Counting the poor: do the poor count?

24 July 2002

Is the world “on the right track” in terms of poverty reduction? The World Bank seems to believe so.

Statements by the World Bank on good policies, aid effectiveness and selectivity give the impression that resources allocated to the World Bank are efficiently contributing to poverty reduction. The case for aid has been abundantly made by James Wolfensohn after 11 September, in the run-up to the Monterrey conference and during negotiations on IDA replenishment.

However Sanjay Reddy and Thomas Pogge of Columbia University, in a paper entitled How not to count the poor give a scathing account of the problems with the World Bank’s poverty numbers. They say an ill-defined poverty line, a misleading and inaccurate measure of purchasing power equivalence, and false precision are the three main errors that may lead to “a large understatement of the extent of global income poverty and to an incorrect inference that it has declined.” This allows the World Bank to insist that the world is indeed “on the right track” in terms of poverty reduction strategy, attributing this ‘success’ to the design and implementation of ‘good’ or ‘better policies’.

In a recent reply, Martin Ravallion of the Bank argues that “Reddy and Pogge have oversimplified the problem of measuring poverty in the world, and exaggerated the supposed faults in the Bank’s methods”. But Reddy and Pogge insist “[World Bank global income poverty estimates] do not stand up to serious scrutiny”. And they are not the only ones to contest the Bank’s methods.

Peter Nunnenkamp, Research Division Head at the Kiel Institute of World Economics contests the World Bank’s claims that its aid goes to countries with ‘good’ policies and institutions. He condemns the Bank for producing “strongly misleading” figures on ‘aid effectiveness’ for the Monterrey summit. He concludes that “little has changed in targeting aid at poor countries with good policies. In the longer run, the World Bank, by playing statistical tricks, may have weakened, rather than strengthened the case for more aid”.

Robert Wade of the London School of Economics argues that the Bank knows very well that the number of absolute poor is politically sensitive. He expresses concern that “it is possible that the people who calculate such numbers – in the Bank or elsewhere – are inclined to make methodological choices that produce a relatively favourable result even as they remain in the bounds of the professionally defensible”. Wade contests the fact that the World Bank, while subject to arm-twisting by its member states, is the world’s principal provider of development statistics. As he put it in a letter to Prospect recently: “We would not want [cigarette transnational] Philip Morris research labs to be the only source of data on the effects of smoking even if the research met professional standards”.

Reddy and Pogge say an alternative methodology is desirable and feasible. It could be done by constructing a basket of commodities required to meet the elementary consumption needs or capabilities of individuals; emphasis would be placed not on the basket mix, but on characteristics such as calorific content. In the meantime they argue that current estimates should stop being used.

World Bank programmes are frequently accused of inflicting “serious damage to health”. Now it seems that the credibility of its poverty figures might well be going up in smoke.

How not to count the poor

Assessing Assessing Aid

“Shooting the Messenger of Good News” Nunnenkamp

The Role and Effectiveness of Development Assistance: Lessons from World Bank Experience

Can the World Cut Poverty in Half? How Policy Reform and Effective Aid Can Meet International Development Goals

Remarks to the G20 Finance Ministers and Central Governors, James Wolfensohn