Recently, the India Meteorological Department (IMD) made the forecast that rainfall in June-September period is likely to be at 97 per cent of long period average (LPA). After the LPA average of 106 per cent in 2013, during the next four years, the figures for actual rainfall were 88, 86, 97, 95 per cent, respectively. In fact, the rainfall figure of 2013 happens to be among the best in almost two decades and also contributed to an impressive agricultural growth rate of 5.6 per cent followed by – 0.2, 0.6, 6.3, and 3 per cent, respectively.
In other words, a “normal” monsoon is supposed to be good news for Indian agriculture and rural economy. Further, in spite of its relatively small share (of approximately 14 per cent) in GDP, agriculture’s performance has, through multiple linkages, significant implications for the economy as a whole. Thus, a normal rainfall is generally greeted with much cheer by economic forecasters. Given that much of India’s agriculture is dependent on rainfall for irrigation, the above noted presumed connections made by the “commoners” as well as the “experts” are hardly surprising.
But, as per the latest economic survey (Economic Survey 2017-18), assured irrigation is still a luxury for large segments of agricultural tracts in the country. Of the total net sown area, estimated to be 141.4 million hectares, only 68.2 million hectares are irrigated and 73.2 million hectares are largely dependent on rainfall; in other words, the former accounts for close to 48 per cent of the net sown area in the country. Furthermore, the net irrigated area as a percentage of total cropped area is even lower at 34.5 per cent. Such figures clearly point to major policy inadequacies since independence and are reminiscent of the old saying that the Indian agriculture is a “gamble of monsoons”! However, it is important to go beyond the simplistic linkages between good and bountiful rains with respect to long period average** (LPA) and levels of average agricultural output. This note is an attempt to examine some of the relevant complexities. Of course, a rigorous academic exercise needs to take into accounts a whole range of factors that influence agriculture’s performance over different time horizons, namely, short, medium and long-run. Indeed, there is substantial academic research, which has sought to examine the role of important determinants (institutional e.g. land reform as well as technocratic interventions e.g. agri-investments, price policies etc.) of the outcomes in the agricultural sector. In this short piece, such an academic exercise is ruled out; basically, what follows is an attempt to go beyond the “common sense” wisdom.
As should be obvious, over the medium to long run several institutional and technocratic variables are likely to play critical roles and it is not easy to segregate their influence on agriculture’s performance. Hence, in this note, I look at a relatively short period of approximately the last decade taking into account the years for which the relevant data is easily available in the public domain. We look at the latest information on state-wise production of food grains, major cereals and a few other significant crops (provided in the Handbook of Statistics on Indian Economy, the Reserve Bank of India) and the aggregate rainfall through the year, for the country as well as state-wise (given in the Statistical Year Book India 2017, MoSPI, Government of India). To make the data from the two sources comparable, we have made appropriate adjustments. In this note, “year” has been defined as April 1 to March 31. Given that we are looking at twelve months of rainfall instead of only the limited period of South-West Monsoon, the reported figures here, on an average, would be higher than LPA as well deficit or excess with respect to it. The findings reported here using the relevant data for the period 2007-08 to 2014-15 is based on a couple of very simple exercises.
The first of these examines the relationship between the total annual food grain production for the country as a whole and annual aggregate rainfall and the graph below depicts the same in a simple visual fashion; the relationship between the two variables appears to be positive and significant. However, if we look at the different crop groupings even at the all India level, substantial variations may be noted. For instance, the value of correlation coefficient for rice is as little as 0.17 whereas for oilseeds, the relevant value is 0.85!
Total Food grain and Aggregate Rainfall.
Values of Correlation Coefficient for Major Food Groupings and Aggregate Rainfall.
However, the Table below, which summarizes the value of correlation coefficients between rainfall and major crop groups, across states and regions for the period 2007-08 to 2014-15, immediately warns us against the simplistic conclusion noted earlier based on the first graph. As may be seen for several states and several crop groups, we get a whole spectrum of results, including inverse relationships, between them. For instance, in case of food grains and rice, four state/regions show an inverse relationship whereas in case of wheat there is only one incidence of such relationship. For other crops grouping, the incidence of inverse relationship turns out to be much higher (between 7 and 8). It seems almost bizarre that for Arunachal Pradesh, the values of correlation coefficients for all food groupings are in the negative whereas for Gujarat, Maharashtra, Andhra Pradesh and Karnataka, the opposite is true. Even for major cereals the values cover an astonishingly large range; for rice the values happen to be between -0.269 (Kerala) and 0.868 (Chhattisgarh) whereas for wheat the range for the relevant values is between -0.179 (Arunachal Pradesh) and 0.786 (Gujarat).
Clearly, these point to a number of other possible factors at work in different parts of the country, even if we are looking at a period of less than a decade. We may also note that further statistical probing throws up additional challenges with respect to neat relationships between the agricultural produce and rainfall levels. In other words, the need for a careful investigation can hardly be overstated and there is a substantial academic literature which has tried to engage with the matter at hand. As stated earlier, it is not possible to get into the relevant issues in any detail for reasons of space.
However, a couple of simple points underscoring the nature of monsoons may be worth noting. The first of these is: a good monsoon must also be well distributed across different agro-climatic and geographical regions. Second, it is also equally important that there is good distribution of the aggregate rainfall (e.g. with respect to the number of precipitation, intensity of each precipitation, etc.) not only across the two major seasons (Kharif and Rabi) but it should be in keeping with the crop-cycle requirements. With respect to both these, the research has shown disturbing and undesirable developments and researchers have often linked these with climate change. There is growing evidence for extreme weather events (such as floods and droughts connected with on-going global warming, with significant implications for farming and farmers. Researchers have also noted that climate change appears to be contributing to new pests, crop diseases, etc., apart from growing incidence of so-called “natural calamities”. As was noted by the latest Economic Survey (2017-18), dramatic environmental changes in the recent years, often attributed to factors associated with climate change, underscore the significance of investing in “climate resilient agriculture”.
Values of Correlation Coefficient for Major Food Groupings and Aggregate Rainfall.
As noted right at the outset, given the overwhelming dependence for irrigation on the monsoons, obviously good rainfalls are important and will continue to remain so. However, we have to understand the inter-temporal changes in the nature of monsoons itself to have a better understanding of their impact on agricultural output. Equally importantly, even in the short-run here is a clutch of policy interventions (e.g. flow of credit, timely backward and forward support to agriculture through provisioning of inputs, marketing, etc.) that can and do impact on performance of agriculture.