India's Missing Agricultural Data

With respect to agriculture, a lot of data is either not available or is released with delay. Often, it is inconsistent with other data sets.

At the recent Vibrant Gujarat summit, Reliance Industries chairman Mukesh Ambani said, “In this new world, data is the new oil. And data is the new wealth. India’s data must be controlled and owned by Indian people and not by corporates, especially global corporations.”

The data of middle class Indians – who use smartphones, shop online and use digital money – is in high demand for Indian and global companies. But agricultural data may not be similarly sought after even if it is necessary for agricultural policy, which impacts two-thirds of the Indian population.

For good data, there are three conditions: it should be available without much delay, should be consistent over time and must be comparable with other data sources.

The directorate of economics and statistics (DES) in the department of agriculture (DAC) does a great job to generate and publish data collected from states. ‘Agriculture Statistics at a Glance’ is the most trusted source of data on agriculture. From 2016, the department has also started publishing ‘Horticulture Statistics at a Glance’.

Also read: Will the PM Kisan Scheme Impress India’s Farmers?

Both the data sets are available online. However, lot of important data is either not available (like district level data) or it is released with delay. Very often, it is not consistent with other data sets.

We will discuss a few examples to illustrate these gaps.

When advance estimates of crop production are announced by the DAC, state-wise estimates are not released. So, it is not possible to assess the damage caused by drought or flood or disease in terms of the production of a particular crop in a state. For example, the production of food grains and other crops in Bihar, Maharashtra and Gujarat, which have been reeling under drought in 2018-19, is not known from second advance estimates released on February 28 this year.

The National Sample Survey Organisation (NSSO) is another major source of data on agriculture. Situation assessment surveys of agriculture households in 2003 (released in 2005) and 2012-13 (released in 2014) provided a rich source of data on income of farmers. But the data was not comparable. In 2003, ownership of land was necessary to define a farmer while in 2012-13, only the right to operate land was considered by the NSSO to define a farmer.

Also read: Life of Labour: Faulty Budget for Farmers; India’s Very Real Unemployment Crisis

In the 2012-13 survey, a cutoff of Rs 3,000 for the value of annual agricultural produce was adopted by the NSSO to define an agricultural household.

In the recent All India Rural Financial Inclusion Survey 2016-17 – by the National Bank for Agriculture and Rural Development (NABARD) – a household was considered agricultural if it earned over Rs 5,000 in value of produce in a year from agricultural activities – including cultivation from agricultural, horticultural, fodder, plantation crops or income from allied activities like animal husbandry, poultry, fisheries, bee keeping etc. Thus, on farmers’ income, NABARD data is not quite comparable with NSSO data.

For information on prices of agricultural commodities, the Agmarknet portal of the DAC is an extremely rich source of data. It captures spot prices from mandis across India. But there are several inconsistencies in the portal which need resolution.

For example, a search on March 5, 2019, for onion prices in Karnataka for September 2018 showed prices for September and August 2018. A search for October 2018 showed prices for October and September 2018. September is common in both searches but data shows the average wholesale onion price as Rs 977.49 per quintal in the October search and Rs 869.42 in the September search.

In 2017-18, the production of horticulture crops was 306.8 million tonnes while the food grain production was 284.8 million tonnes.  District-wise data for production of horticultural crops as given on the DES portal does not match with the data contained in ‘Horticultural Statistics at a Glance’. For example, the DES portal does not show the production of onion in 2015-16 even in major producers like Maharashtra and Madhya Pradesh.

Now that horticulture statistics are published by the horticulture statistics division of the DAC, there is no need for the DES to duplicate it.

A farmer works in wheat field on the outskirts of Ahmedabad, India February 1, 2018. Credit: Reuters

Poultry production grew by about 8% per annum from 2009 to 2014. The Department of Animal Husbandry, Dairy and Fisheries publishes the ‘Basic Animal Husbandry and Fisheries Statistics’ every year. District-level data of poultry production is not available and even state-wise poultry production is not available for 2014-15.

The Planning Commission was an important source of data for various sectors. Since its abolition in 2015, this source has dried up to a considerable extent. For example, the average all India tariff for electricity for the agriculture sector and power subsidy is available only up to 2012-13. The actual power subsidy to the agriculture sector is not known. 

The animal husbandry department is another source of important data. The production of milk in India has increased from 146.31 million tonnes in 2014-15 to 176.35 million tonnes in 2017-18. However, statistics for the district-wise production of milk are not available. Similarly, data on the price realised by milk producers is not available even for co-operatives.

District-wise data of loan waivers can show the extent of benefit to small and marginal farmers in poorer districts. It can also reveal whether farmers in rain-fed areas have benefitted. Similarly, detailed data of PM Kisan and the PM Fasal Bima Yojana can help in analysing these schemes – where government spending in 2019-20 may touch Rs 75,000 crore and Rs 28,000 crore, respectively. 

For sound policy formulation, research into past trends reflected in data is very important. Section 4 of the RTI Act provides that every public authority shall maintain all its records duly catalogued and indexed in a manner which facilitates the right to information under the act. It also mandates that computerisation should be completed within reasonable time. For agricultural data, this mandate remains unfulfilled.

It may be prudent for the Central Information Commission to direct the Central and the state governments to maintain and release detailed data on agriculture – so that policy formulation goes by real data and not tall promises made by political parties.

Siraj Hussain is a retired union agriculture secretary. He is now visiting senior fellow, ICRIER.