Planning Commission’s Poverty Charade

Share this article Share this article
published Published on Apr 5, 2012   modified Modified on Apr 5, 2012

-Economic and Political Weekly

Yojana Bhavan never seems to know how to count India’s poor

That the United Progressive Alliance (UPA) government can on occasion after occasion mishandle a situation and also show insensitivity has been in evidence once again in its handling of the poverty figures estimated from the 66th (2009-10) round of the National Sample Survey (NSS). Although the Planning Commission’s estimates, as measured by the Tendulkar methodology, declined sharply from 37.2% in 2004-05 to 29.8% in 2009-10 during UPA-I’s tenure, this good news turned bad by the way it was presented.

The Planning Commission should not have been surprised that critics would point to the poverty cut-offs as being very low, since this issue was at the centre of the controversy in late 2011 when Yojana Bhavan scored an own-goal by petitioning itself into the Right to Food case in the Supreme Court. In this context, the Planning Commission’s press note on the new poverty figures did not anticipate the obvious: that poverty lines lower than those stated in Court needed explanation. It did not clarify on the extent of inflation adjustment made between 2004-05 and 2009-10 nor explain that the Court had wanted 2011 poverty lines for which subsequent further inflation had also to be factored in. And it did not reiterate the earlier assurance that poverty estimates would not be used to set caps on the number of beneficiaries entitled to government schemes.

Even more inept was the handling of the resulting criticism, leading eventually to the announcement that yet another committee will be set up to go into the measurement of poverty. This not only slurs the late Suresh Tendulkar but also means that the government is likely to postpone any proper official analysis of the data which is now available. Besides important questions regarding what might underlie the evidently wide spatial and temporal differences in the pace of poverty reduction, there is also the obvious issue of how poverty trends from the Tendulkar methodology compare with those obtained if the earlier 1993 Lakdawala methodology is used. That none of this was even mentioned by the Planning Commission is a surprising error of omission, especially when there is a major change in methodology.

It is now up to private researchers to do what the Planning Commission should have done. First, evaluate the robustness of the recently released poverty estimates. And, second, attempt a decomposition of the poverty decline by the relative contributions of four undoubted features of UPA-I’s tenure: faster growth but with much higher food inflation than during 1993-2004; and a continued increase in inequality but along with a very significant step-up in the quantum of redistributive transfers to the poor. These features are likely to have had contradictory effects on whatever happened to poverty during this period, and may be captured differently by dissimilar methodologies. Nonetheless, gauging policy trade-offs is the raison d’être for poverty tracking; and, indeed, much of what is called policy paralysis of UPA-II stems from unresolved internal differences regarding inflation control versus growth and stimulation of investment versus redistribution.

However, this task requires careful scrutiny of survey concepts and clarity on what is comparable and what is not. Controversies have arisen previously on different recall periods used by the NSS and a potentially confusing feature of the 66th round is that it uses three different recalls – two of which have comparable data in earlier thick rounds but one does not. Researchers have in the past been guilty of comparing a recall period from one round with a different recall in another – mostly inadvertently but sometimes also because this suits their priors. Moreover, in this round the NSS has imputed values to some items that households receive free, notably free meals such as those received in schools or from employers. The past definition of private consumption expenditure excluded such items.

This matter of free meals has already been raised elsewhere (“India Undercounts the Poor”, Mint, 26 March). It was pointed out that the 2009-10 poverty ratio increases from 29.8% (33.8% rural and 20.9% urban) as posted by the Planning Commission to 31.5% (35.2% rural and 21.5% urban) on simply excluding this item of imputed expenditure from the 2009-10 numbers. Though small, this difference means that the correct decline in the number of poor between 2004-05 and 2009-10 might have been only 34 million, a third less than the 52 million decline reported by the Planning Commission.

Nonetheless, even with this correction, the pace of poverty reduction accelerated from 0.7 percentage points per year between 1993-94 and 2004-05 to 1.1 percentage points per year between 2004-05 and 2009-10. And, since mid-day meals and other public in-kind transfers do improve welfare of recipients and public expenditure on these did increase substantially after 2004, it is not possible to dismiss a faster pace of poverty reduction. However, if public food transfers are to be factored in, like should be compared with like (imputation done for earlier years) and conceptual consistency maintained (PDS purchases should also be valued at market prices).

The food aspect of poverty reduction is clearly relevant because Tendulkar was criticised for delinking poverty from calories. And, indeed, per capita calorie intake fell and the population consuming less than the calorie norm rose between 2004-05 and 2009-10, despite the significant fall in measured consumption poverty. Perhaps even more interesting is that the poverty estimates for 2009-10 using the earlier Lakdawala method (24.2% rural and 23.5% urban with free meals included and 26.1% rural and 24% urban excluding free meals) show a much smaller reduction from the corresponding 2004-05 estimates (28.3% rural and 25.7% urban). This happened almost entirely because food prices rose much more than non-food prices and the Lakdawala food weights from 1973 are much higher than the Tendulkar weights based on present consumption patterns. At least on this, Tendulkar is an improvement as far as measurement of consumption (as distinct from calorie) poverty is concerned. That, nonetheless, in-kind food transfers might be very important at times of high food prices could be the main lesson from the new data.

The critics who claim that present poverty lines are too low need to ponder whether their battle is to maximise the number of poor so as to entitle as many as possible to 35 kg/month at Rs 2/kg or to get the government to commit to some, possibly lower, near universal in-kind entitlements. And, unless it is really intent on hara-kiri, the UPA-II needs to consider its dithering over the Food Security Bill and also look at how its own state-wise poverty reduction figures correlate with growth of state domestic products and take an immediate call on course correction.

Economic and Political Weekly, Vol XLVII, No. 14, 7 April, 2012,

Related Articles


Write Comments

Your email address will not be published. Required fields are marked *


Video Archives


share on Facebook
Read Later

Contact Form

Please enter security code