The leaders of the Group of 7 declared that poverty in developing countries has declined from 1990 to 1998 and cited data to support this from the MDB Report and the IMF. However, upon carefully examining the yardstick used to measure poverty and the footnotes in the MDB report, it appears that poverty is actually increasing!

by Chakravarthi Raghavan

Geneva, 26 July 2000 -- The percentage of the poor in developing countries have declined from 29% in 1990 to 24% in 1998, the leaders of the Group of 7 industrialized nations declared in their summit communique from Okinawa on 23 July.

They cited as the source for their data, the Report on Poverty Reduction by Multilateral Development Banks (MDBs) and the International Monetary Fund (IMF).

To expect the leaders of the G-7 to read even the MDB report (17 pages plus a page of appendix) is perhaps too much.  But even their sherpas (deputies preparing for the summit and agreeing on the final draft communique) either did not read that report or cited material selectively to speak about the decline in poverty and make their bosses look good.

The report of the MDBs is nuanced, to present a positive picture, but at the same time ensuring that the MDBs can be kept in the poverty business. (Report can be found at

At the time of the Copenhagen Social Summit, it was clear that any line on poverty, based on per capita incomes, even on the imprecisely calculated purchasing power parity terms, would be impossible and would not be valid. Hence the summit decision that each nation must define its poverty yardstick, and on that basis estimate the numbers below the poverty line and the policies and targets to eradicate it.

Uncomfortable with such a yardstick which is not easily manipulable through computer-generated econometric data and models to show that their policies are working, the World Bank came up with the $1 a day per capita PPP, as the line for ‘absolute poverty’ - and this was quickly adopted by the UN system (including the United Nations Secretary-General in signing on to the 'Better World for All Report').

But in drawing up the Global Poverty Report, the World Bank and its staff tried to cover themselves up with some qualifications and footnotes to suggest regional variations.

While making the claim of a reduction in poverty, if the yardstick to measure poverty and the footnotes are read carefully, the report is more about increasing poverty (and the need for the World Bank, and the MDBs to remain in business).

The report says:

“4. Globally, the proportion of people living in poverty declined from 29 percent in 1987 to 26 percent in 1998, although the total number of poor remained almost unchanged at around 1.2 billion. The reduction in the incidence of global poverty is attributed to progress  made in East Asia, most notably in the People’s Republic of China, although progress was somewhat reversed during the recent Asian economic crisis and seems to have stalled in China. The performance in three other regions (Africa, Latin America, and  South Asia) shows only moderate or no declines in the incidence of poverty, while the number of people in poverty in these regions has increased. In Sub-Saharan Africa, an additional 74 million people joined the ranks of the poor reaching a total of 291 million in 1998. In Latin America the figure increased from 64 to 78 million.

[Footnote 3 qualifies this further: “This figure rises to 183 million if a poverty line of $2 per day is used, which is more appropriate in the Latin American context. Using that poverty line for all regions yields a number of poor people equal to 2.8 billion, with the largest concentration of poor still in South Asia].

Neither the report nor the footnote explains why the $2 PPP a day, to measure poverty, is more appropriate for Latin America, but $1 PPP is sufficient for Asia and Africa. Is it a case of moving goal posts?

But on the basis of that $1 a day measure, the report says (continuing on in para 4), “... and in South Asia a total of 522 million people live in poverty.  Countries in Eastern Europe and Central Asia experienced an  increase in both the incidence of poverty and the number of people living below the international poverty line. Here, the incidence rose from 2 percent to 5 percent, translating into 24 million people in poverty."

“5. Experience has shown that in countries with lower inequality, growth has a higher impact on poverty reduction. Income inequality within countries has not changed significantly in most regions over the past thirty years, with the exception of countries in the former Soviet Union where negative GDP growth has been accompanied by a large increase in inequality. But differences between countries have increased: the average income of the richest 20 countries is now 30 times that of the 20 poorest countries, as compared to 15 times 40 years ago.”

As Mr. Roberto Bissio, Director of the Uruguay-based Instituto del Tercer Mundo and coordinator of the NGO Social Watch, tracking Copenhagen commitments worldwide, puts it in a comment on the G-7/G8 communique and claims: “One could as well conclude from the same quotes above that poverty has been INCREASING, particularly in countries following the IMF-World Bank model.”

In fact GNP income calculations, and even estimations of house-hold incomes by national sample surveys are long known to pose serious problems in terms of social indicators, and the old paradox of GNP:

The statements about per capita and GDP income growth estimations, and the rise in per capita incomes and reduction in poverty in China have not been without controversy, and the Chinese officials themselves are the first to admit the problems involved in calculation in an economy that is moving from a non-monetized to a monetized economy, and from a centrally planned, and rural agrarian economy run mostly in communes, changing over to a price-based market economy. There are also other problems they are trying to grapple with—the benefits of trading, modernizing and relatively open markets in the coastal regions and the interior which is still backward and lacking external linkages.

The statistics for 1990 to start with are not too reliable. And a lot of the economic “growth” of China is attributable to accounting for activities that went previously unaccounted for. For example, if a commune was feeding its own people, the only production that figured in national accounts was the surplus going to the market (or for state distribution to other areas). But if the same commune is now a cooperative or private activity and there are paid workers, in turn using their income to buy the food, both the salaries and the buying of the food increase the GNP and are thus registered as an enormous growth in income, even if the total production has not grown at all.

Even the Executive Board of the World Bank in the late 1980s and early 1990s was seized of this problem when the Bank staff produced a report (to upgrade China from the IDA loan soft window) that showed that though India and China in 1947 started with comparable levels of income and poverty, and despite subsequent assessments of unreliability of the Mao-era statistics, China had made spectacular progress and rise in per capita incomes. The issue was fudged over, when China’s access to IDA funds was continued albeit at a reduced level.

And there has been the further problem that opening up to the external world, and effects of the Asian crisis, has led to a situation of increasing income-inequalities, and thus difficulties of measuring poverty incidence.

As Mr. Bissio put it, this (the Chinese example) is a massive application of the old paradox of GNP which is now that two women doing their household work are not “economically active”; if they do the same work at each other's house and pay each other 100 dollars a month, the balance for them is zero. But now they have a per capita income of $1,200 a year and the national GNP has increased by $2,400! -SUNS4717

The above article first appeared in the South-North Development Monitor (SUNS) of which Chakravarthi Raghavan is the Chief Editor.

© 2000, SUNS - All rights reserved. May not be reproduced, reprinted or posted to any system or service without specific permission from SUNS. This limitation includes incorporation into a database, distribution via Usenet News, bulletin board systems, mailing lists, print media or broadcast. For information about reproduction or multi-user subscriptions please e-mail < >