Abstract
Governments in developing countries have attempted to reform inefficient agricultural markets through direct interventions, market-oriented approaches, and institutional mechanisms, with one of the aims being improving the lot of smallholder farmers. However, such interventions often fell short of yielding desired results, perhaps because broad macro objectives failed to take into account micro-level processes. This study examines smallholder farmers’ market selection decisions and attempts to establish if these have a bearing on legislative goals of agricultural marketing reforms. Analysis of survey data using binary logistic regression, reveals that farmers choose between two prominent marketplaces – APMC and farm gate – on the basis of factors such as perishability of produce, quality testing infrastructure at the marketplace, and services provided by buyers such as grading and sorting or transportation. Findings are reinforced by an extended part of the study where farmers ranked assumptive marketplace scenarios developed by incorporating different marketplace features. Results demonstrate the possibility of enabling systemic changes in agricultural markets by catering to grassroots decision mechanisms. Implications for policy and relevant players engaged in agricultural markets are discussed.
Introduction
An integral part of economic expansion is the transformation of agriculture and food system, particularly through diversification, specialization, and commercialization. Individual farms tend to become specialized as farmers move away from subsistence to commercialization, and consumers tend to shift towards diversified food baskets (Timmer 1997). Ideally, policy initiatives ought to accompany these changes to facilitate integration of subsistence agriculture with the market economy, since such harmonization paves way for development. However, despite various attempts by developing country governments to restructure inefficient agricultural markets, challenges continue to exist (Barrett and Mutambatsere 2008), plausibly due to limited understanding of grassroots decision-makers’ unique way of engagement with inefficient and weak markets in rural areas. Research on such themes, based on a practical political economy model, could help connect the missing dots between policy intent and outcomes (Timmer 1997). India is illustrative of a developing country where farm output diversified towards high-value products such as vegetables, fruits, milk, eggs and meat (Sharma and Jain 2011), even as rising incomes are driving consumption shift from cereal-rich diet to one rich in animal products, vegetables and fruits (Gandhi and Zhou 2014). Yet, imperfect agricultural markets burdened farmers with low price realization for their produce whilst consumers complained of high food prices (Chand 2012). Consequently, policymakers initiated progressive reform legislations that tried to overcome inefficiency of agricultural markets. Addressing these issues might require knowledge of agricultural commodity marketing practices prevalent in rural areas, which seem to persist irrespective of policy changes. Stern and Reve (1980) had posited that while governments or regulatory agencies set the external environment for a marketing system, which could affect interactions between actors engaged in a market transaction; the environment itself could get influenced by actors’ combined and institutionalized interactions.
This paper attempts to understand factors that influence farmers’ choice of a marketplace in the backdrop of a policy milieu aiming to induce competition in the agricultural marketing system by encouraging multiple marketplace alternatives to eventually improve systemic efficiency. The next section briefly explains evolution of agricultural marketing in developing countries, and describes the Indian context as an illustration. This is followed by a review of literature on factors influencing farmers’ marketing channel choice, and situating such choice in the political economy framework of distribution channels, proposed by Stern and Reve (1980) for investigation. The study reviews macromarketing literature as it explores decision processes as a component of the larger marketing system (Layton 2007, p. 238), as also micro-level organizational processes under macro-level control structures (Arndt 1981). This exercise is likely to yield granular insights into farmers’ marketplace decisions, which could be inducted as catalytic elements in agricultural marketing policy to benefit different stakeholders in agricultural markets. Outcomes could find application in agricultural marketing systems of developing countries, characterised by a large number of smallholder farmers, and comparable structural issues.
Agricultural Markets in Developing Countries and the Indian Case Study
Agricultural commodity markets in developing countries play a critical role in economic and developmental outcomes, which get undermined by inadequate transport and communication infrastructure, poor law enforcement, scarcity of formal credit, and reliance on personalized interactions for exchange (Barrett and Mutambatsere 2008). Governments in developing countries hence intervened to correct these shortcomings. These interventions unveiled in three phases. In the first phase, reforms focused on stabilizing incomes of growers, supplying food to consuming segments at low prices, and facilitating income redistribution (Giuliano and Scalise 2009, p. 1). Direct public intervention by means of subsidies and administered prices constituted the implementation strategy. However, such price controls often resulted in market distortion (Barrett and Mutambatsere 2008), leading to systemic inefficiencies. Per se, the agenda of the second phase of intervention included correcting market distortion and adoption of market-oriented policies of liberalization. This attracted more private players in the market, leading to greater competition and consequently, a general rise in commodity prices. However, government agencies continued to operate alongside private corporations; and where parastatals exited markets, they could not be substituted by the private sector due to poor physical infrastructure, credit constraints, high transaction costs and bureaucratic hurdles. The third phase of reforms focused on non-price, institutional mechanisms to improve market efficiency. Supermarkets, food processors, restaurants and other downstream actors in the agri-food chain altered market structures; yet oligopsonistic multinationals posed a threat to small growers and traditional processors (Barrett and Mutambatsere 2008). Regulatory focus shifted to protecting interests of smallholder farmers amidst emergence of new marketing channels and changing market structures. However, developing countries in general seemed hesitant in adopting agricultural market reforms due to lack of political will, pressure from trader groups on political actors, and disproportionately large dependence on agriculture for livelihood (Lundberg 2005, p. 156).
The evolution of agricultural marketing in India resonates with the above discussion. Much before formal markets appeared, agricultural produce was exchanged through markets existing in “institutional vacuum” (Thomas 2003, p. 3), and non-market mechanisms such as reciprocity and communal redistribution (Harriss-White 1996, p. 22). Rural weekly markets called “haats” constituted marketplaces for farmers, small traders and consumers. An estimated 22,000 such “haats” still exist, 15 per cent of which are regulated and managed by local governing bodies such as “panchayats” or municipalities (Planning Commission 2011, p. 8). Referred to as primary rural markets, these are similar to spaces owned and operated by municipal agencies where sellers sell fresh food (Visconti, Minowa, and Maclaran 2014, p. 350). Although important for the small producers, these markets suffer from high inefficiencies, lack infrastructure, and have limited number of buyers.
Whereas the earliest regulatory measures for agricultural markets could be traced to the Royal Commission on Agriculture in 1928, the Agricultural Produce Marketing Acts in 1960s laid foundations of formal, institutional mechanisms for farmers to sell their produce to traders in government-regulated marketplaces called Agricultural Produce Marketing Committee yards (APMCs) (Acharya 1998). APMCs were supposed to offer transparent auctions; registered trade participants; sorting, grading and packing facilities; and streamlined payment mechanisms. Each market yard had a notified catchment area and the first trade of any agricultural commodity produced in the notified area had to mandatorily take place in the APMC (Chatterjee and Kapur 2016, p. 3). In 2014-15, there were 2477 APMC market yards and 4843 sub-market yards regulated by the State Agricultural Marketing Boards (Department of Economic Affairs 2015). Parastatals such as Food Corporation of India (FCI) played a major role in food procurement and distribution.
Despite sound intentions, government interventions and regulations on stockholding, price controls and public procurement rendered domestic trade of farm commodities physically, legally, and institutionally restrictive (Bathla 2006). Farmers faced substantial transport costs, low bargaining power, and harassment by labourers while selling at the APMC. As a result, many farmers avoided selling at APMC. For instance, only 29 per cent of the sampled paddy farmers in Karnataka and just about 20 per cent of small and marginal farmers (with landholding less than 2 hectares) in Punjab sold at the APMCs (Chatterjee and Kapur 2016, p. 4). Smallholder producers often sold to various kinds of local buyers, outside the purview of any regulated market yard or regulation. Inefficiencies, shady trade practices, entry barriers for new buyers created by trader associations (Acharya 2006), insufficient facilities and a large number of rent-seeking intermediaries in the value chain (Chand 2012) led to an increasingly dysfunctional APMC system.
India’s Model APMC Act of 2003 sought to liberalize marketing of farm produce and enhance competitiveness in agricultural markets by ending monopsony of the APMCs. The reform agenda incorporated provisions for creation of alternative agricultural market yards by any public or private agency, contract farming, direct purchase of farm produce by buyers without the necessity of trading at APMC market yard, electronic trading of farm commodities, dismantling market yard catchment boundaries, simplifying taxation norms, inviting the private sector to establish market yards, enabling warehouses to be treated as marketplaces, and creating an electronic, country-wide unified agricultural market (Chand 2012). The model State/UT Agricultural Produce and Livestock Marketing (Promotion & Facilitation) Act (APLM Act) of 2017 was a renewed push to the reform agenda (Ministry of Agriculture and Farmers’ Welfare 2017), that brought within its purview marketing of livestock in addition to agricultural commodities.
Contract farming had become more common in India for a wide range of commodities (G. Singh and Asokan 2005); starting with PepsiCo’s famous initiative for tomatoes in 1990s in Punjab (S. Singh 2002). However, production risk typically rested with the farmer and there was limited recourse to mitigate it (G. Singh and Asokan 2005). Various state governments established direct farmers’ markets (such as “Apni Mandi” in Punjab or “Rythu Bazar” in Andhra Pradesh, HOPCOMS in Karnataka); even as corporates (ITC Limited) were developing new models such as e-Choupal (Planning Commission 2011, p. 14). In the late 2000s, Karnataka undertook pilot projects using information and communication technologies for online auction of agricultural commodities (Chengappa, Yadava, and Kumar 2012; Sinha and Ranjit 2010), followed by full-scale electronic integration of all APMCs across the State in 2014 to create a single, state-wide agricultural market (Aggarwal, Jain, and Narayanan 2017). While a slew of new marketing arrangements were being experimented, the reform agenda was generally found to be tardy in promulgating legislation as well as implementation (Ministry of Agriculture 2013).
Thus, from an era of laissez faire, to entrenched public sector involvement, to liberalization, followed by institutional orientation, Indian agricultural marketing emerges representative of the reforms story in developing countries. More pertinent to observe in the above discussion is the simultaneous existence of all old as well as new market models – from the highly localized periodic markets to pan-national e-markets – in the agricultural market ecosystem. These marketplaces, from farmer’s perspective, are nothing but multiple marketing channels. Though the presence of multiple channels is not novel, sellers have been known to use one or more channels for a variety of reasons, including but not limited to, reaching disparate customers or customer groups, differences in service outputs of channels, transaction costs and so on (Coughlan et al. 2007). The present context warrants an investigation into decisions of farmers relating to marketplace choice to evaluate if such decisions could hold a key to understand divergence from intended policy objectives.
Farmers’ Choice of Agricultural Marketplace and the Political Economy Framework for Distribution Channels
Review of literature on marketplace selection by farmers shows immense complexity and diversity in such decisions, typically influenced by factors specific to “country and institutional settings” (Zanello, Srinivasan, and Shankar 2014, p. 1237). Farmers could decide to participate in, or refrain from market-based exchanges (Panda and Sreekumar 2012); trade through one or more channels (Park and Lohr 2006); opt between specialized and conventional channels (Wollni and Zeller 2007) or direct and intermediated channels (Low and Vogel 2011). Farmers’ participation in market-based exchanges could be influenced by access to market information, road infrastructure and transport facilities, training, and grading, as observed in some parts of India (Panda and Sreekumar 2012). Additionally, rural infrastructure such as roads (Bhatia 1999) and market infrastructure such as weighing and grading facilities (Shilpi and Umali-Deininger 2008) also encouraged farmers to sell at specific marketplaces. Conversely, non-availability of product-specific facilities in marketplaces deterred farmers from selling at designated market yards (Manjunath and Kannan 2012).
Incentives offered by new channels in comparison to existing ones, and investment required to participate in certain markets also governed growers’ posture towards contractual arrangements or retail chains (Reardon et al. 2009). For instance, farmers in Thailand preferred marketing arrangements that relied more on trust and less on explicit contracts, though the latter could be made attractive by extending farm inputs and credit (Schipmann and Qaim 2011).
Studies also point to socio-economic conditions of growers governing marketing channel decisions. Factors such as age and education of the farmer, distance to market, quantity of produce, marketing agreements and farmer’s membership of an association affected vegetable growers’ choice of marketing channel in Swaziland (Xaba and Masuku 2012). Growers of organic produce, having less experience were more likely to sell through a single channel, while more experienced farmers with better bargaining and marketing skills were likely to opt for multiple channels (Park and Lohr 2006). Indonesian potato growers favourably considered buyer attributes such as prompt payment in cash, premium price for graded produce, and buyers’ commitment to honour the transaction (Umberger et al. 2015). Tomato farmers of Guatemala selling to supermarkets usually had relatively larger holdings, more capital for investment, and were specialized in commercial horticulture (Hernández, Reardon, and Berdegué 2007). Farmers were found to gain from price information at different markets when empowered with technological advancements, such as access to mobile phones. Though farmers did not sell at marketplaces where higher prices prevailed, price information enabled them to drive a better bargain for themselves at closer markets after considering transaction costs (Zanello, Srinivasan, and Shankar 2014). Smallholders in Benin made channel choices based on their ability to cope with quality compliance norms, particularly for fresh exports and processing of pineapples (Arinloye et al. 2015). Gender too played a role; women were more likely to sell commodities that generated lower revenues, while men sold commodities that fetched higher revenues, and often at formal markets (Njuki et al. 2011).
Thus, numerous micro-level factors seem to influence marketplace choice of growers, which are often dictated, conditioned and shaped by local contexts and buyer-seller dyadic relations, while the macro political environment appears to have only an indirect effect. Grassroots complexities and facilitative local processes could suffer penalization on the assumption of standardized models by centralized policy-making (Harriss-White 1996), leading to non-alignment of policy intent with practices. Comprehending micro-level processes and linking them to policy measures is critical in such situations. Stern and Reve’s (1980) political economy framework for distribution channels (further modified by Achrol, Reve, and Stern (1983)) presents a suitable theoretical framework for such inquiry as it emphasizes on bidirectional linkages between actors in a buy-sell relationship and the broader environment.
The political economy framework for distribution channels captures internal political economy of the channel through economic and socio-political structures and processes within the channel (Stern and Reve 1980). The economic and socio-political environment in which the channel exists and operates constitutes the external political economy. Relationships between various components of the framework are bidirectional. While Stern and Reve deliberated upon interactions and relationships within the channel, Achrol, Reve and Stern discussed the effect of external task environment on channel dyads. The task environment of the dyad could be primary (suppliers and customers of the dyad), secondary (comprised of the circle of suppliers and customers to the immediate set of suppliers and customers of the dyad, and their regulatory agencies), or macro environment (including social, political, economic and technological forces) that affects the primary and secondary task environments.
The external political economy in Indian agriculture, for a long time, focused on raising farm production as means to achieve welfare and poverty alleviation, through a set of public interventions and incentives. Sau (1973) argued that the core of political economy of Indian agriculture should be the Marxist-Leninist philosophy of class relations and state control. Perhaps, as a consequence, land reforms occupied a prominent place in Indian agricultural policy post-Independence for a few decades (c.f., Appu 1997; Binswanger, Deininger, and Feder 1993; Bandyopadhyay 1986). Other discussions in the domain included agricultural yield enhancements for development and poverty alleviation in rural areas (Bhalla and Singh 1997); subsidies and investments in agriculture, especially for fertilizers, irrigation, electricity and other inputs (Fan, Gulati, and Thorat 2008; Gulati and Sharma 1995); access to institutional credit for raising farm income and farm household consumption expenditure (Kumar et al. 2017), and so on. However, plateauing productivity of most crops and fragmenting landholdings redirected thought towards policies for innovative land reforms, access to technology, credit, and marketing (Fan and Chan-Kang 2005). Efficient marketing of agricultural commodities for raising farmers’ incomes and rural development drew attention only in the early years of the current millennium, and gained traction in the past decade (Chand 2012; Vijayshankar and Krishnamurthy 2012). Basis various studies and reports, a strong case arose to overcome systemic inefficiencies, encourage disintermediation and new marketing models, promote greater role for private sector, futures trading, contract farming, and so on (Chand 2017). This led to creation of the Model APMC Act of 2003 and the most recent draft APLM Act of 2017, which targeted ending monopsony of the APMCs by creating alternative marketplaces for farmers.
Positing that marketing channels, “not only adapt to their environments, but also influence and shape them” (Stern and Reve 1980), the political economy framework suggests that the seller-buyer dyadic interactions could be affected by the macro environment (consisting of social, political, economic and technological factors), and vice versa. In the present context, it implies that success of regulatory initiatives to create a competitive marketing system could be subject to favourable response of farmers to alternative marketplace arrangements. Although several prior studies have examined choice of marketing channels and their determinants in the context of smallholder farmers, exploration of the reverse impact of practices on policy outcomes have been largely absent. Insights drawn from this analysis are likely to assist align macro level objectives with micro level processes to enhance competition in the agricultural marketing system.
Method
Choice of a marketplace from a set of available alternatives provides information limited to the current alternatives alone. More recent policy changes, discussed in the previous sections, list new kinds of marketplaces besides features that add value to existing markets such as APMCs. Evaluating farmers’ opinion about these new features and components is likely to help incorporate desirable features, or make appropriate changes to policy. Such insights potentially hold key to facilitate entry of new market participants as well as service providers. Therefore, this study was arranged into two parts. The first part attempts to understand the factors that determine marketplace choice of farmer. The second part evaluates farmers’ perception of marketplace features, both existing as well as proposed. Accordingly, the following research questions were framed: Research question 1: What factors determine the choice of a marketplace by farmers when more than one choice is available? Research question 2: How would farmers rank different existing and new marketplace features, in the event of development of new marketplaces?
To answer the first research question, a large number of variables or factors influencing farmer’s marketplace choice were identified from literature review. This superset of independent variables was narrowed down based on insights obtained during pilot phase of the study, and from inputs of subject experts. While it is advisable to avoid researcher’s personal bias coming in the way of selection of independent variables, it is also essential to exercise utmost caution in understanding variables that should be included in the analysis, and shunning extraneous variables from entering the model (Osborne 2015, pp. 96-97). For instance, basis literature review, “availability of formal and informal credit in the village” constituted one of the independent variables for which data was collected. However, during the survey, it was found that both formal and informal credit could be availed in the village; and that informal borrowing from a buyer could have an impact on marketing decision. Since, this aspect was recorded as one of the services rendered by buyers, data collected on sources of credit was discarded during analysis. Similarly, the survey revealed that all except four respondents had access to irrigation. Hence, the variable recording availability of irrigation was dropped. This process resulted in two broad sets of variables, one set describing the farmer’s demographic and socio-economic condition, and the other set describing the marketplace. Overall, the independent variables which were logically related to the dependent variable of marketplace choice, were identified and organized into four blocks – farmer and farm-related attributes, marketing channel infrastructure, marketing costs, and services provided by buyers. Table 1 presents these vectors and the variables considered under them. Responses were recorded as continuous or categorical data, which were eventually converted into categorical for analysis.
Broad Parameters and Variables to Determine Marketing Channel Choice.
The survey was conducted in Karnataka, a state in south-western part of India, which happens to be among the earliest ones, in 2008, to pass legislation on agricultural marketing reforms. A cross-sectional sample of farmers across three districts (administrative units) of the state – Bidar, Davanagere and Kolar – was covered. Bidar and Davanagere were predominantly into cultivation of field crops such as pigeon pea and corn respectively whereas Kolar was a hub for vegetables such as tomato, cabbage, capsicum, and chili. With an intent to have an equal proportion of farmers growing field crops or staples and vegetables, the total sample was pegged at 200, with 100 farmers in each of the two groups. Within the field crops group, 50 were sampled from Bidar and 50 from Davanagere. Thus about half of the respondents had major income from horticultural crops (primarily, tomato), the rest obtained a major portion of their income from field crops (corn or pigeon pea). Villages at a distance beginning about 5 km from the APMCs of Bidar, Davanagere and Kolar, up to a distance of 30 km were randomly shortlisted. The sample was representative of the population in terms of land holding. Table 2 presents the sample distribution in proportion to landholding. Small and marginal farmers, with holdings less than two hectares constituted the largest group of respondents.
Sample Distribution based on Landholding.
To ensure achieving the planned sample size, a slightly larger number of respondents were interviewed. The final usable responses were 253, which included 55 from Bidar, 58 from Davanagere and 140 from Kolar. Overall, there were 107 farmers whose major source of income came from field crops, and 131 farmers depended on vegetables for a major source of livelihood.
For the second research question, four assumptive marketplace scenarios were developed, which farmers had to rank from most preferred to least preferred option. Eight marketplace features, comprising location; cleaning and drying facilities; certification facilities; warehousing; warehouse receipt financing; mode of sale; mode of payment; and cost of transaction, were drawn from existing marketplaces and the ones proposed in the draft APLM Act of 2017. Since exhaustive combinations of the eight parameters would lead to numerous scenarios, a final set of four scenarios was shortlisted after gauging ability of respondents, their ease of understanding and ranking, and taking expert opinion into consideration. The four scenarios developed from the eight features and options therein are presented in Table 3.
Marketplace Features to Build Marketplace Scenarios.
All the marketplaces were designed to present hypothetical scenarios, building in one way or another on extant markets used by farmers. The simplest marketplace scenario presented farm gate sale with added benefits of warehousing for two months after some manual cleaning and grading, and basic but scientific quality assessment and certification. The scenario indicated a negotiated sale with physical cash payment, even as farmers who wished to avail the warehousing and associated facilities had to pay. The most advanced scenario showed an improvised version of APMC with certification, warehousing, financing, and electronic sale and payment mechanisms. Farmers could avail warehousing facility for three to six months, and could also receive finance or loan against stored goods. Cleaning and grading cost was indicated as a percentage of the price the farmer would receive. The other two scenarios presented a hypothetical market yard at a distance of about five to 10 kilometres from the village and had a mix of different market elements. About 60 farmers, sampled based on their ability to understand and evaluate the scenarios, were asked to rank the marketplace scenarios, which were presented pictorially; eventually yielding 58 usable responses.
Results
Factors Determining Marketplace Choice
The survey brought to fore two prominent marketing channels or marketplaces used by farmers to sell their produce – APMC and Farm Gate (FG). The term farm gate implies a practice of farmers selling their produce right at the farm or in the village to small-time traders who act independently or as agents of registered traders of APMCs. Akinci, Kaynak, Atilgan and Aksoy (2007) in their meta-analysis, identified a host of studies, prominently including but not limited to, evaluation of alternatives or channel choice using binary logistic regression in consumer behaviour and international marketing. Further, several studies that were exploratory in nature and concerned a dichotomous dependent variable employed binary logistic regression as the analytical tool (for example, Peterson and Lambert 2003; Martin et al. 2003; Wieringa and Verhoef 2007; Farashah and Estelami 2014; Fiore et al. 2016). Whereas Gray and Kinnear (2012) advised the use of binary logistic regression using SPSS for such studies, Peterson and Lambert (2003) provided a detailed method of interpretation of results thus obtained. Explanation of the results of binary logistic regression is based on value of the regression coefficient (B), and the odds ratio (Exp[B]), with relevant reference to the significance level of associated Wald chi-square statistic. The odds ratio, which is ratio of probability of the event to the probability of the non-event, represents the change effected by a unit change in a particular independent variable, keeping other variables unchanged. Logistic regression also enables researchers to estimate the probability of occurrence of the event when values of the independent variables are known, using the equation (1/1+e-Z), where Z = B0 + B1X1 + B2X2 +…BnXn (Peterson and Lambert 2003).
In the present case, interpretation is restricted to explaining the relation between the dependent and independent variables, in accordance with the objectives of the study. Using entry significance criteria of p = 0.05 and removal criteria of p = 0.10 for Wald chi-squared statistic, blocks of variables through backward entry in SPSS logistic regression procedure returned a final model with Hosmer-Lemeshow test statistic of 3.072 with 8 df and a p- value of 0.930, indicating the model’s fit to the data. The model correctly classified 72.9% and 92.6% of the cases to predict marketplace choice as FG and APMC respectively.
Table 4 presents the variables that entered the final model along with their regression coefficients and odds. The final model comprised of independent variables from all blocks except those under “Marketing costs”. Of the different variables entered under the block “Farmer and farm-related attributes”, perishability of the major produce turned out to be a significant predictor. The regression coefficient (B) of -3.596 and odds value of 0.027 implies a negative correlation between the predictor and response. Odds of the marketplace choice being FG reduce by 2.7% when the farmer’s major produce is perishable. In the block relating to marketing infrastructure usage, the explanatory variable for sorting and grading infrastructure is significant with B value of -2.908 and Exp(B) value of 0.055. Implying a negative correlation between the predictor and response variables, it indicates that odds of selecting FG as marketplace over APMC reduce by 5.5% when a farmer evaluates his choice based on availability of usable infrastructure for sorting and grading the produce. Two predictors under the block “Services provided by buyers” – sorting and grading services, and transportation services – turned in significant at p=0.05. Odds of choice being FG reduce by 18.9% when the farmer seeks sorting and grading services from the buyer. However, when buyers provide transportation services, it tremendously favours FG being the marketplace option, and the odds in favour of FG improve by the highest factor of 20.221.
Results of Binary Logistic Regression Analysis for Choice of Marketplace.
To summarize, odds of farmers selecting APMC as the marketing channel increase when the produce is perishable; growers expect marketing infrastructure such as scientific quality testing, and buyer services such as sorting and grading services. The odds of choosing FG as marketplace improve enormously when farmers are keen on availing transportation services provided by the buyer. Further details on these results are presented in the discussion section.
Ranking of Marketplace Features
Assessing new marketplace features strengthens the first part of the study as it holds key to improve existing market yards or establishing new ones with such features that might potentially attract farmers. As mentioned previously, eight features of marketplaces were combined to create four marketplace scenarios and respondents were asked to rank them from most preferred to least preferred. The ranking data thus collected was analysed using the Kemeny-Young method. This method estimates the median or consensus ranking, which “minimizes the sum of distances between itself and all input rankings” (D’Ambrosio, Amodio, and Iorio 2015). Mathematical representation of Kemeny distance is as below:
Kemeny median is the distance between ranks Rs and Rt, m, the number of objects to be ranked, “x(s)ij is defined as equal to 1 if object i is preferred to object j in ranking s, equal to -1 if the reverse is true, and equal to 0 if the two objects are tied.” (Heiser & Ambrosio, 2013, p. 24).
What was presented as the most advanced marketplace scenario – APMC with facilities for quality testing, warehousing, certification, financing against stored goods, electronic sale and electronic payment – emerged as the most preferred choice. The second most preferred scenario involved the village as marketplace, with warehousing facility, negotiated sale between buyer and seller, and payment in cash. The other two scenarios which presented a new marketplace at a distance of 5 to 10 km from the village (APMC being about 25 km) were relegated to third and fourth ranks. Likely reasons for such ranking are discussed in the next section.
Discussion and Implications
This study of smallholder growers’ marketplace choice in India reflects an economic antecedent to their decision making even as several local practices and circumstances condition the decisions. Farmers were found to have access to prominently two marketplaces – APMC and Farm Gate. To reiterate, an APMC is a government regulated market yard whereas farm gate sale implies sale of farm commodities by farmers at their farms or at a specific point within the village to a buyer. Perishability of produce, availability of scientific quality testing infrastructure at the marketplace and provision of sorting and grading services by buyers at the APMC influenced farmers to choose selling at APMC. On the other hand, farm gate sale emerged as a preferred option when buyers provided transportation services. In both cases, farmers’ decisions seem to be driven by an economic intent, which is elaborated below.
Perishability necessitates that time between harvest and sale of a farm commodity is narrowed so as to mitigate loss on account of deterioration of quality and wastage. Despite several shortcomings, APMCs come across as an established ecosystem with buyers being present in the market yard and such ready availability of buyers ensures that the produce is easily sold at the APMC. This sounds counterintuitive since it is generally assumed that farmers would prefer farm gate sale of perishables to avoid losses, particularly in transportation. Field observations and enquiry revealed that farm gate buyers are usually small-time traders who act independently or as agents of registered traders at the APMCs. These traders travel from village to village, buy, and aggregate produce from small farmers for subsequent trade. Since buyers are itinerant, farmers do not have prior information about the time of transaction, rendering the condition unsuitable for sale of perishables. However, farm gate sale in case of non-perishables such as grains and pulses, offers upsides such as ability to postpone a sale if the offered price is low. Although farmers typically store locally, often using indigenous methods and storage structures; it improves their bargaining power, as farmers can reject a poor bid, besides saving costs of transport, handling and wastage incurred while transacting at APMC.
Scientific quality testing, sorting and grading facilitate more robust and objective assessment of quality of the commodity. Scientific quality testing implied use of moisture meters, and sorting and grading facilities largely referred to wire mesh sieves for grading and separating grains. Being mechanical and quantifiable, these services tremendously mitigate, if not eliminate malpractices that buyers might be resorting to during the transaction. Findings are supported by Shilpi and Umali-Deininger’s (2008) study where market facilities increase the likelihood of farmers selling at a market yard, particularly in case of smallholders.
On the other hand, growers preferred to sell at farm gate when buyers provided transport service. Logistics are an integral part of marketing, and transporting farm produce constitutes a major transaction cost besides involving drudgery, due to which farmers show inclination towards farm gate sale when buyers support with transport services (Fafchamps and Hill 2005). Some of the respondents had expressed similar views during the survey. Transport facility supported farm gate sale reduced costs of scouting for transport, handling, and wastage for the farmer. Farmers were aware that the price received by them in a farm gate transaction was always adjusted for transport costs, yet the perceived benefits seemed to outweigh perceived costs involved in shipping the produce to the nearest market. Such trade-off between transaction costs and prices has been indicated in other studies as well (for instance, Zanello, Srinivasan, and Shankar 2014). It is pertinent to note here that availability of mobile phones, and various sources of market and price information such as government sources, mass media, traders and so on is perhaps an essential precursor that facilitates the farmer to take a decision of selling outside what constitutes a formal and regulated marketplace. The survey revealed that most farmers or someone in the household did own a mobile phone. However, the availability of market and price information from various sources, including likely phone calls to family or friends or traders, rendered such sources of information statistically non-significant.
The second part of the study examined growers’ choice of marketplaces with different market features. Results corroborate with those of the first part as farmers most preferred a full-fledged APMC with advanced processes and services, followed by farm gate sale with added benefits of warehousing and basic cleaning and grading facilities. A strong case for agricultural commodity warehouses in rural areas was made under older policies, however, such schemes were largely oriented towards meeting national objectives of food grain preservation than market orientation to enhance farmers’ profitability (Dhingra 2016). Despite potential benefits to smallholder farmers from proper warehouses and warehouse receipt financing, such initiatives have not gathered scale (Mor and Fernandes 2009). Similarly, the most advanced scenario, which builds on the concept of electronic national agricultural market for agricultural produce, highlights the necessity of various mechanical, electronic and physical elements required to upgrade an APMC to a full-fledged electronic APMC (Roy, Joshi, and Chandra 2017). Thus, findings point to a felt need of farmers for better marketing infrastructure which has remained unmet despite policy prescriptions. To certain extent, these findings also appear to explain why concepts such as farmers’ markets could not achieve success while others such as contract farming or ITC’s e-Choupal became exemplary. Direct marketing initiatives were mostly aimed at bringing farmers closer to consumers, with hardly any facilitative services, limiting their success and scalability (Mittal 2007). Conversely, success of contract farming can be attributed to marketing side services such as grading and packaging (Eaton and Shepherd 2001). In case of ITC’s e-Choupal, aggregation hubs managed by the agents called “samyojaks” handled all logistics including storage and transportation (Ali and Kumar 2011). These findings echo other studies that have exhorted public investment in physical market infrastructure, including rural roads, warehouses and grading services at marketplaces for ensuring better returns to farmers (Ministry of Agriculture 2013; Sharma and Wardhan 2016). Additionally, extant APMCs may also be revamped to undertake functions of certification and warehousing of produce, than facilitating trading and settlement (Thomas 2003).
While the analysis highlights that smallholder growers, as sellers, accord greater importance to commercial considerations, which though logical, holds some deeper insights. It is explicit that farmers are keen to avail new or advanced methods and practices that could potentially safeguard their interests in a market transaction through lower costs or better returns. The non-explicit aspect is the socio-political context of the buyer-seller dyad. Farmers do not appear to be very confident about processes adopted by buyers, and hence an inclination towards more mechanical or electronic interventions to effect a sale. While farmers may be selling to the same buyer year after year on account of interlocked markets (as described in Bardhan 1980; Subramanian and Qaim 2011), there seems to be trust deficit between the farmer and trader due to conflicting goals or values (Stern and Reve 1980) or perception of dissimilarity by the farmer with the buyer (Mathews, Wilson, and Monoky 1972). Trust is an extremely critical element in marketing relations (Morgan and Hunt 1994), and mutual trust plays a greater role where asymmetry exists in relative influence among transacting parties (Smith and Barclay 1999). In agricultural settings, trust in market transactions has often been based on subjective norms (Batt 2004), or ethnic or religious ties (Lyon and Porter 2009). In a review of agricultural commodity markets in India, Jan and Harriss-White (2012) detailed in a section about the general exploitative nature of traders, particularly while dealing in market transactions with farmers, lending credence to the argument that farmers in general have a trust deficit with their buyers. Therefore, it is necessary to incorporate marketplace features that build trust through greater transparency and objectivity.
Farmers seem to value basic warehousing facility within the village, and more advanced facilities in the APMC. Farmers are likely to respond favourably when organizations communicate and render such services as a part of their offering, be it contract farming, direct markets, or private market yards. An efficient agricultural marketing system, particularly in developing countries, which has remained elusive could thus be achieved, not by more legislations and rules, but by emphasising on specific elements of marketing ecosystem that influence the most fundamental decisions of actors in a market exchange. Thus, a potential roadmap to bring about systemic changes by leveraging on decision processes of actors at the micro-level is presented.
Limitations and Future Research
As mentioned previously, marketing of farm produce is highly contextual and subject to unique local situations. A restricted sample size limited to a specific geography and two sets of crops due to financial and time constraints limits the generalizability of the findings. A larger sample covering a broader variety of crops, such as orchards and plantations, in different cultural and institutional settings could be explored for greater generalizable findings. The surveyed areas presented APMC and farm gate as the two prominent marketplaces, while eliminating instances of other marketplaces such as private market yards, farmers’ markets, or contract farming. Yet, this does not rule out the prevalence of such marketing arrangements. In fact, in the state of Karnataka, where the survey was conducted, contract farming for gherkins, baby corn and chili has been reported (Nagaraj et al. 2008). The presence of such marketplace choices in the survey could have given different results. Typically, researchers have used conjoint analysis in surveys that determine attributes or features. This study employed a few scenarios incorporating attributes of marketplaces and used Kemeny ranking for the scenarios. As anticipated, respondent farmers seemed to face difficulty in fathoming new or hypothetical marketplace scenarios, resulting in low rating to new models as compared to existing ones. However, low ranking could be attributed to local practices as well. For instance, hired transporters charged for fixed distances; a proposed marketplace within a given distance in such cases would not lead to reduction in transport costs. This study opens scope for deeper research in these aspects of agricultural markets. Being carried out from a farmers’ perspective provides only the seller-side economic and socio-political processes without assessing buyers’ preferences or reactions. For instance, creation of new marketing arrangements such as private market yards or e-markets for a buyer or any third party would also depend on the economic viability, besides meeting sellers’ requirements. Bringing about a meeting ground on the basis of smallholder producers’ preferences as evidenced by this study with the predilections of buyers opens scope for further research in this area.
Conclusion
Using Stern and Reve’s (1980) political economy framework for analysis of distribution channels, this paper attempted to examine links between agricultural marketing reform efforts of governments in a developing country context such as India with the decision processes of farmers, positing that deviations from intended outcomes of legislations could be influenced by seller-buyer dyadic interactions. Survey of smallholder farmers in Karnataka, India presented perishability of produce, availability of scientific quality testing infrastructure, and sorting and grading services extended by buyers as statistically significant predictors of APMC as the marketplace choice. Provision of transport services by buyers was a significant predictor of farm gate as preferred marketplace. An extended part of the study also established preference for advanced mechanized and electronic marketing infrastructure. Growers appear disposed towards objective assessment of quality of farm produce instead of extant methods adopted by buyers; possibly and latently conveying deficit of trust with such practices. Marketplace alternatives that provide such facilities and services are more likely to yield a positive attitude of farmers. Accordingly, policy focus has to shift from merely promoting alternative market avenues to emphasising on upgraded facilities that improve effectiveness of a transaction and building trust among farmers. Realigning the focus thus, might hold key to success of the alternatives, consequently inducing competition, and leading to an efficient agricultural marketing system. The study opens up scope to assess links between marketing infrastructure elements and farmer-trader trust in various contexts of agricultural markets defined by geographies and kinds of farm produce.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
