Lines in the sand

The latest poverty figures for India show that it has declined to just 22 percent as of 2011-12, based on data from the National Sample Survey. At a time when information such as this is coming out, many including my colleagues have asked why heavily subsidised food grains are aimed at some 67% of the population, when only 22 percent are poor. This is a question that cannot be asked often enough, but it is important not to make the mistake of linking any government schemes to the poverty line.

The ‘poverty line’ is a line in the sand. There is always a certain amount of arbitrariness to it. However, its primary (and perhaps only) purpose is to see how many people cross it over time. There are two main reasons why this line is tricky to draw: first, we are not a country where all citizens and residents pay income tax. So there are no direct ways to measure income. We measure incomes indirectly, based on surveys of what people consume – through national sample surveys. Warts and all, these surveys still provide the best information we can get on people and their habits. It was this data on consumption that was used by the economist Suresh Tendulkar to revise the ‘poverty line’ based on expenditure on food, education, health and a few other things.

The second reason why the line is tricky to draw is because the cost of these items of expenditure changes. Not only that, but people’s preferences (especially for food) are also changing with time. It helps to look at the poverty line as a rubber band that’s stretched and held taut, while people move across it. Over time, the rubber gets fatigued and droops a little, and needs to be made taut again, to accurately make a line that is comparable to the older one. To look at this as numerical jugglery or falsehood is just plain wrong.

This objective drawing of the line has shown that since the 90s, millions upon millions of Indians have crossed above the poverty line from below. The same was shown to be true for the most recent period of 2004-05 to 2011-12 as well, with the latest numbers.

Linking schemes with entitlements and benefits to this poverty line is an exercise that will ruin the objectivity of the line, and subject it to more political pressures than what it already subject to. The ideal situation is one where using NSS data, economists can come up with robust inclusion and exclusion criteria (like owning a refridgerator, for example) that is capable of doing two things with reasonable accuracy: select ~22 percent of the  population, and have the highest achievable overlap with those considered below the poverty line. This is notionally done even today for most schemes targeted at the poor, but the criteria are updated very slowly and the extent of mis-targeting is immense.

Just today, an argument is made in the Hindu that the government of India is slowly rejecting the legitimacy of the Tendulkar line and various departments are de facto drawing a much higher poverty line, one which includes about 65% of the Indian population below it.

We can argue endlessly about what “true” poverty is, whether the Tendulkar line only represents “kutta-billi” poverty, to use NC Saxena’s colourful phrase, and whether the de facto attempt at redrawing this is more “humane”. What matters is this: even and especially if the criteria for the Tendulkar poverty line is low, 22 percent of India lives below it. And they do so whether we lump them with 45 to 50 percent of the population above them or not. Economic growth, education and better provision of public goods are steadily increasing incomes and prosperity across India.

The simple question is: while people are pulling themselves out of poverty, do we support 1 in 5 people with some form of welfare with the limited resources at the state’s disposal, or do we use those same resources to support 3 out of 5 people? Which is more humane? And let us not forget, it is those same resources that are also used to provide better public goods and services.

We need an objective, consistent poverty line to reliably measure the outcome of poverty reduction. And we need to target welfare schemes only to those who need it the most.

Addendum. My colleague Nitin Pai also writes in The Acorn on the use and misuse of poverty lines.

, , ,


  1. Wanted: objective, consistent poverty line | The Alternative - July 25, 2013

    […] published on the Takshashila blog and re-published with the author’s […]