Global Falsehoods: How the World Bank and the UNDP Distort the Figures on Global Poverty

By Michel Chossudovsky

Professor of Economics, University of Ottawa, author of The Globalisation of Poverty, Impacts of IMF and World Bank Reforms, Zed Books, London, 1997.


© copyright by Michel Chossudovsky, Ottawa, 1999.
To publish this text contact the author at


Until the 1998 financial meltdown ("black September" 1998), the World economy was said to be booming under the impetus of the "free market" reforms.

Without debate or discussion, so-called "sound macro-economic policies" (meaning the gamut of budgetary austerity, deregulation, downsizing and privatisation) continue to be heralded as the key to economic success and poverty alleviation. In turn, both the World Bank and the United Nations Development Programme (UNDP) have asserted authoritatively that economic growth in the late 20th Century has contributed to a reduction in the levels of World poverty. According to the UNDP, "the progress in reducing poverty over the 20th century is remarkable and unprecedented... The key indicators of human development have advanced strongly."1


The Devastating Impacts of Macro-economic Reform are casually denied
The increasing levels of global poverty resulting from macro-economic reform are casually denied by G7 governments and international institutions (including the World Bank and the IMF); social realities are concealed, official statistics are manipulated, economic concepts are turned upside down.


The World Bank Methodology: Defining Poverty at a "Dollar a Day"
The World Bank framework deliberately departs from all established concepts and procedures (eg. by the US Bureau of Census or the United Nations) for measuring poverty. 2 It consists in arbitrarily setting a "poverty threshold" at one dollar a day per capita. It then proceeds (without even measuring) to deciding that population groups with a per capita income "above one dollar a day" are "non-poor".

The World Bank "methodology" conveniently reduces recorded poverty without the need for collecting country-level data. This "subjective" and biased assessment is carried out irrespective of actual conditions at the country level. 3 The one dollar a day procedure is absurd: the evidence amply confirms that population groups with per capita incomes of 2, 3 or even 5 dollars a day remain poverty stricken (ie. unable to meet basic expenditures of food, clothing, shelter, health and education).


Arithmetic Manipulation
Once the one dollar a day poverty threshold has been set (and "plugged into the computer"), the estimation of national and global poverty levels becomes an arithmetical exercise. Poverty indicators are computed in a mechanical fashion from the initial one dollar a day assumption.


"Authoritative" World Bank Numbers
These authoritative World Bank numbers are those which everybody quotes, --ie. 1.3 billion people below the poverty line. But nobody seems to have bothered to examine how the World Bank arrives at these figures.

The data is then tabulated in glossy tables with "forecasts" of declining levels of global poverty into the 21st Century. These World Bank "forecasts" of poverty are based on an assumed rate of growth of per capita income, --ie. growth of the latter implies pari passu a corresponding lowering of the levels of poverty. Its a numerical game!


World Bank "Forecasts": Poverty in China will decline to 2.9 percent
by the Year 2000
According to the World Bank's "simulations", the incidence of poverty in China is to decline from 20 percent in 1985 to 2.9 percent by the year 2000. 4 Similarly, poverty levels in India (where according to official data more than 80 percent of the population (1996) have per capita incomes below one dollar a day), the World Bank's "simulation" (which contradicts its own "one dollar a day" methodology) indicates a lowering of poverty levels from 55 percent in 1985 to 25 percent in the year 2000. 5

The whole framework (stemming from the one dollar a day assumption) is tautological; it is totally removed from an examination of real life situations. No need to analyse household expenditures on food, shelter and social services; no need to observe concrete conditions in impoverished villages or urban slums. In the World Bank framework, the "estimation" of poverty indicators has become numerical exercise.


The UNDP Framework
While the UNDP Human Development Group has in previous years provided the international community with a critical assessment of key issues of global development, the 1997 Human Development Report devoted to the eradication of poverty broadly conveys a similar viewpoint to that heralded by the Bretton Woods institutions. The UNDP's "human poverty index" (HPI) is based on "the most basic dimensions of deprivation: a short life span, lack of basic education and lack of access to public and private resources". 6

Based on the above criteria, the UNDP Human Development Group comes up with estimates of human poverty which are totally inconsistent with country-level realties. The HPI for Colombia, Mexico or Thailand, for instance, is of order of 10-11 percent (see Table 1). The UNDP measurements point to "achievements" in poverty reduction in Sub-Saharan Africa, the Middle East and India which are totally at odds with country-level data.

The human poverty estimates put forth by the UNDP portray an even more distorted and misleading pattern than those of the World Bank). For instance, only 10.9 percent of Mexico's population are categorised by the UNDP as "poor". Yet this estimate contradicts the situation observed in Mexico since the mid-1980s: collapse in social services, impoverishment of small farmers and the massive decline in real earnings triggered by successive currency devaluations. A recent OECD study confirms unequivocally the mounting tide of poverty in Mexico since the signing of the North American Free Trade Agreement (NAFTA). 7


Double Standards in the "Scientific" Measurement of Poverty
"Double standards" prevail in the measurement of poverty: the World Bank's one dollar a day criterion applies only to the "developing countries". Both the Bank and the UNDP fail to acknowledge the existence of poverty in Western Europe and North America. Moreover, the one dollar a day criterion is in overt contradiction with established methodologies used by Western governments and intergovernmental organisations to define and measure poverty in the "developed countries".

In the West, the methods for measuring poverty have been based on minimum levels of household spending required to meet essential expenditures on food, clothing, shelter, health and education. In the United States, for instance, the Social Security Administration (SSA) in the 1960s had set a "poverty threshold"which consisted of "the cost of a minimum adequate diet multiplied by three to allow for other expenses". This measurement was based on a broad consensus within the US Administration. 8


The US Poverty Threshold
The US "poverty threshold" for a family of four (two adults and two children) in 1996 was of the order of $16,036. This figure translates into a per capita income of eleven dollars a day (compared to the one dollar a day criterion of the World Bank used for developing countries). In 1996, 13.1 percent of the US population and 19.6 percent of the population in central cities of metropolitan areas were below the poverty threshold. 9


According to the UNDP Poverty in Mexico is lower than in the United States
Neither the UNDP nor the World Bank undertake comparisons in poverty levels between "developed" and "developing" countries. Comparisons of this nature would no doubt be the source of "scientific embarrassment" --ie. the poverty indicators presented by both organisations for Third World countries are in some cases of the same order of magnitude as (or even below) the official poverty levels in the US, Canada and the European Union. In Canada, heralded by the World community as "a promised land", occupying the first rank among all nations according to the same 1997 Human Development Report, 17.4 percent of the population are below the (official) poverty threshold compared to 10.9 percent for Mexico and 4.1 percent for Trinidad and Tobago. 10

Conversely, if the US Bureau of Census methodology (based on the cost of meeting a minimum diet) were applied to the developing countries, the overwhelming majority of the population would be categorised as "poor". While this exercise of using "Western standards" and definitions has not been applied in a systematic fashion, it should be noted that with the deregulation of commodity markets, retail prices of essential consumer goods are not appreciably lower than in the US or Western Europe. The cost of living in many Third World cities is higher than in the United States.

Moreover, household budget surveys for several Latin American countries suggest that at least sixty percent of the population the region does not meet minimum calorie and protein requirements. In Peru, for instance, following the 1990 IMF sponsored "Fujishock", 83 percent of the Peruvian population according to household census data were unable to meet minimum daily calorie and protein requirements. 11 The prevailing situation in Sub-Saharan Africa and South Asia is more serious where a majority of the population suffer from chronic undernourishment.

The investigation on poverty by both organisations take official statistics at face value. It is largely an "office based exercise" conducted in Washington and New York with few insights or awareness of "what is happening in the field". The 1997 UNDP Report points to a decline of one third to a half in child mortality in selected countries of Sub-Saharan despite the slide in State expenditures and income levels. What it fails to mention, however, is that the closing down of health clinics and the massive lay-offs of health professionals (often replaced by semi-illiterate health volunteers) responsible for compiling mortality data has resulted in a de facto decline in recorded mortality. The IMF-World Bank sponsored macro-economic reforms have also led to a collapse in the process of data collection.


Vindicating the "Free" Market System
These are the realities which are concealed by the World Bank and UNDP poverty studies. The poverty indicators blatantly misrepresent country level situations as well as the seriousness of global poverty. They serve the purpose of portraying the poor as a minority group representing some 20 percent of World population (1.3 billion people).

Declining levels of poverty including forecasts of future trends are derived with a view to vindicating the "free market" policies and upholding the "Washington Consensus" on macro-economic reform. The "free market" system is presented as the "solution", namely as an instrument of poverty alleviation. The impacts of macro-economic reform are denied. Both institutions point to the benefits of the technological revolution and the contribution of foreign investment and trade liberalisation to the eradication of poverty.


Table 1

Selected Developing Countries

Country Poverty Level
(percent of the population below
the poverty line)

Trinidad and Tobago 4.1
Mexico 10.9
Thailand 11.7
Colombia 10.7
Philippines 17.7
Jordan 10.9
Nicaragua 27.2
Jamaica 12.1
Iraq 30.7
Rwanda 37.9
Papua New Guinea 32.0
Nigeria 41.6
Zimbabwe 17.3

Source: Human Development Report 1997, table 1.1, p. 21


Table 2


Country Poverty Level
(percent of the population below
the poverty line)
United States (1996)* 13.7
Canada (1995)** 17.8
United Kingdom (1993)*** 20.0
Italy (1993)*** 17.0
Germany (1993)*** 13.0
France (1993)*** 17.0

*US Bureau of Census,

** Centre for International Statistics, Canadian Council on Social Development
***European Information Service.



1. United Nations Development Programme, Human Development Report, 1997, New York, 1997, p. 2.)

2. For a methodological review on the measurement of poverty see Jan Drewnowski, The Level of living Index, United Nations Institute for Social Research and Development (UNRISD), Geneva, 1965. See also the extensive research on poverty thresholds conducted by the US Bureau of the Census.

3. See World Bank, World Development Report, 1990, Washington DC, 1990.

4. See World Development Report, 1997, table 9.2, chapter 9.

5. Ibid., chapter 9, table 9.2.

6. Ibid., p. 5.

7. See Clement Trudel, Le Mexique subit le choc de

l'internationalisation, Le Devoir, Montreal, 28 March 1998, p.A4.

8. See US Bureau of the Census, Current Population Reports, Series P60-198, Poverty in the United States: 1996, Washington, 1997.

9. US Bureau of the Census, Poverty in the United States: 1996, Washington, 1997, p. 7.

10. According to the official definition of Statistics Canada (1995). For country ranks based on the UNDP's Human Development index, see Table 6, Human Development Report, 1997, p. 161

11. See Michel Chossudovsky, El Ajuste Economico: El Peru Bajo el Dominio del FMI, Mosca Azul Editores, Lima, 1992, p. 83.


Michel Chossudovsky
Department of Economics,
University of Ottawa,
Ottawa, K1N6N5
Voice box: 1-613-562-5800, ext. 1415
Fax: 1-514-425-6224


Recent articles by Chossudovsky on the global economic crisis at:










The Transnational Foundation for Peace and Future Research
Vegagatan 25, S - 224 57 Lund, Sweden
Phone + 46 - 46 - 145909     Fax + 46 - 46 - 144512   E-mail:

Contact the Webmaster at:
Created by Maria Näslund      © 1997, 1998, 1999 TFF