Who is 'poor' in India?

A lot of people are exercised over the Planning Commission's definition of poverty - monthly income in the cities of Rs 965 per head (less in villages).

There has been incensed comment about the inadequate provision for the consumption of items like milk and eggs, school fees and medicines.

...

The official poverty line, which goes back 40 years, focused on a single parameter - what it would cost to get the minimum calorie intake.

But notions of even "absolute" poverty change, as societies become less poor; the Tendulkar committee report commissioned by the Planning Commission was one response.

But the drift of current comment suggests you should not be considered non-poor simply because you own (say) a bicycle or a mobile phone.

Indeed, since 70 per cent of homes now have TV sets, even some TV-owners may be counted among the poor.

We have come a long way from defining poverty as just calorie intake. That is only logical after 40 years in which per capita income has more than quadrupled.

The current ferment is an opportunity to reset the poverty debate, in two ways. First, the focus of poverty measurement has to move from the individual to the family, because the operational unit for living costs is the family, and costs per head are lower for larger families.

In the US, which follows this logic, the poverty line is about $11,000 for an individual, but only twice that (about $22,000) for a four-member family. In India, the realistic norm would be a five-member family.

Second, poverty is now relative and not absolute - if you accept the bicycle and TV argument. If so, India should adopt the European approach, which says you are poor if your income is lower than 60 per cent of the average level.

There is more than one way to define "average", and in this context it is the median; ie, the income of the 51st person in a ranking of 101 people, or the person ranked by income at 600 million, in a population of 1.2 billion.

This would mean that as average incomes rise, the poverty line automatically rises too - as it should, or the US would not have a poverty line that is many multiples of India's.

These two definitional changes would lay the base for a grounded debate on who is poor.

The problem is that there are no reliable income distribution statistics in the country, but a very rough-and-ready projection from the National Council for Applied Economic Research survey numbers suggests that the median five-member Indian family's monthly income would be in the region of Rs 12,000, give or take a bit.

A poverty line defined as 60 per cent of that figure gives Rs 7,200 for a family of five.

This is much higher than the Tendulkar/Planning Commission figure, but in Delhi it matches the government-prescribed minimum wage for a semi-skilled person, and therefore passes one crude reality test.

By definition, a poverty line defined as a percentage of the median would yield a population below the poverty line (or BPL) that is less than 50 per cent the population.

In Germany, it works out to about 15 per cent poor.

In India, the poor might number between 30 per cent and 40 per cent of the total.

Such a definitional "cap" would ensure that benefits meant for the poor go to the really poor, not to others who might be considered poor by flawed identification processes; keep in mind that some not-so-poor states have issued BPL cards that cover 80 per cent of the total families in the state.

The poverty industry has its own rackets, and if public money is to really go to the poor, we need realistic definitions based on proper data.