INTRODUCTION
Worldwide population growth and increasing per capita consumption have amplified the global food demand (Schneider et al., 2011), which is a challenge considering the limited scope of horizontal expansion of agricultural land. The importance of interrelationships among food security, agricultural production, and environmental sustainability in the form of sustainable food production systems is being recognized globally more than ever before. Looking back, the technological progress mobilized by the Green Revolution to enhance food security promoted the extensive use of fertilizer (Gollin et al., 2021). The effects of the Green Revolution were visible in India, where food grain production increased considerably from 82 million tons in 1960 to 316.06 million tons in 2020–21, making the country largely self-sufficient (GoI, 2022).
The Green Revolution, however, led to the use of nitrogenous fertilizers increasing severalfold despite rice and wheat (Triticum aestivum L.) experiencing near-yield stagnation (Ray et al., 2012). Heavy subsidies and low awareness of optimal fertilization rates among farmers have led to excessive consumption of nitrogen fertilizers (Kishore et al., 2019). This inefficient use of nitrogen fertilizers causes severe environmental problems such as soil acidification (Wang et al., 2023), greenhouse gas emissions (nitrous oxide (N2O)) (Davidson, 2009), and groundwater contamination (Bijay-Singh & Craswell, 2021). At the same time, these leakages also reduce the nitrogen (N) available for crops and impose an enormous hidden cost on the exchequer (in the form of subsidies for more fertilizer use). Excessive use of inorganic fertilizers in South Asia has been linked to various health problems. High nitrate levels in drinking water cause methemoglobinemia or “blue baby syndrome” in infants. Additionally, long-term exposure increases cancer risks from bioaccumulated heavy metals and other life-threatening conditions (Wolfe & Patz, 2002; Rojas-Urrea et al., 2023). Further, South Asia and India specifically have emerged as significant nitrogen pollution hotspots due to intense agricultural activities and rapid industrialization (Ul-Haq et al., 2017; Shifa et al., 2022).
Notably, the statistics are not homogenous across India. For example, in the hill states, the application of fertilizer is far below the recommended rates compared to elsewhere in India which are much higher (Chand & Pavithra, 2015). The uneven optimum use of fertilizers across India has implications for agricultural productivity and environmental sustainability. Both overuse and underuse of fertilizers have negative consequences for the soil, yields, and economic returns. Despite these concerns, there are limited studies on regional variations in fertilizer use across India where local practices, climate, and economic conditions may influence farmers' fertilizer use.
Numerous factors influence nitrogen fertilizer use in agriculture, including farm characteristics, economics, agronomics, technology, and external influences (Mujeri et al., 2012; Farnworth et al., 2017; Aryal et al., 2021; Begho et al., 2022). Farmer characteristics, such as age and education, affect practices, with older farmers relying on experience and more educated farmers using synthetic fertilizers (Aryal et al., 2021; Begho et al., 2022). Household size influences labour availability, and gender differences impact decision-making in fertilizer use (Aryal, 2021; Takeshima, 2019). Market demand, input costs, and subsidies also drive fertilizer application, with high prices encouraging efficiency and subsidies encouraging overuse (Holden & Lunduka, 2012; Nasrin et al., 2019). Economic risks from under-fertilizing also motivate overuse (Davidson et al., 2014). Larger farms tend to adopt advanced techniques and use more fertilizer (Wu, 2011; Aryal et al., 2021), while climate, especially rainfall, impacts fertilizer effectiveness, contributing to losses in efficiency (Liu et al., 2016). External factors like training, access to services, and participation in farmer organizations can help reduce overuse by raising awareness about environmental impacts (Pan & Zhang, 2018; Aryal, 2021). However, the impact of socioeconomic, agronomic, technological, and institutional factors on fertilizer use in India remains underexplored. Also, studies jointly assessing organic and inorganic fertilizer use in India are limited.
To address these gaps, this paper examines the determinants of synthetic fertilizer and manure adoption and intensity of use in India.
MATERIALS AND METHODS
Relationship among fertilizer use efficiency, nitrogen use efficiency and fertilizer expenditure
Fertilizer use efficiency (FUE) is a crucial indicator of the effective rate of fertilizer utilization, that is, how effectively fertilizer is converted into crop yields (Bai et al., 2019). A related metric to FUE is Nitrogen use efficiency (NUE), which is used to assess how effectively plants utilize nitrogen from fertilizers (Brentrup & Palliere, 2010). South Asia has one of the lowest NUEs in the world (Ladha et al., 2020). In India, the average crop NUE has declined from 55% to 30–35% in 2010 (Bijay-Singh et al., 2017). Therefore, the challenge is how best to enhance NUE in regions with excessive application and enable a transition from overuse to sustainable use of nitrogen fertilizers.
There is a financial incentive for farmers to maximize FUE as it directly affects their profitability. Overuse of fertilizers can lead to diminishing returns where the cost of additional input exceeds the value of the increased yield. Farms that can maintain or improve crop yields while reducing the expenditure on fertilizers per hectare are likely more efficient. Studies (e.g., Bijay-Singh et al., 2017; Baral et al., 2020; Mirzakhaninafchi et al., 2021; Zou et al., 2024) have shown that FUE and NUE can be improved to a large extent by various precision nutrient tools, techniques and management practices such as adjusting the fertilizer application time, rate, and method as per crop demand, using slow-release fertilizers, urease and nitrification inhibitors, coating with neem oil, and optimizing irrigation. When this is combined with high-yielding, drought-tolerant, and short-duration crop varieties, farmers achieve better NUE (Anik et al., 2023). NUE is also higher when farmers used both organic and inorganic fertilizers efficiently together (Mamuye et al., 2021).
Survey description
The study utilizes a survey data set from the “Situation Assessment of Agricultural Households and Land and Livestock Holdings of Households in Rural India” of the 77th National Sample Survey (NSSO). The data collection for this comprehensive country-wide data set spread over 5,940 villages, covering 45,714 agricultural households, started in January 2019. The first visit was conducted from January to August 2019 to collect the data for the Kharif season (July–December 2018), and the same households were revisited from September to December 2019 for data on the Rabi season (January–June 2019). The reference period for data collection was the agricultural year from July 2018 to June 2019. Since the survey provides information regarding total expenditure on synthetic fertilizer and manure for all crops, data from 14,921 farmers growing only paddy rice in the Kharif season was available. The expenditure on synthetic fertilizers and manures per hectare of paddy area for the selected farmers are then calculated. After removing outliers, the final data set consisted of 14,669 farmers.
Variables included in the model
Guided by the findings and discussions in the literature, data on several socioeconomic characteristics such as age, education, gender, social group, and household size were collected. Economic factors included in the analysis are farm size, irrigation, land and livestock ownership, employment in subsidiary income support activities (MGNREGA), financial inclusion, and expenditure incurred on other inputs. In addition to social factors like access to information and services, membership in farmers' organizations, participation in agricultural training, and access to institutional services such as soil health cards, Kisan credit cards, and crop insurance schemes were also included. Data on climate variables, rainfall, and temperature were collected from the Indian Meteorological Department and added to this data set.
Empirical methods
Binary Choice models (Logit and Probit) have been commonly used for modelling decisions regarding technology adoption or participation behaviour when the dependent variable takes either a 1 (adoption) or 0 (non-adoption) value. However, when the dependent variable (Y) exhibits censoring, meaning specific values of Y are below the lower limit (i.e., zero), OLS or binary choice models fail to yield reliable estimates. Estimations of such regression models where Y is normal but truncated to the left of zero are done using the Tobit model. However, one of the critical limitations of the Tobit model is that it assumes the same variables determine the probability of fertilizer adoption and the level of adoption (intensity of adoption). It assumes the same stochastic process for both the value of continuous observations on the dependent variable and the discrete switch at zero (Blundell & Meghir, 1987).
Heckman's two-step model is an alternative that involves the probit model in the first step (selection). In the second stage, OLS is carried out using the Inverse Mills Ratio (IMR) as an additional explanatory variable to control for selection bias (Wooldridge, 2002). However, the Heckman model does not allow zero in the second stage, which limits its applicability. Therefore, this study applies Cragg's independent double-hurdle model, which allows separate stochastic processes for the incidence and intensity of adoption (Cragg, 1971) and allows for zero in the second stage.
It is hypothesized that variables affecting the fertilizer adoption decision may not necessarily determine the amount of fertilizer used. The expenditure incurred on fertilizers (synthetic/manures) could be explained by various farmer-specific and socioeconomic factors that may be observed and unobserved. A double-hurdle model would be a suitable choice in this case. In the double hurdle model, two distinct decisions regarding the adoption of fertilizer and the intensity of use are determined by two different sets of explanatory variables. In this case, the first hurdle that farmers face is the decision to adopt synthetic/organic fertilizer, which is determined by a binary variable.
Here, is a latent variable indicating whether the farmer adopts the fertilizer/manure or not. It takes the value 1 if the farmer incurred any expenditure on fertilizer/manure, zero otherwise. is a vector of independent covariates that explain the farmer's decision regarding fertilizer adoption, and is an unobserved error term.
The second hurdle is the intensity of use determined by the following equation:
Where depicts the amount of expenditure incurred on synthetic fertilizer/manure per ha and is the vector of covariates that explain the amount used. is the unobserved error term. Both the error terms are assumed to be independent and normally distributed with zero mean and constant variance. This model allows for possible differences between the factors affecting the adoption decision and the intensity of the fertilizer/manure applied. It is worth mentioning that although the study used the amount of expenditure incurred per hectare as a proxy for intensity of use, the extensive data set of 14,921 is representative of Indian farmers and provides substantial scope for achieving the study objective which other existing household surveys in India cannot currently provide.
RESULTS
The description of the variables used in the analysis is provided in the appendix in Table A1. Most of the sample households were male-headed, with an average age of 50 years. Only about 32% of the households had farm sizes greater than 1 hectare, indicating that most were small farmers. Nearly 90 percent of farmers owned land, and the average crop irrigation coverage was reported to be 76%. Households, on average, had five members. Only 7% of the household heads hold graduate or post-graduate degrees. About 20% had access to fertilizer information, the government's employment guarantee scheme, and Kisan Credit Cards (KCC for short-term crop loans). Regarding financial inclusion, about 98% of the households had bank accounts, but only 16% of them had credit. Finally, about 30% of farmers belonged to the disadvantaged section of society.
Description of dependent variables
The histograms in Figure 1 represent the distribution of expenditures on synthetic fertilizers and manures by all farmers in the sample. Figure 1 shows that expenditures on synthetic fertilizers are positively skewed, with most farmers spending less than 10,000 Rs/ha. There is a gradual decline in frequency as expenditure increases, indicating that fewer farmers spent heavily on synthetic fertilizers beyond 20,000 Rs/ha. This distribution suggests a possible variance in the levels of fertilizer inputs. In contrast, manure expenditures show a sharper skew, with the majority of farmers spending under 5,000 Rs/ha. This steeper decline in spending could imply a lower reliance or preference for manure compared to synthetic fertilizers, or could reflect differences in the cost or availability of manure. In Figure 2, the distribution of expenditures on synthetic fertilizers and manures by farmers in the second hurdle indicates a positive skew, but with a broader spread in expenditures compared to Figure 1. Specifically, for manure expenditures, there are more expenditures recorded above 5000 Rs/ha, suggesting a slight increase in manure use among this group of farmers.
[IMAGE OMITTED. SEE PDF]
[IMAGE OMITTED. SEE PDF]
Pearson's correlation coefficient between the continuous variables included in the analysis
Figure 3 summarizing Pearson's correlation coefficients between various agricultural expenditures shows varied correlations among spending categories. There is a moderate positive correlation between expenditures on synthetic fertilizers and labour costs, possibly indicating that more fertilizers also demand more labour. Weak positive correlations are observed between fertilizer costs and both manure and irrigation expenditures, indicating slight tendencies for these costs to increase together.
[IMAGE OMITTED. SEE PDF]
Determinants of synthetic fertilizer adoption and use intensity
Table 1 represents the results of Cragg's double hurdle model, which determines the factors affecting the adoption of synthetic fertilizers and the intensity of use among paddy farmers. In the first hurdle (adoption model), farmers' age, total paddy area, participation in MGNREGA work, holding of a Kisan Credit Card, paddy irrigated area, access to fertilizer information, expenditure on irrigation and labor, loan taken, land ownership, rainfall, temperature, and geographical zones significantly influenced the adoption of synthetic fertilizers. Specifically, older farmers, larger farms, landowners, and zones in the east and west were found to be less likely to adopt synthetic fertilizers, while the other factors positively influenced adoption.
Table 1 Determinants of synthetic fertilizer use among paddy farmers.
Parameters | Adoption of synthetic fertilizers | Intensity of fertilizers use/ha (expenditure level) | ||||
Coef. | Std. Err. | P > z | Coef. | Std. Err. | P > z | |
Gender | 227.56 | 189.15 | 0.229 | 0.189a | 0.064 | 0.003 |
Age | −11.51a | 4.12 | 0.005 | 0.004a | 0.002 | 0.004 |
Household size | 29.78 | 22.86 | 0.193 | |||
Total paddy area (ha) | −972.80a | 77.47 | 0.000 | 0.025 | 0.029 | 0.374 |
Education | −121.54 | 204.75 | 0.553 | −0.051 | 0.087 | 0.562 |
Agricultural training | −476.87 | 426.61 | 0.264 | −0.397a | 0.146 | 0.006 |
Social group | 75.13 | 117.69 | 0.523 | −0.169a | 0.041 | 0.000 |
MGNREG work | 618.64a | 132.00 | 0.000 | −0.024 | 0.049 | 0.626 |
Farmer organisation | −89.79 | 251.12 | 0.721 | 0.067 | 0.122 | 0.582 |
Kisan Credit Card | 491.77a | 140.84 | 0.000 | 0.218a | 0.062 | 0.000 |
Soil Health Card | −136.66 | 461.71 | 0.767 | −0.302 | 0.219 | 0.168 |
Paddy irrigated area percentage | 19.09a | 1.53 | 0.000 | 0.005a | 0.000 | 0.000 |
Fertilizer information | 598.73a | 123.54 | 0.000 | 0.425a | 0.062 | 0.000 |
Expenditure on manure (Rs/ha) | −0.05 | 0.03 | 0.156 | 0.001a | 0.001 | 0.000 |
Expenditure on irrigation (Rs/ha) | 0.23a | 0.01 | 0.000 | 0.001a | 0.000 | 0.000 |
Expenditure on labour (Rs/ha) | 0.22a | 0.01 | 0.000 | 0.001a | 0.000 | 0.000 |
Expenditure on crop insurance (Rs/ha) | 0.24 | 0.16 | 0.151 | |||
Loan taken | 621.73a | 133.86 | 0.000 | 0.311a | 0.070 | 0.000 |
Bovine ownership | 57.58 | 108.27 | 0.595 | −0.220a | 0.042 | 0.000 |
Land ownership | −321.32b | 159.83 | 0.044 | −0.354a | 0.070 | 0.000 |
Rainfall (mm) | 0.55a | 0.11 | 0.000 | 0.001a | 0.000 | 0.000 |
Temperature (max.) | 240.68a | 50.66 | 0.000 | 0.162a | 0.014 | 0.000 |
HH bank account | 748.21 | 416.46 | 0.072 | |||
zone(1=North) | ||||||
2(East) | −577.89a | 150.84 | 0.000 | −0.188a | 0.069 | 0.007 |
3(South) | 1926.02a | 241.06 | 0.000 | −0.041 | 0.125 | 0.744 |
4(West) | −1143.17a | 290.94 | 0.000 | −0.478a | 0.095 | 0.000 |
_cons | −8418.76a | 1737.03 | 0.000 | −3.575a | 0.447 | 0.000 |
lnsigma | ||||||
_cons | 8.387 | 0.011 | 0.000 | |||
/sigma | 4391.814 | 50.040 |
In contrast, fertilizer intensity of use (expenditure model) was influenced by gender, age, training, social group of farmers, expenditure on irrigation, manure and labor, Kisan Credit Card, paddy irrigated area, access to fertilizer information, loan, ownership of livestock, land ownership, temperature, and the region's geographical zones. Among these variables, the drivers of high intensity of fertilizer use were male and older farmers, larger farms, no formal training in agriculture, ownership of a Kisan Credit Card, spending on irrigation, manure and labor, having access to a loan, and a high seasonal average maximum temperature and rainfall (Table 1).
Determinants of manure adoption and use intensity
The determinants of adoption and intensity of use (expenditure model) of manure are presented in Table 2. The adoption of manure was positively influenced by holding a Kisan Credit Card, having access to loans, location, and expenditures on irrigation, labour and fertilizers. Conversely, the area under cultivation, membership in farmer organizations, land ownership, and temperature negatively influenced adoption.
Table 2 Determinants of manure use among paddy farmers.
Parameters | Adoption of manure | Intensity of manure use/ha (expenditure level) | ||||
Coef. | Std. Err. | P > z | Coef. | Std. Err. | P > z | |
Gender | 443.83 | 623.28 | 0.48 | −0.08 | 0.04 | 0.08 |
Age | 11.43 | 14.51 | 0.43 | 0.01b | 0.00 | 0.02 |
Household size | −124.35 | 92.42 | 0.18 | −0.01b | 0.01 | 0.01 |
Total paddy area (ha) | −2595.00a | 353.77 | 0.00 | 0.08a | 0.02 | 0.00 |
Education | 67.46 | 707.88 | 0.92 | 0.06 | 0.05 | 0.18 |
Agricultural training | 103.72 | 1279.41 | 0.94 | 0.04 | 0.10 | 0.71 |
Social group | −1075.04b | 433.42 | 0.01 | −0.01 | 0.03 | 0.75 |
MGNREG work | 124.20 | 441.19 | 0.78 | 0.00 | 0.03 | 0.88 |
Farmer organisation | −2373.77b | 836.02 | 0.01 | 0.20a | 0.06 | 0.00 |
Kisan Credit Card | 1295.94b | 550.65 | 0.02 | −0.02 | 0.03 | 0.63 |
Soil Health Card | 207.38 | 1345.24 | 0.88 | 0.13 | 0.11 | 0.27 |
Paddy irrigated area percentage | −8.48 | 5.15 | 0.10 | 0.01a | 0.00 | 0.00 |
Fertilizer information | 387.90 | 439.31 | 0.38 | −0.06 | 0.03 | 0.06 |
Expenditure on fertilizers (Rs/ha) | 0.25a | 0.05 | 0.00 | 0.01a | 0.00 | 0.00 |
Expenditure on irrigation (Rs/ha) | 0.21a | 0.05 | 0.00 | 0.00 | 0.00 | 0.94 |
Expenditure on labour (Rs/ha) | 0.21a | 0.02 | 0.00 | 0.01a | 0.00 | 0.00 |
Expenditure on crop insurance (Rs/ha) | 0.67 | 0.48 | 0.16 | 0.00 | 0.00 | 0.88 |
Loan taken | 1349.77a | 472.28 | 0.00 | −0.03 | 0.03 | 0.33 |
Bovine ownership | 128.12 | 385.98 | 0.74 | −0.38a | 0.02 | 0.00 |
Land ownership | −1746.63a | 586.98 | 0.00 | 0.02 | 0.04 | 0.61 |
Rainfall (mm) | 0.35 | 0.44 | 0.43 | 0.01a | 0.00 | 0.00 |
Temperature (max.) | 728.23 | 291.24 | 0.01 | 0.00 | 0.02 | 0.83 |
Temperature (min.) | −915.52a | 246.30 | 0.00 | 0.05a | 0.02 | 0.00 |
HH bank account | −122.34 | 1227.44 | 0.92 | −0.21b | 0.08 | 0.01 |
Fasal_bima_yojana | −1670.23 | 898.58 | 0.06 | 0.03 | 0.06 | 0.58 |
zone(1=North) | ||||||
2(East) | 182.03 | 763.61 | 0.81 | 0.43a | 0.04 | 0.00 |
3(South) | 7212.42a | 1086.05 | 0.00 | 0.61a | 0.06 | 0.00 |
4(West) | 1810.36 | 1161.16 | 0.12 | 0.26a | 0.07 | 0.00 |
Constant | −10777.93 | 6622.91 | 0.10 | −1.88a | 0.42 | 0.00 |
Lnsigma | ||||||
_cons | 8.45 | 0.05 | 0.00 | |||
/sigma | 4692.69 | 232.87 |
For the intensity of use of manure, factors such as age, area under cultivation, membership in farmer organizations, paddy irrigated area, expenditures on labour and fertilizers, temperature, and geographical zones were positively associated with the intensity of use. In contrast, owning bovines, having a bank account, and a large area under cultivation negatively influenced the intensity of use.
DISCUSSION
Similarities and differences in factors affecting adoption of fertilizer and manure, and the determinants of use intensity (expenditure)
In terms of the similarities and differences in adoption factors between fertilizer and manure both synthetic fertilizers and manure adoption are influenced by several common factors. Firstly, access to financial resources through instruments like the Kisan Credit Card and loans positively affects the adoption of both synthetic fertilizers and manure. With credit options, farmers can invest in these inputs to improve crop yields and productivity without being constrained by immediate cash flow limitations. Also, credit allow farmers to spread the cost of purchasing fertilizers and manure over time, making it easier to manage their budgets. This reduces the perceived risk of making significant financial investments in agricultural inputs, which could otherwise deter adoption. This finding aligns with previous studies showing that farmers with access to credit facilities are better positioned to invest in agricultural inputs, including fertilizers, which are often capital-intensive. This also corroborates previous findings that farmers with credit access are more likely to adopt modern agricultural technologies and inputs (Aryal et al., 2021; Begho et al., 2022). It also supports findings that suggest such access to credit provides farmers with the flexibility to experiment with different inputs and take up management practices to optimize their farm operations (Jimi et al., 2019).
Secondly, expenditure on inputs such as irrigation and labour emerge as a common determinant for both synthetic fertilizers and manure adoption. This may be attributed to farmers allocating resources overall towards improving agricultural productivity. It suggests that when farmers prioritize their spending on fundamental agricultural resources like water and labour, they are more inclined to adopt inputs that enhance crop yield which reflects an interconnected approach to farm management. Thirdly, the likelihood of adoption of both synthetic fertilizers and manure is lower among landowners and those with larger paddy areas. Fourthly, geographical location plays a crucial role in influencing the adoption of both synthetic fertilizers and manure. The finding that farmers in the southern zone are more likely to adopt synthetic fertilizers compared to those in the northern zone could be influenced by regional factors such as soil type, climate, and historical farming practices (Quemada & Gabriel, 2016; Tefera et al., 2020). Factors such as the prevalence of sustainable agricultural practices or the availability of alternative soil fertility management methods may also contribute to the lower adoption rates of synthetic fertilizers in these areas.
While there are similarities in the factors influencing the adoption of synthetic fertilizers and manure, there are also differences. One significant difference is that younger farmers are more likely to adopt synthetic fertilizers. This could be attributable to factors such as exposure to education, access to information, and a greater willingness to take risks in adopting new methods to enhance productivity. Additionally, younger farmers are often more receptive to government extension programs and initiatives aimed at promoting the use of synthetic fertilizers, further facilitating their adoption, which supports part of the debate on the role of age (Rizzo et al., 2024). It could also be due to older farmers' adherence to traditional farming practices based on indigenous knowledge and experience (Hamadani et al., 2021). This suggests that policies targeting younger farmers may be effective in promoting sustainable synthetic fertilizer use, but that additional efforts to engage older farmers in innovation may be necessary, perhaps by integrating indigenous knowledge with modern techniques. However, age does not have a significant impact on manure adoption. This might indicate that manure, as a more traditional or organic input, is perceived differently to fertilizer across generations, with both older and younger farmers finding it equally familiar and acceptable for their agricultural practices.
Another contrasting factor is membership in farmer organizations, which influences manure adoption but does not significantly affect the adoption of synthetic fertilizers. Farmer organizations may play a crucial role in promoting sustainable agriculture and organic farming practices, thereby influencing farmers' decisions to adopt manure. However, their influence may be less pronounced in the case of synthetic fertilizers, which are often marketed and promoted by agrochemical companies through alternative channels (Sultana et al., 2015; Kishore et al., 2019). This difference highlights the potential of farmer organizations as key drivers of sustainable practices, suggesting that their role could be further strengthened to encourage the adoption of organic inputs such as manure. By contrast, synthetic fertilizer adoption might require partnerships with commercial actors rather than farmer-based organizations. Lastly, undertaking any work under the MGNREGA scheme (the employment guarantee scheme of the government) influenced fertilizer adoption but not manure. This indicates that participation in government schemes might provide additional resources or incentives that specifically encourage the use of synthetic fertilizers, whereas manure adoption may remain tied to traditional or local factors which farmer organizations likely promote.
In regard to the similarities and differences in the determinants of use intensity (expenditure) between fertilizer and manure, older farmers who adopt fertilizers show a higher intensity of use of both synthetic fertilizers and manure. The higher expenditure on fertilizers and manure by older farmers could also be indicative of their accumulated knowledge and reliance on established practices and their tendency to invest in inputs that they have been tested to improve yields. Also, with age comes greater farming experience, allowing them to better understand the benefits and application of these inputs. Arguably, older farmers often have a more stable financial base which enables them to invest more in agricultural inputs without being as constrained by short-term financial limitations. Both synthetic fertilizer and manure adoption are positively influenced by expenditure on labour. This indicates that farmers who can pay for the additional labour required to use both types of fertilizers will likely have a higher intensity of use. Both synthetic fertilizers and manure require significant labour for application. Farmers who can afford labour are better equipped to handle the physical demands of using both fertilizers and allowing for higher intensity of use and improved management of their crops.
Ownership of livestock negatively influences fertilizer and manure expenditure. This could be because farmers with livestock have access to organic manure from their animals, reducing their need to purchase additional fertilizers. Additionally, synthetic fertilizer and manure expenditure are influenced by geographical factors such as temperature and geographical zones. Higher temperatures and being in a specific geographical region, such as the eastern region, positively impact expenditure on both types of fertilizers. This reinforces the idea that climatic conditions and local agro-ecological factors significantly shape farming practices. Farmers in regions with favourable conditions for crop growth may feel incentivized to spend more on fertilizers with the expectation of better returns from their expenditure.
Regarding the differences in the factors influencing expenditure on synthetic fertilizer and manure, while male farmers show a higher intensity of use of synthetic fertilizers, gender does not significantly influence manure expenditure. Formal training in agriculture negatively influences the intensity of synthetic fertilizer use, but there is no association for manure expenditure. This could be explained by the fact that formal training might promote more efficient or precise fertilizer application techniques, reducing the need for high fertilizer input. It also suggests that training programs could further focus on enhancing manure management practices. Lastly, ownership of a Kisan Credit Card and access to loans positively influence the expenditure on synthetic fertilizers but not on manure. The key findings are summarized in Figure 4.
[IMAGE OMITTED. SEE PDF]
Implications for productivity and sustainability
Both underuse and overuse of fertilizer undermine long-term soil fertility and productivity. For example, where fertilizer use is below recommended rates, farms experience lower yields due to nutrient deficiencies. This leads to reduced crop outputs and lower income for farmers. The depletion of soil nutrients can further fuel cycles of food insecurity and poverty in these areas. In contrast, in regions where fertilizers are overused, crop yields do not significantly increase, resulting in resource wastage and increased production costs without corresponding economic benefits. Further, excessive fertilizer use can lead to environmental pollution, degrade soil health, and incur higher costs for soil remediation.
The finding that the decision to adopt fertilizers and the decision on their intensity of use are independent suggests that interventions aimed at increasing adoption may not necessarily result in higher intensity of use if this is not the goal. This understanding is crucial for designing policies that not only encourage adoption but also promote the sustainable and efficient use of fertilizers to maximize productivity. Addressing these issues requires a balanced, region-specific approach, involving tailored fertilizer recommendations, farmer education, integrated nutrient management, and supportive government policies to promote sustainable agricultural practices and protect the environment.
Limitation of the study
Farmers' fertilizer use intensity is assessed through their fertilizer expenditure rather than the traditional weight-based method. It is important to acknowledge that lower expenditure does not necessarily translate to better use efficiency; it often depends on the type of fertilizer, timing of application, and adherence to best agricultural practices. Additionally, changes in fertilizer prices can affect expenditure without reflecting changes in the actual quantity or efficiency of fertilizer use. Future studies can adjust for price fluctuations when using expenditure as a proxy. Also, different fertilizers have different costs and efficiencies (e.g., slow-release vs. quick-release). From this perspective, studies considering the type and formulation of fertilizers used may produce more precise findings.
In India, where subsidies are prevalent for fertilizer products and eligibility is not based on individual farmer characteristics, our expenditure-based analysis remains robust. However, it is essential to acknowledge the availability of free manure to certain farmers, particularly those who engage in livestock farming. Although we attempted to mitigate this by including livestock ownership as an explanatory variable, it remains a limitation worth noting, as it may impact the accuracy of our findings. The results point towards the possible reasons for the high use of synthetic fertilizers and fewer manures in paddy cultivation in India.
Research agenda
Concerted efforts beyond the traditional input promotion channels are needed to effectively push the use of manure in paddy cultivation in India and to achieve a balanced fertilizer use status. However, the challenge lies in identifying the levels of fertilizer and manure expenditure that yield the most favourable outcomes, considering both economic and environmental factors. This challenge falls outside the scope of this paper but can be taken up in future research, especially since we acknowledge that higher expenditure may not always be beneficial. Future studies should focus on determining the optimal expenditure to support or contradict whether a positive correlation is good or bad and whether it serves as a reliable proxy for underapplying or overapplying fertilizer.
This study identified that access to information is a key enabling factor for synthetic fertilizer expenditure and that its role is reversed with respect to manure. There is a need for future studies to investigate whether the reason for this might be that the current information available to farmers from public and private extension agencies promotes synthetic fertilizer use rather than manure.
CONCLUSION
The study demonstrated that farmers' decisions to adopt fertilizer or manure and their decisions on expenditure (an indicator of use intensity) are independent. Collectively, the findings are largely consistent with the literature on farmer characteristics, economic factors, farm characteristics, agronomic factors, technological and institutional factors, and other external factors that influence fertilizer use.
From a policy perspective, the identification of common drivers and barriers allows for targeted interventions to promote sustainable fertilizer use, with a focus on following fertilizer recommendations. To achieve this, extension programs should educate farmers on optimized fertilizer application techniques, which reduces wastage and environmental impacts while maintaining or enhancing productivity. Additionally, supporting the adoption of balanced nutrient practices that integrate organic and synthetic fertilizers would enhance soil health and long-term productivity. Considering the intensity of fertilizer use is influenced by factors such as farmer age, livestock ownership, and regional climatic conditions. Programs should be tailored to the specific needs of different farmer groups. For instance, younger farmers and those with limited livestock should receive additional guidance on organic manure integration. It is also important to promote financial accessibility for farmers, particularly through instruments like the Kisan Credit Card and agricultural loans as access to credit has been identified as a key driver for manure adoption. Ensuring that farmers, particularly younger and small-scale farmers, can access financial resources will enable them to invest in more efficient fertilizer use, thus improving crop productivity.
AUTHOR CONTRIBUTIONS
Girish Kumar Jha: Writing—original draft; writing—review & editing; conceptualization; methodology; supervision. Praveen Koovalamkadu Velayudhan: Conceptualization; writing—original draft; writing—review & editing; formal analysis; software; methodology. Toritseju Begho: Conceptualization; original draft; writing—review & editing; methodology. Vera Eory: Conceptualization; supervision; writing—review & editing. Arti Bhatia: Conceptualization; supervision; writing—review & editing.
ACKNOWLEDGEMENTS
This research was funded by UKRI GCRF South Asian Nitrogen Hub (SANH). The project team includes partners from across South Asia and the UK. Neither UKRI nor any of the partner institutions are responsible for the views advanced here.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
ETHICS STATEMENT
The study utilizes a publicly available survey data set from the “Situation Assessment of Agricultural Households and Land and Livestock Holdings of Households in Rural India” of the 77th National Sample Survey (NSSO).
APPENDIX
Table A1 Descriptive statistics of the variables used in the study.
Variable | Mean/percent | Standard deviation | Definition |
Gender | 91.65% | 1 if the household head is male, 0 otherwise | |
Age | 50.25 | 12.85 | Age of the farmer (in years) |
Education | 6.70% | 1 if a farmer is a graduate or post-graduate, 0 otherwise | |
Household size | 4.98 | 2.40 | Total members of the household |
Land size group | 31.60% | 1 If the land size is > 1 hectare, 0 otherwise | |
Crop area | 0.79 | 0.79 | Area under rice (in hectares) |
Social group | 30.96% | 1 if the farmer belongs to the disadvantaged group,a 0 otherwise | |
Agricultural training | 1.47% | 1 if the farmer attended any formal training in agriculture, 0 otherwise | |
Membership in the Farmer organization | 3.92% | 1 if any of the household member is a member of a registered farmer's organization | |
Benefitted from the Employment guarantee scheme of the government (MGNREGA) | 19.98% | 1 If the farmer undertook any work under the MGNREGA scheme during the last 365 days | |
Crop insurance adoption | 4.52% | 1 if the farmer has taken any insurance, 0 otherwise | |
Beneficiary of Kisan Credit Card | 18.02% | 1 if the household possesses a Kisan Credit card, 0 otherwise | |
Soil health card holder | 1.06% | 1 if the household possesses a soil health card, 0 otherwise | |
Fertilizer information | 20.47% | 1 if the household has accessed fertilizer information, 0 otherwise | |
Crop irrigation | 69.24 | 45.01 | % of crop area under irrigation |
Irrigation expenditure | 1902.76 | 3452.22 | Expenditure on irrigation (Rs/ha) |
Expenditure on labour | 6894.92 | 7484.73 | Expenditure on human labour (Rs/ha) |
Expenditure on crop insurance | 35.57 | 275.08 | Expenditure on crop insurance (Rs/ha) |
Loan taken | 16.35% | 1 if loan was taken, 0 otherwise | |
Bank account | 98.06% | 1 if the household holds a bank account, 0 otherwise | |
Bovine ownership | 60.92% | 1 if the household owns a cow or a buffalo, 0 otherwise | |
Land ownership | 88.73% | 1 if the households own the land, 0 otherwise | |
Rainfall | 1409.97 | 641.96 | Seasonal average rainfall (in mm) |
Max. temperature | 31.36 | 1.47 | Seasonal average maximum temperature (in degrees Celsius) |
Anik AR, Eory V, Begho T, Rahman MM. Determinants of nitrogen use efficiency and gaseous emissions assessed from farm survey: a case of wheat in Bangladesh. Agricult Sys. 2023;206: [eLocator: 103617].
Aryal JP, Sapkota TB, Krupnik TJ, Rahut DB, Jat ML, Stirling CM. Factors affecting farmers' use of organic and inorganic fertilizers in south Asia. Environ Sci Pollut Res. 2021;28(37):51480–51496.
Bai X, Wang Y, Huo X, Salim R, Bloch H, Zhang H. Assessing fertilizer use efficiency and its determinants for apple production in China. Ecological Indicators. 2019;104:268–278.
Baral BR, Pande KR, Gaihre YK, Baral KR, Sah SK, Thapa YB, et al. Increasing nitrogen use efficiency in rice through fertilizer application method under rainfed drought conditions in Nepal. Nutrient Cycling in Agroecosystems. 2020;118(1):103–114.
Begho T, Eory V, Glenk K. Demystifying risk attitudes and fertilizer use: a review focusing on the behavioral factors associated with agricultural nitrogen emissions in south Asia. Frontiers in Sustainable Food Systems. 2022;6: [eLocator: 991185].
Brentrup F, Palliere C. (2010). Nitrogen Use Efficiency as an Agro‐Environmental Indicator OECD Workshop on Agri‐Environmental Indicators. At Leysin, Switzerland, 1–9.
Bijay‐Singh A. Management and use efficiency of fertilizer nitrogen in production of cereals in India—issues and strategiesIn. In: Lal R et al. editors. The Indian Nitrogen Assessment: Sources of Reactive Nitrogen, Environmental and Climate Effects, Management Options and Policies. Duxford, U.K: Woodhead Publishing; 2017. p. 149–162. [DOI: https://dx.doi.org/10.1016/B978-0-12-811836-8.00010-0]
Rojas‐Urrea A, Arias‐Mariño D, Velandia‐Siabato DF, García‐Agudelo L, Gonzalez‐Calderon IC. Urea fertilizer poisoning‐induced multiorgan failure: case report. Asia Pacific Journal of Medical Toxicology. 2023;12(3):113–117.
Rizzo G, Migliore G, Schifani G, Vecchio R. Key factors influencing farmers' adoption of sustainable innovations: a systematic literature review and research agenda. Organic Agriculture. 2024;14(1):57–84.
Schneider UA, Havlík P, Schmid E, Valin H, Mosnier A, Obersteiner M, et al. Impacts of population growth, economic development, and technical change on global food production and consumption. Agricult Sys. 2011;104(2):204–215.
Shifa S, Yang A, Anik AR. Bangladesh Nitrogen Policy Report: Scientific Evidence, Current Initiatives and Policy Landscape. 2022.
Sultana J, Siddique MNA, Abdullah MR. Fertilizer recommendation for agriculture: practice, practicalities and adaptation in Bangladesh and Netherlands. International Journal of Business, Management and Social Research. 2015;1(1):21–40.
Takeshima H. Use of chemical fertilizers in Nepal—issues and implications. Agricultural Transformation in Nepal: Trends, Prospects, and Policy Options. 2019;1:231–260.
Tefera T, Elias E, van Beek C. Determinants of smallholder farmers' decisions on fertilizer use for cereal crops in the Ethiopian highlands. Exp Agric. 2020;56(5):677–687.
Ul‐Haq Z, Tariq S, Ali M. Spatiotemporal patterns of correlation between atmospheric nitrogen dioxide and aerosols over south Asia. Meteorology and Atmospheric Physics. 2017;129:507–527.
Wang Z, Tao T, Wang H, Chen J, Small GE, Johnson D, et al. Forms of nitrogen inputs regulate the intensity of soil acidification. Global Change Biol. 2023;29:4044–4055.
Wooldridge JM. Econometric analysis of cross section and panel data. 108(2). Cambridge, MA: MIT press; 2002. p. 245–254.
Wolfe AH, Patz JA. Reactive nitrogen and human health: acute and long‐term implications. Ambio: a Journal of the Human Environment. 2002;31(2):120–125.
Wu Y. Chemical fertilizer use efficiency and its determinants in China's farming sector: implications for environmental protection. China Agricultural Economic Review. 2011;3(2):117–130.
Zou H, Li D, Ren K, Liu L, Zhang W, Duan Y, et al. Response of maize yield and nitrogen recovery efficiency to nitrogen fertilizer application in field with various soil fertility. Front Plant Sci. 2024;15: [eLocator: 1349180].
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (the "License"). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Introduction
Nitrogen use efficiency (NUE) is lower for South Asia than for most other regions of the world, and average crop NUE is on the decline in India. This inefficient use of nitrogen fertilizers has implications for agricultural productivity and environmental sustainability.
Materials and Methods
Using data from 14,669 farmers in India, this paper examined the determinants of synthetic fertilizer and manure adoption and intensity of use for rice (Oryza sativa L.) production. The latter was assessed through fertilizer expenditure rather than the traditional weight‐based method. A double hurdle model was estimated.
Results
The study showed that farmers' decisions to adopt fertilizer or manure and the decision on use intensity were independent. Both synthetic fertilizers and manure adoption were influenced by common drivers such as access to financial resources through instruments like the Kisan Credit Card and loans, expenditure on irrigation and labour, and geographical location. In terms of barriers, the likelihood of adoption of both synthetic fertilizer and manure was lower among landowners and paddy area cultivated. The intensity of fertilizer and manure use was higher for older farmers and was positively influenced by expenditure on labour but negatively influenced by ownership of livestock. Also, synthetic fertilizers and manure use intensity were determined by regional temperature and geographical zones.
Conclusion
The results of this study are useful for targeted interventions to promote sustainable fertilizer use with a focus on following recommendations in zones or among demographic groups that are currently more likely to have a high intensity of use. Similarly, the findings inform support towards increased adoption and sustainable use where fertilizer is underutilised.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details

1 Division of Bioinformatics, ICAR‐Indian Agricultural Statistics Research Institute, New Delhi, India
2 Division of Agricultural Economics, ICAR‐Indian Agricultural Statistics Research Institute, New Delhi, India
3 Rural Economy, Environment & Society, Scotland's Rural College (SRUC), Edinburgh, United Kingdom
4 Division of Environment Science, ICAR‐Indian Agricultural Statistics Research Institute, New Delhi, India