Are global poverty and inequality getting worse? Dear Martin
22nd January 2002
You have written eloquently in the Financial Times about globalisation. You make three main points. (1) Poverty and inequality on a world scale have both fallen over the past two decades for the first time in more than 150 years. (2) These falls are due to greater global economic integration. (3) The anti-globalisation movement encourages countries to adopt policies that will in fact only intensify their poverty and inequality.
Let us take the first point about trends in poverty and inequality. If you are wrong here, the rest of your argument begins to wobble and, in fact, there are reasons to doubt what you say. On poverty, the World Bank is the main source of numbers. Bank researchers have found that the number of people in absolute poverty (with incomes less than about $1 per day) was roughly constant in 1987 and 1998, at around 1.2 billion. Since world population increased, the proportion of the world's population in absolute poverty fell sharply from around 28 per cent to 24 per cent in only 11 years. This is good news.
But recent research on where the Bank got the 1.2 billion suggests that the method for calculating the numbers is questionable. The effect is probably to understate the true numbers in poverty. How much higher than 1.2 billion we do not yet know.
So what is happening to global inequality? It is widening rapidly, if we compare the average incomes for each country and treat each one as a unit (China = Uganda). Yet income inequality among countries has become more equal, since around 1980, if we compare the average incomes for each country and weight each one by its population. However, this result comes from fast growth in China and India. If they are excluded this measure of inequality shows no obvious trend since 1980.
In any case, this measure-using the average income of each country weighted by population-is interesting only as an approximation to what we are really interested in, which is the income distribution among all the world's people or households, regardless of where they live. The problem is that we do not have good data for the incomes of all the world's people. You say that global inequality amongst households has probably fallen. But the most comprehensive data on world incomes, based on household income and expenditure surveys, find a sharp increase in inequality over as short a time as 1988 to 1993. Some of this may be statistical error; but the results do mean that the balance of probability falls in the direction of increasing global inequality among households.
This conclusion is strengthened by the trends in industrial pay inequality within countries. Pay inequality within countries was stable or declining from the early 1960s to 1982, then sharply increased from 1982 to the present. The year 1982 was a dramatic turning point towards greater inequality within the world's countries.
Doesn't the fast growth of populous China and India create a presumption that world income distribution is becoming more equal? No. At low levels of income, growth has to be fast for a long time before the absolute gap with slow-growing, high income countries begins to fall. The absolute income gap between a developing country with an average income of $1,000 a year, growing at 6 per cent, and a developed country with average income of $30,000, growing at 1 per cent, continues to widen until after the 40th year. China and India are not reducing the gap between their average incomes and the averages of the countries of western Europe, North America and Japan. They are, though, closing the gap with the faltering, middle-income states like Mexico, Brazil, Russia and Argentina, which is why average inequality among countries has become more equal since around 1980. But this reduction in the gap between China and India and the middle-income states is probably offset by widening income inequality within the two giants.
Perhaps all the thunder and lightning about the trends diverts attention from the main issue: the sheer magnitude of poverty and inequality on a world scale. The magnitude is unacceptable, regardless of the trend, and the world development agenda should make inequality reduction (not only poverty reduction) a high priority. Roughly 85 per cent of world income goes to 20 per cent of the world's population and 6 per cent to 60 per cent of the world's population. Can this meet any plausible test of legitimacy? It is difficult to see how it could meet the Rawlsian principle that a given degree of inequality is acceptable if it is necessary for the worst off to become better off.
Integration/globalisation is nothing like the engine of development you say it is. The engine is the advance of technology and the diffusion of technical capacities of people, firms and governments. Some forms of integration may help this, others may hinder it, depending partly on a country's stage of development.
Regards Robert
Dear Robert
25th January 2002
All data on incomes and income distribution are questionable, above all those generated in developing countries. But, contrary to what you say, World Bank researchers have calculated the numbers in extreme poverty-less than $1 a day-on a consistent basis, in recent studies.
The data shows a decline since 1980 of 200m people in the category of the absolutely poor. This is a fall from 31 per cent of the world's population to 20 per cent (not 24 per cent, which is the proportion in developing countries alone). That is a spectacularly rapid fall in poverty by historical standards. It makes a nonsense of the idea that poverty alleviation has been blighted by globalisation.
Now turn to the even murkier area of inequality. Here you argue that if we exclude China and India, there is no obvious trend in inequality. But why would one want to exclude two countries that contained about 60 per cent of the world's poorest people two decades ago and still contain almost 40 per cent of the world's population today? To fail to give these giants their due weight in a discussion of global poverty alleviation or income distribution would be Hamlet without the prince.
You then write that changes in relative average incomes across countries are not what we are really interested in, "which is the income distribution among all the world's people or households." This is wrong in itself. If a country's average income rises rapidly, it does also possess greater means for improving the lot of the poor. Maybe the government refuses to use the opportunity, but a successor government could.
In any case, we do possess data on relative household incomes. In a Foreign Affairs article, David Dollar and Aart Kraay of the World Bank report a big decline in world-wide income inequality since its peak in about 1970. The study builds on work that goes back to 1820. The underlying method is to calculate the percentage gap between a randomly selected individual and the world average. The more unequal the distribution of world income, the bigger that gap becomes. They report that this gap peaked at 88 per cent of world average income in 1970, before falling to 78 per cent in 1995, roughly where it was in 1950.
The chief driver of changes in inequality among households is changes between countries, not within them. This was also the finding of Branko Milanovic's study of global household income distribution between 1988 and 1993, which you cite approvingly. You rely on this study to support the thesis of rising household inequality. But it contains at least four defects. First, there are well-known inconsistencies between data on household expenditures and national accounts. Second, the methods used generate no increase in rural real incomes in China, which is inconsistent with most views of what actually happened. Third, the period of five years is very short. Fourth and most important, this was an atypical period, because India had an economic upheaval in 1991, while China's growth was temporarily slowed by the Tiananmen crisis.
My conclusion is that the last two decades saw a decline not just in absolute poverty but also in world-wide inequality among households. The chief explanation for this was the fast growth of China and, to a lesser extent, of India. This progress was not offset by rising inequality within them. In the case of India there was no such rise. In China there has been a rise in inequality in the more recent period of its growth, largely because of controls on the movement of people from the hinterland to the coastal regions.
Unfortunately, you muddy the waters on inequality by raising the question of growing absolute gaps in incomes between rich and poor countries. If the income of the poor rises faster than the income of the rich, inequality falls, even if absolute gaps rise, since the standard measures of inequality describe relative, not absolute, differences in incomes.
This is vastly more than just a question of definition. China's average incomes per head are only a tenth of those of the US. They would have to grow at around 20 per cent a year to match the absolute increases now prevailing in the US. I see no point in bemoaning the failure to achieve what is impossible. Unless you are suggesting implausibly huge income transfers from taxpayers in rich countries to the world's poor, or complete freedom of migration, absolute gaps in living standards will rise for many decades, even if poor countries now grow very quickly. This is the tyranny of history: we can only start from where we are.
The trends in pay inequality you bring in to support your arguments further cloud the issue. Few of the world's poor earn wages that anybody reports. They work as subsistence farmers or do casual work in informal activities. Almost all reported wage earners are in the upper half of the global income distribution.
Yet this debate is, as you say, not just about measuring poverty and inequality, but about what these trends mean. You write that the magnitude of poverty and inequality are unacceptable. I agree on the former. That is why raising average incomes in poor countries and of poor people in both poor and less poor countries is an urgent goal of public policy.
But your position on the unacceptability of inequality also amounts to saying that the world would be a better place if the rich countries of today had never started rapid development in the 19th and 20th centuries. Maybe you do think this. But almost all citizens of advanced countries do not. They have no intention of doing without what they now have. So bemoaning the magnitude of global inequality, as opposed to the low standards of living of large parts of the world, is just empty rhetoric. It has no significance for action.
Today's global inequality and continuing, though also declining, mass poverty are the outcome of deep-seated historic processes that can be reversed only with vast and sustained improvements in poor countries, supported by rich ones. A start was made in the 1980s and 1990s. But the huge worry concerns those poor countries where there is now no sustained rise in living standards.
Yours Martin
Dear Martin
29th January 2002
On absolute poverty, you take the World Bank figures at face value, I say that they cannot be so taken. On inequality, you cite the work of World Bank economist David Dollar as the main evidence that world income inequality has declined over the past 20-25 years. But his method gives too much weight to what happens in the middle swathe of world population and too little weight to what happens towards the lower and upper ends of the distribution.
Contrary to what you say, inequality in India-rural-urban, intra-urban, intra-rural-increased over the past two decades. And everyone agrees inequality in China has risen rapidly. In a recent year, the ratio of the richest to poorest state or province was 1.9 for the US, 4.2 for India and for China, 7 in the early 1990s, 11 in the late 1990s.
You place too much trust not only in David Dollar's methodology, but also in the Bank's data set on inequality to which it is applied. Anyone applying the "laugh test" would have grounds for doubt: according to the Bank, Spain is the most equal country in Europe; France is much more unequal than Germany; India and Indonesia are in the same equality league as Norway. The data is based on uncoordinated sample surveys, separated in time and space, often conducted by unofficial researchers, in countries with differing concepts of income and differing attitudes towards revealing income to strangers. The Milanovic data based on household income and expenditure surveys around the world also has its problems, as you say, but is not obviously inferior to the Bank's national income data; and it suggests sharply rising inequality.
You say that "bemoaning the magnitude of global inequality, as opposed to the low standards of living of large parts of the world, is just empty rhetoric." No. If the magnitude of inequality is as large and difficult to justify as it seems to be, this greatly fortifies the case for public policy actions-some national, some international-to "tilt the playing field" in favour of the lagging regions. A significantly more equal world is likely to be more stable, peaceful and possibly more prosperous. Also, there's no reason for you to reject measures of absolute income gaps just because these are not relevant to the standard measures of inequality. We should be concerned with both absolute and relative gaps, for both relate to important ethical values, both are relevant to feelings of disempowerment and deprivation. Absolute gaps between, say, the top quintile and the bottom quintile of the world's population, you have to agree, are rising sharply.
So far, all this has been about your first point to do with trends in poverty and inequality. Your second point is about globalisation (or increased economic integration) as the world's best means of reducing poverty and inequality. I doubt it. The most powerful engine of development is the diffusion of technical capacities. This is proceeding at a furious rate in China, more sedately in India, at a snail's pace in most of Latin America, and slower, if at all, in sub-Saharan Africa and much of the middle east. China and India are likely to experience a shift towards more income equality when they come near to full employment, five to ten decades from now. But any such shift in Latin America, sub-Saharan Africa and the middle east will be even further away.
The World Bank studies on which you rely for your conclusions about the benign impacts of globalisation are shot through with problems. They distinguish "globalising" countries from "non-globalising" countries, and find that the former have much better economic performance than the latter. They measure globalising by changes in the ratio of trade to GDP. So globalising countries are ones that had a big rise in their trade/GDP ratio.
Let us accept that the countries the Bank calls globalisers did, indeed, have fast economic growth. The question is: why? At best, the Bank studies show that countries that start being closed, with very low trade/GDP, can have fast growth if they take policy steps that yield more trade/GDP. This sounds plausible, and it matches the experience of South Korea and Taiwan in the 1950s. But the finding does not support the policy prescription that all developing countries should liberalise their trade regimes in order to experience faster growth.
For two decades, the Bank's official view about development has been: adopt a liberal trade policy (low tariff and non-tariff barriers), deregulate markets, privatise state enterprises, welcome foreign firms, maintain fiscal balance and low inflation. The trouble is that there is no evidence that opening to trade does generally result in subsequently faster growth, holding other things like macroeconomic conditions constant.
The best examples of globalising countries are China and India, which are hardly poster-children for globalisation. They have certainly both had fast rises in trade/GDP in the recent period, and also fast economic growth. But the onset of fast growth occurred about a decade prior to their liberalising trade reforms. And the Bank would now be denouncing their trade policies and internal market-restricting policies as growth and efficiency-inhibiting-if they had not been growing so fast. Their policies remain far from those that the Bank seeks to get its borrowers to adopt; in fact, their trade barriers remain amongst the highest in the world. Their experience, and that of Japan, South Korea and Taiwan earlier, shows that countries do not have to have liberal trade policies in order to grow fast. It shows only that as countries become richer they tend to liberalise trade, which is not the same thing. The sensible ones liberalise in line with the growth of domestic capacities-they try to expose domestic producers to enough competition to make them more efficient, but not enough to kill them. China and India suggest a policy regime that is not close to what the Bank says, but nor is it "anti-globalisation."
The China-WTO agreement shows the dangers of pressing free trade upon developing countries. The agreement makes it difficult for China to adopt one of the most powerful inequality-mitigating measures: agricultural subsidies. In Japan, South Korea, Taiwan and, of course, Europe and the US, agricultural subsidies have been an important means of redistributing the fruits of industrialisation. China has had to sign away its rights to all but very low subsidies, with consequences that, given the degree of regional inequality, could be quite explosive. Likewise, China's agreement to give equal access to foreign companies will mean that it cannot protect "inefficient" labour-intensive industries that serve to equalise incomes. It is worrying for the whole world that the Chinese government itself now seems to think it can maintain an urban-rural apartheid state by means of the pass laws, while opening the economy at a pace so fast that unemployment will shoot upwards from its already high levels.
The point is more general. Under WTO rules, developing countries face constraints which prevent them from adopting the measures that already-developed countries (including East Asian ones) deployed to nurture their technological learning. This is outrageous. WTO rules and the Bank's official view need to be revised, soon, in the self-interest of the west, as well as the bulk of the world's population.
Regards Robert
Dear Robert
3rd February 2002
A significant decline has occurred in the proportion of the world population in absolute poverty over the past two decades. I think we agree on this. Whether there has been a fall in absolute numbers is less certain, though also, in my view, highly plausible. Your denial of this latter proposition rests on the view that any data or analyses from the World Bank must be tainted. Yet you rely on another Bank study for the proposition that inequality increased between 1988 and 1993. Your position is that any study which comes to a conclusion you dislike must be rejected (and vice versa).
You stress that absolute gaps between the world's richest and poorest people are rising. I agree. But so what? Even if poor countries grew far faster than rich ones, absolute gaps in living standards would rise for many years. This is the result of two centuries of differential growth. Why bemoan what cannot be helped?
What is needed, you then suggest, is "to tilt the playing field" in favour of lagging regions. There have been many attempts to do this, from carte blanche for protection to substantial aid. None has been notably successful. So what are you proposing? Huge transfers to the world's poor, free migration into rich countries, or a permanent depression in the north? I look forward to your attempts to sell any of these. Yet the very notion that impoverishment of the north might be a good thing shows the absurdity of your obsession with equality in itself. World income distribution was far less unequal two centuries ago, when perhaps 80 per cent of its population lived in extreme poverty. Did this make 1800 better than today?
The only sensible goal must be to raise the standards of living of the world's poorest people as quickly as we can. What is needed for this is faster growth. Look at China and India's own data. Indian data gives a decline of about 100m in absolute poverty between 1980 and 2000. China uses a lower poverty line but reports that the number of extreme poor in China declined from 250m in 1978 to 34m in 1999. Unfortunately, these successes have been offset by calamitous failures elsewhere, notably in sub-Saharan Africa.
I assume you support the need for faster growth in poor countries. Presumably, you also endorse the need for open markets in the north. After all, Taiwan and South Korea developed through exploitation of access to world markets. Yet many in the anti-globalisation movement are against trade liberalisation by the north. Many of them also say that foreign direct investment (FDI) impoverishes the poor. You say that China is not a "poster-child for globalisation." But China has been the biggest recipient of FDI in the developing world. Malaysia, to take another example, has received roughly as much inward FDI as the whole of sub-Saharan Africa.
The causes of developing country growth are complex. But your technological determinism is even more simple-minded than your caricature of the so-called Washington consensus. Many countries devoted much effort to upgrading technological capacity, but have failed to sustain rapid development: the Soviet Union was one and India another. One cannot separate technology from the context in which it is applied.
Among the essential pre-conditions for growth are: a stable state; security of the person and of property; widespread literacy and numeracy; basic health; adequate infrastructure; the ability to develop businesses without suffocation by red tape or corruption; broad acceptance of market forces; macro-economic stability, and a financial system capable of transferring savings to effective uses. In successful countries, these conditions emerge in a mutually reinforcing cycle. There is also evidence, I accept, that some initial equality in income and asset distribution helps.
The tragedy of Africa is that so few of these pre-conditions exist. Do I believe that trade liberalisation would fix this? No. But trade is the handmaiden of growth. There is no country that set out to reduce its reliance on trade, as anti-globalisers propose, and subsequently secured sustained growth. Even in the inauspicious soil of Africa, countries such as Uganda, which tried to exploit market opportunities, have achieved faster growth and poverty reduction.
Finally, you argue that, under WTO rules, developing countries are being forced to forgo their ability to adopt inequality-alleviating or growth-promoting policies. You assume that, in the absence of external constraints on their policy discretion, these countries would choose well-targeted trade policy interventions. This proposition does fail your laugh test. In any case, developing countries can use tariffs if they wish. China could have remained outside the WTO. But the Chinese authorities believed their country would do even better inside. I suspect this judgement will be proved right.
I would invite you to subscribe to the following three propositions. First, the biggest policy challenge is to accelerate economic growth in poor countries. Second, open markets in the north and FDI make an important contribution to such growth. Third, self-sufficiency is a foolish development strategy. If you accept these points, you are on my side of the policy argument with the anti-globalisers, like it or not.
Yours Martin Dear Martin
4th February 2002
I agree with your three propositions and have never argued anything different. You seem wedded to simplistic categorisations of "globalisers" and "anti-globalisers." I am not an anti-globaliser or a friend of protectionist American unions. But there is a fundamental fudge in your argument. It is contained in your category of "globalising" countries-those with fast rising integration into the world economy, like China and India. You take their economic success as evidence of the benefits of integration. You also believe in liberal trade policy with low tariffs and non-tariff barriers; in particular, you believe that trade liberalisation is a powerful force for higher growth, so poor country governments should give high priority in the use of scarce development funds to trade liberalisation and WTO membership. But China and India (Vietnam too) have reaped benefits of integration without having anything like a liberal trade policy, as also have Japan, South Korea and Taiwan before them. So one cannot use their success to support the case for liberalisation.
The single biggest issue in the trend of world income distribution is what has happened in China over the past two decades. It makes a huge difference whether one takes the World Bank's growth rate of 10 per cent at face value, or uses a (more plausible) figure of, say, 7 per cent. If one takes the former figure, global inequality has not got much worse, may have remained constant, and might even have improved. If one uses the latter figure, virtually all the plausible measures of world inequality show a worsening. This underscores the point that the world is ill-served by the Bank being the main producer of development statistics. It has an official view about how to do development and is subject to arm-twisting by its major shareholders. It is under constant pressure to fudge its GDP data base, most glaringly in the case of China and in the case of politically sensitive numbers like the number of absolute poor in the world.
At the heart of our disagreement, I think, is the question about how far rich countries in general should go in using the power our superior resources give us (a) to set the rules of the market so that resource power is translated into market power, and (b) to use that power to the maximum when bargaining with people much poorer than ourselves. You say, "China could have remained outside the WTO." But China faced a choice of taking American terms or facing perpetual uncertainty about the US export market. We need to press the rich country governments to soften the conditions they set for WTO membership, and not just for China.
In relation to countries like China and India, whose ability to absorb and create technology is such that they have a real prospect of reducing the gap with the rich countries in the next five to ten decades, the rich countries should be more generous in making trade and investment concessions to accelerate their growth and allow them to grow in a manner according with their own values. In relation to Africa and the non-oil parts of the middle east and central Asia-countries that lack much of the social technology you list-there has to be more direct aid, much of it in the form of World Bank-type projects on the ground. The FT could help by backing the modest increase of $50 billion in western development aid that Gordon Brown is proposing.
Regards Robert
Dear Robert
6th February 2002
You are convinced that the World Bank has cooked the data on poverty and inequality. You need to produce chapter and verse to substantiate such a serious charge, but have failed to do so. That is not good enough. All you assert is that we do not know what has happened to poverty and inequality, not that they have become worse. I note also that you have not taken up my offer to explain how we are to reduce absolute gaps in living standards in the near future.
I accept that infant-industry promotion, buttressed by trade restrictions, may occasionally accelerate economic growth. However, the record on the use of such policies in developing countries is, with few exceptions, dreadful. Yet even though liberalisation of protectionist trade policy regimes is good for developing countries, I don't claim it is a panacea.
I also fail to see why WTO constraints on policy discretion should be good for rich countries, as we know they are, but not for poor ones. Governments of developing countries are, if anything, more vulnerable to capture by protectionist lobbies than those of advanced countries.
I do accept, however, that developing countries have sometimes been forced to accept inappropriate policies: the trade-related intellectual property agreement is an example. I also agree that the north should liberalise in favour of the south and that more aid, targeted on countries with governments that know how to use it, is a moral and practical necessity.
Yet there is one fundamental matter, in this debate, on which we do disagree. Economic growth is, almost inevitably, uneven. Some countries, regions and people do better than others. The result is growing inequality. To regret that is to regret the growth itself. It is to hold, in effect, that it is better for everyone in the world (or within individual counries) to remain equally poor. You come close to saying just that. It seems to me a morally indefensible and practically untenable position.
Yours Martin
