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

A+| A| A-

Small Farmers and Organised Retail Chains in India

This study compares farmers selling vegetables to Mother Dairy, an organised retail chain, with those selling to the local mandi in Haryana to find out the drivers and constraints determining their participation in these two types of marketing chains, particularly for the small farmers. The findings suggest the significance of farm size in determining farmers’ participation in organised retail chains. Using Heckman selection–correction model, the study found that though the income of participating farmers increases, the increase depends on farm size, while the Ginni coefficient shows that the inequality in income distribution is more among the participants than the non-participating farmers.

The world over rapid rise of organised retail chains (ORCs)1 has been transforming the agricultural food marketing system progressively (Dries et al 2004).2 However, this transformation in India has been slow, both at the upstream and downstream of the supply chain. The size of the food retail market in India was estimated to be large, where the share of agricultural food retailing is growing faster (GoI 2007; NABARD 2011). On the one hand, ORCs are seen as an ­effective channel to link farmers with markets, while, on the other, concerns have also been raised about its impacts on the farmers, particularly the smallholders.

One of the concerns is that agribusiness firms deal mostly with relatively large farmers and exclude the smaller ones. The exclusion of the small farmer from relatively liberalised markets and contract farming can lead to more concentrated landownership and displacement of the rural poor (Key and Runsten 1999).3 The benefits distribution within the rural community by new marketing channels like contract farming can have important implications for their economic and social differentiations (Korovkin 1992). This concern is particularly relevant in case of India where, number of small farmers is on the rise and their farm sizes are shrinking,4 which is resulting in decline of marketable surplus capacity. The small and marginal farmer households earn less than what they spend,5 and half of them are indebted, and most of them live in severe poverty (Kumar et al 2011). Although farmers’ participation neither guarantees its benefits nor insures them against its risks, non-­participation excludes them from its potential benefits, thus increasing inequalities.

Further, the issues, such as high quality standards, high ­rejection rate, procurement of few crops, partial procurement of produce, and delay in payments, may affect income of the participating farmers. The counter-view is that higher prices for better quality, low waste, diversification towards high ­value crops and reduction in marketing cost may improve income of the participating farmers. Besides, inputs and other services by ORCs to farmers can improve their productivity, thus improving their income.

Some of the agriculture produce marketing challenges in ­Indian agriculture have been listed in government reports, ­including the inter-ministerial committee of the government of India and the working groups for the Twelfth Five Year Plan. Inadequate provisioning of regulated markets forces farmers to bear disproportionately higher marketing cost.6 Long distance travels and poor logistic support create huge wastage of agricultural produce. Large number of rural markets are still deprived of weighting, measuring, sorting and packaging ­facilities.7

Moreover, the tax and licencing has increased the trans­action cost and put barriers to entry for market agents.8 The undue regulation of markets has also prevented private investment in marketing infrastructure, post-harvest management, grading and packaging. Regulation has also hampered development of alternative marketing channels in India (Patnaik 2011). Overall, the price efficiency in India has also been low, especially in case of the vegetables.9 The agriculture markets in India are also not well integrated, regionally, vertically and temporally (Acharya 2004). Reforms in agriculture marke­ting in India have been slow.

Since 1950, agriculture and agriculture marketing being a states subject, states have enacted Agricultural Produce Market Committee (APMC) Acts to regulate agriculture ­markets. Its objective was to protect farmers from the exploitation of intermediaries and traders to ensure better prices and timely payment for their produce. The National Com­mission on ­Agriculture, 1976 reviewed the performance of regulations of agriculture markets and found that regulated markets (70% of secondary or terminal markets) have benefitted farmers by preventing trading malpractices, such as unautho­rised market charges, falsification of weights and measures. The commission therefore inter alia recom­mended: (i) establishing a market within a radius of 5 kilometres (km); (ii) bringing unregulated assembly, terminal and even primary markets under regulation; (iii) constitution of a market committee to supervise the market as per rules and regulation; (iv) providing ­facilities of weighting, grading and storing in each regulated market; and (v) licensing market functionaries, like commission agents and trader. But, over the year regulated markets have failed to yield the desired results. As per the Shankerlal Guru Committee, 2001, the regulated agriculture market has res­tricted marketing in India. Later on, efforts have been made to reform the APMC Act, in 2003, and also to promote the direct marketing as an alternative marketing structure. Recently, many organised retail chains, both private and public, have started operations. Of these, Mother Dairy, is a government enterprise, working since 1985.

Against this backdrop, this study compares farmers selling vegetables to Mother Dairy, an organised retail chain, with those selling to local mandi in Haryana, to find out the drivers on their participation in the Mother Dairy chain and consequently the impact on their income, particularly for the small farmers. The study provides evidences that may help in policy decisions on providing institutional mechanisms to make the supply chains inclusive. ­Besides, it contributes in the debate on impact of ORCs on farmers’ income, which may help in policy decisions to devise appropriate institutional framework to restrain the exploitative character of the ORCs. Moreover, the study, unlike many studies in the case of India, uses improved methods to remove the possible selection bias in the sample. The study also discusses the sources of the ­income impact along with a comparison of income distribution between the farmers participating in the farmers market association (FMA farmers henceforth) and those not participating therein (non-FMA farmers henceforth).

Background

Kumar (2006) observed that private agribusiness firms in Punjab operated contract farming more effectively, with positive outcomes for the farmers irrespective of the farm size, that while the state corporation-led contract farming seems to ­favour only those farmers with larger farm who do not benefit as much as direct contract farmers. In absence of representative farmer organisations, contract farming has limited regional and local impact in terms of the inclusion of small farmers (Porter and Howard 1997; Key and Runsten 1999).

The participation of the small farmers in the supply chain depends on their relative advantage or disadvantages. Among advantages, a small farmer operating predominantly with family labour can save on the cost of labour supervision, cost of monitoring, screening of hired labour, cost of contract ­enforcement and cost of negotiation (Key and Runsten 1999). On the other hand, there are disadvantages for small farmers that arise out of their low marketable surplus, low bargaining power and low capacity to invest. Ghezan et al (2002) argued that the factors affecting a small farmer’s access to new marketing channels include low marketable surplus, difficulties in meeting volume, quality and delivery requirements, lack of ­liquidity to withstand the long payment delays and lack of ­access to market information.

The small farmer would be interested in contract farming because it facilitates modern inputs, which are normally unavailable or are more expensively obtained through other sources (Porter and Howard 1997). On the other hand, a firm would prefer dealing with large growers to avoid the complexities of dealing with a large number of small farmers (Glover and Kusterer 1990) and by looking at the large farmers’ investment capacity, risk bearing ability and relatively uniform quality of land.

Ghezan et al (2002) found that in Argentina, supply chains dominated by multinational firms producing frozen French fries, tended to favour medium and large potato farmers, ­excluding the smallholders. High quality standards imposed on the suppliers work as an entry barrier for small growers (Gutman 2002). Deshingkar et al (2003) found that the benefits of government-sponsored schemes in horticulture are reaching the bigger farmers rather than the smaller farmers and landless households. Similar observations about the ­challenges for the small farmers have been made in Costa Rica by Alvarado and Charmel (2002). It has also been witnessed in Africa that producers faced challenges in meeting the tough quality and safety standards, and the requirements to make investments and adopting new practices (Weatherspoon and Reardon 2003; Faiguenbaum et al 2002).

Meeting high quality standards set by ORCs hampers participation of the small producer. The rise in the fixed cost component of the cost of exchange also works as an entry barrier for the small farmer. The exclusion becomes more pronounced when the credit market is imperfect and the cost of borrowing is high for the small farmers (Page and Slater 2003). But new institutions, for example, fair trade companies and cooperatives are helpful in improving the participation of the small producer (Page and Slater 2003).

Reardon and Swinnen (2004) argue that the rise of ORCs brings opportunities for small farmers because these offer a path into high-quality and high-value markets. Their observation also hints that the transformation in the agricultural food system is inclusive of more small farms than it was expected. The exclusion was expected based on the arguments of transaction costs and requirement conditions. The assistance by processing firms to large and small suppliers is overcoming the obstacles in investing and improving quality because few farms can deliver the required quality, which is forcing the ­retail chains to integrate vertically to secure a high-quality supply base (Reardon and Swinnen 2004).

Glover (1984) surveys literature on contract farming to ­examine its bearing on farmers’ welfare, including the issue of participation of the farmer. The study remains inconclusive and argues that in general agribusiness firms prefer large farmers, but most deal with whoever is available, while some look for small farmers. Neven and Reardon (2004) found that supermarkets were not excluding small farmers from supplying to the markets in Kenya in the initial stages of inception. Sutradhar (2014) found that farm size was not a significant entry ­barrier in the participation in Reliance Fresh retail chain in ­Rajasthan in 2011. Miyata et al (2009) conducted a survey of 162 farmer households in Shandong province in China during 2005 to study the impact of contract farming on income of small farmers. They found little evidence to support the ­hypothesis that firms prefer larger farmers over small ones.

Minot (1986) found positive impact of contract farming on income of farmers. Similar observations have also been made by Porter and Howard (1997), in a review of studies conducted on contract farming in Africa. In the Indian context also, a study by Birthal et al (2005) for dairy products found significant improvement in the gross margin of those farmers who participated in ORC. Singh (2002) studied models of contract farming in Punjab and highlighted that despite problems in the models of contract farming, the income of the participants has improved. Studies by the Joseph et al (2008) and Chengappa and Nagraj (2005) provide some leading observations on the impact of ORC on income. In a more rigorous analysis, Sutradhar (2014), found that cauliflower farmers in Rajasthan selling through Reliance Fresh have been able to raise their net revenue per acre significantly, while no such impact was seen for other crops.

The literature on participation of small farmers in ORCs broadly indicates that contracting firms/ORCs prefer to deal with relatively large farmers. However, studies have also indicated that farm size is not a significant barrier in participation. Similarly, there is a broad consensus that income of the ­far­mers participating in the ORCs would improve, however, evidences to support the case for small farmers are not ­prominent.

Data and Methodology

A field survey of 398 farmer households255 linked to Mother Dairy and the rest dependent on local mandifrom 19 villages from Haryana was conducted through structured questionnaire during the summer of 2009 (see Table 1 for details).10 The surveyed districts—Sonipat, Panipat, Karnal and Kurukshetra—are mostly connected to Delhi through National Highway 1 (NH-1), where most of the retail outlets of Mother Dairy are ­located. The surveyed villages are mostly located around 5 km to 30 km distance from NH-1, and in proximity of a town, ­having a vegetable mandi within a maximum radius of 20 km. The state, districts, and villages were purposively selected keeping the procurement operation of Mother Dairy into consideration. The farmer households linked to Mother Dairy and the other farmer households were selected randomly from list of farmer households. Mother Dairy was preferred for the study because of its wide network and long-standing and ­stabilised operations.

A non-FMA village is a nearest located village to FMA village where farmers were supplying vegetables to local mandi. For the selection of the non-FMA farmers, farmers are listed in each non-FMA village recording their basic characteristics and a sample of farmer households was drawn randomly.

Income is defined comprehensively, as net household income (NHI), which includes not only farm business income11 but also subsidiary income or non-farm income.12 Because growing more of contracted crops may result in withdrawal of resour­ces from other crops or non-farm activities, which could result in income forgone. Moreover, it is the overall household ­income which determines expenditure of household on food, clothing, etc, which determines the level of poverty. Since, the expenditure by a household increases with the household size, thus, to pin down the impact of income on poverty, the per capita income of a household is preferred over per acre or per household income.

The econometrics procedure of estimation includes estimation of PROBIT, ordinary least square (OLS) regression and Heckman selection–correction model. The PROBIT model is estimated to identify the factors determining the partici­pation of farmers in ORC. Thereafter, OLS regression and Heckman selection–correction model estimated to know the impact of participation on income of the farmers. The Heckman selection–correction model is used to know the bias, if any, in the results, as the sample is not random. Besides, to overcome the possibility of bias in impact arising out of some unobservable characteristics of the farmers Heckman selection–correction model is used along with regression. The model is specified as follows:

Yi = Xiβ + μ1i outcome equation ... (1)

Ti = (Ziγ + μ2i>0) participation equation .. (2)

where Y is the outcome (per capita income) and X is a vector of the independent variables, while in participation equation Ti is the binary variable take value 1 if participated and 0
otherwise; while Z includes variables that predict whether or not a farmer would participate in ORC. It may be noted that the Z and X may include common variable, and which are taken identical in some studies (Gronau 1974). The selectivity problem is ­defined as:

E[Yi | Xi ,Ti = 1] = Xi β +E[μ1i2i > - Ziγ] ... (3)

Expected value of Yi for observations where farmers have participated into Mother Dairy is defined above. The joint ­distribution of random disturbance term of outcome (μ1i) and participation equation (μ2i) can be written as follows:

μ1i = (σ21/ σ22)* μ2i + υi ... (4)

where σ21 is the covariance of the unobservables of the outcome and participation equations (σ22) is the variance of the unobservable in the participation equation, and υi is assumed to be uncorrelated with the unobservable of participation equation (μ2i). Now since we know the unobservable for outcome equation (μ1i), we can also calculate its expected value which is defined as follow:

E[μ1i | μ2i> - Ziγ] = (σ21/ σ22) E[ (μ2i 22) | (μ2i / σ22 )> - Ziγ/ σ22] = (σ21) ф (Z/ σ22)/ σ22Ф (Ziγ/ σ22) ... (5)

where ф(.) is the standard normal density and Ф(.) is its
cumulative distribution function. The selectivity bias is said to occur wherever σ21 is not zero. The presence of this bias in the models arises due to presence of omitted variables into the original model (1), where the quantity is the omitted variables, also called the Inverse Mills Ratio (IMR), which is defined as:

IMR = ф (Z/ σ22)/ Ф (Z/ σ22) ... (6)

The treated equation, or Heckman selection–correction model, is defined as

Yi = Xiβ + [ф (Ziγ/ σ22)/ Ф (Ziγ/ σ22)]σ ... (7)

where

σ = (σ21/ σ22) which is coefficient of IMR

The estimated coefficients are consistent in Heckman selection–correction model. The Stata software reports lambda, sigma and rho. Rho is correlation coefficient between the unobservable that determines selection equation and the unobservable that determines outcome in outcome equation. Sigma is the adjusted standard error for the outcome equation and lambda is the selection coefficient = sigma * rho. The Average Treatment Effect (ATE) is computed as lambda *average IMR [or exp (ATE) -1)*100 if variable in log form] which is interpreted as how much conditional outcome is shifted up (or down) due to selection or truncated effect. The ATE depends on the statistically significant value of the Chi-square.

The inequality of income distribution is measured using Gini coefficients and Lorenz curve. The value of Gini coeficent ranges between zero and one, where zero shows perfect equality, while one means the most unequal distribution of the variable.

Results and Discussion

Mother Dairy Fruit & Vegetable is a wholly owned subsidiary of National Dairy Development Board (NDDB). It procures large a number of seasonal fruits and vegetables from thousands of farmers across a number of states in India. In Haryana, fruits and vegetables are procured through farmers’ marketing associations (FMA) at the upstream level of the chain, which are sold through Safal outlets spread across National Capital Region (NCR) at downstream. Mother Dairy has distribution centres at Pallabakhtavarpur and Mangolpuri in NCR, which are main coordinating locations having installed a huge infrastructure for storage, processing and logistic facilities.

Most of the procurement centres in villages are maintained by the FMAs in Haryana. Any farmer who has land (no restriction of size), grows fruits or vegetables and is ready to supply, can become member of the association. The objectives of the association are to enhance productivity of fruits and vegetables by provi­ding modern techniques, machines, access to inputs, information, crop protection and crop production programmes. It orga­nises farmers, takes decisions, monitors their actions, enables procurement operations, builds trust and ensures quality. The association is also responsible for procurement of fresh and quality vegetables from growers and transporting it to Mother Dairy. The member farmer of the association elects one president, whereas the secretary, who oversees all procurement operations and maintains records, is appointed by Mother Dairy. On daily basis, the produce brought by farmers is loaded in a vehicle after quality check, weighing and packaging, and then transported to the distribution centre of Mother Dairy. The final quality check is carried out by Mother Dairy at its distribution centre, and the status about rejected percentage and price assigned to the consignment is conveyed to the farmers usually next day of the procurement. Payments are made through the bank account, and usually take more than a week’s time.

Characteristics of surveyed households: The household characteristics are presented in Table 2 (p 17). There are about six persons in an average household; the difference between fma and non-fma farmers is statistically significantly. Proportion of adult (more than 18 years) members is also significantly larger in non-fma group than fma. Average age of fma farmers’ household head is than less compared to the non-fma group. However, education in both the groups is low and does not differ much. These groups also do not differ in terms of agricultural fixed assets (other than land), ownership of cattle and ­vehicle. The fma farmers have some advantages in terms of net operated area, leased in land and area under vegetables. The leased-in area seems to be playing a role in increasing operated area for fma farmers. The cropping intensity is significantly higher for non-fma farmers. The use of inputs such as family labour and biochemicals is higher in fma farmers, while machine labour, irrigation is higher in non-fma farmers, the differences are statistically significant. Marketing cost is lower for fma farmers than the compared group. The value of output is higher and statistically significant for the fma farmers. However, their productivity does not differ significantly from non-fma farmers. The area and value share of vegetables is significantly higher in the case of fma farmers than others. Similarly, the net household income and farm business income are higher for fma farmers, and so is the net household per capita income, and these are statistically significant, too. Off-farm income, however, is higher in non-fma farmers.

Dear Reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here

Updated On : 7th Feb, 2021
Back to Top