Comment on “Debate About Income Inequality” in the UC Atlas of Global Inequality

The “Debate About Income Inequality” is over whether inequality in income across all the world’s citizens (called “global income inequality”) is rising or falling. The UC Atlas discussion of this debate begins by summarizing a 2004 article in the American Journal of Sociology in which Brian Goesling and I used as our point of departure the finding of several earlier studies that global income inequality has declined since about 1980. The aim of our article was not to demonstrate that global inequality had declined but, instead, to identify the leading contributors to change in global inequality from 1980-1998. Here’s what we found (Table 5 of the article):

• Equalizing: The most important factors working to reduce global inequality were faster-than-world average income growth in China and South Asia and slower-than-world average population growth in the Western offshoots (Australia, Canada, and the United States).
• Disequalizing: The most important counter-forces working to worsen global inequality were stagnating economic growth, and faster-than-world average population growth, in sub-Saharan Africa, and faster-than-world average income growth in the Western offshoots.

These findings stand up regardless of whether, on net, the equalizing forces are more powerful than the disequalizing forces.

But it is the “on net” that people want to know about – Is global inequality rising or falling? Hence noted political economist Robert Wade, in his comment on our article, focuses on our reference to what we called the “mounting evidence” in other studies that global income inequality had declined in the last decades of the twentieth century.

The gist of Wade’s comment is that present data simply don’t permit confident claims about the direction of change in global inequality – a rather dramatic shift in his earlier thinking about change in global income inequality, where he had declared confidently that “global inequality is worsening rapidly” (Wade 2001, p. 72) and that “new studies show … a rapid rise in inequality” (p. 73).

Wade is in good company in recanting his earlier claim that global inequality is “worsening rapidly” and, indeed, in providing a note of caution about any unqualified claim about the direction of change in global income inequality. In a recent issue of the Journal of Economic Literature, Sudhir Anand and Paul Segal provide a very thorough and balanced review of evidence on the trend in global income inequality. Their last sentence tells it all: “We conclude that there is insufficient evidence to reject the null hypothesis of no change in global interpersonal inequality over 1970-2000” (Anand and Segal 2008, page 91). In other words, the equalizing and disequalizing forces identified in Firebaugh and Goesling (2004) have offset – or at least the forces are close enough to offsetting that we cannot tell which one is the stronger, given the level of uncertainty in our income estimates.

To appreciate the implications of the last point, note that, if a signal is strong enough, even a weak instrument will tend to pick it up. With respect to global income inequality, then, even with highly imperfect income data we would consistently detect a severe disequilibrium in equalizing and disequalizing forces. Because results across studies are not consistent, we can rule out claims of precipitous change in either direction. That fact is important, because less than a decade ago it was not uncommon to read claims that global inequality was “exploding” or at least “worsening rapidly.”

If the overall trend in global inequality isn’t knowable, what are the right questions?

If we agree that with current data we cannot determine, with any degree of confidence, whether global income inequality is growing or receding, what then? Two good answers are given in the concluding section of the UC Atlas “Debate about income inequality.” First, “the scale of contemporary global inequalities is so large that small changes in inequality are neither here nor there.” Second, “income is only one element of poverty; ill-health, powerlessness and exclusion, and many other factors, are also important.”

I want to offer a third answer. A few years ago I was asked to give a talk at Harvard’s Sociology Department on the evolution of global income inequality. I argued that the fixation on whether global income inequality is rising or falling is misplaced because it misses the singular feature of global income inequality today: Its changing contour. Beginning in the late 18th century and for nearly two centuries thereafter, richer regions of the world led the way in economic growth, resulting in massive between-region and between-nation inequality. Income inequality within nations, by contrast, remained relatively constant or (in some instances) declined with the rise of middle classes.

In today’s world the richer regions are no longer the growth leaders. Large poor countries such as China and India are experiencing faster income growth than rich Western nations. At the same time, inequality is rising within many, but not all, countries in the West and elsewhere.

So we have a reversal of a longstanding trend, from rising inequality across nations and constant or declining inequality within nations, to declining inequality across nations and rising inequality within them. That’s the message of my 2003 book The New Geography of Global Income Inequality.

The contour of this new geography of global income inequality – inequality rising within nations and declining across nations – implies that the link between nationality and income will diminish over time. If present trends persist we will see a continuation of the decline in inequality across nations until average incomes in the large poor nations of Asia begin to approach the world average, so any growth in global inequality in the near future necessarily will be due to rising inequality within countries. Because inequality is declining across nations and rising within them, in tomorrow’s world your income will depend less on your nationality than it does today – but it may well depend even more on your occupation and level of education. Those are important issues that should attract the attention of students of global income inequality (see New Geography for further discussion).

Talks ranged from a discussion of the pathways involved in influencing growing international health disparities to the state of global civil society. In addition, poverty mapping workshops sponsored by Columbia University’s Center for International Earth Science Information Network (CIESIN) were well attended and there was great interest in the use of spatial tools available for exploring and understanding global inequality.

A link to the MGI Book of Abstracts (in PDF format) can be found here.

In addition, selected papers and presentations from the conference have been made available on the California Digital Library (CDL) and can be found at the following link.

review by Brian Fulfrost

There is substantial inequality in access to immunizations between, for example, Egypt, where almost everyone is covered (on the left), and Chad, where more than 40 percent of children are excluded (on the right). Yet the disparities can be as large within some countries as they are across all nations in the sample. In Eritrea, for instance, the richest fifth enjoys almost complete coverage,but almost half of all children in the poorest fifth are excluded.

Part II of the WDR report asks why equity matters. The section primarily focuses on the two consequences of inequality on the development of nations. These include the effects of unequal opportunities in a range of markets. The consequences of unequal opportunities can also impair the development of public institutions. The case study that the report turns to is the situation of children within the caste system in India. The report makes the case that discrimination and stereotyping are mechanisms that reproduce inequality between groups and lower the self-esteem, effort, and performance of individuals. This reduces their potential for individual growth and their ability to contribute to the economy. Figure 5 shows evidence of the detrimental effects of the caste system’s discrimination.

Children from different castes were asked to complete simple exercises, such as solving a maze, with real monetary incentives contingent on performance.The key result of the experiment is that low-caste children perform on par with high-caste children when their caste is not publicly announced by the experimenter. When the caste is made public the children perform significantly worse. If a similar inhibition of talent occurs in the real world, this implies a loss of potential output owing to social stereotyping. The final section of the World Development Report asks how public action can level the political and economic playing fields. For example, the report looks at the way that in many developing countries, state provision of services can magnify—rather than attenuate—inequalities at birth. Because differences in cognitive development start to widen from a very early age,early childhood development initiatives can be central to more equal opportunities.The report refers to an experiment in Jamaica to support this finding.

The experiment focused on undersized children (9 to 24 months) and found that they suffered from lower levels of cognitive development than those of normal height. Nutritional supplements and a program of regular exposure to mental stimulation,helped offset this disadvantage. After 24 months, children who received both better nutrition and more stimulation had virtually caught up developmentally with children who started life at a normal height. This illustrates how decisive and well-designed public action can substantially reduce the opportunity gaps between those least privileged and the societal norm. Investing in the neediest people early in their childhoods can help level the playing field.

Within countries, the Report makes the case for investing in people and expanding access to justice, land, infrastructure, and promoting fairness in markets. Internationally, the report analyses the functioning of global markets and the rules that govern them. It supports the complementary provision of aid to help poor countries and poor people build greater endowments.

On the one hand, the world has become more globalized and much more prosperous in the last 5 centuries. On the other hand, inter-regional inequality has grown. If one tracks per capita GDP of the large regions of the world, the growing disparity is obvious; The richest region was 1.8 times richer than the poorest half a millennium ago, whereas, currently, the richest region has a per capita income that is 20 times the income of the poorest region.

Basu uses the Gini Coefficientto explore income inequality. He finds different results depending on the way the Gini is computed; if we compare per capita income the Gini increased over the last decades. If we compare the income of countries as whole the Gini has been declining almost monotonically since the late 1960s. The latter is driven in large measure by the strong economic growth in China since the late seventies and India since the early nineties.

Basu states that one of the biggest problems of rapid globalization is that the wages of unskilled labor in poor countries will lag behind the wages of skilled labor. Therefore, income inequality within poor nations increases as a consequence of globalization.

He criticizes the traditional objective of policymakers to maximize each country’s per capita income. He proposes another normative criteria which he calls the ‘quintile axiom’. That is, policy makers should be concerned about the income of the poorest 20% of the population. Evaluating an economy using the bottom quintile income makes a large difference in absolute numbers and changes rankings of countries sharply (table below). The first column shows the annual per capita income. The third column shows the annual per capita income of the poorest 20% of the country. For example, using per capita income yields a similar picture for Norway and the US. But the bottom quintile income of the US is almost half that of Norway indicating that the poor in Norway are better off than the poor in the

His defense against critics that this measure, unlike the UNDP the Human Development Index, does not consider the non-income aspects of development is that, in general, quintile incomes will have a closer relation to other wellbeing-indicators, like infant mortality, life expectancy and so on, than per capita incomes. Basu is interested in suggesting a measure that is simple and easy to understand.

Basu urges the need for a new international organization or a division of an existing international organization that helps coordinate inter-country anti-inequality policies. Achieving greater global equality may require the use of policy interventions that are coordinated across countries. “Unilateral effort by a country is likely to cause flight of capital and skilled labor from the country and impoverish those who stay behind”. Economic Game Theory calls this a ‘Prisoner’s Dilemma’. That is, for each country it is individually rational to stay away from actions to reduce global inequality, although the world as a whole would benefit. WTO, ILO, UNEP are agencies designed to reduce such coordination problems. But there is no such agency for anti-poverty and anti-inequality issues.

Kaushik Basu (2006), ‘Globalization, Poverty and Inequality: What is the relationship? What can be done?’ World Development, vol. 34 .

Summary by Christian Lehmann

Objectives of Site

The site aims to provide to the public a unique geographic view of the world for a wide range of data variables. Maps are categorized according to the type of data being mapped. Subjects include, but are not limited to: food, income, wealth, poverty, health, diseases and communication. Each map is described and briefly summarized with adjacent text. There is also very good documentation about the data, definitions, and the process used to create the maps.

Every map is of a different variable, and the sizes of each territory are in proportion to the value of that particular variable. Since cartograms distort boundaries based on the value being mapped, it is often difficult to identify a given country or region. The authors have taken note of this problem and on each map page they provide a link to a color coded and labeled territory map. This territory map assists in identifying the regions and countries in the cartogram maps.

Data and Maps

Maps

The cartogram maps are downloadable as PDF files for printing as well as high resolution PNG image files. The downloadable PDF maps also contain: a table with the top and bottom ranked territories, a bar graph with absolute or relative value of the variable being mapped for each of the twelve regions; and, textual descriptions of the map. Extensive documentation and technical notes on how the maps were created are also provided. At the time of this review (3/8/07) there were 309 maps available for download from the website.

Cartogram of Life Expectancy (2002)

Data

The data used for each map is made available for download as a an Excel spreadsheet or Opendoc format . The source of data used for the Cartograms are from a variety of source, including the World Bank. UN Environmental Programme (UNEP, UN Development Program (UNDP), Central Intelligence Agency (CIA), World Health Organization (WHO) and others.

Innovations and Special Features

The site is unique in that it provides free access to a wide range (and growing) of global cartogram maps on a variety of subjects. These maps provide unique visual insight into the geographic variability of a variety of global datasets. Although cartograms inherently distort territorial boundaries, the authors utilize a recently developed method (Gastner and Newman 2004) that’s preserve the shapes of individual territories and keeps them adjacent to adjacent territories and seas. The code used to cerate the cartograms is also made available for download.

Reference
Michael T. Gastner and M. E. J. Newman (2004) Diffusion-based method for producing density equalizing maps Proc. Natl. Acad. Sci. USA 101, 7499-7504.

Summary by Brian Fulfrost

When comparing to country-specific poverty lines to regional-specific lines country lines have downward biased poverty rates in urban areas and upward biased poverty rates in rural areas as this graph demonstrates.

Despite this difference the empirical results show that the overall poverty level is not significantly affected by this alternative definition of the poverty threshold. The geographic as well as the demographic variability of poverty profiles depend heavily on whether region- or country-specific thresholds are applied.

For region specific poverty lines to be useful there must be identical prices of goods and services as well as uniform norms and consumption habits throughout regions within a given country. Using country specific poverty lines may be inappropriate if there is significant variability in the commodity requirements among regions within a given country, even if prices do not vary.

Norway is used as a case study for the comparison of country v. regional poverty lines. The article found that the poverty rates among young singles and non-western immigrants were significantly higher than what was suggested by previous empirical evidence based on country-specific poverty lines.

Magne Mogstad, Audum Langorgen, Roulf Aaberge (2007): Review essay Journal of Economic Inequality, 2007, 5: 115-122

Summary by Anita Graffius

In section 2, the paper usefully distinguishes different concepts of world income inequality:

1. difference in mean incomes between countries (which continues to diverge)

2. mean national incomes weighted by population (recently converging, but driven by China)

3. interpersonal inequality (rising till 1990s, then falling marginally)

4. vertical (between individuals at different levels of income pyramid) and horizontal (inequality within classes).

The main findings on the effect of global integration on poverty and inequality are these:

i. trade openness generally raises growth but that connection is not universal.

ii. The jury is still out on how growth influences inequality

iii. Opening of trade can effect poor people directly through price changes.

iv. In the 1870-1914 wave of global integration, large scale migration reduced the income gap between rich and poor nations. During the contemporary wave, both labor migration and capital movements have been more constrained. The benefits to capital have been larger than those to labor.

v. Diffusion of technology may widen wage and employment gaps between skilled/semi-skilled and unskilled labor.

Global integration is associated with greater economic and financial volatility, which affects the poorest households most significantly.

In a concluding section on empirical evidence and the policy debate about globalization, inequality and poverty, the authors find there is little evidence of convergence between rich and poor economies. They deduce that globalization can produce winners and losers. A. strategic engagement with globalization can be used to create high skill and high productigvity activities. Passive liberalization, on the other hand may lead to marginalization.

Nissanke and Thorbecke (2006). “Channels and Policy Debate in the Globalization-Inequality-Poverty Nexus.” World Development 34(8): 1338-60.

Summary by Ben Crow

Kapur and McHale analyze four key channels through which international skilled migration affects developing countries through the concepts of prospect, absence, diaspora, and return. The prospect channel focuses on the way in which a prospect or option of emigration affects the decision-making of people in poor countries, e.g., whether or not they’ll actually end up emigrating. Table 1 below shows estimates of skilled workers likely to emigrate and Table 2 shows estimates of emigration rates for the tertiary educated in Africa.

Similarly, the absence channel evaluates the effects on the welfare of “those remaining behind” (TRBs) when skilled individuals actually leave. The prospect of emigration and absence of skilled workers in a developing country may greatly affect their capacity to build domestic institutions, which Kapur and McHale recognize as a critical variable for development. The figure below illustrates the percentage of skilled workers who have emigrated from poor countries to rich countries.

The final two channels, diaspora and return, examine the relationship that skilled workers have with the poor countries that they have left. The diaspora channel shows the effects that overseas resident emigrants have on TRBs, where, depending on the country of origin, emigrants can link domestic residents to international social networks by using their accumulated wealth to invest in home-country projects or act as transnational entrepreneurs. Finally, the return channel looks at how emigrants returning with augmented capital can affect the domestic economy.

Kapur and McHale examine these channels of global migration because they believe these are important factors that have direct correlation with the development of poor countries. They believe that the society of TRBs can be compensated through policy responses. They offer policy options in four broad categories: control, compensation, creation, and connections. The table below shows these classifications. The four policies and roles of rich countries, poor countries, and international organizations are categorized accordingly.

Kapur and McHale fear that the increasing rate of migration from poor countries to rich countries signifies deep problems within developing countries. They offer these solutions to the global migration phenomenon because “the fundamental reality is that [poor] countries need talent to ensure innovation, build institutions, and implement programs – the key pillars of long-term development” (p. 8). They believe that through these policies, migration from poor to rich countries stands a better chance of being beneficial for development.

The Global Migration of Talent: What Does it Mean for Developing Countries? Full Report

Devesh Kapur and John McHale (2005): “The Global Migration of Talent: What Does it Mean for Developing Countries?” Center For Global Development. Washington D.C.

Summary by Ellen Liang

Pranab Bardhan, Professor of Economics at UC Berkeley, explores in this paper the effects that globalization has had on the lives of the rural poor in low-income countries. Bardhan poses the question of how international market structures will ultimately affect the absolute poverty levels for those who live in these rural sectors. He offers supporting evidence for both sides of the globalization debate and insightfully weaves the two together through current events and statistics.

He uses the concepts of trade liberation, export commodity concentration, and comparative advantage to illustrate the impact of globalization on rural poverty.

The Debate:

Bardhan focuses on absolute poverty levels of agricultural and non-agricultural rural sectors in low-income countries and examines the poverty lines of various countries as he introduces the concept of trade liberalization, or free trade. Although rapidly industrializing economies such as China, India, and other Asian countries have seen a large decline in poverty over the last few decades, which many pro-globalization defendants argue, he does not attribute the phenomenon all to globalization. He uses the 1998 post-flood food crisis in Bangladesh as an example. On the one hand, trade liberalization may have helped mitigate the crisis; by introducing private imports, prices on food and commodities were stabilized overall. On the other hand, this caused an end to domestic marketing arrangements and increased the variance of prices. Bardhan questions the outcome of the Bangladesh post-flood food crisis and focuses on those who live in poverty in this country – at what capacity are the poor able to ultimately cope with negative price and income shocks?

This question leads Bardhan to address the notion of the commodity concentration of exports, where countries focus on only a few specialized commodities to export. More than 50 developing countries depend on 3 or fewer primary commodities for more than half of their export. Bardhan points out that exports of such products are often a “curse as well as a blessing for these countries,” (p. 9) because the prices on these exports tend to fluctuate unpredictably and the economy becomes too dependent on them.

Trade liberalization and the commodity concentration of exports leads Bardhan to address the concept of comparative advantage, where countries focus on specific commodities and trade with one another for efficiency. Bardhan expresses the pros and cons to comparative advantage in the global sense. As low-income economies open up and export specialized crops, new opportunities are created that may potentially lift those producing the commodities from poverty. Other crops that the country may lack in comparative advantage, however, will lose out and push the small producers into poverty, assuming that there are no extensive programs of public assistance and services to help these producers reallocate their resources.

Often times the impoverished farmers of traditional crops are ill-equipped to shift by themselves to the new commercial products like fruits, processed foods, etc. Because these products require new storage and transport infrastructures as well as new legal rules, costs would be too high for rural farmers to meet. Bardhan fears that an increasing global network of industrial agri-business will lead to the exploitation of small, rural producers by large corporate marketing chains.

Pranab Bardhan (2004): “Globalization and Rural Poverty” World Institute for Development Economics Research of the United nations University. Helsinki, Finland.

Globalization and Rural Poverty Full Report

Summary by Ellen Liang

In particular, Deaton pointed out that the conversion of life expectancy across countries should be considered with caution. In rich countries, life expectancy gains in the last half-century came largely from reductions in mortality among the middle aged and elderly. In poor countries, however, life expectancy gains are due to reductions in infant and child mortality. Deaton argues that reductions in mortality at different points in the age distribution are not equally valuable. Declining infant mortality could yield the same increase in life expectancy as declining mortality among the elderly. However, do we give both equal weight? For example, welfare gains of declining infant mortality accrue primarily to women; they’d bear fewer children, suffer less from witnessing their children die, and would be free to do other activities. Does declining mortality among the elderly yield similar benefits in wellbeing? Empirically, life expectancy is converging across countries. But without considering the reduction in mortality at different points in the age distribution, this conversion could understate the reduction in inequality.

Concerning the relationship between income growth and health improvements he states that, historically, literature is showing positive correlation between GDP and life expectancy.

According to Deaton, however, economic growth is not the best solution to improving health. There are examples where income growth and relative health improvements are weakly or negatively linked (for instance India and China). Furthermore, the causal linkage remains unclear. Is it health that enables nations to achieve economic growth or is material wealth a precondition to establish health improvements? Deaton argues that people in poor countries are sick, not primarily because they are poor, but because of other social organizational failures. Such failures include health delivery, which is not automatically ameliorated by higher incomes.

http://www.wider.unu.edu/events/wider-annual-lecture-2006-announcement.htm

Angus Deaton (2006):“Global patterns of income and health: facts, interpretations and policies.” 2006 WIDER annual lecture