BackHow Angus Deaton views India
In the Lionel Robbins Memorial lectures in December last year, Deaton spoke about poverty, inequality and difficulties of measuring them in India
Premium
British economist Angus Deaton, awarded the Nobel Prize for Economics this year, has had a long association with India. He has authored several research papers on Indian poverty and on the statistics used to measure poverty. In the Lionel Robbins Memorial lectures across three consecutive evenings on the 9, 10 and 11 December last year, delivered at the London School of Economics, Deaton spoke about poverty and inequality and the difficulties of measuring them. The lectures included many references to India. Here are some of the points made, reproduced from the lecture notes.
On India’s growth
Rapid growth, but slower than warranted poverty decline.
Historically unprecedented economic growth over past 30 years.
Now a (lower) middle-income country.
In India, national accounts are weak, and given the political importance of high growth, little attempt to improve.
My guess is that Indian growth rates are too high.
On India’s poverty
Poverty rate has declined no matter what measure we use, though much more slowly than GDP (gross domestic product) or aggregate consumption growth would seem to warrant.
Growth comes from national accounts, poverty from household surveys.
In 1972-73 survey consumption 5% less than NAS (National Accounts Statistics) consumption.
In 1983-84, 25% less, currently 50% less.
Which is why poverty decline is less than we would expect from the growth of GDP.
(This) Inconsistency is at the heart of the political and statistical debates about poverty in India.
Serious conflict between NAS and NSS (National Sample Survey) consumption figures.
Left says NSS is obviously right, and only complete source of what people actually spend, and poverty is declining slowly.
Right says NAS is obviously right, that surveys are useless, that enumerators sit under trees or in tea shops filling out questionnaires from nothing.
So in late 90s a campaign to discredit the surveys by the rightists and, more constructively, to suggest why they might be wrong.
The following illustrates a perfect example of how politics penetrates deeply into measurement:
How much rice did you buy over the last XX days?
Indian tradition, from Mahalanobis, was XX=30, based on an experiment. In the debate, right argued this was too long, other statistical offices use 7 days, and that people would forget over 30.
This might explain some of the gap between NAS and NSS.
Though not clear that it can say anything about why that gap increases over time, which is the key issue.
One side wanted 7 days, other 30 days, unresolvable.
How to decide? Do a randomized controlled trial (RCT), of course!
Surveyors did a nationwide RCT, randomizing 7 and 30 over villages
RCT showed, indeed, that flow of reported consumption at 7 days was higher than flow at 30 days
Triumph for the right, or was it?
The effect is HUGE: 175 million Indians were removed from poverty in 1998, close to a half.
This is more than 10% of WORLD poverty.
But this didn’t resolve anything: as always RCTs do not tell us WHY the difference occurred, or which was correct.
But they did raise the stakes in the debate, because there was a lot to fight over.
Again, the debate was unresolvable, neither side would concede, so a compromise: do both at once!
This was a disaster, because they cross-contaminated, and the poverty measurement process in India was compromised for a decade, because the major poverty surveys are only done every five years or so.
In the end, yet another Expert Group, which complicated things even more, and has, until recently, been the basis for counts.
Poverty line itself has lost public credibility.
PC (Planning Commission) sent an infamous affidavit to the Supreme Court saying that ₹ 26 per person per day was enough to remove someone being poor.
The media widely condemned such an absurdly low number.
As did their (relatively well-heeled) readers.
What to do? Set up yet another Expert Group, whose report sets new standards of rococo statistical architecture
Then the Planning Commission got abolished!
The standard poverty line based measures in India are now discredited.
In part because events discredited its rhetoric.
So that political squabbling over the measure of success was bound to destroy it.
Leaving one side to proclaim success based on GDP growth, there is little poverty left.
And the other to claim destitution through neoliberal globalization.
If poverty is such an embarrassment, or shame
Deny it!
Growth triumphalism and poverty denialism.
On malnutrition
Remarkable decline in per capita calorie (and protein) consumption in India.
In spite of mass malnutrition in anthropometrics.
In spite of real income and consumption growth, even among the poor, and even according to the surveys.
Happening at all levels of consumption, though more among the rich than the poor.
Why this is happening remains unclear and deeply controversial.
People are being impoverished by neoliberal globalization, CPI(M).
More likely is a sign of progress. Rising real incomes reduce hard labour, as well as better sanitation.
Across areas of India, it is the healthier, richer places, in the south that have lower calorie consumption per capita than do the less healthy poorer places in the north.
Half of children are severely malnourished.
Indian men and women are among the shortest on the planet: at current rates of change it will take 250 years for Indian women to grow as tall as British women
Numbers from a 98/99 survey caused consternation.
Prime Minister Manmohan Singh said they were a national shame.
In recent years, senior officials note that the figures are out of date.
While doing nothing to update them, and new survey has been long delayed.
In the meantime, some eminent economists have pressed the argument that, like the poverty numbers, these numbers are not a problem.
Because Indians are genetically short
To my way of thinking, and understanding of the literature, this is very much an argument of last resort.
Why Indians are so malnourished is a key area on which to focus.
Indian politics
Political problem of how to run an economy that is doing so well for some, while continuing to hold so many poor, malnourished people.
BJP is a religious and business-oriented party, Congress is business-oriented party with a pro-poor rhetoric that sometimes delivered pro-poor policies.
Both are dependent on rapid economic growth for legitimacy.
Important, politically, that growth not only takes place, but be seen to help the poor.
Either through direct schemes (left) or rising tide and boats (right)
Political divide, as elsewhere, whether growth should have priority.
One extreme that growth is sufficient as well as necessary.
Other extreme growth causes poverty, by redistributing upward.
Middle worries about ineffectiveness of pro-poor policies.
On global inequality
Can we really make sensible statements about global inequality when the standard errors of the China and India PPPs (purchasing power parity measures) are 15%?
All the notes for the lectures are available here
Unlock a world of Benefits! From insightful newsletters to real-time stock tracking, breaking news and a personalized newsfeed – it's all here, just a click away! Login Now!
