An economic analysis of crop diversification and dynamics of cropping pattern in Karnataka, India

Crop diversification has long been studied for its linkage with development, risk mitigation, sustainability, and livelihood enhancement. Prior work highlights diversification as a response to climate risks, declining productivity under monocultures, and as a pathway to optimize resource use, recycle nutrients, restore soil fertility, and improve economic viability. Drivers include market dynamics, technological access, agro-climatic conditions, infrastructure (communication, marketing, storage), and institutional factors (policies, protection, risk management). Karnataka, India, is agriculturally important: 64.6% of its area is cultivated, agriculture supports 13.74 million workers, yet the economy has shifted toward services. Agricultural GSDP declined in the early 2000s then rose to Rs. 674.57 billion in 2020-21; operational area fell while number of holdings rose (2015-16 census). Against this backdrop, the study aims to: (1) analyze the nature and extent of crop diversification in Karnataka; (2) identify factors influencing crop diversification; and (3) assess the direction of shifts in cropping patterns.

The paper situates its analysis within a broad literature on agricultural and crop diversification. Studies link diversification to high-value crops and smallholder opportunities (Birthal et al., 2013), micro-level decisions for area shifts (Mehta, 2009), and infrastructure’s role (Utpal and Manabendu, 2010). Methodological precedents for measuring diversification include entropy-based measures (Shiyani and Pandya, 1998). Determinants explored in prior work include irrigation and infrastructure (Gupta and Tewari, 1985), policy incentives such as MSP (Aditya et al., 2017; Singh and Sidhu, 2004), and socio-economic factors like risk-taking and farm size (Calle et al., 2022). Regional analyses (e.g., Ergano et al., 2000; Acharya et al., 2011) and climate impacts (Guiteras, 2007) inform how changing rainfall patterns and market-technological changes shape diversification decisions. This literature motivates the paper’s focus on CEI measurement, econometric determinants, and Markov chain-based dynamics for Karnataka.

Study area: Karnataka, India. Data sources: Directorate of Economics and Statistics (DES), Reserve Bank of India (RBI), and Ministry of Statistics and Programme Implementation (MoSPI). Period: 1998-1999 to 2020-2021. Variables compiled as time series include area, production, productivity by crop; fertilizer consumption (kg/ha); area sown more than once (ha); annual rainfall (mm); average size of landholding (ha); MSP of paddy (Rs./qtl); MSP of coarse cereals (Rs./qtl); percentage of gross irrigated area to gross cultivated area; cropping intensity (%); credit to agriculture by scheduled commercial banks (billion Rs.); per-capita NSDP (Rs.); per-capita power availability (kW); length of national highways (km); and number of factories (units). - Crop diversification (Composite Entropy Index, CEI): Used to assess diversification across crop groups and overall. Crop groups: cereals and millets; pulses; sugar crops; condiments and spices; fresh fruits; dry fruits; vegetables; oilseeds; fibre crops; dyes and tanning materials; drugs, narcotics and plantation crops. Constituents for each group follow standard classifications (e.g., cereals include paddy, jowar, bajra, maize, ragi, wheat, barley, other cereals and millets). CEI computed from the average proportion of area under each crop in total cropped area over time. - Determinants of diversification (Double-log regression): CEI is the dependent variable. Functional form: ln Y = a + b1 ln X1 + ... + bn ln Xn + U. Explanatory variables: fertilizer consumption; area sown more than once; annual rainfall; average landholding size; MSP of paddy; MSP of coarse cereals (analyzed as difference vs paddy); percentage of gross irrigated area to gross cultivated area (irrigation intensity); cropping intensity; credit to agriculture by SCBs; per-capita NSDP; per-capita power availability; length of national highways; number of factories; time; time squared. - Dynamics of cropping pattern (Markov chain): Cropping pattern data (1998-1999 to 2019-2020) used to estimate transition probability matrices among crops/crop groups. LINGO software used for estimation; visualization done in R. The standard Markov specification relates current-period shares to lagged shares via a transition probability matrix, with an error term assumed independent of lagged shares.

- District-level CEI (2019-20): Of 30 districts, 11 are high-diversity, 12 medium, 7 low. Low-diversity districts include Bidar, Davangere, Raichur, Yadgir, Dakshina Kannada, Kalaburagi, and Kodagu (with Kodagu lowest). Average CEI over 1998-99 to 2020-21 also places Kodagu, Dakshina Kannada, Yadgir, Udupi, and Shivamogga among the lowest. Some high-diversity districts have negative CAGR of CEI. - Trend by crop groups (1998-99 to 2020-21): CEI for cereals remained relatively constant, reflecting staple food security priorities. Pulses’ CEI declined (e.g., 0.728 in 2000-01 to 0.689 in 2014-15), attributed to lower profitability and losses from pests/animal attacks. - Cropping pattern shares (2019-20): Cereals and millets 36%; pulses 22%; oilseeds 13%; condiments and spices 7%; sugar crops 7%; fibres 6%; vegetables 4%; fruit crops 3%; drugs, narcotics and plantation crops 3%; dyes and tanning materials 0%. - Food vs non-food area share over time: Food crops increased from 68.45% to 76.46%; non-food crops declined from 31.55% to 23.54%. - Determinants of diversification (double-log model, significance): • Annual rainfall: coefficient −0.025 (1% level) — higher rainfall associated with less diversification (paddy dominance). • Average landholding size: −0.045 (5%) — larger farms more specialized; smaller farms diversify to manage risk. • Per-capita NSDP: −0.016 (5%) — wealthier contexts associated with specialization. • Difference MSP (coarse cereals − paddy): +0.052 (5%) — stronger incentive to diversify toward coarse cereals when MSP differential rises. • Irrigation intensity (gross irrigated area/net irrigated area): +0.030 (5%) — year-round water access supports diversification. • Credit to agriculture by SCBs: +0.021 (5%) — credit availability promotes diversification. • Length of national highways: +0.027 (5%) — better connectivity facilitates diversification. • Time: +0.008 (1%); Time squared: negative and significant — inverted U-shaped relationship, indicating recent slowing/decline despite earlier positive trend. • Other variables (fertilizer, area sown more than once, cropping intensity, per-capita power, number of factories) not statistically significant in this model. - Markov chain: direction and retention probabilities: • Cereals and millets (crop-level retention): maize 0.92; jowar 0.67; paddy 0.64; wheat 0.47; other cereals & millets 0.46; ragi 0.29. Bajra and barley have very low retention; bajra loses area to paddy, ragi, wheat; barley loses area largely to ragi. Net gain positive for ragi, jowar, paddy, maize; negative for barley, other cereals & millets, wheat—contributing to negative diversification trend within cereals/millets. • Pulses: retention highest for other pulses 0.93, then gram 0.61, arhar 0.55. Net gain positive for gram and arhar; other pulses has high retention but negative net gain due to shift to arhar (probability 0.07), explaining recent decline in pulses diversification. • Oilseeds: retention — sunflower 0.94; groundnut 0.83; soybean 0.80; coconut 0.77; niger 0.74; sesamum 0.34; castor 0.31; other oilseeds 0.21; linseed 0.16; safflower 0.16. Except groundnut, sunflower, safflower, and coconut, most oilseeds have negative net gains, contributing to negative diversification trends in oilseeds. • Overall crop-group retention: cereals & millets 0.85; oilseeds 0.81; sugar 0.81; condiments & spices 0.71; pulses 0.66; fibre 0.51; drugs, narcotics & plantation crops 0.48; vegetables 0.32; fruits 0.06. Except pulses, most categories (including cereals & millets) show negative net gains in retention probabilities. - Overall trend: Karnataka’s CEI shows a nominally positive long-run trend, but the near-zero coefficient and inverted U-shape indicate fragility and recent weakening of diversification.

The study addresses its objectives by quantifying diversification (CEI), identifying its drivers (econometric analysis), and mapping directional shifts (Markov chains). The relatively constant CEI for cereals and the decline in pulses’ CEI, combined with Markov evidence of low retention for specific cereals (bajra, barley) and many oilseeds, explain why diversification progress is tenuous. Higher rainfall areas remain paddy-oriented, while smaller farms diversify more—aligning with risk and specialization theories. Policy and infrastructural levers—MSP differentials favoring coarse cereals, irrigation intensity, agricultural credit access, and transport connectivity—exhibit significant positive effects on diversification, suggesting that price incentives and enabling infrastructure can shift cropping choices toward broader crop mixes. However, the inverted U-shaped time effect signals that earlier gains may be tapering, and without targeted interventions (especially for lagging groups like fruits and vegetables with low retention), diversification may stall. The documented shift toward food crops (and away from non-food) and the negative net gains in many crop groups point to structural constraints and risk-return trade-offs that need policy attention to sustain and broaden diversification.

Karnataka exhibits spatial heterogeneity in crop diversification, with Kodagu persistently lowest and districts like Dakshina Kannada, Yadgir, Udupi, and Shivamogga also low. While overall diversification has trended upward, the trend’s coefficient is near zero and recent dynamics suggest an inverted U-shape, indicating emerging headwinds. Markov analysis shows cereals and millets retain large area shares but suffer negative net gains, and many oilseeds also lose ground, which dampens diversification. Determinants analysis underscores that MSP differentials (coarse cereals vs paddy), stronger irrigation intensity, greater agricultural credit, and better highway connectivity positively influence diversification, whereas higher rainfall, larger farm size, and higher per-capita NSDP are associated with specialization. Policy should prioritize reliable irrigation, enhanced credit flows, and transport infrastructure, alongside calibrated price incentives (MSP differentials) to encourage shifts toward diverse, resilient crop portfolios. Future research could examine micro-level heterogeneity, profitability-risk profiles across crops, and the roles of storage, markets, and climate-resilient technologies in sustaining diversification.