World Bank poverty estimates are poor, says government

Says the actual poverty is much higher than suggested by the multilateral lender, adding there is lack of scientific basis in computing the poverty line

The government has contested the World Bank's recently released data that showed only 12.4 per cent of India's population was poor in 2011-12, considering an expenditure cut-off of $1.9 a person a day on purchasing power parity (PPP) terms. It said the actual poverty was much higher than suggested by the multilateral lender, adding there was lack of scientific basis in computing the poverty line.

Before the release of a report, 'Ending Extreme Poverty and Sharing Prosperity', India's Central Statistics Office and the finance ministry held a series of meetings with World Bank teams, contending the consumption basket used to derive PPP did not even reflect the spending pattern of the common man in India, let alone the poor.

"India's poverty can't be this low. PPP is not an appropriate method to measure poverty. It has always been the government's stand that PPP should only be used to compare GDP (gross domestic product) across countries, not poverty. The items in PPP are comparable across the world and, therefore, high-end," said a government official.

In 2011-12, only about an eighth of India's population lived below the poverty line, according to the World Bank's poverty cut-off of $1.9 a day of expenditure per person, if a new methodology of poverty estimation by the National Sample Survey Office (NSSO) was taken into account, the report said. If the previous methodology is considered, the figure stands at 21.2 per cent, also based on the new poverty line.

"We have contested the use of PPP to determine poverty and the use of the dollar line. It is presenting an incorrect picture of the actual situation on the ground and is far from the reality, as far as poverty in India is concerned," said a government official.

According to the World Bank report, the poverty rate in India was one of the lowest among developing nations, even if one used the previous NSSO methodology.

"The items consumed by the poor are not properly represented in the basket. The $1.9 per capita expenditure dollar line is not right in the Indian context and not based on any scientific methodology," the official added.

The PPP based on the common list of items consumed across the world include a range of high-end items, not relevant from an Indian perspective, such as high-end rice, pasta, mineral water, wine and packaged holidays.

The official said the PPP measure might be relevant for the US or Mexico, but not for India, as the consumption and living patterns here were very different.

The government has also argued the weights assigned to items weren't appropriate. "The problem is with the weighting structure and the selection of items. The rice variety consumed by majority of the poor is very different from what is priced in the consumption basket to determine PPP," the official said.

Another official said at Rs 15 a dollar, the PPP for India was too low to reflect the ground reality. PPP of Rs 15 a dollar would mean a poverty line of Rs 28.5 estimated by the World Bank. This is lower than the Rs 33 a day of per capita expenditure estimated by the Suresh Tendulkar methodology, which drew flak from several quarters. However, it is slightly higher than Rs 29 a day for rural areas, also computed through the Tendulkar methodology.

Following criticism, the erstwhile Planning Commission had appointed another panel, headed by economist C Rangarajan, to come out with a set of poverty numbers. The panel estimated the poverty line based on per capita expenditure of Rs 47 a day for urban areas and Rs 33 a day for rural areas.

According to the Tendulkar methodology, as many as 21.5 per cent of Indians were poor in 2011-12. The Rangarajan panel revised this number to 29.5 per cent.

If the older methodology is considered, the poor constituted 21.2 per cent of the country's population, based on the World Bank's poverty line. The bank had raised the line from $1.25 a day of expenditure.

The new methodology is based on modified mixed reference period, the measure of monthly per capita expenditure when household consumer expenditure on most food items is recorded for a reference period of the "last seven days" preceding the survey period. For household consumer expenditure on items of clothing and bedding, footwear, education, institutional medical care, and durable goods is recorded for a reference period of the "last 365 days", while expenditure on all other items is recorded with a reference period of the "last 30 days".

In the earlier methodology, based on uniform reference period, the monthly per capita expenditure is based on household consumer expenditure on each item for a reference period of the "last 30 days".

The new method is considered more accurate, as it converts the 30-day recall period to one of seven days for food items and of a year for low-frequency, non-food consumption items. Respondents are expected to remember how much they spent on food items in the previous seven days more accurately than in the previous month.