The World Bank has just released figures for its latest estimates of individual country and world GDP.  These estimates are based on the International Comparison Program (ICP) of prices conducted in 2005 for 138 countries.

The ICP has been in existence since 1970, when the first such survey was conducted - 10 countries participated in that survey and India was one of three developing countries.

Data on the size of an economy in US dollar terms are routinely available; but these data are subject to temporary fluctuations caused by the currency markets, as well as permanent fluctuations caused by undervaluation or overvaluation of a currency.

Hence, the need for PPP (purchasing power parity) adjustments, which the ICP has provided for several years (1970, 1985, 1993 and now 2005).

PPP estimates provide an estimate of the "true" size of an economy and the welfare of its people. It is precisely the PPP estimates of per capita consumption that allow the World Bank and others to estimate dollar a day world poverty. And the dollar in question is the PPP dollar, not the US dollar.

According to the world and World Bank on December 16, 2007, China's economy was 9.8 trillion in 2005, about 80 per cent of the US. In US dollar terms, it was only 18 per cent. The corresponding figures for India were 33 and 6 per cent (PPP and US dollars, respectively).

Today, we know something different and are presumably wiser. The Chinese economy has just been devalued by 40 per cent. Simply put, the new ICP estimates of China's economy in 2005 have its PPP GDP at 6 trillion, not 10 trillion. Correspondingly, its per capita income and consumption are also lower by at least 40 per cent.

Actually, per capita consumption is estimated by ICP to be 54 per cent lower!  Which implies that absolute poverty in China, previously thought to be around 15 per cent,  is now close to 40 per cent or an increase of more than 300 million people i.e. absolute (below one PPP $ a day) poverty in China was more than 500 million.

When releasing the estimates, in a somewhat unprecedented manner, the World Bank also released a separate document called "Background document - Frequently asked questions". This document attempts to lessen the shock value of the new estimates and also provide credibility to the entire ICP exercise.

There is a specific section explaining the China estimate: "This is the first time China participated in the ICP. Previous estimates were extrapolated � these extrapolations failed to take into account the change in structure and increase in prices over time".

There are further explanations in defence of the ridiculous. In answer to the FAQ question "Why do the PPP data differ from the data in the World Development Indicators" the following explanation is provided: "The data in World Development Indicators [a World Bank annual publication containing PPP estimates] are based on previous benchmark surveys."

Data provided in the previous benchmark survey of 1993 are considered "out of date� So it is not surprising that there will be some changes with the introduction of new and better data. The biggest change is for China which has participated in the ICP for the first time."

The obvious inference from the World Bank FAQ document is that the world remains as we know it, except for China. And China is expected to yield different results because it participated in ICP for the first time. But this is not true for most other countries in the world. India participated in it in the original 1970 ICP survey, as well as in 1985.

Nor is it the case that the price surveys conducted by the ICP program are "unique". Most countries conduct regular consumer price index surveys, and for those that don't, the GDP deflator provides a good average index of price levels at any point in time.

A combination of these data, and data provided by the ICP surveys of 1970, 1985, and 1993, have been used by the World Bank and others e.g. IMF, Penn World Tables and economic historians like Madisson, to generate estimates of PPP for various countries and years.

Thus, for most countries of the world, there should not be much difference between the new and old estimates. Hence, the World Bank going out of its way in making a patently false statement that "the biggest change is for China".

There are 21 countries with a higher negative GDP change (downgrading) than the -40 per cent for China; among these several large population economies like Ghana (-52 per cent), Nepal (-44), Bangladesh (-44), the Philippines(-43), and Uganda (-42 per cent). GDP in India has been reduced by 36 per cent. For the OECD economies, there is virtually zero difference as for most parts of the non-Asian world.

The new GDP estimates have implications - reduction in GDP means an increase in poverty. The Millennium Development Goal of 15% poor in 2015 seemed to have been reached in 2005. Now the World Bank seems to have given itself a new lease of life by estimating the poor in just India and China to be more than 1 billion in that year.

And world poverty in 2005 is estimated to be larger than a decade ago. Globalisation, and World Bank endorsed reforms, and economic growth must really be bad for the poor. Bring on Communism, socialism, planning, and the Hindu rate of growth - and more money for institutions like the World Bank to fight poverty.

In the world of good governance, a self-serving off the wall corrupt (?) estimate like the one produced by the World Bank would mean that heads would roll. But it is much easier to pronounce on governance than to practise it.

surjit.bhalla@oxusinvestments.com