Seasonality, Volatility, and Market Fragmentation: Evidence from India\'s Little Millet Mandis

Abstract

This article examines the scale, seasonality, volatility, and spatial price integration of little millet (Panicum sumatrense) markets in India using Agricultural Marketing Information Network (AGMARKNET) arrivals and price records. National and subnational market structure is assessed over 2007-2024, while district-level price integration is evaluated for Kabirdham, Chitradurga, and Dharwad over January 2017-December 2024. The analysis combines arrival-weighted prices, seasonal-trend decomposition using LOESS, Parkinson range-based volatility, augmented Dickey-Fuller unit-root tests, and static and rolling Johansen trace tests. Cumulative arrivals are highly concentrated: Karnataka accounts for 48.7 percent, the three leading states account for 90.6 percent, and Hosadurga mandi alone accounts for 34.1 percent. Annual arrivals decline from the high-volume early years before a marked 2024 recovery, whereas nominal prices rise substantially. District series display recurring seasonal components but also large irregular shocks, consistent with discontinuous trading. All price levels are non-stationary and log returns are stationary. The static Johansen statistic for no cointegration is 23.74, below the 5 percent critical value of 34.91, and every rolling-window statistic also remains below its rejection threshold. The evidence therefore identifies a concentrated yet spatially fragmented market in which aggregation at a leading hub has not produced a stable interdistrict price system.

Introduction

The text presents an empirical analysis of little millet (Panicum sumatrense) markets in India, focusing on market concentration, seasonality, price volatility, and spatial price integration. Its central argument is that production potential and market presence do not necessarily imply a well-integrated or efficient market.

The study uses AGMARKNET data from 2007–2024 to examine little millet arrivals and mandi prices. It focuses more closely on Chitradurga and Dharwad in Karnataka and Kabirdham in Chhattisgarh, using monthly price data from 2017–2024 for spatial-integration analysis.

Main findings

• Market activity is highly concentrated. Karnataka accounts for about 48.7% of recorded arrivals, followed by Madhya Pradesh (27.1%) and Chhattisgarh (14.8%). These three states together account for more than 90% of recorded arrivals.
• At the mandi level, concentration is even stronger. Hosadurga alone contributes about 34.1% of total recorded arrivals, while the five largest mandis account for approximately 73.7%.
• Some reported markets, such as Morbi and Mallanwala, have substantial recorded quantities but only one active trading day, suggesting that their observations may represent episodic transactions rather than continuously functioning markets.
• Nationally, recorded arrivals fluctuate substantially, while nominal volume-weighted average prices show a broad upward trend. The study cautions that this divergence could reflect scarcity, demand changes, inflation, or changing quality composition and cannot establish causality by itself.

Seasonality and volatility

Using STL decomposition, the study finds recurring seasonal patterns in arrivals and prices, but large irregular or episodic movements are also important.

• Chitradurga is the largest and deepest of the selected markets but still experiences substantial irregular arrival and price movements.
• Dharwad is considerably thinner, with lower and more discontinuous arrivals and relatively pronounced price fluctuations.
• Kabirdham has declining long-run arrival volumes despite rising nominal prices.

The Parkinson range-based volatility analysis shows that volatility is episodic rather than uniformly high. Chitradurga experiences the largest individual volatility spikes, while Dharwad shows instability associated with its thin trading base.

Spatial market integration

The study examines whether prices in Chitradurga, Dharwad, and Kabirdham share a stable long-run relationship.

The ADF tests indicate that the three price series are non-stationary in levels but stationary in first differences. Thus, the series are treated as I(1), making Johansen cointegration analysis appropriate.

The study then uses both:

• full-sample Johansen cointegration, and
• rolling-window Johansen tests

to determine whether spatial integration is stable over time rather than assuming that one full-sample result represents the entire period.

Institutional interpretation

The paper also considers the introduction of Mission Millet Chhattisgarh in 2021, which included procurement and processing functions for kutki. However, it explicitly avoids claiming that the programme caused changes in Kabirdham's arrivals or volatility because the analysis lacks a counterfactual and formal causal identification.

Conclusion

This study finds that India\'s recorded little millet trade is both concentrated and spatially fragmented. Karnataka supplies nearly half of cumulative AGMARKNET arrivals, the three leading states supply more than 90 percent, and Hosadurga alone contributes more than one third. Yet the district price system does not display a stable long-run relationship. ADF tests establish that prices are integrated of order one, while both static and rolling Johansen tests fail to reject no cointegration. The descriptive evidence further shows why physical-market structure matters for interpreting prices. Annual arrivals weaken after the early high-volume years before recovering in 2024, while nominal VWAP rises sharply. STL decompositions show recurring seasonal patterns but also large market-specific remainders, and Parkinson indices reveal episodic volatility even in the dominant trading district. Together, these results describe a market in which cumulative volume is concentrated but continuous liquidity and spatial arbitrage remain limited. The analysis is bounded by the coverage of reporting mandis, nominal prices, and the absence of direct information on trade flows, grades, production, and procurement quantities. These limitations do not weaken the central result; they define its scope. Within the observable mandi system, concentration has not generated a common interdistrict price equilibrium. Deepening transparent, regularly reported trade is therefore essential if growing interest in little millet is to translate into stronger price discovery and more reliable market participation.

References

[1] M. Vetriventhan et al., \"Genetic and genomic resources, and breeding for accelerating improvement of small millets: Current status and future interventions,\" The Nucleus, vol. 63, pp. 217-239, 2020, doi: 10.1007/s13237-020-00322-3. [2] A. Kumar, V. Tomer, A. Kaur, V. Kumar, and K. Gupta, \"Millets: A solution to agrarian and nutritional challenges,\" Agriculture & Food Security, vol. 7, art. no. 31, 2018, doi: 10.1186/s40066-018-0183-3. [3] G. Gruere, L. Nagarajan, and E. D. I. O. King, \"The role of collective action in the marketing of underutilized plant species: Lessons from a case study on minor millets in South India,\" Food Policy, vol. 34, no. 1, pp. 39-45, 2009, doi: 10.1016/j.foodpol.2008.10.006. [4] A. Adekunle, D. Lyew, V. Orsat, and V. Raghavan, \"Helping agribusinesses-small millets value chain-to grow in India,\" Agriculture, vol. 8, no. 3, art. no. 44, 2018, doi: 10.3390/agriculture8030044. [5] P. L. Fackler and B. K. Goodwin, \"Spatial price analysis,\" in Handbook of Agricultural Economics, vol. 1, B. L. Gardner and G. C. Rausser, Eds. Amsterdam, The Netherlands: Elsevier, 2001, pp. 971-1024. [6] B. K. Goodwin and T. C. Schroeder, \"Cointegration tests and spatial price linkages in regional cattle markets,\" American Journal of Agricultural Economics, vol. 73, no. 2, pp. 452-464, 1991, doi: 10.2307/1242730. [7] G. E. Shively, \"Price thresholds, price volatility, and the private costs of investment in a developing country grain market,\" Economic Modelling, vol. 18, no. 3, pp. 399-414, 2001. [8] Directorate of Marketing and Inspection, Government of India, \"AGMARKNET 2.0: Agricultural market price and arrival information.\" [Online]. Available: https://agmarknet.gov.in/. Accessed: Aug. 16, 2026. [9] Chhattisgarh State Minor Forest Produce (Trading and Development) Cooperative Federation Ltd., \"Mission Millet Chhattisgarh: Procurement, processing, packaging, and distribution tender notice,\" Raipur, India, 2024. [Online]. Available: https://www.cgmfpfed.org/. [10] R. B. Cleveland, W. S. Cleveland, J. E. McRae, and I. Terpenning, \"STL: A seasonal-trend decomposition procedure based on LOESS,\" Journal of Official Statistics, vol. 6, no. 1, pp. 3-73, 1990. [11] M. Parkinson, \"The extreme value method for estimating the variance of the rate of return,\" Journal of Business, vol. 53, no. 1, pp. 61-65, 1980. [12] D. A. Dickey and W. A. Fuller, \"Distribution of the estimators for autoregressive time series with a unit root,\" Journal of the American Statistical Association, vol. 74, no. 366a, pp. 427-431, 1979, doi: 10.1080/01621459.1979.10482531. [13] S. Johansen, \"Statistical analysis of cointegration vectors,\" Journal of Economic Dynamics and Control, vol. 12, nos. 2-3, pp. 231-254, 1988, doi: 10.1016/0165-1889(88)90041-3. [14] N. Mylonidis and C. Kollias, \"Dynamic European stock market convergence: Evidence from rolling cointegration analysis in the first euro-decade,\" Journal of Banking & Finance, vol. 34, no. 9, pp. 2056-2064, 2010, doi: 10.1016/j.jbankfin.2010.01.012.

Copyright

Copyright © 2026 Prasanna Surathkal, B. H. Nagoor. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.