Agri ministry’s yieAgri ministry’s yield estimates of same crop show big difference.

New Delhi: On 14 August when the Union ministry of agriculture released the fourth advance estimate of crop production for 2013-14, it gave the country reason to cheer: in 2013-14 India had achieved a record food grain production of 264 million tonnes, beating the previous year’s (2012-13) 257 million tonnes. These production figures, the most commonly used, are collected by the Department of Economics and Statistics (DES) under the Union ministry of agriculture, and are based on numerous crop-cutting experiments (CCEs) conducted across the country. 

However, DES also collects another set of data for the specific purpose of calculating the cost of cultivating major crops such as paddy, wheat, maize, soybean and cotton. This set of data, known as the comprehensive scheme (CS), is used by the Commission for Agricultural Costs and Prices (CACP) to recommend minimum support prices (MSP), the floor price at which the central government agencies procure different crops. Surprisingly, the difference in yield rates between these two estimates is significant for most crops with CS figures tending to be far higher than the CCE ones. For instance, over the decade between 2000-01 to 2009-10, the CS scheme yield rates for paddy were 19.7% higher than CCE estimates.

 For maize and soyabean the divergence was 10.8% and 16.4% respectively. For cotton, during the same period, CS yield rates were a whopping 302% higher than CCE estimates. For 2002-03, the year genetically modified Bt Cotton was introduced in India, the average divergence in yield estimates across nine cotton-growing states was the highest at 375%—varying from 668% for Andhra Pradesh to 527% for Tamil Nadu. 

The matter has attracted the attention of government experts. “It is observed that yield rates generated from the CS scheme are at significant variance with corresponding figures obtained under CCEs, former series generally gives higher levels than latter,” noted the CACP in its Price Policy for Kharif report for 2012-13 before recommending that “the matter needs to be investigated in greater detail”. 

Two years after the issue was flagged, there seem to be no clear answers. One reason, according to Tapas Kumar Dutta, adviser at DES, could be that the procedures of collecting yield data are different under CS and CCE methods. “While the CS scheme takes a smaller sample from about 8,000 farm holdings, the CCE sample size is much larger—over 9 lakh holdings. CS yields are only meant for calculating costs of production and recommending MSPs and are not used for any other purpose,” explained Dutta, adding, “The divergence in cotton yields is rare and unusual. It may be due to small sample sizes.” In fact, for Andhra Pradesh, says Dutta, no more than 10 or 20 farmers may have been selected for measuring yields under the CS scheme. 

CS estimates for cotton yields in the state were 410% more than the CCE estimates, measured over a decade, shows the CACP Kharif price policy report for 2012-13. Surprisingly, cotton yields were substantially higher in Andhra Pradesh, which cultivates the crop under rainfed conditions (in seven out of 10 years between 2000-01 to 2009-10), when compared with Gujarat which grows cotton largely with irrigation. Cotton grown by irrigation should normally have higher yields compared with rainfed cotton. 

Some experts suspect the surprisingly large variance between the two sets of government data indicate a “gross sampling bias”. “Yields under the CS scheme are collected by state agricultural universities. They have a propensity to select plots which show higher yields, thereby underestimating per quintal cost of production,” said G.V. Ramanjaneyulu, director of Centre for Sustainable Agriculture, a Hyderabad-based non-profit organisation; he was formerly a scientist with the Indian Council of Agricultural Research. At the same time, yields for cotton are not derived under the CCE scheme but are sourced by the DES from the Cotton Advisory Board (CAB), said an official at CACP who did not wish to be named.

 “CAB arrives at the yield estimates indirectly, based on market arrivals and area sown which might not give the correct estimates,” he official added. 

This implies that India may not have reliable estimates of cotton yields—from either scheme. This is ironic, as India is expected to topple China as the world’s largest cotton grower this year. “Small samples should ideally not lead to such a divergence. A 2-3% divergence is acceptable, but a 20% divergence in yield estimates, as is the case for paddy, is not acceptable. A committee is presently looking into these aspects,” the CACP official said. Ramesh Chand, director of the National Centre for Agricultural Economics and Policy Research, who is heading the expert committee to examine methodological issues in fixing MSPs, said that “sample plots which give higher yields might be incurring higher input costs”. Therefore, he added, “The fear of underestimating costs may be unfounded.” “CS scheme includes in the yield calculation the weight of cotton seeds in addition to the lint (fibre). Seeds are 2/3rd the weight of Kapas (lint plus seed) and this accounts for the yield divergence.

 It’s remarkable that the data has been wrongly reported and no one in CACP has a clue about this,” adds Chand. Ramanjaneyulu agrees with Chand, but when Mint asked CACP and DES officials, they rejected the ‘lint plus seed’ hypothesis.

RANJAY KUMAR,

PGDM 2^ND SEM,

SOURCE - MINT

Agri ministry’s yield estimates of same crop show big difference

Read more at: http://www.livemint.com/Politics/DMQDidmsUurUpW9rdkue7O/Agri-ministrys-yield-estimates-of-same-crop-show-big-differ.html?utm_source=copy