Increasing inequality-both global and domestic-has become a central concern of social scientists and politicians in the past 20 years, especially in the Anglo-Saxon world.
It is not difficult to see why. From the end of the second world war until the early 1980s, Britain and America enjoyed a long period of rapidly rising prosperity. This prosperity had been shared by all. The distribution of income was stable, with no marked trend either up or down. Then in the early 1980s there was a sharp change. Unemployment soared, reducing the incomes of the poor. As financial markets opened up and Wall Street and the City boomed, earnings differentials widened.
An influential cottage industry has developed, arguing that these trends are permanent features of western society. The articles write themselves, selecting from a list of phrases: "globalisation," "social exclusion," "the 40-30-30 society," "life-long learning," and so on.
The theme of this report is that the conventional wisdom is unduly pessimistic. It exaggerates the extent of the increase in inequality and downplays the countervailing factors-such as the fall in unemployment (on domestic inequality) and, even more important, the rise of east Asia (on global inequality). From a more theoretical perspective, these jeremiads offer a static, fixed view of society which does not correspond to reality.
Measuring inequality
Summarising the degree of equality in a society is tricky. A range of factors might be considered. A familiar one is the snapshot of the degree of inequality in income distribution at a given time. But we could also think about the degree of social mobility. The evidence suggests that a certain amount of mobility does exist. The probability of an individual making a large move in the income distribution is small, but the concept of an immutable stock of "the poor" is at odds with reality.
Most of the debate, however, is focused on measurements of income inequality at a given point in time. The most widely-used measure is the so-called Gini coefficient. The value of the Gini can only lie between zero and 100 (or 0 and 1). In a completely equal society, the Gini coefficient is zero-no inequality-and in a society in which one person has all the income it is 100. So the higher the value, the more unequal the society. The details of its calculation need not concern us here, but what it does is to divide a society up into groups according to their share of total income, and measure how much the result deviates from a completely equal society. The poorest 10 per cent of the population might, for example, have 3 per cent of its total income instead of the 10 per cent they would have in an equal society, the next poorest 5 per cent, and so on. (The Gini measures the degree of inequality in a country: thus, in terms of absolute levels of income, the poor in a country which is rich but unequal can be better off than most people in a poor but more equal society.)
As with many concepts in the social sciences, translating it into practice is not straightforward. The economy and society are not physical systems which we can put on a pair of scales and measure exactly. We have to rely on estimates. The result is that different researchers can and do obtain different values of the Gini coefficient for the same country.
But however measured, its value in Britain and America is now higher than it was 20 years ago. Deininger and Squire reported for the World Bank in 1996 some 693 calculations of Gini coefficients since the war in different countries around the world. For most of the postwar period, the US Gini coefficient was in the low to mid-30s; it now seems to be in the low to mid-40s. Britain is more egalitarian, with a Gini in the mid-20s for many years, now in the low to mid-30s.
But the degree of inequality in the Anglo-Saxon societies must be placed in global context. Taking the world as a whole, the 1996 World Bank study found a minimum value of 18 and a maximum of 63, and offered the following generalisation: "Income inequality is much greater in Latin America and sub-Saharan Africa, which have Gini coefficients in the upper 40s, than in east and south Asia, which have Gini coefficients in the middle-to-upper 30s. The OECD countries, in general, have relatively egalitarian distributions of income, with Gini coefficients around 30, while the eastern European countries have historically had very low Gini coefficients."
In other words, a Gini of about 20 indicates a distribution of income similar to that of the societies of the former Soviet bloc. Most western countries are around 30, and values of 50 or more reveal wide disparities on the South American model. So even after Thatcherism, Britain today does not stand out as a particularly unequal society by world standards. Indeed, we are now at the level which has existed in France, for example, for at least the last 30 years. UN estimates for France in the mid-1970s put the Gini in the mid-40s-similar to the US now. Despite French disdain for Anglo-Saxon capitalism, the signs of inequality in France itself are transparent, from the municipal blocks which surround the main cities, to high unemployment, to big discrepancies in regional income per head. Per capita income in the Ile-de-France is almost 40 per cent more than in nearby Nord-Pas-de-Calais.
From the 1996 World Bank study quoted above, we can extract the estimates made for some 20 western countries. The range of the Gini in developed countries is much narrower than that of the global data set. This suggests that there are stronger constraints on the degree of inequality in a developed democracy than there are in other societies. The single most frequently observed range of the Gini coefficient in the developed world is centred on the value 35, (which is exactly where Britain is). Values above 40 are unusual, suggesting that high degrees of inequality do not persist for long periods. There is, indeed, a powerful force containing inequality in the US at present: full employment. Despite the massive boom on Wall Street, the degree of inequality in the US now is no greater than it was in the early 1990s. It is still high by the standards of western democracies, but it is not getting worse; and on some accounts it is actually falling slightly.
East Asia makes the world more equal
The Jeremiahs are even more wrong when they repeatedly assert that the world as a whole is becoming more unequal. In fact, for the first time in the history of capitalism, the distribution of world income between countries became more equal in the second half of the 20th century. The economic success of east Asia in that period has liberated millions of people from lives of unremitting drudgery and toil, and has sharply reduced world income inequality.
Before looking in more detail at the current distribution of world income we need to go back in time. Estimating Gini coefficients for the early 19th century is not easy. But the most thorough research on national incomes around the world over long periods of time has been carried out by Angus Maddison. Using the data from his monumental work Monitoring the World Economy 1820-1992, we can observe very clear trends in the degree of equality of world income (see chart on p44).
In the early 19th century, average income per head across countries was compressed in a very narrow band: there was a kind of equality of misery. Only a few countries in northern Europe and America had begun to industrialise; the world Gini coefficient appears to have been in the low teens. The subsequent success of capitalism in certain parts of the world led to a marked widening of the degree of world inequality. By the middle of the 20th century, the world Gini coefficient stood at 40.
The club of prosperous nations which had formed by about 1870 had hardly changed by 1950. Scandinavia was an exception as a late entrant, but membership appeared to be fixed. Following its humiliation by the west in the second half of the 19th century, Japan had embarked on a programme of modernisation, but at the outbreak of the second world war it had by no means caught up.
Japanese growth in the 1950s and 1960s was phenomenal, and other, much poorer countries in east Asia gradually began to experience rapid growth in living standards. The net result was a sharp fall in global inequality by the early 1990s. Since then Japan has faltered, but the rest of east Asia, despite the crisis of 1998, has continued to grow. Africa, of course, remains in very serious difficulties. In terms of world inequality, however, events there are dwarfed by developments in Asia. The total population of Africa is well under 1 billion; Asia has a total approaching 4 billion. GDP per head in real terms in Africa as a whole is probably lower now than it was in the late 1960s; over the same period in Asia as a whole-including the two giants, China and India-the figure has almost tripled.
The recent experience of inequality in countries such as South Korea and China mirror the evolution of global inequality over the past two centuries. In the 1950s, Korea was a poor but very egalitarian society. During its phase of phenomenal growth, inequalities opened up and its Gini coefficient approached 40. Now that Korean income is close to the average income levels of the west, prosperity is being more widely shared, and the Gini has dropped back to the low 30s. China is at a much earlier phase of development, and estimates for this country are more than usually unreliable. But it does seem as though the boom in parts of China in recent years is leading to marked increases in inequality.
By far the most important characteristic of capitalist economies, which distinguishes them from all other previously and currently existing societies, is their slow but steady underlying rate of real economic growth. Before the 19th century, increases in living standards of just one percentage point took decades or even centuries to achieve. Since then, per capita growth of around 2 per cent a year has become the norm in the successful market economies.
The impact of moving to the path of sustained growth can be clearly illustrated by a comparison of per capita incomes in South Korea and an African economy. The Ivory Coast is by no means unsuccessful in African terms, but the transformation of South Korea has been stupendous. Even as late as 1970, the two countries had very similar levels of output per head, with Korea being about one-third higher. But it is now nearly nine times higher-around 1,000 per cent.
Predicting the future trend in global inequality is difficult. To some extent, this depends on the outcome of events in eastern Europe. If the former Soviet bloc begins to succeed economically, further downward pressure would be exerted on the world level of inequality. Events in Latin America, too, send mixed signals about future prospects.
Yet this is intellectual speculation, mere "ifs" and "buts" compared to the certainty of what has actually happened. A group of Asian economies has grown rapidly in the closing decades of the 20th century, and world inequality has been reduced. In China and India, average real incomes have approximately doubled in the past 30 years-considerably more in China, on some estimates. Indonesia, with over 200m people, has an average income similar to that of France and Germany in the early years of the 20th century-not rich by contemporary standards, but much, much higher than in most African countries. South Korea and Taiwan are close to current European levels. Further, once a country has managed to create the elusive mix of institutions and motivations which deliver a successful economy, it does not tend to fall back into general poverty. (Argentina in the inter-war period is a possible exception.)
Theories of equality
The difficulties of assessing future movements in the degree of inequality are intensified by the lack of any theoretical framework in which to work. There are, for example, ex post rationalisations of trends over the last couple of decades in Britain and the US. But this is a very particular and limited historical period, and it would be dangerous to try to generalise from it. A range of commonplace economic and social trends can be called up by those who wish to believe that rising inequality is an inherent feature of the new capitalism. For example, economic liberalisation produces higher differentials and more winner-take-all markets; the decline of the public sector and of manufacturing also means, it is argued, a decline in those parts of the economy with narrower income differentials; while the greater spread of market values and perhaps even meritocracy gives greater legitimisation to inequality.
Yet, despite all the reasons why inequality ought to have increased markedly, it has done so only to a limited extent in Britain and America and hardly at all in other developed countries. On the long view, these are just blips.
A general theory which enables systematically accurate predictions to be made of the movements in inequality seems lacking in the social sciences. In economics, general equilibrium theory describes a competitive economy in which prices operate to clear all markets, and supply and demand balance out. In this Platonic market ideal, unemployment is zero and all is well with the world. But the theory allows any degree of inequality to emerge as a solution to its equations.
So far as free market theory considers the concept of equality, it focuses on the Pareto concept. This refers to a state of affairs in which no person can be made better off without someone else being worse off. Enormous intellectual effort has been used in establishing the theoretical conditions under which this so-called Pareto-optimum might obtain. But in the context of most discussions of inequality, the concept itself seems rather bizarre.
From a different perspective, we have Karl Marx. The concept of the immiserisation in absolute terms of the working class, put forward in The Communist Manifesto, has been thoroughly discredited. By the standards of today, even many of the mid-19th century bourgeoisie were paupers. Marx himself noted difficulties with this hypothesis, and postulated that capitalism would result in workers becoming poorer only relative to capitalists. Again, this has not happened.
Yet another thesis on equality between countries over time is provided by the theory of economic growth. Robert Solow formulated this in a seminal article written almost 50 years ago. A strong prediction of this theory is that in the long run, average per capita incomes between countries will converge. But an important recent refinement of this model, the notorious "post neo-classical endogenous growth theory," implies that this will not necessarily be the case. Using postwar data from across the world, economists are unable to decide which of the two theories describes the world better-and a dispassionate observer might conclude that neither is very successful by scientific standards.
Why have no firm theoretical guidelines on the evolution of inequality emerged? The degree of inequality in any given society is the outcome of the interactions of millions of individual agents, both people and companies. Their tastes and preferences are not fixed, as orthodox economics asserts, but depend to varying degrees on the behaviour of others. Agents learn, copy, and adapt their behaviour over time. Institutions alter and change. Grand themes may appear to determine the outcome for many years into the future, but reality is far more complex. Too many factors can influence the degree of equality for it to follow a well-understood, deterministic path.
This helps to explain why so many of the big developments in the past 50 years were not anticipated. The 1930s saw millions of unemployed, but the postwar decades experienced rates of unemployment lower than the wildest dreams of the Keynesian enthusiasts of the 1940s. As late as 1979, conventional wisdom would have dismissed the idea that unemployment might rise to over 3m (in Britain), and would have been even more sceptical about the government under which this happened getting re-elected. In the early 1990s, very few of us imagined that full employment would return to large parts of Britain by the end of the decade. On a more grandiose scale, the sudden collapse of the Soviet bloc came as a surprise even to free market ideologues. The rise of the east Asian economies was sudden and dramatic.
In short, we cannot anticipate how the degree of inequality will move over the next, say, ten years, either within a single country or across the world. Like many phenomena in the social sciences, inequality appears to have the characteristics of what is known as a complex system. There is an inherent lack of predictability. Whatever the method used, it is not possible to make consistently accurate forecasts. The distribution of outcomes over time may be reasonably stable, so that we can make meaningful statements about the range which the system will explore. But we cannot say with confidence where it will be at any moment.
An analogy might help to illuminate the issue. We might think of the problem as, say, attempting to decide whether a particular radio signal received contains genuine information (say, a piece of music) or is simply a combination of random squeals and hisses. Modern mathematical techniques can identify the proportion of the sequence which contained recognisable patterns, which could therefore be presumed to be genuine music (the "signal"), and the proportion which was simply interference (the "noise").
The existence of a relatively high degree of "signal" to "noise" is a necessary condition for reasonable forecasts of the relevant data series to be made. A series dominated by noise is very similar to a purely random series which, by definition, cannot be consistently forecast with any degree of accuracy.
Analysis of, say, annual postwar data on inequality in the US, or year-by-year movements in world inequality, gives very clear results. The data appear to contain a small amount of information, but they are actually dominated by noise rather than signal. The implications for predictions of the degree of inequality are clear. No matter what ex post rationalisation is given for the past, and no matter how derived, it is simply not possible to make accurate forecasts about the future.
The certainty about the end of equality which pervades much of the Anglo-Saxon intelligentsia is thus misplaced. The workings of society are too complex to permit this, no matter how skilfully events can be rationalised ex post. And on a more mundane note, the empirical evidence runs completely against preconceptions. On a global scale, the second half of the 20th century, especially the final quarter, saw a sharp reduction in the degree of world inequality. Even now, by western standards the degree of inequality in Britain is not particularly high.
