Alvin M. Keeling 
I was ‘cooling down’ (a bureaucratic term for spending a minimal period in the state after a central posting) in Ranchi in June 2006, hoping to be back in Delhi in October. One morning, I received a call from the head of the establishment asking me to introduce myself as secretary for food and public distribution the next day. I was parachuted into work and asked to “hit the deck running” from a ship in turbulent waters. The importation of wheat had started and India was trying to avoid a food crisis. How we avoided the global food crisis of 2008 is another story.
While measures to manage the situation were underway, I made efforts to understand how it had come to such an outcome. Figure 1 offers some insights.
Purchases in 2006-07 (April-March) at 9.23 MMT were well below requirements (note that 2006-07 purchase figures relate to the 2005-06 wheat harvest). Buffer stocks were drawn down by 2 MMT, a cardinal error in food management. The stock situation at the end of a poor supply season had put the government in a difficult situation.
What led to this situation? The central pool carried large amounts of inventory and there were widespread criticisms that these were being held for no good reason and costing taxpayers huge sums of money (some even said ‘to feed’ the rats!). The government had, after careful consideration, decided to liquidate certain stocks with the FCI for export. Figure 2 explains the numbers for wheat.
Coincidentally, purchases had started to fall from a high of 20.6 MMT to 15.8 MMT in 2003-04, slightly to 16.7 MMT in 2004-05 and then fell back to 14.8 MMT in 2005- 06. This trend and the resulting depletion of stocks went “unnoticed”.
Looking back, some conclusions can be drawn. The exuberance that India has a food surplus and can feed its people and “the world” has led to the unintentional depletion of public stocks. The drawdown on public stocks without reviewing the production and stock situation each quarter was poorly planned. Ignoring the decline in production almost every two years, especially in 2000-01, 02-03 and 04-05 followed by 05-06 has proven costly.
Not estimating the impact of climate change (high temperatures) on production – grain formation and grain size/weight – has proven to be critical.
The last point above was a game changer. The Ministry of Food, overconfident about buying large quantities, estimating that the size of the crop estimated by the Ministry of Agriculture is more than 75 million tons, began to sell old stocks . By the time the third advance estimate arrived at the end of May (there were no drones or satellite images at that time), the damage was done. I turned to a few private sources to understand the impact of weather conditions and the extent of private stocks. No official data existed on India’s private grain stocks.
The government relied only on production and public stocks data to make policy decisions, ignoring the importance of private stocks in the market. The CEO of a multinational company showed me how their company tracked temperature changes in northwest India on a daily basis and came to the conclusion, using the changes in grain formation and weight, to arrive at a harvest size of 68 MMT. Government agencies had missed this important part. Another CEO had compiled a time series of a “wheat balance sheet” for India, based on production, export, import, supply, consumption (data from ONSS) and private stocks . He had foreseen a shortage. A third, from a market agency, gave me access to daily prices from two dozen markets, a “real-time” ticker on my computer screen. These are revelations.
The current situation seems so familiar. The revised harvest size is 102 mmt. No bonus on the MSP this year despite an expected shortfall. Now an export ban, after claiming it is willing to feed the world, raises questions over the government’s confidence in production figures. With most experts forecasting a weaker harvest at the end of March, where was the need to talk big about exports? A case of overconfidence based on erroneous data?
What can India do to avoid such mistakes?
First, put systems in place to get reliable and timely crop estimates. The second advance estimates arrive in mid-February and the third in mid/late May. Food management requires a better image at the beginning of March (same for kharif). The national crop forecasting system, including “FASAL soft”, will need to be reset. The much-hyped Drone-Artificial Intelligence-Blockchain technologies should be deployed to do one simple thing: prepare a correct harvest estimate well in time, so that the government can plan and act ahead of any crisis.
Reliable price data has always been a missing link in policy planning. Compulsory price reporting (not just APMC price data) of all material transactions (limits can be set) is a must. Price movement is an important indicator of the mismatch between supply and demand.
The government should be aware of the quantum of private stocks, preferably in anonymized and aggregated formats. This requires legal support. A provision requiring the submission of anonymized stock data from all warehouses should be put in place.
The futures market remains largely underutilized. A dynamic futures market can help you plan better. Blaming the futures market for the price rise is over. A futures market should be allowed to operate without knee-jerk government intervention.
A robust system (drones, satellites, ground data) to monitor weather conditions such as temperature, water stress, etc. should be implemented immediately with a focus on key crops and major growing regions. With the expertise available in the country today, algorithms can be developed to assess the impact of weather and pest events on crop size and quality. The government needs this information more than anyone.
It’s time we talked about technology and data.
(The author served as Food Secretary from 2006 to 2008 and Agriculture Secretary in the Government of India from 2008 to 2010.)
