World Bank isn’t telling it like it is

http://indianexpress.com/article/opinion/columns/world-bank-isnt-telling-it-like-it-is/99/

In World Bank language, India has not been a good performer on poverty reduction, especially with regard to China.

Written by Surjit S Bhalla | Posted: October 11, 2014 

Policymakers and politicians are justifiably concerned about the level of absolute poverty in developing economies. But there are strong indications that the institution in charge of assessing global poverty trends — the World Bank — is acting like a dirty monopolist in order to survive. The rules of the game suggest that this manipulation needs to be checked. (Full disclosure — I am an ex-World Bank staff member and loved the work and the institution when I was there in the 1980s.)

The problem is as follows. The World Bank both defines absolute poverty as well as measures it. And it also defines the benchmark for the success or failure of individual country policies. In World Bank language, India has not been a good performer on poverty reduction, especially with regard to China. That may or may not be true — we can only know for sure if there is transparency and an honest debate on the methods and data used by the World Bank.

There is no mystery about the methods or the data and, indeed, there has been considerable transparency. In my 2002 book, Imagine There’s No Country, I pointed out that a major reason for the poor performance of India on poverty alleviation (relative to China and any other country) was because of the suspect data collection on individual consumption expenditure in India. Surveys in all countries capture only a portion of “actual” consumption expenditure. In China, this fraction is around 80 per cent while in India it is below 50 per cent. Correction for this survey data anomaly increased India’s performance in the fight against poverty to almost equal that of the record-breaker China.

This debate has continued since 2002. But today, with the publication of the Global Monitoring Report: Ending Poverty and Sharing Prosperity, the World Bank and its co-author, the IMF, have shed all claims to being objective readers of the poverty scene.

This is revealed by the following fact. Comparison in living standards across countries, and over time, is best done with recourse to purchasing power parity (PPP) estimates reported by the World Bank, jointly with other international institutions like the UN and IMF. PPP estimates are the gold standard for comparing standards of living. In May 2014, PPP 2011 data were released to the public.

This week, the IMF reported the incomes of the world in its latest World Economic Outlook (WEO). In this report, the IMF completely replaced all PPP 2005 income figures with the 2011 estimates. However, the World Bank, in charge of estimating poverty on the basis of these 2011 data, chose to completely ignore them and instead reverted to the old and extremely controversial PPP data constructed in 2005 (and released in 2008). In Shakespearean language, the World Bank spurned the new data like a cur out of its way. (In the World Bank GMR-Poverty report, there isn’t even a reference to the existence of the PPP 2011 survey.)

Normally, new PPP figures do not change one’s estimate or understanding of world poverty. But 2005 PPP was different (see next paragraph) and because 2005 was suspect, the use of 2011 versus 2005 data makes a huge difference to one’s understanding of world poverty.

But first, some background. The PPP estimates are released approximately once a decade, but the 2005 PPP figures were shocking for most observers. How shocking? Well, the aggregate price level for India for January 1, 2005 was approximately 52 per cent higher than what we knew on December 31, 2004. How did this happen? India has more price surveys than there are economists, so how did they all go so horribly wrong on the night of December 31, 2004? For an explanation, you have to look at China, because its consumption price level was also raised by 84 per cent on the same night. Speculation abounds about the origins of this mother of all changes; my favourite and preferred explanation is that China dictated this change. And what possible reason could it have to change the price level by an outlandishly extravagant amount? To show to the world that it was not being a currency manipulator — a stylised “fact” of development is that a currency can be “cheaper” for poorer countries. By manipulating its price level, China was able to prove to the world that it was actually a lot poorer than otherwise believed and hence, accumulating more than $3 trillion in reserves was “economically” justified.

For China, World Bank officials had a ready explanation for the price revision — China had never participated in a PPP survey, so the earlier PPP estimates were wrong. A fair point. But India has been a participant in each of the PPP surveys, including the original 1970 survey that consisted of only 12 countries. So why was there any change in the Indian price level, and especially such a large one? There is no explanation; if ever international officialdom was grotesquely wrong, it was in the construction of the PPP 2005 survey.

Good sense eventually prevailed (checks and balances) and, in superfast time, a new PPP survey was commissioned for 2011. Among the major results of this survey: first, very little change in the price and, therefore, per capita consumption estimates of countries in the developed world. Second, per capita consumption in China was raised (relative to 2005 PPP) by 25 per cent, and in India by nearly double that amount (46 per cent).

Given that India and China are the two most populous “poor” countries in the world, it is easy to infer why the World Bank, a major sponsor and executioner of the PPP project, has chosen to completely ignore the existence of the 2011 PPP survey. Because ending poverty will arrive sooner than the World Bank would like, and then what are its bureaucrats supposed to do — apply for jobs at the IMF or the UN?

The PPP data that one uses makes a considerable amount of difference to estimates of poverty and the evaluation of poverty declines.

Three sets of poverty figures are reported: the GMR-poverty estimates (page 19 of the report), our estimates using the same definition and methods as the World Bank, and the same with PPP 2011 data (see table). That our reproduction of the World Bank data is reasonably accurate is indicated by the fraction of the world in absolute poverty in 2011 (17 per cent World Bank versus 18.3 per cent Oxus).

The 2005 figures are with the poverty line of $1.25 per day. While different methods can be used to derive an equivalent poverty line in 2011 prices, most such methods converge on PPP $1.45 a day. Rather than use this, I have reported poverty figures for the world with a very conservative upper-bound estimate of the poverty line (PPP $1.6 per day). This suggests that poverty in the world in 2011 was 14 per cent. Estimates are that this ratio will decline to 10 per cent in 2015.

Moral of the story? World poverty is a serious problem and honestly we must face it, both in applauding progress and criticising failure to reduce it. How can the World Bank be honest and still exist? Raise the absolute poverty line by 60 per cent of the change in per capita consumption of the poorest 15 countries. If you do that, the suggested poverty line is PPP $0.23 higher in 2011 than in 2005, that is, around PPP $1.7 a day. This will yield approximately the same number of poor in 2011 as “revealed” by the World Bank.

The writer is chairman of Oxus Investments, an emerging market advisory firm, and a senior advisor to Zyfin, a leading financial information company