ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846
-A A +A

The Curious Case of Cocktails, Weedicides, and Tonics

Pest Management Practices of Cotton Farmers in Vidarbha

Thiagu Ranganathan ( teaches at the Indian Institute of Management, Nagpur. Sarthak Gaurav ( teaches at Shailesh J Mehta School of Management, Indian Institute of Technology Bombay.


This study uses primary survey data collected from Vidarbha, Maharashtra, to analyse pest attacks and pest management practices among cotton farmers who are at the risk of long-term exposure to toxic pesticides. We find that despite a reduction in bollworm infestations, secondary pest pressure is high. Farmers use pesticides indiscriminately and there is widespread practice of using “pesticide cocktails.” Farmers also mix pesticides with fertilisers. The use of weedicides and “tonics” is also widely prevalent. Farmers who perceive pesticides as yield-enhancing input spent more per acre on them. These findings raise concerns about the role of agricultural input dealers in filling up the agricultural information void for pest management.

The authors are grateful to the Indian Council of Social Science Research for financial support. The second author would like to acknowledge the financial assistance from IIT Bombay for fieldwork and setting up of weather stations in study villages. The authors have benefited from discussions with T V Venkateswaran, Atul Sharma, Srijit Mishra, Vijay Jawandhia, and Pravin Mankar. Anil Golhar, Gajanand Ahirwal, and the team of investigators in Wardha and Yavatmal provided excellent research assistance. The authors also acknowledge the comments given by the anonymous reviewer.

Pest infestations cause significant yield losses to farmers across the world, and cotton farmers are particularly at risk of frequent losses due to pest attacks (Oerke 2006). Typically, chemical pesticides are used to deal with pest attacks. Pest management practices such as integrated pest management (IPM) and non-pesticide management (NPM) have not gained wide acceptance despite limited success in some regions (Ramanjaneyulu et al 2004).1 The indiscriminate spray of toxic chemical pesticides has been a concern due to the environmental and health hazards associated with them. In the context of cotton farming in India, while the diffusion of Bt cotton has been near universal, issues of insecticide resistance and secondary pests have emerged (Shetty 2004; Shetty et al 2008; Gaurav and Mishra 2015; Taneja 2017).

Long-term exposure to even low levels of organophosphate insecticides such as monocrotophos has been associated with suicidal tendency, depression, and neurobehavioural problems (London et al 2005; Ross et al 2013). This is disconcerting in light of the long-term exposure of farmers and agricultural labourers spraying pesticides. Moreover, consuming insecticides is the most common method used by farmers to commit suicide, and there is also considerable risk of accidental poisoning due to insecticides (Mishra 2006; Kale et al 2014). In 2017 alone, over 20 cotton farmers died due to pesticides-related poisoning in the district of Yavatmal (Public Eye 2018), one of our study districts, whilst over 40 farmers died in the Vidarbha region (Ghadyalpatil 2017). The economic costs are also significant as per hectare expenditure on pesticides by cotton farmers has increased starting from 2002–03 when Bt cotton was introduced in the study region, even though per hectare use of pesticides is lower than other countries (Ranganathan et al 2018). Furthermore, there is lack of evidence on pest management practices among farmers in India in recent times. This study attempts to address this gap.

Study Sample

The study was conducted across 18 villages in two districts of Wardha and Yavatmal in the Vidarbha region of Maharashtra, India.2 We selected three talukas in each of the two districts: Arvi, Seloo, and Wardha in the district of Wardha, and Ralegaon, Kalamb, and Babhulgaon in the district of Yavatmal. The study began as a part of the revisit and extension of a longitudinal study that continuously surveyed risks faced by 120 farmer households in Wardha district from 2009 to 2012 (Gaurav 2014, 2015). The first round of this study was conducted in 2016 in which 576 farmer households were surveyed, and agricultural information about kharif season in 2015 was collected. In 2017, 516 of these households were surveyed to gather information about the 2016–17 agricultural year.

The survey collected information on demographics, land holdings, asset ownership, agricultural information sources, credit, irrigation, livestock, and wages of working members in the households. Since intercropping is common in the study region, we collected detailed plot-level information on the actual pattern of intercropping followed. This enables us to estimate effective areas allocated to specific crops in an intercropping system. Plot-level data on agricultural inputs, crop productivity, marketable surplus, by-products, and prices were also collected. The survey also collected data on knowledge, perceptions, and practices related to the usage of chemical pesticides by cotton farmers. Pesticide usage data indicating the chemical or brand used by the farmer along with pest infestation experienced was also collected. Table 1 reports the summary statistics of the panel households surveyed in both the rounds.

The two rounds of surveys allow us to construct a panel data set comprising 516 households with 279 households from Wardha and 237 households from Yavatmal. There were few female-headed households in the sample—20 from Wardha and nine from Yavatmal. The average age of the household head was 51 years. On average, there were four members in a household. Considering members in the household aged less than 15 years and above 60 years as the dependents, the average dependency ratio (calculated as the ratio of the number of dependents to household size) was around one-third across both the districts. The average land size of farm households in the sample was 5.52 acres. This was 6 acres in Wardha and around 5 acres in Yavatmal. Around 70% of the sample was small and marginal farmers with less than 5 acres of land. Almost 66% of farmers in Wardha and 73% of farmers in Yavatmal were small and marginal farmers. The average landholding of small and marginal farmers was 3.5 acres. The villages were primarily rain-fed, with a little over half of the households in the sample having access to some form of irrigation.

Table 2 reports the acreage of major crops in the study sample over the study period. Cotton, soybean, and tur (pigeon pea/red gram) comprise over 85% of the total cultivated area in the study villages. Over the two years, there was a slight decline in average area under cotton while the decline in average area under soybean and tur was more pronounced. Table 3 reports the average acreage and acreage share by cropping system. Tables 2 and 3 report findings related to all the households surveyed rather than the 516 households in our balanced panel.

Cotton and tur intercropping system is the dominant cropping system in the study area, followed by soybean monocropping. During the two study years, the share of cotton and tur intercropping increased from two-third of total cultivated area in 2015–16 to almost three-fourth in 2016–17. The average area under soybean and tur intercropping was much lower than that of monocropped soybean. However, there was a considerable decline in the share of soybean and tur intercropping in 2016–17 compared to 2015–16. In the study villages, soybean monocropping was more prevalent than cotton monocropping. Tur was not grown as a monocrop in the sample region. We have reported the cropping pattern related to major crops only. Few farmers cultivated sugar cane, jowar (sorghum), groundnut, and vegetables as well in both the years.

Pesticide Cocktails, Weedicides, and Tonics

Since the focus of our study was to understand the pesticide usage among cotton farmers, we discuss results for only the kharif season.3Table 4 (p 55) reports the proportion of farmers cultivating cotton in the panel data by village in each taluka of the study districts.

In the 18 villages in the study region, we found that cotton was cultivated by 440 (85%) farmers in the agricultural year 2015–16 and by 460 (89.1%) farmers in the agricultural year 2016–17. There was considerable variation in terms of the extent of cotton farming across the villages. In one village (Babhulgaon), only 60% of the farmers cultivated cotton in 2015–16 while in another village (Arvi) in the same year, 93.5% of the sampled farmers cultivated cotton. In 2016–17, overall, more farmers cultivated cotton as some farmers shifted from soybean to cotton owing to low soybean yields in 2015–16.

Farmers in the study region have been cultivating Bt cotton for over a decade. However, they started later than the Gujarat farmers, and its diffusion was relatively slow due to poorer irrigation conditions. Ethnography in the region and interviews with subject matter experts reveal that there has been a reduction of American bollworm infestation due to increase in areas under Bt cotton, while there was an increase in pink bollworm attacks in the recent years as they seemed to have developed resistance to the Bt trait. In the context of pest attacks reported by farmers, it is evident that farmers had to deal with a growing infestation of other pests such as sucking pests (including jassids, aphids) and whiteflies. Among these insect pests, whitefly infestation happens during the later stage of cotton growth as compared to aphids and jassids.

We collected data on up to three pest attacks in both the years. Since there were very few farmers who faced three pest attacks for cotton, we present patterns for only two pest attacks. Table 5 provides the details of two pest attacks across the sample in 2015–16 and 2016–17.

As shown in Table 5, more cotton farmers in the panel experienced pest attacks in 2016–17 as compared to 2015–16. However, considerably more households (147) faced two pest attacks on their cotton plant in 2015–16 as compared to those (102) in 2016–17. Furthermore, there were few bollworm infestations in both the years—20 cases in 2015–16 and only two reported cases in 2016–17. Sucking pests such as jassids and aphids were the major pests and these infected a majority of the households in both the years. Whitefly pest infestation was also common, and this was typically the second pest infestation for most of the cotton farmers. This corroborates the previous findings of secondary pests (Menon 2018a; Kranthi 2014). To deal with these pest attacks, farmers used various pesticides—mostly monocrotophos (an organophosphate), acephate, and imidacloprid.4 Monocrotophos is a highly toxic pesticide whose usage is banned in many countries. In India, while the usage of monocrotophos for vegetables and food crops is banned, its usage for cash crops such as cotton has not yet been prohibited. However, there have been demands to ban its usage in the coming years.

A noteworthy observation regarding pesticide usage practice is that it is common for farmers to use “cocktails” of pesticides—typically a mix of two or more pesticides for dealing with pests. Pesticides used by farmers for both the pest attacks in 2015–16 are presented in Table 6a while Table 6b (p 56) presents the information for 2016–17. Another practice was to mix water-soluble fertilisers such as 19.19.19 (mono ammonium phosphate) with pesticides. However, there were a few such cases, and we have not reported mixing with fertilisers in Tables 6a and 6b.


From Table 6a, we find that around 60% of the farmers whose crop was affected by pests used monocrotophos or a combination of monocrotophos and acephate to deal with the first pest infestation. Imidacloprid, which is relatively safer and more effective (Gervais et al 2010), was used without mixing with other pesticides by 28 farmers. Fifty-one farmers used imidacloprid with monocrotophos, and 11 used the same with acephate. Seventeen farmers used all three pesticides together in their cocktail.

We also found that for the first pest attack, among the individual pesticides, the per-acre expenses were highest for imidacloprid than acephate or monocrotophos. This indicates that imidacloprid is costlier even though it might be safer and effective. Farmers might therefore be mixing pesticides to reduce their expenditure. However, in terms of the potency or effectiveness of the cocktails, there is no causal evidence. In a study among farmers in Gujarat, Seo (2013) found the use of a combination of pesticides and suggested that these might be ineffective. We also found that in 2015–16, 12 farmers used diafenthiuron to deal with whiteflies and nine farmers used cocktails consisting of four pesticides. For the second pest infestation (Table 6b), farmers typically used either acephate, imidacloprid, or monocrotophos. Use of flonicamid (insecticide for aphids and whiteflies) was also found.

Among the users of pesticides, average per-acre costs and pesticide expenses are similar for both pest infestations in 2015–16. For the first pest infestation, farmers spent ₹1,390 on an average while for the second they spent ₹1,348. On per-acre of cotton cultivation basis, farmers spent ₹469 for the first pest infestation and ₹479 for the second pest infestation (Table 6a).

Table 7 reports average and per-acre expenditure on pesticides along with expenditure on weedicides and “tonics”—plant growth promoters. In 2016–17, more farmers had their crops infested by pests but they also spent less on pesticides (both total and on per-acre basis). This is likely due to higher usage of the less costly monocrotophos in 2016–17 as 302 farmers used monocrotophos in that year compared to 106 in 2015–16. The use of combinations was lesser in 2016–17. The number of farmers using weedicides and tonics increased in 2016–17 as compared to 2015–16.

From Table 7, we find that only 25 farmers in the sample spent on weedicides in 2015–16 while 156 farmers spent on weedicides in 2016–17. A part of the reason for the increase in the use of weedicides could also be the rising costs of labour. There have been gradual changes in sowing technology whereby oxen labour is being replaced by tractors. Accordingly, changes in the spacing between cotton plants have paved the way for growing use of weedicides. This is similar to the findings from Telangana, where farmers changed the ploughing practices which paved the way for more use of herbicides (Stone and Flachs 2018).

The emergence of tonics is noteworthy. Around 101 farmers used these plant growth promoters in 2015–16 while 339 farmers spent on the same in 2016–17. On average, cotton farmers spent ₹1,561 on insecticides, ₹139 on weedicides, and ₹237 on plant growth promoters, totalling to ₹1,936 on all items in 2015–16. In 2016–17, they spent ₹1,156 on insecticides, ₹565 on weedicides, and ₹799 on plant growth promoters, totalling to ₹2,520. On a per-acre basis, they spent a total of ₹657 in 2015–16 and ₹837 in 2016–17 on pesticide cocktails, weedicides, and plant growth promoters. As a percentage of total input costs, this comprised 6.1% in 2015–16 and 6.8% in 2016–17.

In summary, we found that farmers in the study villages face considerable risk of pest infestation, particularly secondary pests. To manage these pests, farmers used pesticide cocktails, weedicides, and plant growth promoters. Farmers spent more on weedicides and plant growth promoters in 2016–17 compared to 2015–16. This could be an indicator of a shift towards larger use of weedicides and plant growth promoters. It could also be the case that when farmers spent less on insecticides, they spent more on plant growth promoters. Anecdotal evidence supports this claim, but these need not be generalisable.

The set of findings suggests that, in their attempt to reduce yield losses and obtain higher average yields, crop protection through the use of pesticides and enhancement of yield through use of nutritional supplements and fertilisers have become intertwined for farmers. In the absence of timely and effective extension services and crop advisory, farmers rely on agricultural input dealers who fill the informational void (Gaurav 2019). We also explore if perceptions towards pesticides play a role in pesticide spending by farmers. Table 8 presents some of the differences in pesticide expenditure per acre across various parameters.

From Table 8, we find that per-acre pesticide expenditure is higher among small and marginal farmers in 2015–16, but the difference is not statistically significant. The farmers with access to irrigation spent more on pesticides in both the years and the difference is statistically significant in 2016–17. Farmers who reported changing pesticides regularly spent more on pesticides in both the years. The difference was statistically significant in 2016–17. We also found that farmers who bought pesticides on credit spent more on them. The difference was statistically significant in 2015–16. Farmers who reported that pesticide application increased the yield also spent more per acre. This difference was statistically significant in both the years. The findings are in line with Seo (2013) who found that farmers who perceived pesticides as a damage-control input tended to use it lesser than those who perceived it as a yield-enhancing input.

Concluding Remarks

This study explores the patterns of pest management in cotton farming in the Wardha and Yavatmal districts of Vidarbha region during 2015–16 and 2016–17. We find that despite near universal diffusion of Bt cotton that has seemingly reduced infestation of bollworms, there is still a considerable incidence of secondary pests. In the absence of timely advisory on pest management, farmers use a cocktail of pesticides to deal with pest infestations. The use of cocktails is typically considered ineffective both in the short run and the long run when taking pest resistance into account (Seo 2013). We also find that farmers used more of monocrotophos (organophosphate) in 2016–17 as compared to 2015–16. Even though monocrotophos is cheaper, there have been calls to ban the pesticide because of its toxic nature. The practices followed by farmers also are key to the issue of pesticide usage. Furthermore, we found considerable use of weedicides and tonics (plant growth promoters).

We also found that farmers who believed that pesticides are yield-enhancing (and not just damage-control) input spent more on pesticides in both the years. In future studies, there is scope for further analysing linkages among knowledge, perceptions, learning, and usage of pesticides. We also emphasise that the costs of pesticide usage go beyond economic costs and have significant environmental and ecological implications. The intersection of technological advances and current practices of farmers in pesticide usage also needs to be explored further. For example, we found growing popularity of battery-operated knapsack spray pumps manufactured in China. Some of the farmers felt that these pumps sprayed pesticides faster and could have a detrimental effect on health if the person spraying pesticides did not put on
appropriate protective wear. While technology may be labour-saving, its widespread use may result in considerable adverse health impacts.

The findings of the study have some important policy implications in light of the market failures and state failures in pest management in agriculture. In 2017, the farmers in Yavatmal district sprayed huge amounts of mixtures of various highly hazardous pesticides to tackle the secondary pests and pink bollworms (as they developed resistance to the Bt gene). This, along with the use of high-volume spraying pumps, resulted in severe pesticide poisoning, leading to the death of at least 22 farmers and hospitalisation of at least 1,000 farmers (Menon 2018b; Dahat 2018). Such instances reinforce the fact that there has been no internalisation of costs imposed by externalities of pesticide use as farmers continue using more of toxic pesticides. In the absence of timely and effective advisory services, agricultural input dealers influence the pesticides used as well as practices of farmers. There is also considerable deskilling as new chemical compositions keep coming, and farmers are ignorant about scientifically recommended proportions in a cocktail. Similar deskilling has been observed in a selection of genetically modified cotton seeds in the context of cotton farmers in Andhra Pradesh (Stone 2007).

In the context of reduction of chemical use in pest management in cotton, the limitations of IPM as well as lessons from NPM in some parts of the country need to be assessed for successful implementation in other regions. There is a need for filling up vacant positions of agricultural extension workers and greater investment in training farmers in light of the weakness of the farmer field school model.


1 By pesticides, we mean chemical insecticides, weedicides, herbicides, and rodenticides.

2 The study covered 10 villages in Wardha and eight in Yavatmal. We used the land records (“saat bara”—7/12—and “aath a”—8A) to sample farmer households randomly. The talukas were also randomly selected. Wardha and Yavatmal were selected for comparability of agricultural outcomes despite similar agricultural conditions. In 2016, six automated weather stations (AWS) were installed, three each in Wardha and Yavatmal villages, in order to study the relationship between weather data and agricultural outcomes (Gaurav and Chaudhary 2018).

3 Although the extent of rabi cultivation is minimal, with those having irrigation and adequate retained soil moisture choosing to grow wheat and chana (chickpeas), we collected information on rabi cultivation from the second round onwards.

4 Farmers reported brand names as well as chemical compositions to the best of their knowledge. There was considerable variation in the reporting of a standard pesticide, and trained field investigators cross-verified the information provided with packages or bills wherever available.


Dahat, Pavan (2018): “Yavatmal Pesticide Deaths: SIT Blames Farmers for Negligence,” 27 January, Hindu,, viewed on 5 October 2019.

Gaurav, Sarthak (2014): “Risk and Vulnerability of Agricultural Households in India,” unpublished PhD thesis, Indira Gandhi Institute of Development Research (IGIDR), Mumbai.

— (2015): “Are Rainfed Agricultural Households Insured? Evidence from Panel Data in Village Vidarbha, India,” World Development, Vol 66, pp 719–36.

— (2019): “Agriculture or ‘Agreeculture’: Agricultural Extension Dystopia in India,” unpublished manuscript.

Gaurav, Sarthak and S Mishra (2015): “To Bt or Not to Bt: Risk and Uncertainty Considerations in Technology Assessment,” India’s Tryst with Bt Cotton, N Lalitha and P K Viswanathan (eds), New Delhi: Concept, pp 123–55.

Gaurav, Sarthak and V Chaudhary (2018): “Do Farmers Care about Basis Risk: Evidence from a Field Experiment in Maharashtra, India,” unpublished manuscript.

Gervais, J A et al (2010): “Imidacloprid Technical Fact Sheet,” National Pesticide Information Center, Oregon State University Extension Services,, viewed on 24 January 2019.

Ghadyalpatil, Abhiram (2017): “Vidarbha Pesticide Deaths: Probe Panel May File Report This Week,” livemint,, viewed on 1 October 2019.

Kale, N M, S R Khonde and D M Mankar (2014): “Socio-economic, Psychological and Situational Causes of Suicides of Farmers in Vidarbha Region of Maharashtra,” Karnataka Journal of Agricultural Science, Vol 27, pp 40–46.

Kranthi, K R (2014): “Cotton Production Systems—Need for a Change in India,” Cotton Statistics and News, Cotton Association of India, Mumbai, pp 4–7.

London, L et al (2005): “Suicide and Exposure to Organophosphate Insecticides: Cause or Effect?” American Journal of Industrial Medicine, Vol 47, No 4, pp 308–21.

Menon, M (2018a): “Over the Years, Bt Cotton Has Exacerbated Vidarbha’s Existing Vulnerabilities,” Economic & Political Weekly, Vol 53, No 21, pp 1–9.

— (2018b): “Yavatmal Pesticide Poisoning: Lack of Accountability, Negligence and Abysmal Medical Facilities behind Farmers’ Deaths,” Firstpost, 1 November,
, viewed on 5 October 2019.

Mishra, Srijit (2006): “Farmers’ Suicides in Maharashtra,” Economic & Political Weekly, Vol 41, No 16, pp 1538–45.

Oerke, Eric-Christian (2006): “Crop Losses to Pests,” Journal of Agricultural Science, Vol 144, No 1, pp 31–43.

Public Eye (2018): “The Yavatmal Scandal,”, viewed on 1 October 2019.

Ramanjaneyulu, G V et al (2004): “No Pesticides, No Pests: The Remarkable Story of How a Village in Andhra Pradesh Rid Itself of Pesticides,” Centre for Sustainable Agriculture, Secunderabad, India,, viewed on 11 October 2019.

Ranganathan, Thiagu, S Gaurav, and I Halder (2018): “Pesticide Usage by Cotton Farmers in India: Changes over a Decade,” Economic & Political Weekly, Vol 53, No 19, pp 43–51.

Ross, Sarah Mackenzie et al (2013): “Neurobehavioral Problems Following Low-level Exposure to Organophosphate Pesticides: A Systemic and Meta-analytic Review,” Critical Reviews in Toxicology, Vol 43, No 1, pp 21–44.

Seo, Hee Kwon (2013): “Can Behavioral Biases Explain Demand for a Harmful Pesticide? Evidence from India,” Discussion Paper 13–45, Harvard Environmental Economics Program, Harvard Kennedy School.

Shetty, Paddu Krishnappa (2004): “Socio-ecological Implications of Pesticide Use in India,” Economic & Political Weekly, Vol 39, No 49, pp 5261–67.

Shetty, P K, M Murugan and K G Sreeja (2008): “Crop Protection Stewardship in India: Wanted or Unwanted,” Current Science, Vol 95, No 4, pp 457–64.

Stone, Glenn Davis (2007): “Agricultural Deskilling and the Spread of Genetically Modified Cotton in Warangal,” Current Anthropology, Vol 48, No 1, pp 67–103.

Stone, Glenn Davis and A Flachs (2018): “The Ox Fall Down: Path-breaking and Technology Treadmills in Indian Cotton Agriculture,” Journal of Peasant Studies, Vol 45, No 7, pp 1272–96.

Taneja, Sonam (2017): “Why India Continues to Use Lethal Pesticides,” Down to Earth, 27 November,, viewed on 1 October 2019.

Updated On : 11th Nov, 2019


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top