strong ideologies of social solidarity. As the participation of employers, unions and the state in social citizenship institutions remains voluntary, they can withdraw from them without serious legal or institutional consequences. When Mrs Thatcher refused to invite union leaders into 10, Downing Street, she ended one joint bargaining institution at a stroke (her Labour predecessors were famous for solving industrial disputes over beer and sandwiches at No. 10). It was also easier to decentralize Australian than Swedish wage bargaining. Lib-lab institutions more easily broke under pressure to neo-liberalism because most were voluntaristic. In contrast, in both the Nordic and the European countries, the interested parties are more entrenched within the corridors of power. These states are more difficult to shift.

The Anglo region is more globally significant than the number of its states might suggest, as it provided the two successive leaders of the globalizing economy, the UK and the US. The UK was committed at its economic height to the open, unprotected markets nowadays advocated by neo-liberals. The US, we might more cynically say, is committed to neo-liberalism for others. Anglo liberal traditions seem to make them more receptive to neo-liberalism. Yet the USA is something of an outlier within the group, with greater historical autonomy, weaker cultural ties, a weaker labour movement and a unique Superpower status today. Given its importance, it is crucial to underline the exceptional position of the USA even within the Anglo macro-region. According to the census bureau the Gini index increased from above 40 in 1967 to about 46 in 2001. Virtually all of this increased occurred after 1980, and unlike the UK, where income inequality levelled off around 1990, income inequality continued to increase in the USA through 2001 (United States Census Bureau, 2005; Office for National Statistics, 2006). Indeed trends in US income inequality are truly exceptional.

The next two groups of countries that we analyse share a fundamentally similar military situation. Both are protected by the USA. We thus contrast the two primarily in terms of their ideological, economic and political sources of inequality.

5.5 The European continental region
This group of countries possesses a trajectory remarkably similar to the East Asia group, except that inequality here continued to fall after the 1980s, instead of increasing. Ideologically the region is characterized by considerable national cohesiveness, a relative absence of group closure, and Catholicism, either as a majority position or a crucial minority. Economically the Euro-Continentals were marked by corporatism. Agreements among interest groups are institutionalized in the state. European Continental corporatism however appears to be much more egalitarian than its Latin American counterpart. Politically most states in the region possess effective welfare states. Some of the countries in these regions have also pursued effective development strategies.

This region took longer to develop its distinctive tone. Though some of its authoritarian monarchies and churches were the original pioneers of paternalistic welfare policies, World War I then devastated them and led to bloody conflicts between socialism and fascism. Fascism won this battle, but then the USSR and the Anglos bloodily defeated fascism. Though fascism had its own corporatist forms, these were mostly dislodged by military defeat (although debates rage about the importance of the historical legacy of fascist corporatism for its democratic forms). The European Continental region shared one important element with the East Asia case. Like the latter group of countries, their agrarian elites were virtually eliminated through experiences of defeat in mass mobilizing war. Only after 1945 did consolidated liberal democracies emerge in this region, under-girded by a compromise between rightist Christian Democratic paternalism and leftist Social Democracy. This compromise gradually developed similar market, taxation and welfare policies. Given the violence that had earlier enveloped the region, this class-cum-religious compromise remained popular. Countries in the region were linguistically diverse, and there was probably less diffusion of best practices than in either the Nordic or Anglo regions. Nonetheless, the practices were not dissimilar, as their politics centred on rivalry between quasi-socialist and quasi-confessional parties. Italy, France, Spain, Luxembourg and Austria are overwhelmingly Catholic, while Germany, the Netherlands and Belgium are divided between Catholics and Protestants tending to support different parties. Moderate socialists and social Christians combined to develop class conciliation in industry, the labour market and taxation, though redistribution was undercut by religious support for traditional statuses. ‘Christian Social’ would be the most appropriate term for such regimes. While they legitimated moderate levels of inequality, they were uneasy about leaving these to the market. Their response to market pressures might be ambiguous.

It is no surprise that the Nordics proved resilient in the face of recent demographic and global pressures. But continental resilience may seem more surprising. Most of these states are ‘pension-heavy’ with costly pay-as-you-go schemes generating a widespread sense of entitlement (Myles and Pierson, 2001). These welfare states were gradually extended on the insurance principle. As they mature, they reach far more people than do programmes targeted on the poor. It is more difficult to isolate politically the recipients and cut their benefits. Neither the Christian nor the social–democratic parties believe they can win elections with neo-liberal schemes (Swank, 2002, ch. 5). After all, a corporatism embodying a high degree of enforced cooperation is what solved their unhappy history during the first half of the 20th century. Here democracy sustains common resilience among continentals and Nordics—a more general version of the sacred status that the National Health Service still enjoys in Britain and Social Security (for the elderly) enjoys in the US. Obviously, a sustained fiscal crisis produces reform. But these have tended to have more particular causes—Swedish mismanagement in the late 1980s led to rocketing unemployment, German reunification, Italian corruption crises etc. In the Swedish and Italian cases the reforms widened inequality, but only temporarily. If the crisis endures in Germany, this is principally because its Christian Social welfare-state was actually greatly extended over the 1990s to encompass East Germany too. German welfare benefits are now being cut, though it remains unclear whether this will significantly widen inequality.

5.6 Nordics
The countries in the Nordic region experienced a drop in inequality in the 1960s, when the social–democratic state began to reach its maturity. By 1970 they were the most equal countries in the world, and they have remained so since. The only other nations to approach them are very small ones: Taiwan, the Netherlands and Belgium—size also matters. These countries have shared rather similar histories. Ideologically, they became overwhelmingly Protestant yet recently became fairly secular. Except for the Finns, they speak the same Scandinavian mutually intelligible language family. Their cultural unity has a long history but is now reinforced by institutions supervised by the Nordic Council of Ministers. About 30 trans-Nordic institutions coordinate regional activities ranging from folklore to energy needs to statistics gathering—a vast store of information about ‘best practices’. Economically these countries have succeeded in specializing in competitive high-value added goods like cell phones and luxury cars. Politically they have established social–democratic states based on low unemployment and active labour market policies.

Their highly concentrated urban-industrial centres, their historic estate assemblies and radical middle-class Protestant sects assisted the mobilization of labour, farmers and the middle class, compromising class and rural–urban conflict through corporatist states protecting the population from their common high level of exposure to international trade. During the World Wars, most remained neutral but deeply affected, which further assisted class compromise either during or after the wars. From the 1930s labour—including middle-class workers—and farmers were brought inside state committee rooms in legalized rituals of power sharing (Esping-Andersen, 1985; Baldwin, 1990). Agreements were enforced through state ‘corporatist’ agencies embracing the main interest groups. Though each country had its peculiarities, they all gradually embraced social democracy. This took several decades to mature into a comprehensive statist production regime, intervening in markets, redistributing between the classes, and finally moving into redistribution between men and women (Klausen, 1998, ch. 5; Huber and Stephens, 2001). This slowly built up an institutional density that might make the Nordic Region likely to resist later pressures perceived as widening inequality.

The USA exerts relatively little geo-political influence over either the Nordic, or Euro-Continental regions, in contrast to Latin America. It also exerts little cultural influence, unlike the case with the Anglo countries (with perhaps the partial exception of Italy where Berlusconi models himself on Reagan). Though American politicians, bankers and journalists persistently decry the rigidity of the German labour market and the burden of the Swedish welfare-state, their rhetoric is not grounded in power or influence. Though the Bundesbank and the convergence criteria of the EU add their own pressures on welfare-state spending, they also reduce Europe's debt. Unlike Latin America, Europe does not go cap in hand to the IMF. The Washington Consensus crosses the Atlantic but not the English Channel.

Note also the different temporal rhythms of these macro-regions. The Nordics saw fairly steady, cumulative, movement towards social democracy lasting over several decades. The Anglos in two bursts added labour reforms to liberalism. The Continentals saw a sudden mid-century compromise solution to their hitherto violent history, but then a slower maturation of its welfare policies. Latin America saw relatively little change to early extreme exploitation. East Asia also saw post-war reforms reinforcing social cohesion, but under authoritarian statism gradually becoming more democratic. South Asian post-colonial development has been later and more contradictory. Thus macro-regional differences were not constants, but were rather the result of particular historical trajectories.

	6. Conclusion

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
Our macro-sociological analysis seeks to underline two features of income inequality in the contemporary world. First, national level income inequality varies by region. Neither individual nation-states nor the world as a whole are adequate units of analysis. Rather scholars should more explicitly focus on clusters of countries. Second, there are many causal processes that produce inequality. We have identified four: ideological, economic, military and political. In different regions and at different times these causal processes may vary. Thus very different processes in South Asia and Latin America have produced similar high levels of inequality (but different trends). We would suggest then that the literature look at a number of more contingent historical processes, such as victory and defeat in war, the mode of colonial penetration, cultural diffusion and contingent policy changes to explain income inequality. The rather narrow focus of most debates leads to two questionable conclusions in our view. First, many equate global economy and society with the experience of only countries in the Anglo region. Yet convergence to an Anglo model has not occurred. Second, many see Nordic social democracy as the only egalitarian alternative to neo-liberalism. Yet the record of East Asian and continental European countries shows social democracy and even democracy as such is not the only levelling force in the modern world. All is not lost for equality if socialism declines.

In fact, differences between macro-regions remain strong in the new millennium. The range among the average Gini indices for the regions has not lessened. There is neither convergence among regions, nor simply greater polarization between North and South. All this contradicts most neo-classical, leftist and globalization theories. Against neo-classical theorists, the highly unequal Latin American regimes remain so despite opening up to international trade; and relatively equal regimes in both North and South remain so (Swank, 2002, produces parallel arguments). Politically entrenched macro-regional ideologies affect income inequality independently of the economic or demographic pressures. We do not doubt that the latter have put pressure on labour markets and states. But states have adapted policies to changed circumstances without abandoning past conceptions of best practices.

How long can states withstand current pressures on labour markets and budgets? Perhaps inequalities showed up first among the Anglos and are now diffusing elsewhere. This is an especially plausible argument given the (perhaps now waning) power of the US. Trends from the mid-1980s to the mid-1990s would be consistent with this, but the trend may then have petered out at least in the UK (Smeeding, 2002; Atkinson, 2003). Using an EU Household Panel study of post-tax income in the 15 EU member states during 1993–96, Alvarez-Garcia et al. (2002) found four countries in which inequality monotonically decreased (Germany, Luxembourg, Portugal and Austria), four in which it decreased less consistently (Belgium, Italy, France and Greece), three seeing no significant change (Denmark, the UK and Spain), two with an unsteady increase in inequality (Ireland and the Netherlands), and none with a monotonic increase. All the countries of declining inequality were continentals (including Portugal and Greece, which we excluded). By 1996 the most equal country was Sweden, followed by Denmark and Finland—their three Nordic countries. Then came Germany, the Netherlands, Austria, France, Luxembourg (1995 data only), Italy and Belgium, all continentals. Then came the UK, the only Anglo country in the study.⁵ Macro-regions also persist in their data.

Recent economic and demographic pressures have generated moderate-level fiscal crises whatever the regime (though indebtedness worsened things). A new neo-liberal solution was proffered, initially in the USA. Two macro-regions, the Anglos and the Latin Americans, were ideologically or geo-politically vulnerable and implemented much of it quite early. Fiscally it mostly worked, though it widened inequality. Other macro-regions experienced similar pressures but were not so sympathetic to the remedy. For a time, they poured money into counteracting market-driven inequality, but the ensuing crisis forced them to backtrack, though without abandoning their overall tax and welfare regime. This widened inequality later and less than in the Anglo cases. Some countries experienced sharper crises for more particular reasons (Antipodean raw materials price falls, Anglo–American stagflation, German unification, Italian corruption scandals, Argentine political divisions, Korean opening up to foreign capital etc.). But such crises have not usually endured nor are they remotely comparable in scale to earlier crises consequent upon war, conquest or the clash of political extremes. That is why recent trends towards greater inequality have been much weaker than the earlier reverse trends. It is also why the ensuing reforms have been so ‘path-dependent’. Current demographic pressures will almost certainly bring further fiscal difficulties in most OECD countries, as their populations age and as young people stay in education longer, but political reactions may remain as varied as in the past. It remains open whether and to what extent inequalities will widen in consequence.

There is no global consensus on a single ‘correct’ regime of political economy. Neo-liberalism, the latest contender, seems to have peaked in influence. In any case, regimes serve power constellations more than they do ‘efficiency’. Each contains what foreigners might consider excess baggage—labour market rigidities, mañana culture, corruption, too many lawyers or prisoners, too much or too little inequality, etc. As their practices converge towards a few macro-regional cultures, there is insufficient global competition to force them to adopt the kind of super-efficiency, which economists often expect. Countries and regions will continue muddling through moderate-level crises, adapting current conceptions of best practices to new pressures until more serious crises come along. Then we might see the kind of deeper macro-sociological restructuring which 60 years ago sparked off our macro-regions along their varying trajectories.

We would like to thank Juan Fernandez for his assistance on the data displays.

	Notes

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 
¹ Some have modified Esping-Andersen' categories, distinguishing ‘northern’ and ‘Southern’ (or ‘Latin’ or ‘Mediterranean’) tiers within his conservative corporatist group. Others say that Australia and New Zealand do not fit and suggest an ‘Antipodean’ or ‘wage earner’ welfare-state type. Individual countries are also sometimes identified as deviants, like the Netherlands, Italy or Canada (Castles, 1985; Huber and Stephens, 2001, 85–112, 169–180). We broadly follow Esping-Andersen's types, noting deviant countries where we find them.

² We constructed a table with the descriptive statistics and distributions of observations by country on which these box-plots are based. We have also compiled a similar table using the UTIP imputed Gini coefficients. These tables are available upon request from Dylan Riley.

³ The distribution of the F-statistic can be used to test the ‘null hypothesis’ that the samples are derived from an identical underlying population. If the P-value is less than 0.05 we can, by convention, reject the null hypothesis that our samples were drawn from an identical underlying population. Thus we report P-values. However strictly speaking there is no sample here, and so inferential statistics are not, strictly speaking, appropriate.

⁴ Some say one cause of greater inequality in the Anglo countries is more centralized, unicameral and majoritarian political systems able to force through neo-liberal reforms without having to overcome the numerous political veto-groups entrenched in less centralized polities. This might account for both the Anglo surges, towards equality in the 1950s and away from it in the 1980s. However, the argument is undermined by the fact that Canada and the USA have decentralized polities. The Anglos actually have varied polities on these dimensions.

⁵ The most unequal were Ireland, Spain, Greece and Portugal, which were all excluded from our analysis.

	References

TOP 1. Introduction 2. International income... Notes 3. Method 4. Data 5. Explaining macro-regional... 6. Conclusion References

 

Alderson, A. S. and Nielsen, F. (1999) ‘Income Inequality, Development, and Dependence: A Reconsideration’, American Sociological Review, 64, 606–631.[CrossRef]

Alderson, A. S. and Nielsen, F. (2002) ‘Globalization and the Great U-Turn: Income Inequality Trends in 16 OECD Countries’, American Journal of Sociology, 107, 1244–1299.[CrossRef]

Alvarez-Garcia, S., Rodriguez, J. P. and Salas, F. (2002) ‘The evolution of income inequality in the European Union during the period 1993–96’ [cited March 31, 2006], Working Paper of the Instituto de Estudios Fiscales. Available at http://www.ief.es/.

Arrighi, G. (1994) The Long Twentieth Century, London, Verso Books.

Arrighi, G. and Silver, B. J. (1999) Chaos and Governance in the World System, Minneapolis, University of Minnesota Press.

Arrighi, G. and Silver B. J. (2003) ‘Industrial Convergence, Globalization and the Persistence of the North-South Divide’, Studies in Comparative International Development, 38, 3–31.

Atkinson, A. B. (1975) The Economics of Inequality, Oxford, Clarendon Press.

Atkinson, A. B. (2003) ‘Income Inequality in OECD Countries: Data and Explanations’, CESifo Economic Studies, 49, 479–513.[Medline]

Baldwin P. (1990) The Politics of Social Solidarity: Class Bases of the European Welfare State 1875–1975, Cambridge, Cambridge University Press.

Bauman, Z. (1998) Globalization: The Human Consequences, Cambridge, Polity Press.

Baumann, R. (2001) ‘Brasil en los años novena: una economia en transicion’, Revista de la CEPAL, 73, 149–171.

Béteille, A. (1977) Inequality Among Men, Oxford, Basil Blackwell.

Béteille, A. (1983) The Idea of Natural Inequality and Other Essays, Delhi, Oxford University Press.

Berry, A. (ed.) (1998) Poverty, Economic Reform and Income Distribution in Latin America, Boulder, CO, Lynne Rienner.

Boix, C. (1998) Political Parties, Growth and Equality : Conservative and Social Democratic Economic Strategies in the World Economy, Cambridge, NY, Cambridge University Press.

Brenner, R. (1998) ‘The Economics of Global Turbulence’, New Left Review, 229, 1–264.

Brenner, R. (2002) The Boom and the Bubble. The US in the World Economy, London, Verso.

Bulmer-Thomas, V. (ed.) (1996) ‘Introduction’. In The New Economic Model in Latin America and its Impact on Income Distribution and Poverty, New York, St Martin's Press, pp. 7–26.

Burtless, G. (1995) ‘International Trade and the Rise in Earnings Inequality’, Journal of Economic Literature, 33, 800–816.

Castells, M. (1998) The Information Age. Vol 3: End of Millenium, Oxford, Blackwell.

Castles, F. (1985) The Working Class and Welfare, Sydney, Allen & Unwin.

Centeno, M. (2002) Blood and Debt: War and the Nation-State in Latin America, University Park, Pennsylvania, The Pennsylvania State University Press.

Chan, S., Clarck, C. and Danny Lam. (1998) Beyond the Developmental State: East Asia's Political Economy Reconsidered, New York, St Martin's Press.

Cornia, G. A. and Addison, T. with Kiiski, S. (2003) ‘Income Distribution Changes and their Impact in the Post-World War II Period’, Discussion Paper of UNU-WIDER, No. 2003/28 [cited September 9, 2005]. Available at http://www.wider.unu.edu/publications/publications.htm.

Cornia, G. A. with Kiiski, S. (2001) ‘Trends in Income Distribution in the Post-World War II Period: Evidence and Interpretation’, Discussion Paper of UNU-WIDER No. 2001/89 [cited September 9, 2005]. Available at http://www.wider.unu.edu/publications/publications.htm.

De Ferranti, D. M., Perry, G. E., Ferreira, F. H. G. and Walton, M. (2004) Inequality in Latin America: Breaking with History?, Washington, DC, The World Bank. Available at http://www-wds.worldbank.org.

Deininger, K. and Squire L. (1996) ‘A new data set measuring income inequality’, The World Bank Economic Review, 10, 3.

Desphande, A. (2000) ‘Does Caste Still Define Disparity? A Look at Inequality in Kerala, India’, The American Economic Review, 90, 322–325.

Dobb, M. (1963) Studies in the Development of Capitalism, New York, International Publishers.

Dollar, D. and Kraay, A. (2004) ‘Trade, Growth and Poverty’, Economic Journal, 493, 22–49.[CrossRef]

Esping-Andersen, G. (1985) Politics Against Markets, Princeton, Princeton University Press.

Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism, Princeton, Princeton University Press.

Esping-Andersen, G. (1999) Social Foundations of Postindustrial Economies, Oxford, Oxford University Press.

Evans, P. (1995) Embedded Autonomy: States and Industrial Transformation, Princeton, NJ, Princeton University Press.

Ferreira, F. and Litchfield, J. (1996) ‘Inequality and Poverty in the Lost Decade: Brazilian Income Distribution in the 1980s’. In Bulmer-Thomas, V. (ed.) The New Economic Model in Latin America and its Impact on Income Distribution and Poverty, New York, St Martin's Press, pp. 249–270.

Firebaugh, G. (1999) ‘Empirics of World Income Inequality’, American Journal of Sociology, 104, 1597–1630.[CrossRef]

Firebaugh, G. (2000) ‘The Trend in Between-Nation Income Inequality.’ Annual Review of Sociology, 26, 323–339.[CrossRef][Web of Science]

Fourcade-Gourinchas, M. and Babb, S. L. (2002) ‘The Rebirth of the Liberal Creed: Paths to Neoliberalism in Four Countries’, American Journal of Sociology, 108, 533–579.[CrossRef]

Galbraith, J. K. and Kum, H. ‘Estimating the inequality of household incomes: a statistical approach to the creation of a dense and consistent global data set’, Review of Income and Wealth, 51, 115–143.

Garrett, G. (1998) ‘Shrinking States: Globalization and National Autonomy in the OECD’, Oxford Developmental Studies, 26, 71–98.

Goesling, B. (2001) ‘Changing Income Inequalities within and between Nations: New Evidence’, American Sociological Review, 66, 745–761.[CrossRef]

Goldthorpe, J. H. (2002) ‘Globalisation and Social Class’, West European Politics, 25, 1–28.[Abstract/Free Full Text]

Goodin, R., Heady, B. Muffels, R. and Dirven, H. J. (1999) The Real Worlds of Welfare Capitalism, Cambridge, Cambridge University Press.

Gottschalk, P. and Joyce, M. (1998) ‘Cross-national differences in the Rise in Earnings Inequality-Market and Institutional Factors’, Review of Economics and Statistics, 80, 489–502.[CrossRef]

Gottschalk, P. and Smeeding, T. M. (1997) ‘Cross-National Comparisons of Earnings and Income Inequality’, Journal of Economic Literature, 35, 633–687.[Web of Science]

Gowan, P. (1999) Global Gamble. Washington's Faustian Bid for World Dominance, London, Verso.

Gustafsson, B. and Johansson, M. (1999) ‘In Search of Smoking Guns: What Makes Inequality Vary over Time in Different Countries?’ American Sociological Review, 64, 585–605.[CrossRef]

Hall, P. and Soskice, D. (2001) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage, New York, Oxford University Press.

Held, D. and McGrew, A. (2000) ‘The Great Globalization Debate: An Introduction’. In The Global Transformations Reader, Cambridge, Polity, pp. 1–45.

Hicks, A. (1999) Social Democracy and Welfare Capitalism: A Century of Income Securities Politics, Ithaca and London, Cornell University Press.

Higgins, M. and Williamson, J. G. (1999) ‘Explaining Income Inequality the World Round: Cohort Size, Kuznets Curves, and Openness’, NBER Working Paper Series, w7224 [cited March 30, 2006]. Available at http://www.nber.org/papers/.

Hobsbawm, E. (1994) The Age of Extremes, New York, Pantheon.

Huber, E. and Stephens, D. J. (2001) Development and Crisis of the Welfare State, Chicago, University of Chicago Press.

Inter-American Development Bank. (1999) ‘Facing up to Inequality in Latin America’, Economic and Social Progress in Latin America, 1998–1999 Report, Washington, DC, Inter-American Development Bank.

Jayaraman, R. and Peter L. (1998) The Evolution of Poverty and Inequality in Indian Villages, Washington, DC, World Bank, Development Research Group, Poverty and Human Resources.

Johnson, C. (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975, Stanford, Stanford University Press.

Johnson, C. (1995) Japan: Who Governs? The Rise of the Developmental State, New York and London, W.W. Norton.

Kentor, J. (2001) ‘The Long Term Effects of Globalization on Income Inequality, Population Growth, and Economic Development’, Social Problems, 48, 435–455.[CrossRef][Web of Science]

Kenworthy, L. (2004) Egalitarian Capitalism? Jobs, Income and Inequality in Affluent Countries, New York, Russell Sage Foundation.

Kersbergen, K. (1995) Social Capitalism: A Study of Christian Democracy and the Welfare State, London, Routledge.

King, D. and Wood S. (1999) ‘The Political Economy of Neoliberalism: Britain and the United States in the 1980s’. In Kitschelt, H., Lange P. Marks, G. and Stephens, J. D. (eds) Continuity and Change in Contemporary Capitalism, New York, Cambridge University Press.

Klausen, J. (1998) War and Welfare. Europe and the United States, 1945 to the Present. New York, St Martin's Press.

Korpi, W. (2000) ‘Faces of Inequality: Gender, Class and Patterns of Inequalities in Different Types of Welfare States’, Social Politics, 7, 127–191.[Medline]

Korzeniewicz, R. P. and Moran, T. P. (1997) ‘World-Economic Trends in the Distribution of Income, 1965–1992’, American Journal of Sociology, 102, 1000–1039.[CrossRef]

Mann, M. (1986) The Sources of Social Power, Vol I, Cambridge, Cambridge University Press.

Marshall T. H. (1963) Sociology at the Crossroads and Other Essays, London, Heinemann.

Marx, K. (1984) Capital: A Critique of Political Economy, Volume III, Frederick E. (ed.), New York, International Publishers.

Milanovic, B. (2003) ‘The Two Faces of Globalization: against Globalization as We Know It’, World Development, 31, 667–684.[CrossRef][Web of Science]

Morley, S. (2001) The Income Distribution Problem in Latin America and the Caribbean, Santiago, Chile, CEPAL/ ECLAC

Myles, J. and Pierson, P. (2001) ‘The Comparative Political Economy of Pension Reform’. In Pierson, P. (ed.) The New Politics of the Welfare State, Oxford, Oxford University Press.

Ocampo, J. A., Sanchez, F. and Tovar, C. E. (2000) ‘Mercardo laboral y distribucion del ingreso en Colombia en los anos noventa’, Revista de la CEPAL, 73, 53–78.

Office for National Statistics. (2006) ‘Income Inequality: Fall in Inequality since 2001/02’, [cited May 31, 2006]. Available at http://www.statistics.gov.uk/cci/nugget.asp?id=332.

Panuco-Laguette, H. and Szekely, M. (1996) ‘Income Distribution and Poverty in Mexico’. In Bulmer-Thomas, V. (ed.) The New Economic Model in Latin America and its Impact on Income Distribution and Poverty, New York, St Martin's Press.

Pierson, P. (1998) "Irresistible Forces, Immovable Objects: Post-industrial Welfare States Confront Permanent Austerity", Journal of European Public Policy, 5, 539–560.

Pierson, P. (2001) ‘Coping with Permanent Austerity: Welfare State Restructuring in Affluent Democracies’. In Pierson, P. (ed.) The New Politics of the Welfare State, Oxford, Oxford University Press.

Rhodes, M. (2001) "The Political Economy of Social Pacts: ‘Competitive Corporatism’ and European Welfare Reforms". In Pierson P. (ed.) The New Politics of the Welfare State, Oxford, Oxford University Press.

Rodrik, D. (2002) ‘Feasible Globalizations’, Harvard University Department of Economics [cited March 31, 2006]. Available at http://ksghome.harvard.edu/~drodrik/papers.html.

Rowen, H. (ed.) (1998) Behind East Asian Growth: The Political and Social Foundations of Prosperity, London, Routledge.

Rueda, D. and Pontusson, J. (2000) ‘Wage Inequality and Varieties of Capitalism’, World Politics, 53, 350–383.

Saunders, P. (2002) The Ends and Means of Welfare. Coping with Economic and Social Change in Australia, Port Melbourne, Cambridge University Press.

Smeeding, T. M. (2002) ‘Globalization, Inequality and the Rich Countries of the G-20: Evidence from the Luxembourg Income Study (LIS)’, LIS Working Papers #320 [cited March 31, 2006]. Available at http://www.lisproject.org/publications/wpapersg.htm.

Smith, A. (1976) An Inquiry into the Nature and Causes of The Wealth of Nations, Chicago, The University of Chicago Press.

Stallings, B. (1992) ‘International Influence on Economic Policy: Debt, Stabilization, and Structural Reform’. In Haggard, S. and Kaufman, R. R. (eds) The Politics of Economic Adjustment: International Constraints, Distributive Conflicts, and the State, Princeton, NJ, Princeton University Press.

Swank, D. (2002) Global Capital, Political Institutions, and Policy Change in Developed Welfare States, New York, Cambridge University Press.

Tsai, P. L. (1995) ‘Foreign Direct Investment and Income Inequality: Further Evidence’, World Development, 23, 469–483.[CrossRef]

United Nations University WIDER (World Institute for Development Economics Research). (2000) World Income Inequality Database, Version 1.0, 12 September 2000 [cited January 15, 2002]. Available at http://www.wider.unu.edu/wiid/wiid.htm.

United States Census Bureau. (2005) Historical Income Tables: Income Inequality Table IE-I. [cited September 9, 2005]. Available at http://www.census.gov/hhes/www/income/histinc/ie1.html.

University of Texas Inequality Project. (2000) UTIP World Time Series 2000 [cited January 15, 2002]. Available at http://utip.gov.utexas.edu/.

Western, B. (1993) ‘Postwar Unionization in 18 Advanced Capitalist Countries’, American Sociological Review, 58, 266–282.[CrossRef]

Weiss, L. (1999) The Myth of the Powerless State, Oxford, Polity Press.

Weiss, L. and Hobson, J. (1995) States and Economic Development, Oxford, Polity Press.

Wildeboer Schut, J. M., Vrooman, J. C. and de Beer, P. T. (2001) On Worlds of Welfare. The Hague, Dutch Government, Social and Cultural Planning Office.

Wood, A. (1994) North-South Trade, Employment and Inequality: Changing Fortunes in a Skill-Driven World, Oxford, Clarendon Press.

CiteULike    Connotea    Del.icio.us    What's this?

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 5/1/81    most recent mwl017v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Google Scholar Articles by Mann, M. Articles by Riley, D. Search for Related Content Related Collections N30 - General, International, or Comparative D31 - Personal Income, Wealth, and Their Distributions Social Bookmarking          What's this?

Online ISSN 1475-147X - Print ISSN 1475-1461

Copyright © 2009 Society for the Advancement of Socio-Economics and Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics