Reducing global poverty the World Bank way

S. Subramanian

Story [:] …A usually fictional prose or verse narrative intended to interest or amuse the hearer or reader; a tale. — The Free Dictionary 

“He - had - it - up - his - sleeve”. —Stephen Leacock: “The Conjurer's Revenge.”

THE World Bank has announced its latest estimates of global poverty and its projections for the future in a new World Bank publication, Policy Research Working Paper 7432. By the Bank's reckoning, the proportion of the world's population living in extreme poverty in 2011 was of the order of 14.5 per cent, which dipped further to 12.8 per cent in 2012 — all the way down from 37.1 per cent in 1990, thus ensuring that the world stayed ahead of schedule in meeting the target of the First Millennium Development Goal. (This goal — that of reducing the share of the world's population in extreme poverty by one-half from 1990 to 2015 — is curiously called the goal of “Eradicating Extreme Poverty and Hunger”). The Bank's own goal for the world is to bring the headcount ratio of extreme poverty down to 3 per cent by 2030. It notes in its publication, with a certain becoming gravitas, that reality may fall a little short of this heroic goal, with global extreme poverty rates probably turning out to be of the order of between 4.2 per cent and 5.7 per cent in 2030. 

Far be it from me to suggest that these spectacular results have been achieved by sleight-of-hand: this is the age of scientific expertise, so we can rely on Purchasing Power Parity Dollars and the Global Poverty Line, instead of on mirrors and magic lanterns, to achieve the Great Illusion. Specifically, the proportion of a population in poverty must be expected to increase as the poverty line is raised; and it turns out that trends in the headcount ratio of poverty become less and less flattering as the poverty line is raised. The conjurer's first trick, therefore, would be to pitch the poverty line as low as he can possibly get away with. 

Second, how do we ensure that the poverty standard is “comparable” across countries at any point of time? An answer, which is plausible at first blush, is to have a unique global poverty line that has the same “real” value across countries: this requires that the line should be measured in such a way as to preserve “purchasing power parity” (PPP) across the countries. If we treat the U.S. dollar as the “standard” currency (or what economists would call the “numeraire currency”), and if x US dollars is taken to be the global poverty line, then country-specific exchange rates — called purchasing power parity rates — must be such, for instance, that x US dollars, when converted into the Indonesian rupaiah, must have the same purchasing power as when converted into the Indian rupee or the Thai baht. If the Indian rupee is valued at 40 rupees to a dollar, then the global poverty line of x dollars, translated into Indian rupees, is 40x rupees; at 20 rupees to a dollar, it is only 20x rupees. If a smaller poverty line is “better” for the magnitudes and trends of headcount ratios of poverty, then prudence would dictate a list of country-specific PPP exchange rates that are closer to, rather than further away above, the US dollar. The conjurer's second trick, that is, would be to pitch PPP exchange rates “low”.

The first trick is well-served by the World Bank's choice of a global poverty line which is the average of the national poverty lines (some of them fixed by the Bank itself!) of the 15  poorest (in terms of per capita consumption) countries in the world in 2005. In 2005, and employing PPP exchange rates yielded by the Bank's 2005 International Comparison Programme (ICP) data on country-specific price levels, the global poverty line was pitched at 1.25 PPP$ per person per day. By 2014, a new set of country-specific PPP exchange rates became available from the Bank's 2011 ICP data on country-specific price levels. 

The 2011 PPP exchange rates are, in general, substantially lower than the 2005 PPP exchange rates. The conjurer's second trick has been well-served by the Bank's decision to “update” the 15 poorest countries' 2005 national poverty lines in domestic currency to their levels in 2011 by employing their domestic rates of inflation, as captured by the countries' consumer price indices; converting these national poverty lines into dollar terms by using the relatively more “benign” 2011 PPP exchange rates; taking the average of these lines — which works out to 1.90 PPP$; and treating this as the global poverty line.

One would have thought that the “poverty line” is a level of income that enables people to secure the minimum necessities of life. One would also have thought that invariance of the poverty norm ought to be preserved in the space of “human functionings”, rather than in the space of  “real incomes”. The two “tricks” we have alluded to above successfully avoid both of these common-sense outcomes. What is the result? Consider only the case of India for a moment. It is a matter of common knowledge that the poverty lines prescribed by the Rangarajan Committee in 2014 are way short of what one might reasonably expect a poverty line to secure, by way of some of the basic necessities of life. It is worth recalling that in 2014, and in response to the massive public outcry against the 2011 Tendulkar Committee's very low poverty lines (themselves an improvement on the 1993 Expert Group's recommendations), the Rangarajan Committee came up with rural and urban poverty lines for 2011, in 2011 prices, of Rs 32.40 and Rs 46.90 respectively, per person per day. The all-India headcount ratio of poverty corresponding to these poverty lines is of the order of 29.5 per cent, accounting for some 363 million people in poverty. However, corresponding to the 2011 global poverty line of 1.90PPP$ at 2011 PPP exchange rates, the headcount ratio of poverty in India for 2011 works out to only around 18.2 per cent, accounting for some 224 million poor persons. Our own national standards of money-metric poverty, pitiful as they are, have been bested by the World Bank: the number of people in poverty, according to the former, is nearly 140 million more than according to the latter! These 'missing numbers', for India alone, account for about 15.5 per cent of the Bank's estimated number (902 million) of globally poor persons in 2011.

With such low levels of poverty, set well to reduce further and rapidly at that, supra-national institutions do not have to deal with troublesome issues of trade, aid, austerity, conditionalities, debt and strife. The World Bank way of reducing poverty assures us that there really is nothing very much to worry about on the poverty front. I hope many of us are now enabled to appreciate that the Bank's President, Jim Yong Kim, never spoke a truer word than when he said, after the World Bank released its global poverty figures:  “This is the best story in the world today…” (A helpful definition of the word “story” is supplied at the head of this article.)

The writer is  National Fellow, Indian Council of Social Science Research