We think of erratic monsoons and droughts as terrible acts of god, events that don?t happen usually, as terrible rare outcomes. This is plain wrong. Bad rains and droughts are a recurrent phenomenon in India. But because we treat them as abnormal events, we don?t have a plan for them. There should be a template for monsoon management. And that template should become relevant from the time the IMD makes its first forecast.
Rainfall forecast by the IMD gives only broad idea about the amount of total rainfall for very large geographic regions. There is a probability associated with these predictions that is often ignored in reading the number.
For instance, IMD forecast for monsoon in North West India in 2009 was 81% of long period average with model error of ?8%. This implies that NW region could get 73% of normal rainfall; which involves 27% deficiency in June to September rainfall. As the rainfall is rarely distributed according to long-term monthly average, the actual rainfall was likely to fall more than 27% in the kharif sowing period. But this information, which was available as early as June 24, was not used in advance to prepare for deficient rainfall.
Was the reason a lack of faith in IMD forecasts? But that has been guiding the official policy in every other way. The stronger reason was the official thinking that deficient rainfall is an unlikely event, not something that should be woven into regular policymaking.
The net result is that kharif crops could not be sown on more than six million hectare area in the country till fourth week of July. The maximum damage is reported to area under paddy which may translate into more than 15% reduction in rice output during 2009-10. The effect on farm income would be much greater due to increase in cost of irrigation and labour.
There are four simple steps that need to become a part of regular monsoon policy. Taken effectively, they can insulate agriculture and the food economy from most of the effects of poor monsoon. The first step is to improve usefulness and accuracy of IMD forecasts and to follow them. This requires forecasts at disaggregate levels, that is, state or regions within large states. The exercise should be done by fortnights rather than by quarters. Obviously, IMD must develop database and a new model to accomplish this.
The next step is preparation of alternative crop plans under different scenarios of rainfall distribution. Implementation of such plans entails keeping ready adequate quantities of seeds of alternative crops. Only then can cropping patterns be changed quickly. Effectiveness of such plans requires very close coordination between state and central governments and agencies. Agriculture is a state subject. So, any plan to combat drought will be implemented by state governments. Some states? first response is to blame the Centre for inadequate relief. This typically holds up coordinating a production strategy. Blaming is easier than planning. But every year that has seen poor rainfall has seen ineffective response in part because of state-level political wrangling.
The third step is stock management. Sometimes the severity of such events is so large that moderate decline in domestic production is bound to take place. This can be dealt with scientific management of inventories and imports. One reason India need not worry about the food security at national level despite looming threat of drought this year is the comfortable stock of rice and wheat with the government.
Incidentally, this stock was not built with an intention to face the drought; it resulted from policies that have nothing to do with drought management, for example, hikes in minimum support prices. There is a need to maintain stock of major food commodities targeted towards eventualities like drought. Inventory building must be extended to pulses and sugar. We also need to synchronise our trade policy to the monsoon forecast as early as possible. Rather than changing tariffs and deciding about imports when the production shortfall becomes evident to the whole world, we should import early so that prices don?t move against us sharply. The government also needs to involve parastatals in import of pulses whenever there is a likelihood of poor harvest.
The fourth response should be to use science. There are varieties of paddy that can be grown in short duration, there are aerobic rice varieties that require much less water than common varieties. It?s not rocket science to get them to farmers in time. Water stress affects paddy output severely?it need not. For pulses, there are varieties that mature in less than sixty days, a great advantage in poor monsoon. But their field level adoption is very low. There?s no reason that this should be the case.
?The author is ICAR national professor, National Centre for Agricultural Economics & Policy Research